Flower Development: Methods and Protocols (Methods in Molecular Biology, 2686) [2 ed.] 1071632981, 9781071632987

Table of contents :
Preface
Contents
Contributors
Part I: Review and Overview Chapters
Chapter 1: Flower Development in Arabidopsis
1 Part 1: The Floral Transition
1.1 External Parameters Influencing Flowering-Temperature and Photoperiod
1.1.1 Temperature
1.1.2 Photoperiod
1.2 Internal Cues-Metabolism and Age-Dependent Pathways
1.2.1 Sugar and Nitrate Metabolism
1.2.2 Age-Dependent Pathway
1.3 Integration of the Different Input to Promote Flowering and Flower Formation
2 Part 2: The First Hours of Arabidopsis Flowers
2.1 Primordium Positioning and Auxin Signaling
2.2 Auxin Biosynthesis and Transport
2.3 Downstream of Auxin Signaling
2.4 Boundaries and Polarization Set the Limit of the Floral Primordium
2.5 Establishment of a Transient Stem Cell Identity
2.6 Modifications in Cell Wall Composition Trigger Floral Outgrowth
3 Part 3: Determination of the Floral Primordia Identity
3.1 Activation of the Floral Meristem Identity
3.2 Repression of the Inflorescence Trait
3.3 A Common Early Regulatory Network Between LFY and AP1
3.4 The Genesis of the Boundary FM-SAM
4 Part 4: Establishment of the Floral Patterning-from the Historical ABC to the Revised ABCDE Model
4.1 Sepal Identity and Maintenance
4.2 Sepal Growth
4.3 Petal Identity and Maintenance
4.4 Petal Growth
4.5 Stamen Identity and Maintenance
4.6 Anther and Stamen Development
4.7 Nectary Formation
4.8 Carpel and Ovule Identity and Maintenance
4.9 MADS TF Target Gene Specificity
4.10 Boundaries and Polarity Genes Are Key to Define Proper Organ Identity Domains
5 Part 5: Floral Meristem Termination
6 Conclusions and Perspectives
6.1 Coordination and Timing Are Key
6.2 Rethinking MADS-Box TF Specificity
6.3 Reaching the Single Cell Level
6.4 The Cost of Making a Flower
References
Chapter 2: Flower Development in the Solanaceae
1 The Solanaceae Family (Nightshades): Crops and Model Species
2 Specification of Floral Organ Identity: Revisions to the ABC Model
2.1 Redefining the A-Function
2.2 Variations on the B-Function: Specializing for a Single Floral Organ
2.3 Multiple Functions for C-Class Genes
2.4 Uncovering the E-Function
2.5 Divergence to the Textbook ABC Model
3 Initiation and Fusion of Floral Organs
4 Growth and Maturation of Floral Organs: Chinese Lanterns and Petal Colors
5 Conclusion
References
Chapter 3: The ABC of Flower Development in Monocots: The Model of Rice Spikelet
1 Introduction
2 The Spikelet and the Flower in Grasses
3 The Spikelet and the Flower in Rice
4 Genetic Control of Spikelet and Floral Organ Identity in Rice: An ABC View
4.1 C and D Function Genes of Rice
4.2 B Function Genes of Rice
4.3 Rice SEP, AP1/SQUA, AGL6, and OsMADS32 Genes Specify Together the (A) Function
References
Chapter 4: Model Species to Investigate the Origin of Flowers
1 Introduction
1.1 What Is a Flower?
1.2 The Rise of Flowering Plants: When, Where, and Why?
1.3 A Family Tree of the Flowering Plants
1.4 A Portrait of the Ancestral Flower
2 Models and Molecular Approaches to Study the Origin of Flowers
2.1 Amborellales
2.2 Nymphaeales
2.3 Austrobaileyales
2.4 Magnoliids
2.5 Gymnosperms
3 Missing Links and New Approaches
References
Chapter 5: Hormones and Flower Development in Arabidopsis
1 Introduction
2 Floral Meristem Initiation
3 Sepals
4 Petals
5 Stamens
6 Gynoecium
7 Conclusions
References
Part II: Genetic and Phenotypic Analyses
Chapter 6: Genetic Screens for Floral Mutants in Arabidopsis thaliana: Enhancers and Suppressors
1 Introduction
1.1 Mutagenesis of Arabidopsis
1.1.1 EMS Mutagenesis of Arabidopsis
1.1.2 T-DNA Insertional Mutagenesis
1.2 Identification of the Mutation Responsible for the Observed Mutant Phenotype
1.2.1 Map-Based Positional Cloning
1.2.2 Thermal Asymmetric Interlaced Polymerase Chain Reaction (TAIL-PCR)
1.2.3 Mapping by Deep Sequencing
2 Materials
2.1 Mutagenesis, Mutant Screening, and Initial Mapping of Mutations
2.1.1 EMS Mutagenesis of Arabidopsis
2.1.2 Planting EMS Mutagenized M1, M2, and Mapping Population Seeds
2.1.3 Preparing DNA for Map-Based Positional Cloning
2.1.4 CTAB DNA Extraction
2.1.5 Quick and Dirty PCR
2.1.6 T-DNA Insertional Mutagenesis: Agrobacterium-Mediated Transformation
2.2 Pinpointing the Mutation That Causes the Phenotype
2.2.1 Map-Based Positional Cloning: PCR
2.2.2 TAIL-PCR
2.2.3 Mapping by Deep Sequencing
3 Methods
3.1 Mutagenesis, Mutant Screening, and Initial Mapping of Mutations
3.1.1 EMS Mutagenesis of Arabidopsis
3.1.2 Planting EMS Mutagenized M1, M2, and Mapping Population Seeds
3.1.3 Preparing DNA for Map-Based Positional Cloning
3.1.4 CTAB DNA Extraction
3.1.5 Quick and Dirty Extraction
3.1.6 T-DNA Insertional Mutagenesis: Transformation and Selection of Agrobacterium
3.1.7 T-DNA Insertional Mutagenesis: Preparation of Agrobacterium
3.1.8 T-DNA Insertional Mutagenesis: Floral Dip
3.1.9 T-DNA Insertional Mutagenesis- Selection of Transformants
3.2 Pinpointing the Mutation That Causes the Phenotype
3.2.1 Map-Based Positional Cloning-PCR
3.2.2 Map-Based Positional Cloning-Analysis
3.2.3 mhiTAIL-PCR
3.2.4 Mapping by Deep Sequencing: Library Preparation
4 Notes
References
Chapter 7: Genetic and Phenotypic Analysis of Shoot Apical and Floral Meristem Development
1 Introduction
2 Materials
2.1 Confocal Laser Scanning Microscopy of the Embryonic Meristem
2.2 Histological Sectioning of the Vegetative Meristem
2.3 Meristem Size Measurement
2.4 Vegetative Meristem RNA In Situ Hybridization
2.5 Confocal Laser Scanning Microscopy of the Inflorescence Meristem
2.6 Live Imaging Confocal Laser Scanning Microscopy of the Inflorescence Meristem
2.7 Scanning Electron Microscopy of the Inflorescence Meristem
2.8 Floral Organ Number Counting
3 Methods
3.1 Confocal Laser Scanning Microscopy of the Embryonic Meristem
3.1.1 Embryo Dissection
3.1.2 Tissue Staining and Rinsing
3.1.3 Tissue Dehydration and Clearing
3.1.4 Mounting and Imaging
3.2 Histological Sectioning of the Vegetative Meristem
3.2.1 Tissue Dissection and Fixation
3.2.2 Tissue Dehydration and Infiltration
3.2.3 Tissue Staining and Embedding
3.2.4 Tissue Sectioning
3.2.5 Toluidine Blue Staining
3.2.6 Mounting and Visualization
3.3 Meristem Size Measurement
3.4 Vegetative Meristem RNA In Situ Hybridization
3.4.1 Fixation of Tissue Sections
3.4.2 Tissue Dehydration
3.4.3 Tissue Embedding in Paraffin
3.4.4 Tissue Sectioning and Mounting
3.4.5 Riboprobe Preparation (See Note 40)
3.4.6 Riboprobe Synthesis
3.4.7 Slide Pre-Hybridization and Hybridization Treatments
3.4.8 Post-Hybridization Washes
3.4.9 Detection
3.5 Confocal Laser Scanning Microscopy of the Inflorescence Meristem
3.5.1 Tissue Fixation
3.5.2 Tissue Staining and Rinsing
3.5.3 Tissue Dehydration and Clearing
3.5.4 Mounting and Imaging
3.6 Live Imaging Confocal Laser Scanning Microscopy of the Inflorescence Meristem
3.6.1 Tissue Harvesting
3.6.2 Tissue Staining and Mounting
3.7 Scanning Electron Microscopy of the Inflorescence Meristem
3.7.1 Tissue Fixation
3.7.2 Tissue Rinsing and Dehydration
3.7.3 Critical Point Drying
3.7.4 Tissue Mounting
3.8 Floral Organ Number Counting
3.8.1 Flower Dissection and Counting
3.8.2 Data Analysis
4 Notes
References
Chapter 8: Cell Biological Analyses of Anther Morphogenesis and Pollen Viability in Arabidopsis and Rice
1 Introduction
2 Materials
2.1 Alexander Red Staining
2.2 Iodine Pollen Starch Detection
2.3 FDA Staining and Imaging
2.4 Scanning Electron Microscopy for Anther Structure, Anther Dehiscence, and Pollen Wall Structure
2.5 Examination of Anther Anatomy Using Semi-Thin Sections
2.6 Callose Staining
2.7 Staining of the Pollen Exine and Intine
2.8 Detection of Programmed Cell Death Using TUNEL Assay
2.9 Ultrathin Section and Transmission Electron Microscopy (TEM) for Observation of Tapetum and Pollen Morphology
2.10 Sample Preparation for Laser Capture Microdissection
3 Methods
3.1 Alexander Staining and Photography
3.2 Iodine Pollen Starch Detection
3.3 FDA Staining and Imaging
3.4 Scanning Electron Microscopy for Anther Structure, Anther Dehiscence, and Pollen Wall Structure
3.5 Anther Anatomy Using Semi-Thin Sections
3.6 Callose Staining
3.7 Staining of the Pollen Exine and Intine
3.8 Detection of Programmed Cell Death Using TUNEL Assay
3.9 Ultrathin Section and Transmission Electron Microscopy (TEM) for Observation of Tapetum and Pollen Morphology
3.10 Sample Preparation for Laser Capture Microdissection
4 Notes
References
Chapter 9: Isolation of Meiocytes and Cytological Analyses of Male Meiotic Chromosomes in Soybean, Lettuce, and Maize
1 Introduction
2 Materials
2.1 Plants
2.2 Isolation of Male Meiocytes from Soybean and Lettuce
2.3 Chromosome Spread and Immunostaining in Arabidopsis, Soybean, and Lettuce
2.4 Light Microscopy of Maize Meiosis
2.5 Fluorescent In Situ Hybridization (FISH) of Maize Chromosomes
2.5.1 Maize Meiotic Chromosome Spreading
2.5.2 Labeled Probes
2.5.3 Pretreatment of Chromosome Spreads
2.5.4 Hybridization
2.5.5 Stringent Wash and Mounting
3 Methods
3.1 Isolation of Meiocytes from Soybean and Lettuce
3.2 Chromosome Spread for Meiotic Chromosomes of Soybean and Lettuce
3.3 Immunostaining of Meiotic Proteins in Lettuce
3.4 Immunolocalization of DNA Methylation and Histone Modifications in Arabidopsis Meiocytes
3.5 Isolation and Observation of Maize Meiosis by Light Microscopy
3.6 Fluorescent In Situ Hybridization (FISH) for Maize
3.6.1 Maize Meiotic Chromosome Spreading
3.6.2 Pretreatment of Chromosome Spreads
3.6.3 Hybridization
3.6.4 Stringent Wash and Mounting
4 Notes
References
Chapter 10: Genetic and Phenotypic Analyses of Carpel Development in Arabidopsis
1 Introduction
2 Materials
2.1 Aniline Blue Staining of Arabidopsis Pollen Tubes
2.2 Cleared Tissue for Observation of Vascular Development
2.3 NPA Treatment
2.4 Lignin Staining
2.5 Genetic Analyses
3 Methods
3.1 Aniline Blue Staining of Arabidopsis Pollen Tubes
3.1.1 Material Collection
3.1.2 Tissue Fixation
3.1.3 Pistil Softening
3.1.4 Pistil Staining
3.1.5 Pistil Mounting and Visualization
3.2 Cleared Tissue for Observation of Vascular Development
3.2.1 Material Collection
3.2.2 Tissue Fixation
3.2.3 Tissue Clearing
3.2.4 Pistil Mounting and Visualization
3.3 NPA Treatment
3.3.1 Plant Growth and Preparation
3.3.2 Plant Treatment
3.3.3 Phenotype Visualization
3.4 Lignin Staining: Whole Mount Phloroglucinol Staining (Wiesner Stain)
3.4.1 Tissue Fixation
3.4.2 Fruit Staining
3.4.3 Fruit Lignin Visualization
3.5 Lignin Staining: Tissue Section
3.5.1 Tissue Fixation
3.5.2 Tissue Dehydration and Paraplast Embedding
3.5.3 Tissue Sectioning
3.5.4 Tissue Staining
3.6 Genetic Analyses
4 Notes
References
Chapter 11: Genetic and Phenotypic Analysis of Ovule Development in Arabidopsis
1 Introduction
2 Materials
2.1 Clearing of Ovules for Wholemount Analysis
2.2 Staining of Ovules for Confocal Analysis
2.2.1 Aniline Blue Staining for Callose
2.2.2 Renaissance Staining for Analysis of Cell Morphology
2.3 Embedding and Sectioning Ovule Tissues for Immunolabelling
2.4 Crosses to Marker Lines for Assessment of Ovule Cell Identity
2.5 Laser Dissection of Ovule Tissues for Transcriptomic Analysis
3 Methods
3.1 Clearing of Ovules for Wholemount Analysis
3.2 Staining of Ovules for Confocal Analysis
3.2.1 Aniline Blue Staining for Callose
3.2.2 Renaissance Staining for Analysis of Cell Morphology
3.3 Embedding and Sectioning Ovule Tissues for Immunolabelling
3.4 Crosses to Marker Lines for Assessment of Ovule Cell Identity
3.5 Laser Dissection of Ovule Tissues for Transcriptomic Analysis
4 Notes
References
Part III: Experimental Systems
Chapter 12: Floral Induction Systems for the Study of Arabidopsis Flower Development
1 Introduction
2 Materials
2.1 Plant Lines and Growth
2.2 Reagents for Induction of Flower Formation
2.3 Reagents for Agrobacterium-Mediated Transformation Using the Floral Dip Method
3 Methods
3.1 Plant Growth and Treatment
3.2 Agrobacterium-Mediated Transformation Using the Floral Dip Method
4 Notes
References
Chapter 13: Protoplasting and Fluorescence-Activated Cell Sorting of the Shoot Apical Meristem Cell Types
1 Introduction
2 Materials
2.1 Protoplasting and Cell Sorting
2.2 Isolation of Total RNA from Sorted Cells
3 Methods
3.1 Protoplasting and Cell Sorting
3.2 Isolation of Total RNA from Sorted Cells
4 Notes
References
Chapter 14: Protoplast Isolation for Plant Single-Cell RNA-seq
1 Introduction
2 Materials
2.1 Plant Growth and Tissue Collection
2.2 Protoplast Isolation
3 Methods
4 Notes
References
Chapter 15: Plant Nuclei Isolation for Single-Nucleus RNA Sequencing
1 Introduction
2 Materials
3 Method
4 Notes
References
Chapter 16: Isolation of Nuclei Tagged in Specific Cell Types (INTACT) in Arabidopsis
1 Introduction
2 Materials
2.1 Generation of INTACT Reporter Lines
2.2 Verification of INTACT Reporter Lines by Microscopy and In Situ Histochemistry
2.3 INTACT
2.4 Assessment of the Enrichment of SAM-Specific Nuclei by Low Input Smart-qPCR
3 Methods
3.1 Establishment of INTACT Reporter Plant Lines
3.1.1 Generation of INTACT Reporter Lines
3.1.2 Verification of the INTACT Reporter Lines
3.2 INTACT Procedure
3.2.1 Harvesting Starting Material
3.2.2 INTACT
3.2.3 Assessment of the Enrichment of Tissue-Specific Nuclei by Low-Input Smart-PCR
4 Notes
References
Part IV: Molecular Biology, Genomics, and Systems Biology
Chapter 17: RNA In Situ Hybridization on Plant Tissue Sections: Expression Analysis at Cellular Resolution
1 Introduction
2 Materials
2.1 Probe Synthesis and Dot Blot
2.2 Tissue Embedding and Sectioning
2.3 Probe Hybridization
2.4 Microscopy
3 Method
3.1 Probe Synthesis
3.2 Tissue Embedding and Sectioning
3.3 Probe Hybridization
3.4 Microscopy: The Secret of Good Picture-Taking
4 Notes
References
Chapter 18: The GUS Reporter System in Flower Development Studies
1 Introduction
2 Materials
2.1 Histochemical GUS Assay
2.1.1 GUS Staining
2.1.2 Embedding/Sectioning/Mounting
2.2 Fluorometric GUS Assay
2.2.1 Protein Extracts
2.2.2 Fluorometric Assay
3 Methods
3.1 Histochemical GUS Assay
3.1.1 GUS Staining
3.1.2 Embedding
3.1.3 Sectioning and Mounting
3.1.4 Alternative Simplified GUS Screening Assay
3.2 Fluorometric GUS Assay
3.2.1 Preparation of Protein Extracts
3.2.2 MU Standard Curve
3.2.3 MUG Assay on Plant Extracts
3.2.4 Bradford Assay
3.2.5 MUG Assay on Intact Inflorescences or Flowers
4 Notes
References
Chapter 19: Expression and Functional Studies of Leaf, Floral, and Fruit Developmental Genes in Non-model Species
1 Introduction
2 Materials
2.1 Small-Scale Expression Analyses
2.1.1 Plant Material Collection
2.1.2 Total RNA Isolation
2.1.3 RNA Quality Test
2.1.4 Sample Submission to the Sequencing Facilities
2.1.5 Primer Design
2.1.6 cDNA Synthesis
2.1.7 RT-PCR
2.1.8 Visualization and Result Readings
2.1.9 qRT-PCR
2.2 Virus-Induced Gene Silencing (VIGS)
2.2.1 Vector Construction
2.2.2 Plant Growth and Agroinfiltration
2.3 RNA-seq Data Analyses
2.4 Differentially Expressed Genes (DEGs) Identification
3 Methods
3.1 Experimental Design
3.2 Purification of Total RNA
3.2.1 Plant Material Collection
3.2.2 Total RNA Isolation
3.2.3 RNA Quality Test
3.2.4 Sample Submission to the Sequencing Facilities
3.3 RNA-seq Data Analyses
3.3.1 Read Quality Control and Preprocessing
3.3.2 Quality Trimming
3.3.3 Transcriptome Assembly
3.3.4 Orthologous Gene Identification
3.4 Differentially Expressed Genes (DEGs) Identification
3.4.1 Normalized Quantification of Transcripts from Reads with Kallisto (TPM)
3.4.2 Counts Generation
3.4.3 Results from Kallisto
3.4.4 Differential Gene Expression Analysis Using DESeq2
3.4.5 Figures on Differentially Expressed Genes (plotPCA)
3.5 Small Scale Expression Analyses
3.5.1 cDNA Synthesis
3.5.2 Primer Design
3.5.3 RT-PCR
3.5.4 Visualization and Analysis Results
3.5.5 qRT-PCR
3.6 Virus-Induced Gene Silencing
3.6.1 Vector Construction
3.6.2 Plant Growth and Agrobacterium Culture Preparation
3.6.3 Agroinfiltration
4 Notes
References
Chapter 20: Gene Expression Analysis by Quantitative Real-Time PCR for Floral Tissues
1 Introduction
2 Materials
2.1 Tissue Collection and RNA Extraction
2.2 Reverse Transcription Reaction
2.3 Quantitative Real Time PCR-LightCycler 480 System
2.4 Quantitative Real Time PCR-BioMark System
3 Methods
3.1 Tissue Collection and RNA Extraction
3.2 Reverse Transcription Reaction
3.3 Quantitative Real Time PCR: LightCycler 480 System
3.4 Quantitative Real Time PCR: BioMark System
3.5 Data Analysis
3.5.1 Absolute Quantification
3.5.2 Relative Quantification
4 Notes
References
Chapter 21: Misexpression Approaches for the Manipulation of Flower Development
1 Introduction
2 Materials
2.1 Construction of Expression Cassettes
2.2 Construction of CRISPR-Based Transcription Regulators
2.3 Agrobacterium-Mediated Transformation of Arabidopsis
2.4 Screening of Transgenic Plants
2.5 Dexamethasone Induction and Protein Synthesis Inhibition Treatment
2.6 Screening and Identification of Activation-Tagged Genes
2.7 Observation of Phenotypes
3 Methods
3.1 Construction of Expression Cassettes
3.1.1 Choice of Promoters
3.1.2 Choices of Inducible Systems
3.1.3 Cloning Step of Misexpression Vectors
3.1.4 Activation Tagging
3.2 Construction of CRISPR-Based Transcription Regulators
3.2.1 Preparation of CRISPR-Based Artificial Transcription Activator
3.2.2 Preparation of CRISPR-Based Artificial Transcription Repressor
3.3 Transformation of Arabidopsis (Agrobacterium-Mediated Transformation)
3.4 Screening of Transformants
3.5 Dexamethasone Induction and Protein Synthesis Inhibition Treatment
3.6 Screening and Determination of the Activation Tag
3.7 Observation of Phenotypes
4 Notes
References
Chapter 22: Genomic Approaches for the Study of Flower Development in Floriculture Crops
1 Introduction
1.1 Plant Material to Develop a Genome Draft Assembly
1.2 Plant Material Used to Develop a Mapping Population for the Phenotype of Interest
1.3 Plant Material Used for the Characterization of the Transcriptomic Landscape of the Phenotype of Interest
1.4 Genome Sequence Assembly
1.5 Genotyping and QTL Analysis of the Population
1.6 Transcriptomic Analysis
2 Materials
2.1 Generation of a Draft Genome Sequence
2.1.1 HMW DNA Extraction, Size Selection, and Size/Quality Evaluation
2.1.2 Library Preparation and DNA Sequencing
2.1.3 Base Calling and Genome Sequence Assembly
2.1.4 Genome Annotation
2.2 Genotyping and QTL Analysis of the Population
2.2.1 DNA Extraction and Quantification
2.2.2 Genotyping-By-Sequencing Library Preparation
2.2.3 Genotyping-By-Sequencing Data Processing
2.2.4 Genetic Map and QTL Analysis
2.3 Transcriptomic Analysis
2.3.1 RNA Extraction and Quality Evaluation
2.3.2 RNA-Seq Analysis
3 Methods
3.1 Generation of a Draft Genome Sequence
3.1.1 HMW DNA Extraction
3.1.2 Size Selection
3.1.3 DNA Quantification and Quality Evaluation
3.1.4 DNA Library Preparation
3.1.5 MinION DNA Sequencing and Base Calling
3.1.6 Genome Sequence Assembly
3.1.7 Genome Annotation
3.2 QTL Analysis
3.2.1 DNA Extraction
3.2.2 Genotyping-By-Sequencing Library Preparation
3.2.3 Genotyping-By-Sequencing Data Processing and Variant Calling
3.2.4 Development of the Genetic Map
3.2.5 QTL Analysis
3.3 Transcriptomic Analysis
3.3.1 RNA Extraction and Quality Evaluation
3.3.2 RNA-Seq Library Preparation
3.3.3 RNA-Seq Data Processing and Transcript Quantification
4 Notes
References
Chapter 23: Multi-Omics Methods Applied to Flower Development
1 Introduction
2 Materials
2.1 Protein Extraction
2.2 RNA Extraction
2.3 LC-MS/MS
3 Methods
3.1 Protein Extraction
3.2 RNA Extraction
3.3 LC-MS/MS
3.3.1 Sample Preparation
3.3.2 Chromatographic and Mass Spectrometric Analysis
3.3.3 Data Analysis
3.3.4 Treatment of Missing Values and Data Imputation
3.3.5 Example: Treatment of Missing Values in a Time Series Experiment
4 Notes
References
Chapter 24: Peptidomics Methods Applied to the Study of Flower Development
1 Introduction
2 Materials
2.1 General
2.2 Ultrafiltration
2.3 Ammonium Sulphate Precipitation
2.4 Reverse-Phase Chromatography Peptide Extraction
2.5 LC-MS/MS
3 Methods
3.1 Ultrafiltration
3.2 Ammonium Sulphate Precipitation
3.3 Reverse-Phase Chromatography Peptide Extraction
3.4 LC-MS/MS
3.4.1 Sample Preparation
3.4.2 Chromatographic and Mass Spectrometric Analysis
3.4.3 Data Analysis for Database-Search Peptide Identification
4 Notes
References
Chapter 25: Quantifying Gene Expression Domains in Plant Shoot Apical Meristems
1 Introduction
2 Method
2.1 Characterization of a Fluorescence Domain
2.2 Applications
2.2.1 Inflorescence Meristem Pipeline
2.2.2 Floral Meristems Pipeline
3 Notes
References
Chapter 26: A NanoLuc-Based Transactivation Assay in Plants
1 Introduction
2 Materials
2.1 NanoLuc and Overexpression Vectors
2.2 Agroinfiltration
2.3 NanoLuc Assay
2.4 Luminescence Detection
3 Methods
3.1 NanoLuc and Overexpression Vectors
3.2 Agroinfiltration
3.3 NanoLuc Assay in Plant Extracts
3.4 NanoLuc Assay in Entire N. benthamiana Leaves
4 Notes
References
Chapter 27: A Hands-On Guide to Generate Spatial Gene Expression Profiles by Integrating scRNA-seq and 3D-Reconstructed Micros...
1 Introduction
2 Materials
2.1 Computational Infrastructure
2.2 Required Datasets
3 Methods
3.1 Computational Environment Setup
3.1.1 Download Scripts and Data
3.1.2 Install Docker
3.1.3 Build the JupyterLab Docker Image
3.1.4 Start the JupyterLab Docker Image
3.1.5 Access the JupyterLab Docker Image
3.1.6 Stop the Docker Container
3.2 Predicting 3D Gene Expression Profiles and Cell-to-Cell Mappings
3.2.1 Loading Libraries and Setting Paths to the Input Data
3.2.2 Setting the Parameters
3.2.3 Loading the Required Data
3.2.4 Preprocessing the Data
3.2.5 Preparation for novoSpaRc-Based Expression Reconstruction
3.2.6 Predict Cell-to-Cell Mappings and 3D Gene Expression Profiles
3.2.7 Format and Save the Results
3.2.8 Visualizing 3D Gene Expression Profiles
3.3 Projecting scRNA-seq Clusters onto the 3D Meristem
3.4 Visualizing scRNA-seq Clusters in the 3D Flower Meristem
3.4.1 Load the Data
3.4.2 Visualize the UMAP Plot of scRNA-seq Cells
3.4.3 Plot the Projection of UMAP Cluster in the 3D Meristem
3.5 Evaluating Prediction Performance (AUCROC and PEP-Score)
4 Notes
References
Index

Citation preview

Methods in Molecular Biology 2686

José Luis Riechmann Cristina Ferrándiz  Editors

Flower Development Methods and Protocols Second Edition

METHODS

IN

MOLECULAR BIOLOGY

Series Editor John M. Walker School of Life and Medical Sciences University of Hertfordshire Hatfield, Hertfordshire, UK

For further volumes: http://www.springer.com/series/7651

For over 35 years, biological scientists have come to rely on the research protocols and methodologies in the critically acclaimed Methods in Molecular Biology series. The series was the first to introduce the step-by-step protocols approach that has become the standard in all biomedical protocol publishing. Each protocol is provided in readily-reproducible step-by step fashion, opening with an introductory overview, a list of the materials and reagents needed to complete the experiment, and followed by a detailed procedure that is supported with a helpful notes section offering tips and tricks of the trade as well as troubleshooting advice. These hallmark features were introduced by series editor Dr. John Walker and constitute the key ingredient in each and every volume of the Methods in Molecular Biology series. Tested and trusted, comprehensive and reliable, all protocols from the series are indexed in PubMed.

Flower Development Methods and Protocols Second Edition

Edited by

José Luis Riechmann Centre for Research in Agricultural Genomics CSIC-IRTA-UAB-UB, Barcelona, Spain; Institució Catalana de Recerca i Estudis Avancats (ICREA), Barcelona, Spain

Cristina Ferrándiz Instituto de Biología Molecular y Celular de Plantas CSIC-UPV, Valencia, Spain

Editors Jose´ Luis Riechmann Centre for Research in Agricultural Genomics CSIC-IRTA-UAB-UB Barcelona, Spain

Cristina Ferra´ndiz Instituto de Biologı´a Molecular y Celular de Plantas CSIC-UPV Valencia, Spain

Institucio´ Catalana de Recerca i Estudis Avancats (ICREA) Barcelona, Spain

ISSN 1064-3745 ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-0716-3298-7 ISBN 978-1-0716-3299-4 (eBook) https://doi.org/10.1007/978-1-0716-3299-4 © Springer Science+Business Media, LLC, part of Springer Nature 2023 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Humana imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer Nature. The registered company address is: 1 New York Plaza, New York, NY 10004, U.S.A.

Preface Over the past 35 years, detailed insights into the genetic and molecular mechanisms that control flower development in different angiosperm species have been obtained, making great progress in the identification of key regulators of flower morphogenesis, in elucidating their organization in pathways and networks, and in characterizing the developmental process at multiple levels, from organismal to single cells. These advances have relied on the continuous development of methodologies and techniques, from molecular genetics to advanced microscopy or single-cell biology, and their implementation into plant research. To facilitate further progress in the field of flower development, this book provides a collection of protocols for many of the experimental approaches that are currently used to study the formation of flowers, from genetic methods and phenotypic analyses, to genomewide experiments, modeling, and system-wide approaches. An effort has been made to facilitate the incorporation of non-model species that can be useful to study specific developmental processes or the origin of evolutionary innovations. In addition, several introductory chapters provide an overview of the current status on the field of flower development, also highlighting open questions and future directions. Methods chapters are organized in three major sections: genetic and phenotypic analyses; experimental systems; and molecular biology, genomics, and systems biology. Each chapter contains a brief introduction, step-bystep methods, a list of necessary materials, and a Notes section with tips on troubleshooting, as well as extensive reference lists. Comprehensive and up to date, we hope that this book on flower development will become an essential guide for plant developmental biologists, from the more novice to the experienced researcher, and for those considering venturing into the field. Barcelona, Spain Valencia, Spain

Jose´ Luis Riechmann Cristina Ferra´ndiz

v

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

PART I

v xi

REVIEW AND OVERVIEW CHAPTERS

1 Flower Development in Arabidopsis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Hicham Chahtane, Xuelei Lai, Gabrielle Tichtinsky, Philippe Rieu, Moı¨ra Arnoux-Courseaux, Coralie Cance´, Claudius Marondedze, and Franc¸ois Parcy 2 Flower Development in the Solanaceae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Marie Monniaux and Michiel Vandenbussche 3 The ABC of Flower Development in Monocots: The Model of Rice Spikelet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Ludovico Dreni 4 Model Species to Investigate the Origin of Flowers . . . . . . . . . . . . . . . . . . . . . . . . . 83 Charles P. Scutt 5 Hormones and Flower Development in Arabidopsis. . . . . . . . . . . . . . . . . . . . . . . . . 111 ˜ iga-Mayo, Yolanda Dura´n-Medina, Victor M. Zu´n Nayelli Marsch-Martı´nez, and Stefan de Folter

PART II

GENETIC AND PHENOTYPIC ANALYSES

6 Genetic Screens for Floral Mutants in Arabidopsis thaliana: Enhancers and Suppressors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zhigang Huang, Thanh Theresa Dinh, Elizabeth Luscher, Shaofang Li, Xigang Liu, So Youn Won, and Xuemei Chen 7 Genetic and Phenotypic Analysis of Shoot Apical and Floral Meristem Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mona M. Monfared, Thai Q. Dao, and Jennifer C. Fletcher 8 Cell Biological Analyses of Anther Morphogenesis and Pollen Viability in Arabidopsis and Rice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fang Chang, Shuangshuang Wang, Zesen Lai, Zaibao Zhang, Yue Jin, and Hong Ma 9 Isolation of Meiocytes and Cytological Analyses of Male Meiotic Chromosomes in Soybean, Lettuce, and Maize . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cong Wang, Xiang Li, Jiyue Huang, Hong Ma, Chung-Ju Rachel Wang, and Yingxiang Wang 10 Genetic and Phenotypic Analyses of Carpel Development in Arabidopsis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ` , Patricia Ballester, Monica Colombo, Chloe´ Fourquin, Vicente Balanza ´ Irene Martınez-Ferna´ndez, Clara I. Ortiz-Ramı´rez, and Cristina Ferra´ndiz

vii

131

163

199

219

241

viii

11

Contents

Genetic and Phenotypic Analysis of Ovule Development in Arabidopsis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 Dayton C. Bird, Chao Ma, Sara Pinto, Weng Herng Leong, and Matthew R. Tucker

PART III 12

13

14 15 16

Floral Induction Systems for the Study of Arabidopsis Flower Development. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ´ ’Maoile´idigh, Bennett Thomson, and Frank Wellmer Diarmuid O Protoplasting and Fluorescence-Activated Cell Sorting of the Shoot Apical Meristem Cell Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . G. Venugopala Reddy Protoplast Isolation for Plant Single-Cell RNA-seq . . . . . . . . . . . . . . . . . . . . . . . . . Shulin Ren and Ying Wang Plant Nuclei Isolation for Single-Nucleus RNA Sequencing . . . . . . . . . . . . . . . . . . Xu Xin, Fei Du, and Yuling Jiao Isolation of Nuclei Tagged in Specific Cell Types (INTACT) in Arabidopsis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ruben M. Benstein, Markus Schmid, and Yuan You

PART IV 17

18 19

20

21 22

23 24

EXPERIMENTAL SYSTEMS 285

293 301 307

313

MOLECULAR BIOLOGY, GENOMICS, AND SYSTEMS BIOLOGY

RNA In Situ Hybridization on Plant Tissue Sections: Expression Analysis at Cellular Resolution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vladislav Gramma and Vanessa Wahl The GUS Reporter System in Flower Development Studies . . . . . . . . . . . . . . . . . . Janaki S. Mudunkothge, C. Nathan Hancock, and Beth A. Krizek Expression and Functional Studies of Leaf, Floral, and Fruit Developmental Genes in Non-model Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Natalia Pab
on-Mora, Harold Sua´rez-Baron, Yesenia Madrigal, Juan F. Alzate, and Favio Gonza´lez Gene Expression Analysis by Quantitative Real-Time PCR for Floral Tissues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ´ lvarez-Urdiola, Mariana Bustamante, Joana Ribes, Raquel A and Jose´ Luis Riechmann Misexpression Approaches for the Manipulation of Flower Development . . . . . . Yifeng Xu, Eng-Seng Gan, and Toshiro Ito Genomic Approaches for the Study of Flower Development in Floriculture Crops . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tomas Hasing and Aureliano Bombarely Multi-Omics Methods Applied to Flower Development . . . . . . . . . . . . . . . . . . . . . ´ lvarez-Urdiola, Jose´ Toma´s Matus, and Jose´ Luis Riechmann Raquel A Peptidomics Methods Applied to the Study of Flower Development . . . . . . . . . . ´ lvarez-Urdiola, Eva Borra ` s, Federico Valverde, Jose´ Toma´s Matus, Raquel A Eduard Sabid
o , and Jose´ Luis Riechmann

331 351

365

403

429

453 495 509

Contents

ix

25

Quantifying Gene Expression Domains in Plant Shoot Apical Meristems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537 Pau Formosa-Jordan and Benoit Landrein 26 A NanoLuc-Based Transactivation Assay in Plants . . . . . . . . . . . . . . . . . . . . . . . . . . 553 Rosa Esmeralda Becerra-Garcı´a, Jose´ Erik Cruz-Valderrama, Vincent E. Cerbantez-Bueno, Nayelli Marsch-Martı´nez, and Stefan de Folter 27 A Hands-On Guide to Generate Spatial Gene Expression Profiles by Integrating scRNA-seq and 3D-Reconstructed Microscope-Based Plant Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 567 Manuel Neumann and Jose M. Muino Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

581

Contributors ´ LVAREZ-URDIOLA • Centre for Research in Agricultural Genomics (CRAG) CSICRAQUEL A IRTA-UAB-UB, Edifici CRAG, Campus UAB, Cerdanyola del Valle`s, Barcelona, Spain JUAN F. ALZATE • Centro Nacional de Secuenciaci
on Gen
omica–CNSG, Sede de Investigaci
on Universitaria–SIU, Medellı´n, Antioquia, Colombia MOI¨RA ARNOUX-COURSEAUX • CNRS, Universite´ Grenoble Alpes, CEA, INRAE, IRIG, BIG-LPCV, Grenoble, France VICENTE BALANZA` • Instituto de Biologı´a Molecular y Celular de Plantas CSIC-UPV, Campus de la Universidad Polite´cnica de Valencia, Valencia, Spain PATRICIA BALLESTER • Instituto de Biologı´a Molecular y Celular de Plantas CSIC-UPV, Campus de la Universidad Polite´cnica de Valencia, Valencia, Spain ROSA ESMERALDA BECERRA-GARCI´A • Unidad de Gen
omica Avanzada (UGA-LANGEBIO), Centro de Investigaci
on y de Estudios Avanzados del Instituto Polite´cnico Nacional (CINVESTAV-IPN), Irapuato, Guanajuato, Mexico; Departamento de Biotecnologı´a y Bioquı´mica, Unidad Irapuato, CINVESTAV-IPN, Irapuato, Guanajuato, Mexico RUBEN M. BENSTEIN • Umea˚ Plant Science Centre, Department of Plant Physiology, Umea˚ University, Umea˚, Sweden DAYTON C. BIRD • School of Agriculture, Food and Wine, University of Adelaide, Waite Campus, Urrbrae, SA, Australia AURELIANO BOMBARELY • Instituto de Biologı´a Molecular y Celular de Plantas (IBMCP) (UPV-CSIC), Valencia, Spain EVA BORRA`S • Centre for Genomic Regulation (CRG), Barcelona Institute of Science and Technology, Barcelona, Spain; Universitat Pompeu Fabra, Barcelona, Spain MARIANA BUSTAMANTE • Centre for Research in Agricultural Genomics (CRAG) CSICIRTA-UAB-UB, Edifici CRAG, Campus UAB, Cerdanyola del Valle`s, Barcelona, Spain; Roche Pharma Research and Early Development, Roche Innovation Center Basel, F. Hoffmann-La Roche Ltd., Basel, Switzerland CORALIE CANCE´ • CNRS, Universite´ Grenoble Alpes, CEA, INRAE, IRIG, BIG-LPCV, Grenoble, France VINCENT E. CERBANTEZ-BUENO • Unidad de Gen
omica Avanzada (UGA-LANGEBIO), Centro de Investigaci
on y de Estudios Avanzados del Instituto Polite´cnico Nacional (CINVESTAV-IPN), Irapuato, Guanajuato, Mexico HICHAM CHAHTANE • CNRS, Universite´ Grenoble Alpes, CEA, INRAE, IRIG, BIG-LPCV, Grenoble, France; Institut de Recherche Pierre Fabre, Green Mission Pierre Fabre, Conservatoire Botanique Pierre Fabre, Soual, France FANG CHANG • State Key Laboratory of Genetic Engineering, Ministry of Education Key Laboratory of Biodiversity Sciences and Ecological Engineering and Institute of Biodiversity Sciences, Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai, China XUEMEI CHEN • Department of Botany and Plant Sciences, University of California, Riverside, CA, USA MONICA COLOMBO • Instituto de Biologı´a Molecular y Celular de Plantas CSIC-UPV, Campus de la Universidad Polite´cnica de Valencia, Valencia, Spain; CREA Research Centre for Genomics and Bioinformatics, Fiorenzuola d’Arda, Italy

xi

xii

Contributors

JOSE´ ERIK CRUZ-VALDERRAMA • Unidad de Gen
omica Avanzada (UGA-LANGEBIO), Centro de Investigaci
on y de Estudios Avanzados del Instituto Polite´cnico Nacional (CINVESTAV-IPN), Irapuato, Guanajuato, Mexico THAI Q. DAO • Plant Gene Expression Center, USDA-ARS/UC Berkeley, Albany, CA, USA; Department of Plant and Microbial Biology, University of California, Berkeley, CA, USA; Department of Botany and Plant Pathology, Oregon State University, Corvallis, OR, USA STEFAN DE FOLTER • Unidad de Gen
omica Avanzada (UGA-LANGEBIO), Centro de Investigaci
on y de Estudios Avanzados del Instituto Polite´cnico Nacional (CINVESTAVIPN), Irapuato, Guanajuato, Mexico THANH THERESA DINH • Department of Botany and Plant Sciences, University of California, Riverside, CA, USA LUDOVICO DRENI • Instituto de Biologı´a Molecular y Celular de Plantas (IBMCP), Consejo Superior de Investigaciones Cientı´ficas-Universidad Polite´cnica de Valencia, Valencia, Spain FEI DU • State Key Laboratory of Plant Genomics, Institute of Genetics and Developmental Biology, The Innovative Academy of Seed Design, Chinese Academy of Sciences, Beijing, China YOLANDA DURA´N-MEDINA • Departamento de Biotecnologı´a y Bioquı´mica, Centro de Investigaci
on y de Estudios Avanzados del Instituto Polite´cnico Nacional (CINVESTAVIPN), Irapuato, Guanajuato, Mexico CRISTINA FERRA´NDIZ • Instituto de Biologı´a Molecular y Celular de Plantas CSIC-UPV, Campus de la Universidad Polite´cnica de Valencia, Valencia, Spain JENNIFER C. FLETCHER • Plant Gene Expression Center, USDA-ARS/UC Berkeley, Albany, CA, USA; Department of Plant and Microbial Biology, University of California, Berkeley, CA, USA PAU FORMOSA-JORDAN • Sainsbury Laboratory, University of Cambridge, Cambridge, UK; Department of Plant Developmental Biology, Max Planck Institute for Plant Breeding Research, Cologne, Germany; Cluster of Excellence on Plant Science (CEPLAS), Max Planck Institute for Plant Breeding Research, Cologne, Germany CHLOE´ FOURQUIN • Instituto de Biologı´a Molecular y Celular de Plantas CSIC-UPV, Campus de la Universidad Polite´cnica de Valencia, Valencia, Spain ENG-SENG GAN • Republic Polytechnic, School of Applied Science (SAS), Singapore, Singapore FAVIO GONZA´LEZ • Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Sede Bogota´, Bogota´, Colombia VLADISLAV GRAMMA • Max Planck Institute of Plant Physiology, Potsdam, Germany C. NATHAN HANCOCK • Department of Biology and Geology, University of South Carolina Aiken, Aiken, SC, USA TOMAS HASING • ELO Life Systems, Durham, NC, USA JIYUE HUANG • College of Life Sciences, South China Agricultural University, Guangzhou, China; Guangdong Laboratory for Lingnan Modern Agriculture, Guangzhou, China ZHIGANG HUANG • Hunan Provincial Key Laboratory of Phytohormones and Growth Development, Hunan Agricultural University, Changsha, China TOSHIRO ITO • Nara Institute of Science and Technology, Biological Sciences, Plant Stem Cell Regulation and Floral Patterning Laboratory, Ikoma, Nara, Japan YULING JIAO • State Key Laboratory of Plant Genomics, Institute of Genetics and Developmental Biology, The Innovative Academy of Seed Design, Chinese Academy of Sciences, Beijing, China; State Key Laboratory of Protein and Plant Gene Research, Peking-

Contributors

xiii

Tsinghua Center for Life Sciences, Center for Quantitative Biology, School of Life Sciences, Peking University, Beijing, China YUE JIN • State Key Laboratory of Genetic Engineering, Ministry of Education Key Laboratory of Biodiversity Sciences and Ecological Engineering and Institute of Biodiversity Sciences, Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai, China BETH A. KRIZEK • Department of Biological Sciences, University of South Carolina, Columbia, SC, USA XUELEI LAI • CNRS, Universite´ Grenoble Alpes, CEA, INRAE, IRIG, BIG-LPCV, Grenoble, France; Huazhong Agricultural University, National Key Laboratory of Crop Genetic Improvement, Hubei Hongshan Laboratory, Wuhan, China ZESEN LAI • State Key Laboratory of Genetic Engineering, Ministry of Education Key Laboratory of Biodiversity Sciences and Ecological Engineering and Institute of Biodiversity Sciences, Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai, China BENOIT LANDREIN • Sainsbury Laboratory, University of Cambridge, Cambridge, UK; Laboratoire Reproduction et De´veloppement des Plantes, Universite´ de Lyon, ENS de Lyon, UCB Lyon 1, CNRS, INRAE, INRIA, Lyon, France WENG HERNG LEONG • School of Agriculture, Food and Wine, University of Adelaide, Waite Campus, Urrbrae, SA, Australia SHAOFANG LI • Department of Botany and Plant Sciences, University of California, Riverside, CA, USA XIANG LI • College of Horticulture, Henan Agricultural University, Zhengzhou, China XIGANG LIU • Department of Botany and Plant Sciences, University of California, Riverside, CA, USA ELIZABETH LUSCHER • Department of Botany and Plant Sciences, University of California, Riverside, CA, USA CHAO MA • School of Agriculture, Food and Wine, University of Adelaide, Waite Campus, Urrbrae, SA, Australia HONG MA • Department of Biology, The Huck Institutes of the Life Sciences, The Pennsylvania State University, University Park, PA, USA YESENIA MADRIGAL • Instituto de Biologı´a, Universidad de Antioquia, Medellı´n, Colombia CLAUDIUS MARONDEDZE • CNRS, Universite´ Grenoble Alpes, CEA, INRAE, IRIG, BIG-LPCV, Grenoble, France; Department of Biochemistry, Faculty of Medicine, Midlands State University, Senga, Gweru, Zimbabwe NAYELLI MARSCH-MARTI´NEZ • Departamento de Biotecnologı´a y Bioquı´mica, Centro de Investigaci
on y de Estudios Avanzados del Instituto Polite´cnico Nacional (CINVESTAVIPN), Irapuato, Guanajuato, Mexico IRENE MARTI´NEZ-FERNA´NDEZ • Instituto de Biologı´a Molecular y Celular de Plantas CSICUPV, Campus de la Universidad Polite´cnica de Valencia, Valencia, Spain JOSE´ TOMA´S MATUS • Centre for Research in Agricultural Genomics (CRAG) CSIC-IRTAUAB-UB, Edifici CRAG, Campus UAB, Cerdanyola del Valle`s, Barcelona, Spain; Institute for Integrative Systems Biology (I2SysBio), Universitat de Vale`ncia-CSIC, Paterna, Valencia, Spain MONA M. MONFARED • Plant Gene Expression Center, USDA-ARS/UC Berkeley, Albany, CA, USA; Department of Plant and Microbial Biology, University of California, Berkeley, CA, USA; Department of Molecular and Cellular Biology, University of California, Davis, CA, USA

xiv

Contributors

MARIE MONNIAUX • Laboratoire Reproduction et De´veloppement des Plantes, Universite´ de Lyon, ENS de Lyon, UCB Lyon 1, CNRS, INRAE, Lyon, France JANAKI S. MUDUNKOTHGE • Department of Biological Sciences, University of South Carolina, Columbia, SC, USA JOSE M. MUINO • Institute for Biology, Humboldt-Universit€ a t zu Berlin, Berlin, Germany MANUEL NEUMANN • Institute for Biology, Humboldt-Universit€ a t zu Berlin, Berlin, Germany ´ ’MAOILE´IDIGH • Department of Biology, Maynooth University, Maynooth, DIARMUID O County Kildare, Ireland CLARA I. ORTIZ-RAMI´REZ • Instituto de Biologı´a Molecular y Celular de Plantas CSIC-UPV, Campus de la Universidad Polite´cnica de Valencia, Valencia, Spain NATALIA PABO´N-MORA • Instituto de Biologı´a, Universidad de Antioquia, Medellı´n, Colombia FRANC¸OIS PARCY • CNRS, Universite´ Grenoble Alpes, CEA, INRAE, IRIG, BIG-LPCV, Grenoble, France SARA PINTO • LAQV REQUIMTE, Departamento de Biologia, Faculdade de Cieˆncias da Universidade do Porto, Porto, Portugal G. VENUGOPALA REDDY • Department of Botany and Plant Sciences, Center for Plant Cell Biology (CEPCEB), Institute of Integrative Genome Biology (IIGB), University of California, Riverside, CA, USA SHULIN REN • College of Life Sciences, University of Chinese Academy of Sciences, Beijing, China JOANA RIBES • Centre for Research in Agricultural Genomics (CRAG) CSIC-IRTA-UABUB, Edifici CRAG, Campus UAB, Cerdanyola del Valle`s, Barcelona, Spain JOSE´ LUIS RIECHMANN • Centre for Research in Agricultural Genomics (CRAG) CSICIRTA-UAB-UB, Edifici CRAG, Campus UAB, Cerdanyola del Valle`s, Barcelona, Spain; Instituci
o Catalana de Recerca i Estudis Avanc¸ats (ICREA), Barcelona, Spain PHILIPPE RIEU • CNRS, Universite´ Grenoble Alpes, CEA, INRAE, IRIG, BIG-LPCV, Grenoble, France; Structural Plant Biology Laboratory, Department of Botany and Plant Biology, University of Geneva, Geneva, Switzerland EDUARD SABIDO´ • Centre for Genomic Regulation (CRG), Barcelona Institute of Science and Technology, Barcelona, Spain; Universitat Pompeu Fabra, Barcelona, Spain MARKUS SCHMID • Umea˚ Plant Science Centre, Department of Plant Physiology, Umea˚ University, Umea˚, Sweden; Department of Plant Biology, Linnean Centre for Plant Biology, Swedish University of Agricultural Sciences, Uppsala, Sweden CHARLES P. SCUTT • Laboratoire Reproduction et De´veloppement des Plantes, Universite´ de Lyon, ENS de Lyon, UCB Lyon-1, CNRS, INRA, Lyon, France HAROLD SUA´REZ-BARON • Instituto de Biologı´a, Universidad de Antioquia, Medellı´n, Colombia; Departamento de Ciencias Naturales y Matema´ticas, Pontificia Universidad Javeriana, Cali, Colombia BENNETT THOMSON • Smurfit Institute of Genetics, Trinity College Dublin, Dublin, Ireland GABRIELLE TICHTINSKY • CNRS, Universite´ Grenoble Alpes, CEA, INRAE, IRIG, BIG-LPCV, Grenoble, France MATTHEW R. TUCKER • School of Agriculture, Food and Wine, University of Adelaide, Waite Campus, Urrbrae, SA, Australia FEDERICO VALVERDE • Institute for Plant Biochemistry and Photosynthesis CSIC – University of Seville, Seville, Spain

Contributors

xv

MICHIEL VANDENBUSSCHE • Laboratoire Reproduction et De´veloppement des Plantes, Universite´ de Lyon, ENS de Lyon, UCB Lyon 1, CNRS, INRAE, Lyon, France VANESSA WAHL • Max Planck Institute of Plant Physiology, Potsdam, Germany; The James Hutton Institute, Dundee, UK CHUNG-JU RACHEL WANG • Institute of Plant and Microbial Biology, Academia Sinica, Taipei, Taiwan CONG WANG • College of Life Sciences, South China Agricultural University, Guangzhou, China; Guangdong Laboratory for Lingnan Modern Agriculture, Guangzhou, China SHUANGSHUANG WANG • State Key Laboratory of Genetic Engineering, Ministry of Education Key Laboratory of Biodiversity Sciences and Ecological Engineering and Institute of Biodiversity Sciences, Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai, China; School of Life Sciences, East China Normal University, Shanghai, China YING WANG • College of Life Sciences, University of Chinese Academy of Sciences, Beijing, China YINGXIANG WANG • College of Life Sciences, South China Agricultural University, Guangzhou, China; Guangdong Laboratory for Lingnan Modern Agriculture, Guangzhou, China FRANK WELLMER • Smurfit Institute of Genetics, Trinity College Dublin, Dublin, Ireland SO YOUN WON • Department of Botany and Plant Sciences, University of California, Riverside, CA, USA XU XIN • State Key Laboratory of Plant Genomics, Institute of Genetics and Developmental Biology, The Innovative Academy of Seed Design, Chinese Academy of Sciences, Beijing, China; College of Advanced Agricultural Sciences, University of Chinese Academy of Sciences, Beijing, China YIFENG XU • College of Life Sciences, Nanjing Agricultural University, Nanjing, Jiangsu, China YUAN YOU • Center for Plant Molecular Biology (ZMBP), Department of General Genetics, Eberhard Karls University of Tu¨bingen, Tu¨bingen, Germany; Department of Molecular Life Sciences, Technical University of Munich, Freising, Germany ZAIBAO ZHANG • State Key Laboratory of Genetic Engineering, Ministry of Education Key Laboratory of Biodiversity Sciences and Ecological Engineering and Institute of Biodiversity Sciences, Institute of Plant Biology, School of Life Sciences, Fudan University, Shanghai, China VICTOR M. ZU´N˜IGA-MAYO • CONACyT – Postgrado en Fitosanidad-Fitopatologı´a, Colegio de Postgraduados, Campus Montecillo, Montecillo, Estado de Me´xico, Mexico

Part I Review and Overview Chapters

Chapter 1 Flower Development in Arabidopsis Hicham Chahtane, Xuelei Lai, Gabrielle Tichtinsky, Philippe Rieu, Moı¨ra Arnoux-Courseaux, Coralie Cance´, Claudius Marondedze, and Franc¸ois Parcy Abstract Like in other angiosperms, the development of flowers in Arabidopsis starts right after the floral transition, when the shoot apical meristem (SAM) stops producing leaves and makes flowers instead. On the flanks of the SAM emerge the flower meristems (FM) that will soon differentiate into the four main floral organs, sepals, petals, stamens, and pistil, stereotypically arranged in concentric whorls. Each phase of flower development—floral transition, floral bud initiation, and floral organ development—is under the control of specific gene networks. In this chapter, we describe these different phases and the gene regulatory networks involved, from the floral transition to the floral termination. Key words Flower development, Arabidopsis, Meristem, MADS-box

1

Part 1: The Floral Transition The moment at which the floral transition occurs in plants is key for reproductive success and is fine-tuned by external and internal cues. Eventually, the signaling pathways of these cues converge on a handful of genes called floral integrators that make the link between the regulation of flowering time and the triggering of flower development [1]. Among these integrators, the gene FLOWERING LOCUS T (FT) plays a fundamental role [2] (see Fig. 1). It encodes a mobile protein produced in the leaves and that travels through the phloem to activate other floral integrators at the SAM [3]. These include LEAFY (LFY), an orphan and plant-specific transcription factor (TF) of the Helix-Loop-Helix class and SUPPRESSOR OF CONSTANS OVEREXPRESSION 1 (SOC1), a TF belonging to the MADS-box TF family [4, 5]. As a result, the onset of expression of the florigen FT greatly influences the passage from the vegetative

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_1, © Springer Science+Business Media, LLC, part of Springer Nature 2023

3

4

Hicham Chahtane et al.

Fig. 1 Simplified representation of the gene network involved in floral transition. Close up on the vegetative SAM containing leaves (in green) before the floral transition. Internal (grey panel) and external (yellow panel) cues regulate the expression of the florigen FT. Sugar content represented as T6P, and nitrates, are both positive regulators of flowering and activate indirectly FT expression. The age-dependent pathway is under the control of SPL and miRNA156. When the plant ages, SPL accumulates and negatively regulates AP2-like transcription through mir172 to promote FT transcription. Temperature responses are mediated at least by SVP, FLC, and PIF4. Increased ambient temperature induces chromatin accessibility allowing PIF4 to activate FT. Upon cold response, SVP (in complex with FLMβ) represses FT transcription. Vernalization response is mediated by the repressive activity of FLC on FT. In daylight conditions such as in the afternoon, GI promotes CO by modulating CDF activity. PHYA promotes CO stability during the day. During the night, the complex COP1/SPA/ELF3 represses CO activity. Once FT is promoted, the florigen complex FT/FD acts in the SAM to activate several floral integrators such as SOC1, AP1 and indirectly LFY. Accumulation of the floral integrators induces the SAM to switch to the reproductive phase. Please note that the crosstalk between pathways exist and that this simplified representation does not include all the regulation processes discussed in the text

Flower Development in Arabidopsis

5

to the reproductive stage. This peculiar gene acts as a hub to integrate several external and internal signals and its regulation is highly controlled by a myriad of mechanisms [6] (see Fig. 1). We will mention a non-exhaustive list of the major actors controlling FT expression. 1.1 External Parameters Influencing Flowering—Temperature and Photoperiod 1.1.1

Temperature

Temperature is a major determinant of flowering and is effective through two distinct mechanisms (see review by Susila, et al. [7] for details). One is a prolonged exposure of seedlings to cold called vernalization, whereas the other results in the internalization of the ambient temperature [8–10]. The central gene in vernalization is FLOWERING LOCUS C (FLC), which encodes a MADS-box TF [11] (see Fig. 1). Genetic, transcriptomic and genome-wide binding profiles of FLC revealed its main activities in the direct repression of two major floral integrators, FT and SOC1 [12]. The action of FLC on FT and SOC1 is partially under the control of gibberellic acid (GA) sensitive proteins DELLA, where DELLA and FLC form a repressor complex to downregulate the expression of floral integrators, revealing the crosstalk between vernalization and hormonal pathways [13]. The expression of FLC is strongly and negatively correlated with flowering time and FT expression [11]. The main positive regulator of FLC is FRIGIDA (FRI), which maintains high FLC mRNA level and conditions the plant to remain in the vegetative state. On the other hand, FLC expression is suppressed by the chromatin remodelling protein VERNALIZATION INSENSITIVE3 (VIN3), whose expression is rapidly induced after a cold period [14]. More recently, another mode of regulation of FLC has been discovered, involving post-transcriptional mechanisms (see review for details [15, 16]). It has been shown that cold exposure induces the formation of two types of non-coding RNA variants of FLC, called COOLAIR (for COLD INDUCED LONG ANTISENSE INTRAGENIC RNA) and COLDAIR (for COLD ASSISTED INTRONIC NONCODING RNA) [17, 18]. These two non-coding RNAs repress FLC by either preventing its translation (in the case of COOLAIR) or by promoting epigenetic repression at the locus [18–21]. In addition to the effect of temperature through vernalization, the ambient temperature also strongly affects flowering time. This control likely involves a wealth of pathways and mechanisms and research is still ongoing to understand what are the major ones [10]. For example, a recent study described how ambient temperature controls the degree of unsaturation of phosphatidyl glycerol, that in turn affects the retention of the FT protein in leaf cells [22]. Higher temperature reduces unsaturation leading to increased release of FT in phloem companion cells and accelerated flowering [22, 23].

6

Hicham Chahtane et al.

The SHORT VEGETATIVE PHASE (SVP) gene, which also belongs to the MADS-box family, acts as a repressor of flowering, particularly on FT transcription [24, 25]. How ambient temperature regulates SVP has been recently unravelled but is still under debate (see discussion in [26]). Like other MIKC-type MADS box proteins, SVP acts in complex with homologous proteins to repress FT transcription [27]. One of them is FLOWERING LOCUS M (FLM) that is transcribed as several splice variants in a temperaturedependent manner [27, 28]. One splice variant, called FLM-β, is preferentially produced when the temperature is cooler. At low temperature, SVP preferentially interacts with FLM-β, which allows SVP to be translocated to the nucleus and to inhibit suppresses FT expression [27–29] (see Fig. 1). In addition to the transcriptional regulation of FT by SVP, the stability of SVP protein is also temperature-dependent: it is degraded via the proteasome when the ambient temperature increases, reducing further the SVP repressive activity on FT level under elevated temperature [30]. PHYTOCHROME INTERACTING FACTOR4 (PIF4) protein has also been proposed to regulate FT expression in a temperature-dependent manner [31]. PIF4 binds directly to a regulatory region of the FT promoter that is only accessible at high temperature (see Fig. 1). More recent shreds of evidence suggest that PIF4 activity is highly dependent on external temperature, mainly due to its interaction with the thermosensor protein PHYTOCHROME B (PHYB) (see review for details [32]). In addition, others temperature-dependent mechanism regulating PIF4 expressions have recently been identified in response to warm temperature. One involved the TF CYCLOIDEA AND PCF TRANSCRIPTION FACTOR 5 (TCP5), which is stabilized under heat stress, and both positively regulates PIF4 expression and modulates its activity to integrate thermomorphogenesis responses [33]. Other mechanisms mediated by the chromatin remodelling POWERDRESS protein participates in the regulation of histone accessibility at the PIF4 locus after elevated temperature and thus is a critical component of the temperature responses [34]. Recently, it has been shown that EARLY FLOWERING3 (ELF3), a member of the evening complex (EC; Consists of ELF3, ELF4 and LUX ARRYTHMO (LUX)) involved in circadian clock regulation, also controls the response to ambient temperature. elf3 mutant display a strong early flowering phenotype, due to ELF3 repressive action on FT expression [35, 36]. The activity of the prion-like domain protein ELF3, and its binding to specific genomic regions is dependent on temperature, suggesting that the EC could act as a sensor of temperature to control flowering [35, 37–40]. Interestingly, several light signaling components such as PHYB and ELF3 bind common genomic regions, suggesting that these two proteins act additively to integrate external cues (temperature and light) to modulate flowering [38].

Flower Development in Arabidopsis 1.1.2

Photoperiod

7

The onset of flowering is also largely regulated by the duration of light exposure, a mechanism called photoperiod perception, as revealed by the earlier Arabidopsis blooming in long day (LD) relative to short day (SD) conditions (see review [41]). The major actor of the photoperiodic pathway is the CONSTANS (CO) gene, which encodes a B-box TF directly responsible for the transcription of FT. In growth chamber conditions, the CO gene is repressed in the early morning, then its transcription increases at the end of the day and during the night [42, 43]. This rhythmic regulation is controlled by two complexes: CYCLING DOF FACTOR (CDF) and GIGANTEA (GI)/FLAVIN-BINDING KELCH REPEAT-F-BOX 1 (FKF1). During the morning, the CDFs directly represses the transcription of CO while the rest of the day and night the GI/FKF1 complex strongly represses the CDFs transcription, allowing indirect activation of CO [44] (see Fig. 1). Other factors such as light quality influence the activity of GI/FKF1 complex and add another layer of complexity to the regulation of CO (see review for details [41, 44]). Regulation of CO also occurs at the post-transcriptional level. During the day, perception of light involves PHYTOCHROMES (PHY) such as PHYA, which stabilizes CO and thus affects FT regulation [42] (see Fig. 1). During the night, CO protein is degraded by the proteasome through the action of an E3 ubiquitin ligase complex involving CONSTITUTIVE PHOTOMORPHOGENESIS1 (COP1) and SUPPRESSOR OF PHYA-105 protein (SPA1) [43]. ELF3 also physically interacts with COP1/SPA complex and is key to negatively affect CO stability [36, 45] (see Fig. 1). Importantly, several studies highlight different genetic responses to flowering when plants were grown in artificially controlled conditions compared to natural conditions. For instance, a morning peak of FT expression is observed only under natural environments but not in lab-controlled conditions [36]. This is probably due to changes in CO stability, where the protein is slowly degraded in natural environments due to different light and circadian clock crosstalk [36]. Another example is illustrated by the light spectrum used in classic lab growth conditions, which most of the time lack ultraviolet-B (UV-B) wavelength. Plants sense UV-B thanks to UV RESISTANT LOCUS 8 (UVR8) photoreceptor and respond by upregulating several genes, including the REPRESSOR OF UV-B PHOTOMORPHOGENESIS 2 (RUP2) (see review for details [46]). Under SD conditions and in response to light containing UV-B, RUP2 is induced and acts as a repressor of flowering by preventing FT expression via physical interaction with CO and inactivation of CO activity [47]. This study reveals an UVR8-dependent flowering pathway (under specific photoperiod conditions) and again challenges the classical use of standardized conditions found in many labs to study flowering time regulation.

8

Hicham Chahtane et al.

1.2 Internal Cues— Metabolism and AgeDependent Pathways

Besides the above-mentioned external factors that play an important role in flowering, plants also use endogenous signals such as metabolic cues to trigger the energy-consuming production of flowers, fruits, and seeds. As a result, internal signals such as energy reserves or the perception of age allow a genetic control of genes involved in flowering.

1.2.1 Sugar and Nitrate Metabolism

Trehalose 6-phosphate (T6P) is a disaccharide that acts as a sensor of the plant’s energy reserves [48]. T6P is synthesized by TREHALOSE-6-PHOSPHATE SYNTHASE1 (TPS1), an enzyme presents in leaves and SAM. Mutation in the TPS1 gene leads to the death of the embryo, demonstrating the key function of this gene in the plant life cycle [49]. An ingenious complementation experiment to produce plants with reduced T6P level by driving TPS1 expression under a seed-specific promoter to promote embryogenesis and seedling development shows that those plants are late flowering [49]. This experiment strongly suggests that T6P is an endogenous signal that positively regulates flowering. Further study confirmed the fundamental role of T6P in the positive regulation of FT and also genes involved in the age-dependent pathway (which will be discussed later in the review) [50] (see Fig. 1). However, it is still unclear how T6P triggers downstream events leading to gene regulation of FT and other targets, including its perception, integration, and activation of transcriptional targets. Nitrate availability is another key component that positively regulates floral transition, for example, a delayed of flowering is often observed when a suboptimal concentration of nitrate is used [51]. Master regulators of nitrates signaling, the TFs NIN-LIKE PROTEIN 6 and 7 (NLP6 and 7) probably directly regulate SQUAMOSA PROMOTER BINDING PROTEIN-LIKE 3 (SPL3) expression in a nitrate-dependent manner. This leads to the upregulation of SOC1 expression, a key floral integrator in the nitrate response, thus promoting bolting [51] (see Fig. 1). Interestingly, crosstalk between nitrate signaling, T6P metabolism and photoperiod pathway exists, highlighting the importance of the metabolic state to flowering regulation [51].

1.2.2 Age-Dependent Pathway

During the vegetative growth, the morphology of successive leaves evolves with age, with changes in size, shape, petiole length, and trichome distribution [52]. This is due to a juvenile-to-adult phase transition conditioning flowering (see [53] for details). This transition is mainly under the control of an interplay between microRNAs and TFs. Vegetative growth is controlled by miR156 and its targets, the members of the SPL TF family [54]. SPL protein levels increase with leaf age, whereas the miR156 is abundant during the juvenile phase. In young leaves, high levels of miR156 repress SPL translation [55]. Thus, the miR156/SPL ratio represents a sensor of the age of the plant. Upon miR156 decrease, SPL proteins are

Flower Development in Arabidopsis

9

produced and induce the synthesis of miR172, which targets several members of the APETALA2 (AP2) TF family [56] (see Fig. 1). Among them, TARGET OF EAT 1 binds to CO and inhibits its activity, leading to FT repression [57]. Some SPLs, such as SPL9 or SPL3, also activate flowering by acting directly on the transcription of integrators such as LFY and SOC1 in the SAM [54, 55, 58]. Genetic analyses revealed that the age-dependent pathway controlling flowering is highly dependent on environmental conditions or plant species. For instance, multiple spl mutants flowered similarly later in both LD or SD conditions in Arabidopsis Col-0 ecotype [54], whereas in others plants such as Arabis alpina or Cardamine flexuosa the interplay between the photoperiod or vernalization pathways revealed the importance of the SPL/miRNA156/miR172 pathway on flowering [59, 60]. 1.3 Integration of the Different Input to Promote Flowering and Flower Formation

2

The different signaling pathways described above converge towards the regulation of FT either in leaves or in the SAM. Signals in favor of blooming result in an accumulation of FT protein in leaves, where it is loaded via the phloem companion cells and transported to the SAM. In association with the bZIP TF FLOWERING LOCUS D (FD), the complex FT/FD activates the transcription of other floral integrators such as SOC1 and APETALA1 (AP1) [2] (see Fig. 1). Using genome-wide analyses, it has been shown that FD binds not only G-boxes (as expected for a bZIP TF) but also non-canonical DNA motifs [61]. Once activated by FT/FD, AP1 and SOC1 (in association with AGL24) directly activate LFY in the SAM to promote flowering [62–64] (see Fig. 1). The FT/FD complex must act transiently to promote the expression of the floral integrators AP1 and LFY since a prolonged FT/FD activity prevents the development of normal flowers [65]. Once the floral integrators accumulate in SAM, the plant starts to produce floral primordia, the earliest recognizable stage of a flower.

Part 2: The First Hours of Arabidopsis Flowers Whether a primordium will become a leaf or a flower, its emergence involves a set of basic regulators including the auxin phytohormone. In both cases, auxin distribution and signaling are critical but the auxin-driven floral primordia emergence requires the action of a few floral regulators, which will confer the floral fate (reviewed in [66]).

2.1 Primordium Positioning and Auxin Signaling

Clear evidence from the role of auxin in flower development came from the study of the pin-formed 1 (pin1) and pinoid (pid) mutants that lack floral primordia: they are a phenocopy of auxin polar transport inhibitor application and can be rescued by exogenous

10

Hicham Chahtane et al.

auxin application [67–70]. Auxin indeed controls the regular and predictable arrangement of leaves and flowers around the stem, called phyllotaxis. Several genetic and mathematical evidences showed that the regulation of auxin localization leads to a selforganized pattern involving local auxin accumulation surrounded by auxin depletion, thereby creating the heterogeneity triggering organ initiation [70–73]. This process is made more robust by the interplay between auxin and cytokinin signaling [74]. Deciphering the roles of auxin distribution in flower initiation implies to detail auxin biosynthesis, transport, and signaling. 2.2 Auxin Biosynthesis and Transport

Local auxin synthesis, in addition to polar auxin transport, also contributes to flower emergence (see [75] for review). The major actors in the tryptophan-dependent auxin biosynthesis are proteins encoded by TRYPTOPHAN AMINOTRANSFERASE OF ARABIDOPSIS/YUCCA (YUC) genes. Their role is most obvious in the yuc1 yuc2 yuc4 yuc6 quadruple mutant, which makes almost no floral bud or only rudimentary incomplete flowers [76]. Auxin dynamic transport, named polar auxin transport, is driven at the plasma membrane by the specific influx and efflux carriers, respectively encoded by the AUXIN RESISTANT1 (AUX1) and PIN FORMED1 (PIN1) gene families [77] (see Fig. 2). Local auxin accumulation and depletion are mediated by asymmetric localization of PIN proteins, which drive auxin in the epidermis and generate an auxin sink below the primordia after they have formed [70, 78]. PIN1 polarity undergoes a dynamic change during organ initiation due to the action of PID, a serine-threonine protein kinase, and of a PP2A-phosphatase responsible for the phosphorylation status and the asymmetric distribution of PIN1 in the young floral meristem [79, 80]. More recently, D6 PROTEIN KINASE (D6PK) has been shown to also phosphorylate PIN1 proteins and be required for efficient PIN1 distribution in the inflorescence [81]. However, the two kinases have non-redundant roles as their phosphosites on PIN1 are different and d6pk and pid display distinct phenotypes. PIN1 dynamics in the cell also involve the MACCHI-BOU4 (MAB4) gene family, which encode NONPHOTOTROPIC HYPOCOTYL 3-like proteins and regulate PIN1 endocytosis. Multiple mutants for MAB4 genes (such as the triple mab4 mel1 mel2 mutant) have pin-like inflorescences [82].

2.3 Downstream of Auxin Signaling

Cells transcriptionally respond to auxin through the nuclear auxin pathway, which involves AUXIN RESPONSE FACTORs (ARFs), Aux/IAA repressors, and the TIR1/AFB auxin co-receptors [70, 83]. ARF5 (also referred to as MONOPTEROS, MP) is the major ARF acting in the floral meristem initiation, as some mp/arf5 mutant alleles lack floral buds on the stem [84, 85] (see Fig. 2). MP induces floral meristem emergence in various ways, such as

Flower Development in Arabidopsis

11

Fig. 2 Key regulators involved in floral meristem initiation. Close up on the nascent FM (stage 2) on the flank of the SAM. Locally produced auxin (for instance IAA) by YUC biosynthesis enzymes is canalized in the epidermis (L1) of the flank of the SAM by the PIN transporters (under the control of PID kinase) to form an auxin maximum. This auxin peak triggers the activation of ARF5/MP TF through degradation of the associated Aux/IAA repressors. ARF5/MP is the main factor between auxin and downstream events. Associated with SYD and BRM chromatin remodelers, it activates the AINT/AIL6 and LFY/RAX1 cascades that control redundantly meristem initiation. ARF5/MP also induces FIL, involved in meristem initiation and in specifying the abaxial territory of the FM. The WUS/CLV meristem homeostasis loop is indicated as central to allow the formation of the floral meristem, but the way it is controlled by upstream actors is still unclear. In collaboration with ARF3 and HDA19, FIL also represses the expression of the SAM marker STM. On the other hand, STM is induced by mechanical stress at the border between the SAM and the FM. This spatial regulation of STM, in collaboration with CUC genes and other partners mentioned in the text, allows the establishment of a clear border between the SAM and the FM

12

Hicham Chahtane et al.

reinforcing local auxin maxima by orienting PIN polarity in a non-cell autonomous manner, possibly via the activation of MAB4 [82, 86]. Auxin activated ARF5 also regulates the expression of two early acting regulators of floral meristem identity, AINTEGUMENTA (ANT), an AP2/ERF TF, and LFY [87–90]. ARF5 recruits the chromatin remodelers BRAHMA (BRM) or SPLAYED (SYD), which are subunits of the SWI/SNF (SWITCHING DEFECTIVE/SUCROSE NON FERMENTING) complex that enables chromatin opening, thus leading to gene activation of ANT and LFY [91] (see Fig. 2). Both ANT and LFY participate in floral primordium emergence, a role that remains cryptic in ant or lfy single mutants but is revealed in mutant combinations such as pin ant ail, lfy pid or lfy pin [92–94]. Moreover, ANT and its homolog AINTEGUMENTA like 6 (AIL6) bind common genomic regions to activate (likely directly) LFY expression in parallel to ARF5 as well as a wide range of genes involved in floral development [95, 96]. The few ARF5 direct targets identified in FM include MAB4 and ARABIDOPSIS HISTIDINE PHOSPHOTRANSFER PROTEIN6 (AHP6), a negative regulator of cytokinin signaling that reinforces the auxin pattern adding robustness to the phyllotaxis [5, 74]. Further efforts such as elucidating the ARF5 binding landscape in floral tissue will shed light on the gene network involved in this step [97]. 2.4 Boundaries and Polarization Set the Limit of the Floral Primordium

Genes involved in abaxial/adaxial polarity such as REVOLUTA (REV) and FILAMENTOUS FLOWER (FIL) are also necessary during flower meristem development as attested by floral defects in the corresponding mutants or the fil rev double [98–101]. REV may function by modulating local response to auxin during floral primordium formation [102]. Other important factors involved in early meristem development include several TFs such as LATERAL SUPPRESSOR (LAS), CUP-SHAPED COTYLEDON 1 to 3 (CUC1–3), or REGULATOR OF AXILLARY MERISTEMS 1 (RAX1) [103–107]. Particularly, RAX1 is expressed in the earliest stage of floral primordium and directly activated by LFY [108]. Conversely, the expression of the SHOOT MERISTEMLESS (STM), a SAM marker gene, is repressed at the site of auxin maxima, where the floral bud forms. This repression is important to promote the initiation of floral primordium and is controlled by at least two convergent pathways [109] (see Fig. 2). First, STM expression is downregulated directly by the repressive class B ARF3 TF that binds to STM second intron. In parallel, ARF5 directly activates FIL, which in turn participates in STM repression [91, 109]. Based on protein-protein interactions and genetic data, it has been proposed that FIL/ARF3 complex jointly recruits the histone deacetylase HDA19 to promote chromatin compaction on STM locus

Flower Development in Arabidopsis

13

and ensure its silencing [109]. This dual function of ARF proteins warrants stable repression of STM in the early floral primordium steps, to allow proper development of floral primordia initiation. 2.5 Establishment of a Transient Stem Cell Identity

Floral meristem initiation also involves re-establishing a stem cell niche in the flower with a combined expression of CLAVATA3 (CLV3), a stem cell marker gene, and WUSCHEL (WUS), a master regulator of the organizing center [110, 111]. As in the SAM, these two regulators form a minimal genetic circuitry that ensures a transient stem cell niche formation. Models have been proposed on how a combination of signals (hormonal or derived from the L1 layer) should re-establish an organizing center when the flower meristem reaches a certain size [112]. However, the necessary connection to the early FM regulators (MP, LFY, ANT) remains elusive (see Fig. 2).

2.6 Modifications in Cell Wall Composition Trigger Floral Outgrowth

Whereas gene regulatory networks between TFs have been unravelled, few connections have been made with terminal actors bearing the enzymatic activities that shape cells. This gap started to be filled when ANT and AIL6 were shown to promote cell wall polysaccharide modification leading to change in cell growth [113]. The importance of cell wall composition is illustrated by mutations in several GLYCOSYLTRANSFERASES OF THE CELLULOSE SYNTHASE LIKE D (CSLD) genes that strongly affect the number of floral primordia [114]. This likely affects auxin distribution, as revealed by the reduced pattern of the auxin transporter PIN1 in some cell wall cellulose biosynthesis mutant, such as cesa3 [115]. Also, flower primordia express a specific set of glycosyltransferases, suggesting that the FM outgrowth is facilitated by local modification of the physical properties of the cell wall [114]. The regulation of plant cell growth is indeed closely linked to the cell wall structure [116]. Combining use of the naked pin1 stem with chemical and physical treatments, Sassi et al. illustrated how auxin affects cell wall stiffness and microtubule orientation. Before floral outgrowth, cell cortical microtubules are anisotropic thereby promoting vertical growth and inhibiting bulging. An auxin maximum reduces anisotropy and cell wall stiffness, promoting the emergence of the FM [117]. This change in cell wall properties and tissue shape generates local mechanical perturbations that rapidly induces STM in FM boundaries [118] (see Fig. 2).

3

Part 3: Determination of the Floral Primordia Identity

3.1 Activation of the Floral Meristem Identity

While mutants affected in floral primordiaoutgrowth show naked stems, other mutants develop lateral structures but they lack a proper floral identity. This is typically the case of lfy, where flowers are usually replaced by inflorescences subtended by leaves or by

14

Hicham Chahtane et al.

abnormal flowers often bearing bracts. LFY encodes an orphan plant-specific TF which controls FM identity together with the MADS-Box gene AP1 or CAULIFLOWER (CAL), a close homolog of AP1 [5, 119, 120]. They are both expressed early in the FM, LFY from the floral stage 0 on and immediately inducing AP1. LFY and AP1 orchestrate the robust acquisition of floral meristem identity thanks to a feed-forward and positive feedback loop between them and with the class I HD-Zip TF LATE MERISTEM IDENTITY2 (LMI2) [63, 121–123]. In parallel to the direct AP1 activation, LFY also modulates the GA hormone pathway: it induces expression of ELA1, which encodes a gibberellin catabolic enzyme thus leading to reduced GA levels in the FM [124]. This GA reduction frees the AP1 activator SPL9 from the repression by the DELLA proteins. Therefore, multiple activations of AP1 by LFY ensure an irreversible acquisition of floral identity (see Fig. 3). Several new LFY and AP1 upstream regulators have been ¨ SCHEN (DRN), known as a recently identified: the DORNRO direct target of MP/ARF5 in the embryo [85], and DRNL and PUCHI TFs (all belonging to the AP2/EREBP family) act redundantly with the BTB/POZ proteins BOP1/2 to regulate LFY expression. The quintuple drn drnl puchi bop1 bop2 exhibits such a strong decrease in LFY expression that it phenocopies the lfy mutant [125]. BOP1/2 proteins not only affect LFY expression, but these components of an E3 ligase complex also seem to promote LFY activity through a post-transcriptional mechanism [126, 127]. 3.2 Repression of the Inflorescence Trait

The irreversible switch towards flower identity requires the repression of the inflorescence identity promoted by TERMINAL FLOWER 1 (TFL1) and other genes such as SOC1, SVP, and AGL24. TFL1 belongs to the same family as FT but acts oppositely: while FT promotes flowering, TFL1 represses flowering due to a small difference in 4 residues critical for the protein function [128, 129]. TFL1 plays a double role: it represses LFY and AP1 expression and promotes the vegetative state of the meristem (see [130] for details) (see Fig. 3). As FT, TFL1 (devoid of DNA binding capability) acts in complex with other TFs, such as the bZIP protein FD to regulate gene expression [61, 131–133]. To allow a stable switch to floral identity, AP1 directly binds CArG boxes in a remote 3′ regulatory region of TFL1 and represses its expression. AP1 also represses the vegetative MADS-box TFs, including SOC1, AGL24, and SVP [63, 134] (see Fig. 3). Until recently, LFY was also thought to directly repress TFL1 [135]. However, recent evidence suggests that LFY directly activates TFL1, and only represses it indirectly by activating AP1 [5, 136, 137]. The functional relevance of this incoherent feedback loop between LFY, AP1, and TFL1 is still unclear but it has been proposed that it could act as a sensor of

Flower Development in Arabidopsis

15

Fig. 3 Gene network specifying floral meristem identity. Close up on the nascent FM (stage 2–3) on the flank of the SAM. Inflorescence and floral genetic identity are antagonistic. Whereas the inflorescence meristem is specified by TFL1, as well as SVP and SOC1, the floral meristem is under the control of LFY and AP1 TFs. LFY is early expressed in the floral anlagen below ARF5/MP (see Fig. 2), but is also controlled by other factors such as PUC, DRN/DRNL, and BOP1/2. LFY induces the expression of AP1 directly, or indirectly through LMI2 and the GA hormonal pathway. AP1 also induces LFY in a positive feedback regulatory loop stabilizing the genetic floral identity. Once induced, AP1 inhibits the flowering time genes and the inflorescence identity genes. Reciprocal inhibition between TFL1 and AP1 is considered as central to maintain exclusive inflorescence and floral identities

floral meristem marker and make sure that floral patterning starts only when the levels of LFY and AP1 are sufficient [136, 138]. A model of the early floral meristem regulatory network has been proposed that captures the different gene expressions observed in wild-type and several mutant backgrounds [139].

16

Hicham Chahtane et al.

3.3 A Common Early Regulatory Network Between LFY and AP1

Despite their completely different DNA binding specificity, LFY and AP1 share several common target genes [140, 141]. Almost 30% (769 genes) of all genomic regions bound by LFY are also bound by AP1. Among them, 25% are both regulated by LFY and AP1. Many of them are linked to flower development, such as genes involved in auxin (MAB4, ARF6, ARF11, and IAA18), the GA catabolism enzymes (ELA1, GA2ox20) and other regulators of floral development (SEP3, RAX1, LMI1, FD, and TFL1) [63, 140, 142, 143]. Another feature shared by LFY and AP1 is their ability to bind to closed chromatin region, a property characteristic of pioneer factors [144–148]. The fact that LFY and AP1 share many bound regions and regulate a common set of target genes suggests that they may act in the same complex. This could be direct physical interaction or indirect association mediated by common co-factors, such as chromatin remodelers BRM and SYD, which are recruited by LFY and to interact with AP1 as shown by mass spectrometry experiments [149, 150]. However, the only MADS-box TF interacting with LFY is SEP3, and available proteomics data do not suggest a direct interaction between LFY and other MADS-box TFs [150, 151]. Another possibility could explain why LFY and AP1 shared common targets. Genes bound by multiple TFs show more dynamic expression during flower development [152]. Thus, genes with multiple different TF binding sites in their regulatory sequence tend to be more rapidly and more efficiently regulated to promote the switch to flower development.

3.4 The Genesis of the Boundary FM-SAM

Emerging flower primordia are separated from the SAM by a group of specialized cells that form a boundary, characterized by reduced growth [153]. This boundary interface separates and influences two territories of different developmental fates, the SAM and the FP. Several TFs belonging to the TALE superfamily contribute redundantly to SAM maintenance, including the KNOX class protein STM, and the BELL class proteins PENNYWISE (PNY) and the closely related POUND-FOOLISH (PNF) [154–156]. Both KNOX and BELL class proteins specify meristems and define boundaries likely by forming functional heterodimers [156]. Several genes involved in boundary formation are regulated, directly or indirectly by the TALE proteins. Some of them are encoded by NAC family TFs CUP-SHAPED COTYLEDON (CUC1–3). CUC1 and 2 act redundantly to establish boundaries of floral primordia [157, 158]. CUC1 directly activates STM expression in a positive feedback loop to generate the boundary between the SAM and the FP, a function probably shared by the closed relatives CUC2 and CUC3 proteins [104, 105, 159]. Similarly to CUCs, BOP1 and BOP2 are expressed in the boundary and also define boundaries of the floral meristem in addition to their role in floral identity [160]. PNY and PNF downregulate BOP1 and BOP2 to

Flower Development in Arabidopsis

17

ensure that their expression domain is restricted to the boundary zone [161]. On the other hand, BOP1, together with the bZIP TF TGACG MOTIF-BINDING FACTOR 4 (TGA4), activate the expression of the BELL class gene ARABIDOPSIS HOMEOBOX 1 (ATH1) in different boundary regions [162]. Other TALE genes are also regulated by BOP and may involve interaction with TGA [163, 164]. Another typical boundary protein is HANABA TARANU (HAN), a GATA-3 type TF expressed in several boundary regions, including in early stage of floral primordium [165, 166]. HAN interacts with proteins specific to the SAM and the boundary, respectively. HAN likely directly regulates both BOP1 and BOP2, but also activates and interacts physically with PINHEAD (PNH), a member of ARGONAUTE family involved in SAM maintenance by sequestering miR166/165 [166, 167]. Boundaries are also characterized by a specific hormonal context that downregulates growth and differentiation. While high CK to low GA promotes SAM maintenance, the opposite hormonal balance (low CK to high GA) prevents differentiation in the boundary zones [168]. HAN reduces CK levels in the boundary by directly activating CYTOKININ OXYDASE 3 (CKK3), a gene involved in cytokinin degradation [166]. Besides, CUC1 and 2 also modulate cytokinin homeostasis in inflorescences, how it acts during early flower development is not yet clear [169].

4 Part 4: Establishment of the Floral Patterning—from the Historical ABC to the Revised ABCDE Model In Arabidopsis, floral organs are arranged into a whorled pattern, with, from outside to inside, four sepals, four petals, six stamens, and two fused carpels. The identity of these organs is specified by the combined actions of floral homeotic genes, namely A-, B-, C-, D- and E-function genes. Their expressions are initiated by floral meristem identity genes, such as LFY and AP1/CAL, and maintained throughout most of the floral organ development. These genes were discovered based on homeotic organ transformation occurring in the respective mutants. The combinatorial action of these genes led to the formulation of the ABC model, later revised to ABCDE model when the role of E-function genes (SEPALLATA1 to 4, SEP1 to SEP4) in all four whorls was discovered [170] (see Fig. 4a). In the ABCDE model, A (AP1 and AP2)- and E-function genes, together, specify sepal formation and development. Genes from class A-, B (APETALA3, AP3; and PISTILLATA, PI)- and E-function genes initiate petal formation and development. Class B-, C (AGAMOUS, AG)- and E-function genes are involved in stamen formation and development. Class

18

Hicham Chahtane et al.

Fig. 4 Floral organs development. (a) Floral organ identity. The Arabidopsis flower is composed of the four main floral organs: sepals (in green), petals (in white), stamens (in blue) and an ovule-containing pistil (salmon and red), as well as nectaries (in grey) located between petals and stamens. The floral quartet model indicates that tetramers of MADS-box TFs specify the identity of the four main organ types. The A function protein AP1 forms part of both sepal and petal tetramers. B function factors, AP3 and PI, forms part of petals and stamen tetramers, whereas C function factor AG participates in stamen, carpel and ovule specific tetramers. All floral quartets include the E function SEP protein. All MADS-Box TFs bind to CArG-boxes on DNA with different specificities. Structural parameters such as the flanking region (represented as dark grey boxes), but also the central region of the CArG box (illustrated as salmon CArG-box*) differ, especially for the complex containing Class E (SEP) and Class C proteins (AG but likely also STK and SHP). Nectaries are mainly under the control of CRC TF, which is upregulated by AG, a C function gene, as well as by UFO, expressed in the flower in whorls 2 and 3. (b) Floral organ growth. Sepal growth is dependent on differential endopolyploidy, whereas JAG TF

Flower Development in Arabidopsis

19

C- and E-function genes are required for carpel formation; and class C-, D- (SHATTERPROOF, SHP; and STEEDSTICK, STK) and E-function genes initiate ovule formation and development. A class genes are induced by LFY and MADS-box TFs such as SOC1 or FUL initially uniformly in the early floral meristem and later cleared from its center when AG induced in the center has been attributed to the synergistic action of LFY and WUS, even if the precise molecular mechanism for this synergy remains to be established as a LFY-WUS complex on AG regulatory sequence has never yet been characterized [171]. AP3 and possibly PI gene expressions are induced by the synergistic action of LFY and UNUSUAL FLORAL ORGANS (UFO). Recent work showed that UFO function as a LFY co-factor that redirects LFY to novel genomic sites where they form a regulatory complex [172]. All homeotic genes encode MADS-box TFs, except for AP2, which belongs to AP2/EREBP class of TF. Using protein interaction assays, such as yeast two-hybrid, pull-down assays, and proteomic approaches, it was later proposed that these MADS-box proteins formed the so-called flower quartet complexes to carry out their floral organ specification function [173]. The crystallographic structure of a SEP3 tetramerization domain was solved, revealing a coiled-coil tetramerization mode [174]. These MADSbox TFs function as hetero quartets in vivo [150, 173, 175] and it will be important to structurally characterize biologically relevant complexes to better understand how different quartets trigger different developmental programs. The importance of tetramerization was probed using plants expressing a version of SEP3 with impaired tetramerization capacity: these plants had altered carpel identity demonstrating the importance of the coil-coiled domain [176]. Recent work showed that different quartets slightly differ in DNA binding specificity due to a key role of the Intervening domain, particularly to specify the preferred spacing between the two binding sites bound by a MADS quartet [177, 178]. Whether AP2 is involved in the complex formation in the floral quartet model is also enigmatic. In recent years, extensive reviews are available in floral quartet model [138, 173, 179, 180]. Here, we mainly focus on the genes directly regulated by A-, B-, C- D- and E-function genes and/or their complexes, and how these target genes lead to the downstream events. Major debates and contradictory findings between Arabidopsis and other plant species are also discussed. ä Fig. 4 (continued) modulates petal growth through the cell wall and cell cycle control. (c) Floral organ boundaries. Boundaries between organs are also genetically specified. PTL controls petal emergence and forms a boundary between sepal and petal domains through RBE regulation. RBE, also controlled by UFO, specifies the border between petals and stamens and SUP between stamens and carpels. CUC TFs control mainly inter-organs boundaries, with other proteins such as HWS, PTL, LSH, or RBE

20

Hicham Chahtane et al.

4.1 Sepal Identity and Maintenance

Mutations in class A genes lead to defects in the first two whorls of the flower. In ap1, petals are absent and sepals replaced by leaves with axillary flowers. In ap2 mutant, sepals have carpelloid characteristics due to AG ectopic expression, petals are absent or replaced by stamenoid structures [181, 182]. Because the A function is strongly questioned outside of the Arabidopsis model [183, 184], it was proposed that A- and E-function genes both play roles in floral meristem identity specification and specify the ground state of a floral organ, the sepal. This led to the hypothesis of (A)BC model, where A- and E-function genes are fused into a single extended A-function [138]. The A/E complex thus fulfils function in early FM identity specification and possibly in sepal identity since leaves replace sepals in the ap1 single or the sep1–4 quadruple mutant [64, 185]. The direct targets of AP1 or AP1/SEP3 complex in sepal specification are still largely unknown and how this process is impacted by AP2 also remains to be determined. Y2H and genetic analyses showed that transcriptional co-repressor complex LEUNIG (LUG) and SEUSS (SEU) can form complexes with AP1 or SEP3 to suppress AG expression in the outer two floral whorls [186–188]. These interactions were confirmed by proteomic analysis, which showed interaction of AP1 with SEU and LUG homolog LEUNIG-HOMOLOG, while also revealed several other co-factors, including chromatinassociated factors such as BRM, SYD and RELATIVE OF EARLY FLOWERING 6 (REF6), and several TFs from the TALE superfamily such as BELLRINGER (BLR) [150]. Interestingly, it has been shown that BLR represses AG in floral and inflorescence meristems, and seems to act synergistically with the general corepressors LUG and SEU [189]. Taken together, it seems that AP1 likely act by forming higher-order complexes with not only MADS TFs but also other TF families, together with recruiting chromatinassociating factors for its function in floral patterning. AP2 suppresses AG in the outer two floral whorls by recruiting the transcriptional co-repressor family TOPLESS/TOPLESS RELATED (TPL/TPR) and HISTONE DEACETYLASE 19 (HDA19) through direct physical interactions [190]. Interestingly, it has been shown that expression of a fusion protein containing AP2 DNA binding domain and the TPL under the control of the TPL regulatory sequence was sufficient to restore AP2 function in ap2 mutant background [190]. Furthermore, AP2-TPLHDA19 complex binds to the regulatory regions of AP3 and SEP3, and expression of AP3, PI and SEP3 was detected in the outer whorls in plants that lack functional copies of AP2, TPL or HDA19 [190]. Therefore, it seems that AP2-TPL-HDA19 complex suppresses both B- and C- function genes in sepals. On the other hand, AP2 does not seem to be repressed by AG directly, rather, this function is mediated by the microRNA miR172, which prevents the accumulation of AP2 mRNA and protein in the inner two whorls [191, 192].

Flower Development in Arabidopsis

4.2

Sepal Growth

21

The outer epidermis of the sepal is characterized by heterogeneous cells population, in shape and size (some cells are almost 40 times longer than other) [193]. Cells with different size express distinct gene markers, such as the membrane protein DEFECTIVE KERNEL1 (DEK1) expressed only in giant cells. The major mechanism of sepal growth is based on endopolyploidy, a process increasing ploidy in individual cells (see Fig. 4b). This process is controlled by LOSS OF GIANT CELLS FROM ORGANS (LGO), a cyclindependent kinase inhibitor promoting cell commitment to endoreduplication [194]. Other mechanisms such as mechanical feedback involving microtubules act also as growth regulator mechanisms [195]. However, genes known to be involved in sepal growth (related to cell size or microtubule dynamics) do not seem to be regulated directly by SEP3, AP2, or AP1, suggesting that another downstream gene network fills the gap between organ identity genes and genes involved in sepal growth [152, 196, 197].

4.3 Petal Identity and Maintenance

B-function gene mutants, i.e., ap3 and pi, have sepals in the second whorl instead of petals, and stamens are replaced by carpels (or filaments) in the third whorl. With the role of E-function genes in all floral whorls, petal identity is thus proposed to be specified by the combinatorial action of A-, B- and E-function genes. Indeed, it has been shown that AP3 and PI physically interact with AP1 and SEP3, and that vegetative leaves of transgenic plants overexpressing AP3, PI and SEP3 are transformed into petaloid organs [198]. Genome-wide identification of target genes by ChIP-seq and transcriptome analysis revealed that AP3/PI act as bimodal TF by activating genes involved in organogenesis and repressing key regulators of carpel formation and ovule development [199]. For example, a key regulator of carpel development, CRABS CLAW (CRC), is precociously expressed in the third whorl of B-function mutants [200]. Furthermore, artificial microRNAs targeting the transcripts of AP3 or PI at various stages of flower development could lead to the transformation of stamens into carpels [199]. Thus, B-function genes seem to determine whether female or male reproductive organs are specified.

4.4

Petal growth is modulated by JAGGED (JAG), which promotes growth in the distal part of the petal [201]. Genome-wide analysis revealed that JAG plays a dual function during the petal development. First, it directly represses several genes involved in meristem development such as HAN, BOP, or LFY. Second, JAG governs cellular activities by controlling the expression of many cell wall synthesis genes as well as genes involved in the cell cycle [202] (see Fig. 4b). Interestingly, AG directly upregulates JAG expression [203], suggesting that JAG acts as a hub between organ identity and growth. The link between petal identity and petal growth could

Petal Growth

22

Hicham Chahtane et al.

also be indirect. BOP1/2 act upstream of JAG and negatively regulate its expression through a negative feedback loop [202, 204]. Interestingly, BOP1/2 are preferentially expressed on the AP3 domain, and are regulated directly by several MADS box TFs including AP3/PI [152]. Recently, a novel mechanism linking MADS-box TFs and petal development was discovered. Several TCP TFs contribute to petal and stamen development [205, 206]. Among them, TCP4, which is directly activated by SEP3 [152]. SEP3 also regulates indirectly TCP4 by repressing miR319a, a major repressor of TCP translation into protein [152, 205]. However, direct targets of TCP during petal development remain to be uncovered. 4.5 Stamen Identity and Maintenance

Stamen identify in Arabidopsis is conferred by B-, C- and E-function genes (see Fig. 4a). For example, in a C-function mutant, i.e., ag mutant, stamens are converted into petals and carpels are replaced by a whole again abnormal flower. It has been shown that AP3, PI (B-function) interact with AG (C-function) and SEP3 (E-function) in yeast and in vitro, and that floral organs of transgenic plants overexpressing AP3, PI, SEP3 and AG are transformed into staminoid organs [198].

4.6 Anther and Stamen Development

AG directly regulates SPOROCYTELESS/NOZZLE (SPL/NZZ), encoding a TF crucial for both anther and ovule developments [207, 208]. Despite its central function in microsporogenesis, little is known on the gene network downstream of SPL/NZZ. Recent findings suggest that miR166 and its target PHABULOSA (PHB) control the spatiotemporal expression of SPL/NZZ during anther development [209]. The phosphorylation of SPL/NZZ is also important for the stability and the activity of the protein during anther development, which adds another mechanism regulating the activity of this protein [210]. SPL/NZZ is also able to confer stamen identity probably via the upregulation of AG and SEP3 genes [211].

4.7 Nectary Formation

Nectaries are small secretory glands responsible for the production of nectar, a sweet reward for insects [212]. Even if interaction with insects is not key in a self-pollinating species, Arabidopsis flowers do possess nectaries. They are present at the base of the stamens facing the petal junction [213]. Genetic analyses revealed that mutations in a dozen of genes (CRC, BOP1/2, AG, SHP1/2, SEP1/2/3, UFO, LFY, AP3, and AG) affect nectary formation. The major actor in nectary formation is CRC, which is essential but not sufficient to trigger ectopic nectaries [214, 215] (see Fig. 4a). The identity of factors that act together with CRC to induce nectary formation still have to be determined. CRC expression is redundantly regulated by class C clade (AG) and class E (SEP1/2/3), likely thanks to the presence of several CArG boxes in its promoter region [214] suggesting that several

Flower Development in Arabidopsis

23

MADS-box TF heterotetramers direct nectary formation by inducing CRC expression [203, 216]. The development of nectaries is also dependent on genes involved in third whorl identity: ectopic nectaries are found in the fourth whorl when carpels are converted into stamens by ectopic UFO expression but not when B genes are ectopically expressed. This suggests that UFO, probably in concert with LFY, define the third whorl that contains both the stamen and the nectaries [215] (see Fig. 4a). Surprisingly, ectopic nectaries can also be observed at the base of the pedicel when UFO, LFY or AP1 activities are modified [215]. 4.8 Carpel and Ovule Identity and Maintenance

We will only briefly introduce the major factors involved in carpel and ovule formation since a description is presented in another chapter of this book. In the Arabidopsis mature flower, two fused carpels (the female reproductive organs) constitute the gynoecium which protects the ovules and ensures the normal propagation of the progeny. The developmental program leading to carpel initiation is induced by the class C gene AG, which is locally activated at the end of stage 3 FM by the LFY/WUS complex [171]. ag mutant has anomalies in the third and fourth whorls: stamens are replaced by petals, and the fourth whorl, normally composed of carpels, is replaced by a new flower due to the indeterminate state of the floral meristem [181, 217] (see Subheading 5 for details in floral determinacy). AG acts in concert with Class E SEP proteins to promote carpel vs leaf development, as revealed by the homeotic conversion of carpels into leaves in the multiple sep mutant [170, 185] (see Fig. 4a). Among the plethora of direct target genes, AG represses leaf identity by directly controlling a set of genes involved in trichome formation, but also modulates cytokinin signaling to control the growth of the gynoecium, probably in a heterocomplex with other MADS-box proteins [203, 218]. Ovule identity is redundantly conferred by the closest paralogs of AG, the homeotic class D genes SHP1, SHP2, and STK (see Fig. 4a). The triple mutant shp1 shp2 stk exhibits a homeotic conversion of ovules into leaf or carpel-like structures [120, 219]. Importantly, SHP2 is a direct target of AG, implying that AG controls different steps of gynoecium development, from its initiation to the formation of normal ovules [203]. Recently, a new mode of regulation of Class C and D genes has been described. The HUA-PEP activity is composed of different genes encoding RNA-binding proteins controlling pre-mRNA processing and maturation. Mutants in the HUA-PEP complex exhibit strong defects in floral organ identity reminiscent to those of ag or shp1 shp2 stk mutants. These phenotypes are due to the production of aberrant AG, SHP1, SHP2, and STK mRNAs prematurely terminated, and an absence of functional class C and D proteins

24

Hicham Chahtane et al.

[220, 221]. Thus, the HUA-PEP activity is crucial for functional messengers and the global development of the gynoecium. The boundary CUC TFs also play a role in ovules separations in a partially redundant manner [222]. 4.9 MADS TF Target Gene Specificity

Despite their DNA binding domains of very similar sequence, MADS-box TFs have different functions in vivo and their genome-wide bound regions are overlapping but not identical (reviewed in [141]). This suggests that the presence of CArG-box signatures is not sufficient to explain the behavior of MADS protein on DNA. Some divergent features might explain their specificity: first, some DNA motifs with specific shapes exist around the CArGbox that might distinguish between them [223]. Second, recent in vitro studies and computational analyses using k-mers pinpointed that MADS-box TFs have different DNA binding preferences [224]. Whereas it is not so easy to visualize in a standard logo representation, the analysis of overrepresented sequences of k-mers does detect differences, especially for MADS complexes containing AG [224] (see Fig. 4a). In addition to the sequence itself, some information key for binding is also encoded in the DNA shape within the binding site, particularly the shape of the minor groove at the center of the CArG-box [223, 225]. These structural parameters could also have consequences on DNA bending. Finally, the distance between CArG-boxes might affect the cooperative binding (looping) of tetramers and depend on its components [176–178, 226–228] as suggested by predictions based on modelling dimerization and tetramerization interfaces of SEP3 [174]. Overall, several parameters have been proposed to explain the differences in tetramer binding and regulation specificity, but it is not yet clear whether those parameters can be used in a highly predictive manner to faithfully predict MADS-quartet target genes.

4.10 Boundaries and Polarity Genes Are Key to Define Proper Organ Identity Domains

Floral whorls and organs are separated by different types of boundaries (inter- and intra-whorl), that are set up early during flower development in coordination with organ development (see Fig. 4c). Boundaries are defined as a group of cells with reduced growth. Several genes have been described as required for boundary formation in flowers; some are flower-specific but others are also required for setting up other boundaries during the vegetative phase. Loss of function mutants of boundary genes are mainly characterized by fused floral organs, and these genes are considered as the housekeepers of floral organ patterning. Some of the best-described boundary genes are the previouslymentioned CUC TFs, belonging to the NAC family, that specify boundaries (see Fig. 4c). Those TFs have many boundary-related roles during the vegetative phase and are known to be required for

Flower Development in Arabidopsis

25

the establishment of the SAM through their direct activation of STM. The three CUC proteins are expressed in the FM boundaries and act partially redundantly. cuc double mutants are characterized by the fusion of organs from the same whorl [104]. CUC activates several genes from the ALOG family such as LIGHT DEPENDENT SHORT HYPOCOTYL 3 and 4 (LSH3/LSH4) to specify boundaries and downregulate growth [229]. CUCs are also posttranscriptionally regulated by the miR164 family, which target CUC mRNA for degradation [230, 231]. miR164c, encoded by EARLY EXTRA PETALS 1 (EEP1), induces CUC1 and CUC2 downregulation. This was further demonstrated by expressing miR164-resistant versions of CUC1 and CUC2, yielding a eep1like phenotype. Interestingly, CUC2 is directly regulated by LFY at stage 3 [232]. Another crucial gene for boundary formation is PETAL LOSS (PTL) (see Fig. 4c). PTL is the first tri-helix TF shown to be required for morphogenesis. ptl mutants are characterized by fused sepals and a highly reduced number of petals. Remaining petals also show several defects [233]. PTL is expressed from early stages in inter-sepal regions where, together with but independently of CUC genes, it down-regulates growth and thus prevents fusion between whorl 1 organs [234]. PTL jointly regulates petal initiation despite not being expressed in petal primordia. PTLinduced reduced growth in inter-sepal region regulates auxin maximum in adjacent petal primordia, required for growth [235]. Furthermore, PTL was shown to be required for activation of downstream genes such as RABBIT EARS (RBE) [236]. RBE is a TF that belongs to the C2H2 zinc finger TF family. Analysis of rbe loss of function mutants showed that RBE is required for petal initiation and development, sepal dissociation and boundary maintenance between the second and third whorl [236, 237]. RBE was shown to bind miR164c promoter and to repress it, enabling CUC1 and CUC2 expression [238]. Furthermore, RBE downregulates TCP genes to initiate petal development [206]. Finally, RBE represses AG in the first and second whorl, as rbe mutant defects in the first and second whorls are removed when AG is mutated [237]. RBE is also regulated by another boundary protein: UFO [172, 237] (see Fig. 4c). UFO is a F-box protein containing a C-terminal Kelch domain which has several roles in flower development such as AP3 activation together with LFY [172]. UFO may also be required for proper boundary formation. In addition to defects observed in second and third whorl, ufo shows fused organs, representative of boundary defects [239]. UFO is expressed in a cup-shape territory during stage 3, and may specify a boundary. A genetic screen on ufo mutant allowed to identified other genes required for proper boundary formation called FUSED FLORAL ORGANS (FFO1,2,3) [240]. FFO1 was later found to correspond to HAWAIIAN SKIRT (HWS) [241]. HWS belongs to the

26

Hicham Chahtane et al.

large F-box family and is thought to contain a Kelch domain. hws mutants are characterized by delayed floral organ abscission, fused sepals and fused stamens (see Fig. 4c). Recently it was shown that hws1 phenotype is due to miRNA164-mediated CUC mRNA downregulation [242]. It also appears that HWS regulates other miRNA pathways, likely through degradation of unknown factors [243]. SUPERMAN (SUP) is another boundary specific gene that specifies the limit between the third and fourth whorls [244, 245]. SUP is a C2H2 zinc finger TF with a C-terminal EAR-like repression domain. sup is characterized by additional stamens or stamen/carpel organs in the fourth whorl instead of carpels. This phenotype is due to AP3 ectopic expression in the fourth whorl, and it was recently shown that extra stamens are produced from whorl 4 cells [246]. SUP-mediated growth suppression may be mediated by negative regulation of auxin biosynthesis genes in the whorl 3-whorl 4 boundary [247].

5

Part 5: Floral Meristem Termination The floral meristem is a determinate structure that stops growing after producing the carpels. This implies that the population of stem cells necessarily disappears from the center of flowers and is reinitiated only in ovules. Several TFs and chromatin remodelling factors [248] are in charge of FM termination including the MADS-box TF AG that represses the stem cell regulator WUS by two distinct mechanisms. First, AG binds directly to CArG boxes present downstream of the WUS gene and represses its expression via the recruitment of polycomb complex [249]. In parallel, AG in complex with SEP3 [176] directly promotes the transcription of KNUCKLES (KNU), encoding a C2H2-zinc finger TF, and participates in the removal of repressive marks in its promoter region [250] (see Fig. 5). Once KNU is expressed, it binds to WUS promoter, causes the eviction of the SWI/SNF ATPase SYD, and interacts with FIE (a member of the PRC2 complex) responsible for the deposition of the H3K27me3 repressive marks to silence WUS transcription and ensure a stable termination of floral identity [251] (see Fig. 5). AG also likely directly regulates MINI ZINC FINGER2 (MIF2), encoding a short protein with an unusual zinc finger domain [252]. MIF2 contributes to floral determinacy, as attested by the multicarpellate fruit developing when MIF2 expression is silenced [252]. MIF2 binds in vivo to the first intron of WUS in a KNU-dependent manner and represses its expression, suggesting that MIF2/KNU form a repressive complex. The repressive activity of this complex is probably conferred by the presence of TOPLESS (TPL) and the histone deacetylase HDA19 [252].

Flower Development in Arabidopsis

27

Fig. 5 Genetic regulation of floral meristem termination. Close up on the FM at stage 6. In the meristem, indeterminate growth depends on the maintenance of the WUS/CLV loop. In the Arabidopsis developing flower bud, WUS is down-regulated, leading to a determinate growth, through a complex genetic and epigenetic network controlled by AG. AG acts either directly, in association with the polycomb repressive complex (PcG), or through KNU TF. KNU prevents WUS expression in association with FIE, a member of PRC2 repressive complex, or with MIF2, an AG target, in association with HD19 and TPL. Another pathway goes through CRC and TRN2 and modulates WUS expression

In parallel to its action on WUS and KNU, AG promotes CRC expression. CRC modulates auxin homeostasis through the direct repression of the plasma membrane protein TORNADO2 (TRN2) and represses WUS repression [175, 253] (see Fig. 5). In addition to AG, the C2H2 transcriptional repressor SUP also represses WUS at the periphery of the WUS expression domain [246].

28

Hicham Chahtane et al.

Those pathways leading to WUS repression, act against AP2, which antagonizes the transcriptional activity of AG by promoting KNU repression [254]. Interestingly, AP2 and AG share common direct targets but oppositely affect their gene expression, suggesting a possible competition due to closed regulatory elements [141, 152, 254].

6

Conclusions and Perspectives

6.1 Coordination and Timing Are Key

As discussed here, flower development requires dozens of key regulators, some of which are involved in several key developmental steps. Thus, at a given time, to induced floral meristem identity and not carpel formation, for example, several regulators coordinate and transiently or suddenly switch from one developmental program to other thanks to feedback loops, epigenetic controls, changes in chromatin accessibility and formation of boundaries. Both positive and negative feedbacks ensure the proper development of the flower, which will end up with a precise and stereotypic architecture of the well-organized Arabidopsis flower. While genetic analyses revealed the major players involved in the different developmental phases discussed here, a lot of mysteries remains on how this coordination is controlled temporally. For example, direct local activation of the stem cell regulatory loop in the early event of FM is largely unknown, as well as the link between organ identity and organ growth.

6.2 Rethinking MADS-Box TF Specificity

Several interesting questions remain on how MADS-box TFs act together in vivo to specify floral identity and patterning. While the floral quartet model represents an attractive hypothesis consistent with genetic conclusions, in vitro biochemical and structural characterizations of the different floral quartet complexes remains challenging. Probably due to the intrinsic problematic of A-function (as discussed earlier), the hypothetical floral quartet involved in petal identity (AP1/SEP3/AP3/PI) was never reconstructed in vitro. So far, only the carpel specification complex (SEP3-AG heterotetramer) has been reconstructed in vitro [150, 176], while the stamen specification complex (AP3-PI-SEP3-AG) was validated in yeast [198]. How the different floral quartet complexes target a specific set of genes remain poorly understood. In a given tissue several floral quartet complexes can be present and in vivo analyses could not discriminate between the different captured floral quartet complexes. Recent advances shed light on the importance of DNA intrinsic properties, including DNA structure features and flanking region for MADS TF specificity. The study of individual floral quartet reconstructed in vitro on a native DNA using recent methods such as DAP-Seq represents an appealing tool to uncover specificity for each floral quartet [177, 255, 256].

Flower Development in Arabidopsis

29

Interestingly, non-flowering plants such as gymnosperms have also paralogues to class B or C genes, and several models proposed a functional homotetramer of MADS-box proteins to specify non-floral organ identity [173]. It raises the question of whether homodimers or homotetramers observed in vitro (as observed for AP1, AG and SEP3) could exist and be functional in planta in flowering plants [150]. Understanding how MADS box complexes behave in plants devoid of flowers should provide a better understanding of the floral quartet complexes and specificity. 6.3 Reaching the Single Cell Level

As for many developmental processes, performing genomic studies and gene regulatory network reconstruction in the floral tissue is challenging. This is mainly due to all the diversity of cellular identities present in complex tissues or organs. Although still in its infancy, the field of single-cell genomics offers promising perspectives to access to events occurring locally, in a subgroups of specialized cells [257, 258].

6.4 The Cost of Making a Flower

Plant fitness is highly dependent on its ability to allocate resources during the reproductive phase to promote the growth of the flower. The activation of costly developmental programs, such as flower development, requires fine-tuning the balance between growth and other energy-consuming developmental aspects. Different transcriptomic data revealed a link between some floral regulators and defence responses. While ANT/AIL6 repress plant defence by modulating SA and JA signaling pathways, LFY reduces defence responses and callose deposition triggered by pathogen attacks [113, 143]. Other floral regulators such as AG also participate in the modulation of plant defences, as revealed by transcriptome analyses [254]. Even if the molecular mechanisms mediating the trade-off between floral development and plant defence crosstalk are not well documented, plants may need to reallocate resources dedicated to defence to fulfil the formation of the flower.

References 1. Quiroz S et al (2021) Beyond the genetic pathways, flowering regulation complexity in Arabidopsis thaliana. Int J Mol Sci 22:5716 2. Wigge PA et al (2005) Integration of spatial and temporal information during floral induction in Arabidopsis. Science 309:1056–1059 3. Jaeger KE, Wigge PA (2007) FT protein acts as a long-range signal in Arabidopsis. Curr Biol 17:1050–1054 4. Wigge PA, Thomas D (2011) FT, a mobile developmental signal in plants. Curr Biol 21: R374–R378

5. Denay G, Chahtane H, Tichtinsky G, Parcy F (2017) A flower is born: an update on Arabidopsis floral meristem formation. Curr Opin Plant Biol 35:15–22 6. Pin PA, Nilsson O (2012) The multifaceted roles of FLOWERING LOCUS T in plant development. Plant Cell Environ 35:1742– 1755 7. Susila H, Nasim Z, Ahn JH (2018) Ambient temperature-responsive mechanisms coordinate regulation of flowering time. Int J Mol Sci 19:3196

30

Hicham Chahtane et al.

8. Antoniou-Kourounioti RL, Zhao Y, Dean C, Howard M (2021) Feeling every bit of winter – distributed temperature sensitivity in vernalization. Front Plant Sci 12:628726 9. Jin S, Ahn JH (2021) Regulation of flowering time by ambient temperature: repressing the repressors and activating the activators. New Phytol 230:938–942 10. Brightbill CM, Sung S (2022) Temperaturemediated regulation of flowering time in Arabidopsis thaliana. aBIOTECH 3:78–84 11. Michaels SD, Amasino RM (1999) FLOWERING LOCUS C encodes a novel MADS domain protein that acts as a repressor of flowering. Plant Cell 11:949–956 12. Deng W et al (2011) FLOWERING LOCUS C (FLC) regulates development pathways throughout the life cycle of Arabidopsis. Proc Natl Acad Sci U S A 108:6680–6685 13. Li M et al (2016) DELLA proteins interact with FLC to repress flowering transition. J Integr Plant Biol 58:642–655 14. Sung S, Amasino RM (2004) Vernalization in Arabidopsis thaliana is mediated by the PHD finger protein VIN3. Nature 427:159–164 15. Whittaker C, Dean C (2017) The FLC locus: a platform for discoveries in epigenetics and adaptation. Annu Rev Cell Dev Biol 33:555– 575 16. Hepworth J, Dean C (2015) Flowering locus C’s lessons: conserved chromatin switches underpinning developmental timing and adaptation. Plant Physiol 168:1237–1245 17. Swiezewski S, Liu F, Magusin A, Dean C (2009) Cold-induced silencing by long antisense transcripts of an Arabidopsis Polycomb target. Nature 462:799–802 18. Heo JB, Sung S (2011) Vernalizationmediated epigenetic silencing by a long intronic noncoding RNA. Science 331:76–79 19. Srikanth A, Schmid M (2011) Regulation of flowering time: all roads lead to Rome. Cell Mol Life Sci 68:2013–2037 20. Andres F, Coupland G (2012) The genetic basis of flowering responses to seasonal cues. Nat Rev Genet 13:627–639 21. Marquardt S et al (2014) Functional consequences of splicing of the antisense transcript COOLAIR on FLC transcription. Mol Cell 54:156–165 22. Susila H et al (2021) Florigen sequestration in cellular membranes modulates temperatureresponsive flowering. Science 373:1137– 1142 23. Jaillais Y, Parcy F (2021) Lipid-mediated regulation of flowering time. Science 373:1086– 1087

24. Lee JH et al (2007) Role of SVP in the control of flowering time by ambient temperature in Arabidopsis. Genes Dev 21:397–402 25. Li D et al (2008) A repressor complex governs the integration of flowering signals in Arabidopsis. Dev Cell 15:110–120 26. Melzer R (2017) Regulation of flowering time: a splicy business. J Exp Bot 68:5017–5020 27. Pose D et al (2013) Temperature-dependent regulation of flowering by antagonistic FLM variants. Nature 503:414–417 28. Capovilla G, Symeonidi E, Wu R, Schmid M (2017) Contribution of major FLM isoforms to temperature-dependent flowering in Arabidopsis thaliana. J Exp Bot 68:5117–5127 29. Jin S et al (2022) FLOWERING LOCUS M isoforms differentially affect the subcellular localization and stability of SHORT VEGETATIVE PHASE to regulate temperatureresponsive flowering in Arabidopsis. Mol Plant 15:1696–1709 30. Lee JH et al (2013) Regulation of temperature-responsive flowering by MADSbox transcription factor repressors. Science 342:628–632 31. Kumar SV et al (2012) Transcription factor PIF4 controls the thermosensory activation of flowering. Nature 484:242–245 32. Sakamoto T, Kimura S (2018) Plant temperature sensors. Sensors 18:4365 33. Han X et al (2019) Arabidopsis transcription factor TCP5 controls plant thermomorphogenesis by positively regulating PIF4 activity. iScience 15:611–622 34. Tasset C et al (2018) POWERDRESSmediated histone deacetylation is essential for thermomorphogenesis in Arabidopsis thaliana. PLoS Genet 14:e1007280 35. Reed JW et al (2000) Independent action of ELF3 and phyB to control hypocotyl elongation and flowering time. Plant Physiol 122: 1149–1160 36. Song YH et al (2018) Molecular basis of flowering under natural long-day conditions in Arabidopsis. Nat Plants 4:824–835 37. Box MS et al (2015) ELF3 controls thermoresponsive growth in Arabidopsis. Curr Biol 25:194–199 38. Ezer D et al (2017) The evening complex coordinates environmental and endogenous signals in Arabidopsis. Nat Plants 3:17087 39. Silva CS et al (2020) Molecular mechanisms of evening complex activity in Arabidopsis. Proc Natl Acad Sci U S A 117:6901–6909

Flower Development in Arabidopsis 40. Jung J-H et al (2020) A prion-like domain in ELF3 functions as a thermosensor in Arabidopsis. Nature 585:256–260 41. Shim JS, Kubota A, Imaizumi T (2017) Circadian clock and photoperiodic flowering in Arabidopsis: CONSTANS is a hub for signal integration. Plant Physiol 173:5–15 42. Valverde F et al (2004) Photoreceptor regulation of CONSTANS protein in photoperiodic flowering. Science 303:1003–1006 43. Jang S et al (2008) Arabidopsis COP1 shapes the temporal pattern of CO accumulation conferring a photoperiodic flowering response. EMBO J 27:1277–1288 44. Song YH, Ito S, Imaizumi T (2013) Flowering time regulation: photoperiod- and temperature-sensing in leaves. Trends Plant Sci 18:575–583 45. Yu J-W et al (2008) COP1 and ELF3 control circadian function and photoperiodic flowering by regulating GI stability. Mol Cell 32: 617–630 46. Yin R, Ulm R (2017) How plants cope with UV-B: from perception to response. Curr Opin Plant Biol 37:42–48 47. Arongaus AB et al (2018) Arabidopsis RUP2 represses UVR8-mediated flowering in noninductive photoperiods. Genes Dev 32:1332– 1343 48. Lunn JE et al (2006) Sugar-induced increases in trehalose 6-phosphate are correlated with redox activation of ADPglucose pyrophosphorylase and higher rates of starch synthesis in Arabidopsis thaliana. Biochem J 397:139– 148 49. Gomez LD, Gilday A, Feil R, Lunn JE, Graham IA (2010) AtTPS1-mediated trehalose 6-phosphate synthesis is essential for embryogenic and vegetative growth and responsiveness to ABA in germinating seeds and stomatal guard cells. Plant J 64:1–13 50. Wahl V et al (2013) Regulation of flowering by trehalose-6-phosphate signaling in Arabidopsis thaliana. Science 339:704–707 51. Olas JJ et al (2019) Nitrate acts at the Arabidopsis thaliana shoot apical meristem to regulate flowering time. New Phytol 223:814– 827 52. Huijser P, Schmid M (2011) The control of developmental phase transitions in plants. Development 138:4117–4129 53. Hyun Y, Richter R, Coupland G (2017) Competence to flower: age-controlled sensitivity to environmental cues. Plant Physiol 173:36–46 54. Xu M et al (2016) Developmental functions of miR156-regulated SQUAMOSA PROMOTER BINDING PROTEIN-LIKE

31

(SPL) genes in Arabidopsis thaliana. PLoS Genet 12:e1006263 55. Wang J, Czech B, Weigel D (2009) miR156regulated SPL transcription factors define an endogenous flowering pathway in Arabidopsis thaliana. Cell 138:738–749 ´ ’Maoile´idigh DS et al (2021) Systematic 56. O analyses of the MIR172 family members of Arabidopsis define their distinct roles in regulation of APETALA2 during floral transition. PLoS Biol 19:e3001043 57. Zhang B, Wang L, Zeng L, Zhang C, Ma H (2015) Arabidopsis TOE proteins convey a photoperiodic signal to antagonize CONSTANS and regulate flowering time. Genes Dev 29:975–987 58. Yamaguchi A et al (2009) The microRNAregulated SBP-Box transcription factor SPL3 is a direct upstream activator of LEAFY, FRUITFULL, and APETALA1. Dev Cell 17:268–278 59. Zhou C-M et al (2013) Molecular basis of age-dependent vernalization in Cardamine flexuosa. Science 340:1097–1100 60. Bergonzi S et al (2013) Mechanisms of age-dependent response to winter temperature in perennial flowering of Arabis alpina. Science 340:1094–1097 61. Collani S, Neumann M, Yant L, Schmid M (2019) FT modulates genome-wide DNA-binding of the bZIP transcription factor FD. Plant Physiol 180:367–380 62. Lee J, Oh M, Park H, Lee I (2008) SOC1 translocated to the nucleus by interaction with AGL24 directly regulates leafy. Plant J 55: 832–843 63. Kaufmann K et al (2010) Orchestration of floral initiation by APETALA1. Science 328: 85–89 64. Liljegren SJ, Gustafson-Brown C, Pinyopich A, Ditta GS, Yanofsky MF (1999) Interactions among APETALA1, LEAFY, and TERMINAL FLOWER1 specify meristem fate. Plant Cell 11:1007–1018 65. Abe M et al (2019) Transient activity of the florigen complex during the floral transition in Arabidopsis thaliana. Development 146: dev171504 66. Wang Y, Jiao Y (2018) Auxin and aboveground meristems. J Exp Bot 69:147–154 67. Okada K, Ueda J, Komaki MK, Bell CJ, Shimura Y (1991) Requirement of the auxin polar transport system in early stages of Arabidopsis floral bud formation. Plant Cell 3: 677–684 68. Bennett SRM, Alvarez J, Bossinger G, Smyth DR (1995) Morphogenesis in pinoid mutants of Arabidopsis thaliana. Plant J 8:505–520

32

Hicham Chahtane et al.

69. Reinhardt D, Mandel T, Kuhlemeier C (2000) Auxin regulates the initiation and radial position of plant lateral organs. Plant Cell 12:507–518 70. Reinhardt D et al (2003) Regulation of phyllotaxis by polar auxin transport. Nature 426: 255–260 71. Jonsson H, Heisler MG, Shapiro BE, Meyerowitz EM, Mjolsness E (2006) An auxindriven polarized transport model for phyllotaxis. Proc Natl Acad Sci U S A 103:1633– 1638 72. Smith RS et al (2006) A plausible model of phyllotaxis. Proc Natl Acad Sci U S A 103: 1301–1306 73. Truskina J, Vernoux T (2018) The growth of a stable stationary structure: coordinating cell behavior and patterning at the shoot apical meristem. Curr Opin Plant Biol 41:83–88 74. Besnard F et al (2014) Cytokinin signalling inhibitory fields provide robustness to phyllotaxis. Nature 505:417–421 75. Zhao Y (2018) Essential roles of local auxin biosynthesis in plant development and in adaptation to environmental changes. Annu Rev Plant Biol 69:417–435 76. Cheng Y, Dai X, Zhao Y (2006) Auxin biosynthesis by the YUCCA flavin monooxygenases controls the formation of floral organs and vascular tissues in Arabidopsis. Genes Dev 20:1790–1799 77. Petrasek J, Friml J (2009) Auxin transport routes in plant development. Development 136:2675–2688 78. Benkova E et al (2003) Local, effluxdependent auxin gradients as a common module for plant organ formation. Cell 115:591– 602 79. Christensen SK, Dagenais N, Chory J, Weigel D (2000) Regulation of auxin response by the protein kinase PINOID. Cell 100:469–478 80. Michniewicz M et al (2007) Antagonistic regulation of PIN phosphorylation by PP2A and PINOID directs auxin flux. Cell 130:1044– 1056 81. Zourelidou M et al (2014) Auxin efflux by PIN-FORMED proteins is activated by two different protein kinases, D6 PROTEIN KINASE and PINOID. Elife 3:e02860 82. Furutani M, Nakano Y, Tasaka M (2014) MAB4-induced auxin sink generates local auxin gradients in Arabidopsis organ formation. Proc Natl Acad Sci U S A 111:1198– 1203 83. Leyser O (2018) Auxin signaling. Plant Physiol 176:465–479 84. Przemeck GK, Mattsson J, Hardtke CS, Sung ZR, Berleth T (1996) Studies on the role of

the Arabidopsis gene MONOPTEROS in vascular development and plant cell axialization. Planta 200:229–237 85. Cole M et al (2009) DORNROSCHEN is a direct target of the auxin response factor MONOPTEROS in the Arabidopsis embryo. Development 136:1643–1651 86. Bhatia N et al (2016) Auxin acts through MONOPTEROS to regulate plant cell polarity and pattern phyllotaxis. Curr Biol 26: 3202–3208 87. Yamaguchi N et al (2013) A molecular framework for auxin-mediated initiation of flower primordia. Dev Cell 24:271–282 88. Blazquez MA, Soowal LN, Lee I, Weigel D (1997) LEAFY expression and flower initiation in Arabidopsis. Development 124:3835– 3844 89. Krizek B (2009) AINTEGUMENTA and AINTEGUMENTA-LIKE6 act redundantly to regulate Arabidopsis floral growth and patterning. Plant Physiol 150:1916–1929 90. Long J, Barton MK (2000) Initiation of axillary and floral meristems in Arabidopsis. Dev Biol 218:341–353 91. Wu MF et al (2015) Auxin-regulated chromatin switch directs acquisition of flower primordium founder fate. Elife 4:e09269 92. Krizek BA (2011) Aintegumenta and Aintegumenta-Like6 regulate auxin-mediated flower development in Arabidopsis. BMC Res Notes 4:176 93. Ezhova TA, Soldatova OP, Kalinina AI, Medvedev SS (2000) Interaction of ABRUPTUS/ PINOID and LEAFY genes during floral morphogenesis in Arabidopsis thaliana (L) Heynh. Genetika 36:1682–1687 94. Yamaguchi N, Wu M-F, Winter CM, Wagner D (2014) LEAFY and polar auxin transport coordinately regulate Arabidopsis flower development. Plants 3:251–265 95. Yamaguchi N, Jeong CW, Nole-Wilson S, Krizek BA, Wagner D (2016) AINTEGUMENTA and AINTEGUMENTA-LIKE6/ PLETHORA3 induce LEAFY expression in response to auxin to promote the onset of flower formation in Arabidopsis. Plant Physiol 170:283–293 96. Krizek BA, Blakley IC, Ho Y-Y, Freese N, Loraine AE (2020) The Arabidopsis transcription factor AINTEGUMENTA orchestrates patterning genes and auxin signaling in the establishment of floral growth and form. Plant J 103:752–768 97. Stigliani A et al (2019) Capturing auxin response factors syntax using DNA binding models. Mol Plant 12:822–832

Flower Development in Arabidopsis 98. Talbert PB, Adler HT, Parks DW, Comai L (1995) The REVOLUTA gene is necessary for apical meristem development and for limiting cell divisions in the leaves and stems of Arabidopsis thaliana. Development 121: 2723–2735 99. Chen Q et al (1999) The Arabidopsis FILAMENTOUS FLOWER gene is required for flower formation. Development 126:2715– 2726 100. Sawa S et al (1999) FILAMENTOUS FLOWER, a meristem and organ identity gene of Arabidopsis, encodes a protein with a zinc finger and HMG-related domains. Genes Dev 13:1079–1088 101. Otsuga D, DeGuzman B, Prigge MJ, Drews GN, Clark SE (2001) REVOLUTA regulates meristem initiation at lateral positions. Plant J 25:223–236 102. Ram H et al (2020) An integrated analysis of cell-type specific gene expression reveals genes regulated by REVOLUTA and KANADI1 in the Arabidopsis shoot apical meristem. PLoS Genet 16:e1008661 103. Greb T et al (2003) Molecular analysis of the LATERAL SUPPRESSOR gene in Arabidopsis reveals a conserved control mechanism for axillary meristem formation. Genes Dev 17: 1175–1187 104. Aida M, Ishida T, Fukaki H, Fujisawa H, Tasaka M (1997) Genes involved in organ separation in Arabidopsis: an analysis of the cup-shaped cotyledon mutant. Plant Cell 9: 841–857 105. Hibara K et al (2006) Arabidopsis CUP-SHAPED COTYLEDON3 regulates postembryonic shoot meristem and organ boundary formation. Plant Cell 18:2946– 2957 106. Keller T, Abbott J, Moritz T, Doerner P (2006) Arabidopsis REGULATOR OF AXILLARY MERISTEMS1 controls a leaf axil stem cell niche and modulates vegetative development. Plant Cell 18:598–611 107. Muller D, Schmitz G, Theres K (2006) Blind homologous R2R3 Myb genes control the pattern of lateral meristem initiation in Arabidopsis. Plant Cell 18:586–597 108. Chahtane H et al (2013) A variant of LEAFY reveals its capacity to stimulate meristem development by inducing RAX1. Plant J 74: 678–689 109. Chung Y et al (2019) Auxin response factors promote organogenesis by chromatinmediated repression of the pluripotency gene SHOOTMERISTEMLESS. Nat Commun 10:886

33

110. Brand U, Fletcher JC, Hobe M, Meyerowitz EM, Simon R (2000) Dependence of stem cell fate in Arabidopsis on a feedback loop regulated by CLV3 activity. Science 289: 617–619 111. Yadav RK et al (2011) WUSCHEL protein movement mediates stem cell homeostasis in the Arabidopsis shoot apex. Genes Dev 25: 2025–2030 112. Gruel J, Deichmann J, Landrein B, Hitchcock T, Jonsson H (2018) The interaction of transcription factors controls the spatial layout of plant aerial stem cell niches. NPJ Syst Biol Appl 4:36 113. Krizek BA et al (2016) RNA-Seq links the transcription factors AINTEGUMENTA and AINTEGUMENTA-LIKE6 to cell wall remodeling and plant defense pathways. Plant Physiol 171:2069–2084 114. Yang W et al (2016) Regulation of meristem morphogenesis by cell wall synthases in Arabidopsis. Curr Biol 26:1404–1415 115. Sampathkumar A et al (2019) Primary wall cellulose synthase regulates shoot apical meristem mechanics and growth. Development 146:dev179036 116. Traas J (2018) Organogenesis at the shoot apical meristem. Plants (Basel) 8(1):6 117. Sassi M et al (2014) An auxin-mediated shift toward growth isotropy promotes organ formation at the shoot meristem in Arabidopsis. Curr Biol 24:2335–2342 118. Landrein B et al (2015) Mechanical stress contributes to the expression of the STM homeobox gene in Arabidopsis shoot meristems. Elife 4:e07811 119. Huala E, Sussex IM (1992) LEAFY interacts with floral homeotic genes to regulate Arabidopsis floral development. Plant Cell 4:901– 913 120. Pinyopich A et al (2003) Assessing the redundancy of MADS-box genes during carpel and ovule development. Nature 424:85–88 121. Parcy F, Nilsson O, Busch MA, Lee I, Weigel D (1998) A genetic framework for floral patterning. Nature 395:561–566 122. Wagner D, Sablowski RW, Meyerowitz EM (1999) Transcriptional activation of APETALA1 by LEAFY. Science 285:582–584 123. Pastore JJ et al (2011) LATE MERISTEM IDENTITY2 acts together with LEAFY to activate APETALA1. Development 138: 3189–3198 124. Yamaguchi N et al (2014) Gibberellin acts positively then negatively to control onset of flower formation in Arabidopsis. Science 344: 638–641

34

Hicham Chahtane et al.

125. Chandler JW, Werr W (2017) DORNROSCHEN, DORNROSCHEN-LIKE, and PUCHI redundantly control floral meristem identity and organ initiation in Arabidopsis. J Exp Bot 68:3457–3472 126. Zhang B et al (2017) BLADE-ON-PETIOLE proteins act in an E3 ubiquitin ligase complex to regulate PHYTOCHROME INTERACTING FACTOR 4 abundance. Elife 6:e26759 127. Chahtane H et al (2018) LEAFY activity is post-transcriptionally regulated by BLADE ON PETIOLE2 and CULLIN3 in Arabidopsis. New Phytol 220:579–592 128. Ahn JH et al (2006) A divergent external loop confers antagonistic activity on floral regulators FT and TFL1. EMBO J 25:605–614 129. Ho WWH, Weigel D (2014) Structural features determining flower-promoting activity of Arabidopsis FLOWERING LOCUS T. Plant Cell 26:552–564 130. Pe´rilleux C, Bouche´ F, Randoux M, Ormanligeza B (2019) Turning meristems into fortresses. Trends Plant Sci 24:431–442 131. Hanano S, Goto K (2011) Arabidopsis TERMINAL FLOWER1 is involved in the regulation of flowering time and inflorescence development through transcriptional repression. Plant Cell 23:3172–3184 132. Goretti D et al (2020) TERMINAL FLOWER1 functions as a mobile transcriptional cofactor in the shoot apical meristem. Plant Physiol 182:2081–2095 133. Zhu Y et al (2020) TERMINAL FLOWER 1-FD complex target genes and competition with FLOWERING LOCUS T. Nat Commun 11:5118 134. Liu C et al (2007) Specification of Arabidopsis floral meristem identity by repression of flowering time genes. Development 134: 1901–1910 135. Parcy F, Bomblies K, Weigel D (2002) Interaction of LEAFY, AGAMOUS and TERMINAL FLOWER1 in maintaining floral meristem identity in Arabidopsis. Development 129:2519–2527 136. Goslin K et al (2017) Transcription factor interplay between LEAFY and APETALA1/ CAULIFLOWER during floral initiation. Plant Physiol 174:1097–1109 137. Serrano-Mislata A et al (2017) Regulatory interplay between LEAFY, APETALA1/ CAULIFLOWER and TERMINAL FLOWER1: new insights into an old relationship. Plant Signal Behav 12:e1370164 138. Thomson B, Wellmer F (2019) Molecular regulation of flower development. In: Grossniklaus U (ed) Plant development and

evolution, Current topics in developmental biology, vol 131. Elsevier, Amsterdam, pp 185–210 139. Azpeitia E et al (2021) Cauliflower fractal forms arise from perturbations of floral gene networks. Science 373:192–197 140. Winter CM, Yamaguchi N, Wu M, Wagner D (2015) Transcriptional programs regulated by both LEAFY and APETALA1 at the time of flower formation. Physiol Plant 155:55–73 141. Yan W, Chen D, Kaufmann K (2016) Molecular mechanisms of floral organ specification by MADS domain proteins. Curr Opin Plant Biol 29:154–162 142. Moyroud E et al (2011) Prediction of regulatory interactions from genome sequences using a biophysical model for the Arabidopsis LEAFY transcription factor. Plant Cell 23: 1293–1306 143. Winter CM et al (2011) LEAFY target genes reveal floral regulatory logic, cis motifs, and a link to biotic stimulus response. Dev Cell 20: 430–443 144. Pajoro A et al (2014) Dynamics of chromatin accessibility and gene regulation by MADSdomain transcription factors in flower development. Genome Biol 15:R41 145. Sayou C et al (2016) A SAM oligomerization domain shapes the genomic binding landscape of the LEAFY transcription factor. Nat Commun 7:11222 146. Lai X, Verhage L, Hugouvieux V, Zubieta C (2018) Pioneer factors in animals and plantscolonizing chromatin for gene regulation. Molecules 23:1914 147. Jin R et al (2021) LEAFY is a pioneer transcription factor and licenses cell reprogramming to floral fate. Nat Commun 12:626 148. Lai X et al (2021) The LEAFY floral regulator displays pioneer transcription factor properties. Mol Plant 14:829–837 149. Wu M-F et al (2012) SWI2/SNF2 chromatin remodeling ATPases overcome polycomb repression and control floral organ identity with the LEAFY and SEPALLATA3 transcription factors. Proc Natl Acad Sci U S A 109: 3576–3581 150. Smaczniak C et al (2012) Characterization of MADS-domain transcription factor complexes in Arabidopsis flower development. Proc Natl Acad Sci U S A 109:1560–1565 151. Liu C, Xi W, Shen L, Tan C, Yu H (2009) Regulation of floral patterning by flowering time genes. Dev Cell 16:711–722 152. Chen D, Yan W, Fu L-Y, Kaufmann K (2018) Architecture of gene regulatory networks controlling flower development in Arabidopsis thaliana. Nat Commun 9:4534

Flower Development in Arabidopsis 153. Breuil-Broyer S et al (2004) High-resolution boundary analysis during Arabidopsis thaliana flower development. Plant J 38:182–192 154. Smith HMS, Campbell BC, Hake S (2004) Competence to respond to floral inductive signals requires the homeobox genes PENNYWISE and POUND-FOOLISH. Curr Biol 14:812–817 155. Ung N, Lal S, Smith HMS (2011) The role of PENNYWISE and POUND-FOOLISH in the maintenance of the shoot apical meristem in Arabidopsis. Plant Physiol 156:605–614 156. Hamant O, Pautot V (2010) Plant development: a TALE story. C R Biol 333:371–381 157. Aida M, Tasaka M (2006) Genetic control of shoot organ boundaries. Curr Opin Plant Biol 9:72–77 158. Yu H, Huang T (2016) Molecular mechanisms of floral boundary formation in Arabidopsis. Int J Mol Sci 17:317 159. Spinelli SV, Martin AP, Viola IL, Gonzalez DH, Palatnik JF (2011) A mechanistic link between STM and CUC1 during Arabidopsis development. Plant Physiol 156:1894–1904 160. Xu M et al (2010) Arabidopsis BLADE-ONPETIOLE1 and 2 promote floral meristem fate and determinacy in a previously undefined pathway targeting APETALA1 and AGAMOUS-LIKE24. Plant J 63:974–989 161. Khan M et al (2015) Repression of lateral organ boundary genes by PENNYWISE and POUND-FOOLISH is essential for meristem maintenance and flowering in Arabidopsis. Plant Physiol 169:2166–2186 162. Wang Y et al (2019) Clade I TGACG-motif binding basic leucine zipper transcription factors mediate BLADE-ON-PETIOLE-dependent regulation of development. Plant Physiol 180:937–951 163. Khan M et al (2012) Antagonistic interaction of BLADE-ON-PETIOLE1 and 2 with BREVIPEDICELLUS and PENNYWISE regulates Arabidopsis inflorescence architecture. Plant Physiol 158:946–960 164. Khan M, Tabb P, Hepworth SR (2012) BLADE-ON-PETIOLE1 and 2 regulate Arabidopsis inflorescence architecture in conjunction with homeobox genes KNAT6 and ATH1. Plant Signal Behav 7:788–792 165. Zhao Y et al (2004) HANABA TARANU is a GATA transcription factor that regulates shoot apical meristem and flower development in Arabidopsis. Plant Cell 16:2586– 2600 166. Ding L et al (2015) HANABA TARANU (HAN) bridges meristem and organ primordia boundaries through PINHEAD, JAGGED, BLADE-ON-PETIOLE2 and

35

CYTOKININ OXIDASE 3 during flower development in Arabidopsis. PLoS Genet 11:e1005479 167. Ji L et al (2011) ARGONAUTE10 and ARGONAUTE1 regulate the termination of floral stem cells through two microRNAs in Arabidopsis. PLoS Genet 7:e1001358 168. Jasinski S et al (2005) KNOX action in Arabidopsis is mediated by coordinate regulation of cytokinin and gibberellin activities. Curr Biol 15:1560–1565 169. Cucinotta M et al (2018) CUP-SHAPED COTYLEDON1 (CUC1) and CUC2 regulate cytokinin homeostasis to determine ovule number in Arabidopsis. J Exp Bot 69:5169– 5176 170. Pelaz S, Ditta GS, Baumann E, Wisman E, Yanofsky MF (2000) B and C floral organ identity functions require SEPALLATA MADS-box genes. Nature 405:200–203 171. Lohmann JU et al (2001) A molecular link between stem cell regulation and floral patterning in Arabidopsis. Cell 105:793–803 172. Rieu P et al (2022) The F-box cofactor UFO redirects the LEAFY floral regulator to novel cis-elements. bioRxiv. https://doi.org/10. 1101/20220614495942 173. Theissen G, Melzer R, Rumpler F (2016) MADS-domain transcription factors and the floral quartet model of flower development: linking plant development and evolution. Development 143:3259–3271 174. Puranik S et al (2014) Structural basis for the oligomerization of the MADS domain transcription factor SEPALLATA3 in Arabidopsis. Plant Cell 26:3603–3615 175. Hugouvieux V, Zubieta C (2018) MADS transcription factors cooperate: complexities of complex formation. J Exp Bot 69:1821– 1823 176. Hugouvieux V et al (2018) Tetramerization of MADS family transcription factors SEPALLATA3 and AGAMOUS is required for floral meristem determinacy in Arabidopsis. Nucleic Acids Res 46:4966–4977 177. Lai X et al (2020) Genome-wide binding of SEPALLATA3 and AGAMOUS complexes determined by sequential DNA-affinity purification sequencing. Nucleic Acids Res 48: 9637–9648 178. Lai X et al (2021) The intervening domain is required for DNA-binding and functional identity of plant MADS transcription factors. Nat Commun 12:4760 179. Prunet N, Jack TP (2014) Flower development in Arabidopsis: there is more to it than learning your ABCs. Methods Mol Biol 1110: 3–33

36

Hicham Chahtane et al.

180. Kramer EM (2019) Plus ca change, plus c’est la meme chose: the developmental evolution of flowers. Curr Top Dev Biol 131:211–238 181. Bowman JL, Smyth DR, Meyerowitz EM (1989) Genes directing flower development in Arabidopsis. Plant Cell 1:37–52 182. Irish VF, Sussex IM (1990) Function of the apetala-1 gene during Arabidopsis floral development. Plant Cell 2:741–753 183. Morel P et al (2017) Divergence of the floral A-function between an asterid and a rosid species. Plant Cell 29:1605–1621 184. Monniaux M, Vandenbussche M (2018) How to evolve a perianth: a review of cadastral mechanisms for perianth identity. Front Plant Sci 9:1573 185. Ditta G, Pinyopich A, Robles P, Pelaz S, Yanofsky MF (2004) The SEP4 gene of Arabidopsis thaliana functions in floral organ and meristem identity. Curr Biol 14:1935–1940 186. Conner J, Liu Z (2000) LEUNIG, a putative transcriptional corepressor that regulates AGAMOUS expression during flower development. Proc Natl Acad Sci U S A 97:12902– 12907 187. Franks RG, Wang C, Levin JZ, Liu Z (2002) SEUSS, a member of a novel family of plant regulatory proteins, represses floral homeotic gene expression with LEUNIG. Development 129:253–263 188. Sridhar VV, Surendrarao A, Liu Z (2006) APETALA1 and SEPALLATA3 interact with SEUSS to mediate transcription repression during flower development. Development 133:3159–3166 189. Bao X, Franks RG, Levin JZ, Liu Z (2004) Repression of AGAMOUS by BELLRINGER in floral and inflorescence meristems. Plant Cell 16:1478–1489 190. Krogan NT, Hogan K, Long JA (2012) APETALA2 negatively regulates multiple floral organ identity genes in Arabidopsis by recruiting the co-repressor TOPLESS and the histone deacetylase HDA19. Development 139:4180–4190 191. Chen X (2004) A microRNA as a translational repressor of APETALA2 in Arabidopsis flower development. Science 303:2022–2025 192. Wollmann H, Mica E, Todesco M, Long JA, Weigel D (2010) On reconciling the interactions between APETALA2, miR172 and AGAMOUS with the ABC model of flower development. Development 137:3633–3642 193. Roeder AHK, Cunha A, Ohno CK, Meyerowitz EM (2012) Cell cycle regulates cell type in the Arabidopsis sepal. Development 139: 4416–4427

194. Robinson DO et al (2018) Ploidy and size at multiple scales in the Arabidopsis sepal. Plant Cell 30:2308–2329 195. Hervieux N et al (2016) A mechanical feedback restricts sepal growth and shape in Arabidopsis. Curr Biol 26:1019–1028 196. Kaufmann K et al (2009) Target genes of the MADS transcription factor SEPALLATA3: integration of developmental and hormonal pathways in the Arabidopsis flower. PLoS Biol 7:e1000090 197. Yant L et al (2010) Orchestration of the floral transition and floral development in Arabidopsis by the bifunctional transcription factor APETALA2. Plant Cell 22:2156–2170 198. Honma T, Goto K (2001) Complexes of MADS-box proteins are sufficient to convert leaves into floral organs. Nature 409:525– 529 199. Wuest SE et al (2012) Molecular basis for the specification of floral organs by APETALA3 and PISTILLATA. Proc Natl Acad Sci U S A 109:13452–13457 200. Bowman JL, Smyth DR (1999) CRABS CLAW, a gene that regulates carpel and nectary development in Arabidopsis, encodes a novel protein with zinc finger and helixloop-helix domains. Development 126: 2387–2396 201. Sauret-Gueto S, Schiessl K, Bangham A, Sablowski R, Coen E (2013) JAGGED controls Arabidopsis petal growth and shape by interacting with a divergent polarity field. PLoS Biol 11:e1001550 ˜ o JM, Sablowski R (2014) 202. Schiessl K, Muin Arabidopsis JAGGED links floral organ patterning to tissue growth by repressing Kip-related cell cycle inhibitors. Proc Natl Acad Sci U S A 111:2830–2835 203. OMaoileidigh DS et al (2013) Control of reproductive floral organ identity specification in Arabidopsis by the C function regulator AGAMOUS. Plant Cell 25:2482–2503 204. Norberg M, Holmlund M, Nilsson O (2005) The BLADE ON PETIOLE genes act redundantly to control the growth and development of lateral organs. Development 132: 2203–2213 205. Nag A, King S, Jack T (2009) miR319a targeting of TCP4 is critical for petal growth and development in Arabidopsis. Proc Natl Acad Sci U S A 106:22534–22539 206. Huang T, Irish VF (2015) Temporal control of plant organ growth by TCP transcription factors. Curr Biol 25:1765–1770 207. Ito T et al (2004) The homeotic protein AGAMOUS controls microsporogenesis by

Flower Development in Arabidopsis regulation of SPOROCYTELESS. Nature 430:356–360 208. Wei B et al (2015) The molecular mechanism of sporocyteless/nozzle in controlling Arabidopsis ovule development. Cell Res 25:121– 134 209. Li X, Lian H, Zhao Q, He Y (2019) microRNA166 monitors SPOROCYTELESS/ NOZZLE (SPL/NZZ) for building of the anther internal boundary. Plant Physiol 181: 208–220 210. Zhao F et al (2017) Phosphorylation of SPOROCYTELESS/NOZZLE by the MPK3/6 kinase is required for anther development. Plant Physiol 173:2265–2277 211. Liu X et al (2009) The SPOROCYTELESS/ NOZZLE gene is involved in controlling stamen identity in Arabidopsis. Plant Physiol 151:1401–1411 212. Kram BW, Carter CJ (2009) Arabidopsis thaliana as a model for functional nectary analysis. Sex Plant Reprod 22:235–246 213. Alvarez-Buylla ER et al (2010) Flower development. Arabidopsis Book 8:e0127 214. Lee J-Y et al (2005) Recruitment of CRABS CLAW to promote nectary development within the eudicot clade. Development 132: 5021–5032 215. Baum SF, Eshed Y, Bowman JL (2001) The Arabidopsis nectary is an ABC-independent floral structure. Development 128:4657– 4667 216. Morel P et al (2018) The floral C-lineage genes trigger nectary development in petunia and Arabidopsis. Plant Cell 30:2020–2037 217. Yanofsky MF et al (1990) The protein encoded by the Arabidopsis homeotic gene agamous resembles transcription factors. Nature 346:35–39 218. O’Maoileidigh DS, Stewart D, Zheng B, Coupland G, Wellmer F (2018) Floral homeotic proteins modulate the genetic program for leaf development to suppress trichome formation in flowers. Development 145: dev157784 219. Favaro R et al (2003) MADS-box protein complexes control carpel and ovule development in Arabidopsis. Plant Cell 15:2603– 2611 220. Rodriguez-Cazorla E et al (2015) K-homology nuclear ribonucleoproteins regulate floral organ identity and determinacy in arabidopsis. PLoS Genet 11:e1004983 221. Rodriguez-Cazorla E et al (2018) Ovule identity mediated by pre-mRNA processing in Arabidopsis. PLoS Genet 14:e1007182

37

222. Goncalves B et al (2015) A conserved role for CUP-SHAPED COTYLEDON genes during ovule development. Plant J 83:732–742 ˜ o JM, Smaczniak C, Angenent GC, 223. Muin Kaufmann K, van Dijk ADJ (2014) Structural determinants of DNA recognition by plant MADS-domain transcription factors. Nucleic Acids Res 42:2138–2146 ˜ o JM, Chen D, Angenent 224. Smaczniak C, Muin GC, Kaufmann K (2017) Differences in DNA binding specificity of floral homeotic protein complexes predict organ-specific target genes. Plant Cell 29:1822–1835 225. Kappel S, Melzer R, Rumpler F, Gafert C, Theissen G (2018) The floral homeotic protein SEPALLATA3 recognizes target DNA sequences by shape readout involving a conserved arginine residue in the MADS-domain. Plant J 95:341–357 226. Melzer R, Theissen G (2009) Reconstitution of “floral quartets” in vitro involving class B and class E floral homeotic proteins. Nucleic Acids Res 37:2723–2736 227. Melzer R, Verelst W, Theissen G (2009) The class E floral homeotic protein SEPALLATA3 is sufficient to loop DNA in ‘floral quartet’-like complexes in vitro. Nucleic Acids Res 37: 144–157 228. Jetha K, Theissen G, Melzer R (2014) Arabidopsis SEPALLATA proteins differ in cooperative DNA-binding during the formation of floral quartet-like complexes. Nucleic Acids Res 42:10927–10942 229. Takeda S et al (2011) CUP-SHAPED COTYLEDON1 transcription factor activates the expression of LSH4 and LSH3, two members of the ALOG gene family, in shoot organ boundary cells. Plant J 66:1066–1077 230. Laufs P, Peaucelle A, Morin H, Traas J (2004) MicroRNA regulation of the CUC genes is required for boundary size control in Arabidopsis meristems. Development 131:4311– 4322 231. Mallory AC, Dugas DV, Bartel DP, Bartel B MicroRNA regulation of (2004) NAC-domain targets is required for proper formation and separation of adjacent embryonic, vegetative, and floral organs. Curr Biol 14:1035–1046 232. Yamaguchi N, Wu M-F, Winter CM, Wagner D (2014) LEAFY and polar auxin transport coordinately regulate Arabidopsis flower development. Plants (Basel) 3:251–265 233. Griffith ME, da Silva Conceicao A, Smyth DR (1999) PETAL LOSS gene regulates initiation and orientation of second whorl organs in the Arabidopsis flower. Development 126: 5635–5644

38

Hicham Chahtane et al.

234. Brewer PB et al (2004) PETAL LOSS, a trihelix transcription factor gene, regulates perianth architecture in the Arabidopsis flower. Development 131:4035–4045 235. Lampugnani ER, Kilinc A, Smyth DR (2013) Auxin controls petal initiation in Arabidopsis. Development 140:185–194 236. Takeda S, Matsumoto N, Okada K (2004) RABBIT EARS, encoding a SUPERMANlike zinc finger protein, regulates petal development in Arabidopsis thaliana. Development 131:425–434 237. Krizek BA, Lewis MW, Fletcher JC (2006) RABBIT EARS is a second-whorl repressor of AGAMOUS that maintains spatial boundaries in Arabidopsis flowers. Plant J 45:369– 383 238. Huang T, Lopez-Giraldez F, Townsend JP, Irish VF (2012) RBE controls microRNA164 expression to effect floral organogenesis. Development 139:2161–2169 239. Levin JZ, Meyerowitz EM (1995) UFO: an Arabidopsis gene involved in both floral meristem and floral organ development. Plant Cell 7:529–548 240. Levin JZ, Fletcher JC, Chen X, Meyerowitz EM (1998) A genetic screen for modifiers of UFO meristem activity identifies three novel FUSED FLORAL ORGANS genes required for early flower development in Arabidopsis. Genetics 149:579–595 241. Gonzalez-Carranza ZH et al (2007) Hawaiian skirt: an F-box gene that regulates organ fusion and growth in Arabidopsis. Plant Physiol 144:1370–1382 242. Gonzalez-Carranza ZH et al (2017) HAWAIIAN SKIRT controls size and floral organ number by modulating CUC1 and CUC2 expression. PLoS One 12:e0185106 243. Zhang X et al (2017) The Arabidopsis thaliana F-box gene HAWAIIAN SKIRT is a new player in the microRNA pathway. PLoS One 12:e0189788 244. Sakai H, Medrano LJ, Meyerowitz EM (1995) Role of SUPERMAN in maintaining Arabidopsis floral whorl boundaries. Nature 378:199–203 245. Bowman JL et al (1992) SUPERMAN, a regulator of floral homeotic genes in Arabidopsis. Development 114:599–615 246. Prunet N, Yang W, Das P, Meyerowitz EM, Jack TP (2017) SUPERMAN prevents class B gene expression and promotes stem cell

termination in the fourth whorl of Arabidopsis thaliana flowers. Proc Natl Acad Sci U S A 114:7166–7171 247. Xu Y et al (2018) SUPERMAN regulates floral whorl boundaries through control of auxin biosynthesis. EMBO J 37:e97499 248. Sun B, Ito T (2015) Regulation of floral stem cell termination in Arabidopsis. Front Plant Sci 6:17 249. Liu X et al (2011) AGAMOUS terminates floral stem cell maintenance in Arabidopsis by directly repressing WUSCHEL through recruitment of Polycomb Group proteins. Plant Cell 23:3654–3670 250. Sun B, Xu Y, Ng K-H, Ito T (2009) A timing mechanism for stem cell maintenance and differentiation in the Arabidopsis floral meristem. Genes Dev 23:1791–1804 251. Sun B et al (2019) Integration of transcriptional repression and Polycomb-mediated silencing of WUSCHEL in floral meristems. Plant Cell 31:1488–1505 252. Bollier N et al (2018) At-MINI ZINC FINGER2 and Sl-INHIBITOR OF MERISTEM ACTIVITY, a conserved missing link in the regulation of floral meristem termination in Arabidopsis and tomato. Plant Cell 30:83– 100 253. Yamaguchi N, Huang J, Xu Y, Tanoi K, Ito T (2017) Fine-tuning of auxin homeostasis governs the transition from floral stem cell maintenance to gynoecium formation. Nat Commun 8:1125 254. Huang Z et al (2017) APETALA2 antagonizes the transcriptional activity of AGAMOUS in regulating floral stem cells in Arabidopsis thaliana. New Phytol 215:1197– 1209 255. O’Malley RC et al (2016) Cistrome and epicistrome features shape the regulatory DNA landscape. Cell 166:1598 256. Bartlett A et al (2017) Mapping genome-wide transcription-factor binding sites using DAP-seq. Nat Protoc 12:1659–1672 257. Neumann M et al (2022) A 3D gene expression atlas of the floral meristem based on spatial reconstruction of single nucleus RNA sequencing data. Nat Commun 13:2838 ˜ o JC, Kaufmann K 258. Xu X, Smaczniak C, Muin (2021) Cell identity specification in plants: lessons from flower development. J Exp Bot 72:4202–4217

Chapter 2 Flower Development in the Solanaceae Marie Monniaux and Michiel Vandenbussche Abstract Flower development is the process leading from a reproductive meristem to a mature flower with fully developed floral organs. This multi-step process is complex and involves thousands of genes in intertwined regulatory pathways; navigating through the FLOR-ID website will give an impression of this complexity and of the astonishing amount of work that has been carried on the topic (Bouche´ et al., Nucleic Acids Res 44:D1167–D1171, 2016). Our understanding of flower development mostly comes from the model species Arabidopsis thaliana, but numerous other studies outside of Brassicaceae have helped apprehend the conservation of these mechanisms in a large evolutionary context (Moyroud and Glover, Curr Biol 27: R941–R951, 2017; Smyth, New Phytol 220:70–86, 2018; Soltis et al., Ann Bot 100:155–163, 2007). Integrating additional species and families to the research on this topic can only advance our understanding of flower development and its evolution. In this chapter, we review the contribution that the Solanaceae family has made to the comprehension of flower development. While many of the general features of flower development (i.e., the key molecular players involved in flower meristem identity, inflorescence architecture or floral organ development) are similar to Arabidopsis, our main objective in this chapter is to highlight the points of divergence and emphasize specificities of the Solanaceae. We will not discuss the large topics of flowering time regulation, inflorescence architecture and fruit development, and we will restrict ourselves to the mechanisms included in a time window after the floral transition and before the fertilization. Moreover, this review will not be exhaustive of the large amount of work carried on the topic, and the choices that we made to describe in large details some stories from the literature are based on the soundness of the functional work performed, and surely as well on our own preferences and expertise. First, we will give a brief overview of the Solanaceae family and some of its specificities. Then, our focus will be on the molecular mechanisms controlling floral organ identity, for which extended functional work in petunia led to substantial revisions to the famous ABC model. Finally, after reviewing some studies on floral organ initiation and growth, we will discuss floral organ maturation, using the examples of the inflated calyx of the Chinese lantern Physalis and petunia petal pigmentation. Key words Flower, Development, Solanaceae, Nightshades, ABC model, Floral organ identity, Petunia, Tomato, Physalis

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_2, © Springer Science+Business Media, LLC, part of Springer Nature 2023

39

40

1

Marie Monniaux and Michiel Vandenbussche

The Solanaceae Family (Nightshades): Crops and Model Species The Solanaceae, also called nightshades, belong to the Asteridsclade in the eudicots and gather more than 2,500 species and 115 genera [1]. The family has a world-wide distribution but the greatest diversity in species is found in south and central America, probably reflecting the origin of the family. The Solanaceae include many agronomically important crops, such as tomato (Solanum lycopersicum), tobacco (Nicotiana tabacum), potato (Solanum tuberosum), petunia (Petunia × hybrida), pepper (Capsicum annuum) or eggplant (Solanum melongena) (see Fig. 1). Several members of the family produce alkaloid compounds with an agronomical interest, such as nicotine in tobacco, or psychoactive or poisonous substances such as those found in belladonna, stramonium or mandrake. Indeed, the name Solanaceae might come from the Latin solare, meaning “to soothe,” in reference to the pharmaceutical properties of many members of the family. The Solanaceae display a large variety of inflorescence structures but the family is typified by the characteristic cymose inflorescence, where the terminal flower dies out and new flowers grow from lateral buds. The flowers generally have a type-5 symmetry with five petals fused at different degrees and five stamens partly fused to the corolla, and the ovary can develop into either a fleshy (e.g., tomato) or dry (e.g., petunia) fruit.

Fig. 1 Flowers and inflorescences from various Solanaceae members: Solanum lycopersicum (tomato, a), Nicotiana tabacum (tobacco, b), Solanum tuberosum (potato, c), Petunia × hybrida (petunia, d), Capsicum annuum (pepper, e), Brugmansia sp. (f), fruits from the arlequin tomato mutant (g), fruit from Physalis alkegengi (h). (Picture credits respectively, from (a) to (h): Niek Willems, H. Zell, Keith Weller, Michiel Vandenbussche, Simone Stibbe, Tom Morphy, arlequin tomato picture reproduced with authorization from Rafael Lozano [2], and Michael Gasperl. Licenses Creative Commons Attribution-Share Alike 2.5 or 3.0 Generic or Unported)

Molecular Mechanisms of Flower Development in the Nightshades (Solanaceae). . .

41

The Solanaceae contain model species such as petunia, tomato, potato, and tobacco that can be easily transformed with Agrobacterium tumefaciens, and whose genomes have been sequenced (https://solgenomics.net/) [3]. These species of agronomical interest have been bred to improve specific traits, and therefore constitute good model systems to understand the genetic processes of domestication [4]. Petunia is a famous bedding plant with high agronomic value, and hundreds of floral varieties selected over the years provide a large repertoire for variation in flower morphology. Particularly, petunia has been instrumental to decipher the anthocyanin biosynthesis pathway involved in petal pigmentation [5] and evolution of flower color in relation to changes in pollinators [6]. Moreover, a large transposon-insertion database allows for reverse genetics to be carried out easily, which distinguishes it from the other model Solanaceae species [7, 8]. In tobacco, the famous Maryland Mammoth short-days flowering mutant plant revealed the role of photoperiodism on the floral transition [9, 10], but afterwards tobacco has been little used for research on flower development [3]. The same stands true for potato, although this species has revealed the parallel between flowering and tuberization that uses a similar molecular toolkit with daylength-dependent traveling proteins [11]. Tomato was a major model species for classical genetics before the rise of Arabidopsis and is still widely used as a model fruit-bearing crop [3, 12]. In the next paragraphs, we will review how members of the Solanaceae family, particularly petunia, have contributed to a better understanding of how flowers are built.

2

Specification of Floral Organ Identity: Revisions to the ABC Model Environmental and endogenous signals inform the plant when to flower, and at that point, the vegetative meristem that produces leaves will turn into an inflorescence meristem that produces flowers. At an early stage, groups of cells in the flower will acquire a specific identity depending on their position, by the apposition of a precise molecular identity. In Arabidopsis, this floral organ identity is specified by the combinatorial action of A-, B- and C-class proteins, as summarized in the classical ABC model and in the floral quartet model that derives from it [13–15] (see Fig. 2). Although the ABC model is generally considered valid across a wide variety of flowering plants, this often relies on expression data only and the tedious genetic work that is needed to validate the function of ABC genes has only been done in a few species. For this reason, extended work in petunia, where forward and reverse genetics can be carried at a large scale [8], has been key to extend the ABC model to species outside of Brassicaceae. In the following paragraphs, we will review this work, together with pieces of evidence from other nightshades,

42

Marie Monniaux and Michiel Vandenbussche

Fig. 2 The textbook ABC model and its variation in petunia, as we currently understand it. In the textbook ABC model, A-class genes and C-class genes mutually repress their expression domains. A-class genes alone and in combination with B-class genes specify the identities of sepals (se) and petals (pe) respectively, C-class genes alone and in combination with B-class genes specify the identities of carpels (ca) and stamens (st) respectively. In the petunia model, the two C-class genes PMADS3 and FBP6 redundantly share the C-function and specify carpel identity alone, and stamen identity together with the four B-class genes PhDEF, PhTM6, PhGLO1 and PhGLO2. These four genes show additional patterns of subfunctionalization detailed in Fig. 4. C-class genes additionally specify nectary (nect) and ovule (ov) development. PhTM6 is expressed in the stamens and carpels and its expression is activated by C-class genes; however, the function of PhTM6 in the carpels is unknown so far. The (A)-function is ensured by the genes BL, BEN, ROBs and AP1-like: BL and BEN repress C-class genes expression in sepals and petals, and BEN, ROBs and AP1-like genes repress B-class genes expression in the sepal whorl. In this model, the colored boxes represent the region of action of the genes and not necessarily their domain of expression (BL, BEN, and the ROB genes are for instance expressed in all floral organs)

to illustrate how a detailed floral organ identity patterning model, quite distant to the idealized ABC model found in textbooks, has been built over the years (see Fig. 2). To help the reader mostly familiar with Arabidopsis genes, we have included a phylogeny of the ABC MADS-box transcription factors discussed in this chapter, showing orthologs in petunia and tomato (see Fig. 3). 2.1 Redefining the AFunction

In contrast to the B- and C-functions that are generally well conserved in flowering plants, the molecular identity of the players encoding the A-function, and even the existence of the A-function itself, have been debated [19–21]. A-class genes, as formulated in the classical ABC model, have a dual role: on one hand they antagonize the expression of C-class genes in the two first whorls of the flower, and on the other hand they specify alone the identity of sepals, and together with B-class genes the identity of petals [13, 14]. As a result, A-class mutants are expected to develop carpels in the first whorl and stamens in the second whorl of the flower. In Arabidopsis, APETALA1 (AP1) and AP2 are classified as A-class genes. The ap1 mutant flowers lack petals and have sepals that display bract features [22, 23], while the ap2

Molecular Mechanisms of Flower Development in the Nightshades (Solanaceae). . .

43

Fig. 3 Neighbor joining trees of euAP2 transcription factors (left), B/C/D class (middle) and AP1/SEP/AGL6 superclade (right) MADS-box transcription factors. Prefixes At, Ph and Sl represent Arabidopsis thaliana, Petunia x hybrida and Solanum lycopersicum respectively. The euAP2 phylogenetic analysis was obtained using the pipeline offered by https://ngphylogeny.fr, and the tree was rooted with Arabidopsis ANT, an AP2 transcription factor not belonging to the euAP2 lineage. Both MADS-box trees were rooted with Ph-UNS/ FBP20, a SOC1 subfamily member. Bootstrap values (based on 1000 replicates) supporting tree branching above 70% are indicated near the branching points. Scalebars correspond with 0.1 substitions per site. The MADS trees were computed with Treecon software [18] using the Tajima & Nei Distance Calculation method and further default settings

mutant sepals are converted into carpel- or leaf-like organs, and its petals are absent or transformed into stamen-like structures [24, 25], suggesting indeed that AP2 (and AP1 to some extent) is necessary for sepals and petals to form correctly and to repress C-class gene expression from these organs. However, such expected A-class mutants were never clearly found outside of Arabidopsis, suggesting that the A-function as initially defined is not universal [19]. Instead, in Antirrhinum only a gain-of-function mutant of the C-class gene PLE produces the expected A-class mutant phenotype [26], showing that the wild-type sepal and petal identity is rather due to the repression of B- and C-class gene expression from the outer whorls of the flower, than by an “added value” of A-class genes [20, 27]. For these reasons, variations to the ABC model have been proposed, particularly the (A)BC model [13, 20, 28]. In this model, the (A) function merely provides a floral context in which the B- and C-class genes are active to specify petal, stamen, and carpel identity, while sepals represent the ground floral organ identity. In other words, the A-function plays a cadastral role, setting the outside boundaries of B- and C-genes expression, rather than specifying a precise floral organ identity. In petunia, the cadastral function of A-class genes on C-class gene expression is clearly apparent in the double mutant for BLIND (BL) and BLIND ENHANCER (BEN), in which petals are absent or stamenoid, and sepals are converted into carpel-like structures [29]. BL encodes a microRNA from the miR169 family [30] while BEN is a member of the euAP2 lineage [29], to which the

44

Marie Monniaux and Michiel Vandenbussche

Arabidopsis A-class gene AP2 also belong. Therefore, in petunia the C-repression role of the (A) function is fulfilled by the combinatorial action of two distinct molecular players: a microRNA and an AP2-type transcription factor. Although BEN encodes the functional equivalent of Arabidopsis AP2, they are not orthologs since BEN belongs to the TOE-type genes of the euAP2 lineage (see Fig. 3), members of which in Arabidopsis redundantly act as floral repressors [31–34]. Combining the ben mutant with mutations in the euAP2-type genes ROB1, ROB2, and ROB3 (orthologous to Arabidopsis AP2) causes a complete homeotic conversion of sepals into petals, witnessing the full derepression of B-class genes in the first whorl [29]. In addition, mutations in the four members of the AP1/SQUA subfamily (euAP1, PFG, FLORAL BINDING PROTEIN 26 (FBP26) and FBP29, see Fig. 3) results in terminal flowers with normal petals and sepals with petaloid sectors, this last feature being strongly enhanced when combined with mutations in ROBs genes [35]. This demonstrates the existence of a cadastral function to restrict B-class gene expression to its correct domain, fulfilled by BEN, ROBs and AP1/SQUA genes in petunia. In contrast, in Arabidopsis only indirect evidence for a repressive role of AP2 on B-class genes had been reported [36]. Therefore, in petunia a combination of different molecular players restrict B- and C-class gene expression to their respective domains, fulfilling altogether the cadastral part of the (A) function (see Fig. 2). It is now clear in the Solanaceae family that AP1- and AP2-like genes are not needed to specify petal identity. In tomato, a knockout insertion mutant in the gene MACROCALYX (MC), orthologous to the A-function gene AP1, shows no defects in petal identity, with sepals being enlarged like bracts [37, 38]. However, this could be masked by redundancy with other AP1-like genes. But as shown previously, in petunia a quadruple mutant of the four AP1/SQUA genes leads to terminal flowers with perfectly normal petals, and with enlarged sepals with petaloid sectors, as a result of B-class gene derepression [35]. Similarly, the rob1 rob2 rob3 mutant (ROBs being orthologous to Arabidopsis AP2) forms petals, although with some growth and pigmentation defects [29]. To conclude, neither AP1-like nor AP2-like genes are necessary for basic specification of petal identity in petunia flowers, in sharp contrast to what is known in Arabidopsis and generalized in the textbook ABC model. 2.2 Variations on the B-Function: Specializing for a Single Floral Organ

The classical ABC model states that B-genes specify petals and stamens, in combination with A- and C-genes respectively. Indeed, B-class mutants in eudicots usually show a phenotype affecting both the petal and stamen whorls and converting them into sepals and carpels respectively; this is for instance the case in tomato when the APETALA3 (AP3) ortholog STAMENLESS is knocked-out [39, 40]. However, the situation is different in petunia: mutant in

Molecular Mechanisms of Flower Development in the Nightshades (Solanaceae). . .

45

the B-class gene PhDEFICIENS (PhDEF), also known as green petals, has a full conversion of petals into sepals, but stamens remain unaffected [41]. PhDEF is expressed in petal and stamen primordia, suggesting that another gene redundantly controls stamen identity [16]. This gene was found to be its paralog PhTM6, which resembles the ancestral paleoAP3 type of B-class genes, rather than the classical euAP3 type B-class genes to which PhDEF belongs (see Fig. 3) [42]. Surprisingly, PhTM6 appeared to be expressed rather as a C-class gene in whorls 3 and 4 [17]. This unconventional expression pattern for a B-class gene is caused by the fact that C-class genes activate PhTM6 expression [21]. The phtm6 mutant has no visible phenotype, while the phdef phtm6 double mutant displays full homeotic conversion of petals into sepals and stamens into carpels, as would be expected for a B-class mutant [17]. Thus, in petunia, B-class genes from the AP3/DEF clade have specialized into controlling petal and stamen, or only stamen, identity in a partially redundant fashion (see Fig. 4). Furthermore, it was shown that the AP3/DEF proteins act in obligate heterocomplexes with PISTILLATA/GLOBOSA (PI/ GLO) proteins, also belonging to the B-class family (see Fig. 3), and that this complex activates its own expression [43, 44]. For instance, Arabidopsis AP3 interacts with PI, and Antirrhinum DEF interacts with GLO [45, 46]. In petunia, there are two PI/GLO proteins, both expressed in petals and stamens, and their interaction pattern with the two AP3/DEF proteins is logically more complex than in Arabidopsis or Antirrhinum: PhDEF interacts with both PhGLO1 and PhGLO2, while PhTM6 only interacts with PhGLO2, which was confirmed genetically in the corresponding mutants [16]. In addition, fusion of the stamen filaments with the inner petal tube is specifically regulated by the PhDEF/PhGLO1 heterodimer [16]. In summary, the increased complexity in gene number and protein interaction pattern in petunia led to subtle subfunctionalization of these genes in specifying petal and stamen identity and development (see Fig. 4). But how the different protein complexes divide up tasks to regulate target genes and specify correct organ identity at the molecular level remains to be understood. 2.3 Multiple Functions for C-Class Genes

The C-function, i.e., the specification of stamen and carpel identity, generally coupled with the control of floral determinacy, is controlled by members of the AGAMOUS (AG) family. This family is subdivided into the euAG and PLENA (PLE) clades (see Fig. 3), whose names come from the rosid species Arabidopsis and the asterid species Antirrhinum, where AG and PLE respectively specify the C-function [13, 14]. In contrast, the Arabidopsis PLE-like genes SHATTERPROOF1 and 2 (SHP1/2) play a late role in fruit shattering but are not essential for the C-function [47], while the Antirrhinum euAG-like gene FARINELLI (FAR) only has little

46

Marie Monniaux and Michiel Vandenbussche

Fig. 4 Subfunctionalization of the B-class genes for specification of petal and stamen identity in petunia. (a) Expression domains of the PhDEF, PhTM6, PhGLO1 and PhGLO2 genes in petal and stamen initiation domains. (b) Protein complexes formed between AP3/DEF- and PI/GLO-type proteins in each floral organ, based on [16, 17]

contribution to stamen development [48]. These observations led to build a model where, after the gene duplication that generated the euAG- and PLE-clades, control of the C-function has been randomly allocated to either member of the two clades, with the other member adopting a distinct function after changes in gene expression pattern or protein biochemical properties [48]. In Solanaceae, several evidences suggest that the C-function is largely redundantly controlled by both members of the euAG and PLE clades, which might reflect the ancestral state just after the

Molecular Mechanisms of Flower Development in the Nightshades (Solanaceae). . .

47

euAG/PLE duplication [48, 49]. Indeed in petunia, single and double knock-out mutants in the euAG-like gene PETUNIA MADS BOX GENE3 (PMADS3) and the PLE-like gene FBP6 demonstrate that the two genes redundantly control stamen and carpel identity and floral determinacy [21]. In tomato, different RNAi lines against the euAG-like gene TOMATO AGAMOUS1 (TAG1) and the PLE-like gene ARLEQUIN/ TOMATO AGAMOUS-LIKE1 (TAGL1) generated slightly conflicting results, likely due to different degrees of gene down-regulation and possible co-silencing of paralogous genes [50, 51]. Still, these studies suggest a partially redundant function of TAG1 and TAGL1 in the control of the C-function [50, 51]. Consistently, the beautiful arlequin tomato (see Fig. 1g), a semi-dominant mutant of TAGL1, forms carpelloid sepals appearing fleshy and bright red when the fruit is mature [2]. Both genes also appear to play a role in fruit development and ripening [50, 51]. In tobacco, VIGS lines suggest that the euAG- and PLE-like genes NbAG and NbSHP redundantly control the C-function, while NbSHP has an additional role in fruit dehiscence [49]. Similarly, in Physalis, downregulation of the euAG- and PLE-like genes PFAG1 and PFAG2 by VIGS revealed their partially redundant role in regulating the C-function [52]. Although work in tomato, tobacco, and Physalis is not fully clear since gene function was assessed by downregulation and not by complete knock-out, overall, it appears that in Solanaceae control of the C-function is largely shared between euAG- and PLE-like genes, and that PLE-like genes can have an additional role in fruit development or ripening (similarly to the SHP genes in Arabidopsis). Together with previous studies in Arabidopsis and Antirrhinum, this supports an evolutionary model with several independent subfunctionalization events between AG-family members for the control of the C-function [48, 49]. B-class genes have also been reported to participate in the determination of carpel identity in Solanaceae: as discussed previously, in petunia the paleoAP3-type gene PhTM6 is expressed in the carpel [17], suggesting it plays a role in its establishment, although this has not been formally demonstrated yet. In Physalis, it was recently shown that the GLO-like gene DOLL1 is expressed in the carpel whorl where it activates the expression of the ortholog of the carpel regulator CRABS CLAW (CRC), which ensures proper carpel development and fertility [53, 54]. C-class genes can have other roles in addition to the C-function: these genes also trigger nectary development [55]. Interestingly, this newly discovered function is true both in petunia where nectaries are found at the base of the ovary, and in Arabidopsis where nectaries are found at the base of the stamens. In both species, the AG-family genes (i.e., AG and SHP1/2 in Arabidopsis, and FBP6 and PMADS3 in petunia) redundantly activate expression of the YABBY transcription factor CRC (or its orthologs

48

Marie Monniaux and Michiel Vandenbussche

PhCRC1/2 in petunia), that are essential for nectary development [54, 55]. This work revealed that the same AG-CRC genetic module is involved in nectary development in two distantly related species with different nectary positioning, suggesting the possibility of an ancestral mechanism for nectary specification before the asterids-rosids divergence [55]. Finally, the petunia C-class genes FBP6 and PMADS3 also participate to the D-function, specifying ovule identity [21]. This function was first identified in petunia, where co-silencing of the D-class genes FBP7 and FBP11 converted ovules into carpelloid organs [56]. However, full knock-out lines revealed later that both genes were dispensable for ovule identity (suggesting that other genes had been silenced in the cosuppression lines), but that combining the fbp7 fbp11 mutations with additional mutation or silencing in FBP6 or PMADS3 (therefore creating a partial-C/full-D class mutant) leads to a strong loss of ovule identity [21]. Indeed, this is in accordance with what is observed in Arabidopsis, where the C- and D-class genes AG, SHP1/2 and SEEDSTICK also participate in defining ovule identity [57]. In tomato, the D-class genes Sl-AGL11 and Sl-MBP3 are expressed both in the carpel, the seeds and the fruit, and overexpression of Sl-AGL11 leads to early fruit ripening together with a dramatic conversion of sepals into fleshy fruit tissue [58]. This suggests that at least one of the D-class genes in tomato has maintained the capacity to specify carpel identity when ectopically expressed, and may have also developed an additional role during fruit development. 2.4 Uncovering the E-Function

The Solanaceae flowers brought the first evidence for the existence of an extra floral function in addition to the (A)BC(D) functions. Indeed, although clearly not gene-specific, co-suppression lines of the petunia FBP2 or the tomato TM5 genes (both MADS-box genes orthologous to Arabidopsis SEP3) exhibited flowers with homeotic conversion of petals, stamens, and carpels into sepaloid organs, together with floral indeterminacy [59, 60]. At that time, this was interpreted as the involvement of FBP2 and TM5 in floral meristem identity or in the repression of sepaloid identity [59, 60]. Both interpretations still stand true today, although we tend not to view sepaloid identity as being repressed but rather that petal, stamen, and carpel identities are activated on a sepaloid background, and that this activation of the B- and C-functions requires genes from the E-function, such as petunia FBP2, tomato TM5 or Arabidopsis SEP3. Indeed, it was later found in Arabidopsis that petals, stamens, and carpels in the sep1 sep2 sep3 triple mutant were transformed into sepals [61], while all floral organs in a sep1 sep2 sep3 sep4 mutant developed as leaf-like organs [62], leading to the idea that the SEP genes are required for the identity of all floral organs in a largely redundant fashion, thereby incarnating the E-function. Molecular support for this additional function came

Molecular Mechanisms of Flower Development in the Nightshades (Solanaceae). . .

49

from the identification of “floral quartets,” where E-class proteins bridge other MADS-box proteins together in a floral organ specific manner, providing a physical explanation for the implication of SEP proteins in the identity of all floral organs [15, 63, 64]. In addition, work in petunia and Arabidopsis further showed that ovule identity also requires SEP activity, as illustrated by the homeotic conversions of all ovules into leaf-like organs in the petunia fbp2 fbp5 mutant [65] and of part of the ovules into leaf-like or carpel-like organs in the SEP1/sep1 sep2 sep3 mutant [66]. Furthermore, it was later shown that Petunia AGL6, a member of the MADS-box AGL6 subfamily closely related to the SEP subfamily (see Fig. 3), also performs SEP-like functions, redundantly with some of the petunia SEP genes [67], adding further genetic complexity to the SEP function in these Solanaceous species. More recently, a large genetic study was performed aimed at revealing all individual and redundant functions of Petunia AGL6 and its six SEP-like genes, based on the analysis of single and higher order mutants [35]. This study revealed that the petunia SEP1/2/3 orthologs (see Fig. 3) together with AGL6 encode the classical SEP floral organ identity and floral termination functions, with a master role for the petunia SEP3 ortholog FBP2. Remarkably however, it was found that the FBP9 subclade members FBP9 and FBP23, for which no clear ortholog is present in Arabidopsis, play a major role in determining floral meristem identity together with FBP4, while contributing only very moderately to floral organ identity. Indeed, triple fbp4 fbp9 fbp23 mutants completely lack flowers and exhibit a highly branched inflorescence structure due to the homeotic transformation of its floral meristems into inflorescence meristems. This shows that in contrast to Arabidopsis, a subset of the Petunia SEP genes (FBP4, -9 and -23) have evolved a specific function as floral meristem identity genes rather than encoding the classical organ identity SEP function. This is remarkably similar to the earlier proposed roles for three orthologous tomato genes LIN (LONG INFLORESCENCE), J2 (JOINTLESS 2) and EJ2 (ENHANCER OF JOINTLESS 2) which are responsible for the transition from the inflorescence to the floral meristem identity [68], suggesting that this is conserved within the Solanaceae. Interestingly, mutations in J2 and EJ2 were individually selected during tomato breeding as they both caused beneficial traits (elimination of the fruit abscission zone and increased calyx size), but when combined affect inflorescence branching and fertility [68]. Finely controlling the expression levels of J2 and EJ2 by genetic engineering allows to generate tomato varieties with a better combination of beneficial traits [68, 69]. This reveals the wide and complex roles of the tomato SEP-like genes in the control of inflorescence branching, calyx size, flower abscission and flower fertility. Finally, silencing approaches suggest that the two tomato SEP1/2/3 members TM5 and TM29 and the tomato AGL6 gene exhibit classical SEP organ identity functions [60, 70, 71] similar to what we found in petunia.

50

Marie Monniaux and Michiel Vandenbussche

In Arabidopsis, the floral meristem identity function is attributed to members of the AP1/SQUA MADS-box subfamily (AP1, CAL and FUL genes in Arabidopsis) [72, 73] rather than to a specific subclass of SEP genes as described above for petunia and tomato. Interestingly, AP1, SEP and AGL6 genes all form a superclade with shared ancestry (see Fig. 3). This suggests that the floral meristem identity function might have been ancestral to the duplication event that generated those three different gene families, and has been distributed primarily to AP1-like genes in Arabidopsis and to a subclass of SEP-like genes in petunia and tomato. 2.5 Divergence to the Textbook ABC Model

3

Research in the Solanaceae, particularly in petunia, has led to the construction of a floral organ identity patterning model quite distinct to the textbook ABC model (see Fig. 2). Why so many differences? Is this a specificity of the Solanaceae? The ABC model is by essence, a simplification of the reality, and even the founder species of the model, Arabidopsis and Antirrhinum, do not fit perfectly in the frame. For instance, no complete A-function mutant was ever found in Antirrhinum, and the A-class gene AP1 is fully dispensable for petal identity in Arabidopsis [19, 20]. Therefore, it is only normal that, as we discover increased details about the molecular players of floral organ identity, the model that was proposed more than 30 years ago becomes less adequate and needs to be complexified [74]. Moreover, the number of ABC genes within a species necessarily complexifies the model, as each gene copy can evolve new functions (neofunctionalization) or retain part of the ancestral function (subfunctionalization) [75]. Therefore, the increased complexity of the model in Fig. 3, as compared to the textbook ABC model, is due to the large number of (A)BC genes in petunia, the high power of functional analysis that can be performed in this species and the large amount of work that was carried on the topic over the years. This only advocates for the need of many more model species where such a fine analysis can be performed to find more general principles and specificities of floral organ identity patterning.

Initiation and Fusion of Floral Organs While floral organ identity is continuously specified by homeotic genes throughout organ development, other molecular players direct the pattern of floral organ initiation and the hormone auxin appears as a key player in this process ([76], and references therein). The synthetic promoter DR5, containing auxin-responsive elements and fused to a reporter gene such as GFP, is often used as a late reporter for auxin signaling [77]. During floral development, DR5:GFP expression peaks successively at incipient floral organ primordia, first marking the sepals initiation domain, followed by petals, stamens, and carpels [78, 79].

Molecular Mechanisms of Flower Development in the Nightshades (Solanaceae). . .

51

So far, Solanaceae have brought relatively little contribution to the understanding of floral organ initiation and growth. Yet, some specific features from the family could strongly enrich the field of research. For instance, flowers have a type-5 symmetry, with five sepals, five petals and five stamens initiating successively, and almost jointly within each whorl, in contrast to the type-4 symmetry of Arabidopsis flowers. Models, based on periodic auxin accumulation and increasing space in the floral meristem, generate self-organizing patterns of primordia initiation [80, 81]. Researchers have attempted to explain the emergence of different floral organ numbers per whorl [82], but these models would strongly benefit from additional data in species with a type-5 symmetry, such as the detailed spatio-temporal pattern of DR5 expression that was characterized during tomato flower and fruit development [83]. Petal fusion (sympetaly) is a trait of major evolutionary importance, that led to new floral structures and possibly accelerated speciation rates. Some molecular players involved in petal fusion have been found in Arabidopsis, based on mutants with fused petals [84]. Using a model species with fused petals, such as, e.g., petunia, should allow to identify other actors of petal fusion, particularly those involved in the evolution of the trait. The petunia mutants maewest (maw) and choripetala suzanne (chsu) were isolated in a genetic screen for petal fusion defects [85]. These mutants form partly fused and narrow petals, together with narrow leaves, suggesting general defects in organ laminar lateral growth. Petal fusion is congenital in petunia (i.e., petals are fused by the confluence of their individual primordia), and the defects observed in maw or chsu appear to be due to defects in petal primordia lateral expansion, suggesting that fusion and lateral growth are inherently coupled during petunia petal development. MAW is a WUS-like homeobox (WOX) gene, to which the famous gene WUSCHEL belongs, that controls stem cell maintenance in the meristem [86, 87]. In tomato flowers, whose petals are partly fused at the base, mutation in the MAW ortholog UN-FUSED FLOWER (UF) also causes narrow and unfused petals, and narrow leaves [88]. MAW and UF are homologous to Arabidopsis WOX1, whose mutation causes no obvious phenotypic defects in the flower [85]. However, combining mutations in wox1 and in another WOX gene, PRESSED FLOWER (PRS), results in plants with very narrow leaves and petals, reminiscent of the petunia maw phenotype [85, 89]. This indicates that petal lateral growth (and fusion in petunia) is controlled by similar classes of genes from the WOX subfamily in Arabidopsis and petunia, but different members have been recruited to fulfill this function in the two species [85, 86]. Although Arabidopsis WOX1 and PRS genes appear to fulfill together the same function as MAW in petunia, Arabidopsis petals are not fused, suggesting that other differences in the regulatory network caused divergence in this trait during evolution.

52

Marie Monniaux and Michiel Vandenbussche

Sympetaly likely has a single origin in asterids [84], hence identifying the key event leading to the evolution of this trait might require comparative work involving a sister group to the asterids, such as Caryophyllaceae family where most species form flowers with unfused petals.

4

Growth and Maturation of Floral Organs: Chinese Lanterns and Petal Colors The elegant Chinese lanterns from Physalis and Whitania form an encapsulated fruit after pollination, named the inflated calyx syndrome (ICS), due to late sepal growth (see Fig. 1h). Indeed, sepals of Chinese lanterns resume growth after pollination, and this trait was shown to be an advantageous morphological novelty, because the inflated calyx has photosynthetic capability and provides a micro-environment improving fruit maturation [90]. It had been proposed that the MADS-box gene MPF2, orthologous to Arabidopsis AGAMOUS-LIKE 24, and genes from the MPF2-like-A family were involved in ICS: down-regulating MPF2 expression by RNA interference in Physalis resulted in small sepals, and ectopic expression of MPF2 in tomato enlarged its calyx by increasing cell division rates [91, 92]. In tomato, the MPF2 ortholog STMADS16 is only expressed in leaves, while in Physalis MPF2 is expressed both in leaves and sepals, suggesting that during evolution MPF2 was recruited in the calyx, by heterotopic expression, to form the Chinese lantern phenotype [91]. However, these results were recently reevaluated, because knocking-out MPF2 by CRISPR was not sufficient to disrupt the ICS, and neither was the individual knock-out of ten other MADS-box genes from the AP1, SEP4, AG or DEF/ GLO clades [93]. RNA interference notoriously silences both onand off-targets with partial efficiencies, which probably explains the inconsistency of phenotypes observed between the RNA interference and CRISPR lines. This suggests that multiple MADS-box genes may play a role in the formation of the inflated calyx, but the exact genetic change that caused the emergence of this novelty has not been pinpointed yet [94]. Mature floral organs acquire a whole set of tissue and cell properties, and in particular petals display a specific pigmentation, crucial for their interaction with pollinators. Since petunia petals are big and showy, and because insertion mutants frequently arise in the line with an active transposase, a mutagenesis screen for the production of anthocyanins (the main pigments accumulating in petunia petals) entails a beautiful and easy phenotyping process. Consequently, petunia has been instrumental in identifying the molecular players of flower color, with the majority of enzymes and regulators involved in anthocyanin production cloned and characterized in this species [5]. Production of anthocyanins first relies on the production of flavonols, later modified into

Molecular Mechanisms of Flower Development in the Nightshades (Solanaceae). . .

53

anthocyanins through the action of specific enzymes and regulators [5]. Interestingly, the mutant phenotypes of these different regulators are distinct: mutant in the gene AN1(ANTHOCYANIN1) loses pigmentation in all tissues, while the an2 and an8 mutations only affect pigmentation of the limb (the upper part of the corolla) and the tube respectively [5, 95, 96]. This suggests that the wildtype petal pigmentation is the result of the global and local action of a combination of regulatory genes. The action mechanism of the regulator AN1 has been elucidated in more detail: additional to the regulation of anthocyanin biosynthesis, AN1 regulates the pH of the vacuole of petal epidermal cells, which directly affects the absorption spectra of anthocyanins and hence the resulting petal color [95]. Particularly, AN1 regulates the expression of PH1 and PH5, two vacuolar P-ATPases that pump protons into the vacuole and therefore acidify it [97, 98]. Surprisingly, it was recently found that fruit acidity in lemon and oranges is caused by the same mechanism: the genes CitPH1 and CitPH5 acidify the vacuoles in the juice vesicle cells of the fruit and can thereby generate a pH as low as 2 [99]. The regulator of anthocyanin production AN2 was also found to be involved in the evolution of petal color between wild petunia species. Indeed, in the white-colored flowers of Petunia axillaris, five independent losses of function in AN2, some of them by frame-shift mutations, were found in the wild, affecting flower color and pollinator preferences [100]. Moreover, the surprising “resurrection” of the AN2 gene after pseudogenization, by a secondary mutation that restored the original reading frame of the gene, has been reported in the purple-colored petals of P. secreta [101]. The study of petal pigmentation has proven to be a rich field of research in the Solanaceae, bridging gaps between the subcellular and the microevolutionary scale.

5

Conclusion Flower development is a massive field of research that can be apprehended under different angles, at different scales and in different species. For the last 30 years, research on Arabidopsis has led to the characterization of most key concepts and molecular players of flower development. However, each species has its specificities and Arabidopsis is no exception; therefore, some of the mechanisms uncovered in Arabidopsis are highly divergent in comparison to the common ancestor of eudicots. Moreover, due to the random nature of gene duplication and subfunctionalization in different species, gene functions can appear hidden in redundancy in Arabidopsis but be revealed by gene subfunctionalization in other species (or vice versa). Finally, some important features of flower development simply do not exist in Arabidopsis, for instance petal fusion or petal pigmentation. For these reasons, including several model

54

Marie Monniaux and Michiel Vandenbussche

species in the research on flower development is now crucial to further improve our understanding of this process [8]. Solanaceae have brought their share of results on this topic and surely this family will continue bringing new insights on flower development, also with potential application for agronomical purposes. References 1. The Plant List (2013) http://www. theplantlist.org/ 2. Gime´nez E, Pineda B, Capel J et al (2010) Functional analysis of the Arlequin mutant corroborates the essential role of the Arlequin/TAGL1 gene during reproductive development of tomato. PLoS One 5: e14427. https://doi.org/10.1371/journal. pone.0014427 3. Gebhardt C (2016) The historical role of species from the Solanaceae plant family in genetic research. Theor Appl Genet 129: 2281–2294. https://doi.org/10.1007/ s00122-016-2804-1 4. Lin T, Zhu G, Zhang J et al (2014) Genomic analyses provide insights into the history of tomato breeding. Nat Genet 46:1220–1226. https://doi.org/10.1038/ng.3117 5. Tornielli G, Koes R, Quattrocchio F (2009) The genetics of flower color. In: Petunia: evolutionary, developmental and physiological genetics. Springer International Publishing, pp 269–299 6. Galliot C, Stuurman J, Kuhlemeier C (2006) The genetic dissection of floral pollination syndromes. Curr Opin Plant Biol 9:78–82. https://doi.org/10.1016/j.pbi.2005. 11.003 7. Vandenbussche M, Janssen A, Zethof J et al (2008) Generation of a 3D indexed Petunia insertion database for reverse genetics. Plant J 54:1105–1114. https://doi.org/10.1111/j. 1365-313X.2008.03482.x 8. Vandenbussche M, Chambrier P, Rodrigues Bento S, Morel P (2016) Petunia, your next supermodel? Front Plant Sci 7:72. https:// doi.org/10.3389/fpls.2016.00072 9. Amasino RM (2013) My favourite flowering image: Maryland Mammoth tobacco. J Exp Bot 64:5817–5818. https://doi.org/10. 1093/jxb/ert083 10. Parcy F (2019) L’histoire secre`te des fleurs. Humensciences 11. Abelenda JA, Navarro C, Prat S (2014) Flowering and tuberization: a tale of two nightshades. Trends Plant Sci 19:115–122. https://doi.org/10.1016/j.tplants.2013. 09.010

12. Kimura S, Sinha N (2008) Tomato (Solanum lycopersicum): a model fruit-bearing crop. CSH Protoc 2008:pdb.emo105. https://doi. org/10.1101/pdb.emo105 13. Schwarz-Sommer Z, Huijser P, Nacken W et al (1990) Genetic control of flower development by homeotic genes in Antirrhinum majus. Science 250:931–936. https://doi. org/10.1126/science.250.4983.931 14. Coen ES, Meyerowitz EM (1991) The war of the whorls: genetic interactions controlling flower development. Nature 353:31–37. https://doi.org/10.1038/353031a0 15. Theissen G, Saedler H (2001) Plant biology. Floral quartets. Nature 409:469–471. https://doi.org/10.1038/35054172 16. Vandenbussche M, Zethof J, Royaert S et al (2004) The duplicated B-class heterodimer model: whorl-specific effects and complex genetic interactions in Petunia hybrida flower development. Plant Cell 16:741–754. https://doi.org/10.1105/tpc.019166 17. Rijpkema AS, Royaert S, Zethof J et al (2006) Analysis of the Petunia TM6 MADS box gene reveals functional divergence within the DEF/AP3 lineage. Plant Cell 18:1819– 1832. https://doi.org/10.1105/tpc.106. 042937 18. Van de Peer Y, De Wachter R (1994) TREECON for Windows: a software package for the construction and drawing of evolutionary trees for the Microsoft Windows environment. Comput Appl Biosci 10:569–570 19. Litt A (2007) An evaluation of A-function: evidence from the APETALA1 and APETALA2 gene lineages. Int J Plant Sci 168: 73–91. https://doi.org/10.1086/509662 20. Causier B, Schwarz-Sommer Z, Davies B (2010) Floral organ identity: 20 years of ABCs. Semin Cell Dev Biol 21:73–79. https://doi.org/10.1016/j.semcdb.2009. 10.005 21. Heijmans K, Ament K, Rijpkema AS et al (2012) Redefining C and D in the petunia ABC. Plant Cell 24:2305–2317. https://doi. org/10.1105/tpc.112.097030 22. Irish VF, Sussex IM (1990) Function of the apetala-1 gene during Arabidopsis floral

Molecular Mechanisms of Flower Development in the Nightshades (Solanaceae). . . development. Plant Cell 2:741–753. https:// doi.org/10.1105/tpc.2.8.741 23. Mandel MA, Gustafson-Brown C, Savidge B, Yanofsky MF (1992) Molecular characterization of the Arabidopsis floral homeotic gene APETALA1. Nature 360:273–277. https:// doi.org/10.1038/360273a0 24. Kunst L, Klenz JE, Martinez-Zapater J, Haughn GW (1989) AP2 gene determines the identity of perianth organs in flowers of Arabidopsis thaliana. Plant Cell 1:1195– 1208. https://doi.org/10.1105/tpc.1.12. 1195 25. Bowman JL, Smyth DR, Meyerowitz EM (1991) Genetic interactions among floral homeotic genes of Arabidopsis. Development 112:1–20 26. Bradley D, Carpenter R, Sommer H et al (1993) Complementary floral homeotic phenotypes result from opposite orientations of a transposon at the plena locus of Antirrhinum. Cell 72:85–95 27. Monniaux M, Vandenbussche M (2018) How to evolve a perianth: a review of cadastral mechanisms for perianth identity. Front Plant Sci 9:1573. https://doi.org/10.3389/ fpls.2018.01573 28. Baum DA, Hileman LC (2006) A developmental genetic model for the origin of the flower. In: Annual plant reviews volume 20: flowering and its manipulation. Wiley, pp 1–27 29. Morel P, Heijmans K, Rozier F et al (2017) Divergence of the floral A-function between an asterid and a rosid species. Plant Cell 29: 1605–1621. https://doi.org/10.1105/tpc. 17.00098 30. Cartolano M, Castillo R, Efremova N et al (2007) A conserved microRNA module exerts homeotic control over Petunia hybrida and Antirrhinum majus floral organ identity. Nat Genet 39:901–905. https://doi.org/10. 1038/ng2056 31. Aukerman MJ, Sakai H (2003) Regulation of flowering time and floral organ identity by a microRNA and its APETALA2-like target genes. Plant Cell 15:2730–2741. https:// doi.org/10.1105/tpc.016238 32. Jung J-H, Seo Y-H, Seo PJ et al (2007) The GIGANTEA-regulated microRNA172 mediates photoperiodic flowering independent of CONSTANS in Arabidopsis. Plant Cell 19: 2736–2748. https://doi.org/10.1105/tpc. 107.054528 33. Mathieu J, Yant LJ, Mu¨rdter F et al (2009) Repression of flowering by the miR172 target

55

SMZ. PLoS Biol 7:e1000148. https://doi. org/10.1371/journal.pbio.1000148 34. Yant L, Mathieu J, Dinh TT et al (2010) Orchestration of the floral transition and floral development in Arabidopsis by the bifunctional transcription factor APETALA2. Plant Cell 22:2156–2170. https://doi.org/10. 1105/tpc.110.075606 35. Morel P, Chambrier P, Boltz V et al (2019) Divergent functional diversification patterns in the SEP/AGL6/AP1 MADS-box transcription factor superclade. Plant Cell 31: 3033–3056. https://doi.org/10.1105/tpc. 19.00162 36. Krogan NT, Hogan K, Long JA (2012) APETALA2 negatively regulates multiple floral organ identity genes in Arabidopsis by recruiting the co-repressor TOPLESS and the histone deacetylase HDA19. Development 139: 4180–4190. https://doi.org/10.1242/dev. 085407 37. Vrebalov J, Ruezinsky D, Padmanabhan V et al (2002) A MADS-box gene necessary for fruit ripening at the tomato ripening-inhibitor (rin) locus. Science 296:343–346. https:// doi.org/10.1126/science.1068181 38. Yuste-Lisbona FJ, Quinet M, Ferna´ndezLozano A et al (2016) Characterization of vegetative inflorescence (mc-vin) mutant provides new insight into the role of MACROCALYX in regulating inflorescence development of tomato. Sci Rep 6:18796. https://doi.org/10.1038/srep18796 39. Quinet M, Bataille G, Dobrev PI et al (2014) Transcriptional and hormonal regulation of petal and stamen development by STAMENLESS, the tomato (Solanum lycopersicum L.) orthologue to the B-class APETALA3 gene. J Exp Bot 65:2243–2256. https://doi.org/10. 1093/jxb/eru089 40. de Martino G, Pan I, Emmanuel E et al (2006) Functional analyses of two tomato APETALA3 genes demonstrate diversification in their roles in regulating floral development. Plant Cell 18:1833–1845. https://doi.org/ 10.1105/tpc.106.042978 41. van der Krol AR, Brunelle A, Tsuchimoto S, Chua NH (1993) Functional analysis of petunia floral homeotic MADS box gene pMADS1. Genes Dev 7:1214–1228 42. Vandenbussche M, Theissen G, Van de Peer Y, Gerats T (2003) Structural diversification and neo-functionalization during floral MADS-box gene evolution by C-terminal frameshift mutations. Nucleic Acids Res 31: 4401–4409. https://doi.org/10.1093/nar/ gkg642

56

Marie Monniaux and Michiel Vandenbussche

43. Schwarz-Sommer Z, Hue I, Huijser P et al (1992) Characterization of the Antirrhinum floral homeotic MADS-box gene deficiens: evidence for DNA binding and autoregulation of its persistent expression throughout flower development. EMBO J 11:251–263 44. McGonigle B, Bouhidel K, Irish VF (1996) Nuclear localization of the Arabidopsis APETALA3 and PISTILLATA homeotic gene products depends on their simultaneous expression. Genes Dev 10:1812–1821. https://doi.org/10.1101/gad.10.14.1812 45. Tro¨bner W, Ramirez L, Motte P et al (1992) GLOBOSA: a homeotic gene which interacts with DEFICIENS in the control of Antirrhinum floral organogenesis. EMBO J 11:4693– 4704 46. Goto K, Meyerowitz EM (1994) Function and regulation of the Arabidopsis floral homeotic gene PISTILLATA. Genes Dev 8:1548– 1560. https://doi.org/10.1101/gad.8.13. 1548 47. Liljegren SJ, Ditta GS, Eshed Y et al (2000) SHATTERPROOF MADS-box genes control seed dispersal in Arabidopsis. Nature 404:766–770. https://doi.org/10.1038/ 35008089 48. Causier B, Castillo R, Zhou J et al (2005) Evolution in action: following function in duplicated floral homeotic genes. Curr Biol 15:1508–1512. https://doi.org/10.1016/j. cub.2005.07.063 49. Fourquin C, Ferra´ndiz C (2012) Functional analyses of AGAMOUS family members in Nicotiana benthamiana clarify the evolution of early and late roles of C-function genes in eudicots. Plant J 71:990–1001. https://doi. org/10.1111/j.1365-313X.2012.05046.x 50. Pan IL, McQuinn R, Giovannoni JJ, Irish VF (2010) Functional diversification of AGAMOUS lineage genes in regulating tomato flower and fruit development. J Exp Bot 61: 1795–1806. https://doi.org/10.1093/jxb/ erq046 ˜ eda L, Pineda B et al 51. Gimenez E, Castan (2016) TOMATO AGAMOUS1 and ARLEQUIN/TOMATO AGAMOUS-LIKE1 MADS-box genes have redundant and divergent functions required for tomato reproductive development. Plant Mol Biol 91:513– 531. https://doi.org/10.1007/s11103016-0485-4 52. Zhao J, Gong P, Liu H et al (2021) Multiple and integrated functions of floral C-class MADS-box genes in flower and fruit development of Physalis floridana. Plant Mol Biol 107:101–116. https://doi.org/10.1007/ s11103-021-01182-4

53. Gong P, Song C, Liu H et al (2021) Physalis floridana CRABS CLAW mediates neofunctionalization of GLOBOSA genes in carpel development. J Exp Bot 72:6882–6903. https://doi.org/10.1093/jxb/erab309 54. Bowman JL, Smyth DR (1999) CRABS CLAW, a gene that regulates carpel and nectary development in Arabidopsis, encodes a novel protein with zinc finger and helixloop-helix domains. Development 126: 2387–2396 55. Morel P, Heijmans K, Ament K et al (2018) The floral C-lineage genes trigger nectary development in Petunia and Arabidopsis. Plant Cell 30:2020–2037. https://doi.org/ 10.1105/tpc.18.00425 56. Angenent GC, Franken J, Busscher M et al (1995) A novel class of MADS box genes is involved in ovule development in petunia. Plant Cell 7:1569–1582. https://doi.org/ 10.1105/tpc.7.10.1569 57. Pinyopich A, Ditta GS, Savidge B et al (2003) Assessing the redundancy of MADS-box genes during carpel and ovule development. Nature 424:85–88. https://doi.org/10. 1038/nature01741 58. Huang B, Routaboul J-M, Liu M et al (2017) Overexpression of the class D MADS-box gene Sl-AGL11 impacts fleshy tissue differentiation and structure in tomato fruits. J Exp Bot 68:4869–4884. https://doi.org/10. 1093/jxb/erx303 59. Angenent GC, Franken J, Busscher M et al (1994) Co-suppression of the petunia homeotic gene fbp2 affects the identity of the generative meristem. Plant J 5:33–44 60. Pnueli L, Hareven D, Broday L et al (1994) The TM5 MADS box gene mediates organ differentiation in the three inner whorls of tomato flowers. Plant Cell 6:175–186. https://doi.org/10.1105/tpc.6.2.175 61. Pelaz S, Ditta GS, Baumann E et al (2000) B and C floral organ identity functions require SEPALLATA MADS-box genes. Nature 405: 200–203. https://doi.org/10.1038/ 35012103 62. Ditta G, Pinyopich A, Robles P et al (2004) The SEP4 gene of Arabidopsis thaliana functions in floral organ and meristem identity. Curr Biol 14:1935–1940. https://doi.org/ 10.1016/j.cub.2004.10.028 63. Melzer R, Theissen G (2009) Reconstitution of “floral quartets” in vitro involving class B and class E floral homeotic proteins. Nucleic Acids Res 37:2723–2736. https://doi.org/ 10.1093/nar/gkp129

Molecular Mechanisms of Flower Development in the Nightshades (Solanaceae). . . 64. Immink RGH, Tonaco IAN, de Folter S et al (2009) SEPALLATA3: the “glue” for MADS box transcription factor complex formation. Genome Biol 10:R24. https://doi.org/10. 1186/gb-2009-10-2-r24 65. Vandenbussche M, Zethof J, Souer E et al (2003) Toward the analysis of the petunia MADS box gene family by reverse and forward transposon insertion mutagenesis approaches: B, C, and D floral organ identity functions require SEPALLATA-like MADS box genes in petunia. Plant Cell 15:2680– 2693. https://doi.org/10.1105/tpc.017376 66. Favaro R, Pinyopich A, Battaglia R et al (2003) MADS-box protein complexes control carpel and ovule development in Arabidopsis. Plant Cell 15:2603–2611. https:// doi.org/10.1105/tpc.015123 67. Rijpkema AS, Zethof J, Gerats T, Vandenbussche M (2009) The petunia AGL6 gene has a SEPALLATA-like function in floral patterning. Plant J 60:1–9. https://doi.org/10. 1111/j.1365-313X.2009.03917.x 68. Soyk S, Lemmon ZH, Oved M et al (2017) Bypassing negative epistasis on yield in tomato imposed by a domestication gene. Cell 169:1142–1155.e12. https://doi.org/ 10.1016/j.cell.2017.04.032 69. Soyk S, Lemmon ZH, Sedlazeck FJ et al (2019) Duplication of a domestication locus neutralized a cryptic variant that caused a breeding barrier in tomato. Nat Plants 5: 471. https://doi.org/10.1038/s41477019-0422-z 70. Ampomah-Dwamena C, Morris BA, Sutherland P et al (2002) Down-regulation of TM29, a tomato SEPALLATA homolog, causes parthenocarpic fruit development and floral reversion. Plant Physiol 130(2): 605–617 71. Yu X, Chen G, Guo X et al (2017) Silencing SlAGL6, a tomato AGAMOUS-LIKE6 lineage gene, generates fused sepal and green petal. Plant Cell Rep 36(6):959–969 72. Bowman JL, Alvarez J, Weigel D et al (1993) Control of flower development in Arabidopsis thaliana by APETALA1 and interacting genes. Development 119:721–743. https://doi. org/10.1242/dev.119.3.721 73. Kempin SA, Savidge B, Yanofsky MF (1995) Molecular basis of the cauliflower phenotype in Arabidopsis. Science 267:522–525 74. Bowman JL, Smyth DR, Meyerowitz EM (2012) The ABC model of flower development: then and now. Development 139: 4095–4098. https://doi.org/10.1242/dev. 083972

57

75. Zhang R, Guo C, Zhang W et al (2013) Disruption of the petal identity gene APETALA3-3 is highly correlated with loss of petals within the buttercup family (Ranunculaceae). Proc Natl Acad Sci U S A 110: 5074–5079. https://doi.org/10.1073/pnas. 1219690110 76. Smyth DR (2018) Evolution and genetic control of the floral ground plan. New Phytol 220:70–86. https://doi.org/10.1111/nph. 15282 77. Ulmasov T, Murfett J, Hagen G, Guilfoyle TJ (1997) Aux/IAA proteins repress expression of reporter genes containing natural and highly active synthetic auxin response elements. Plant Cell 9:1963–1971. https://doi. org/10.1105/tpc.9.11.1963 78. Aloni R, Aloni E, Langhans M, Ullrich CI (2006) Role of auxin in regulating Arabidopsis flower development. Planta 223:315–328. https://doi.org/10.1007/s00425-0050088-9 79. Lampugnani ER, Kilinc A, Smyth DR (2013) Auxin controls petal initiation in Arabidopsis. Development 140:185–194. https://doi. org/10.1242/dev.084582 80. Smith RS, Guyomarc’h S, Mandel T et al (2006) A plausible model of phyllotaxis. Proc Natl Acad Sci U S A 103:1301–1306. h t t p s : // d o i . o r g / 1 0 . 1 0 7 3 / p n a s . 0510457103 81. Jo¨nsson H, Heisler MG, Shapiro BE et al (2006) An auxin-driven polarized transport model for phyllotaxis. Proc Natl Acad Sci U S A 103:1633–1638. https://doi.org/10. 1073/pnas.0509839103 82. Kitazawa MS, Fujimoto K (2015) A dynamical phyllotaxis model to determine floral organ number. PLoS Comput Biol 11: e1004145. https://doi.org/10.1371/jour nal.pcbi.1004145 83. Goldental-Cohen S, Israeli A, Ori N, Yasuor H (2017) Auxin response dynamics during wild-type and entire flower development in tomato. Plant Cell Physiol 58:1661–1672. https://doi.org/10.1093/pcp/pcx102 84. Zhong J, Preston JC (2015) Bridging the gaps: evolution and development of perianth fusion. New Phytol 208:330–335. https://doi.org/10.1111/nph.13517 85. Vandenbussche M, Horstman A, Zethof J et al (2009) Differential recruitment of WOX transcription factors for lateral development and organ fusion in Petunia and Arabidopsis. Plant Cell 21:2269–2283. https://doi.org/ 10.1105/tpc.109.065862

58

Marie Monniaux and Michiel Vandenbussche

86. Costanzo E, Trehin C, Vandenbussche M (2014) The role of WOX genes in flower development. Ann Bot 114:1545–1553. https://doi.org/10.1093/aob/mcu123 87. Laux T, Mayer KF, Berger J, Ju¨rgens G (1996) The WUSCHEL gene is required for shoot and floral meristem integrity in Arabidopsis. Development 122:87–96 88. Zhang C, Wang J, Wang X et al (2020) UF, a WOX gene, regulates a novel phenotype of un-fused flower in tomato. Plant Sci 297: 110523. https://doi.org/10.1016/j.pla ntsci.2020.110523 89. Nakata M, Matsumoto N, Tsugeki R et al (2012) Roles of the middle domain-specific WUSCHEL-RELATED HOMEOBOX genes in early development of leaves in Arabidopsis. Plant Cell 24:519–535. https://doi. org/10.1105/tpc.111.092858 90. Li J, Song C, He C (2019) Chinese lantern in Physalis is an advantageous morphological novelty and improves plant fitness. Sci Rep 9: 596. https://doi.org/10.1038/s41598-01836436-7 91. He C, Saedler H (2005) Heterotopic expression of MPF2 is the key to the evolution of the Chinese lantern of Physalis, a morphological novelty in Solanaceae. Proc Natl Acad Sci U S A 102:5779–5784. https://doi.org/10. 1073/pnas.0501877102 92. Khan MR, Hu J-Y, Riss S et al (2009) MPF2like-a MADS-box genes control the inflated Calyx syndrome in Withania (Solanaceae): roles of Darwinian selection. Mol Biol Evol 26:2463–2473. https://doi.org/10.1093/ molbev/msp159 93. He J, Alonge M, Ramakrishnan S et al (2023) Establishing Physalis as a Solanaceae model system enables genetic reevaluation of the inflated calyx syndrome. Plant Cell 35:351– 368. https://doi.org/10.1093/plcell/ koac305

94. He C, Saedler H (2007) Hormonal control of the inflated calyx syndrome, a morphological novelty, in Physalis. Plant J 49:935–946. https://doi.org/10.1111/j.1365-313X. 2006.03008.x 95. Spelt C, Quattrocchio F, Mol JN, Koes R (2000) anthocyanin1 of petunia encodes a basic helix-loop-helix protein that directly activates transcription of structural anthocyanin genes. Plant Cell 12:1619–1632. https:// doi.org/10.1105/tpc.12.9.1619 96. Quattrocchio F, Wing J, van der Woude K et al (1999) Molecular analysis of the anthocyanin2 gene of petunia and its role in the evolution of flower color. Plant Cell 11: 1433–1444. https://doi.org/10.1105/tpc. 11.8.1433 97. Verweij W, Spelt C, Di Sansebastiano G-P et al (2008) An H+ P-ATPase on the tonoplast determines vacuolar pH and flower colour. Nat Cell Biol 10:1456–1462. https://doi. org/10.1038/ncb1805 98. Faraco M, Spelt C, Bliek M et al (2014) Hyperacidification of vacuoles by the combined action of two different P-ATPases in the tonoplast determines flower color. Cell Rep 6:32–43. https://doi.org/10.1016/j. celrep.2013.12.009 99. Strazzer P, Spelt CE, Li S et al (2019) Hyperacidification of Citrus fruits by a vacuolar proton-pumping P-ATPase complex. Nat Commun 10:744. https://doi.org/10. 1038/s41467-019-08516-3 100. Hoballah ME, Gubitz T, Stuurman J et al (2007) Single gene-mediated shift in pollinator attraction in Petunia. Plant Cell 19:779– 790. https://doi.org/10.1105/tpc.106. 048694 101. Esfeld K, Berardi AE, Moser M et al (2018) Pseudogenization and resurrection of a speciation gene. Curr Biol 28:3776–3786.e7. https://doi.org/10.1016/j.cub.2018. 10.019

Chapter 3 The ABC of Flower Development in Monocots: The Model of Rice Spikelet Ludovico Dreni Abstract The initial seminal studies of flower developmental genetics were made from observations in several eudicot model species, particularly Arabidopsis and Antirrhinum. However, an increasing amount of research in monocot model and crop species is finally giving the credit that monocots deserve for their position in the evolutionary history of Angiosperms, their astonishing diversification and adaptation, their diversified floral structures, their pivotal function in most ecosystems on Earth and, finally, their importance in agriculture and farming, economy, landscaping and feeding mankind. Rice is a staple crop and the major monocot model to study the reproductive phase and flower evolution. Inspired by this, this chapter reviews a story of highly conserved functions related to the ABC model of flower development. Nevertheless, this model is complicated in rice by cases of gene neofunctionalization, like the recruitment of MADS-box genes for the development of the unique organs known as lemma and palea, subfunctionalization, and rewiring of conserved molecular pathways. Key words ABC model, (A)BC model, Angiosperms, Core eudicots, Monocots, Rice (Oryza sativa)

1

Introduction In terms of Angiosperm evolution, monocots can be regarded as the counterpart of core eudicots. With over 69,000 species currently documented (Angiosperm Phylogeny Website [1]), monocots are the second largest taxon of Angiosperms, comprising about 25% of all the species. But monocots are not second but first in terms of importance for mankind, and possibly also for their degree of adaptation and diversification, including their extremely evolved and diversified floral structures. Generally, four types of floral organs arranged in four distinct whorls are found in core eudicots: two perianth whorls of sepals and petals, and two inner whorls of reproductive organs, consisting of stamens and pistils. Monocots are a monophyletic group presenting an amazing range of floral models, from very showy to dull, the number of each floral organ type being generally three or a multiple of three and including a

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_3, © Springer Science+Business Media, LLC, part of Springer Nature 2023

59

60

Ludovico Dreni

perianth consisting of two whorls of three tepals each. Often, the first and second whorl tepals are similar in appearance, although they might correspond to eudicot sepals and petals, respectively. This trimerous developmental model is ancestral and already found in the earliest diverged monocots, Acorales [2]. However, it is hard to still recognize it in the highly evolved and derived flowers of orchids and grasses, for example, which hardly resemble each other [3]. The closely related monocots Lacandonia schismatica and Triuris brevistylis offer the unique existing examples of “reversed flowers” in their natural populations, with female reproductive structures arranged in the third floral whorl and stamens in the central fourth whorl [4, 5], which is just the opposite than predicted by any flower developmental model. It is therefore impossible to define a monocot flower model to represent them all. The monocot family of grasses, or Poaceae, is one of the largest plant families with >11,000 accepted species (Grass Phylogeny Working Group II [6]), and its several domesticated cereal species are by far the major source of calory intake for mankind. Several grass species serve as models or emerging models and, above all, rice (Oryza sativa) has been extensively used to study floral development and its molecular regulation. Currently, the majority of the fully sequenced monocot genomes belong to grasses, and pan-genome projects have been already developed for rice and other grasses such as Brachypodium distachyon, maize (Zea mays), and wheat (Triticum sp.) [7–12]. Among these grasses, rice does not possess the shortest life cycle nor the smaller size, yet it quickly became the main grass and monocot model plant for molecular studies in the last two decades. Indeed, rice was the second plant for which a whole genome sequence was made available [13, 14], only 2 years later than the Arabidopsis genome [15]. Of course, much of the interest on rice derives from its role as a major staple food crop for human consumption, together with two other grasses, wheat and maize (www.fao.org).

2

The Spikelet and the Flower in Grasses As indicated above, the grass flower is highly modified from the ancestral trimerous model of monocots. In addition, grass flowers group into clusters named spikelets, which can be seen as small inflorescence units which in turn, depending on the species, organize in much different types of higher order inflorescences, e.g., panicles, spikes, or racemes (see Fig. 1) [16–18]. The main axis of such inflorescences, termed rachis, is often indeterminate, with known exceptions like wheat. Since spikelets are themselves inflorescences, the grass inflorescence is a compound structure, and is therefore technically a synflorescence [18].

Flower Development in Monocots

61

Fig. 1 Schematic representations of reproductive structures of the grass family, Poaceae. Up: inflorescence types commonly encountered in grasses. Bottom left: Basic model of a grass spikelet. Bottom right: model of the modified rice spikelet producing only a terminal flower. r rudimentary glume, sl sterile lemma, l lemma, p palea, MRP marginal tissue of palea, BOP body of palea, lo lodicule

62

Ludovico Dreni

Although the term spikelet is also used to define similar inflorescence units occurring in other Poales, the spikelet of true grasses has a distinctive structure and it has been proposed to have evolved independently near the base of their lineage [19]. It may have evolved even after the most recent common ancestor of Poaceae, as true spikelets are not observed in their earliest diverged subfamily Anomochlooideae [20, 21]. Within Anomochlooideae, Streptochaeta produces so called pseudospikelets that comprise, depending on the author, 11 or 12 bracts arranged in two nodes, which have been used as a model to try to reconstruct the evolution of grass spikelet [22, 23]. Unfortunately, these works did not agree on the number of bracts found in Streptochaeta pseudospikelets, and led to the publication of many different models describing the possible origin of the grass spikelet, although one of them seems supported by the expression analysis of MADS-box marker genes [23]. A true grass spikelet comprises two leafy, papery glumes subtending a short axis termed rachilla, that bears lateral flowers and a terminal flower (see Fig. 1). Each flower, also referred as floret, starts with a bract-like organ, the lemma, followed by a similar organ at the opposite site, known as palea. Enclosed by these two glumes, two or three lodicules are generally found in the second whorl, three or six stamens in the third whorl, and one pistil in the fourth and last whorl. The pistil is compound tricarpellate, although not visibly, as its three carpel primordia start the process of fusion immediately after their inception from the floral meristem (FM), but the number of locules is reduced to one containing a single ovule [20, 24, 25]. In the basalmost subfamily Anomochlooideae, the style in Streptochaeta has three stylar branches each ending in a stigma, while Anomochloa has a single structure that presumably corresponds to a single stylar branch and stigma [26]. In the monophyletic lineage formed by the small subfamily Puelioideae and the large BOP and PACMAD clades, which include >99.8% of the grass species, two stylar branches and two stigmas are mostly found, so this lineage has been referred as the bistigmatic clade, which assumes that this is their synapomorphic character, and that the three stigmas character observed in species such as Puelia and several bamboos is derived. Further reductions are found in some species. For example, in corn the two branch initials are congenitally fused to form the long silk with bipartite tip [6, 26, 27]. Upon successful fertilization, the grass ovary develops into a dry indehiscent fruit termed caryopsis, which is similar to an achene except that the pericarp is fused to the seed coat. Especially in domesticated cereals, the caryopsis is commonly referred as grain or kernel.

Flower Development in Monocots

3

63

The Spikelet and the Flower in Rice In the subfamily Oryzoideae (previously known as Ehrhartoideae), to which rice belongs, the spikelet is reduced and develops only the terminal flower. Thus, the rice spikelet consists of two extremely reduced rudimentary glumes subtending a terminal flower with its lemma and palea. Additionally, two small scales are attached to the rachilla, in between the rudimentary glumes and the lemma and palea, which are sometimes referred as sterile glumes or, more properly, as sterile lemmas (see Figs. 1 and 2a). These sterile lemmas are generally interpreted as the remnants of the lemmas of two lateral repressed flowers which implies that, at some point, the ancestor lineage of Oryzoideae had a three-flowered spikelet [27, 28]. The genetic mechanism that led to the putative reduction from three- to mono-flowered spikelet in Oryzoideae is not clear yet. Several genes are involved in the repression of the two lateral sterile lemmas in rice, among which are the SEP gene OsMADS34/PANICLE PHYTOMER2 (PAP2) [29–31], the ALOG gene LONG STERILE LEMMA1 (G1)/ELONGATED EMPTY GLUME (ELE) [32, 33], the putative lipase gene EXTRA GLUME1 (EG1) [34], the C2H2 zinc finger gene NONSTOP GLUMES1 (NSG) [35], and the ABERRANT SPIKELET AND PANICLE1 (ASP1) gene encoding a putative TOPLESS-like transcriptional co-repressor [36]. However, loss of function of any of these genes simply leads to the sterile lemmas partially or fully developing into ectopic lemmas, yet a lateral FM fails to form at their axil. Only recently, a mutant forming such lateral FMs has been discovered [37]. This is the dominant gain-of-function mutant lateral floret 1 (lf1), in which relatively complete lateral flowers develop, with a palea and all inner floral organs, although the sterile lemmas remain reduced as in the wild type or are even more degenerated. It would be very interesting to test the effect of combining lf1 with mutations of the genes repressing sterile lemmas cited above. In any case, it must be noticed that the “resurrection” of both lateral flowers occurs in only 4% of lf1 mutant spikelets, suggesting that other unknown molecular mechanisms provide robustness to the monoflowered spikelet trait of rice. LF1 encodes the rice HD-ZIP III transcriptional activator factor OsHB1, homologous to Arabidopsis REVOLUTA (REV), and positive regulator of the meristem maintenance gene ORYZA SATIVA HOMEOBOX1 (OSH1) [38]. The causal mutation of lf1 is a single nucleotide substitution making its transcript resistant to miRNA165/166. The resistance to miRNA165/166 explains the ls1 ectopic activity in the region of sterile lemmas, where it probably activates the lateral FMs in an OSH1-dependent manner. Furthermore, this pioneer study also suggests that a miRNA165/166–LF1–OSH1 pathway exists in

64

Ludovico Dreni

Fig. 2 Examples of MADS-box floral homeotic mutants of rice. (a) Wild type spikelets, from the left to the right: a whole spikelet; a spikelet where the frontal part of lemma has been removed to expose the palea MRP (arrows and inset); a spikelet where the frontal parts of lemma and palea have been removed; a spikelet where lemma and palea have been completely removed; a dissected pistil. (b) Osmads6 mutant spikelets gradually dissected from the left to the right, as the wild type samples above; the MRP tissues are absent (arrows). (c) Osmads1 mutant spikelets, from the left to the right: a strongly affected indeterminate spikelet with another spikelet initiated inside; a mildly affected spikelet with defective lemma and palea and inner floral organs; another mildly affected spikelet where lemma and palea have been removed to show an ectopic glume (g), the conversion of lodicules into MRP tissue (arrows), and one ectopic lodicule on the palea side (arrowhead). (d) An osmads16/spw1 mutant spikelet where lemma and palea have been removed to show the conversion of lodicules into MRP tissue (arrows) and of each stamen into an ectopic pistil. (e) An osmads3 osmads13 osmads58 triple mutant spikelet where lemma and palea have been removed to show that the reproductive organs have been replaced by an indeterminate axis bearing an ectopic palea-like organ (ep) and multiple ectopic lodicules (arrowheads). Bar represents 1 mm. r rudimentary glume, sl sterile lemma, l lemma, p palea, MRP marginal tissue of palea, BOP body of palea, lo lodicule, a anther, pi pistil, lp lemma-like palea, gl glumelodicule chimeric organ, g ectopic glume, ep ectopic palea, osm osmads

Flower Development in Monocots

65

rice, similar to the miRNA165/166–REV–STM pathway identified in Arabidopsis [39]. Indeed, the generation of three-flowered spikelet varieties seems a promising way to boost rice productivity, and the topic has been addressed in a recent review [40]. Rice rudimentary glumes, sterile lemmas, lemma and palea follow an alternate distichous pattern (see Figs. 1 and 2a). The rice lemma is characterized by five parallel vascular strands, the absence of stomata, an abaxial (inner) smooth epidermis, the adaxial (outer) epidermis decorated by unique regular cuticular thickenings forming protrusions or bulges, and by trichomes arranged in lines. The palea is formed by a body of palea (BOP) resembling the lemma but with only three vascular strands and slightly smaller adaxial cuticular bulges, and by two smooth, thin, papery and semi-transparent marginal regions of palea (MRPs) [41] (see Figs. 2a and 3). The two lemma borders are hooked inward to form an interlocking structure with the MRPs, so that the floral organs of the three inner whorls, lodicules, stamens, and gynoecium, are completely sealed by lemma and palea. The two fleshy lodicules, which develop on the lemma side, are probably modified second whorl floral organs homologous to the inner tepal whorl of other monocots [26]. Six stamens and a central pistil occupy the third and fourth whorls, respectively (see Figs. 2a and 3). The last part of the FM terminates into one ovule primordium which, at the end of its development, is hemianatropous with the micropyle and egg apparatus positioned at the base of the ovary toward the lemma side [42, 43]. When the flower is mature, the swelling of the lodicules causes the opening of lemma and palea for a short time, during which pollen is shed from the anthers and the pollination of the plumose stigma occurs. Upon successful fertilization of the embryo sac, the pistil starts to grow into a caryopsis. Rice is predominantly autogamous, which facilitates its use as model species in growth chambers and open field, although out-crossing does occur at a low frequency. In the studies aimed to compare grass and eudicot flowers, lemma and palea are often referred as the first whorl organs, which is highly questionable. A direct observation of grass mature flowers, and of early floral buds by electron microscopy, reveals that lemma and palea develop sequentially from the FM as two separate whorls [44]. Nevertheless, their possible homology to eudicot sepals, or at least to monocot outer tepals, has not been demonstrated so far. Several hypotheses can be encountered in the literature, from lemma and palea being derived from a bract and a prophyll, respectively, to them being actually outer perianth organs [45, 46]. However, the lemma arises on the main axis of the spikelet, the rachilla (see Fig. 1), which is distinct from the floral axis, thus supporting it being a modified leaf or bract which subtends the floral meristem in its axil. The palea is the first organ that originates on the floral axis (see Fig. 1) and then, it is commonly

66

Ludovico Dreni

Fig. 3 Diagrammatic representation of wild-type and several types of MADS-box mutants. sl sterile lemma, l lemma, p palea, MRP marginal tissue of palea, BOP body of palea, lo lodicule, osm osmads

considered a prophyll [46]. Interestingly, another hypothesis is that the palea is a far more complex organ in which, essentially, the BOP is derived by a prophyll, that became fused laterally with two of the three first whorl sepals/tepals to form the two MRPs, and the third tepal having been suppressed [47]. Alternatively, the palea has been interpreted as two directly fused sepals (adaxial tepals) [46, 48]. The debate addressing the ancestry and origin of these spikelet organs was best described by Prof. Elizabeth Kellogg in 2001 as “vast and largely inconclusive” [20]. Genetic studies

Flower Development in Monocots

67

conducted in the last two decades, some of which I review below, have provided more clues but not yet conclusive proofs. Further analysis of existing and future mutants is needed to test these models, including in other grass species where the distinction of BOP and MRP regions is much less pronounced than in rice.

4

Genetic Control of Spikelet and Floral Organ Identity in Rice: An ABC View There is a vast body of literature addressing the genes and molecular networks regulating floral development in rice and other model grasses. This chapter focuses in reviewing the role of MADS-box genes and to clarify if, and how, the ABC model can be applied to rice and, by extension, to other monocots. Nor do I discuss the complex mechanisms of floral meristem regulation in grasses, which is a topic already covered by excellent research articles and reviews [18, 49–54]. MIKCC (type II) MADS-box transcription factors (TFs) are master regulators of reproductive habit and organ development in flowering plants. The ABC model [55], including its further adaptations [56–58] (see also Chapters 1 and 2), sums up the combined activity of five classes of MIKCC TFs to specify each type of floral organ of core eudicots: A-class + E-class, sepal identity and repression of C-class in whorl 1; A + B + E, petal identity in whorl 2; B + C + E, stamen identity in whorl 3; C + E, specification of carpel identity and of FM determinacy, and repression of A-class in whorl 4; D + E, ovule identity inside the gynoecium and, in some species such as petunia, contribution to FM determinacy. These functional classes belong to five conserved subfamilies of MIKCC proteins: the A function proteins belong to the AP1/SQUA subfamily; B belongs to the two closely related subfamilies PISTILLATA/GLOBOSA (PI/GLO) and APETALA3/DEFICIENS (AP3/DEF); both C and D proteins belong to the AGAMOUS subfamily; and E-class proteins belong to the SEPALLATA (SEP) subfamily. Besides MIKCC TFs, only one different type of TF is ascribed to the ABC model, the Arabidopsis APETALA2 (AP2) which supports the A function of sepal identity together with APETALA1 (AP1) [55]. However, this role of AP2-like TFs has only been observed in Arabidopsis, while their conserved functions seem limited to the repression of C-class genes. Thus, the general roles of A-function genes are strongly questioned outside Brassicaceae [59–61]. Therefore, a modified version of the ABC model which better applies to all core eudicots has been proposed, and named (A)BC model, predicting that AP1/SQUA and SEP together play the (A) function, consisting in establishing the FM identity and the “basal floral state,” i.e., the identity of first whorl organ (sepal), over which the complexes formed by B-class, C-class and SEP TFs can establish the identity of the other three types of floral organs [61].

68

Ludovico Dreni

Another model, known as the quartet model, aims to explain the ABC model and its updated versions from a molecular perspective. MIKCC TFs have a strong affinity to form dimeric complexes, especially heterodimers; in the form of dimers, they can interact with their genomic target sites. The quartet model predicts that those MIKCC dimers involved in flower development will further interact to form tetramers [62], which regulate target gene promoters by inducing conformational changes and also by recruiting several other classes of transcriptional and chromatin regulators [62–64]. While there is solid experimental evidence of MADSbox quartets forming in vitro, it cannot be excluded that simple dimers might be functional at least in some of the target genes. At least 38 MIKCC type MADS-box genes have been reported in the rice genome [65], including four AP1/SQUA subfamily genes, five SEP subfamily genes, one AP3/DEF subfamily gene, two PI/GLO subfamily genes, and five AG subfamily genes (Table 1). Most of these rice genes have homologues in other grasses, except for OsMADS66, which is found only in rice and might be a non-functional, tandem duplication of the AG gene OsMADS58. The reproductive floral organs are remarkably conserved across Angiosperms. In agreement with this, there is growing evidence that the B, C and D function MADS-box genes are always their master regulators [66], although functional studies in the earliest diverged Angiosperm lineages are still missing. On the contrary, the possible homology of perianth organs of dicots and monocots and the concept of a conserved A function are highly debated, as discussed above. Therefore, it is simpler here to discuss the “rice ABC” starting from the much more conserved inner reproductive whorls. 4.1 C and D Function Genes of Rice

In rice, OsMADS3 and OsMADS58 belong to the AG subfamily, and are phylogenetically homologous to Arabidopsis AG, with which they share a similar expression pattern in the terminal FM and during stamen and carpel development [67–70]. The osmads3 single mutants are only partially defective, with an arrest in anther development causing male sterility, and the anther filaments partially converted into second whorl organs, i.e., lodicules. However, FM determinacy and carpel identity are only weakly affected in osmads3. Despite earlier reports of OsMADS58 RNAi knockdown plants showing strong agamous-like phenotypes [67], which might have been caused by nonspecific co-silencing of OsMADS3, stable transposon knock-down lines and CRISPRCas9 knock-outs of OsMADS58 have no conspicuous phenotypes, except for an additional stigma at low frequency [68, 71]. However, the osmads3 osmads58 double mutants show a complete loss of reproductive organ identity and FM determinacy, with anthers homeotically replaced by second whorl lodicules, and carpels replaced by small bract-like structures representing miniatures of

Flower Development in Monocots

69

Table 1 Rice genes encoding the MIKCC-type MADS-box TFs that form the multifarious complexes that regulate the identity of spikelet and floral organs and the determinacy of the floral meristem MSU annotation Rice gene

ID

RAP-DB annotation ID

Homologs in Arabidopsis

Subfamily

OsMADS14

LOC_Os03g54160 Os03g0752800

FUL, AP1, CAL, AGL79

AP1/SQUA

OsMADS15/DEP

LOC_Os07g01820 Os07g0108900

FUL, AP1, CAL, AGL79

AP1/SQUA

OsMADS18

LOC_Os07g41370 Os07g0605200

FUL, AP1, CAL, AGL79

AP1/SQUA

OsMADS20

LOC_Os12g31748 Os12g0501700

FUL, AP1, CAL, AGL79

AP1/SQUA

OsMADS1/LHS1

LOC_Os03g11614 Os03g0215400

SEP1, SEP2, SEP4

SEPALLATA

OsMADS5

LOC_Os06g06750 Os06g0162800

SEP1, SEP2, SEP4

SEPALLATA

OsMADS34/PAP2

LOC_Os03g54170 Os03g0753100

SEP1, SEP2, SEP4

SEPALLATA

OsMADS7/ OsMADS45

LOC_Os08g41950 Os08g0531700

SEP3

SEPALLATA

OsMADS8/ OsMADS24

LOC_Os09g32948 Os09g0507200

SEP3

SEPALLATA

OsMADS2

LOC_Os01g66030 Os01g0883100

PI

PI/GLO

OsMADS4

LOC_Os05g34940 Os05g0423400

PI

PI/GLO

OsMADS16/SPW1

LOC_Os06g49840 Os06g0712700

AP3

AP3/DEF

OsMADS3

LOC_Os01g10504 Os01g0201700

AG, SHP1, SHP2

AGAMOUS

OsMADS58

LOC_Os05g11414 Os05g0203800

AG, SHP1, SHP2

AGAMOUS

OsMADS66 (only in Oryza)

LOC_Os05g11380 –

AG, SHP1, SHP2

AGAMOUS

OsMADS13

LOC_Os12g10540 Os12g0207000

STK (AGL11)

AGAMOUS

OsMADS21

LOC_Os01g66290 Os01g0886200

STK (AGL11)

AGAMOUS

OsMADS6

LOC_Os02g45770 Os02g0682200

AGL6, AGL13

AGL6

OsMADS17

LOC_Os04g49150 Os04g0580700

AGL6, AGL13

AGL6

OsMADS32

LOC_Os01g52680 Os01g0726400

Lost in eudicots

OsMADS32

palea, with smooth semi-transparent marginal tissue and a body covered by cuticular bulges and rigid trichomes. Only the adaxial inner epidermis of these bracts remained similar to the corresponding one in the wild type carpels [68]. From the rice AGAMOUS subfamily, OsMADS13 is homologous to SEEDSTICK (STK); however, besides its conserved expression in the developing

70

Ludovico Dreni

ovule, OsMADS13 is also expressed in the adaxial epidermis of the carpel. In triple mutants osmads3 osmads13 osmads58 (see Figs. 2e and 3), the carpel is still replaced by a palea-like organ, similar to that of osmads3 osmads58 mutants, but growing even more and where also its adaxial epidermis is completely converted to the type that is normally found in the lemma and in the palea BOP [68]. These results support that in rice, the C function is conserved and provided by OsMADS3 and OsMADS58 in a largely redundant fashion, particularly carpel identity and FM determinacy, although OsMADS3 is necessary for a complete development of the anther and for pollen formation [72]. In addition, OsMADS13 redundantly contributes to carpel identity, at least in the adaxial epidermis, and to FM determinacy [68]. However, the main role of OsMADS13 is to confer ovule identity (D function), as in the osmads13 single mutant, ovules are completely converted into ectopic carpelloid tissue [73], just as it happens in Arabidopsis when the three AG homologous genes STK, SHATTERPROOF1 (SHP1) and SHP2 are knocked out to generate stk shp1 shp2 triple mutants [58]. In Arabidopsis, several different transcription factors are involved in pistil morphogenesis in parallel or downstream to MADS-box complexes. In rice, only the homolog of the YABBY gene CRABS CLAW (CRC) [74, 75], which is named DROOPING LEAF (DL), has been studied so far [76]. While the expression of Arabidopsis CRC is completely dependent on the AG-lineage genes AG, SHP1 and SHP2 [77], rice DL stays uniformly expressed in the gynoecium of osmads3 osmads58 double mutant and osmads3 osmads13 osmads58 triple mutant, despite that the pistil is completely replaced by a palea-like organ showing all the typical cells of a palea [68]. Therefore, not only the C function is not strictly necessary to activate DL in rice, but DL is a marker of the gynoecium position rather than of carpel identity. While the crc mutant of Arabidopsis is mainly defective in carpel morphogenesis and fusion, the rice dl mutant has a much more drastic effect, since the gynoecium is completely replaced by ectopic stamens, suggesting that DL is directly or indirectly involved in the repression of B function genes in the fourth whorl [76]. The dl phenotype is just the equivalent of an ectopic expression of B-class genes in the fourth whorl as predicted by the ABC model. For this reason, the analysis of dl mutants cannot tell if DL has other roles in carpel development and morphogenesis such as CRC, which seems, however, very likely. The phenotype of dl osmads16 double mutants, consisting of ectopic tissues of unidentified identity [76], gave no conclusive response either. The repression of B genes in the gynoecium and the ectopic conversion of carpels into stamens observed in dl mutants is not explained by the ABC model. Despite this, based on the analysis of more mutants, other authors have proposed DL as a C function gene and also claimed that rice AG genes may not be

Flower Development in Monocots

71

indispensable to specify the identity of the carpel abaxial outer epidermis [71]. More in-depth studies of the molecular regulation of rice gynoecium specification are clearly needed. 4.2 B Function Genes of Rice

Rice has only one AP3-like gene, designated as OsMADS16 or SUPERWOMAN1 (SPW1), but two genes from the PI/GLO subfamily, namely OsMADS2 and OsMADS4, which show a conserved expression pattern in the second and third whorls, starting from their founder cells in the FM [76, 78]. Studies conducted so far also pointed to a significant conservation of the B function. Indeed, in rice there is evidence of a functional AP3-PI heterodimer [79, 80], and both osmads16 loss-of-function mutants (see Figs. 2d and 3), or OsMADS2-OsMADS4 RNAi co-suppression transgenic lines, showed the homeotic conversion of second whorl lodicules into small scales or glume-like organs similar to palea MRP, and the development of homeotic carpels in the third whorl in place of stamens [76]. More recent RNAi gene-specific suppression experiments suggested a partial subfunctionalization of the two PI/GLO genes: OsMADS2 is more important for lodicule development, while they are roughly equally important for stamen identity [81]. Therefore, in Arabidopsis, the loss of B function causes petalto-sepal conversion in whorl 2 while, in rice, it causes lodicules to convert into MRPs; the ectopic expression of C function genes causes both Arabidopsis whorl 1 sepals and rice MRPs to convert into carpelloid organs, while the rice BOP is less affected [55, 76, 82–84]. Considering also the relative position of MRP and lodicule in the rice spikelet, this parallelism is perhaps the strongest available molecular evidence supporting that rice MRPs and lodicules evolved from outer and inner tepals, respectively. In whorled flowers, different than flowers with spiral phyllotaxis, a sharp restriction of homeotic gene expression to the correct whorls is essential to maintain robust separation of organ identities, which is orchestrated by cadastral genes such as Arabidopsis SUPERMAN (SUP) [85–87] and RABBIT EARS (RBE) [88]. RBE and SUP encode two closely related C2H2 zinc finger transcriptional repressors which repress C-class MADS-box genes in petals and B-class genes in the gynoecium, respectively. SUP cellautonomously represses B-class genes across whorls 3 and 4 to maintain a sharp boundary between androecium and gynoecium [89]. Because of this cadastral activity, SUP has been proposed as a candidate canalization factor for the increased developmental robustness of the male/female boundary evolved by core eudicots [90]. However, the rice ortholog of SUP, SMALL REPRODUCTIVE ORGANS (SRO) has been recently characterized [91] suggesting that these pathways may have evolved quite differently in monocots. Indeed, rice SRO is essential for the development of male and female reproductive organs, but does not show any obvious cadastral function on MADS-box genes, and it seems rather to

72

Ludovico Dreni

act downstream of them [91]. Interestingly, a SUP-like function in excluding the B-class genes from the gynoecium is carried out in rice by the CRC ortholog, YABBY gene DL [76]. 4.3 Rice SEP, AP1/ SQUA, AGL6, and OsMADS32 Genes Specify Together the (A) Function

These highly conserved functions of rice B-, C- and D-class MADS TFs seem to depend on their physical interactions with SEP TFs, just like their eudicot counterparts [66]. There are five SEP proteins in rice, showing a wide spectrum of possible interaction partners. In all Angiosperms, SEPs divide into two subgroups, those orthologous to Arabidopsis SEP1/2/4 (or LOFSEP) and those orthologous to SEP3 [92, 93]. The rice OsMADS1/ LEAFY HULL STERILE1 (LHS1), OsMADS5 and OsMADS34/PANICLE PHYTOMER2 (PAP2) are orthologous to Arabidopsis SEP1, SEP2, and SEP4 in the LOFSEP clade, while OsMADS24 and OsMADS45, which are also known as OsMADS8 and OsMADS7, respectively, are orthologous to SEP3 [94] (Table 1). The two rice SEP3 genes show, like several Angiosperm homologs, an expression domain restricted to the FM and the three inner whorls, suggesting a conserved function of their encoded proteins as main interaction partners for B, C and D identity complexes [94]. Indeed, in Arabidopsis, the protein product of SEP3 is the most abundant and effective E function subunit in floral identity complexes [62, 95], yet the four Arabidopsis SEPs show a high degree of functional redundancy, with single and double mutants displaying no or minimal floral phenotypes under standard growing conditions, the sep1 sep2 sep3 triple mutant combination generating “sepallata” indeterminate flowers [57], and the sep1 sep2 sep3 sep4 quadruple mutant causing a complete floral reversion into an indeterminate vegetative shoot-like axis producing only leafy lateral organs [96]. The high redundancy of SEP genes observed in Arabidopsis, however, might be uncommon in Angiosperms [94], and RNA-interference knockdown of both rice SEP3 genes was enough to generate significant loss of identity in the three inner whorls [97]. Significant redundancy is shared anyway among rice SEPs, as knocking down all but OsMADS34/PAP2 was enough to lead an almost complete floral reversion into an indeterminate leafy branch, similar to the Arabidopsis sep quadruple mutant [97]. However, a osmads1 osmads5 osmads34 triple mutant was also enough to produce a strong floral reversion (see Fig. 3), that was also associated with the decrease and delay in the activation of expression of B-class, C-class, and also SEP3-like genes [98], despite the fact that only osmads1 could be considered a complete knock out allele. This result could indicate that in rice, the LOFSEP genes act partially upstream to SEP3, B and C function genes to establish the correct onset of their expression and the identity of the FM, a function that has not been reported for their Arabidopsis counterparts so far. Indeed, contrary to SEP3 homologs, the LOFSEP genes are discontinuously expressed or not expressed at all in the

Flower Development in Monocots

73

three inner whorls of the rice flower. The most obvious case is OsMADS1/LHS1, which first activates in the spikelet meristem when it turns into a terminal FM, being actually considered a marker gene of this fundamental transition [99], and then it stays expressed in the primordia of lemma and palea, but not in the FM nor in the inner three whorls of organ primordia, except for the differentiating ovary wall [100, 101]. However, its loss of function causes homeotic conversions of lodicules into MRP-like organs, similar to B-class mutants, and loss of FM determinacy after the differentiation of glumes (see Figs. 2c and 3). Therefore, the rice LOFSEP TFs may have a lesser function than SEP3 in forming floral identity complexes, but a major non-cell-autonomous effect derived from their earlier action in establishing the FM. Indeed, the loss-of-function allele osmads1-z alone or combined with other MADS-box mutants evidenced several classes of flower phenotypes, including premature arrest of the FM or its loss of determinacy [100]. Such opposite phenotypes are unusual in MADS-box homeotic mutants, which tend to be very uniform and stable, further suggesting that OsMADS1/LHS1 acts like a “hub” in the regulation and balancing of different molecular pathways for the proper establishment of the FM. Indeed, OsMADS1/LHS1 is supposed to regulate and coordinate transcription factors and hormone signalling pathways [102–104], and to specify lemma identity, at least in part, by repressing miR172s, therefore ensuring a proper transcript amount of the AP2 targets of miR172 [103]. All the three LOFSEP genes mediate the transition of the spikelet meristem to FM [98]. Lemma and palea are new organs that specifically arose in grass spikelets, and LOFSEP genes OsMADS1, OsMADS5 and OsMADS34 have been likely recruited to regulate their identity. In a “lofsep” triple mutant spikelets are characterized by rudimentary glumes, sterile lemmas, lemma and palea being all converted into leaf-like organs. Therefore, these three SEP-like genes have a general role in promoting FM identity and outer organ identity, whose loss causes a partial reversion to a shoot meristem and a branch (see Fig. 3) [98]. Besides this functional redundancy, OsMADS34 and OsMADS1 have specific roles in regulating the unique features of sterile lemma and lemma/palea identity, respectively. Rice sterile lemmas are the vestigial of lemmas of two repressed lateral flowers. In the osmads34 loss-of-function mutants [29, 30], as well as in transgenic lines constitutively expressing OsMADS1 [101], sterile lemmas turn into well-developed lemma-like organs (see Fig. 3). Despite that the lemma identity gene OsMADS1 is ectopically expressed in sterile lemmas of the osmads34 mutant [105], the phenotype of osmads34 is not alleviated in osmads34 osmads1 double mutants where, on the contrary, the identity of sterile lemmas is further lost versus more leaf-like organs, and even more when all LOFSEP genes are mutated [98]. While lemma and palea are similar in

74

Ludovico Dreni

appearance, several genetic pathways act specifically in the palea [46]. However, OsMADS1 is expressed in both, and the lack of its function triggers their partial loss of identity to become more leaflike and not interlocked anymore due to the disappearance of the MRPs (see Fig. 2c and 3). Therefore, OsMADS1 is a master regulator of lemma and palea development [41, 101, 106]. From the AP1/SQUA MADS-box subfamily, whose AP1 member provides the A function in Arabidopsis, the rice paralogous genes OsMADS14 and OsMADS15 are consistently expressed during floral organogenesis. However, because in osmads14 osmads15 double knock-out plants the formation of spikelet meristems from the inflorescence primary branches is completely impaired [82], their full functions and redundancy in flower development cannot be revealed by stable mutants. The analysis of single mutants and of osmads14/+ osmads15 and osmads14 osmads15/+ double mutants showed that OsMADS14 and OsMADS15 are necessary for palea development and to repress C function genes in it, while promoting FM identity and B and C genes in their correct expression domains [82]. The rudimentary glumes, sterile lemmas and lemma of osmads14/+ osmads15 double mutants were converted into leaflike organs similar to lofsep triple mutants, indicating that rice AP1/FUL and LOFSEP may work together in spikelet and flower specification. Indeed, OsMADS14, OsMADS15 and LOFSEPs are widely co-expressed and their protein products are strong dimerization partners [98, 107]. In conclusion, the AP1/SQUA and LOFSEP genes of grasses have been essential for the evolution of the new grass spikelet organs, lemma and palea, a clear process of neofunctionalization, but they also keep a function in the establishment of FM identity and in orchestrating the function of the more conserved SEP3, B, C and D floral identity genes. Although not complete, the conversion of osmads14/+ osmads15 and lofsep mutant spikelets into vegetative organs corresponds to the effect seen in sep1 sep2 sep3 sep4 quadruple mutant flowers. Therefore, it definitely makes sense to apply to rice the modified (A)BC model [61] (see Fig. 4). It is important to mention that rice flower development is also regulated by at least two other MIKCC TFs, which are not homologous to any of the typical eudicot ABCDE model TFs: OsMADS6 or MOSAIC FLORAL ORGANS 1 (MFO1) [47, 108] and OsMADS32/CHIMERIC FLORAL ORGANS1 (CFO1) [109]. OsMADS6/MFO1 is a member of the AGL6 subfamily, which is sister to SEPALLATA [93, 110–112]. Its expression starts in the FM and palea primordium, and is then detected in the lodicule, pistil, and ovule primordia, especially in the ovule integuments, but not in anther primordia. In the developing palea, its expression ceases soon in the BOP and persists only in the MRPs. When the function of OsMADS6 is lost, not only the MRPs disappear and the BOP just converts into another lemma, but the two

Flower Development in Monocots

75

Fig. 4 The proposed (A)BC model of rice. The activity of rice homeotic MADS-box genes in floral organs specification can be best summarized by a modified ABC model with widespread applicability that has been proposed previously [57]. However, in rice the (A) function should also include the AGL6 and OsMADS32 subfamilies, and FUL genes prevent the expression of both B and C genes to expand outwards, while in Arabidopsis they only restrict C genes. Moreover, the (A) function genes, except SEP3 homologs, have also neofunctionalized to specify the identity of the grass floral bracts lemma and palea outside the floral whorls

lodicules in the second whorl lose their identity and turn into small glume- or glume-lodicule chimeric organs, and similar additional organs appear on the palea side of the second whorl (see Fig. 2b). Although OsMADS6 is not expressed in anther primordia, osmads6 anthers develop mostly as chimeric organs, particularly lodicule/ glume/anther mosaic organs, probably reflecting a partial loss of FM identity and organization. In agreement with this hypothesis, the floral defects of osmads1 and osmads6 increase much when they are combined into double mutants [108], suggesting that the function in FM establishment is shared between rice AP1/SQUA, LOFSEP and AGL6-like genes. Also the carpels and ovules are defective in osmads6, similar to a partial loss of C and D function [47, 108]. These observations, and the wide ability of OsMADS6 to dimerize with most of the other floral identity MADS-box proteins [79, 113–115], led the rice scientific community to assign to OsMADS6 a SEP-like function in promoting the identity of floral meristem and floral organs [111]. So, the rice AGL6 family should be included into the (A) function (see Fig. 4), just like in petunia [60, 110]. OsMADS32/CFO1 belongs to a subfamily which is only found in Amborella and monocots [109, 116], but lost in eudicots, and the osmads32 mutant phenotype in rice is quite comparable to that of osmads6 [109]. Interestingly, the OsMADS32 subfamily is sister to AP3/DEF and PI/GLO subfamilies [117], however, its larger expression pattern and the wider interaction abilities of its encoded protein [118, 119], together with the osmads32 mutant phenotypes, make OsMADS32 apparently more similar to SEPs. In eudicots, MADS-box tetramers are mostly possible thanks to the ability

76

Ludovico Dreni

of SEP proteins to bridge their interaction partners. This apparent SEP-like function of rice OsMADS6 and OsMADS32 seems to further complicate this scenario, raising the question of which subunit combinations the putative rice floral identity tetramers are exactly made of. It would be very interesting to isolate the rice floral identity complexes formed in vivo, and to study them by mass spectrometry, just like it has been done in Arabidopsis [62]. Meanwhile, I propose to include also OsMADS32 in the (A) function (see Fig. 4).

Acknowledgements I thank the editors for inviting my contribution, and the European Commission and CSIC for funding (Marie Sklodowska–Curie Individual Fellowship RI No. 661678 “GainGrain” and “convocatoria extensio´n MSCA IF ERC,” respectively). References 1. Stevens PF (2001 onwards) Angiosperm phylogeny website. Version 14 July 2017 [and more or less continuously updated since]. h t t p : // w w w. m o b o t . o r g / M O B O T / research/APweb/ 2. Buzgo M, Endress PK (2000) Floral structure and development of acoraceae and its systematic relationships with basal angiosperms. Int J Plant Sci 161:23–41 3. Rudall PJ, Bateman RM (2004) Evolution of zygomorphy in monocot flowers: iterative patterns and developmental constraints. New Phytol 162:25–44. https://doi.org/10. 1111/J.1469-8137.2004.01032.X 4. Vergara-Silva F, Espinosa-Matı´as S, Ambrose BA, Va´zquez-Santana S, Martı´nez-Mena A, Ma´rquez-Guzma´n J, Martı´nez E, Meyerowitz EM, Alvarez-Buylla ER (2003) Inside-out flowers characteristic of Lacandonia schismatica evolved at least before its divergence from a closely related taxon, Triuris brevistylis. Int J Plant Sci 164:345–357. https://doi.org/10. 1086/368235 ´ lvarez-Buylla ER, Ambrose BA, Flores5. A Sandoval E, Vergara-Silva F, Englund M, Garay-Arroyo A, Garcı´a-Ponce B, de la Torre-Ba´rcena E, Espinosa-Matı´as S, Martı´nez E et al (2010) B-function expression in the flower center underlies the homeotic phenotype of Lacandonia schismatica (Triuridaceae). Plant Cell 22:3543–3559. https://doi. org/10.1105/tpc.109.069153 6. Aliscioni S, Bell HL, Besnard G, Christin PA, Columbus JT, Duvall MR, Edwards EJ,

Giussani L, Hasenstab-Lehman K, Hilu KW et al (2012) New grass phylogeny resolves deep evolutionary relationships and discovers C4 origins. New Phytol 193:304–312. https://doi.org/10.1111/J.1469-8137. 2011.03972.X 7. Wang W, Mauleon R, Hu Z, Chebotarov D, Tai S, Wu Z, Li M, Zheng T, Fuentes RR, Zhang F et al (2018) Genomic variation in 3,010 diverse accessions of Asian cultivated Rice. Nature 557:43–49. https://doi.org/ 10.1038/s41586-018-0063-9 8. Gordon SP, Contreras-Moreira B, Woods DP, Des Marais DL, Burgess D, Shu S, Stritt C, Roulin AC, Schackwitz W, Tyler L (2017) Extensive gene content variation in the Brachypodium distachyon pan-genome correlates with population structure. Nat Commun 8:2184. https://doi.org/10. 1038/s41467-017-02292-8 9. Chapman MA, He Y, Zhou M (2022) Beyond a reference genome: pangenomes and population genomics of underutilized and orphan crops for future food and nutrition security. New Phytol 234:1583–1597 10. Gui S, Wei W, Jiang C, Luo J, Chen L, Wu S, Li W, Wang Y, Li S, Yang N et al (2022) A pan-Zea genome map for enhancing maize improvement. Genome Biol 23:178. https://doi.org/10.1186/s13059-02202742-7 11. Hufford MB, Seetharam AS, Woodhouse MR, Chougule KM, Ou S, Liu J, Ricci WA, Guo T, Olson A, Qiu Y et al (2021) De novo

Flower Development in Monocots assembly, annotation, and comparative analysis of 26 diverse maize genomes. Science 373: 655–662. https://doi.org/10.1126/science. abg5289 12. Walkowiak S, Gao L, Monat C, Haberer G, Kassa MT, Brinton J, Ramirez-Gonzalez RH, Kolodziej MC, Delorean E, Thambugala D et al (2020) Multiple wheat genomes reveal global variation in modern breeding. Nature 588:277–283. https://doi.org/10.1038/ s41586-020-2961-x 13. Goff SA, Ricke D, Lan TH, Presting G, Wang R, Dunn M, Glazebrook J, Sessions A, Oeller P, Varma H et al (2002) A draft sequence of the rice genome (Oryza sativa L. ssp. japonica). Science 296:92–100. https://doi.org/10.1126/science.1068275 14. Yu J, Hu S, Wang J, Wong GKS, Li S, Liu B, Deng Y, Dai L, Zhou Y, Zhang X et al (2002) A draft sequence of the rice genome (Oryza sativa L. ssp. indica). Science 296:79–92. https://doi.org/10.1126/science.1068037 15. Kaul S, Koo HL, Jenkins J, Rizzo M, Rooney T, Tallon LJ, Feldblyum T, Nierman W, Benito MI, Lin X et al (2000) Analysis of the genome sequence of the flowering plant Arabidopsis thaliana. Nature 408: 796–815. https://doi.org/10.1038/ 35048692 16. Bommert P, Whipple C (2018) Grass inflorescence architecture and meristem determinacy. Semin Cell Dev Biol 79:37–47 17. Zhang D, Yuan Z (2014) Molecular control of grass inflorescence development. Annu Rev Plant Biol 65:553–578. https://doi.org/10. 1146/annurev-arplant-050213-040104 18. Kellogg EA (2022) Genetic control of branching patterns in grass inflorescences. Plant Cell 34:3518–2533. https://doi.org/ 10.1093/plcell/koac080 19. Soreng RJ, Davis JI (1998) Phylogenetics and character evolution in the grass family (Poaceae): simultaneous analysis of morphological and chloroplast DNA restriction site character sets. Bot Rev 64:1–85. https://doi.org/10. 1007/BF02868851 20. Kellogg EA (2001) Evolutionary history of the grasses. Plant Physiol 125:1198–1205. https://doi.org/10.1104/pp.125.3.1198 21. Grac¸a Sajo M, Pabo´n-Mora N, Jardim J, Stevenson DW, Rudall PJ (2012) Homologies of the flower and inflorescence in the earlydivergent grass Anomochloa (Poaceae). Am J Bot 99:614–628. https://doi.org/10.3732/ ajb.1100290 22. Sajo MG, Longhi-Wagner HM, Rudall PJ (2008) Reproductive morphology of the

77

early-divergent grass Streptochaeta and its bearing on the homologies of the grass spikelet. Plant Syst Evol 275:245–255. https:// doi.org/10.1007/s00606-008-0080-5 23. Preston JC, Christensen A, Malcomber ST, Kellogg EA (2009) MADS-box gene expression and implications for developmental origins of the grass spikelet. Am J Bot 96:1419– 1429. https://doi.org/10.3732/ajb. 0900062 24. Dreni L, Osnato M, Kater MM (2013) The ins and outs of the rice AGAMOUS subfamily. Mol Plant 6:650–664. https://doi.org/10. 1093/mp/sst019 25. Rudall PJ, Stuppy W, Cunniff J, Kellogg EA, Briggs BG (2005) Evolution of reproductive structures in grasses (Poaceae) inferred by sister-group comparison with their putative closest living relatives, Ecdeiocoleaceae. Am J Bot 92:1432–1443. https://doi.org/10. 3732/ajb.92.9.1432 26. Kellogg EA (2015) Flowering plants. Monocots: Poaceae. In: Kubitzki K (ed) Vol XIII of: The families and genera of vascular plants. Springer, New York. ISBN 9783319153322 27. Grass Phylogeny Working Group, Barker NP, Clark LG, Davis JI, Duvall MR, Guala GF, Hsiao C, Kellogg EA, Linder HP (2001) Phylogeny and subfamilial classification of the grasses (Poaceae). Ann Mo Bot Gard 88: 373. https://doi.org/10.2307/3298585 28. Kellogg EA (2009) The evolutionary history of Ehrhartoideae, Oryzeae, and Oryza. Rice 2:1–14 29. Kobayashi K, Maekawa M, Miyao A, Hirochika H, Kyozuka J (2010) PANICLE PHYTOMER2 (PAP2), encoding a SEPALLATA subfamily MADS-box protein, positively controls spikelet meristem identity in rice. Plant Cell Physiol 51:47–57. https:// doi.org/10.1093/pcp/pcp166 30. Gao X, Liang W, Yin C, Ji S, Wang H, Su X, Guo C, Kong H, Xue H, Zhang D (2019) The SEPALLATA-like gene OsMADS34 is required for Rice inflorescence and spikelet development. Plant Physiol 153:728–740. https://doi.org/10.1104/pp.110.156711 31. Lin X, Wu F, Du X, Shi X, Liu Y, Liu S, Hu Y, Theißen G, Meng Z (2014) The pleiotropic SEPALLATA-like gene OsMADS34 reveals that the “empty glumes” of rice (Oryza sativa) spikelets are in fact rudimentary lemmas. New Phytol 202:689–702. https://doi.org/10. 1111/nph.12657 32. Yoshida A, Suzaki T, Tanaka W, Hirano HY (2009) The homeotic gene long sterile Lemma (G1) specifies sterile lemma identity

78

Ludovico Dreni

in the rice spikelet. Proc Natl Acad Sci USA 106:20103–22108. https://doi.org/10. 1073/pnas.0907896106 33. Hong L, Qian Q, Zhu K, Tang D, Huang Z, Gao L, Li M, Gu M, Cheng Z (2010) ELE restrains empty glumes from developing into lemmas. J Genet Genomics 37:101–115. https://doi.org/10.1016/S1673-8527(09) 60029-1 34. Li H, Xue D, Gao Z, Yan M, Xu W, Xing Z, Huang D, Qian Q, Xue Y (2009) A putative lipase gene EXTRA GLUME1 regulates both empty-glume fate and spikelet development in rice. Plant J 57:593–605. https://doi.org/ 10.1111/j.1365-313X.2008.03710.x 35. Zhuang H, Wang HL, Zhang T, Zeng XQ, Chen H, Wang ZW, Zhang J, Zheng H, Tang J, Ling YH et al (2020) NONSTOP GLumes1 encodes a C2H2 zinc finger protein that regulates spikelet development in rice. Plant Cell 32:392–413. https://doi.org/10. 1105/tpc.19.00682 36. Yoshida A, Ohmori Y, Kitano H, TaguchiShiobara F, Hirano HY (2012) ABERRANT SPIKELET and PANICLE1, encoding a TOPLESS-related transcriptional co-repressor, is involved in the regulation of meristem fate in rice. Plant J 70:327–339. https://doi.org/10.1111/j.1365-313X. 2011.04872.x 37. Zhang T, Li Y, Ma L, Sang X, Ling Y, Wang Y, Yu P, Zhuang H, Huang J, Wang N et al (2017) LATERAL FLORET 1 induced the three-florets spikelet in rice. Proc Natl Acad Sci U S A 114:9984–9989. https://doi.org/ 10.1073/pnas.1700504114 38. Matsuoka M, Ichikawa H, Saito A, Tada Y, Fujimura T, Kano-Murakami Y (1993) Expression of a rice homeobox gene causes altered morphology of transgenic plants. Plant Cell 5:1039–1048. https://doi.org/ 10.2307/3869625 39. Shi B, Zhang C, Tian C, Wang J, Wang Q, Xu T, Xu Y, Ohno C, Sablowsk R, Heisler MG et al (2016) Two-step regulation of a meristematic cell population acting in shoot branching in Arabidopsis. PLoS Genet 12: e1006168. https://doi.org/10.1371/jour nal.pgen.1006168 40. Ren D, Li Y, He G, Qian Q (2019) Multifloret spikelet improves rice yield. New Phytol 225:2301–2306. https://doi.org/10.1111/ nph.16303 41. Prasad K, Parameswaran S, Vijayraghavan U (2005) OsMADS1, a rice MADS-box factor, controls differentiation of specific cell types in the lemma and palea and is an early-acting regulator of inner floral organs. Plant J 43:

915–928. https://doi.org/10.1111/j. 1365-313X.2005.02504.x 42. Lopez-Dee ZP, Wittich P, Pe` ME, Rigola D, Del Buono I, Sari Gorla M, Kater MM, Colombo L (1999) OsMADS13, a novel rice MADS-box gene expressed during ovule development. Dev Genet 25:237–244. https://doi.org/10.1002/(SICI)1520-6408 (1999)25:33.0.CO;2-L 43. Itoh J-I, Nonomura K-I, Ikeda K, Yamaki S, Inukai Y, Yamagishi H, Kitano H, Nagato Y (2005) Rice plant development: from zygote to spikelet. Plant Cell Physiol 46:23–47. https://doi.org/10.1093/pcp/pci501 44. Ambrose BA, Lerner DR, Ciceri P, Padilla CM, Yanofsky MF, Schmidt RJ (2000) Molecular and genetic analyses of the silky1 gene reveal conservation in floral organ specification between eudicots and monocots. Mol Cell 5:569–579. https://doi.org/10.1016/ S1097-2765(00)80450-5 45. Schrager-Lavelle A, Klein H, Fisher A, Bartlett M (2017) Grass flowers: an untapped resource for floral Evo-devo. J Syst Evol 55: 525–541 46. Lombardo F, Yoshida H (2015) Interpreting lemma and palea homologies: a point of view from rice floral mutants. Front Plant Sci 6:61. https://doi.org/10.3389/fpls.2015.00061 47. Ohmori S, Kimizu M, Sugita M, Miyao A, Hirochika H, Uchida E, Nagato Y, Yoshida H (2009) MOSAIC FLORAL ORGANS1, an AGL6-like MADS box gene, regulates floral organ identity and meristem fate in rice. Plant Cell 21:3008–3025. https://doi.org/10. 1105/tpc.109.068742 48. Reinheimer R, Kellogg EA (2009) Evolution of AGL6-like MADS box genes in grasses (Poaceae): ovule expression is ancient and palea expression is new. Plant Cell 21:2591– 2605. https://doi.org/10.1105/tpc.109. 068239 49. Pautler M, Tanaka W, Hirano HY, Jackson D (2013) Grass meristems I: shoot apical meristem maintenance, axillary meristem determinacy and the floral transition. Plant Cell Physiol 54:302–312 50. Tanaka W, Pautler M, Jackson D, Hirano HY (2013) Grass meristems II: inflorescence architecture, flower development and meristem fate. Plant Cell Physiol 54:313–324 51. Zong J, Wang L, Zhu L, Bian L, Zhang B, Chen X, Huang G, Zhang X, Fan J, Cao L et al (2022) A rice single cell transcriptomic atlas defines the developmental trajectories of rice floret and inflorescence meristems. New

Flower Development in Monocots Phytol 234:494–512. https://doi.org/10. 1111/nph.18008 52. Tanaka W, Ohmori S, Kawakami N, Hirano HY (2021) Flower meristem maintenance by TILLERS ABSENT 1 is essential for ovule development in rice. Development 148: dev199932. https://doi.org/10.1242/dev. 199932 53. Shen C, Li G, Dreni L, Zhang D (2021) Molecular control of carpel development in the grass family. Front Plant Sci 12:635500 54. Chongloi GL, Prakash S, Vijayraghavan U (2019) Regulation of meristem maintenance and organ identity during rice reproductive development. J Exp Bot 70:1719–1736. https://doi.org/10.1093/jxb/erz046 55. Coen E, Meyerowitz E (1991) The war of the whorls: genetic interactions controlling flower development. Nature 353:31–37 56. Colombo L, Franken J, Koetje E, Van Went J, Dons HJM, Angenent GC, Van Tunen AJ (1995) The petunia MADS box gene FBP11 determines ovule identity. Plant Cell 7:1859– 1868. https://doi.org/10.2307/3870193 57. Pelaz S, Ditta GS, Baumann E, Wisman E, Yanofsky MF (2000) B and C floral organ identity functions require SEPALLATA MADS-box genes. Nature 405:200–203. https://doi.org/10.1038/35012103 58. Pinyopich A, Ditta GS, Savidge B, Liljegren SJ, Baumann E, Wisman E, Yanofsky MF (2003) Assessing the redundancy of MADSbox genes during carpel and ovule development. Nature 424:85–88. https://doi.org/ 10.1038/nature01741 59. Thomson B, Wellmer F (2019) Molecular regulation of flower development. In: Grossniklaus U (ed) Plant development and evolution. Current topics in developmental biology, vol 131. Academic Press, New York 60. Monniaux M, Vandenbussche M (2018) How to evolve a perianth: a review of cadastral mechanisms for perianth identity. Front Plant Sci 9:1573 61. Causier B, Schwarz-Sommer Z, Davies B (2010) Floral organ identity: 20 years of ABCs. Semin Cell Dev Biol 21:73–79. https://doi.org/10.1016/j.semcdb.2009. 10.005 ˜ o JM, 62. Smaczniak C, Immink RGH, Muin Blanvillain R, Busscher M, Busscher-Lange J, Dinh QD, Liu S, Westphal AH, Boeren S et al (2012) Characterization of MADS-domain transcription factor complexes in Arabidopsis flower development. Proc Natl Acad Sci U S A 109:1560–1565. https://doi.org/10.1073/ pnas.1112871109

79

63. Theißen G, Melzer R, Ruu¨mpler F (2016) MADS-domain transcription factors and the floral quartet model of flower development: linking plant development and evolution. Development 143:3259–3271 64. Hugouvieux V, Silva CS, Jourdain A, Stigliani A, Charras Q, Conn V, Conn SJ, Carles CC, Parcy F, Zubieta C (2018) Tetramerization of MADS family transcription factors SEPALLATA3 and AGAMOUS is required for floral meristem determinacy in Arabidopsis. Nucleic Acids Res 46:4966– 4977. https://doi.org/10.1093/nar/ gky205 65. Arora R, Agarwal P, Ray S, Singh AK, Singh VP, Tyagi AK, Kapoor S (2007) MADS-box gene family in rice: genome-wide identification, organization and expression profiling during reproductive development and stress. BMC Genomics 8:242. https://doi.org/10. 1186/1471-2164-8-242 66. Dreni L, Kater MM (2014) MADS reloaded: evolution of the AGAMOUS subfamily genes. New Phytol 201:717–732. https://doi.org/ 10.1111/nph.12555 67. Yamaguchi T, Lee DY, Miyao A, Hirochika H, An G, Hirano H-Y (2006) Functional diversification of the two C-class MADS box genes OSMADS3 and OSMADS58 in Oryza sativa. Plant Cell 18:15–28. https://doi.org/10. 1105/tpc.105.037200 68. Dreni L, Pilatone A, Yun D, Erreni S, Pajoro A, Caporali E, Zhang D, Kater MM (2011) Functional analysis of all AGAMOUS subfamily members in rice reveals their roles in reproductive organ identity determination and meristem determinacy. Plant Cell 23: 2850–2863. https://doi.org/10.1105/tpc. 111.087007 69. Zahn LM, Leebens-Mack JH, Arrington JM, Hu Y, Landherr LL, DePamphilis CW, Becker A, Theissen G, Ma H (2006) Conservation and divergence in the AGAMOUS subfamily of MADS-box genes: evidence of independent sub- and neofunctionalization events. Evol Dev 8:30–45. https://doi.org/ 10.1111/j.1525-142X.2006.05073.x 70. Kramer EM, Jaramillo MA, Di Stilio VS (2004) Patterns of gene duplication and functional evolution during the diversification of the AGAMOUS subfamily of MADS box genes in angiosperms. Genetics 166:1011– 1023. https://doi.org/10.1534/genetics. 166.2.1011 71. Sugiyama S-H, Yasui Y, Ohmori S, Tanaka W, Hirano H-Y (2019) Rice flower development revisited: regulation of carpel specification and flower meristem determinacy. Plant Cell

80

Ludovico Dreni

Physiol 60:1284–1295. https://doi.org/10. 1093/pcp/pcz020 72. Hu L, Liang W, Yin C, Cui X, Zong J, Wang X, Hu J, Zhang D (2011) Rice MADS3 regulates ROS homeostasis during late anther development. Plant Cell 23:515– 533. https://doi.org/10.1105/tpc.110. 074369 73. Dreni L, Jacchia S, Fornara F, Fornari M, Ouwerkerk PBF, An G, Colombo L, Kater MM (2007) The D-lineage MADS-box gene OsMADS13 controls ovule identity in rice. Plant J 52:690–699. https://doi.org/10. 1111/j.1365-313X.2007.03272.x 74. Alvarez J, Smyth DR (1999) CRABS CLAW and SPATULA, two Arabidopsis genes that control carpel development in parallel with AGAMOUS. Development 126:2377–2386 75. Alvarez J, Smyth DR (2002) CRABS CLAW and SPATULA genes regulate growth and pattern formation during gynoecium development in Arabidopsis thaliana. Int J Plant Sci 163:17–41. https://doi.org/10.1086/ 324178 76. Nagasawa N, Miyoshi M, Sano Y, Satoh H, Hirano H, Sakai H, Nagato Y (2003) SUPERWOMAN1 and DROOPING LEAF genes control floral organ identity in rice. Development 130:705–718 77. Morel P, Heijmans K, Ament K, Chopy M, Trehin C, Chambrier P, Bento SR, Bimbo A, Vandenbussche M (2018) The floral C-lineage genes trigger nectary development in Petunia and Arabidopsis. Plant Cell 30: 2020–2037. https://doi.org/10.1105/tpc. 18.00425 78. Yun D, Liang W, Dreni L, Yin C, Zhou Z, Kater MM, Zhang D (2013) OsMADS16 genetically interacts with OsMADS3 and OsMADS58 in specifying floral patterning in rice. Mol Plant 6:743–756. https://doi.org/ 10.1093/mp/sst003 79. Seok HY, Park HY, Park JI, Lee YM, Lee SY, An G, Moon YH (2010) Rice ternary MADS protein complexes containing class B MADS heterodimer. Biochem Biophys Res Commun 401:598–604. https://doi.org/10.1016/j. bbrc.2010.09.108 80. Moon YH, Jung JY, Kang HG, An G (1999) Identification of a rice APETALA3 homologue by yeast two-hybrid screening. Plant Mol Biol 40:167–177. https://doi.org/10. 1023/A:1026429922616 81. Yao S-G, Ohmori S, Kimizu M, Yoshida H (2008) Unequal genetic redundancy of rice PISTILLATA orthologs, OsMADS2 and OsMADS4, in lodicule and stamen

development. Plant Cell Physiol 49:853– 857. https://doi.org/10.1093/pcp/pcn050 82. Wu F, Shi X, Lin X, Liu Y, Chong K, Theißen G, Meng Z (2017) The ABCs of flower development: mutational analysis of AP1/FUL-like genes in rice provides evidence for a homeotic (A)-function in grasses. Plant J 89:310–324. https://doi.org/10. 1111/tpj.13386 83. Zheng M, Wang Y, Wang Y, Wang C, Ren Y, Lv J, Peng C, Wu T, Liu K, Zhao S et al (2015) DEFORMED FLORAL ORGAN1 (DFO1) regulates floral organ identity by epigenetically repressing the expression of OsMADS58 in rice (Oryza sativa). New Phytol 206:1476–1490. https://doi.org/10. 1111/nph.13318 84. Yan D, Zhang X, Zhang L, Ye S, Zeng L, Liu J, Li Q, He Z (2015) CURVED CHIMERIC PALEA 1 encoding an EMF1-like protein maintains epigenetic repression of OsMADS58 in rice palea development. Plant J 82:12–24. https://doi.org/10.1111/tpj. 12784 85. Schultz EA, Pickett FB, Haughn GW (1991) The FLO10 gene product regulates the expression domain of homeotic genes AP3 and PI in Arabidopsis flowers. Plant Cell 3: 1221–1237. https://doi.org/10.1105/tpc. 3.11.1221 86. Bowman JL, Sakai H, Jack T, Weigel D, Mayer U, Meyerowitz EM (1992) SUPERMAN, a regulator of floral homeotic genes in Arabidopsis. Development 114:599–615. https:// doi.org/10.1242/dev.114.3.599 87. Sakai H, Medrano LJ, Meyerowitz EM (1995) Role of SUPERMAN in maintaining Arabidopsis floral whorl boundaries. Nature 378:199–203. https://doi.org/10.1038/ 378199a0 88. Krizek BA, Lewis MW, Fletcher JC (2006) RABBIT EARS is a second-whorl repressor of AGAMOUS that maintains spatial boundaries in Arabidopsis flowers. Plant J 45:369– 3 8 3 . h t t p s : // d o i . o r g / 1 0 . 1 1 1 1 / j . 1365-313X.2005.02633.x 89. Prunet N, Yang W, Das P, Meyerowitz EM, Jack TP (2017) SUPERMAN prevents class B gene expression and promotes stem cell termination in the fourth whorl of Arabidopsis thaliana flowers. Proc Natl Acad Sci U S A 114:7166–7171. https://doi.org/10.1073/ pnas.1705977114 90. Breuil-Broyer S, Trehin C, Morel P, Boltz V, Sun B, Chambrier P, Ito T, Negrutiu I (2016) Analysis of the Arabidopsis superman allelic series and the interactions with other genes

Flower Development in Monocots demonstrate developmental robustness and joint specification of male-female boundary, flower meristem termination and carpel compartmentalization. Ann Bot 117:905–923. https://doi.org/10.1093/aob/mcw023 91. Xu W, Zhu W, Yang L, Liang W, Li H, Yang L, Chen M, Luo Z, Huang G, Duan L et al (2022) SMALL REPRODUCTIVE ORGANS, a SUPERMAN-like transcription factor, regulates stamen and pistil growth in rice. New Phytol 233:1701–1718. https:// doi.org/10.1111/nph.17849 92. Malcomber ST, Kellogg EA (2005) SEPALLATA gene diversification: brave new whorls. Trends Plant Sci 10:427–435. https://doi. org/10.1016/j.tplants.2005.07.008 93. Zahn LM, Kong H, Leebens-Mack JH, Kim S, Soltis PS, Landherr LL, Soltis DE, DePamphilis CW, Ma H (2005) The evolution of the SEPALLATA subfamily of MADSbox genes: a preangiosperm origin with multiple duplications throughout angiosperm history. Genetics 169:2209–2223. https://doi. org/10.1534/genetics.104.037770 94. Dreni L, Ferra´ndiz C (2022) Tracing the evolution of the SEPALLATA subfamily across angiosperms associated with neo- and sub-functionalization for reproductive and agronomically relevant traits. Plan Theory 1 1 : 2 93 4 . h t t p s : // do i . o r g / 1 0. 33 9 0/ plants11212934 95. Melzer R, Theissen G (2009) Reconstitution of “floral quartets” in vitro involving class B and class E floral homeotic proteins. Nucleic Acids Res 37:2723–2736. https://doi.org/ 10.1093/nar/gkp129 96. Ditta G, Pinyopich A, Robles P, Pelaz S, Yanofsky MF (2004) The SEP4 gene of Arabidopsis thaliana functions in floral organ and meristem identity. Curr Biol 14:1935–1940. https://doi.org/10.1016/j.cub.2004. 10.028 97. Cui R, Han J, Zhao S, Su K, Wu F, Du X, Xu Q, Chong K, Theißen G, Meng Z (2010) Functional conservation and diversification of class e floral homeotic genes in rice (Oryza sativa). Plant J 61:767–781. https://doi. org/10.1111/j.1365-313X.2009.04101.x 98. Wu D, Liang W, Zhu W, Chen M, Ferra´ndiz C, Burton RA, Dreni L, Zhang D (2018) Loss of LOFSEP transcription factor function converts spikelet to leaf-like structures in rice. Plant Physiol 176:1646–1664. https://doi.org/10.1104/pp.17.00704 99. Ta KN, Sabot F, Adam H, Vigouroux Y, De Mita S, Ghesquie`re A, Do NV, Gantet P, Jouannic S (2016) MiR2118-triggered phased siRNAs are differentially expressed

81

during the panicle development of wild and domesticated African rice species. Rice 9:10. https://doi.org/10.1186/s12284-0160082-9 100. Hu Y, Liang W, Yin C, Yang X, Ping B, Li A, Jia R, Chen M, Luo Z, Cai Q et al (2015) Interactions of OsMADS1 with floral homeotic genes in rice flower development. Mol Plant 8:1366–1384 101. Prasad K, Sriram P, Santhosh Kumar C, Kushalappa K, Vijayraghavan U (2001) Ectopic expression of rice OsMADS1 reveals a role in specifying the lemma and palea, grass floral organs analogous to sepals. Dev Genes Evol 211:281–290. https://doi.org/10. 1007/s004270100153 102. Khanday I, Das S, Chongloi GL, Bansal M, Grossniklaus U, Vijayraghavan U (2016) Genome-wide targets regulated by the OsMADS1 transcription factor reveals its DNA recognition properties. Plant Physiol 172:372–388. https://doi.org/10.1104/ pp.16.00789 103. Dai Z, Wang J, Zhu M, Miao X, Shi Z (2016) OsMADS1 represses MicroRNA172 in elongation of palea/lemma development in rice. Front Plant Sci 7:1891. https://doi.org/10. 3389/fpls.2016.01891 104. Khanday I, Ram Yadav S, Vijayraghavan U (2013) Rice LHS1/OsMADS1 controls floret meristem specification by coordinated regulation of transcription factors and hormone signaling pathways. Plant Physiol 161:1970– 1983. https://doi.org/10.1104/pp.112. 212423 105. Ren D, Rao Y, Leng Y, Li Z, Xu Q, Wu L, Qiu Z, Xue D, Zeng D, Hu J et al (2016) Regulatory role of OsMADS34 in the determination of glumes fate, grain yield, and quality in rice. Front Plant Sci 7:1853. https:// doi.org/10.3389/fpls.2016.01853 106. Agrawal GK, Abe K, Yamazaki M, Miyao A, Hirochika H (2005) Conservation of the E-function for floral organ identity in rice revealed by the analysis of tissue cultureinduced loss-of-function mutants of the OsMADS1 gene. Plant Mol Biol 59:125– 135. https://doi.org/10.1007/s11103005-2161-y 107. Lim J, Moon YH, An G, Jang SK (2000) Two rice MADS domain proteins interact with OsMADS1. Plant Mol Biol 44:513–527. h t t p s : // d o i . o r g / 1 0 . 1 0 2 3 / A:1026517111843 108. Li H, Liang W, Jia R, Yin C, Zong J, Kong H, Zhang D (2010) The AGL6-like gene OsMADS6 regulates floral organ and

82

Ludovico Dreni

meristem identities in rice. Cell Res 20:299– 313. https://doi.org/10.1038/cr.2009.143 109. Sang X, Li Y, Luo Z, Ren D, Fang L, Wang N, Zhao F, Ling Y, Yang Z, Liu Y et al (2012) CHIMERIC FLORAL ORGANS1, encoding a monocot-specific MADS box protein, regulates floral organ identity in rice. Plant Physiol 160:788–807. https://doi.org/10. 1104/pp.112.200980 110. Rijpkema AS, Zethof J, Gerats T, Vandenbussche M (2009) The petunia AGL6 gene has a SEPALLATA-like function in floral patterning. Plant J 60:1–9. https://doi.org/10. 1111/j.1365-313X.2009.03917.x 111. Dreni L, Zhang D (2016) Flower development: the evolutionary history and functions of the AGL6 subfamily MADS-box genes. J Exp Bot 67:1625–1638. https://doi.org/10. 1093/jxb/erw046 112. Viaene T, Vekemans D, Becker A, Melzer S, Geuten K (2010) Expression divergence of the AGL6 MADS domain transcription factor lineage after a core eudicot duplication suggests functional diversification. BMC Plant Biol 10:148. https://doi.org/10.1186/ 1471-2229-10-148 113. Moon YH, Kang HC, Jung JY, Jeon JS, Sung SK, An G (1999) Determination of the motif responsible for interaction between the rice APETALA1/AGAMOUS-LIKE9 family proteins using a yeast two-hybrid system. Plant Physiol 120:1193–1203. https://doi.org/ 10.1104/pp.120.4.1193 114. Favaro R, Immink RGH, Ferioli V, Bernasconi B, Byzova M, Angenent GC, Kater M, Colombo L (2002) Ovule-specific

MADS-box proteins have conserved proteinprotein interactions in monocot and dicot plants. Mol Gen Genomics 268:152–159. https://doi.org/10.1007/s00438-0020746-6 115. Lee S, Jeon JS, An K, Moon YH, Lee S, Chung YY, An G (2003) Alteration of floral organ identity in rice through ectopic expression of OsMADS16. Planta 217:904–911. https://doi.org/10.1007/s00425-0031066-8 116. DePamphilis CW, Palmer JD, Rounsley S, Sankoff D, Schuster SC, Ammiraju JSS, Barbazuk WB, Chamala S, Chanderbali AS, Determann R et al (2013) The Amborella genome and the evolution of flowering plants. Science 342:1241089. https://doi.org/10. 1126/science.1241089 117. Gramzow L, Weilandt L, Theißen G (2014) MADS goes genomic in conifers: towards determining the ancestral set of MADS-box genes in seed plants. Ann Bot 114:1407– 1429 118. Hu Y, Wang L, Jia R, Liang W, Zhang X, Xu J, Chen X, Lu D, Chen M, Luo Z et al (2021) Rice transcription factor MADS32 regulates floral patterning through interactions with multiple floral homeotic genes. J Exp Bot 72:2434–2449. https://doi.org/10.1093/ jxb/eraa588 119. Wang H, Zhang L, Cai Q, Hu Y, Jin Z, Zhao X, Fan W, Huang Q, Luo Z, Chen M et al (2015) OsMADS32 interacts with PI-like proteins and regulates rice flower development. J Integr Plant Biol 57:504– 513. https://doi.org/10.1111/jipb.12248

Chapter 4 Model Species to Investigate the Origin of Flowers Charles P. Scutt Abstract The angiosperms, or flowering plants, arose at least 135 million years ago (Ma) and rapidly diversified to form over 300,000 species alive today. This group appears, however, to have separated from its closest living relatives, the extant gymnosperms, much earlier: over 300 Ma. Representatives of basally-diverging angiosperm lineages are of key importance to studies aimed at reconstructing the most recent common ancestor of living angiosperms, including its morphological, anatomical, eco-physiological and molecular aspects. Furthermore, evo-devo comparisons of angiosperms with living gymnosperms may help to determine how the many novel aspects of angiosperms, including those of the flower, first came about. This chapter reviews literature on the origin of angiosperms and focusses on basally-diverging angiosperms and gymnosperms that show advantages as potential experimental models, reviewing information and protocols for the use of these species in an evo-devo context. The final section suggests a means by which data from living and fossil groups could be integrated to better elucidate evolutionary events that took place on the long stem-lineage that apparently preceded the radiation of living angiosperms. Key words ANA grade, Magnoliids, Gymnosperms, Origin of angiosperms, Origin of flowers, Evodevo, Amborella trichopoda, Nymphaea thermarum, Trithuria, Austrobaileyales

1

Introduction

1.1 What Is a Flower?

For this chapter, flowers are defined as the sexual reproductive axes of the flowering plants, or angiosperms. In most angiosperms, flowers take the form of compact, bisexual axes that bear ovulecontaining carpels centrally, surrounded in turn by pollenproducing stamens and sterile perianth organs: typically petals and sepals. This bauplan is sometimes reduced or simplified, such as in the many dioecious or monoecious species that contain unisexual flowers, or in taxa that lack a perianth, such as Salix (willows) and Ulmus (elms). In extreme cases, the flower may be reduced to a single floral organ, as in some Chloranthaceae.

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_4, © Springer Science+Business Media, LLC, part of Springer Nature 2023

83

84

Charles P. Scutt

1.2 The Rise of Flowering Plants: When, Where, and Why?

The flowering plants are the latest major clade of land plants to emerge. A recent molecular clock-based study by Barba-Montoya et al. [1] gives five possible ranges of dates for the most recent common ancestor (MRCA) of living flowering plants, depending on the fossil calibration strategy used, which correspond to a combined range of 256–149 million years ago (Ma). The more recent end of this range, near the start of the Cretaceous (~145 Ma), is perhaps not too incongruent with the oldest known fossil pollen grains of clear angiosperm affinity, which date from around 135 Ma [2]. However, recent molecular-clock studies that give estimates for the radiation of living angiosperms long before the start of the Cretaceous (e.g., 214 Ma [3], 275 Ma [4], and the earlier part of the range given by Barba-Montoya et al. [1]) are clearly highly incongruous with the pollen fossil evidence. The reasons for this discrepancy are not yet entirely clear, but molecular studies that give very early dates for the MRCA of living angiosperms may have been biased by factors related to the origin of very large clades [5]. Meanwhile, the possible angiosperm affinities of numerous Jurassic flower-like macrofossils, including the recently discovered Nanjinganthus from 174 Ma [6], have been widely called into question [7–9]. Indeed, the oldest known convincing fossilized angiosperm flower dates from only ~125 Ma, and has been interpreted as a relative of the living genus Ceratophyllum [10], which is likely sister to eudicots (see Fig. 1). The geographic radiation and evolutionary diversification of the flowering plants can be traced efficiently through the appearance of novel pollen types, as discussed by Coiro et al. [8]. The earliest known angiosperm pollen is monosulcate: having one pore for the emergence of the pollen tube. This type of pollen first appears in N. Africa, the Middle East and Western Europe, territories which, in the early Cretaceous, were centered around northern Gondwana and situated at tropical latitudes of the northern hemisphere. However, within approximately ten million years, novel pollen types, including the tricolpate pollen characteristic of eudicots (the group comprising the majority of living angiosperms) begin to appear in sediments expanding away from the northern paleotropics. These detailed paleogeographic data lend strong support to a late Jurassic/early Cretaceous origin of angiosperms in Northern Gondwana, followed by a rapid geographic radiation and biological diversification. According to most recent molecular phylogenies (e.g., Refs. [1, 11]), the closest living relatives of the angiosperms comprise a clade containing all extant gymnosperms (see Fig. 1). Whatever absolute date-estimates such studies propose, they consistently indicate that the angiosperm and gymnosperm lineages separated considerably before the most recent common ancestor of living angiosperms. Accordingly, the angiosperms appear to have radiated

Model Species to Investigate the Origin of Flowers

85

Fig. 1 Experimental features of basally-diverging angiosperms and gymnosperms. Useful features (see key) for experimental purposes are listed after boxed species names. The schematic phylogeny is based on Doyle [95] and Byng et al. [16], but collapses Gnetales, Pinaceae and other conifers to a polytomy. The placement of putative angiosperm stem-lineage relatives (shown in grey) from the fossil record is based on Doyle [95]

from a stem lineage of perhaps some 150 million years in length, which means there are no living close relatives of the flowering plants that could help elucidate the origin of the flower. Evidence exists that a whole genome duplication, termed epsilon, preceded the diversification of living angiosperms [12], although a study by Zwaenepoel and Van der Peer [13] has questioned this finding. The extra sequences generated in the epsilon duplication (if it occurred) may have formed the raw material for the large-scale neofunctionalization of developmental regulators and thereby enabled the rapid evolution of angiosperm-specific features, including those of the flower. Several biotic and abiotic factors have been proposed to have played a role in the rapid initial expansion and diversification of the angiosperms, as reviewed by Willis and McElwain [2]. These include coevolution with insect pollinators, more efficient photosynthesis as a response to falling atmospheric CO2 concentration, a reduction in genome size, faster growth rates, and a shorter juvenile phase. This proposed acceleration of the angiosperm life cycle may have enabled these plants to adapt to new ecological niches formed

86

Charles P. Scutt

in shaded, damp and disturbed habitats [14], or even to more efficiently escape from newly-evolved groups of low-browsing, herbivorous dinosaurs, a hypothesis critically discussed by Barrett and Willis [15]. 1.3 A Family Tree of the Flowering Plants

The phylogenetic tree of living flowering plants is remarkably asymmetrical [16]. It contains a grade of three lineages, termed the ANA grade after the initials of its component orders Amborellales, Nymphaeales, and Austrobaileyales, which diverge basally from a remaining angiosperm lineage (see Fig. 1). The ANA grade contains a total of only around 200 living species, while the remaining angiosperm lineage has diversified to form a clade of over 300,000 living species, termed the euangiosperms or mesangiosperms. The euangiosperms include an early-diverging magnoliid clade of five orders, which contain a total of around 11,000 species. The remaining euangiosperms include the two major clades of eudicots and monocots, which together contain the vast majority of living angiosperms. The small aquatic order Ceratophyllales, containing only Ceratophyllum, is probably sister to eudicots.

1.4 A Portrait of the Ancestral Flower

Numerical reconstruction methods, based on present-day character states, have enabled the reconstruction of many features of the MRCA of living angiosperms. According to these analyses, this ancestral species was probably a woody plant, perhaps a scrambling shrub with possible liana-like tendencies and a relatively short life cycle, that grew in shaded and disturbed environments such as the banks of fast-flowing streams in dense forest [14]. Its flowers were probably small [17], bisexual [18], protogynous [19], actinomorphic, and contained an undifferentiated perianth of tepals [18– 20]. Sauquet et al. [18] suggest that the perianth and androecium of the MRCA of living angiosperms were whorled, but that its gynoecium was of spiral phyllotaxy. However, other authors have argued that such transitions in symmetry between the androecium and gynoecium are not found in present-day angiosperms and may be prevented by developmental constraints [21, 22]. The gynoecium of the MRCA of living angiosperms was almost certainly superior and contained several free, ascidiate (bottleshaped) carpels. These carpels probably contained an aperture or canal at the apex through which pollen tubes could grow, which was filled by substances secreted from the adjacent cells [23]. Each of these carpels probably contained a single pendant ovule, or a small number of such ovules [19], and these were likely to have been of either anatropous or orthotropous symmetry [24]. The ovules of the MRCA of living angiosperms almost certainly possessed two integuments and an extensive nucellus, or maternal nutritive tissue. These ovules also contained an embryo sac which probably consisted of four cells, and double fertilization would have led to the production of a zygote and a biparental endosperm, both of which were diploid [25].

Model Species to Investigate the Origin of Flowers

2

87

Models and Molecular Approaches to Study the Origin of Flowers Most of what is currently known of the molecular mechanisms of flower development derives from the study of well-adapted plant models from the eudicots and monocots, examples of which are given in Fig. 1. To better understand the origin and early evolution of flowers, a wider spectrum of model species is required, chosen for their key phylogenetic positions. Of particular importance are the living lineages that diverged before and shortly after the initial radiation of the living angiosperms, which correspond to the gymnosperms and basally-diverging angiosperms, respectively. Models chosen for molecular studies should be amenable to standard molecular biology procedures such as nucleic acid extraction and RNA in situ hybridization. It is helpful if study species can be cultivated easily, providing simple access to material at all developmental stages for laboratory work. It is also a huge advantage if the species chosen are amenable to functional-genetic approaches such as the production and long-term storage of mutant collections, and/or stable genetic transformation and the use of RNAi and gene-editing technologies [26]. Recent advances using developmental transcription factors to facilitate transformation and gene-editing in a wide range of species may prove particularly useful [27]. Most functional-genetic approaches are much simplified in species that are diploid, rather than polyploid. Large-scale approaches typically further require plants to be self-fertile, have short generation times and small physical size at maturity. The species chosen should ideally produce copious amounts of seed with orthodox storage characteristics and simple germination requirements. An alternative functional-genetic approach is available in species amenable to Virus Induced Gene Silencing, in which gene knock-downs can be performed using transgenic viruses modified to contain a fragment of a plant gene of interest [28]. It was previously a significant advantage for models to have small genomes, which greatly facilitated whole-genome sequencing and assembly, although recent technical advances have made possible the sequencing and assembly of very large genomes such as those of numerous gymnosperms. Nonetheless, compact genomes with relatively short intergenic regions and introns remain advantageous in model species. Long intergenic regions can complicate both in-vitro and in-vivo analyses, partly because it is difficult to estimate, in such circumstances, the size of functionally significant upstream and/or downstream cis-regulatory regions. Similarly, very long introns (e.g., an average of 30.8 kb in the recently sequenced Cycas gymnosperm [29]), which may or may not contain important cis-regulatory sequences, inevitably complicate the use of entire gene sequences in experiments involving plant transformation. Fixed publications such as book chapters are not very

88

Charles P. Scutt

Table 1 Online databases of genomic and transcriptomic resources and other information useful for studies of the origin of flowers Name

Web address

Description

PlaBiPD

https://www.plabipd.de/

Continually updated listings of published plant genomes, presented as cladograms with links to the relevant publications, among other resources

OneKP Database [96]

https://db.cngb.org/ onekp/

Transcriptomes of >1000 plant species available for downloading or BLAST searching

CoGe

https://genomevolution. org/coge/

Comparative genomics platform covering numerous plant genomes and containing many useful features

PHYTOZOME

https://phytozome-next. jgi.doe.gov/

Comparative plant genomics platform with many useful features for data-searching and bulkdownloading, etc.

PLAZA

https://bioinformatics.psb. ugent.be/plaza/

Comparative plant genomics platform with many useful features including micro-synteny analyses and phylogenetic pipeline construction. Separate instances for dicots, monocots, gymnosperms, etc.

PlantGenie

https://plantgenie.org/

Plant genomics/transcriptomics platform including several gymnosperms

World Flora Online

http://www. worldfloraonline.org/

Authoritative and comprehensive resource for plant taxonomy

Angiosperm Phylogeny Group [16]

https://www.mobot.org/ MOBOT/research/ APweb/

Authoritative, comprehensive, periodically updated consensus view of angiosperm phylogeny

Kew C-value Database

https://cvalues.science.kew. World-leading database of plant genome size org/ estimations

Kew Millennium Seed Bank Database

https://data.kew.org/sid/ sidsearch.html

World-leading database of seed dormancy types and optimum storage conditions

TimeTree

http://www.timetree.org/

Convenient resource to generate dated species phylogenies

efficient means for cataloguing genomic or transcriptomic resources. Therefore, besides the resources and information specifically mentioned in this chapter, the reader is directed to regularly updated web-sites dedicated to genomics, transcriptomics, and the collation of other useful information, such as those listed in Table 1. Species that are not readily amenable to functional genetics can be studied using a wide range of in-vitro and heterologous in-planta methods. These include the complementation-testing of mutants in well-established model systems using orthologous coding sequences from species of interest [30], the study of

Model Species to Investigate the Origin of Flowers

89

protein-protein interactions by a wide range of methods, and the study of protein-DNA interactions using SELEX, Protein Binding Microarrays [31], or DAP-seq [32]. Conservation of epigenetic marks can also be compared between established models and other species chosen for their phylogenetic positions, conditionally on the availability of sequenced genomes [31]. The following sections detail some of the most promising plant models for study among basally diverging angiosperm and gymnosperm lineages. By comparing the molecular networks that control reproductive development in these two groups, it may be possible to identify key molecular changes that were responsible for the origin and early evolution of the flower and its component organs. 2.1

Amborellales

Amborella trichopoda (referred to below as Amborella) is the only known member of Amborellales and hence probable sister to all other living flowering plants [16] (see Fig. 1). This dioecious, scrambling shrub (see Fig. 2a), endemic to the understory of sub-tropical cloud forests in New Caledonia, has been widely used in studies aimed at establishing the pleisiomorphic features of living flowering plants and their underlying molecular mechanisms [33]. Likely pleisiomorphic features of the Amborella flower include its undifferentiated perianth of free tepals (see Fig. 2b, c) and its gynoecium of free, ascidiate carpels (see Fig. 2c), each of which contains an apical secretion-filled canal for pollen-tube growth (see Fig. 2d) and a pendent, bitegmic ovule. Probably derived features of Amborella include its dioecy [18, 32], which recent studies [34] indicate to be due to the presence of a pair of ZW sex chromosomes (ZW = female, ZZ = male) that likely originated less than 16.5 Ma, i.e., much more recently that the origin of flowering plants. The non-recombining, sex-determining region is located on Chromosome 9 of Amborella and extends over ~4 Mb [34]. Another derived feature of Amborella is its unique eight-celled embryo sac that gives rise, after double fertilization, to a triploid endosperm [35], this type of endosperm ploidy having apparently evolved twice independently in Amborellales and euangiosperms. Features of Amborella that correspond to ambiguous character states in the MRCA of living angiosperms include its nearorthotropous ovule symmetry [24] and entirely spiral floral phyllotaxy. Amborella can be propagated by seed or cuttings and thrives in a temperate, moist atmosphere at relatively low light-intensities [36]. Its seed, which is enclosed within a drupe, typically matures over several months. Germination of Amborella seed is furthermore subject to a period of morphophysiological dormancy of over 3 months, although this can be shortened by mechanical or chemical weakening of the pericarp [37]. Such morphophysiological seed dormancy is an inferred ancestral characteristic in angiosperms.

90

Charles P. Scutt

Fig. 2 ANA-grade angiosperms and putative angiosperm stem-lineage relatives. (a –d) Amborella trichopoda: 3-year-old plant in flower (a), male flower (b), female flower (c), and median longitudinal section of developing carpel (d). (e –f) Nymphaea thermarum: newly opened flower (e) and young plant (f). (g –h) Trithuria submersa: reproductive unit (g) and adult plants growing on agar (f). (i –k) Cabomba caroliniana: submerged leaves and stems (i), numbered succession of young floating leaves subtending flower buds present in the mass of tissues at the centre (j), newly opened flower (k). Illicium anisatum, mature flower (l). Caytonia (Caytoniales) female cupule-bearing rachis (m). Dictyopteridium (Glossopteridales), female phytomer (n). Williamsoniella (Bennettitales), flower-like bisexual reproductive axis in longitudinal section (o). (Panels a [98], b, c, g [125], i –k [59], and m –o [93] are reproduced with permission from published sources). b bract, c carpel, ca canal (for pollen tube growth), cu cupule, fl fertile leaf, m microsporophyll, o ovule, or ovuliferous receptacle, p bract-like phyllome, r rachis, t tepal. Scale bars: 10 cm in a; 1 mm in b, c, and g; 50 μm in d; 1 cm in e, f, h, and i –l

Although partially resistant to desiccation, Amborella seed loses viability through long-term storage, as reviewed by Poncet et al. [33]. Amborella is readily amenable to standard molecular-biology procedures such as in situ hybridization, and this technique has been used in several studies to assess conservation of functions of

Model Species to Investigate the Origin of Flowers

91

flower development genes since the MRCA of living angiosperms (e.g., Refs. [38–41]). However, Amborella’s seed characteristics, dioecious breeding system and woody habit make it poorly adapted to functional-genetic studies, and it seems relatively unlikely that a large-scale mutagenesis program could be undertaken in this species. Few data are available on tissue-culture in Amborella, and no protocols for its stable genetic transformation have yet been reported. The ~870 Mb-long genome of Amborella has been entirely sequenced [42], and more recently a chromosome-level assembly has been made available (Table 2). Interestingly, the individual used for genomic sequencing was phenotypically male, though had been reported occasionally to change sex during vegetative propagation [43] and is now known to be genetically female [34], carrying both the Z and W sex-chromosome haplotypes. Analyses of this genome sequence indicate that no whole genome duplications have occurred in the Amborella lineage since the hypothesized epsilon event. The phylogenetic position of Amborella, and the absence of lineage-specific duplications, make this genome a very important reference for the study of genome evolution in the angiosperms as a whole, and it has already made a major contribution to studies aimed at reconstructing gene-content and gene-order in the MRCA of living angiosperms [3]. 2.2

Nymphaeales

Nymphaeales, the probable second-earliest diverging ANA-grade order, form a widely distributed group consisting of three families of aquatic and semi-aquatic herbs: Nymphaeaceae (water-lilies), Cabombaceae and Hydatellaceae, of which the latter is basally diverging [16] (see Fig. 1). Nymphaeaceae have a cosmopolitan distribution and include the five genera Nymphaea, Nuphar, Barclaya, Victoria, and Euryale. However, a phylogeny by Lo¨hne et al. [44] indicates that Victoria and Euryale may form a clade nested within Nymphaea, suggesting a possible need for taxonomic revision. The genome sequence of the unique species of Euryale, E. ferox, has recently been published [45]. Cabombaceae include five species of Cabomba, which are native to the Americas, and the unique species of Brasenia, B. schreberi, which is found in N. America, Africa, and Asia, among other territories. Hydatellaceae contain the single genus Trithuria (>12 species [46, 47]), and are native to Australasia and the southern tip of India. Within Nymphaeaceae, several species or cultivars of Nuphar and Nymphaea have been used as evolutionary-developmental models to assess the conservation of expression patterns of developmental regulators with established model species (e.g., Refs. [39, 48]). In addition, genome sequences have been published from two Nymphaea species: N. colorata (see PlaBiDP, Table 1) and N. thermarum [49]. Of these, N. thermarum, perhaps

92

Charles P. Scutt

Table 2 Transcriptomic resources including reproductive tissues from basally-diverging angiosperms and gymnosperms, with corresponding genomic data where available Nuclear Nuclear genome Transcriptomic resources genome size sequence

Species

Taxonomy

Amborella trichopda

ANA-grade angiosperms, Amborellales, Amborellaceae

RNA-seq data of tepal, leaf, 01c = .89 pg Genome V1 [42] [99]; https:// root and egg-apparatus 870 Mb [42] phytozome-next. [97], male and female jgi.doe.gov/ flower buds [34], and Genome V6. laser-microdissected (chromosomefemale flower tissues level assembly) [98] https:// genomevolution. org/coge/

Nymphaea thermarum

ANA-grade angiosperms, Nymphaeales, Nymphaeaceae

RNA-seq of mature ovules 1c = 0.51 pg Povilus et al. [49] [101]; and developing seeds 497 Mb [49] [100]

Nymphaea ‘King ANA-grade of Siam’ angiosperms, Nymphaeales, Nymphaeaceae

RNA-seq of petals [102]

Nuphar advena

ANA-grade angiosperms, Nymphaeales, Nymphaeaceae

Flower EST microarray analyses [103]

1c = 2.78 pg [101]

Trithuria submersa

ANA-grade angiosperms, Nymphaeales, Hydatellaceae

RNA-seq of whole plants, including reproductive tissues [104]

1c = 1.37 pg [57], but probably octoploid [47]

Aristolochia fimbriata

Qin et al. [71] RNA-seq of dissected floral 1c = 0.50 Magnollids, [105]; organs and other tissues Piperales, 258 Mb [71] [71] Aristolochiaceae

Persea americana

Magnollids, Laurales, Lauraceae

Rendon-Anaya et al. Microarrays of flower ESTs 1c = 0.92 [75] [103] [106]; 840 Mb [75]

Liriodendron chinense

Magnollids, Laurales, Magnoliaceae

Petal and leaf transcriptome [107]; Flower-expressed microRNAs [108]

1.75 Gb [76] Chen et al. [76]

Pinus taeda

Gymnosperms, Pinaceae

RNA-seq of mixed tissues including female cones [109]; https:// plantgenie.org/

1c = 22.10 [110]; 11.6 Gb

https://plantgenie. org/

(continued)

Model Species to Investigate the Origin of Flowers

93

Table 2 (continued) Nuclear Nuclear genome Transcriptomic resources genome size sequence

Species

Taxonomy

Pinus tabuliformis

Gymnosperms, Pinaceae

RNA-seq and microarrays 1c = 25.70 [111]; of male and female cone 25.4 Gb stages [85] [112]

Pinus koraiensis

Gymnosperms, Pinaceae

RNA-seq of mixed tissues including cones [113]

Picea abies

Gymnosperms, Pinaceae

Transcriptome profiling of 1c = 20.01 female cones [86] [114]; 19.8 Gb [115]

Pseudotsuga menziesii

Gymnosperms, Pinaceae

RNA-seq of megagametophyte stages [116]

Cryptomeria japonica

Gymnosperms, Cupressaceae

RNA-seq including cone 1c = 11.05 tissues https://db.cngb. [118] org/onekp/

Platycladus orientalis

Gymnosperms, Cupressaceae

RNA-seq of female cones [119]

Cephalotaxus sinensis

Gymnosperms, Cephalotaxaceae

Ovule RNA-seq [120]

Ginkgo biloba

Gymnosperms, Ginkgoales, Ginkgoaceae

Ovule RNA-seq [121]; Ovule small RNA-seq [122]

1c = 9.39 [123]; 11.75 Gb [124]

Cycas Gymnosperms, panzhihuaensis Cycadales, Cycadaceae

RNA-seq of both reproductive and vegetative tissues [29]

10.5 Gb [29] Lui et al. [29]

Niu et al. [112]

1c = 28.20 [111]

1c = 19.05 [110]; 15.7 Gb [117]

https://plantgenie. org/; Nystedt et al. [115] Neale et al. [117]

1c = 10.46 [118]

Guan et al. [124]

uniquely in the ANA grade, regroups a wide range of characteristics that make it highly suitable as a molecular-genetic model. N. thermarum is the smallest known water lily, with adult plants measuring around 10–20 cm in diameter. The species was known from only one location in Rwanda, but is now considered extinct in the wild [50] and is conserved mainly in botanic gardens and research collections. Flower structure in N. thermarum has been described by Fischer and Magdalena Rodriguez [50], and by Povilus et al. [51]. Mature, open flowers of N. thermarum measure up to 2 cm in diameter (see Fig. 2e). These are of whorled phyllotaxy and contain 4 greenish outer tepals and 6–8 whitish inner tepals. The N. thermarum gynoecium consists of 7–9 carpels

94

Charles P. Scutt

which fuse together basally during flower development. Angiospermy (enclosure of the ovules) in all Nymphaea spp. is completed by post-genital fusion at the carpel margins, rather than by secretion of substances into an apical aperture or canal [23]. Numerous bitegmic, anatropous ovules form in each carpel of N. thermarum, each of which contains a four-celled embryo sac that contributes, after double fertilization, to produce a diploid endosperm [51]. The bisexual axis, bitegmic ovule, four-celled embryo sac and diploid endosperm of N. thermarum are likely pleisiomorphic features of living angiosperms, whereas the complete closure of its carpels by cellular structures, its partially syncarpic gynoecium, and its high number of ovules-per-carpel all appear to be derived features. Anatropous ovule symmetry in N. thermarum contrasts with the near-orthotropous symmetry in Amborella, and one of these two symmetry types is likely to be ancestral in living angiosperms. N. thermarum therefore possesses a list of likely pleisiomorphic features that is highly complementary to those of Amborella. N. thermarum is a perennial herb that can be grown from seed to flowering in around 2–3 months [50]. It thrives in a warm, humid environment under high light intensities and can be easily cultivated in pots of wet compost (see Fig. 2f) as the plants do not need to grow completely immersed in water. The small size of N. thermarum flowers and their component organs make these highly practical for in-situ hybridization and other microscopic procedures. A recent study included genetic crosses in N. thermarum, demonstrating the practicality of this species for such standard genetic procedures [52]. N. thermarum is self-fertile, and approximately 150 seeds are produced in each fruit. Optimal seed storage conditions have not yet been published, although seeds of some other water lilies are known to be of the orthodox type (Kew Millennium Seed Bank database, Table 1). N. thermarum seeds germinate rapidly under water, but the young seedlings, which are filiform in structure, should then be placed horizontally on the surface of a substrate, covered by a thin film of water, until the first true leaves appear [50]. Unlike some water lilies, N. thermarum cannot be propagated by rhizomes and no methods of vegetative propagation have yet been reported. The scientific literature on N. thermarum is currently very limited, but this is an obvious candidate for full-scale development as an ANA-grade angiosperm model. Its diploid genome is among the smallest in Nymphaeaceae and among the ANA-grade angiosperms as a whole (Table 2), and some tissue/stage-specific transcriptomic data are also available (Table 2). Growth of N. thermarum in tissue culture should be facilitated by the ready availability of seed that can easily be surface-sterilized and germinated in vitro, and further studies in this species may pave the way to the development of protocols for its stable genetic transformation.

Model Species to Investigate the Origin of Flowers

95

Trithuria, the only genus of Hydatellaceae, includes very small aquatic/semi-aquatic annuals and perennials [46]. These plants produce tufts of linear leaves, each containing a single vascular bundle, and apical reproductive units (RUs), also termed “nonflowers”, that consist of compact axes on which reproductive organs are surrounded by a small number of bract-like phyllomes [53]. The genus includes dioecious species in which male and female RUs occur on separate individuals, monoecious species in which both male and female RUs form on the same individuals, and bisexual species in which both carpels and stamens form on the same RUs. In bisexual Trithuria spp., the carpels generally form externally to the stamens (see Fig. 2g), and it is possible that the Trithuria RU represents an inflorescence, each flower of which is reduced to a single reproductive organ, as discussed by Rudall et al. [53]. Trithuria appears therefore to have either a highly derived or radically simplified floral structure, compared to the reconstructed MRCA of living angiosperms. Trithuria’s carpels, however, appear relatively underived. These occur separately, are acidiate in shape, contain a short apical canal for pollen growth [54], and enclose a single pendant, bitegmic, anatropous ovule. The Trithura ovule includes a large perisperm, derived from the nucellus, which has been suggested to be a relictual character that has persisted from gymnosperm-like ancestors [55]. The embryo sac structure and endosperm ploidy in Trithuria [55] are similar to those of N. thermarum, described above. As in Amborella, the likely pleisiomorphic condition of morphophysiological seed dormancy is present in Trithuria [56], in contrast to the more recently-evolved physiological dormancy present in other Nymphaeales [37]. The relatively underived structure of the individual floral organs of Trithuria make this a potentially useful model for the origin and early evolution of angiosperms, whereas its highly derived organization into RUs make it an essential model for studies of the evolution of apparently “inside-out” flower-like structures, which are very rare among the angiosperms. Few molecular data have yet been reported in Trithuria, and the tractability of these species as potential molecular-genetic models remains uncertain. Plants of the bisexual T. submersa can easily be grown to maturity under in-vitro conditions [57] (see Fig. 2h), although seed production under these conditions has not been reported. T. konkanensis has also been grown in vitro [58]. Genome sizes in Trithuria are modest (Kew Millennium Seed Bank database, Table 1), though T. submersa, which might otherwise make a potential genetic model, appears to be octoploid [47], a feature which would likely complicate any attempt at functional genetics. A trancriptomic analysis of entire T. submersa plants has been performed (Table 2), although no full genomic sequence has yet been published.

96

Charles P. Scutt

Cabomba is probably better adapted to represent Cabombaceae in molecular-based evo-devo studies than its sister taxon Brasenia schreberi, as Cabomba flowers are the simpler in structure, and because the leaves, stems and floral organs of B. schreberi are coated in a layer of highly viscous mucilage, which might complicate some molecular-biology procedures. Cabomba plants form horizontal rhizomes and sparsely branched submerged stems that give rise to submerged leaves that are highly dissected to reduce water-drag (see Fig. 2i). Adventitious roots can emerge at the submerged leaf nodes [59]. After induction to flowering, however, a series of floating leaves are produced at the Cabomba stem apex (see Fig. 2j), each of which subtends a floral bud. These floating leaves are sagittate to orbicular in shape, with entire margins, and function to support the subtended flowers at the water’s surface. Cabomba flowers contain a largely undifferentiated perianth of two trimerous whorls of tepals, a doubled trimerous androecium of six stamens, and a gynoecium, typically of two or three free carpels (see Fig. 2k). The Cabomba carpel includes the likely pleisiomorphic angiosperm features of an ascidiate shape, an apical canal for pollen tube growth, and a low number of pendant, bitegmic ovules [23]. These ovules are anatropous in symmetry. Embryo sac structure and endosperm ploidy have not been investigated in Cabombaceae. Cabomba plants can be grown without difficulty in a mediumsized aquarium and induced to flower by adding far-red light (~740 nm, from light-emitting diodes, for example) to standard lighting for aquarium plants [59]. Vegetative propagation of Cabomba can be easily performed by planting bunches of cut stems in the aquarium substrate to encourage the production of adventitious roots. C. caroliniana, the most temperate species of the genus, is a significant invasive weed in many regions and is now subject to strict controls in several jurisdictions [60]. However, less hardy species such as C. aquatica may substitute for C. caroliniana in evo-devo studies. Cabomba flowers and other tissues are readily amenable to standard molecular biology procedures such as in situ hybridization (e.g., Refs. [38, 59, 61]). A modest-sized Cabomba flower EST database is available [59], although no large-scale RNAseq data have been produced. Genetic transformation has not been reported in Cabomba, while its aquatic habit, fragile stems and uncharacterized seed-storage characteristics make the genus unlikely to be suitable for large-scale molecular-genetic procedures such as the generation of mutant collections. 2.3

Austrobaileyales

Austrobaileyales, the probable third-earliest diverging ANA-grade order, includes three families of woody plants containing a total of around 100 species in five genera [16]. The most basally-diverging of these families, Austrobaileyaceae, contains only Austrobaileya scandens, a liana endemic to Northern Queensland. Trimeniaceae

Model Species to Investigate the Origin of Flowers

97

contains only the small genus Trimenia, whose members are small trees, shrubs, or lianas, and are native to Eastern Australia, SE Asia, and islands in the Pacific. Schisandraceae, the third family of Austrobaileyales, contains the three small-to-medium sized genera Illicium (see Fig. 2l), Schisandra and Kadsura, whose members are shrubs or lianas, native mainly to SE Asia, although with some species present in N. America and the Caribbean. Austrobaileyales include plants with bisexual and/or unisexual flowers and a range of breeding systems (discussed by Endress [17]), including protogyny, e.g., in Austrobaileya scandens and Illicium spp.; andromonoecy, e.g., in Trimenia spp.; dioecy, e.g., in Schisandra chinensis; mixed monoecy and dioecy, e.g., in Kadsura japonica; and self-incompatibility (SI) in Illicium floridanum [62], Austrobaileya scandens [63], and Trimenia moorii [64]. Of these various breeding systems, bisexuality with protogyny is thought to be the pleisiomorphic condition of living angiosperms, and SI may also have been present very early in angiosperm evolution [65]. Indeed, almost all of the likely pleisiomorphic features of flowers can be found in Austrobaileyales, including bisexuality (e.g., Austrobaileya scandens and Illicium), an undifferentiated perianth of tepals (all taxa), an unfused gynoecium of carpels, each containing a canal or aperture for pollen tube growth (all taxa except Illicium), and a single or low number of bitegmic ovules (all taxa). The presumed pleisiomorphic arrangement of a 4-celled embryo sac and diploid endosperm is known to occur in Illicium [66] and may be present throughout Austrobaileyales. An interesting and potentially ancient phenomenon revealed in Trimenia shows that multiple embryo sacs may persist in each ovule, and compete for fertilization by growing towards the oncoming pollen tubes [67]. Austrobaileyales have spiral floral phyllotaxy (although this is pseudo-whorled in Illicium), which may or may not be a pleisiomorphic character in angiosperms (as discussed in the Introduction, above). Austrobaileyales have been extensively included in analyses to reconstruct morphological character states in the MRCA of living angiosperms. However, members of this order have been little used to investigate the molecular bases of flower evolution. Their woody habit makes these plants, like Amborella, relatively poorly adapted as potential molecular-genetic models, and there is no published literature on methods for their stable genetic transformation. Some Austrobaileyales, such as Austrobaileya scandens, are known to flower unpredictably in cultivation, though others, such as species of Schisandra, Kadsura and Illicium, are ornamental or commercially important plants that flower reliably and can be readily obtained from botanic gardens or commercial sources. Genome sizes of Austrobaileyales are larger than that of Amborella and much larger than those of most Nymphaeales [59], and no genome of this order has yet been sequenced. Transcriptomic

98

Charles P. Scutt

resources are, however, currently available for Austrobaileya scandens, Illicium spp., Kadsura heteroclita (OneKP database, Table 1), and Schisandra spp. [68, 69]. It is likely, as studies of ANA-grade angiosperms progress, that more attention will be focused on Austrobaileyales. Despite their large genomes and lack of ideal features as molecular-genetic models, many of these species retain a high proportion of likely pleisiomorphic angiosperm characters and, as woody, terrestrial plants, may have undergone a more conservative evolutionary history than did the aquatic and herbaceous Nymphaeales [17]. In addition, Amborella, the only other woody, non-aquatic member of the ANA-grade, is likely to be derived in numerous respects (described above), underlining the need for multiple models and a comparative approach in evolutionary analyses. 2.4

Magnoliids

The magnoliids form a diverse clade of around 11,000 species, classified in five orders (see Fig. 1), and include both woody and herbaceous taxa [16]. The magnoliid stem lineage emerges within the living angiosperm tree, shortly after those of the ANA grade (see Fig. 1). Herbaceous magnoliids might make more convenient, if slightly less basally-diverging models than the woody plants or fully aquatic herbs that make up most of the ANA grade. In addition to potentially providing further insights into the origin of angiosperms, magnoliid models can be expected to constitute an external reference to better reconstruct the origin of the later-emerging major clades of the eudicots and monocots (see Fig. 1). One species suggested as a promising herbaceous magnoliid model is Aristolochia fimbriata [70]. Aristolochia (Piperales, Aristolochiaceae; also known as pipevines) is a large genus containing both woody and herbaceous species. A. fimbriata is a vine that attracts pollinating flies into a chamber formed by its bilaterally symmetrical perianth of fused sepals. Bilateral perianth symmetry is a novel feature that has arisen several times independently in the angiosperms and may have occurred first within magnoliids [70]. Other floral features of Aristolochia that are distinct from those of earlier-diverging lineages include its inferior ovary, in which the stamens are developmentally fused to the style to form a gynostemium. A. fimbriata is highly suitable as a model magnoliid, in part for its small size at maturity, rapid life cycle, self-compatibility and small genome [70] (Table 2), which is now entirely sequenced [71]. Interestingly, A. fimbriata is thought, like Amborella, to have undergone no lineage-specific whole genome duplication event since the epsilon duplication at the base of the angiosperm cade. Crucially for its use as a model basal angiosperm, tissue culture procedures have been established in A. fimbriata [72], leading to protocols for its stable genetic transformation using Agrobacterium tumefaciens [70]. A. fimbriata is readily amenable to standard molecular procedures such as in-situ hybridization, and

Model Species to Investigate the Origin of Flowers

99

this technique has already been used to study the expression of numerous classes of MADS-box floral homeotic genes [73]. Interactions between the corresponding proteins have also been studied in yeast-two-hybrid experiments [74]. In addition to that of A. fimbriata, 13 further magnoliid genome sequences are now available, details of which are given on the PlaBiDB database (Table 1). Among the first magnoliid species with published genomes were Persea americana (Laurales, Lauraceae; avocado; [75]), a woody crop species of high economic importance, and Liriodendron chinense (Magnoliales, Magnoliaceae; Chinese tulip tree; [76]), an ornamental tree from E. Asia. In contrast to A. fimbriata, analyses of the P. americana genome suggest the presence of two lineage-specific whole genome duplications that occurred subsequently to the hypothesized epsilon duplication [75], one of which appears to be shared with the L. chinense lineage [29, 76]. An analysis of the expression patterns of P. americana MADS-box floral regulators has been performed [39], while numerous floral transcriptomic resources are available for L. chinense and its unique sister species L. tulipifera (Table 2). 2.5

Gymnosperms

The gymnosperms form the closest living out-group to the flowering plants and therefore constitute a vital external reference to understand the evolutionary events that contributed to the flower and other angiosperm-specific traits. The reproductive structures of gymnosperms differ from most angiosperm flowers by, among other features, being unisexual and lacking a perianth. Gymnosperm ovules have a single integument, rather than the two integuments present in most angiosperm groups, and are not enclosed within carpels. However, further tissue layers surround the ovules in certain gymnosperms (e.g., Taxus, Juniperus, and Gnetales), forming fleshy structures that persist around the seed. Pollination in gymnosperms occurs mostly by wind, although some Cycadales and Gnetales are pollinated by both wind and insects [77]. Double fertilization to produce a biparental endosperm is absent in gymnosperms, although a supernumerary zygote is formed through a second fertilization event in Gnetales [78]. Most recent molecular phylogenies of seed plants indicate the extant gymnosperms to be sister to angiosperms, although two studies, dating from some years ago, suggested an alternative topology in which a clade containing Cycadales and angiosperms was sister to remaining gymnosperms [79, 80]. By contrast, the internal topology of the gymnosperm clade remains unresolved, and several alternative topologies have been proposed (reviewed by Doyle [81]). Most of these topologies include Cycadales and Ginkgoales in basally-diverging positions relative to a remaining clade containing all other living gymnosperms, as shown in Fig. 1. Within this remaining gymnosperm clade, however, the position of the small order Gnetales appears highly labile, as various studied have placed

100

Charles P. Scutt

this as sister to Pinaceae, sister to all conifers, or sister to all conifers excluding Pinaceae. Some phylogenies have even shown Gnetales as basally-diverging within living gymnosperms (e.g., Ref. [82]), although such conclusions may reflect bias due to the inclusion of particularly fast-evolving Gnetales sequences in phylogenomic analyses (see Zhong et al. [83]). Topologies showing Gnetales in a basal position in gymnosperms have also been criticized as inconsistent with the fossil record, which indicates a relatively recent origin of this group [81]. The lack of complete phylogenetic resolution in gymnosperms leads to problems for their classification into higher taxa. A classification has, however, been proposed which circumvents this problem by placing several gymnosperm families, including Pinaceae and the three families within Gnetales (see Fig. 1), within separate monotypic orders [84]. Gymnosperms form a vital external reference to enable the reconstruction of the molecular-evolutionary processes that generated the angiosperm flower. One potentially useful aim might be to reconstruct gene-order in the genome of the MRCA of living seed plants (angiosperms+gymnosperms), from which the angiosperm stem lineage originally emerged. This goal would no doubt benefit from a representative taxonomic sampling of genomes, and indeed complete genome sequences are now available from all major gymnosperm clades. Some of these genomes are listed in Table 2, along with transcriptomic resources in reproductive tissues, while publications describing the genomes of Abies alba, Sequoia sempervirens, Sequoiadendron giganteum, Taxus chinensis and T. walichiana (conifers), Gnetum montanum and Welwitschia mirabilis (Gnetales), and Gingko biloba (Ginkgoales), are given in the PlaBiDB database (Table 1). The recent publication of a chromosome-level assembly of the Cycas panzhihuaensis genome [29] has revealed numerous interesting features, including a likely whole-genome duplication at the base of the living gymnosperm clade and a potentially common molecular basis of chromosomal sex-determination shared between Cycas and Gingko. Transcriptomic resources that include reproductive tissues are currently available from a number of gymnosperms (Table 2). Of particular interest, analyses of a complete developmental series of male and female cones have been performed in Pinus tabuliformis [85], while a recently developed spatially-resolved transcriptome profiling technique has been successfully applied to female cones of Picea abies [86]. Gymnosperm orthologs of numerous flower development genes have been studied using a wide range of methods. Major elements of conservation between mechanisms of development in gymnosperm cones and angiosperm flowers have been concluded from in situ hybridization data (e.g., Refs. [87, 88]) protein-DNA interaction data (e.g., Refs. [88, 89]) and from the

Model Species to Investigate the Origin of Flowers

101

complementation of angiosperm floral homeotic mutants using orthologous sequences from gymnosperms (e.g., Refs. [87, 90]). One practical problem for transcriptomics relates to the slow growth of many gymnosperm reproductive tissues, which can complicate sampling procedures. However, certain taxa such as Welwitschia mirabilis (Gnetales, Welwitschiaceae) possess a developmental series in each cone, with younger stages occurring nearer the apex [88], while others such as Ephedra spp. (Gnetales, Ephedraceae) complete their reproductive development relatively quickly, facilitating sampling from all developmental stages in each reproductive season. Sampling from Picea abies is complicated by the form of the adult trees, on which cones typically emerge at a height of several meters on slender and fragile branches. However, the acrocona mutant of P. abies [91] produces cones at low level and can be grown from seed to reproductive maturity in under 3 years. The molecular basis of the acrocona mutation is not yet known, although the wild-type locus may act to downregulate a specific MADS-box gene [92]. Numerous gymnosperms are amenable to genetic transformation using Agrobacterium tumefaciens through the generation of somatic embryos (reviewed by Uddenberg et al. [91], see Fig. 1). However, transformation methods have not yet been used to analyze the roles of genes controlling reproductive development in gymnosperms, probably due to the typically long juvenile phase of these plants (e.g., over 25 years in wild-type P. abies). It is possible, however, that the acrocona mutant of P. abies, and perhaps in the future equivalent mutants in other gymnosperms [91] may provide ways to speed-up this process. The use of gymnosperms with shorter juvenile phases, such as Welwitschia mirabilis and Ephedra spp., could also help to accelerate functional-genetic analyses of gymnosperm reproductive development.

3

Missing Links and New Approaches This chapter mainly concerns living species of basally diverging angiosperms and gymnosperms that can be used to better understand the origin of the flower. However, the MRCA of all living seed plants is separated from that of living angiosperms by a stem lineage of perhaps some 150 million years, from which no species have survived other than the angiosperms themselves. In the absence, therefore, of close living relatives of the flowering plants, the epsilon whole genome duplication may provide a point of reference to help reconstruct the lost ancestors of the angiosperms (notwithstanding the recent conclusions of Zwaenepoel and Van der Peer [13] on the absence of this duplication!). The combined analysis of molecular phylogeny and synteny in multiple angiosperm genomes (e.g., Ref. [3]), may soon enable the

102

Charles P. Scutt

systematic identification of paralogs whose gene lineages separated at the (hypothesized) epsilon duplication. Certain gene lineages are known, from model angiosperms, to be necessary for the development of angiosperm-specific characters such as the perianth, bisexual axis, carpel and outer ovule integument (reviewed by Scutt [93]). Protein sequences from nodes in any molecular phylogeny can be reconstructed using ancestral sequence reconstruction or “protein resurrection,” as discussed by Vialette-Guiraud et al. [31], and the corresponding genes and proteins can then be generated in the laboratory for both in-vitro and in-vivo studies [31]. Reconstructed regulators of reproductive development from the time of the epsilon duplication may, therefore, help to shed light on the morphological features of the plants in which these molecules functioned. Notably, information on the biochemical properties of ancestralized regulatory molecules, and a comparison of their rates and modes of sequence evolution from before and after the hypothesized epsilon event, may provide clues on the order-ofacquisition of novel angiosperm features, including the perianth, bisexual axis, carpel, and outer integument. Although no living, close relative of the angiosperms is available for study, there is an extensive fossil record of extinct gymnosperm groups, some of which show features that suggest affinities with angiosperms. Several groups of gymnosperms have been suggested as potential stem-lineage relatives of the angiosperms and tentatively placed, using phylogenetic analyses based on morphological features, on the seed-plant phylogenetic tree ([81, 94, 95], see Fig. 1). Interestingly, several of these potential angiosperm relatives show contrasting sets of angiosperm-like features. For example, Caytonia (Caytoniales) are fossilized female reproductive structures in which multiple unitegmic ovules are enclosed within cupules that are pinnately arranged on a radially symmetrical rachis (see Fig. 2m). The Caytonia cupule and rachis have been suggested to be potentially homologous to the angiosperm outer integument and carpel, respectively (reviewed by Doyle [95], and see also Shi et al. [94]). Like Caytoniales, Glossopteridales have unisexual female reproductive structures composed of ovule-containing cupules, although these typically form in the axil of a fertile leaf or sporophyll (see Fig. 2n). In the case of Glossopteridales, the cupule and sporophyll may be homologous to the angiosperm outer integument and carpel, respectively. Bennettitales, by contrast, do not possess cupules, but some species of this order have a bisexual axis that bears male organs externally to female organs, as in the angiosperm flower (see Fig. 2o). The reproductive axes of Bennettitiales also possess a perianth-like structure, giving them a distinctly flower-like appearance. The predicted order-of-acquisition of angiosperm-specific features will vary depending on which fossil gymnosperm is regarded as the closest relative to angiosperms. If, for example, Bennettitales

Model Species to Investigate the Origin of Flowers

103

were considered as the closest angiosperm relatives, it would be logical to conclude that the origin of the angiosperm perianth and bisexual axis preceded that of the outer integument and carpel. If, by contrast, Caytoniales or Glossopteridales were regarded as closest to angiosperms, the opposite order-of-acquisition of the abovelisted features would be concluded. Such contrasting predictions could be compared with the timeline (notably, before and after epsilon) of molecular signatures of neofunctionalization in gene lineages known to be involved in the development of the angiosperm characters under consideration. In this way, it may be possible to combine molecular data from living groups to choose between alternative potential ancestors of the angiosperms, and thus shine some light into the dark ages surrounding the origin of the flower.

Acknowledgements Thanks are due to Ame´lie Andres-Robin for help in producing plant photos and to the editors for inviting this contribution. References 1. Barba-Montoya J, dos Reis M, Schneider H et al (2018) Constraining uncertainty in the timescale of angiosperm evolution and the veracity of a Cretaceous Terrestrial Revolution. New Phytol 218:819–834. https://doi. org/10.1111/nph.15011 2. Willis K, McElwain J (2013) The evolution of plants, 2nd edn. OUP, Oxford 3. Murat F, Armero A, Pont C et al (2017) Reconstructing the genome of the most recent common ancestor of flowering plants. Nat Genet 49:490–496. https://doi.org/10. 1038/ng.3813 4. Salomo K, Smith JF, Feild TS et al (2017) The emergence of earliest angiosperms may be earlier than fossil evidence indicates. Syst Bot 42:607–619. https://doi.org/10.1600/ 036364417X696438 5. Budd GE, Mann RP (2018) History is written by the victors: the effect of the push of the past on the fossil record. Evolution 72:2276– 2291. https://doi.org/10.1111/evo.13593 6. Fu Q, Bienvenido Diez J, Pole M et al (2018) An unexpected noncarpellate epigynous flower from the Jurassic of China. elife 7: e38827. https://doi.org/10.7554/eLife. 38827 7. Bateman RM (2020) Hunting the Snark: the flawed search for mythical Jurassic

angiosperms. J Exp Bot 71:22–35. https:// doi.org/10.1093/jxb/erz411 8. Coiro M, Doyle JA, Hilton J (2019) How deep is the conflict between molecular and fossil evidence on the age of angiosperms? New Phytol 223:83–99. https://doi.org/10. 1111/nph.15708 9. Sokoloff DD, Remizowa MV, El ES et al (2019) Supposed Jurassic angiosperms lack pentamery, an important angiosperm-specific feature. New Phytol 228:420–426. https:// doi.org/10.1111/nph.15974 10. Gomez B, Daviero-Gomez V, Coiffard C et al (2015) Montsechia, an ancient aquatic angiosperm. Proc Natl Acad Sci U S A 112:10985– 10988. https://doi.org/10.1073/pnas. 1509241112 11. Magallon S, Hilu KW, Quandt D (2013) Land plant evolutionary timeline: gene effects are secondary to fossil constraints in relaxed clock estimation of age and substitution rates. Am J Bot 100:556–573. https://doi.org/10. 3732/ajb.1200416 12. Jiao Y, Wickett NJ, Ayyampalayam S et al (2011) Ancestral polyploidy in seed plants and angiosperms. Nature 473:97–100. https://doi.org/10.1038/nature09916 13. Zwaenepoel A, Van de Peer Y (2019) Inference of ancient whole-genome duplications and the evolution of gene duplication and

104

Charles P. Scutt

loss rates. Mol Biol Evol 36:1384–1404. https://doi.org/10.1093/molbev/msz088 14. Feild TS, Arens NC, Doyle JA et al (2004) Dark and disturbed: a new image of early angiosperm ecology. Paleobiology 30:82– 107. https://doi.org/10.1666/0094-8373 (2004)0302.0.CO;2 15. Barrett PM, Willis KJ (2001) Did dinosaurs invent flowers? Dinosaur-angiosperm coevolution revisited. Biol Rev 76:411–447. h t t p s : // d o i . o r g / 1 0 . 1 0 1 7 / S1464793101005735 16. Byng JW, Chase MW, Christenhusz MJM et al (2016) An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG IV. Bot J Linn Soc 181:1–20. https://doi.org/10. 1111/boj.12385 17. Endress PK (2001) The flowers in extant basal angiosperms and inferences on ancestral flowers. Int J Plant Sci 162:1111–1140. https:// doi.org/10.1086/321919 18. Sauquet H, von Balthazar M, Magallon S et al (2017) The ancestral flower of angiosperms and its early diversification. Nat Commun 8: 1 6 0 4 7 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 / ncomms16047 19. Endress PK, Doyle JA (2015) Ancestral traits and specializations in the flowers of the basal grade of living angiosperms. Taxon 64:1093– 1116. https://doi.org/10.12705/646.1 20. Endress PK, Doyle JA (2009) Reconstructing the ancestral angiosperm flower and its initial specializations. Am J Bot 96:22–66. https:// doi.org/10.3732/ajb.0800047 21. De-Paula OC, Assis LCS, de Craene LPR (2018) Unbuttoning the ancestral flower of angiosperms. Trends Plant Sci 23:551–554. https://doi.org/10.1016/j.tplants.2018. 05.006 22. Sokoloff DD, Remizowa MV, Bateman RM, Rudall PJ (2018) Was the ancestral angiosperm flower whorled throughout? Am J Bot 105:5–15. https://doi.org/10.1002/ajb2. 1003 23. Endress PK, Igersheim A (2000) Gynoecium structure and evolution in basal angiosperms. Int J Plant Sci 161:S211–S223. https://doi. org/10.1086/317572 24. Endress PK (2011) Angiosperm ovules: diversity, development, evolution. Ann Bot 107: 1465–1489. https://doi.org/10.1093/aob/ mcr120 25. Friedman WE, Ryerson KC (2009) Reconstructing the ancestral female gametophyte of angiosperms: insights from Amborella and other ancient lineages of flowering plants. Am

J Bot 96:129–143. https://doi.org/10. 3732/ajb.0800311 26. Mao Y, Botella JR, Liu Y, Zhu J-K (2019) Gene editing in plants: progress and challenges. Natl Sci Rev 6:421–437. https://doi. org/10.1093/nsr/nwz005 27. Maher MF, Nasti RA, Vollbrecht M et al (2020) Plant gene editing through de novo induction of meristems. Nat Biotechnol 38: 84–89. https://doi.org/10.1038/s41587019-0337-2 28. Di Stilio VS (2011) Empowering plant evo-devo: virus induced gene silencing validates new and emerging model systems. BioEssays 33:711–718. https://doi.org/10. 1002/bies.201100040 29. Liu Y, Wang S, Li L et al (2022) The Cycas genome and the early evolution of seed plants. Nat Plants 8:389–401. https://doi.org/10. 1038/s41477-022-01129-7 30. Kramer EM (2015) A stranger in a strange land: the utility and interpretation of heterologous expression. Front Plant Sci 6: 734. https://doi.org/10.3389/fpls.2015. 00734 31. Vialette-Guiraud ACM, Andres-Robin A, Chambrier P et al (2016) The analysis of Gene Regulatory Networks in plant evo-devo. J Exp Bot 67:2549–2563. https://doi.org/10.1093/jxb/erw119 32. Bartlett A, O’Malley RC, Huang SC et al (2017) Mapping genome-wide transcriptionfactor binding sites using DAP-seq. Nat Protoc 12:1659–1672. https://doi.org/10. 1038/nprot.2017.055 33. Poncet V, Birnbaum P, Burtet-Sarramegna V et al (2019) Amborella– bearing witness to the past? Annu Plant Rev Online 2(3). https://doi.org/10.1002/9781119312994. apr0689 34. Kafer J, Bewick A, Andres-Robin A et al (2022) A derived ZW chromosome system in Amborella trichopoda, representing the sister lineage to all other extant flowering plants. New Phytol 233:1636–1642. https://doi. org/10.1111/nph.17662 35. Friedman WE (2006) Embryological evidence for developmental lability during early angiosperm evolution. Nature 441:337–340. https://doi.org/10.1038/nature04690 36. Grosse-Veldmann B, Korotkova N, Reinken B et al (2011) Amborella trichopoda- cultivation of the most ancestral angiosperm in botanic gardens. Sibbaldi 9:143–155 37. Fogliani B, Gateble G, Villegente M et al (2017) The morphophysiological dormancy in Amborella trichopoda seeds is a

Model Species to Investigate the Origin of Flowers pleisiomorphic trait in angiosperms. Ann Bot 119:581–590. https://doi.org/10.1093/ aob/mcw244 38. Fourquin C, Vinauger-Douard M, Fogliani B et al (2005) Evidence that CRABS CLAW and TOUSLED have conserved their roles in carpel development since the ancestor of the extant angiosperms. Proc Natl Acad Sci U S A 102:4649–4654. https://doi.org/10. 1073/pnas.0409577102 39. Kim S, Koh J, Yoo MJ et al (2005) Expression of floral MADS-box genes in basal angiosperms: implications for the evolution of floral regulators. Plant J 43:724–744. https://doi. org/10.1111/j.1365-313X.2005.02487.x 40. Vialette-Guiraud ACM, Adam H, Finet C et al (2011) Insights from ANA-grade angiosperms into the early evolution of CUP-SHAPED COTYLEDON genes. Ann Bot 107:1511–1519. https://doi.org/10. 1093/aob/mcr024 41. Arnault G, Vialette ACM, Andres-Robin A et al (2018) Evidence for the extensive conservation of mechanisms of ovule integument development since the most recent common ancestor of living angiosperms. Front Plant Sci 9:1352. https://doi.org/10.3389/fpls. 2018.01352 42. Albert VA, Barbazuk WB, dePamphilis CW et al (2013) The Amborella genome and the evolution of flowering plants. Science 342: 1241089. https://doi.org/10.1126/science. 1241089 43. Buzgo M, Soltis PS, Soltis DE (2004) Floral developmental morphology of Amborella trichopoda (Amborellaceae). Int J Plant Sci 165: 925–947. https://doi.org/10.1086/424024 44. Loehne C, Borsch T, Wiersema JH (2007) Phylogenetic analysis of nymphaeales using fast-evolving and noncoding chloroplast markers. Bot J Linn Soc 154:141–163. https:// doi.org/10.1111/j.1095-8339.2007. 00659.x 45. Yang Y, Sun P, Lv L et al (2020) Prickly waterlily and rigid hornwort genomes shed light on early angiosperm evolution. Nat Plants 6: 215–222. https://doi.org/10.1038/ s41477-020-0594-6 46. Iles WJD, Rudall PJ, Sokoloff DD et al (2012) Molecular phylogenetics of hydatellaceae (nymphaeales): sexual-system homoplasy and a new sectional classification. Am J Bot 99:663–676. https://doi.org/10.3732/ajb. 1100524 47. Sokoloff DD, Marques I, Macfarlane TD et al (2019) Cryptic species in an ancient

105

flowering-plant lineage (Hydatellaceae, Nymphaeales) revealed by molecular and micromorphological data. Taxon 68:1–19. https://doi.org/10.1002/tax.12026 48. Yamada T, Ito M, Kato M (2003) Expression pattern of INNER NOOUTER homologue in Nymphaea (water lily family, Nymphaeaceae). Dev Genes Evol 213:510–513. https://doi.org/10.1007/s00427-0030350-8 49. Povilus RA, DaCosta JM, Grassa C et al (2020) Water lily (Nymphaea thermarum) genome reveals variable genomic signatures of ancient vascular cambium losses. Proc Natl Acad Sci U S A 117:8649–8656. h t t p s : // d o i . o r g / 1 0 . 1 0 7 3 / p n a s . 1922873117 50. Fischer E, Rodriguez CM (2010) 690. Nymphaea thermarum. Curtis’s Bot Mag 27:318– 3 2 7 . h t t p s : // d o i . o r g / 1 0 . 1 1 1 1 / j . 1467-8748.2010.01711.x 51. Povilus RA, Losada JM, Friedman WE (2015) Floral biology and ovule and seed ontogeny of Nymphaea thermarum, a water lily at the brink of extinction with potential as a model system for basal angiosperms. Ann Bot 115: 211–226. https://doi.org/10.1093/aob/ mcu235 52. Povilus RA, Diggle PK, Friedman WE (2018) Evidence for parent-of-origin effects and interparental conflict in seeds of an ancient flowering plant lineage. Proc R Soc B Biol Sci 285:20172491. https://doi.org/10. 1098/rspb.2017.2491 53. Rudall PJ, Remizowa MV, Prenner G et al (2009) Nonflowers near the base of extant angiosperms? Spatiotemporal arrangement of organs in reproductive units of hydatellaceae and its bearing on the origin of the flower. Am J Bot 96:67–82. https://doi.org/10.3732/ ajb.0800027 54. Rudall PJ, Sokoloff DD, Remizowa MV et al (2007) Morphology of Hydatellaceae, an anomalous aquatic family recently recognized as an early-divergent angiosperm lineage. Am J Bot 94:1073–1092. https://doi.org/10. 3732/ajb.94.7.1073 55. Friedman WE, Bachelier JB, Hormaza JI (2012) Embryology in trithuria submersa (hydatellaceae) and relationships between embryo, endosperm, and perisperm in earlydiverging flowering plants. Am J Bot 99: 1083–1095. https://doi.org/10.3732/ajb. 1200066 56. Tuckett RE, Merritt DJ, Rudall PJ et al (2010) A new type of specialized

106

Charles P. Scutt

morphophysiological dormancy and seed storage behaviour in Hydatellaceae, an earlydivergent angiosperm family. Ann Bot 105: 1053–1061. https://doi.org/10.1093/aob/ mcq062 57. Kynast RG, Joseph JA, Pellicer J et al (2014) Chromosome behavior at the base of the angiosperm radiation: karyology. Am J Bot 101:1447–1455. https://doi.org/10.3732/ ajb.1400050 58. Adsul AA, Patil SM, Yadav SR, Bapat VA (2012) In vitro culture of Trithuria konkanensis, one of the smallest angiosperms. Curr Sci 103:979–980 59. Vialette-Guiraud ACM, Alaux M, Legeai F et al (2011) Cabomba as a model for studies of early angiosperm evolution. Ann Bot 108: 589–598. https://doi.org/10.1093/aob/ mcr088 60. Scheers K, Denys L, Jacobs I et al (2019) Cabomba caroliniana Gray (Cabombaceae) invades major waterways in Belgium. Knowl Manag Aquat Ecosyst 420:22. https://doi. org/10.1051/kmae/2019014 61. Yamada T, Yokota S, Hirayama Y et al (2011) Ancestral expression patterns and evolutionary diversification of YABBY genes in angiosperms. Plant J 67:26–36. https://doi.org/ 10.1111/j.1365-313X.2011.04570.x 62. Thien L, White D, Yatsu L (1983) The reproductive-biology of a relict Illiciumfloridanum Ellis. Am J Bot 70:719–727. https://doi.org/10.2307/2443126 63. Prakash N, Alexander JH (1984) Selfincompatibility in Austrobaileya scandens. In: Williams EG, Knox RB (eds) Pollination ’84. University of Melbourne, Melbourne, pp 214–216 64. Bernhardt P, Sage T, Weston P et al (2003) The pollination of Trimenia moorei (Trimeniaceae): floral volatiles, insect/wind pollen vectors and stigmatic self-incompatibility in a basal angiosperm. Ann Bot 92:445–458. https://doi.org/10.1093/aob/mcg157 65. Allen AM, Hiscock SJ (2008) Evolution and phylogeny of self-incompatibility systems in angiosperms. In: Franklin-Tong VE (ed) Self-incompatibility in flowering plants. Springer, Berlin-Heidelberg, pp 73–101 66. Williams JH, Friedman WE (2004) The fourcelled female gametophyte of Illicium (Illiciaceae; Austrobaileyales): implications for understanding the origin and early evolution of monocots, eumagnoliids, and eudicots. Am J Bot 91:332–351. https://doi.org/10. 3732/ajb.91.3.332

67. Bachelier JB, Friedman WE (2011) Female gamete competition in an ancient angiosperm lineage. Proc Natl Acad Sci U S A 108: 12360–12365. https://doi.org/10.1073/ pnas.1104697108 68. Qing-tian Z, Shu-tian F, Yi-ming Y et al (2016) De novo transcriptome assembly of Schisandra chinensis Turcz. (Baill.). Genom Data 10:153–154. https://doi.org/10. 1016/j.gdata.2016.11.008 69. Wang S, Zhang Z, Song W-L et al (2018) Transcriptome analysis of Schisandra sphenanthera discovers putative lignan biosynthesis genes and genetic markers. Pak J Bot 50: 1047–1059 70. Bliss BJ, Wanke S, Barakat A et al (2013) Characterization of the basal angiosperm Aristolochia fimbriata: a potential experimental system for genetic studies. BMC Plant Biol 13:13. https://doi.org/10.1186/14712229-13-13 71. Qin L, Hu Y, Wang J et al (2021) Insights into angiosperm evolution, floral development and chemical biosynthesis from the Aristolochia fimbriata genome. Nat Plants 7: 1239–1253. https://doi.org/10.1038/ s41477-021-00990-2 72. Bliss BJ, Landherr L, dePamphilis CW et al (2009) Regeneration and plantlet development from somatic tissues of Aristolochia fimbriata. Plant Cell Tissue Organ Cult 98:105– 114. https://doi.org/10.1007/s11240-0099543-9 73. Pabon-Mora N, Suarez-Baron H, Ambrose BA, Gonzalez F (2015) Flower development and perianth identity candidate genes in the basal angiosperm Aristolochia fimbriata (Piperales: Aristolochiaceae). Front Plant Sci 6:1095. https://doi.org/10.3389/fpls. 2015.01095 74. Perez-Mesa P, Suarez-Baron H, Ambrose BA et al (2019) Floral MADS-box protein interactions in the early diverging angiosperm Aristolochia fimbriata Cham. (Aristolochiaceae: Piperales). Evol Dev 21:96–110. https://doi.org/10.1111/ede.12282 75. Rendon-Anaya M, Ibarra-Laclette E, Mendez-Bravo A et al (2019) The avocado genome informs deep angiosperm phylogeny, highlights introgressive hybridization, and reveals pathogen-influenced gene space adaptation. Proc Natl Acad Sci U S A 116:17081– 17089. https://doi.org/10.1073/pnas. 1822129116 76. Chen J, Hao Z, Guang X et al (2019) Liriodendron genome sheds light on angiosperm phylogeny and species-pair differentiation.

Model Species to Investigate the Origin of Flowers Nat Plants 5:18–25. https://doi.org/10. 1038/s41477-018-0323-6 77. Nepi M, Little S, Guarnieri M et al (2017) Phylogenetic and functional signals in gymnosperm ovular secretions. Ann Bot 120:923– 936. https://doi.org/10.1093/aob/ mcx103 78. Friedman WE, Carmichael JS (1996) Double fertilization in gnetales: implications for understanding reproductive diversification among seed plants. Int J Plant Sci 157:S77– S94. https://doi.org/10.1086/297405 79. Magallon S, Sanderson MJ (2002) Relationships among seed plants inferred from highly conserved genes: sorting conflicting phylogenetic signals among ancient lineages. Am J Bot 89:1991–2006. https://doi.org/10. 3732/ajb.89.12.1991 80. Mathews S, Clements MD, Beilstein MA (2010) A duplicate gene rooting of seed plants and the phylogenetic position of flowering plants. Philos Trans R Soc B Biol Sci 365:383–395. https://doi.org/10.1098/ rstb.2009.0233 81. Doyle JA (2012) Molecular and fossil evidence on the origin of angiosperms. In: Jeanloz R (ed) Annual review of earth and planetary sciences, vol 40. Annual Reviews, Palo Alto, pp 301–326 82. Lee EK, Cibrian-Jaramillo A, Kolokotronis S-O et al (2011) A functional phylogenomic view of the seed plants. PLoS Genet 7: e1002411. https://doi.org/10.1371/jour nal.pgen.1002411 83. Zhong B, Deusch O, Goremykin VV et al (2011) Systematic error in seed plant phylogenomics. Genome Biol Evol 3:1340–1348. https://doi.org/10.1093/gbe/evr105 84. Christenhusz MJM, Reveal JL, Farjon A et al (2011) A new classification and linear sequence of extant gymnosperms. Phytotaxa 19:55–70. https://doi.org/10.11646/ phytotaxa.19.1.3 85. Niu S, Yuan H, Sun X et al (2016) A transcriptomics investigation into pine reproductive organ development. New Phytol 209: 1278–1289. https://doi.org/10.1111/nph. 13680 86. Giacomello S, Salmen F, Terebieniec BK et al (2017) Spatially resolved transcriptome profiling in model plant species. Nat Plants 3:17061. https://doi.org/10.1038/nplants. 2017.61 87. Zhang PY, Tan HTW, Pwee KH, Kumar PP (2004) Conservation of class C function of floral organ development during 300 million years of evolution from gymnosperms to

107

angiosperms. Plant J 37:566–577. https:// doi.org/10.1046/j.1365-313X.2003. 01983.x 88. Moyroud E, Monniaux M, Thevenon E et al (2017) A link between LEAFY and B-gene homologues in Welwitschia mirabilis sheds light on ancestral mechanisms prefiguring floral development. New Phytol 216:469–481. https://doi.org/10.1111/nph.14483 89. Wang Y-Q, Melzer R, Theissen G (2010) Molecular interactions of orthologues of floral homeotic proteins from the gymnosperm Gnetum gnemon provide a clue to the evolutionary origin of “floral quartets”. Plant J 64: 177–190. https://doi.org/10.1111/j. 1365-313X.2010.04325.x 90. Winter KU, Saedler H, Theissen G (2002) On the origin of class B floral homeotic genes: functional substitution and dominant inhibition in Arabidopsis by expression of an orthologue from the gymnosperm Gnetum. Plant J 31:457–475. https://doi.org/10.1046/j. 1365-313X.2002.01375.x 91. Uddenberg D, Akhter S, Ramachandran P et al (2015) Sequenced genomes and rapidly emerging technologies pave the way for conifer evolutionary developmental biology. Front Plant Sci 6:970. https://doi.org/10.3389/ fpls.2015.00970 92. Uddenberg D, Reimegard J, Clapham D et al (2013) Early cone setting in Picea abies acrocona is associated with increased transcriptional activity of a MADS box transcription factor. Plant Physiol 161:813–823. https:// doi.org/10.1104/pp.112.207746 93. Scutt CP (2018) The origin of angiosperms. Evol Dev Biol Ref Guide:1–20 94. Shi G, Herrera F, Herendeen PS et al (2021) Mesozoic cupules and the origin of the angiosperm second integument. Nature 594:223– 226. https://doi.org/10.1038/s41586-02103598-w 95. Doyle JA (2008) Integrating molecular phylogenetic and paleobotanical evidence on origin of the flower. Int J Plant Sci 169:816– 843. https://doi.org/10.1086/589887 96. Leebens-Mack JH, Barker MS, Carpenter EJ et al (2019) One thousand plant transcriptomes and the phylogenomics of green plants. Nature 574:679–685. https://doi.org/10. 1038/s41586-019-1693-2 97. Flores-Tornero M, Proost S, Mutwil M et al (2019) Transcriptomics of manually isolated Amborella trichopoda egg apparatus cells. Plant Reprod 32:15–27. https://doi.org/10. 1007/s00497-019-00361-0

108

Charles P. Scutt

98. Rivarola Sena AC, Andres-Robin A, Vialette AC et al (2022) Custom methods to identify conserved genetic modules applied to novel transcriptomic data from Amborella trichopoda. J Exp Bot 73:2487–2498. https://doi. org/10.1093/jxb/erac044 99. Leitch IJ, Hanson L (2002) DNA C-values in seven families fill phylogenetic gaps in the basal angiosperms. Bot J Linn Soc 140:175– 1 7 9 . h t t p s : // d o i . o r g / 1 0 . 1 0 4 6 / j . 1095-8339.2002.00096.x 100. Povilus RA, Friedman WE (2022) Transcriptomes across fertilization and seed development in the water lily Nymphaea thermarum (Nymphaeales): evidence for epigenetic patterning during reproduction. Plant Reprod 35:161–178. https://doi.org/10.1007/ s00497-022-00438-3 101. Pellicer J, Kelly LJ, Magdalena C, Leitch IJ (2013) Insights into the dynamics of genome size and chromosome evolution in the early diverging angiosperm lineage Nymphaeales (water lilies). Genome 56:437–449. https:// doi.org/10.1139/gen-2013-0039 102. Wu Q, Wu J, Li S-S et al (2016) Transcriptome sequencing and metabolite analysis for revealing the blue flower formation in waterlily. BMC Genomics 17:897. https://doi. org/10.1186/s12864-016-3226-9 103. Yoo M-J, Chanderbali AS, Altman NS et al (2010) Evolutionary trends in the floral transcriptome: insights from one of the basalmost angiosperms, the water lily Nuphar advena (Nymphaeaceae). Plant J 64:687–698. https://doi.org/10.1111/j.1365-313X. 2010.04357.x 104. Marques I, Montgomery SA, Barker MS et al (2016) Transcriptome-derived evidence supports recent polyploidization and a major phylogeographic division in Trithuria submersa (Hydatellaceae, Nymphaeales). New Phytol 210:310–323. https://doi.org/10. 1111/nph.13755 105. Bharathan G, Lambert G, Galbraith D (1994) Nuclear-DNA content of monocotyledons and related taxa. Am J Bot 81:381–386. https://doi.org/10.2307/2445466 106. Arumuganathan K, Earle ED (1991) Nuclear DNA content of some important plant species. Plant Mol Biol Report 9:208–218 107. Yang Y, Xu M, Luo Q et al (2014) De novo transcriptome analysis of Liriodendron chinense petals and leaves by Illumina sequencing. Gene 534:155–162. https://doi.org/ 10.1016/j.gene.2013.10.073 108. Wang K, Li M, Gao F et al (2012) Identification of conserved and novel microRNAs from Liriodendron chinense floral tissues. PLoS

One 7:e44696. https://doi.org/10.1371/ journal.pone.0044696 109. Lorenz WW, Ayyampalayam S, Bordeaux JM et al (2012) Conifer DBMagic: a database housing multiple de novo transcriptome assemblies for 12 diverse conifer species. Tree Genet Genomes 8:1477–1485. https://doi.org/10.1007/s11295-0120547-y 110. O’Brien IEW, Smith DR, Gardner RC, Murray BG (1996) Flow cytometric determination of genome size in Pinus. Plant Sci 115: 91–99 111. Joyner KL, Wang XR, Johnston JS et al (2001) DNA content for Asian pines parallels New World relatives. Can J Bot Rev Can Bot 79:192–196. https://doi.org/10.1139/ b00-151 112. Niu S, Li J, Bo W et al (2022) The Chinese pine genome and methylome unveil key features of conifer evolution. Cell 185:204–217. https://doi.org/10.1016/j.cell.2021. 12.006 113. Du J, Zhang Z, Zhang H, Junhong T (2017) EST-SSR marker development and transcriptome sequencing analysis of different tissues of Korean pine (Pinus koraiensis Sieb. et Zucc.). Biotechnol Biotechnol Equip 31: 679–689. https://doi.org/10.1080/ 13102818.2017.1331755 114. Fuchs J, Jovtchev G, Schubert I (2008) The chromosomal distribution of histone methylation marks in gymnosperms differs from that of angiosperms. Chromosom Res 16:891– 898. https://doi.org/10.1007/s10577008-1252-4 115. Nystedt B, Street NR, Wetterbom A et al (2013) The Norway spruce genome sequence and conifer genome evolution. Nature 497: 579–584. https://doi.org/10.1038/ nature12211 116. Little SA, Boyes IG, Donaleshen K et al (2016) A transcriptomic resource for Douglas-fir seed development and analysis of transcription during late megagametophyte development. Plant Reprod 29:273–286. https://doi.org/10.1007/s00497-0160291-9 117. Neale DB, McGuire PE, Wheeler NC et al (2017) The Douglas-fir genome sequence reveals specialization of the photosynthetic apparatus in Pinaceae. G3-Genes Genomes Genet 7:3157–3167. https://doi.org/10. 1534/g3.117.300078 118. Hizume M, Kondo T, Shibata F, Ishizuka R (2001) Flow cytometric determination of genome size in the Taxodiaceae, Cuprssaceae

Model Species to Investigate the Origin of Flowers sensu stricto and Sciadopityaceae. Cytologia (Tokyo) 66:307–311 119. Zhou W, Chen Q, Wang X-B et al (2019) De novo assembly of the Platycladus orientalis (L.) Franco transcriptome provides insight into the development and pollination mechanism of female cone based on RNA-Seq data. Sci Rep 9:10191. https://doi.org/10.1038/ s41598-019-46696-6 120. Pirone-Davies C, Prior N, von Aderkas P et al (2016) Insights from the pollination drop proteome and the ovule transcriptome of Cephalotaxus at the time of pollination drop production. Ann Bot 117:973–984. https:// doi.org/10.1093/aob/mcw026 121. Wang L, Lu Z, Li W et al (2016) Global comparative analysis of expressed genes in ovules and leaves of Ginkgo biloba L. Tree Genet Genomes 12:29. https://doi.org/10. 1007/s11295-016-0989-8

109

122. Wang L, Zhao J, Luo K et al (2016) Deep sequencing discovery and profiling of conserved and novel miRNAs in the ovule of Ginkgo biloba. Trees Struct Funct 30:1557– 1567. https://doi.org/10.1007/s00468016-1389-2 123. Smarda P, Vesely P, Smerda J et al (2016) Polyploidy in a ‘living fossil’ Ginkgo biloba. New Phytol 212:11–14. https://doi.org/10. 1111/nph.14062 124. Guan R, Zhao Y, Zhang H et al (2016) Draft genome of the living fossil Ginkgo biloba. Gigascience 5:49. https://doi.org/10. 1186/s13742-016-0154-1 125. Scutt CP, Vandenbussche M (2014) Current trends and future directions in flower development research. Ann Bot 114:1399–1406. https://doi.org/10.1093/aob/mcu224

Chapter 5 Hormones and Flower Development in Arabidopsis Victor M. Zu´n˜iga-Mayo, Yolanda Durán-Medina, Nayelli Marsch-Martı´nez, and Stefan de Folter Abstract Sexual reproduction requires the participation of two gametes, female and male. In angiosperms, gametes develop in specialized organs, pollen (containing the male gametes) develops in the stamens, and the ovule (containing the female gamete) develops in the gynoecium. In Arabidopsis thaliana, the female and male sexual organs are found within the same structure called flower, surrounded by the perianth, which is composed of petals and sepals. During flower development, different organs emerge in an established order and throughout their development distinct tissues within each organ are differentiated. All this requires the coordination and synchronization of several biological processes. To achieve this, hormones and genes work together. These components can interact at different levels generating hormonal interplay and both positive and negative feedback loops, which in turn, gives robustness, stability, and flexibility to flower development. Here, we summarize the progress made on elucidating the role of different hormonal pathways during flower development in Arabidopsis thaliana. Key words Flower development, Floral organs, Transcription factors, Hormones

1

Introduction The flower is the plant structure that includes the reproductive organs and defines the angiosperms. In Arabidopsis thaliana during the reproductive stage, the inflorescence meristem generates floral meristems (FMs). The FM gives rise to the flower, where floral organs are arranged in concentric whorls, with a fixed number of whorls and organs per whorl. The Arabidopsis flower has four sepals in the first whorl and four petals in the second (both organs are sterile), six stamens in the third (male reproductive organs), and two fused carpels in the fourth, forming the gynoecium (female reproductive organ) [1, 2]. The initiation of floral organs takes place in a specific order. The sepals emerge first, followed by the medial stamens and the petals, shortly thereafter, the lateral stamens, and finally the gynoecium [1, 3]. At stage 6, the floral meristem stops producing new organs.

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_5, © Springer Science+Business Media, LLC, part of Springer Nature 2023

111

112

Victor M. Zu´n˜iga-Mayo et al.

Fig. 1 Flower development. (a) Scheme representing an inflorescence meristem (IM), viewed from the side. The floral meristems (FMs) arise from the periphery of the IM, in an ordered manner. (b) The floral organs emerge, from the FM, as primordia arranged in four concentric rings or whorls. The most external organs, the sepals, are initiated first. Afterwards petal formation initiates, followed by stamens. Finally, the gynoecium initiates as two fused carpels. When all the organs are fully formed, the flower opens, and fertilization occurs. IM inflorescence meristem, FM floral meristem. (Modified from Alvarez-Buylla et al. [2])

At the following stages, floral organs continue to grow and differentiate. Finally, at anthesis the flower is fully developed and ready for the pollination process (see Fig. 1) [1, 3]. Flower development requires the precise spatio-temporal coordination of a series of events, which needs the cooperative action of hormones and genes. Hormones can affect the transcriptional regulation of genes but also gene regulation can impact hormonal pathways at different levels, and they can connect different hormonal pathways generating hormonal crosstalk. In addition, hormones need different kinds of proteins to carry out their biosynthesis, signaling and transport, furthermore, most hormones are regulated by feedback mechanisms. Altogether, flower development is a stable and robust process [4–7]. In this chapter, we will focus on work done using the model plant Arabidopsis thaliana that has helped to understand the role of hormones during flower development.

2

Floral Meristem Initiation Plants produce lateral organs throughout their life from the flanks of specialized tissues called meristems [8]. After the switch from the vegetative to the reproductive phase, the vegetative shoot apical meristem (SAM) becomes the inflorescence meristem (IM). This transition is triggered by different stimuli in A. thaliana and involves two phases: the first phase generates branching subtended by cauline leaves and during the second phase flowers are formed [9]. The early stages of flower development include initiation, identity acquisition and emergence of the FM, which are connected and synchronized processes [10]. Studies in A. thaliana indicate that at least three hormones participate in the early stages of flower development. Gibberellin levels have a critical role in the two phases of floral transition. First,

Hormones and Flower Development

113

high levels of gibberellins promote termination of the vegetative phase and beginning of the reproductive phase. Once the transition is achieved, the plant begins to produce branches and cauline leaves but not floral meristems. A subsequent decrease in gibberellin levels allows the generation of lateral organs from the IM to acquire FM identity [9, 11–13]. Cytokinins can also induce the transition to the flowering phase under short day conditions [14], while jasmonates delay it [15]. On the other hand, auxin plays a central role in FM initiation and positioning at the IM (phyllotaxis). Different studies indicate that the establishment of a local auxin maximum due to its polarized transport is required to induce FM formation. Thus, the auxin transport system regulates the site of FM initiation through auxin accumulation, but also prevents new FM initiation around the new primordium where auxin levels are low [16–22]. Besides the place where a new FM is initiated, the time in which it is initiated is an important aspect for IM phyllotaxis. Recent studies indicate that cytokinin signaling (or the presence of inhibitory fields) is important for sequential FM initiation. Differential cytokinin signaling inactivation between the next two sites where a new FM will be generated prevents the co-initiation of more FMs. The FM with higher cytokinin signaling will grow first. Thus, the data suggest that a synergistic interaction between auxins and cytokinins is necessary for proper FM initiation, where auxin signaling determines the place of FM initiation and cytokinin signaling modulation controls the sequence of FM emergence [23, 24]. After FM initiation and emergence, an inward auxin flow is formed at the incipient primordium leading to FM outgrowth [25]. Following an initial growth phase, a stem cell population is established at the FM. Modeling studies propose that the combination of epidermal signals gives a mechanism for the establishment of stem cell domains in FM primordia, where cytokinin acts as a longrange signal, which, in combination with other signals, control gene expression at the center of the FM [26, 27]. Currently, the FM is a group of morphologically undifferentiated cells, but soon after, floral organs begin to form in an established order where hormones also play important roles.

3

Sepals The sepals are the first floral organs that form from the floral meristem (see Fig. 1). They form the outer organ layer of the flower bud and protect the developing flower. To date, auxins are the main hormone linked to sepal development in A. thaliana (see Fig. 2). Data generated from molecular markers suggest that auxins could have different roles in floral organ initiation with respect to FM initiation. At early stages of FM development both markers, the auxin signaling (DR5) and the floral organ founder cell

114

Victor M. Zu´n˜iga-Mayo et al.

Fig. 2 Overview of hormonal action through flower development. The different hormonal effects and interactions described for each floral organ at different developmental stages are depicted. In some cases, the effects of hormones will depend on the developmental stage of the organ, for example, promoting or repressing cell expansion

¨ SCHEN-LIKE; DRNL) markers, appear simulta(DORNRO neously with partially overlapping expression domains. In contrast, some reports indicate that, during floral organ development the expression of the floral organ founder cell marker precedes the expression of the auxin signaling marker. Therefore, the authors suggest that at least canonical auxin signaling might regulate floral organ outgrowth, instead of their initiation [28–31].

Hormones and Flower Development

115

The sepals are initiated in the following sequential order: one abaxial, then two lateral, and the fourth sepal forms adaxially. However, some reports indicate that auxin signaling appears in a different order: first, the abaxial and adaxial sepals and then the lateral sepals, which correlates with the sepal emergence. The authors of these reports suggest that auxin signaling coordinates the emergence of sepals, although it might not be involved in the establishment of their founder cells [29–32]. On the other hand, a modeling study predicts the participation of cytokinins during sepal development, possibly in polarity establishment and vascular development (see Fig. 2) [33]. A recent study sheds new light on the involvement of both hormones in sepal initiation; overlapping signaling of both hormonal pathways in focused regions for a small time is needed for the timing of sepal initiation [34]. Senescence of sepals is affected by jasmonic acid. In the agamous loss of function mutant, petal senescence is delayed which can be rescued by applying exogenous methyl jasmonate [35].

4

Petals Petals initially protect the organs developing in the inner whorls of the flower, and when development is complete, their growth contributes to open the flower and, in some species, they attract pollinators (see Fig. 1). The emergence of petals and medial stamens marks the beginning of stage 5. After petal emergence, its outgrowth can be divided into two phases. During early developmental stages petal growth depends on cell division, while cell expansion controls its growth during later developmental stages [1, 36]. So far, the participation of four hormones in petal development has been reported. These hormones interact at different levels during the cell expansion phase to achieve a proper flower maturation process (see Fig. 2). Local auxin maxima precede petal primordia outgrowth. Lampugnani and colleagues propose that growing sepals act as auxin sinks, draining it from their surroundings but leaving auxin available in the inter-sepal zones. This auxin is directed towards the petal initiation zones, through its polar transport [37]. Although petal and medial stamen primordia are visible at the same time, the local auxin maxima for petal emergence are established before those of the medial stamens [30, 32]. These variations between auxin accumulation and organ outgrowth suggest that petals and stamens have different thresholds for auxin responses or requirements [31]. Once petal primordia emerge, auxin is distributed in the developing provasculature [38] and the development continues with a period of cell proliferation. It has been proposed that auxin participates at this early stage through the regulation of transcription

116

Victor M. Zu´n˜iga-Mayo et al.

factors of the AINTEGUMENTA (ANT) family [39, 40]. There is information that suggests that auxin negatively regulates cell proliferation through ARF8 [39, 41]. However, it also has been reported that auxin activates ARF8 and ARF6 which participate at late stages of petal development [42]. Therefore, auxin participates in the regulation of both phases (cell proliferation and cell expansion); however, the regulation by this hormone is complex, and it depends on the stage of flower development. Around anthesis, auxins and gibberellins positively regulate jasmonate biosynthesis. They contribute to the synchronized growth of petals and stamens that results in the opening of the flower and pollination of the pistil. In this regulatory network, feedback loops regulate these hormonal pathways positively for auxin, and positively or negatively for jasmonates depending on the developmental stage of the petals (before or after anthesis) [42, 43]. However, some critical components of this transcriptional response can be regulated by jasmonate-independent pathways [42]. Ethylene also participates in petal development through the regulation of cell expansion. Evidence shows that either high or low ethylene signaling, or ethylene level can result in an increase of petal cell expansion [44, 45]. This suggests that ethylene levels must be fine-tuned to achieve optimal growth of the petal, since an imbalance in its concentration or signaling results in increased cell expansion [45]. At late stages of petal development, a deficiency in jasmonate biosynthesis or mutations in specific components of auxin signaling (for example, ARF8) have the same effect on petals, increasing both their size, and the number of their secondary veins. This suggests that that both hormones negatively modulate petal growth and vasculature formation at the end of development, when wild type petals stop growing but mutant petals do not (stage 14) [41, 46].

5

Stamens Stamens are the male reproductive organs of flowers (see Fig. 1). They are formed by two different structures: the anther and the filament. The anther consists of a four-locule structure, each locule is composed of four distinct concentric tissues, from outer to inner: epidermis, endothecium, middle layer and tapetum. These tissues support pollen grain development and provide protection. In addition, the anther helps to release the pollen at anthesis. The filament connects the anther to the flower and functions as a channel through which water and nutrients are transported between the plant and the anther [47–49]. Stamen development can be divided into two phases. During early developmental stages (stages 5–9 of flower development), stamen primordia emerge, tissue patterning is established, and

Hormones and Flower Development

117

microsporogenesis (the first phase of pollen grain development) occurs. While pollen grain mature, stamen filaments elongate and subsequently, anther dehiscence takes place during the late phase (stages 10–13) [49–51]. Stamens grow slowly at early stages and grow faster at stages 12 and 13 [42]. Different studies have reported the contribution of five hormones during stamen development (see Fig. 2). Arabidopsis flowers have four long medial and two short lateral stamens. All stamens emerge during stage 5. However, the primordia of the two lateral stamens are visible slightly later than the primordia of the four medial stamens. Auxins are necessary for stamen development throughout their entire life cycle. Mutants in different aspects related to the auxin pathway show an alteration in stamen number, indicating that stamen primordia initiation requires auxin biosynthesis, transport and signaling to establish local auxin maxima [18, 52–54]. On the other hand, auxin biosynthesis needs to be actively repressed at the boundary between stamen and carpel to control the number of stamen primordia that emerge in each flower [55]. Once the stamen primordium has been established, it begins to grow and at the same time, new organ patterning is determined. During this early morphogenetic phase, local auxin biosynthesis and fine-tuned auxin transport are necessary for the synchronized growth of the four locules within the same stamen, the synchronization of the meiotic divisions that microsporocytes undergo, and repression of cellular proliferation and expansion in the tapetum [56–58]. The proposed model suggests that in normal conditions, auxin is synthesized in microsporocytes and tapetal cells. Then, a fraction of the auxin is transported away from these tissues and the rest is necessary to ensure proper male gametophyte development at early stages. Thus, when transport is affected, auxin is accumulated in these tissues causing alterations of these processes [57, 58]. At the beginning of the late phase, auxin biosynthetic genes are expressed in different stamen tissues. However, experimental data indicate that, at this stage, auxin is mainly transported from the tapetum to the middle layer forming a local auxin maximum in the latter. Once established, this auxin maximum is critical for correct distribution of this hormone to other tissues [52, 59–61]. In addition to the middle layer, auxin also accumulates in the pollen. Moreover, the stamen filament serves as a channel through which auxin is transported both acropetally and basipetally [62, 63]. These studies indicate that auxin synthesis, transport, accumulation and perception are involved in the three main processes that take place in the late phase of stamen development: (1) auxins positively regulate filament elongation, (2) coordinate pollen maturation, and (3) negatively regulate anther dehiscence [57, 59– 63]. The data suggest that these processes require a precise communication between them through auxin, since an alteration in the

118

Victor M. Zu´n˜iga-Mayo et al.

accumulation or distribution of this hormone in any of the three tissues (filament, pollen, or middle layer) has a similar effect in the three processes mentioned above. Gibberellins also regulate stamen development. Expression patterns of gibberellin biosynthetic genes suggest that this hormone is synthesized in most stamen tissues including filament, pollen, tapetum, and mature anther walls [64–66]. At the end of the early phase, gibberellin signaling affects cell expansion in the tapetum and is necessary for male gametophyte development allowing microsporocytes to undergo meiosis. Gibberellin signaling is also necessary at later stages, during the mitosis phase [66–69]. So far, it is not clear how gibberellins affect anther dehiscence, although it has been proposed that both anther and fruit patterning that allows dehiscence could occur through similar gibberellindependent processes [66, 70]. On the other hand, it has been reported that gibberellins promote stamen filament elongation through the regulation of jasmonate biosynthesis [43]. During the late phase, jasmonates integrate signaling cascades triggered by auxins and gibberellins to carry out anther dehiscence and filament elongation. Different studies suggest that auxins can affect jasmonate biosynthesis both negatively and positively [42, 43, 71, 72]. During late anther development, auxin regulates anther dehiscence by blocking the premature lignification of the endothecium and through the negative regulation of jasmonate biosynthesis. Once the auxin concentration decreases, the jasmonate biosynthetic genes are released allowing an increase in jasmonate concentration, which triggers the transcriptional response necessary for stomium break and anther opening [73]. On stamen filament elongation, gibberellin acts through jasmonate by positively regulating its biosynthesis. However, the effect of jasmonates on filament elongation (see Fig. 3a) requires auxin signaling trough ARF6 and ARF8, suggesting a positive feedback loop between jasmonates and auxins that allows anther filament elongation. However, it has also been suggested that these hormones could regulate this process independently [42, 43, 72]. Pollen is an abundant source of brassinosteroids and this hormone plays an important role in stamen development. Deficiencies of this hormone cause male sterility. During the early phase of stamen development, brassinosteroids contribute to the adequate development of the tapetum. They also regulate microspore mother cell formation and its later development, affecting the production of pollen grains significantly. At the late phase, brassinosteroids are necessary for anther filament elongation and pollen wall formation which affects their release from the anther [74–77]. Phenotypic analysis shows that cytokinin signaling is required for the development of different stamen tissues including the number of locules per anther, tapetum degeneration, the number of pollen grains and anther dehiscence. Cytokinins negatively regulate

Hormones and Flower Development

119

Fig. 3 Hormones and anthers and pistil development. (a) On the left side, a representation of the anther phenotypes of wild type and jasmonate defective flowers. Anther growth and dehiscence, and pollen maturation are affected in mutants with reduced jasmonate biosynthesis, perception or signaling. Auxins and gibberellins are involved in anther development. On the right side, a representation of a wild type, untreated gynoecium, and gynoecia presenting increasing severity of defects in organs along the apical-basal axis. Mutants affected in auxin signaling, transport or biosynthesis components, and wild type gynoecia treated with auxin transport inhibitors or cytokinin present apical-basal defects. (b) A graphical representation of the different developmental axes, and the tissues that compose a gynoecium, externally (on the left) and internally (right) at different stages of development. CMM carpel margin meristem, vm valve margin, r replum, tt transmitting tract. (Modified from Marsch-Martı´nez and de Folter [85])

secondary wall lignification in the anther endothecium which affects the break of the septum and stomium at anther opening [78–81].

6

Gynoecium The female reproductive structure at the inner whorl of a flower is called the gynoecium (see Figs. 1 and 3b). In A. thaliana, the gynoecium is constituted of two congenitally fused carpels. The gynoecium, together with the stamens, is crucial for plant sexual

120

Victor M. Zu´n˜iga-Mayo et al.

reproduction. After fertilization, the gynoecium develops into a fruit, protecting the seeds that develop inside it and releasing them at maturation [3, 4, 6, 7, 82, 83]. The gynoecium primordium is established at the beginning of stage 6. Throughout its development different tissues are differentiated along three developmental axes (see Fig. 3b). The abaxialadaxial axis refers to the identity of cell types that develop on the outer or inner side of the gynoecium (ovary), respectively; the medial-lateral axis that includes the valves (lateral domain), and the carpel margin meristem (CMM) and the other tissues generated from it, which constitute the medial domain; and the apical-basal axis composed, from top to bottom of stigma, style, ovary and gynophore [3, 4, 6, 7, 82, 84]. The gynoecium structure is complex and requires the coordinated and synchronized development of many tissues. To achieve this, the collective action of different hormones is necessary (see Fig. 2) [7, 85]. During gynoecium development, auxins and cytokinins show complementary signaling patterns. However, these two hormonal pathways are closely interconnected at different levels, including the synthesis and signaling of both hormones and transport in the case of auxins [86–91]. Just before the gynoecium primordium emerges, auxin biosynthesis and transport are necessary for the formation of two lateral signaling foci, while cytokinin signaling is active at the medial domain of the presumptive gynoecium primordium [87, 90]. Shortly thereafter, auxins affect the cytokinin pathway through the repression of both cytokinin biosynthesis and signaling. This auxin-cytokinin interaction at the FM center allows the gynoecium to emerge and the FM to cease its activity [91–94]. Once the gynoecium primordium is initiated, cytokinin signaling is present at the medial domain, and furthermore, two new medial auxin signaling foci are established. It has been shown that during early gynoecium development, cytokinin signaling activates auxin biosynthesis and transport, while auxins activate cytokinin signaling repression [87, 88, 90, 95]. During the transition from stage 8–9, auxin response increases and a ring is formed at the gynoecium apex, which is important for style and stigma development [95, 96]. On the other hand, cytokinin signaling is maintained in the medial domain, where it is required for CMM development and participates in other processes, including ovule number determination and replum development [84, 86, 88, 90, 97–102]. It has also been shown that cytokinin induces carpeloid features in an AGAMOUS-dependent manner [103]. Moreover, AGAMOUS expression is dependent on the cytokinin signaling pathway [103, 104]. Throughout gynoecium development, auxin is synthesized in the medial domain where cytokinin signaling plays a positive role by activating auxin biosynthetic genes, then auxin is

Hormones and Flower Development

121

transported to different gynoecium tissues including those at its apex through a distribution system known as the “reverse fountain,” a model proposed by Benkova and colleagues in 2003 [18, 87, 88]. In the young gynoecium, auxins play an important role in the formation of medial domain tissues including placenta, septum, transmitting tract and ovule initiation [18, 105, 106]. In addition, auxin and cytokinin influence apical-basal patterning of the gynoecium (see Fig. 3a). All aspects of the auxin pathway participate in this developmental process, since mutants affected in synthesis, transport and signaling show apical-basal patterning defects [52–54, 105, 107]. These defects are alterations in the normal proportions of the tissues formed along the apical-basal axis (see Fig. 3a). Initially, it was proposed that an auxin gradient directed the apical-basal patterning of the gynoecium, although more recent results suggest that a local auxin maximum at the gynoecium apex, rather than a gradient, is important for this process [105, 108]. On the other hand, the data suggest that cytokinin also affects the apical-basal patterning of the gynoecium through the auxin pathway. Besides, cytokinin perception (in the medial domain presumably) and cytokinin signaling repression in the lateral domain seem to have a buffering effect on perturbations in auxin and cytokinin homeostasis [88, 109, 110]. Furthermore, it was recently reported that auxin is also capable of directly affecting the activity of a protein complex (proteinprotein interactions of ETTIN/ARF3) that participates in gynoecium development through a mechanism known as non-canonical auxin sensing [111, 112]. Recently, a study on another ARF, reported that MONOPTEROS/ARF5 is also able to function in regions with an auxin minimum in ovule development. Furthermore, an ARF5 splice form that encodes a biologically functional isoform that lacks the Aux/IAA interaction domain can complement the mp/arf5 mutant, restoring gynoecium and ovule development [113]. Brassinosteroids regulate the development of different tissues in the medial domain. This hormone positively regulates the number of ovules produced per gynoecium and affects style and stigma development [114–116]. Furthermore, genetic analyses suggest that brassinosteroids act downstream of auxins in the regulation of transmitting tract development and differentiation, as well as in the process of cell death in this tissue and the stigma [117]. Another study suggests that brassinosteroids produced by the transmitting tract aid pollen tube to grow faster and more efficiently through this tissue [118]. Cytokinin signaling is also needed for transmitting tract development [86, 88, 102]. Gibberellins negatively affect the proliferative activity of the CMM, negatively affecting the number of ovules per gynoecium. This effect seems to be auxinindependent [116, 119, 120].

122

Victor M. Zu´n˜iga-Mayo et al.

After fertilization, the valve margin begins to differentiate. This tissue is composed of two cell layers: the separation layer and the lignification layer, both necessary for fruit dehiscence [83]. Before anthesis, cytokinin signaling appears in the presumptive valve margin and is maintained during fruit development. This cytokinin signaling is important for valve margin formation [86]. In addition, auxins and gibberellins also participate in this process. During fruit development, auxin signaling is dynamic. First, at early fruit development an auxin maximum is required in the valve margin, then an auxin minimum is necessary at late stages [121, 122]. Gibberellins specifically participate in the separation layer differentiation [70].

7

Conclusions In the last decades, different research groups around the world have made considerable progress that has helped to better understand the roles that different hormones play during flower development. Undoubtedly, the most studied hormone so far is auxin; this hormone participates throughout flower development, and it is necessary for the development of all floral organs. As expected, the floral organs in which most research has been made are the reproductive organs. Stamen development is the process for which more hormones have been described in detail. Particularly, the last stages of development where auxins and gibberellins interact, at least partially, with the jasmonate pathway, have been intensively studied. On the other hand, the study of auxin-cytokinin interaction and its participation during different stages of the gynoecium development has been addressed in recent years. Although progress has been made, we still have a lot to learn, especially on the interactions that may occur between different hormonal pathways and at different levels, including synthesis, transport, signaling, and response. A key point to consider is the identification of elements that connect two or more hormones and their impact on flower development. Another important aspect that begins to be explored and will be crucial in the coming years is the study of the interactions between hormones and other genetic regulatory “layers” such as epigenetic modifications. The combination of molecular approaches and in silico modeling will allow us to generate knowledge to better understand the role of hormonal networks during flower development.

Acknowledgements VMZM research is supported by the Cátedras-CONACyT program, grant number 5016 and Colegio de Postgraduados. Work in the SDF laboratory was financed by the Mexican National

Hormones and Flower Development

123

Council of Science and Technology (CONACyT) grants CB-2012177739, FC-2015-2/1061, and CB-2017-2018/A1-S-10126, and NMM and YDM by the CONACyT grant CB-2015-255069. SDF also acknowledges support of the Marcos Moshinsky Foundation and participation in the European Union H2020-MSCARISE-2015 project ExpoSEED (grant no. 691109), H2020MSCA-RISE-2019 project MAD (grant no. 872417), and H2020-MSCA-RISE-2020 project EVOfruland (grant no. 101007738). References 1. Smyth DR, Bowman JL, Meyerowitz EM (1990) Early flower development in Arabidopsis. Plant Cell 2:755–767 2. Alvarez-Buylla ER, Benı´tez M, Corvera-Poire´ A et al (2010) Flower development. Arabidopsis Book 8:e0127 3. Roeder AHK, Yanofsky MF (2006) Fruit development in Arabidopsis Arabidopsis Book 4:e0075 4. Simonini S, Østergaard L (2018) Female reproductive organ formation: a multitasking endeavor. Curr Top Dev Biol 131:337–371 5. Thomson B, Wellmer F (2019) Molecular regulation of flower development. Curr Top Dev Biol 131:185–210 ˜ iga-Mayo VM, Go´mez-Felipe A, Herrera6. Zu´n Ubaldo H, de Folter S (2019) Gynoecium development: networks in Arabidopsis and beyond. J Exp Bot 70:1447–1460 7. Herrera-Ubaldo H, de Folter S (2022) Gynoecium and fruit development in Arabidopsis. Development 149:dev200120 8. Gaillochet C, Lohmann JU (2015) The never-ending story: from pluripotency to plant developmental plasticity. Development 142:2237–2249 9. Yamaguchi N, Winter CM, Wu MF et al (2014) Gibberellin acts positively then negatively to control onset of flower formation in Arabidopsis. Science 344:638–642 10. Chandler JW (2012) Floral meristem initiation and emergence in plants. Cell Mol Life Sci 69:3807–3818 11. Galvao VC, Horrer D, Kuttner F, Schmid M (2012) Spatial control of flowering by DELLA proteins in Arabidopsis thaliana. Development 139:4072–4082 12. Yu S, Galva˜o VC, Zhang YC et al (2012) Gibberellin regulates the Arabidopsis floral transition through miR156-targeted SQUAMOSA PROMOTER BINDING–LIKE transcription factors. Plant Cell 24:3320–3332

13. Porri A, Torti S, Romera-Branchat M, Coupland G (2012) Spatially distinct regulatory roles for gibberellins in the promotion of flowering of Arabidopsis under long photoperiods. Development 139:2198–2209 14. D’Aloia M, Bonhomme D, Bouche´ F et al (2011) Cytokinin promotes flowering of Arabidopsis via transcriptional activation of the FT paralogue TSF. Plant J 65:972–979 15. Zhai Q, Zhang X, Wu F et al (2015) Transcriptional mechanism of jasmonate receptor COI1-mediated delay of flowering time in Arabidopsis. Plant Cell 27:2814–2828 16. Reinhardt D, Mandel T, Kuhlemeier C (2000) Auxin regulates the initiation and radial position of plant lateral organs. Plant Cell 12:507–518 17. Reinhardt D, Pesce ER, Stieger P et al (2003) Regulation of phyllotaxis by polar auxin transport. Nature 426:255–260 18. Benková E, Michniewicz M, Sauer M et al (2003) Local, efflux-dependent auxin gradients as a common module for plant organ formation. Cell 115:591–602 19. Heisler MG, Ohno C, Das P et al (2005) Patterns of auxin transport and gene expression during primordium development revealed by live imaging of the Arabidopsis inflorescence meristem. Curr Biol 15:1899– 1911 20. de Reuille PB, Bohn-Courseau I, Ljung K et al (2006) Computer simulations reveal properties of the cell-cell signaling network at the shoot apex in Arabidopsis. Proc Natl Acad Sci U S A 103:1627–1632 21. Jo¨nsson H, Heisler MG, Shapiro BE et al (2006) An auxin-driven polarized transport model for phyllotaxis. Proc Natl Acad Sci U S A 103:1633–1638 22. Smith RS, Guyomarc’h S, Mandel T et al (2006) A plausible model of phyllotaxis. Proc Natl Acad Sci U S A 103:1301–1306

124

Victor M. Zu´n˜iga-Mayo et al.

23. Besnard F, Refahi Y, Morin V et al (2014) Cytokinin signalling inhibitory fields provide robustness to phyllotaxis. Nature 505:417– 421 24. Besnard F, Rozier F, Vernoux T (2014) The AHP6 cytokinin signaling inhibitor mediates an auxin-cytokinin crosstalk that regulates the timing of organ initiation at the shoot apical meristem. Plant Signal Behav 9:e28788 25. Furutani M, Nakano Y, Tasaka M (2014) MAB4-induced auxin sink generates local auxin gradients in Arabidopsis organ formation. Proc Natl Acad Sci U S A 111:1198– 1203 26. Adibi M, Yoshida S, Weijers D, Fleck C (2016) Centering the organizing center in the Arabidopsis thaliana shoot apical meristem by a combination of cytokinin signaling and self-organization. PLoS One 12(3): e0174751 27. Gruel J, Landrein B, Tarr P et al (2016) An epidermis-driven mechanism positions and scales stem cell niches in plants. Sci Adv 2: e1500989 28. Aloni R, Aloni E, Langhans M, Ullrich CI (2006) Role of auxin in regulating Arabidopsis flower development. Planta 223:315–328 29. Chandler JW (2011) Founder cell specification. Trends Plant Sci 16:607–613 30. Chandler JW, Jacobs B, Cole M et al (2011) ¨ SCHEN-LIKE expression marks DORNRO Arabidopsis floral organ founder cells and precedes auxin response maxima. Plant Mol Biol 76:171–185 31. Chandler JW, Werr W (2014) Arabidopsis floral phytomer development: auxin response relative to biphasic modes of organ initiation. J Exp Bot 65:3097–3110 32. van Mourik S, Kaufmann K, van Dijk ADJ et al (2012) Simulation of organ patterning on the floral meristem using a polar auxin transport model. PLoS One 7:e28762 33. La Rota C, Chopard J, Das P et al (2011) A data-driven integrative model of sepal primordium polarity in Arabidopsis. Plant Cell 23: 4318–4333 34. Zhu M, Chen W, Mirabet V et al (2020) Robust organ size requires robust timing of initiation orchestrated by focused auxin and cytokinin signalling. Nat Plants 6:686–698 35. Jibran R, Tahir J, Cooney J et al (2017) Arabidopsis AGAMOUS regulates sepal senescence by driving jasmonate production. Front Plant Sci 8:2101 36. Hill JP, Lord EM (1989) Floral development in Arabidopsis thaliana: a comparison of the

wild type and the homeotic pistillata mutant. Can J Bot 67:2922–2936 37. Lampugnani ER, Kilinc A, Smyth DR (2013) Auxin controls petal initiation in Arabidopsis. Development 140:185–194 38. Sauret-Gu¨eto S, Schiessl K, Bangham A et al (2013) JAGGED controls Arabidopsis petal growth and shape by interacting with a divergent polarity field. PLoS Biol 11:e1001550 39. Krizek BA, Anderson JT (2013) Control of flower size. J Exp Bot 64:1427–1437 40. Huang T, Irish VF (2016) Gene networks controlling petal organogenesis. J Exp Bot 67:61–68 41. Varaud E, Brioudes F, Sze´csi J et al (2011) AUXIN RESPONSE FACTOR8 regulates Arabidopsis petal growth by interacting with the bHLH transcription factor BIGPETALp. Plant Cell 23:973–983 42. Reeves PH, Ellis CM, Ploense SE et al (2012) A regulatory network for coordinated flower maturation. PLoS Genet 8:e1002506 43. Cheng H, Song S, Xiao L et al (2009) Gibberellin acts through jasmonate to control the expression of MYB21, MYB24, and MYB57 to promote stamen filament growth in Arabidopsis. PLoS Genet 5:e1000440 44. Pei H, Ma N, Tian J et al (2013) A NAC transcription factor controls ethyleneregulated cell expansion in flower petals. Plant Physiol 163:775–791 45. van Es SW, Silveira SR, Rocha DI et al (2018) Novel functions of the Arabidopsis transcription factor TCP5 in petal development and ethylene biosynthesis. Plant J 94:867–879 46. Brioudes F, Joly C, Sze´csi J et al (2009) Jasmonate controls late development stages of petal growth in Arabidopsis thaliana. Plant J 60:1070–1080 47. Sanders P, Bui AQ, Weterings K et al (1999) Anther developmental defects in Arabidopsis thaliana male-sterile mutants. Sex Plant Reprod 11:297–322 48. Ma H (2005) Molecular genetic analyses of microsporogenesis and microgametogenesis in flowering plants. Annu Rev Plant Biol 56: 393–434 49. Cardarelli M, Cecchetti V (2014) Auxin polar transport in stamen formation and development: how many actors? Front Plant Sci 5:333 50. Goldberg RB, Beals TP, Sanders PM (1993) Anther development: basic principles and practical applications. Plant Cell 5:1217– 1229

Hormones and Flower Development 51. Feng X, Dickinson HG (2010) Tapetal cell fate, lineage and proliferation in the Arabidopsis anther. Development 137:2409–2416 52. Cheng Y, Dai X, Zhao Y (2006) Auxin biosynthesis by the YUCCA flavin monooxygenases controls the formation of floral organs and vascular tissues in Arabidopsis. Genes Dev 20:1790–1799 53. Sessions A, Nemhauser JL, McCall A et al (1997) ETTIN patterns the Arabidopsis floral meristem and reproductive organs. Development 124:4481–4491 54. Okada K, Ueda J, Komaki MK et al (1991) Requirement of the auxin polar transport system in early stages of Arabidopsis floral bud formation. Plant Cell 3:677–684 55. Xu Y, Prunet N, Gan ES et al (2018) SUPERMAN regulates floral whorl boundaries through control of auxin biosynthesis. EMBO J 37:e97499 56. Bennett SRM, Alvarez J, Bossinger G, Smyth DR (1995) Morphogenesis in pinoid mutants of Arabidopsis thaliana. Plant J 8:505–520 57. Cecchetti V, Brunetti P, Napoli N et al (2015) ABCB1 and ABCB19 auxin transporters have synergistic effects on early and late Arabidopsis anther development. J Integr Plant Biol 57: 1089–1098 58. Yao X, Tian L, Yang J et al (2018) Auxin production in diploid microsporocytes is necessary and sufficient for early stages of pollen development. PLoS Genet 14:e1007397 59. Cecchetti V, Altamura MM, Falasca G et al (2008) Auxin regulates Arabidopsis anther dehiscence, pollen maturation, and filament elongation. Plant Cell 20:1760–1774 60. Cecchetti V, Celebrin D, Napoli N et al (2017) An auxin maximum in the middle layer controls stamen development and pollen maturation in Arabidopsis. New Phytol 213: 1194–1207 61. Ghelli R, Brunetti P, Napoli N et al (2018) A newly identified flower-specific splice variant of AUXIN RESPONSE FACTOR8 regulates stamen elongation and endothecium lignification in Arabidopsis. Plant Cell 30:620–637 62. Feng XL, Ni WM, Elge S et al (2006) Auxin flow in anther filaments is critical for pollen grain development through regulating pollen mitosis. Plant Mol Biol 61:215–226 63. Salinas-Grenet H, Herrera-Vásquez A, Parra S et al (2018) Modulation of auxin levels in pollen grains afects stamen development and anther dehiscence in Arabidopsis. Int J Mol Sci 19:2480 64. Mitchum MG, Yamaguchi S, Hanada A et al (2006) Distinct and overlapping roles of two

125

gibberellin 3-oxidases in Arabidopsis development. Plant J 45:804–818 65. Hu J, Mitchum MG, Barnaby N et al (2008) Potential sites of bioactive gibberellin production during reproductive growth in Arabidopsis. Plant Cell 20:320–336 66. Plackett ARG, Thomas SG, Wilson ZA, Hedden P (2011) Gibberellin control of stamen development: a fertile field. Trends Plant Sci 16:568–578 67. Cheng H, Qin L, Lee S et al (2004) Gibberellin regulates Arabidopsis floral development via suppression of DELLA protein function. Development 131:1055–1064 68. Millar AA, Gubler F (2005) The Arabidopsis GAMYB-like genes, MYB33 and MYB65, are microRNA-regulated genes that redundantly facilitate anther development. Plant Cell 17: 705–721 69. Plackett ARG, Ferguson AC, Powers SJ et al (2014) DELLA activity is required for successful pollen development in the Columbia ecotype of Arabidopsis. New Phytol 201:825– 836 70. Arnaud N, Girin T, Sorefan K et al (2010) Gibberellins control fruit patterning in Arabidopsis thaliana. Genes Dev 24:2127–2132 71. Tabata R, Ikezaki M, Fujibe T et al (2010) Arabidopsis AUXIN RESPONSE FACTOR6 and 8 regulate jasmonic acid biosynthesis and floral organ development via repression of class 1 KNOX genes. Plant Cell Physiol 51: 164–175 72. Nagpal P, Ellis CM, Weber H et al (2005) Auxin response factors ARF6 and ARF8 promote jasmonic acid production and flower maturation. Development 132:4107–4118 73. Cecchetti V, Altamura MM, Brunetti P et al (2013) Auxin controls arabidopsis anther dehiscence by regulating endothecium lignification and jasmonic acid biosynthesis. Plant J 74:411–422 74. Grove MD, Spencer GF, Rohwedder WK et al (1979) Brassinolide, a plant growthpromoting steroid isolated from Brassica napus pollen. Nature 281:216–217 75. Szekeres M, Ne´meth K, Koncz-Kálmán Z et al (1996) Brassinosteroids rescue the deficiency of CYP90, a cytochrome P450, controlling cell elongation and de-etiolation in Arabidopsis. Cell 85:171–182 76. Bouquin T, Meier C, Foster R et al (2001) Control of specific gene expression by gibberellin and brassinosteroid. Plant Physiol 127: 450–458 77. Ye Q, Zhu W, Li L et al (2010) Brassinosteroids control male fertility by regulating the

126

Victor M. Zu´n˜iga-Mayo et al.

expression of key genes involved in Arabidopsis anther and pollen development. Proc Natl Acad Sci U S A 107:6100–6105 78. Higuchi M, Pischke MS, Mahonen AP et al (2004) In planta functions of the Arabidopsis cytokinin receptor family. Proc Natl Acad Sci U S A 101:8821–8826 79. Nishimura C, Ohashi Y, Sato S et al (2004) Histidine kinase homologs that act as cytokinin receptors possess overlapping functions in the regulation of shoot and root growth in Arabidopsis. Plant Cell 16:1365–1377 80. Jung KW, Oh SI, Kim YY et al (2008) Arabidopsis histidine-containing phosphotransfer factor 4 (AHP4) negatively regulates secondary wall thickening of the anther endothecium during flowering. Mol Cells 25:294–300 81. Kinoshita-Tsujimura K, Kakimoto T (2011) Cytokinin receptors in sporophytes are essential for male and female functions in Arabidopsis thaliana. Plant Signal Behav 6:66–71 ˜ iga-Mayo VM, Chávez 82. Reyes-Olalde JI, Zun Montes RA et al (2013) Inside the gynoecium: at the carpel margin. Trends Plant Sci 18:644–655 83. Ballester P, Ferrándiz C (2017) Shattering fruits: variations on a dehiscent theme. Curr Opin Plant Biol 35:68–75 84. Reyes-Olalde JI, de Folter S (2019) Control of stem cell activity in the carpel margin meristem (CMM) in Arabidopsis. Plant Reprod 32:123–136 85. Marsch-Martı´nez N, de Folter S (2016) Hormonal control of the development of the gynoecium. Curr Opin Plant Biol 29:104– 114 86. Marsch-Martı´nez N, Ramos-Cruz D, ReyesOlalde JI et al (2012) The role of cytokinin during Arabidopsis gynoecia and fruit morphogenesis and patterning. Plant J 72:222– 234 87. Larsson E, Roberts CJ, Claes AR et al (2014) Polar auxin transport is essential for medial versus lateral tissue specification and vascular-mediated valve outgrowth in Arabidopsis gynoecia. Plant Physiol 166:1998– 2012 ˜iga-Mayo VM, Serwa88. Reyes-Olalde JI, Zu´n towska J et al (2017) The bHLH transcription factor SPATULA enables cytokinin signaling, and both activate auxin biosynthesis and transport genes at the medial domain of the gynoecium. PLoS Genet 13:e1006726 ˜ iga-Mayo VM, Marsch89. Reyes-Olalde JI, Zu´n Martı´nez N, de Folter S (2017) Synergistic relationship between auxin and cytokinin in the ovary and the participation of the

transcription factor SPATULA. Plant Signal Behav 12:e1376158 90. Mu¨ller CJ, Larsson E, Spı´chal L, Sundberg E (2017) Cytokinin-auxin crosstalk in the gynoecial primordium ensures correct domain patterning. Plant Physiol 175:1144–1157 91. Zhang K, Wang R, Zi H et al (2018) AUXIN RESPONSE FACTOR3 regulates floral meristem determinacy by repressing cytokinin biosynthesis and signaling. Plant Cell 30:324– 346 92. Yamaguchi N, Huang J, Xu Y et al (2017) Fine-tuning of auxin homeostasis governs the transition from floral stem cell maintenance to gynoecium formation. Nat Commun 8:1125 93. Yamaguchi N, Huang J, Tatsumi Y et al (2018) Chromatin-mediated feed-forward auxin biosynthesis in floral meristem determinacy. Nat Commun 9:5290 94. Xu Y, Yamaguchi N, Gan ES, Ito T (2019) When to stop: an update on molecular mechanisms of floral meristem termination. J Exp Bot 70:1711–1718 95. Moubayidin L, Ostergaard L (2014) Dynamic control of auxin distribution imposes a bilateral-to-radial symmetry switch during gynoecium development. Curr Biol 24: 2743–2748 96. Schuster C, Gaillochet C, Lohmann JU (2015) Arabidopsis HECATE genes function in phytohormone control during gynoecium development. Development 142:3343–3350 97. Bartrina I, Otto E, Strnad M et al (2011) Cytokinin regulates the activity of reproductive meristems, flower organ size, ovule formation, and, thus, seed yield in Arabidopsis thaliana. Plant Cell 23:69–80 98. Cucinotta M, Manrique S, Guazzotti A et al (2016) Cytokinin response factors integrate auxin and cytokinin pathways for female reproductive organ development. Development 143:4419–4424 ˜ iga-Mayo VM, Ban ˜ os-Bayardo CR, Dı´az99. Zun Ramı´rez D et al (2018) Conserved and novel responses to cytokinin treatments during flower and fruit development in Brassica napus and Arabidopsis thaliana. Sci Rep 8: 6836 100. Barro-Trastoy D, Dolores Gomez M, Tornero P, Perez-Amador MA (2020) On the way to ovules: the hormonal regulation of ovule development. Crit Rev Plant Sci 39: 431–456 101. Cucinotta M, Di Marzo M, Guazzotti A et al (2020) Gynoecium size and ovule number are

Hormones and Flower Development interconnected traits that impact seed yield. J Exp Bot 71:2479–2489 ˜ iga-Mayo VM, 102. Cerbantez-Bueno VE, Zu´n Reyes-Olalde JI et al (2020) Redundant and non-redundant functions of the AHK cytokinin receptors during gynoecium development. Front Plant Sci 11:568277 103. Go´mez-Felipe A, Kierzkowski D, de Folter S (2021) The relationship between AGAMOUS and cytokinin signaling in the establishment of carpeloid features. Plan Theory 10:827–836 104. Rong XF, Sang YL, Wang L et al (2018) Type-B ARRs control carpel regeneration through mediating AGAMOUS expression in Arabidopsis. Plant Cell Physiol 59:761– 769 105. Nemhauser JL, Feldman LJ, Zambryski PC (2000) Auxin and ETTIN in Arabidopsis gynoecium morphogenesis. Development 127:3877–3888 106. Nole-Wilson S, Azhakanandam S, Franks RG (2010) Polar auxin transport together with AINTEGUMENTA and REVOLUTA coordinate early Arabidopsis gynoecium development. Dev Biol 346:181–195 107. Stepanova AN, Robertson-Hoyt J, Yun J et al (2008) TAA1-mediated auxin biosynthesis is essential for hormone crosstalk and plant development. Cell 133:177–191 108. Larsson E, Franks RG, Sundberg E (2013) Auxin and the Arabidopsis thaliana gynoecium. J Exp Bot 64:2619–2627 ˜iga-Mayo VM, Reyes-Olalde JI, Marsch109. Zu´n Martinez N, de Folter S (2014) Cytokinin treatments affect the apical-basal patterning of the Arabidopsis gynoecium and resemble the effects of polar auxin transport inhibition. Front Plant Sci 5:191 110. Durán-Medina Y, Serwatowska J, ReyesOlalde JI et al (2017) The AP2/ERF transcription factor DRNL modulates gynoecium development and affects its response to cytokinin. Front Plant Sci 8:1841 111. Simonini S, Deb J, Moubayidin L et al (2016) A noncanonical auxin-sensing mechanism is required for organ morphogenesis in Arabidopsis. Genes Dev 30:2286–2296

127

112. Simonini S, Mas PJ, Mas CMVS et al (2018) Auxin sensing is a property of an unstructured domain in the Auxin Response Factor ETTIN of Arabidopsis thaliana. Sci Rep 8:13563 113. Cucinotta M, Cavalleri A, Guazzotti A et al (2021) Alternative splicing generates a MONOPTEROS isoform required for ovule development. Curr Biol 31:892–899 114. Nole-Wilson S, Rueschhoff EE, Bhatti H, Franks RG (2010) Synergistic disruptions in seuss cyp85A2 double mutants reveal a role for brassinolide synthesis during gynoecium and ovule development. BMC Plant Biol 10:198 115. Huang HY, Jiang WB, Hu YW et al (2013) BR signal influences Arabidopsis ovule and seed number through regulating related genes expression by BZR1. Mol Plant 6: 456–469 ˜ os J et al 116. Barro-Trastoy D, Carrera E, Ban (2020) Regulation of ovule initiation by gibberellins and brassinosteroids in tomato and Arabidopsis: two plant species, two molecular mechanisms. Plant J 102:1026–1041 117. Crawford BCW, Yanofsky MF (2011) HALF FILLED promotes reproductive tract development and fertilization efficiency in Arabidopsis thaliana. Development 138:2999– 3009 118. Vogler F, Schmalzl C, Englhart M et al (2014) Brassinosteroids promote Arabidopsis pollen germination and growth. Plant Reprod 27:153–167 119. Gomez MD, Barro-Trastoy D, Escoms E et al (2018) Gibberellins negatively modulate ovule number in plants. Development 145: dev163865 120. Barro-Trastoy D, Gomez MD, Blanco-Tour˜ án N et al (2022) Gibberellins regulate in ovule number through a DELLA–CUC2 complex in Arabidopsis. Plant J 110:43–57 121. Sorefan K, Girin T, Liljegren SJ et al (2009) A regulated auxin minimum is required for seed dispersal in Arabidopsis. Nature 459:583– 586 122. van Gelderen K, van Rongen M, Liu A et al (2016) An INDEHISCENT-controlled auxin response specifies the separation layer in early arabidopsis fruit. Mol Plant 9:857–869

Part II Genetic and Phenotypic Analyses

Chapter 6 Genetic Screens for Floral Mutants in Arabidopsis thaliana: Enhancers and Suppressors Zhigang Huang, Thanh Theresa Dinh, Elizabeth Luscher, Shaofang Li, Xigang Liu, So Youn Won, and Xuemei Chen Abstract The flower is a hallmark feature that has contributed to the evolutionary success of land plants. Diverse mutagenic agents have been employed as a tool to genetically perturb flower development and identify genes involved in floral patterning and morphogenesis. Since the initial studies to identify genes governing processes such as floral organ specification, mutagenesis in sensitized backgrounds has been used to isolate enhancers and suppressors to further probe the molecular basis of floral development. Here, we first describe two commonly employed methods for mutagenesis (using ethyl methanesulfonate (EMS) or T-DNAs as mutagens), and then describe three methods for identifying a mutation that leads to phenotypic alterations: traditional map-based cloning, modified high-efficiency thermal asymmetric interlaced PCR (mhiTAIL-PCR), and deep sequencing in the plant model Arabidopsis thaliana. Key words EMS, T-DNA, Floral development, Map-based cloning, mhiTAIL-PCR

Mutagenesis,

Genetic screen,

Arabidopsis,

1 Introduction Plants, being sessile organisms, have to cope with many different stresses, both biotic and abiotic. The flower is a major contributing factor to the success and robustness of angiosperms. Until the 1980s, studies on plants had been largely limited to agriculturally important species; however, it is often difficult to work with these plants due to their large, in some cases, polyploid genomes. The small weed Arabidopsis thaliana was first purported as a genetic model system for plants by Friedrich Laibach [1]. Arabidopsis is amenable to genetic analyses due to its small size, small genome (125 Mb), rapid generation time (5–6 weeks from seed to seed),

This chapter is an updated version of Chapter 6 of Flower Development: Methods and Protocols (1st edition, 2014), Methods in Molecular Biology, vol 1110. Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_6, © Springer Science+Business Media, LLC, part of Springer Nature 2023

131

132

Zhigang Huang et al.

high fecundity (up to 10,000 seeds per plant), low repetitive sequences, and the ability to self-fertilize [1–5]. The power of Arabidopsis as a model system was evidenced, in part, by the identification of a series of genes involved in floral development in the late 1980s and early 1990s [6–14]. Since then, Arabidopsis has been the genetic model organism for plants and has been used to study diverse biological pathways. Chemical mutagenesis has been extensively used to genetically perturb biological pathways in Arabidopsis to identify components of these pathways. Initial genetic screens are often conducted in a wild-type background to identify major, nonredundant factors. However, such screens often fail to identify regulatory genes that have overlapping functions with other (often closely related) genes. Genetic modifier screens can be conducted to circumvent this genetic redundancy. A genetic modifier screen is performed in a mutant background, usually a weak allele in a major player in a biological process, to isolate other genes in that biological pathway. The second-site mutations resulting from this screen can either lead to an enhancement or to a suppression of the phenotype of the mutant plants that are being mutagenized. There have been several notable examples in which modifier screens have allowed researchers to effectively dissect complex genetic pathways in floral development. For instance, AGAMOUS (AG) controls carpel and stamen development in Arabidopsis, and loss of AG results in floral patterning defects and a reiterative flower-in-flower phenotype indicative of loss of floral determinacy [7, 14, 15]. Many alleles of AG have been identified, ranging from plants with a weak (ag-10) to strong (ag-1) phenotype. One of the alleles, ag-4, generates a partially functional protein that is able to confer stamen identity, but floral determinacy or carpel identity is compromised [16]. Dr. Xuemei Chen in Dr. Elliot Meyerowitz’s laboratory performed an EMS mutagenesis in the ag-4 background and identified a mutant with an enhanced phenotype, reminiscent of ag-1 (or flowers in which the stamens have been converted to petals). Map-based cloning revealed that the enhanced phenotype was due to mutations in two genes, which she named HUA1 and HUA2 (HUA means flower in Chinese) [17]. HUA1 is an RNA-binding protein and HUA2 is a novel protein involved in RNA processing [17–19]. hua1 hua2 double mutants have a weak phenotype, while the single mutants are phenotypically normal [17]. Using the hua1 hua2 double mutant, Dr. Chen and her group performed another EMS mutagenesis screen to identify mutants with enhanced floral determinacy and organ identity defects. From this screen, many mutants were isolated, including another weak allele of AG, ag-10; mutant alleles of HEN4, a gene encoding a KH domain RNA-binding protein; mutants in HEN1, a gene involved in small RNA biogenesis; and a mutant in ARGONAUTE10 (AGO10), an effector in microRNA function

Genetic Screens: Enhancers and Suppressors

133

[20]. Studies with the hua1 hua2 hen4 mutants revealed the redundant functions of the three genes in promoting the transcription elongation or splicing of AG pre-mRNA [18]. Subsequent work on HEN1 eventually led to the discovery that miR172 promotes floral determinacy by repressing its target gene APETALA2 (AP2) [21]. Furthermore, Dr. Xigang Liu, a postdoctoral scholar in Dr. Chen’s laboratory, performed an enhancer screen using the weak ag-10 allele. From that screen, CURLY LEAF (CLF), AGO10, POWERDRESS (PWR), AUXIN RESPONSE FACTOR 3 (ARF3), TOPOISOMERASE1a (TOP1a) and FAR-RED ELONGATED HYPOCOTYL3 (FHY3) were identified and shown to be involved in floral determinacy [20, 22–26]. In addition to this, an intermediate-strength AG allele, ag-11, was also isolated from the genetic screen in the ag-10 background [27]. Figure 1 depicts the discovery of players that specify the identities of the reproductive organs or confer floral determinacy through genetic screens. Another well-known example of the usefulness of modifier screens involves the gene CRABS CLAW (CRC) and the pathway concerning polarity in Arabidopsis carpels. CRC is the founding member of the YABBY family of transcription factors, in which members are involved in establishing polar differentiation of lateral organs [28, 29]. Loss-of-function crc alleles were shown to be defective in carpel and nectary development, but none of the phenotypes in the mutant alleles indicated a defect in polar differentiation [15]. A modifier screen of crc led to the identification of PICKLE (PKL) and KANADI1 (KAN1), and the discovery that CRC promotes abaxial identity in Arabidopsis carpels. This function is normally masked by KAN1, an abaxial identity promoting gene, and PKL, a gene that finely regulates meristematic activity [30]. Another modifier screen (using EMS) was done on kan12 pkl-12 plants to search for other redundant genes in establishing carpel polarity. From this screen, four different enhancer loci were identified: crc, hasty (hst), splayed (spl), and kanadi2 (kan2) [31–33]. Although most genetic screens in floral development have been used to identify enhancers, a good example of a suppressor screen is the identification of CORYNE (CRN) as an essential component of the stem-cell restricting CLAVATA3 (CLV3) signaling pathway [34]. Since increased CLV3 signaling arrests meristem function leading to a facile phenotypic output, Muller and colleagues performed an EMS mutagenesis experiment with a mild CLV3-overexpressing line such that the meristem arrest is not that severe, so the plants are able to set seed [34, 35]. A mutant was isolated and found to be a crn-1 allele, which has an aberrant silique shape reminiscent of clv mutants. Interestingly, the flowers in crn-1 have enlarged gynoecium and some flowers have additional sepals or petals, probably caused by an increase in floral meristem size [34]. Initially, CRN was thought to be a receptor kinase that aids

134

Zhigang Huang et al.

EMS

ag-4 ag-4 hua1 hua2: stamens petals

ag-4: Indeterminate floral meristem: no carpel, stamen present

EMS

hua1 hua2 hua1 hua2: No obvious phenotype

hua1 hua2 hen1: Loss of floral determinacy; stamens petals

hua1 hua2 ago10: Loss of floral determinacy

hua1 hua2 ag-10: Loss of floral determinacy; stamens petals

ag-10 ago10: Loss of floral determinacy

ag-10 clf: Loss of floral determinacy

ag-11: Loss of floral determinacy

EMS

ag-10 ag-10: Similar to wild type

EMS

ag-11

ag-11: Loss of floral determinacy

ag-11 ap2: partially suppressed the floral determinacy defect

Fig. 1 EMS Mutagenesis is a Powerful Tool to Dissect Pathways Governing Floral Development. An outline of several EMS-based enhancer and suppressor screens used to dissect the floral determinacy pathway governed by AG. A relatively weak allele of AG, ag-4, is defective in carpel identity and floral determinacy; however, stamens are still present. EMS mutagenesis was performed on ag-4 seeds, resulting in the identification of the genes HUA1 and HUA2, mutations that convert the stamens in ag-4 to petals. The hua1 hua2 mutant has weak organ identity defects but is normal in floral determinacy. Another EMS mutagenesis experiment was performed on hua1 hua2 and another ag allele was isolated (ag-10), as well as mutations in HEN1 and AGO10. ag-10 has a very weak phenotype, only a few siliques in a plant are slightly bulged, indicative of loss of determinacy. Another EMS mutagenesis experiment was performed on ag-10, and mutants in genes such as CLF and AGO10 were isolated. Thus, CLF and AGO10 are involved in floral determinacy. In addition to this, an intermediate-strength AG allele, ag-11, was also isolated and further used for suppressor screens. Two suppressors with longer and thinner gynoecia were found to harbor loss-offunction mutations in AP2. This suggests AP2 antagonizes AG activity in controlling floral determinacy

in the transmission of the CLV3 signal [34]; however, it was later found to be a pseudokinase and is hypothesized to play a scaffolding role, perhaps to aid the export of CLV2 to the plasma membrane and/or to assemble higher-order CLV1 or CLV1-substrate complexes [36]. Another suppressor screen was done on the

Genetic Screens: Enhancers and Suppressors

135

mutant of HAWAIIAN SKIRT (HWS), which exhibits fused sepals and increased organ size. From this screen, a mutant that carries a nucleotide substitution in the miR164 binding site of CUPSHAPED COTYLEDON 1 (CUC1) mRNA was isolated. Further genetic analyses demonstrate that HWS controls size and floral organ number by modulating CUC1 and CUC2 expression [37]. ag-11 is an intermediate-strength AG allele, which compromises the floral determinacy but not the organ identity functions of AG. Dr. Chen and her group performed an ag-11 genetic screen for mutations that suppressed the floral determinacy defect. Two suppressors with longer and thinner gynoecia were found to harbor loss-of-function mutations in AP2. This suggests AP2 antagonizes AG activity in controlling floral determinacy [27]. In addition to the three pathways mentioned above, modifier screens have been used for the isolation of mutants in DORN¨ SCHEN-LIKE from the pi-5 background [38]; mutants in RO three FUSED FLORAL ORGANS genes from a ufo (unusual floral organs) allele [39]; and mutants in SPLAYED (SYD) from a weak leafy mutant background [32]. Thus, the aforementioned experiments highlight the power of modifier screens in teasing apart the molecular mechanisms governing different aspects of floral development. There have been several diverse reviews [40, 41] as well as protocols for EMS mutagenesis [42]; the present chapter therefore offers a comprehensive view of both EMS and insertional mutagenesis and outlines methods used to identify the mutation (s) responsible for the aberrant floral phenotypes. The chapter is divided into two parts. First, we provide facile protocols for two methods of mutagenesis using EMS and T-DNA as mutagens. Second, we describe three protocols used to identify the molecular lesion responsible for the mutant phenotypes. 1.1 Mutagenesis of Arabidopsis

Due to its small genome size, fecundity, and ease of transformation, Arabidopsis is the ideal organism to perform genetic mutagenesis screens. First, we discuss two major, divergent agents of mutagenesis: EMS and T-DNA.

1.1.1 EMS Mutagenesis of Arabidopsis

EMS is an alkylating agent that, ~90% of the time, induces C/G to T/A substitutions [43, 44]. At a low frequency, EMS can also generate G/C to C/G or G/C to T/A transversions by 7-ethylguanine hydrolysis or A/T to G/C transition by 3-ethyladenine pairing errors [43]. In Arabidopsis, the frequencies of EMS-induced stop codon and missense mutations are about ~5% and ~65%, respectively [45]. EMS mutagenesis allows the identification of loss- and gain-of-function mutants and can help researchers to understand the function of specific amino acids within a protein. EMS mutagenesis tends to lead to plants containing more than one mutation (this issue will be addressed later). However, to ensure adequate coverage of the genome, meaning that one will

136

Zhigang Huang et al.

LB

Selectable Marker

RB

RB-0a RB-1a RB-2a

Unknown target flanking sequence mLAD AC1 AC0

Fig. 2 Schematic diagram depicting the identification of a T-DNA insertion site. A basic schematic of the T-DNA insertion region and location of all primers used to identify the site of integration is shown. The dark grey boxes represent the right and left border (RB and LB, respectively). The light grey portion represents the region where the selectable marker can be placed. The T-DNA region (the region that will be integrated into the host genome) encompasses everything from the LB to RB (with the selectable marker). The mLAD1!4 primers are each used with RB-0a and AC0 in the four initial mhiTail-PCR reactions. AC1 in conjunction with RB-1a and RB-2a primers are used in subsequent nested PCR reactions

get a mutation in every single gene, a large screening population must be obtained. A study showed that saturation could theoretically be achieved if 135,000 M1 lines (five-fold coverage) are obtained [46]. In addition, the desired phenotypic output plays a role in saturation levels as well as the amount and time the seeds were exposed to EMS. 1.1.2 T-DNA Insertional Mutagenesis

Another common mode of mutagenesis is the use of Agrobacterium-mediated transformation via the floral dip method [47]. Transfer-DNA (T-DNA) transformation is a phenomenon by which Agrobacterium tumefaciens inserts a portion of its Ti plasmid into the host (plant) genome, usually causing infection. Scientists have used this natural machinery and manipulated it to transfer their gene of interest or, in cases of random mutagenesis, a selectable marker placed in lieu of the normal T-DNA, into Arabidopsis. The T-DNA region is flanked by two 25-bp repeat regions called the left border (LB) and right border (RB) (see Fig. 2). T-DNA integration occurs through illegal recombination by utilizing the plant DNA double-stranded break repair system. Several models of integration have been proposed; however, there is a consensus that integration is random and is mediated by the LB and RB [48]. T-DNA integration is higher at gene-rich regions than centromeric regions. Moreover, in actively transcribed genes, integration is higher around the transcription initiation and termination sites than in coding regions [49–51]. Both EMS and T-DNA insertional mutagenesis have been widely used as tools to probe diverse biological pathways in Arabidopsis. However, there are advantages as well as caveats for each technique. EMS mutagenesis is performed on seeds, making the procedure easy to perform; however, there is a low statistical probability that any one gene will be mutated in any one plant. Therefore, a copious amount of seeds needs to be processed in each mutagenesis. Furthermore, the kill rate increases exponentially with dose, while the mutation rate for any single gene rises linearly. Hence, achieving saturation in a genetic screen is quite difficult. In

Genetic Screens: Enhancers and Suppressors

137

addition, multiple mutations may incur within a single plant, so that backcrossing is necessary to ensure the elimination of background mutations unrelated to the phenotypes of interest. One advantage of EMS mutagenesis is that it can produce viable, weak alleles in genes whose function is essential to the plant. In contrast, T-DNA mutagenesis often produces loss-of-function mutants as integration of the T-DNA usually perturbs the gene’s function. T-DNA mutagenesis can also be used to overexpress genes that are located close to the T-DNA insertion if the T-DNA harbors transcriptional enhancers [52]. An advantage of T-DNA mutagenesis compared to the use of EMS is that T-DNA insertion sites can often be readily identified using PCR-based approaches. A possible disadvantage is that it can result in chromosomal rearrangements, such as inversions or deletions. Moreover, multiple insertions and complex T-DNA loci may incur [53]. In addition, not all mutations generated through T-DNA transformation are actually caused by T-DNA insertion events [53]. Linkage studies or rescue experiments are necessary to determine whether a mutant phenotype is caused by a T-DNA insertion. After mutagenesis and isolation of a mutant with the desired phenotype, the next step is to identify the mutation(s) responsible for the phenotype. 1.2 Identification of the Mutation Responsible for the Observed Mutant Phenotype

There are several experimental strategies that can be employed for the identification of a mutation, which causes a mutant phenotype. These include: map-based cloning (mainly used for EMS-induced mutations), TAIL-PCR (for T-DNA insertion mutants), as well as genome re-sequencing (for both EMS and T-DNA induced mutations).

1.2.1 Map-Based Positional Cloning

Map-based positional cloning is the most common method used to identify mutations after mutagenesis. It is a PCR-based method relying on markers, such as simple sequence length polymorphisms (SSLPs); cleaved amplified polymorphic sequences (CAPS); and derived CAPS (dCAPS). Many such markers have been described [54], but several new markers have been designed by our laboratory, which allow a better distinction between sequences derived from Ler and Col-0 accessions (see Table 1). The underlying premise of this technique is to cross the mutant of interest into a different accession. In the F2 generation, in which the mutant phenotype is observed, one can determine where the mutation lies due to innate sequence differences between the accessions as markers in genetic mapping. This technique is often laborious and time-consuming. Theoretically, 300–400 plants are sufficient to narrow down the region where a mutated gene is located. However, in practice, thousands of plants may need to be processed.

F21M12-F 1CER458676F F28H19-F F24O1-F 1CER470018F

T23K3-F 2CER449854F T20P8-F T3D7-F

F24P17 K1G2-F CIW4-F T20O10-F

CIW5-F CIW6-F 4CER450255F AP22-F

T31P16-F CIW8-F ATPHYC-F MJB21-38F MRB17-60F

CHR1

CHR2

CHR3

CHR4

CHR5

Primer name

TCGAAGTAACTTACTTTCTA TAGTGAAACCTTTCTCAGAT CTCAGAGAATTCCCAGAAAAATCT CGATGCTCAGGTTCTACATT CGAGCAAATGAATCTGAAGG

GGTTAAAAATTAGGGTTACGA CTCGTAGTGCACTTTCATCA CACAAGACAACACCAAAAAC ATTATGTTAGGAAAATGAGAT

ACTGCACATTGCACGAACA ATGAGCTTTAGGAGTGTGTA GTTCATTAAACTTGCGTGTGT AAATGCCAGGGGAATAGA

CGTGTTTACCGGGTCGGA GACGACTTCGAGAAAGTTACAAAAC TCCGATTCGATTAAACTC GGTATCGATTGAGCAAATAA

GGCTTTCTCGAAATCTGTCC CGTTTGAAACACCTACAGGATTAC TGCGGGAGTGTGATAGAATA TCACAAAAATGCAACATTTA GATCATATTCTATTGCACCCATCAG

Sequence (50 !30 )

Table 1 Oligonucleotide sequences for map-based positional cloning

T31P16-R CIW8-R ATPHYC-R MJB21-38R MRB17-60R

CIW5-R CIW6-R 4CER450255R AP22-R

F24P17 K1G2-R CIW4-R T20O10-R

T23K3-R 2CER449854R T20P8-R T3D7-R

F21M12-R 1CER458676R F28H19-R F24O1-R 1CER470018R

Primer name

AATGTCGCAAAGACTTCC TTATGTTTTCTTCAATCAGTT AAACTCGAGAGTTTTGTCTAGATC ACTAAAATATCATCTCGTTGTA GTGATAATTCGTAAATATGGACT

AGATTTACGTGGAAGCAAT CACATGGTTAGGGAAACAATA AGAAGGAATGGCTTCATCTA GCGTTGTAAGAATTAAGAA

GGATGGCAACTTAGGCTGAA AATTTTGTCCCAAAAGAATA TACGGTCAGATTGAGTGATTC CAAACCATGCAATGATGC

AAAACCCTTGAAGAATACG AACGGAGTAATAACTCCAATCTCATC TTATTTCCTATTTCAAGACT ACATGCGTCTGCTTGGAG

TTACTTTTTGCCTCTTGTCATTG AGCTCAAACTTAAGCATAGAAACC TCCTCGAAAGATTCATTGAT TAATGGCTCCAATCAATA CTTACGCGTTTACTGTCTCATGTTT

Sequence (50 !30 )

AT5G10040 AT5G22550 AT5G35840 AT5G42720 AT5G54630

AT4G01710 AT4G13575 AT4G20095 AT4G36780

AT3G06400 AT3G27460 AT3G50820 AT3G62988

AT2G01860 AT2G13851 AT2G27130 AT2G47160

AT1G09880 AT1G27520 AT1G43763 AT1G62370 AT1G71930

Position

138 Zhigang Huang et al.

Genetic Screens: Enhancers and Suppressors

139

Table 2 Oligonucleotide sequences for mhiTAIL-PCR Name

Sequence (50 !30 )

mLAD1

GCTCACGATGGACGCTGAGTGGCACCTG(G/C/A)N(G/C/A)NNGGAA

mLAD2

GCTCACGATGGACGCTGAGTGGCACCTG(G/C/T)N(G/C/T)NNCCTT

mLAD3

GCTCACGATGGACGCTGAGTGGCACCTG(C/T/A)N(C/T/A)NNAACC

mLAD4

GCTCACGATGGACGCTGAGTGGCACCTG(G/T/A)N(G/T/A)NNTTGG

AC0

GAGCTCACGATGGACTGC

AC1

CGATGGACTGCTGAGT

RB-0a

ccc cga tcg ttc aaa cat ttg gc

RB-1a

CGATGGACTGCTGAGTGGCACCTGttgccggtcttgcgatgat

RB-2a

GCgcatgacgttatttatgagatgggt

1.2.2 Thermal Asymmetric Interlaced Polymerase Chain Reaction (TAIL-PCR)

One of the advantages of T-DNA insertional mutagenesis is that since the RB and LB sequences are known, TAIL-PCR can be performed to discover the genomic position of the T-DNA insertion. TAIL-PCR was first described in 1995 by Liu and Whittier [55], and the latest updated version is mhiTAIL-PCR [56]. In summary, mhiTAIL-PCR utilizes nested primers in three consecutive reactions to find the insertion site [56]. Four initial PCR reactions are conducted, each with a modified long arbitrary degenerate (mLAD) primer (one each from mLAD1!4) in combination with primer AC0 and RB-0a. Products from these reactions are then utilized as a template in a second round of PCR reactions with a primer termed AC1 (AC1 partially overlaps with the mLAD primers, except it does not contain the degenerate sequences) and the primer RB-1a from the T-DNA region (see Fig. 2 and Table 2). A third round of PCR reactions are conducted with products from the second round as templates and the AC1/RB-2a primer pair (see Fig. 2 and Table 2). The products from the second and third rounds of PCR are resolved in an agarose gel and products containing a single band and with the correct band patterning (the product from the third round of PCR should run faster than that from the second round of PCR) are sequenced using RB-2a or a new primer nested to RB-2a; the other side can be sequenced using AC1. The advantages of TAIL-PCR are that it is a simple, sensitive, efficient, and fast method; that it has high specificity; and that the PCR product can be directly sequenced [55]. However, the major disadvantage is that this procedure will not work if multiple copies of the T-DNA are inserted in different positions in the genome, are tandemly inverted at the same position, or if the T-DNA is only partially inserted. Thus, it is important that genetic analysis be performed to ensure single locus insertion prior to performing TAIL-PCR. However, this does not help if multiple copies are

140

Zhigang Huang et al.

inserted within a single locus and/or the copies are inverted. Further, the vector backbone may also get transferred or the insertion may cause chromosomal rearrangements. In addition, there may not be linkage between the T-DNA and the phenotype, or the T-DNA may be partially inserted [53]. Thus, in these cases, map-based positional cloning or deep sequencing is utilized to identify the gene that, when mutated, leads to the observed phenotype. 1.2.3 Mapping by Deep Sequencing

As mentioned above, the map-based cloning method is commonly used to identify mutations causing a phenotype. However, there are some problems that prevent this method from being used in all cases. One, if the phenotype studied is very sensitive to the genetic background, then map-based cloning would not be applicable. For example, the phenotype may become suppressed when crossed to another accession, so that F2 mutant plants cannot be identified. Also, if the mutation occurs in a region where the recombination rate is low, thousands of plants will be needed to locate the mutation, which makes the process very tedious and time-consuming. With technological advances, another method is available to identify genes from mutagenesis experiments: high-throughput DNA sequencing. The identification of EMS-induced mutations responsible for the observed phenotype can be accomplished in three manners using high-throughput sequencing. First, bulk analyses of an F2 mapping population can be utilized by following the same rules as map-based cloning. DNA from pooled F2 plants of the mutant phenotype is used to produce a library for deep sequencing, which provides information on the segregation of various SNPs relative to the mutation. Linkage to certain SNPs helps to narrow down the region containing the mutation. Searching for genes with point mutations in this region then helps identify candidate genes. Three pipelines, SHOREmap (http://bioinfo.mpipz.mpg.de/ shoremap/), NGM (http://bar.utoronto.ca/NGM/) and Mutmap (https://genome-e.ibrc.or.jp/resource/mutmap), have been developed to locate the mutation caused by EMS using bulk analysis of an F2 mapping population [57–59]. It is suggested to use around ~500 plants when using SHOREMAP, around 80 plants when using NGM and around 30 plants when using Mutmap. Second, bulk analysis of an F2 backcross population can also be used to identify the mutation in question. Locating the position of the mutation responsible for the mutant phenotype relies on the fact that other EMS-generated mutations in the genome segregate according to their linkage (or not) with the mutation in question. SHOREMAP provides an algorithm to locate a mutation using an F2 backcross population. To rule out the non-causal mutations, it is suggested to sequence pooled DNA from wild-type-looking plants as well [60].

Genetic Screens: Enhancers and Suppressors

141

Finally, the phenotype-causing mutation can also be found through deep sequencing of the mutant itself, which has already been used in our and several other laboratories [61]. Since the rough location of the mutation is not known, there may be too many candidates. Several practices may help to filter the candidates. Backcrosses prior to sequencing can reduce the number of noncausal mutations. Preliminary rough mapping information helps to limit the causal mutation to a region. Sequencing the parental line from which the mutant was derived also helps to eliminate nonphenotype-causing mutations. Sequencing multiple mutant alleles in the same gene also helps pinpointing the gene of interest. To identify the mutations caused by T-DNA transformation, there are two possible scenarios. For mutations caused by the T-DNA insertion, the affected gene could be easily located by the reads containing both genomic DNA and T-DNA sequences. To confidently identify such chimeric reads, it is better to sequence at 100 cycles (rather than at 50). For mutations caused by T-DNA transformation but that are not linked to a T-DNA insertion, the identification of candidate mutations is similar but more complicated than for EMS-generated mutations since three types of mutations: big INDELs (insertion and deletion), small INDELs, or SNPs (single nucleotide polymorphisms) need to be considered. To locate the region containing the mutation, the mapping can be done in the same manner as that for EMS-generated mutations. For the identification of the mutation from the region of interest, the three types of potential changes all need to be considered. Using deep sequencing technology to find a mutation is not limited to the accessions with sequenced genomes. As long as the species has a reference genome, deep sequencing can be applied to mutants derived from un-sequenced accessions. Which methods to use and how to analyze the data depend on the situation and the specific experiment at hand. Despite these variables, the process of building the DNA sequencing libraries can be done in the same way, which will be described in this chapter.

2

Materials

2.1 Mutagenesis, Mutant Screening, and Initial Mapping of Mutations

1. 1 mL of Arabidopsis seeds (see Note 1).

2.1.1 EMS Mutagenesis of Arabidopsis

5. 5 M NaOH. Prepare 0.5 M NaOH by mixing 146.1 g of NaOH with 400 mL of water in a beaker containing a magnetic stir bar. Stir until dissolved, and bring volume up to 500 mL with water. Autoclave the solution for 25 min.

2. Twenty trays of soil with 12 pots per tray. 3. Disposable 50 mL conical tubes. 4. Parafilm.

142

Zhigang Huang et al.

6. Tween-20 solution: 0.1% (v/v) Tween-20. Prepare 50 mL using autoclaved H2O. 7. Ethyl methanesulfonate (EMS) solution: 0.2% EMS (e.g., from Sigma-Aldrich, M0880) in autoclaved H2O (see Note 2). 8. 0.1% agar. Add 0.9 g of agar to 90 mL of autoclaved H2O. Microwave to dissolve the agar and leave to cool at room temperature prior to use (see Note 3). 2.1.2 Planting EMS Mutagenized M1, M2, and Mapping Population Seeds

1. Soil.

2.1.3 Preparing DNA for Map-Based Positional Cloning

1. Toothpicks and tape for labeling.

2.1.4 CTAB DNA Extraction

1. CTAB extraction buffer: 2% (w/v) cetyltrimethylammonium bromide, 1.4 M NaCl, 20 mM EDTA, 0.2% (v/v) β-mercaptoethanol, 200 mM Tris-HCl pH 8.0. For 500 mL buffer: 10 g of CTAB, 100 mL of Tris-HCl pH 8.0 (to make a 500 mL 1 M Tris-HCl stock solution, add 60.55 g of Tris base to 400 mL of H2O, add ~21.1 mL of concentrated HCl (pH 8.0), add water to a final volume of 500 mL and autoclave for 25 min), 40.95 g of NaCl, 20 mL of 0.5 M EDTA (to make a 500 mL 0.5 M EDTA stock solution, add 84.05 g of EDTA to 250 mL H2O, add 5 M NaOH (~71 mL) slowly while stirring until the EDTA dissolves, pH should be 8, bring up to volume with H2O and sterilize by autoclaving for 25 min). Prior to use, calculate how much CTAB buffer you will need, transfer that volume to another tube and add 1/500 volume of β-mercaptoethanol (see Note 4).

2. Mortar and pestle. 3. Liquid nitrogen.

2. Chloroform. 3. Isopropanol. 4. Nanodrop spectrophotometer or any means to quantify the DNA. 2.1.5 PCR

Quick and Dirty

1. Quick and Dirty PCR Extraction Buffer: 200 mM Tris-HCl pH 7.6 (to make a 500 mL 1 M Tris-HCl stock solution, add 60.55 g of Tris base to 400 mL of H2O, add ~28.5 mL of concentrated HCl (pH 7.6), add water to final volume of 500 mL and autoclave for 25 min), 250 mM NaCl, 25 mM EDTA, 0.5% (w/v) SDS (to make a 10% (w/v) SDS stock solution, add 50 g in 500 mL of distilled water, stir overnight) (see Note 5). 2. Small pestles for grinding tissue in 1.5 mL centrifuge tubes.

Genetic Screens: Enhancers and Suppressors 2.1.6 T-DNA Insertional Mutagenesis: Agrobacterium-Mediated Transformation

143

1. Flowering Arabidopsis thaliana (age is dependent on growth conditions) (see Note 6). 2. Electrically competent Agrobacterium tumefaciens cells. 3. Plant transformation vector containing the T-DNA construct of interest. 4. Plates for plant growth, and antibiotics/herbicides needed for selection of transgenic lines. 5. 5% (w/v) sucrose solution 6. Silwet L-77. 7. Trays for floral dipping. 8. Spectrophotometer or other method of measuring cell densities. 9. MS medium, LB medium, and agar or other growth media. 10. Dark area and/or cover to keep plants away from bright light. 11. Gene PulserTm (Bio-Rad) or another suitable instrument for Agrobacterium transformation, and corresponding electroporation cuvettes.

2.2 Pinpointing the Mutation That Causes the Phenotype

1. 3–4% agarose gels (see Note 7).

2.2.1 Map-Based Positional Cloning: PCR

3. 2.5 mM dNTPs

2. 10 PCR buffer: 15 mM MgCl2, 500 mM KCl, 100 mM TrisHCl (pH 8.3) 4. Taq DNA polymerase (see Note 8). 5. 10 DNA loading buffer: mix 3.9 mL of glycerol, 500 μL of 10% (w/v) SDS, 200 μL of 0.5 M EDTA, 0.025 g of bromophenol blue, and 0.025 g of xylene cyanol, bring to 10 mL final volume with distilled H2O.

2.2.2

TAIL-PCR

1. Phanta Super-Fidelity DNA Polymerase (Vazyme, cat#P501d1). 2. Gel DNA Recovery Kit (e.g., Zymo cat#D4001).

2.2.3 Mapping by Deep Sequencing

1. 5 μg RNase-treated, high-quality genomic DNA (see Note 9). 2. Bioruptor (Diagenode, #UCD-200 TO). 3. E-Gel™ Power Snap (Invitrogen, G8100). 4. E-Gel™ 1% Agarose with SYBR™ Safe (Invitrogen, A45202). 5. E-Gel® 1 kb Plus DNA Ladder (Invitrogen, 10488-090). 6. Agencourt AMPure XP 60 mL kit (Beckman Coulter Genomics, # A63881). 7. TruSeq DNA PCR-Free Sample Preparation Kit (Illumina, FC-121-3001).

144

Zhigang Huang et al.

8. Magnetic stand-96 (Invitrogen, # AM10027). 9. 96-well thermal cycler with heated lid. 10. 96-well 0.3 mL DNase-free PCR plate. 11. Ethanol (200 proof).

3

Methods

3.1 Mutagenesis, Mutant Screening, and Initial Mapping of Mutations 3.1.1 EMS Mutagenesis of Arabidopsis

1. Place seeds (100 μL; corresponding to ~1000 seeds) in a 50 mL conical tube and wash in 30 mL of 0.1% (v/v) Tween-20 for 15 min at room temperature (RT). Keep the tube shaking constantly (see Note 10). 2. Spin down at 400  g for 1 min at RT. 3. Remove as much solution as possible first using a transfer pipet and then a pipetman. 4. Add up to 15 mL of autoclaved water. 5. Add 30 μL of EMS to the tube and wrap the cap with parafilm (see Note 11). 6. Mix by inverting the tube several times, and incubate at RT (while shaking) for 8–12 h. 7. Spin the tube at 400  g for 1 min at RT. 8. Remove the EMS solution, and put it in a separate 50 mL conical tube. 9. Add 5 M NaOH to the tube so that the final concentration is 0.5 M NaOH. Put parafilm on the tube to seal it and incubate, shaking at RT, overnight. Discard the parafilm and solution into a biohazardous waste container. 10. Rinse the seeds ten times in 20 mL of autoclaved water each (see Note 12). 11. After the last rinse, add 10 mL of autoclaved water and incubate the seeds at RT, while shaking, for 2–4.5 h. 12. Pour off the water, add an additional 10 mL of water to the tube, and using a transfer pipet, transfer water and seeds to a flask containing 90 mL of 0.1% agar (see Note 13). 13. Mix and sow the seeds at 500 μL per pot with a pipetman using a 1000 μL tip. 14. Cover the trays of soil with a lid and leave at 4  C for 1 week. 15. Transfer the trays to the growth chamber and grow at 23  C under continuous light. 16. After the plants have developed two leaves, remove the lid (see Note 14).

Genetic Screens: Enhancers and Suppressors 3.1.2 Planting EMS Mutagenized M1, M2, and Mapping Population Seeds

145

1. As the M1 plants are growing, check the plants and look for dominant mutations. If a dominant mutant is identified, proceed to step 10. 2. Harvest M2 seeds from the individual M1 plants. Place seeds from each plant into a separate tube (see Note 15). 3. Poke a small hole in the lid of the tube and leave the seeds to dry for 1 week at RT. 4. Plant the seeds one by one (24–30 seeds per pot), one M2 family for one pot. 5. Place trays at 4  C for 1 week so the seeds can stratify. 6. Transfer the trays to the growth chamber at 23  C under continuous light (see Note 16). 7. When the plants grow to the two-leaf stage, remove the lid. 8. As the plants flower, assay the plants for the desired floral phenotype (either enhancer or suppressor). 9. Once a mutant phenotype is observed, note the segregation ratio of that phenotype in the pot (see Note 17). 10. Backcross the mutant plants to the parental line at least twice before any molecular analysis is performed (see Note 18). 11. Cross the mutant plant to another plant (from a different ecotype) to generate the mapping population (see Note 19). 12. Generate a “Het” (heterozygous) sample as a control. For example, if the screen was performed in the Ler background, cross the mutant plant with another accession (that you will use for mapping, such as Col-0). With the F1 plants, collect the tissue (several leaves) and extract DNA via the CTAB method or the “quick and dirty method” (for Quick and Dirty PCR prep) (see Note 20). This sample will be included in subsequent mapping reactions as a control. 13. Using the F2 seeds (from the cross with another accession (same accession that you chose for step 12)), plant (24 seeds per pot) the mapping population. Keep in mind that you will, on average, get only 1 out of 4 plants that exhibit the phenotype. For a population of 50 mutant plants, which is needed for a rough mapping of the mutation, at least 200 seeds need to be planted. 14. Place trays at 4  C for 1 week so the seeds can stratify. 15. Transfer trays to the growth chamber at 23  C under continuous light. 16. When the plants grow to the two-leaf stage, remove the lid.

146

Zhigang Huang et al.

3.1.3 Preparing DNA for Map-Based Positional Cloning

1. Within the F2 mapping population, look for plants with your desired phenotype. If it is a dominant mutation, look for and perform the map-based positional cloning on plants without the desired phenotype (see Note 21). 2. Label the plants with toothpicks, and number them. For the rough mapping, 50 plants are needed, which can be pooled (put small samples from many plants in one tube and isolate DNA at the same time). For the pooled method, CTAB DNA extraction is performed for rough mapping. Otherwise, use the Quick & Dirty method on individual plants. For fine mapping, collect tissue from each plant separately, using the Quick & Dirty method. 3. Collect one leaf and place it either in a 15 mL conical tube for the CTAB DNA extraction (in liquid nitrogen) or individually in 1.5 mL centrifuge tubes. 4. Remember to collect tissue from both ecotypes for use as controls (see Note 22).

3.1.4 CTAB DNA Extraction

1. Cool the mortar and pestle in liquid nitrogen. 2. Add frozen tissue to the cold mortar and pestle. 3. Grind the tissue until the powder is light green. 4. Add the tissue to 10 mL of CTAB extraction buffer (with β-mercaptoethanol added) and invert tube several times (see Note 23). 5. Place the tubes at 65  C for 40–60 min. Label two sets of tubes while waiting. 6. Centrifuge for 7 min at 11,648  g at RT. 7. Transfer the supernatant into a new 15 mL tube. 8. Add an equal volume of chloroform and vortex for 10 s (see Note 24). 9. Centrifuge for 15 min at RT at 11,648  g. 10. Transfer the upper phase to a new tube (see Note 25). 11. Add an equal volume of cold isopropanol (see Note 26). 12. Place the tubes in 80  C for 30 min or 20  C overnight (see Note 27). 13. Centrifuge tubes for 30 min at 4  C at 11,648  g. 14. Decant the supernatant into a liquid waste container. 15. Wash pellet twice with 4 mL of 70% EtOH (see Note 28). 16. Invert the tube, and leave the tube in the hood to air dry (see Note 29). 17. Once the pellet is dry, add 100 μL of autoclaved water for resuspension (see Note 30).

Genetic Screens: Enhancers and Suppressors

147

18. If you wish to remove RNA, add RNase A to a final concentration of 20 μg/mL and incubate at 37  C for 1 h. If not, continue with step 29. 19. Add water to a final volume of 300 μL. 20. Add 150 μL phenol (high pH) and 150 μL chloroform. 21. Briefly vortex. 22. Spin in a tabletop microfuge at 16,100  g for 15 min. 23. Transfer supernatant to a new, clean tube and add an equal volume of isopropanol. 24. Leave at 80  C for 30 min (see Note 31). 25. Spin in a tabletop microfuge at 16,100  g for 25 min at 4  C. 26. Wash twice with 70% EtOH, as in step 15. 27. Invert and leave the tube to air dry in the hood. 28. Resuspend in 100 μL of autoclaved water, as in step 17. 29. Quantify the DNA. 30. Store DNA at 20  C (see Note 32). 3.1.5 Quick and Dirty Extraction

1. Grind tissue in a 1.5 mL Eppendorf tube (see Note 33). 2. Add 400 μL of Quick and Dirty extraction buffer to the tube. 3. Grind a little more until you do not see chunks of tissue. 4. Spin in a tabletop microfuge at 16,100  g for 5 min at RT. 5. Transfer the supernatant to a new tube (see Note 34). 6. Add an equal volume of isopropanol. 7. Leave the tubes at RT for 15 min (see Note 35). 8. Spin down the samples in a tabletop microfuge for 10 min at RT at 16,100  g. 9. Decant the supernatant. 10. Wash pellet twice with 150 μL of 70% EtOH. 11. For the last wash, spin the tubes in a tabletop microfuge for 5 min at RT at 16,100  g. 12. Decant the supernatant and leave the inverted tubes/plates to dry in the hood. 13. Add 50 μL of autoclaved water to the pellet. 14. Leave DNA at RT for 1 h if performing the PCR on the same day. Use 1 μL for a PCR reaction with a 10 μL final volume. 15. Store the DNA at 20  C.

3.1.6 T-DNA Insertional Mutagenesis: Transformation and Selection of Agrobacterium

1. Add 0.5 μL (2–10 ng) plasmid to 50 μL of electrically competent Agrobacterium cells on ice (see Note 36). 2. Cool a 0.2 cm electroporation cuvette on ice.

148

Zhigang Huang et al.

3. Add the entire plasmid and competent cell mixture to the cuvette. 4. Place the electroporation cuvette in the Gene PulserTm. Make sure to select the setting for bacteria, and press the pulse button. Remove the cuvette from the machine and place it on ice. 5. Pipette the mixture carefully out of the cuvette and into a 1.5 mL tube containing medium (LB or other). This should be done in the flow hood to help reduce contamination. 6. Shake for 1–3 h in a 28  C shaker. 7. Pipette the mixture carefully out of the tube and onto a selection plate containing the appropriate medium (LB or other) and antibiotic. Make sure the mixture is spread evenly on the surface of the plate. Let it air dry in the flow hood. 8. Incubate this plate overnight at 28  C or until colonies form (see Note 37). 9. Prepare a test tube of 5 mL of medium of choice and antibiotic. Inoculate a colony into the liquid medium using a pipette tip by touching the colony. Then, dip the tip end into the medium, pipetting if desired. The tip can also be placed into the medium if necessary. 10. Incubate the 5 mL tube on a 28  C shaker overnight. 11. The presence of the desired DNA construct can be verified by colony PCR (see Note 38). 12. Store Agrobacterium if needed (see Note 39). 3.1.7 T-DNA Insertional Mutagenesis: Preparation of Agrobacterium

1. The next morning, pour the 5 mL Agrobacterium culture into 500 mL (1:100 dilution) of LB, or other growth medium. Make sure to add the antibiotics for selection. Grow at 28  C until OD600 ¼ 0.8 (see Note 40). 2. Spin down the Agrobacterium (5500  g, 15 min at RT) and resuspend in a 5% sucrose, 0.5 MS solution, pH 5.7. For example, if the OD600 ¼ 0.8 in 500 mL LB, the OD600 will also be 0.8 in 500 mL of 5% sucrose solution. However, only about 200–300 mL of solution is needed for dipping of 1–3 trays of Arabidopsis (see Note 41). 3. Add 0.05% (v/v) of Silwet L-77 to the solution and mix well (see Note 42). 4. Pour the solution into a low, flat tray. This solution does not need to autoclaved if used quickly.

3.1.8 T-DNA Insertional Mutagenesis: Floral Dip

1. Dip the shoots of plants into the solution, and let sit for 5 min. Do not let the soil touch the solution (see Note 43).

Genetic Screens: Enhancers and Suppressors

149

2. If transforming Ler plants, you must put the solution and pot (from step 1) into a vacuum for 5 min. Release the vacuum slowly after 5 min. This step can be skipped for Col-0 plants, which are easier to transform. 3. Spray the shoots of the plants with water, using a spray bottle, for several seconds to remove the Sucrose solution. 4. Cover the plants with a plastic dome or other covering for about 24 h. Plants are usually stored on their sides during this time. 5. Return plants to the upright position and grow them until they are ready for seed harvesting. 6. Collect seeds from the plants and seeds from many plants can be pooled. 3.1.9 T-DNA Insertional Mutagenesis— Selection of Transformants

1. Grow the seeds on soil and spray with herbicide such as ammonium-glufosinate (BASTA) to select transformants. Remember to stratify the seeds for 2–5 days after planting. 2. To select transformants with an antibiotic, prepare plates of 0.5 MS and 0.6% agar with the antibiotic added in the correct amount. Grow under continuous light for 7–10 d. 3. Positive transformants will appear larger than seedlings that were not transformed. 4. Transfer transformants to soil, and collect seeds from either a single plant or a few plants (5–50).

3.2 Pinpointing the Mutation That Causes the Phenotype 3.2.1 Map-Based Positional Cloning—PCR

1. The PCR reaction mix is as follows for a 10 μL reaction: 1 μL of Quick and Dirty DNA (or 50 ng of CTAB DNA), 0.1 μL each of 10 mM forward and reverse primer, 0.4 μL of 2.5 mM dNTPs, 1 μL of 10 PCR buffer (contains 18 mM MgCl2), 0.1 μL of Taq DNA polymerase, and 7.3 μL of autoclaved water (see Note 44). 2. The PCR reaction conditions are as follows: step 1: 94  C for 3 min; step 2: 94  C for 30 s; step 3: 54  C for 30 s; step 4: 72  C for 30 s; repeat steps 2–4 for an additional 34 times; step 5: 72  C for 10 min. 3. Run the PCR product on a 3% or 4% agarose gel. Make sure to add EtBr before pouring the gel since EtBr is not easily absorbed into a higher percentage gel after the gel is made. Be careful when handling EtBr, it is a mutagen. The lanes should be set up as follows: pooled DNA from mutant plants; Ler; Col-0; DNA from F1 plants or Ler and Col DNA mixed in a 1:1 ratio.

150

Zhigang Huang et al.

Table 3 Example of analysis of F2 mapping population Marker 1 Marker 2 Marker 3 Ciw8 Marker 4 Marker 5

3.2.2 Map-Based Positional Cloning— Analysis

Plant 1

Het

Ler

Ler

Ler

Ler

Ler

Plant 2

Het

Het

Ler

Ler

Ler

Het

Plant 3

Het

Ler

Ler

Ler

Het

Het

Plant 4

Het

Het

Het

Ler

Het

Ler

1. For the rough mapping via the pooled method, if the mutant is in the Ler background, one should look for the set of primers that show a predominant Ler pattern (see Note 45). 2. Once the region that is linked to the mutation is identified, fine mapping can be started. 3. Isolate the DNA individually for each mutant (as described above). 4. Go to www.arabidopsis.org. Click “sequence viewer” and identify BACs flanking the region in which the mutation is linked. 5. Find or design SSLP, CAPS, or dCAPS markers in the BAC. One useful resource is http://amp.genomics.org.cn/, which lists such markers. Alternatively, use the SNP information from http://signal.salk.edu/atg1001/3.0/gebrowser.php. Design primers that flank the SNPs (see Note 46). 6. Run a PCR reaction, and resolve the product on an agarose gel. 7. For the analysis, find the region flanked by recombinant markers. An example is shown in Table 3. Assume ciw8 was the marker to which the mutant was linked. Markers 1–5 are in the vicinity of the linked region. Results from plant 1 show that the mutation is below Marker 1. Results from plant 2 tell us that the mutation is between Marker 2 and Marker 5. Results from plant 3 tell us that the mutation is between Marker 1 and 4. Results from plant 4 tell us that the mutation is between Marker 3 and 4. In summary, the mutation is between Marker 3 and Marker 4. The mutation is localized to the region flanked by two markers, one on each side, with the least number of recombinants in the population. 8. Once a region of 200 kb or less is identified, go to www. arabidopsis.org, find the region in Seqviewer, and scan the region to see if there are any genes of interest. If so, sequence the gene from the mutant to find any mutations. To narrow down the region further, more plants will be needed for fine mapping. 9. If a mutation is identified, complementation analysis should be performed.

Genetic Screens: Enhancers and Suppressors

151

Table 4 mhiTAIL-PCR set-up for the pre-amplification reaction Reagent

Final concentration

H2O

4.8

2 PCR buffer

10

dNTP

200 μM (2.5 mM stock) 1.6

mLAD 1–4 primer

0.5 μM (10 μM stock)

1

AC0 primer

0.3 μM (10 μM stock)

0.6

RB-0a primer

0.3 μM (10 μM stock)

0.6

Phanta DNA polymerase 0.4 units

0.4

DNA template

1

40 ng

Total 3.2.3

mhiTAIL-PCR

Amount to add (μL)

20

The mhiTAIL-PCR is comprised of three separate PCR reactions as outlined below. 1. The first reaction is the pre-amplification reaction. See Table 4 for the set-up (see Note 47). 2. The PCR conditions for the first reaction are shown in Table 5. 3. The second reaction is the Primary mhiTAIL-PCR reaction. See Table 6 for the set-up (see Note 48). 4. The PCR condition for the second reaction is shown in Table 7. 5. The third reaction is the Secondary TAIL-PCR reaction. See Table 8 for the set-up (see Note 49). 6. The PCR condition for the third reaction is shown in Table 9. 7. Run the Primary and Secondary mhiTAIL-PCR reactions on a 1.2% agarose gel. 8. Choose the PCR product with clean bands and a proper ladder effect (see Note 50). 9. Weigh an empty tube, and record the weight (see Note 51). 10. Take an image of the gel (see Note 52). 11. Put the gel on top of plastic wrap, and place it on top of a UV light source (see Note 53). 12. Cut out the band (see Note 54). 13. Place the band into the pre-weighed 1.5 mL centrifuge tube. 14. Weigh the tube, and subtract the weight of the tube without the gel (see Note 55).

152

Zhigang Huang et al.

Table 5 mhiTAIL-PCR conditions for the pre-amplification reaction Step

Temperature ( C)

Time (min:sec)

1

93

2:00

2

95

1:00

3

95

0:30

4

62

0:40

5

72

2:00

6

Go to step 3

10 times

7

95

0:30

8

25

2:00

9

Ramping to 72

0.5  C/s

10

72

2:00

11

95

0:20

12

58

0:40

13

72

2:00

14

Go to step 11

25 times

15

72

2:30

16

End

Table 6 mhiTAIL-PCR set-up for the primary reaction Reagent

Final concentration

Amount to add (μL)

H2O

6.2

2 PCR buffer

10

dNTP

200 μM (2.5 mM stock)

1.6

AC1 primer

0.15 μM (10 μM stock)

0.3

RB-1a primer

0.25 μM (10 μM stock)

0.5

Phanta DNA polymerase

0.4 units

0.4

DNA template

1/50 dilution of the pre-amplification product

1

Total

20

Genetic Screens: Enhancers and Suppressors

153

Table 7 mhiTAIL-PCR conditions for the primary reaction Step

Temperature ( C)

Time (min:sec)

1

95

0:20

2

60

0:40

3

72

2:30

4

Go to step 1

4 time

5

95

0:20

6

68

0:40

7

72

2:00

8

95

0:20

9

68

0:40

10

72

2:00

11

95

0:20

12

50

0:40

13

72

2:00

14

Go to step 5

12 times

15

72

2:30

16

End

Table 8 mhiTAIL-PCR set-up for the secondary reaction Reagent

Final concentration

Amount to add (μL)

H2O

9.5

2 PCR buffer

15

dNTP

200 μM (2.5 mM stock)

2.4

AC1 primer

0.2 μM (10 μM stock)

0.6

RB-2a primer

0.3 μM (10 μM stock)

0.9

Phanta DNA polymerase

0.6 units

0.6

DNA template

1/20 dilution of the primary reaction PCR product

1

Total

30

154

Zhigang Huang et al.

Table 9 mhiTAIL-PCR conditions for the secondary reaction Step

Temperature ( C)

Time (min:sec)

1

95

0:20

2

65

0:40

3

72

2:00

4

95

0:20

5

65

0:40

6

72

2:00

7

95

0:20

8

50

0:40

9

72

2:00

10

Go to step 1

7–8 times

11

72

2:30

12

End

15. Multiply the weight of the agarose by three. Add that volume (in μL) of ADB buffer (see Note 56). 16. Leave the tube at 37–55  C for 5–10 min (see Note 57). 17. Transfer the solution containing the melted gel to a ZymoSpin™ Column in the collection tube. 18. Leave the solution in the column for 3 min. 19. Spin the column in a tabletop microfuge for 30 s at RT at 16,100  g. 20. Discard the flow-through. 21. Add 200 μL of wash buffer to the column (see Note 58). 22. Let the column sit at RT for 3 min. 23. Centrifuge for 30 s at 16,100  g in a tabletop microfuge. Discard the flow-through. 24. Repeat the wash step. 25. Centrifuge the empty column for 1 min at 16,100  g in a tabletop microfuge. 26. Transfer the column to a new, labeled centrifuge tube. 27. Add 25 μL water to the column (see Note 59). 28. Let the column sit at RT for 3 min. 29. Centrifuge the column for 2 min at 16,100  g in a tabletop microfuge.

Genetic Screens: Enhancers and Suppressors

155

30. Quantify the DNA, and run ~2–5 μL of DNA on the agarose gel to make sure there is a clean, single band. 31. Send the required amount to the sequencing facility per their specifications. 3.2.4 Mapping by Deep Sequencing: Library Preparation

1. Obtain 5 μg of RNase-treated genomic DNA via the CTAB method described above with the exception of resuspending in 50 μL of autoclaved water at the final step. 2. Use the Bioruptor to sonicate the DNA. Perform two replicates of 15-min sonication with 30 s on at maximum speed and 30 s off (see Note 60). 3. Recycle the DNA between 200–300 bp through the E-gel system (see Note 61). 4. Build a DNA sequencing library using the TruSeq DNA Sample Preparation kit. 5. Sequence the library using the Illumina Genome Analyzer (see Note 62).

4

Notes 1. Since this protocol is written for modifier screens, use a mutant allele with a weak phenotype to identify enhancers and a mutant allele with a stronger phenotype to identify suppressors. A seed volume of 100 μL corresponds to ~1000 seeds. 2. EMS is a highly hazardous and volatile compound. Wear double gloves, a lab coat, goggles, and closed-toe shoes when working with this compound. In addition, make sure to prepare all the solutions inside a fume hood. Decontaminate everything that has come in contact with EMS or is touched while working with EMS by washing it with 1 M NaOH and discard waste into the appropriate biohazard container. 3. Suspending the seeds in 0.1% agar will allow for better dispersal of the seeds. 4. All reagents must be added in the order listed. EDTA will not dissolve quickly, so adding the NaOH slowly will allow the EDTA to dissolve when the pH reaches 8.0. Before using the solution, make sure it has cooled down after autoclaving. Make sure to add the β-mercaptoethanol in a hood. 5. SDS is potentially harmful to the respiratory system. Thus, wear a mask and weigh SDS in the hood. 6. It is optional to clip off the first bolts of flowering Arabidopsis to promote more secondary bolts. Four to six days after clipping, plants can be transformed. Since immature flowers are the organs transformed, plants can be dipped multiple times (2–3 times).

156

Zhigang Huang et al.

7. Depending on the size difference between the two bands, 3 or 4% gels can be used. The gels can be used up to four times by melting and re-pouring. 8. Any Taq DNA polymerase should be suitable. 9. Although the DNA will be fragmented afterwards, DNA integrity is important to prepare high quality DNA sequencing libraries. 10. The purpose of using Tween-20 is to allow the EMS to penetrate the seeds more easily. Therefore, this protocol requires less EMS than other protocols. 11. Wrapping with Parafilm helps prevent contamination. 12. Rinsing a copious amount of times helps to get rid of the EMS. Treat the water from the rinse by adding NaOH (to get a 0.5 M final concentration) using the 5 M stock solution. 13. The agar helps to suspend the seeds in the solution more evenly, making planting much easier. Remember to allow the 0.1% agar to cool prior to adding the seeds. 14. The clear, dome-shaped lid maintains moisture to promote germination. 15. Dominant mutations are rare, so most plants will not exhibit an abnormal phenotype. Seeds are collected from individual plants so that mutations in the M2 can be easily maintained in the heterozygous situation if the homozygous plant is sterile. 16. Keeping plants under continuous light allows them to grow faster. Flowering can be accelerated by growing plants under continuous light, which will expedite the genetic screen aimed at the isolation of floral mutants. In addition, the plants should be grown in their own section so as to avoid contamination by other plants in the growth chamber. 17. In the M1 plant, usually only one of two cells that give rise to the germ line gets mutated. The segregation ratio for a recessive mutation in the M2 should be close to 1 out of 8. 18. This is very important as molecular analyses of the mutant should only be performed on a clean line. 19. For example, if our mutant were in the Ler background, we would cross it to Col-0 to generate the mapping population. Oftentimes, mutants have a stronger or weaker phenotype in the Col-0 background. Thus, there could be an enhancer or suppressor in Col-0, so one could possibly map the modifier. Also, if the mutagenesis is performed in a given mutant background (such as ag-10), it is a good idea to introgress the mutation into a different ecotype several times as the mapping parental line (in our case, we crossed ag-10 to Col-0 six times

Genetic Screens: Enhancers and Suppressors

157

and used the ag-10Col-0 as the other parent to generate the mapping population). This is important, as we want to keep two copies of ag-10 in the subsequent population so that we are able to see the “enhancer” phenotype in a reasonable proportion in the F2 population. 20. This DNA is used as a control for future mapping experiments (called Het hereafter). Collect at least four samples for each method because a lot is used for positional cloning. 21. Positional cloning can also be used to map dominant mutations. In the F2 population, the plants with no mutant phenotypes should be wild type at both alleles, and thus are used for mapping. 22. For simplicity purposes, the controls will be referred to as Ler and Col-0. For pooled tissue, make sure the leaves are about the same size for each plant. The tissue can be stored at 80  C (for the pooled tissue) or 4  C (for the Quick and Dirty PCR) for several days prior to using. 23. The mixture should be dark green and viscous but not too thick. If it is too thick, the solution is hard to move when inverted. In this case, split the samples into different tubes and add more extraction buffer. If CTAB extraction is being performed on one plant (several leaves), no liquid nitrogen is needed. Use a blue pestle (for microcentrifuge tubes) and grind the tissue in a 1.5 mL centrifuge tube. All spins can be done in a tabletop microfuge at 16,100  g. 24. The solution should be cloudy. 25. Be careful not to jostle the tube to disrupt the separation. If the tube is jostled, re-spin. This phase should be clear. 26. Be careful not to transfer any chloroform at this step. 27. Theoretically, the DNA can be saved at 20  C indefinitely as it is stable; however, it is best to use the DNA as soon as possible. 28. For the first wash, add 4 mL of 70% EtOH very slowly and invert carefully so as to avoid dislodging the pellet. If the pellet is dislodged, spin for 5 min and wash again. If you do not dislodge the pellet, decant 70% EtOH, add more 70% EtOH, and invert the tube vigorously. Spin down the tube at 11,648  g, 4  C for 5 min. 29. Since the pellet may disappear after drying, make a mark on the outside of the tube to show the pellet’s location. That way, when the pellet is resuspended, autoclaved water can be added up to that point. 30. Use a tip to help resuspend the pellet. Warming up the water (37  C) also helps with the resuspension. If there was a small amount of tissue used, resuspend in 30 μL of autoclaved water.

158

Zhigang Huang et al.

31. The solution can also be left at 20  C for a few hours to overnight. If left overnight, add a 1/10 volume of 3 M sodium acetate and 1 μL of glycogen, as a carrier, to the solution. 32. If the DNA is used the same day, leave the tubes at room temperature for 1 h. 33. Keep elbow above the wrist when grinding. This will allow better grinding without using that much energy and/or causing wrist injury. When performing extraction for hundreds of plants, this helps. 34. If hundreds of preps are being performed, it might be easier to use the plate method. In this case, add 140 μL of the supernatant to each well of a 96-well PCR plate, which contain 140 μL of isopropanol. Because this leads to the well-being almost completely full, be careful to avoid cross-contaminations between individual wells. As a precaution and to allow you to repeat the experiment in case of a contamination, keep the tubes with the leftover supernatant. 35. Do not leave the tubes in the freezer or at 4  C as this will result in DNA preparations with a high level of impurities. 36. There is an optional step of keeping the mixture of competent cells and plasmid on ice 30 min prior to electroporation. 37. Colonies should form within a couple of days. If they do not form, the transformation was not successful and should be repeated. These plates can be stored at 4  C for a few months at a time. 38. To make sure a positive colony is obtained, multiple colonies can be inoculated into test tubes (5 mL of medium with antibiotics). Multiple colonies can also be verified using colony PCR to save time. 39. To store bacteria, inoculate another 5 mL of medium and antibiotics in a test tube with a positive colony. Grow overnight in a 28  C shaker. Pipette 150 μL sterile glycerol into a 1.5 mL plastic tube. Pipette 850 μL bacterial culture into the same tube and mix well. It will take some effort to mix the two completely. Label the tube and flash freeze it in liquid nitrogen. Store in 80  C until needed for future transformations. 40. This OD600 requirement is not strict. Generally, 0.6–0.8 is used. However, as long as one can suspend the Agrobacterium to about an OD600 ¼ 0.8 in a solution of 5% sucrose, then efficiency of transformation will not be adversely affected. The time will be about 18–24 h of incubation. Alternatively, the Agrobacterium can be grown to 1.0–1.2 and diluted to 0.8. 41. Each tray is sized 2100  1100  200 with 12 pots, in which multiple plants of the same genotype are grown.

Genetic Screens: Enhancers and Suppressors

159

42. Standard is 0.05% (v/v) Silwet L-77, but as low as 0.005% (v/v) can be used successfully. Less should only be used when toxicity is a risk to plants. 43. If the solution touches the soil, the high sugar concentration will change soil pH, leading to unhealthy plants. 44. Make sure to run both ecotypes and 1:1 mix of the two accessions as controls for all reactions. 45. If the mutant is in the Col-0 background, identify the marker that is linked to Col-0. 46. Go to the markers tab and insert BAC number. A set of primers will be displayed. The most commonly used polymorphism information between Col-0 and Ler is http://www.ara bidopsis.org/browse/Cereon/index.jsp; however, the one mentioned in the actual protocol is more comprehensive. 47. There will be a total of four different PCR tubes for this reaction (one for each one of the mLAD primers). 48. Once again, there will be a total of four different PCR tubes for this reaction (one for each one of the mLAD primers). 49. Once again, there will be a total of four different PCR tubes for this reaction (one for each one of the mLAD primers). 50. The band from the Secondary PCR reaction should run faster than the band from the Primary PCR reaction. As the product will be used for sequencing, make sure everything is clean. For instance, clean the gel apparatus, use new running buffer, etc. 51. Use a balance that measures the weight to the precision of at least 1 mg. 52. Do this as fast as possible as long exposure to the UV light will induce thymine dimers that affect subsequent PCR reactions. 53. Make sure to wear a UV mask to protect eyes and skin from burns. 54. Make sure to use a new razor blade and get rid of as much agarose as possible. 55. This is to find the weight of the gel containing the band. 56. For example, if the weight of the agarose band is 100 mg, add 300 μL of ADB buffer. 57. Every few minutes, briefly vortex the tube to help the agarose dissolve. Running the PCR product on a higher percentage gel will cause the gel to be harder to dissolve. It may take longer than 10 min for the gel to fully dissolve. 58. Make sure to add the required amount of ethanol to the wash buffer prior to using. 59. Be careful not to let the tip touch the column.

160

Zhigang Huang et al.

60. The genomic DNA can be fragmented by any kind of sonication method as long as the DNA is sheared to 200–300 bp. Run a 1% agarose gel to check. A smear is expected, but the majority of the band intensity should be between 200 and 300 bp. 61. This step can be skipped if the gel method is used to purify the ligation product before enrichment of the DNA fragment when preparing the DNA sequencing libraries. Also, this can be done through regular gel purification methods (as mentioned above with the Zymo kit). If purifying the DNA at this step, there is no need to purify the ligated DNA during the next part of the procedure. 62. Multiple samples can be barcoded in a single lane to lower costs of sequencing. However, the coverage for the genome should be higher than 10. References 1. Laibach F (1943) Arabidopsis thaliana (L.) Heynh als objekt fur genetische und entwick lungsphysiologische Untersuchungen. Bot Archiv 44:439–455 2. Leutwiler LS, Hough-Evans BR, Meyerowitz EM (1984) The DNA of Arabidopsis thaliana. Mol Gen Genet 194:15–23 3. Meyerowitz EM (1989) Arabidopsis, a useful weed. Cell 56(2):263–269 4. Meyerowitz EM, Pruitt RE (1985) Arabidopsis thaliana and plant molecular genetics. Science 229(4719):1214–1218 5. Pruitt RE, Meyerowitz EM (1986) Characterization of the genome of Arabidopsis thaliana. J Mol Biol 187(2):169–183 6. Bowman JL, Meyerowitz EM (1991) Genetic control of pattern formation during flower development in Arabidopsis. Symp Soc Exp Biol 45:89–115 7. Bowman JL, Smyth DR, Meyerowitz EM (1989) Genes directing flower development in Arabidopsis. Plant Cell 1(1):37–52 8. Coen ES, Meyerowitz EM (1991) The war of the whorls: genetic interactions controlling flower development. Nature 353(6339):31–37 9. Meyerowitz EM et al (1991) A genetic and molecular model for flower development in Arabidopsis thaliana. Dev Suppl 1:157–167 10. Drews GN, Weigel D, Meyerowitz EM (1991) Floral patterning. Curr Opin Genet Dev 1(2): 174–178 11. Jack T, Brockman LL, Meyerowitz EM (1992) The homeotic gene APETALA3 of Arabidopsis

thaliana encodes a MADS box and is expressed in petals and stamens. Cell 68(4):683–697 12. Smyth DR, Bowman JL, Meyerowitz EM (1990) Early flower development in Arabidopsis. Plant Cell 2(8):755–767 13. Weigel D et al (1992) LEAFY controls floral meristem identity in Arabidopsis. Cell 69(5): 843–859 14. Yanofsky MF et al (1990) The protein encoded by the Arabidopsis homeotic gene agamous resembles transcription factors. Nature 346(6279):35–39 15. Bowman JL, Smyth DR (1999) CRABS CLAW, a gene that regulates carpel and nectary development in Arabidopsis, encodes a novel protein with zinc finger and helix-loop-helix domains. Development 126(11):2387–2396 16. Sieburth LE, Running MP, Meyerowitz EM (1995) Genetic separation of third and fourth whorl functions of AGAMOUS. Plant Cell 7(8):1249–1258 17. Chen X, Meyerowitz EM (1999) HUA1 and HUA2 are two members of the floral homeotic AGAMOUS pathway. Mol Cell 3(3):349–360 18. Cheng Y et al (2003) Two RNA binding proteins, HEN4 and HUA1, act in the processing of AGAMOUS pre-mRNA in Arabidopsis thaliana. Dev Cell 4(1):53–66 19. Li J, Jia D, Chen X (2001) HUA1, a regulator of stamen and carpel identities in Arabidopsis, codes for a nuclear RNA binding protein. Plant Cell 13(10):2269–2281

Genetic Screens: Enhancers and Suppressors 20. Ji L et al (2011) ARGONAUTE10 and ARGONAUTE1 regulate the termination of floral stem cells through two microRNAs in Arabidopsis. PLoS Genet 7(3):e1001358 21. Chen X (2004) A microRNA as a translational repressor of APETALA2 in Arabidopsis flower development. Science 303(5666):2022–2025 22. Liu X et al (2011) AGAMOUS terminates floral stem cell maintenance in Arabidopsis by directly repressing WUSCHEL through recruitment of Polycomb Group proteins. Plant Cell 23(10):3654–3670 23. Yumul RE et al (2013) POWERDRESS and diversified expression of the MIR172 gene family bolster the floral stem cell network. PLoS Genet 9(1):e1003218 24. Liu X et al (2014) AUXIN RESPONSE FACTOR 3 integrates the functions of AGAMOUS and APETALA2 in floral meristem determinacy. Plant J 80(4):629–641 25. Liu X et al (2014) DNA topoisomerase I affects polycomb group protein-mediated epigenetic regulation and plant development by altering nucleosome distribution in Arabidopsis. Plant Cell 26(7):2803–2817 26. Li D et al (2016) FAR-RED ELONGATED HYPOCOTYL3 activates SEPALLATA2 but inhibits CLAVATA3 to regulate meristem determinacy and maintenance in Arabidopsis. Proc Natl Acad Sci U S A 113(33):9375–9380 27. Huang Z et al (2017) APETALA2 antagonizes the transcriptional activity of AGAMOUS in regulating floral stem cells in Arabidopsis thaliana. New Phytol 215(3):1197–1209 28. Siegfried KR et al (1999) Members of the YABBY gene family specify abaxial cell fate in Arabidopsis. Development 126(18): 4117–4128 29. Sawa S et al (1999) FILAMENTOUS FLOWER, a meristem and organ identity gene of Arabidopsis, encodes a protein with a zinc finger and HMG-related domains. Genes Dev 13(9):1079–1088 30. Eshed Y, Baum SF, Bowman JL (1999) Distinct mechanisms promote polarity establishment in carpels of Arabidopsis. Cell 99(2): 199–209 31. Telfer A, Poethig RS (1998) HASTY: a gene that regulates the timing of shoot maturation in Arabidopsis thaliana. Development 125(10): 1889–1898 32. Wagner D, Meyerowitz EM (2002) SPLAYED, a novel SWI/SNF ATPase homolog, controls reproductive development in Arabidopsis. Curr Biol 12(2):85–94

161

33. Eshed Y et al (2001) Establishment of polarity in lateral organs of plants. Curr Biol 11(16): 1251–1260 34. Muller R, Bleckmann A, Simon R (2008) The receptor kinase CORYNE of Arabidopsis transmits the stem cell-limiting signal CLAVATA3 independently of CLAVATA1. Plant Cell 20(4):934–946 35. Brand U et al (2000) Dependence of stem cell fate in Arabidopsis on a feedback loop regulated by CLV3 activity. Science 289(5479): 617–619 36. Nimchuk ZL, Tarr PT, Meyerowitz EM (2011) An evolutionarily conserved pseudokinase mediates stem cell production in plants. Plant Cell 23(3):851–854 37. González-Carranza ZH et al (2017) HAWAIIAN SKIRT controls size and floral organ number by modulating CUC1 and CUC2 expression. PLoS One 12(9):e0185106 38. Nag A, Yang Y, Jack T (2007) DORNROSCHEN-LIKE, an AP2 gene, is necessary for stamen emergence in Arabidopsis. Plant Mol Biol 65(3):219–232 39. Levin JZ et al (1998) A genetic screen for modifiers of UFO meristem activity identifies three novel FUSED FLORAL ORGANS genes required for early flower development in Arabidopsis. Genetics 149(2):579–595 40. Page DR, Grossniklaus U (2002) The art and design of genetic screens: Arabidopsis thaliana. Nat Rev Genet 3(2):124–136 41. Papdi C et al (2010) Genetic screens to identify plant stress genes. Methods Mol Biol 639:121– 139 42. Kim Y, Schumaker KS, Zhu JK (2006) EMS mutagenesis of Arabidopsis. Methods Mol Biol 323:101–103 43. Krieg DR (1963) Ethyl methanesulfonate induced reversion of bacteriophage T4rII mutants. Genetics 48:561–580 44. Greene EA et al (2003) Spectrum of chemically induced mutations from a large-scale reversegenetic screen in Arabidopsis. Genetics 164(2): 731–740 45. McCallum CM et al (2000) Targeted screening for induced mutations. Nat Biotechnol 18(4): 455–457 46. Jander G et al (2003) Ethylmethanesulfonate saturation mutagenesis in Arabidopsis to deter mine frequency of herbicide resistance. Plant Physiol 131(1):139–146 47. Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J 16(6):735–743

162

Zhigang Huang et al.

48. Tzfira T et al (2004) Agrobacterium T-DNA integration: molecules and models. Trends Genet 20(8):375–383 49. Szabados L et al (2002) Distribution of 1000 sequenced T-DNA tags in the Arabidopsis genome. Plant J 32(2):233–242 50. Wang Y (2008) How effective is T-DNA insertional mutagenesis in Arabidopsis? J Biochem Technol 1(1):11–20 51. Li Y et al (2006) Analysis of T-DNA insertion site distribution patterns in Arabidopsis thaliana reveals special features of genes without insertions. Genomics 87(5):645–652 52. Weigel D et al (2000) Activation tagging in Arabidopsis. Plant Physiol 122(4):1003–1013 53. Radhamony RN, Prasad AM, Srinivasan R (2005) T-DNA insertional mutagenesis in Arabidopsis: a tool for functional genomics. Electron J Biotechnol 8(1). https://doi.org/10. 2225/vol8-issue1-fulltext-4 54. Lukowitz W, Gillmor CS, Scheible WR (2000) Positional cloning in Arabidopsis. Why it feels good to have a genome initiative working for you. Plant Physiol 123(3):795–805 55. Liu YG, Whittier RF (1995) Thermal asymmetric interlaced PCR: automatable

amplification and sequencing of insert end fragments from P1 and YAC clones for chromosome walking. Genomics 25(3):674–681 56. Tan J et al (2019) A modified high-efficiency thermal asymmetric interlaced PCR method for amplifying long unknown flanking sequences. J Genet Genomics 46(7):363–366 57. Austin RS et al (2011) Next-generation mapping of Arabidopsis genes. Plant J 67(4): 715–725 58. Schneeberger K et al (2009) SHOREmap: simultaneous mapping and mutation identification by deep sequencing. Nat Methods 6(8): 550–551 59. Sugihara Y et al (2022) High-performance pipeline for MutMap and QTL-seq. PeerJ 10: e13170 60. Zhu Y et al (2012) Gene discovery using mutagen-induced polymorphisms and deep sequencing: application to plant disease resistance. Genetics 192(1):139–146 61. Sarin S et al (2008) Caenorhabditis elegans mutant allele identification by whole-genome sequencing. Nat Methods 5(10):865–867

Chapter 7 Genetic and Phenotypic Analysis of Shoot Apical and Floral Meristem Development Mona M. Monfared, Thai Q. Dao, and Jennifer C. Fletcher Abstract The shoot apical and floral meristems (SAM and FM, respectively) of Arabidopsis thaliana contain reservoirs of self-renewing stem cells that function as sources of progenitor cells for organ formation during development. The primary SAM produces all the aerial structures of the adult plant, while the FMs generate the four types of floral organs. Consequently, aberrant SAM and FM activity can profoundly affect vegetative and reproductive plant morphology. The embedded location and small size of Arabidopsis meristems make accessing these structures difficult, so specialized techniques have been developed to facilitate their analysis. Microscopic, histological, and molecular techniques provide both qualitative and quantitative data on meristem organization and function, which are crucial for the normal growth and development of the entire plant. Key words Shoot apical meristem, Floral meristem, Inflorescence meristem, Stem cells, Confocal laser scanning microscopy, Histology, in situ hybridization, Scanning electron microscopy, Live imaging

1

Introduction One of the main distinctions between plant and animal development is that, unlike animals, the embryonic form of plants does not closely resemble the adult body plan. A plant embryo is a simple structure containing a pool of pluripotent stem cells at its shoot apex that post-embryonically generates all aerial organs, such as leaves and flowers [1]. The shoot apical meristem (SAM) maintains this central stem cell reservoir throughout the plant life cycle while simultaneously providing progeny cells for continuous organ formation on its flanks [2]. During early development leaves are produced from the vegetative SAM, which later transitions into an inflorescence meristem (IM) when endogenous and environmental cues trigger flowering. Floral meristems (FM) form on the

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_7, © Springer Science+Business Media, LLC, part of Springer Nature 2023

163

164

Mona M. Monfared et al.

flanks of the IM and contain transient stem cells as well as their descendant cells that produce the four types of flower organs: sepals, petals, stamens and carpels [3]. Analyzing the morphology and histology of the SAM and FM and characterizing their gene expression patterns can shed light on many developmental processes [4]. Because the majority of development occurs post-embryonically, overall plant growth and architecture depends on the maintenance of stable apical meristems. Thus defects in growth rate, organ initiation, organ number, and/or stem size can often be traced back to perturbations in meristem structure or function that may be undetectable by the naked eye [5, 6]. As the stable source of cells for organogenesis, plant meristems are an excellent model system for studying stem cell maintenance and termination, cell fate specification, organ morphogenesis, and pattern formation. In this chapter we explain methods and approaches for the analysis of SAM and FM development: (i, v, and vi) Confocal laser scanning microscopy (CLSM) permits the imaging of Arabidopsis SAM and FM meristem tissue from whole mount samples [7–9] (see Subheadings 3.1, 3.5, and 3.6, respectively). CLSM is a powerful method for examining the topology, cell layering and cell number of embryonic SAMs, IMs, and FMs without the need for physically sectioning the specimens, as the optical sections can be combined into a three-dimensional digital image of the sample. In addition, fluorescent reporter activity patterns in IMs and FMs can be visualized at high resolution using live imaging [7, 10]. (ii) Histological sectioning of resin-embedded samples allows the analysis of vegetative SAM tissues, using cell-specific stains that permit high resolution imaging of the internal cellular morphology (see Subheading 3.2). (iii) Meristem size is a key indicator of stem cell activity in Arabidopsis and other plant species [11–15]. The use of publicly available image analysis software provides a fast, convenient, and highly accurate method of measuring the height and diameter of meristem sections from saved image files (see Subheading 3.3). (iv) In situ hybridization uses a labeled nucleotide probe to localize a specific RNA transcription pattern within a whole mount or tissue section [16]. First introduced in the late 1960s for RNA detection in Xenopus oocytes [17], this technique has been successfully adapted to study gene transcription patterns in plants such as Arabidopsis, maize, rice and tomato at all developmental stages [18–20]. However, vegetative SAMs are surrounded by leaf primordia, making it very difficult to achieve full tissue penetration of the probe and obtain a satisfactory signal-to-noise ratio. A modified RNA in situ hybridization protocol has been developed to effectively detect and localize mRNA transcripts to specific regions within vegetative shoot apices [21–23] (see Subheading 3.4). (vii) Scanning electron microscopy (SEM) allows high resolution imaging by measuring the angle and energies of electrons scattered by

Analysis of Meristem Development

165

atoms on the surface of a sample. The extreme level of surface detail that can be acquired using SEM is applicable to plants for analyzing the number and arrangement of floral meristems, the position and structure of floral organ primordia, and the cell surface morphology of individual flower organs [24] (see Subheading 3.7). (viii) Floral organ number counting is a simple method to quantify and statistically analyze the phenotypes of mutants from any plant species that display altered floral organ number and/or produce mosaic floral organs. Such phenotypes are often caused by an underlying defect in the regulation of floral meristem size [11, 25] (see Subheading 3.8).

2

Materials Arabidopsis thaliana seedlings are grown at 21 °C on plates containing Murashige and Skoog (MS) medium [26] under cool-white fluorescent lights (100–140 μmol/m2 s) for 4–10 days. MS medium: 4.33 g MS salts, 3 g sucrose, 900 mL distilled water. Adjust pH to 6.0 with 1 M KOH. Add 9 g bactoagar and adjust volume to 1 L with distilled water. For analysis of Arabidopsis plants during the reproductive phase, seeds are sown in a 1:1:1 mixture of perlite:vermiculite:topsoil and grown under cool-white fluorescent lights (100–140 μmol/m2 sec) at 21–22 °C. A minimum of 10 samples from each genotype should be analyzed per experiment to obtain statistical robustness. For RNA in situ hybridization experiments, at least 50 samples per genotype or experimental batch should be prepared to achieve a good representation of the expression pattern of each gene of interest.

2.1 Confocal Laser Scanning Microscopy of the Embryonic Meristem

1. Whatman or 3MM paper. 2. Small Petri dishes (15 × 60 mm). 3. Glass scintillation vials or similar containers. 4. Pasteur pipettes or P100 micropipette. 5. Microscope depression slides. 6. Cover glass (18 × 18 mm). 7. Fine forceps (number 5, Ted Pella Inc.). 8. Stereo-microscope. 9. Nail polish. 10. Cold distilled water. 11. Pseudo-Schiff reagent: 100 mM Sodium Metabisulphite, 0.15 N HCl in distilled water. 12. Propidium Iodide (PI) red-fluorescent counterstain (see Note 1).

166

Mona M. Monfared et al.

13. Propidium Iodide staining solution: Prepare fresh by adding Propidium Iodide to pseudo-Schiff reagent at a final concentration of 10 μg/mL. Prepare a solution of ~1 mL for every 25 embryos in the vial. 14. 100% ethanol. 15. 50% ethanol in distilled water. 16. Methyl salicylate (see Note 2). 17. Dehydration solution: 50% ethanol, 50% methyl salicylate. 2.2 Histological Sectioning of the Vegetative Meristem

1. Single-edge razor blades. 2. Scintillation vials. 3. Vacuum bell or vacuum oven. 4. Microtome. 5. 42–55 °C slide-warmer. 6. Histoform S (Teflon embedding mould; Heraeus Kulzer, Wehrheim/Ts., Germany). 7. Histobloc (Heraeus Kulzer, Wehrheim/Ts., Germany). 8. Tungsten Carbide Microtome Knife. 9. FAA fixation solution: 50% ethanol, 5% acetic acid, 3.7% formaldehyde. Prepare a fresh mixture of ~10 mL per vial (see Note 3). 10. Graded ethanol series: Prepare 500 mL solutions of 15%, 30%, 50%, 70%, 85%, 90%, 95% ethanol in distilled water. 11. Technovit Glycol Methacrylate (GMA; also known as 2-hydroxyethyl methacrylate) Kit 7100: Each kit contains 500 mL of GMA monomer base liquid, five 1 g packs of Hardener I, and 40 mL of Hardener II (see Note 4). 12. Graded Technovit series: Prepare 50% (v/v), 70% (v/v), 90% (v/v) Technovit 7100 resin in 100% ethanol. 13. Technovit 7100 Hardener I solution: Prepare 1 g Technovit 7100 Hardener I in 100 mL GMA monomer base liquid. Leftover solution can be stored at 4 °C for up to 4 weeks. 14. Toluidine blue solution: Prepare 0.1% (w/v) Toluidine blue O in 0.1% aqueous sodium tetraborate (see Note 5). 15. Neutral Red solution: Prepare 0.01% (w/v) solution by melting neutral red powder in Technovit 7100 Hardener I solution. 16. ImmunoHistoMount (Sigma).

2.3 Meristem Size Measurement

1. Computer equipped with NCBI Image J or Fiji software for image processing and analysis (see Note 6). 2. Digital images of shoot apical meristem sections (see Note 7).

Analysis of Meristem Development

2.4 Vegetative Meristem RNA In Situ Hybridization

167

1. Single-edge razor blades. 2. Scintillation vials. 3. Vacuum bell or vacuum oven. 4. 56 °C oven for flasks. 5. Bench-top microscope slide warming table. 6. Weighing boats (8 cm × 8 cm). 7. Microtome. 8. Paraffin block holder. 9. Positively charged adhesion slides (ProbeOn Plus slides, FisherBrand). 10. Cover glass (24 × 50 mm). 11. 42 °C water bath for slides. 12. Fine paintbrushes. 13. 42–55 °C bench for slides (slide-warmer). 14. Racks for slides or Copeland jars: Glassware (racks or Copeland jars) that will hold the slides must be baked (see Note 8). 15. 37–60 °C incubator with a shaker for the slide racks/ Copeland jars. 16. UV crosslinker. 17. Plastic microscope slide boxes with latch, approximately 21 × 16.5 × 3 cm. 18. Stereo-microscope. 19. Eosin Y dye. 20. HistoClear. 21. Paraplast X-tra paraffin chips. 22. RNase-free Milli-Q water. 23. RNA molecular weight marker. 24. T7/T3/SP6 RNA Polymerases (see Note 9). 25. RNase out/ RNasin (see Note 9). 26. DNaseI (see Note 9). 27. Anti-Digoxigenin-AP Conjugate antibody (see Note 9) 28. Roche Blocking reagent (see Note 9): A 5× Blocking reagent (2.5%) stock can be prepared in advance and stored at -20 °C. 29. Nitroblue Tetrazolium (NBT) (see Note 9). 30. 5-bromo-4-chloro-3-indolyl phosphate (BCIP) (see Note 9). 31. 10× Phosphate Buffer Saline (PBS): 1.3 M NaCl, 70 mM Na2HPO4, 30 mM NaH2PO4 pH 7 (NaOH). Prepare 1 L and autoclave it.

168

Mona M. Monfared et al.

32. 4% Paraformaldehyde fixation solution (see Note 10): Take 100 mL of 1× PBS and adjust pH to 11 with 1 M NaOH. Heat to 60 °C, add 4 g of fresh paraformaldehyde and stir to dissolve. Cool on ice and adjust pH to 7 with 1 M H2SO4. Add 1 mL of Triton X-100 and 1 mL of Dimethyl Sulfoxide (DMSO) (see Note 11). Make this solution fresh each time and keep on ice. 33. 100% ethanol, 15% ethanol in distilled water, and graded series of 30%, 50%, 75%, 95%, and of 30%, 50%, 70%, 80%, 90% and 95% ethanol in distilled water. 34. 2× Probe hydrolysis solution: Make fresh each time by mixing 30 μL of 21 mg/mL Na2CO3 and 20 μL of 16.8 mg/mL NaHCO3 per probe. 35. 10% glacial acetic acid. 36. 0.2 M HCl. 37. Proteinase K buffer: 50 mL of 100 mM Tris-HCl, pH 8.0, 50 mM EDTA. 38. Proteinase K stock: 5 mg/mL proteinase K in 1 mL of proteinase K buffer. The stock can be aliquoted and stored at -20 °C for subsequent use. 39. 2 mg/mL glycine in 1 × PBS 40. 37% formaldehyde 41. 20× Sodium Salt Citrate (SSC) stock solution: 3 M NaCl, 0.3 M Na2C6H5O7 2H2O. Prepare 1 L and autoclave solution. 42. Hybridization buffer: For 4 mL, mix 2 ml of pure deionized formamide, 1.2 mL of 20× SSC, 600 μL of 20% SDS, 40 μL of 10 mg/mL yeast tRNA and 158 μL of RNAse-free Milli-Q water. Leftover buffer can be frozen at -20 °C for further use (see Note 12). The hybridization buffer precipitates at room temperature. Keep it at hybridization temperature (55 °C) before aliquoting it onto the slides. 43. Wash solution: 0.2 × SSC, 0.1% SDS. 44. RNase solution: 10 μg/mL RNase in 2 × SSC. 45. 1× Tris Buffer Saline (TBS) solution: 0.4 M NaCl, 0.1 M TrisHCl, pH 7.5. Prepare 2 L and autoclave. A 10× TBS stock can be prepared, although the powders will not dissolve easily. 46. Wash buffer: 1 × TBS containing 0.5% Bovine Serum Albumin (BSA) and 0.1% Triton-X100. 47. Wash buffer: 1 × TBS containing 0.5% BSA. 48. Detection buffer: For 250 mL, mix 25 mL of 1 M Tris-HCl, pH 9.6, 6.25 mL of 4 M NaCl, 12.5 mL of 1 M MgCl2, 207 mL of RNase-free Milli-Q water. Prepare the detection

Analysis of Meristem Development

169

buffer with the substrates by mixing 1.6 μL of 50 mg/mL BCIP and 2.2 μL of 100 mg/mL NBT per mL of detection buffer. 2.5 Confocal Laser Scanning Microscopy of the Inflorescence Meristem

1. Glass scintillation vials or similar containers. 2. Microscope depression slides. 3. Cover glass (18 × 18 mm). 4. Fine forceps (number 5, Ted Pella Inc.). 5. Pasteur pipettes or micropipettes. 6. Stereo-microscope. 7. Nail polish. 8. Methanol Acetic Acid (MAA) fixation solution: 50% methanol, 10% acetic acid, 40% distilled water (see Note 13). Prepare a fresh mixture of ~10 mL per vial. 9. 50% ethanol in distilled water. 10. 80% ethanol in distilled water. 11. 100% ethanol. 12. 1% periodic acid in distilled water. 13. Pseudo-Schiff reagent: 100 mM Sodium Metabisulphite, 0.15 N HCl in distilled water. 14. Propidium Iodide (PI) red-fluorescent counterstain (see Note 1). 15. Propidium Iodide staining solution: Prepare fresh by adding Propidium Iodide to pseudo-Schiff reagent at a final concentration of 10 μg/mL. Prepare a solution of ~3 mL per vial. 16. Methyl salicylate (see Note 2). 17. Dehydration solution: 50% ethanol, 50% methyl salicylate.

2.6 Live Imaging Confocal Laser Scanning Microscopy of the Inflorescence Meristem

1. Small Petri dishes (15 × 60 mm). 2. Fine forceps (number 5, Ted Pella Inc.). 3. 1.5 mL Eppendorf tubes. 4. Dissecting scissors. 5. Stereo-microscope. 6. Distilled water. 7. 1% agarose (w/v) in distilled water. 8. Propidium Iodide (PI) red-fluorescent counterstain (see Note 1). 9. 1 mg/mL (w/v) Propidium Iodide staining solution: Dissolve 0.1 g Propidium Iodide in 1 mL of water. Store at 4 °C and protect from light. Solution can be stored for several months and is no longer good when the color fades to a dull orange.

170

Mona M. Monfared et al.

2.7 Scanning Electron Microscopy of the Inflorescence Meristem

1. Glass scintillation vials. 2. Plastic conical tubes. 3. Pasteur pipets. 4. Fine forceps (number 5, Ted Pella Inc.). 5. Stereo-microscope. 6. Cylinder mount gripper (Ted Pella, Inc.). 7. Mounting bases (see Note 14). 8. Mounting stub. 9. Conductive stickers. 10. White index cards. 11. Small Petri dishes (15 × 60 mm). 12. Critical point dryer. 13. Sputter coater apparatus. 14. 0.1 M Sodium Phosphate Buffer (PB) buffer: combine 200 mL of 0.1 M Sodium Phosphate Monobasic NaH2PO4 (12 g/L) and 800 mL of 0.1 M Sodium Phosphate Dibasic Na2HPO4 (14.2 g/L). The pH should be between 7.2–7.4. 15. 25 mM Sodium Phosphate Buffer (PB) wash solution: dilute 100 mL of 0.1 M Sodium Phosphate Buffer (PB) to 25 mM in distilled water. 16. Glutaraldehyde fixation solution: freshly prepare 6 mL 0.1 M PB, 3 mL 25% Glutaraldehyde, 16 mL distilled water in a plastic conical tube in the hood (see Note 15). 17. Graded ethanol series: prepare 30%, 50%, 65%, 75%, 89% and 95% ethanol in distilled water. 18. High quality 100% ethanol.

2.8 Floral Organ Number Counting

1. Fine forceps (number 5, Ted Pella Inc.). 2. Dissecting scissors or a sharp blade. 3. Stereo-microscope with 10× objective. 4. Statistical analysis software.

3

Methods

3.1 Confocal Laser Scanning Microscopy of the Embryonic Meristem 3.1.1

Embryo Dissection

1. Place two discs of Whatman or 3 MM paper in a Petri dish and let them absorb as much water as they can hold. 2. Distribute mature and dry Arabidopsis seeds on the surface of the paper, close the Petri dish and let the seeds imbibe overnight at 4 °C in the dark (see Note 16).

Analysis of Meristem Development

171

3. The next day place a small piece of wet Whatman or 3MM paper on a microscope slide. Transfer a few seeds onto the slide and dissect the embryos out of the seed coat under a stereomicroscope using fine forceps and needles (see Note 17). Hold the seeds with fine forceps by the micropylar end and make an incision on the other end of the seed coat with a needle or another pair of fine forceps. Apply gentle pressure on the micropylar side of the seed using the forceps slanted to one side. The embryo should pop out of the scar in the seed coat. Add more water if the paper on the slide dries. 4. With a fine needle immediately transfer the isolated embryo into a glass scintillation vial containing a few mL of cold distilled water and keep the vial on ice (see Note 18). 3.1.2 Tissue Staining and Rinsing

1. Replace the water (see Note 19) with 1 mL of the 10 μg/mL Propidium Iodide staining solution for every 25 embryos in the vial. Stain overnight at room temperature (see Note 20). After the staining incubation the tissue appears pale orange. 2. Replace the staining solution with 2–3 mL of distilled water. 3. Rinse the tissues twice with distilled water.

3.1.3 Tissue Dehydration and Clearing

1. Incubate the tissues in 2–3 mL of 50% ethanol for 30 min to 1 h. 2. Incubate in 100% ethanol for 30 min to 1 h. 3. Incubate in 50% ethanol: 50% methyl salicylate for 30 min to 1 h. 4. Incubate in 100% methyl salicylate for 10 min.

3.1.4 Mounting and Imaging

1. Pipet a drop of 100% methyl salicylate in the center of a cover glass. 2. Pipet some embryos with a P100 micropipette (see Note 21) into the drop of 100% methyl salicylate. Each embryo should be lying on its side with the root and the two cotyledons touching the cover glass, and be fully covered with methyl salicylate to prevent the tissue drying out. At least 6–8 Arabidopsis embryos can be placed on the same cover glass. 3. Flip the cover glass over a depression slide, such that the embryo side is down, and seal the four corners of the cover glass with nail polish (see Note 22). 4. Visualize each slide using a Confocal Laser Scanning Microscope (see Fig. 1). Propidium Iodide can be excited by a 514 nm argon laser beam and emits between 580 and 610 nm.

172

Mona M. Monfared et al.

Fig. 1 Confocal laser scanning micrographs of Arabidopsis mature embryo and embryonic shoot apical meristem. (a) Optical longitudinal section of a wild-type Columbia-0 (Col-0) mature embryo. The shoot apical meristem is boxed. (b) Optical longitudinal section of an embryonic shoot apical meristem from a Col-0 mature embryo. The apical meristem cell nuclei stain brightly with Propidium Iodide. Scale bars: 20 μm 3.2 Histological Sectioning of the Vegetative Meristem 3.2.1 Tissue Dissection and Fixation

Plant materials for histological analysis should be grown on slanted agar plates to facilitate tissue harvesting. Depending on the growing conditions, 4 to 10-day-old Arabidopsis seedlings are appropriate for vegetative SAM observation. 1. Remove the roots of each seedling at the base of the hypocotyl using clean forceps or small scissors. 2. Immediately place each dissected sample into a glass scintillation vial containing FAA fixation solution in a fume hood (see Note 3). 3. Loosen the caps of the scintillation vials and place them in a vacuum chamber. 4. Pull the vacuum slowly to 25 psi and let the samples sit for 20–30 min. This step removes air bubbles from the samples to allow the penetration of the fixative into the tissue. The samples will begin to sink. Slowly release the vacuum to return the samples to air. 5. Repeat the above step once to permit all the samples to sink to the bottom of the vials (see Note 23). 6. Slowly release the vacuum and remove the vials from chamber. Keep the sample vials in the fume hood overnight to complete the fixation process.

3.2.2 Tissue Dehydration and Infiltration

1. The next day, remove the FAA fixation solution using a Pasteur pipette and replace with 50% ethanol. Incubate for 30 min at room temperature.

Analysis of Meristem Development

173

2. Dehydrate the samples through a graded ethanol series (60%, 70%, 80%, 90%, 99.5%), leaving the tissues in each solution for 30 min. 3. Wash the samples twice with 100% (v/v) ethanol for 1 h each. 4. Transfer the dehydrated samples into 50% (v/v) Technovit 7100 resin. Keep the samples at room temperature for 30 min. 5. Repeat the above step with 70% (v/v) and 90% (v/v) resin. 6. Replace with 100% resin (v/v) two times for 1 h each. 7. Replace with 100% resin (v/v) and store the samples at least overnight at 4 °C (see Note 24). 3.2.3 Tissue Staining and Embedding

1. Place a drop of neutral red solution onto a slide and place the sample in it for 3–4 s (see Note 25). 2. Add Technovit 7100 Hardener II to Hardener I solution at a ratio of 1 mL to 15 mL, mix the solution and pour it into the embedding mold. Place the neutral red-stained samples into the mold (see Note 26). 3. Allow polymerization for 1 h. 4. Place Histobloc on the top of each mold in an upside-down position. 5. Store the samples at least overnight at room temperature to allow full polymerization.

3.2.4

Tissue Sectioning

1. After removing the sample from the mold, place it on a rotary microtome at an 8° knife inclination angle. Use a very sharp microtome knife. 2. Section the tissue at 1–4 μm thickness. The sections will be released by the microtome one by one, unattached to each other. 3. Check continuously for sections that contain the SAM tissue (see Note 27). 4. Remove each SAM tissue section from the microtome with tweezers and float it in a petri dish filled with distilled water (see Note 28). 5. Put one section on a cover glass and dry it on a slide warmer at 42 °C for 10–15 min (see Note 29).

3.2.5 Toluidine Blue Staining

1. Soak the cover glass with ribbon in Toluidine blue solution for 1–2 min. 2. Soak the cover glass with ribbon in distilled water for 3–4 min to remove the excess stain. Check the sections under the microscope. Repeat step 1 if the sections are too lightly stained, and repeat step 2 if they are too strongly stained (see Note 30). 3. Place the cover glass with the ribbon on the slide warmer at 42 °C for 10–15 min.

174

Mona M. Monfared et al.

Fig. 2 Histological sections of Arabidopsis vegetative shoot apical meristems. (a) Longitudinal section. (b) Transverse section. The shoot apical meristem is marked with an asterisk (*). Ten-day-old wild-type Col-0 seedlings were fixed in FAA solution and infiltrated with Technovit 7100 resin before sectioning at 2 μm thickness. Scale bars: 50 μm 3.2.6 Mounting and Visualization

1. Place a drop of ImmunoHistoMount onto a glass slide. 2. Place the cover glass with the completely dried ribbon upsidedown on top of the ImmunoHistoMount, ribbon side down (see Note 31). 3. Place the slide on the slide warmer overnight to allow complete fixation of the cover glass to the slide. 4. Visualize the stained sections using a stereo-microscope (see Fig. 2).

3.3 Meristem Size Measurement

1. Launch the Image J software and open a meristem image file. 2. Set the scale by choosing the “straight line selections” tool from the tool bar below the main menu and drawing a straight line along the scale bar in the raw image (see Note 32). Go to the Analyze scroll down menu and select “set scale.” In the window that appears, fill in the “known distance” and “unit of length” customizable space. After entering the correct values click “ok” to close the calibration window. 3. Draw lines on the image to make the desired measurements. Two parameters can be measured on longitudinal shoot apical meristem images: width and height (see Fig. 3). The width is measured by drawing a horizontal line between the interior boundaries of the visible ridges that will form the next primordia. The height is measured by drawing a perpendicular line from the point equidistant along the width up to the tip of the meristem. 4. Once an accurate line is drawn, use the Ctrl+M keyboard combination to open a results window where all the measurements will be saved.

Analysis of Meristem Development

175

Fig. 3 Measurement of the Arabidopsis vegetative shoot apical meristem. (a) Schematic of the shoot apex showing the meristem as a dome of cells between the developing leaf primordia (lp). (b) Longitudinal section through a seven-day-old shoot apex. The meristem width (w) and height (h) measurements are made as shown

5. To make a line permanent on the image (useful for example to measure SAM height in longitudinal sections), click the “Edit” scroll down menu in the menu bar and select “draw.” 6. Open the next image file and repeat steps 2, 3, and 4. The new measurement will be saved in the same results window, which can be saved later as a Microsoft Excel table (see Note 33). 7. Perform statistical analysis using a program such as Microsoft Excel or OpenOffice SpreadSheet. 3.4 Vegetative Meristem RNA In Situ Hybridization 3.4.1 Fixation of Tissue Sections

For section preparation, the seedling tissues must be embedded in paraffin. To preserve morphology the tissues should first be fixed. Next, they are dehydrated and stained with a dye to facilitate sectioning. For tissue embedding, an organic solvent gradually replaces the ethanol present in the tissue. The non-toxic solvent Histoclear is typically used in paraffin-based embedding procedures. Finally, paraffin is slowly introduced into the solvent solution until it reaches 100% concentration. Numerous solution changes are needed for the paraffin to fully penetrate the tissue. Fixation is one of the most critical steps for successful in situ hybridization, for which a compromise must be found between tissue morphology preservation and good probe penetration. A 4% paraformaldehyde fixation solution penetrates 2 mm of tissue in about 1 h at room temperature. Because of the structure formed by the cotyledons, a bubble of air is usually trapped on the top of the shoot apex, preventing full penetration of the fixative into the SAM tissue. Moreover, the stacking of leaf primordia over the SAM creates a layer of tissue the fixative must pass through before reaching the meristematic cells. Chopping the tip of the seedling greatly enhances the SAM fixation process and allows all tissues to be crosslinked at the same pace.

176

Mona M. Monfared et al.

Fig. 4 Dissection of the Arabidopsis vegetative shoot apex for in situ hybridization. Lay a seedling sideways on a slide under a dissecting microscope. Holding the base of the hypocotyl firmly, make a sharp cut to remove the cotyledons and the oldest leaf primordia (arrow 1). Make another cut midway along the hypocotyl to remove the root system (arrow 2). Immediately place the dissected tissue in cold fixation solution

1. Prepare the 4% paraformaldehyde fixation solution fresh. Aliquot ~10 mL of the 4% paraformaldehyde fixation solution into each scintillation vial kept on ice (see Note 18). Preserve half of the fixation solution on ice for step 7. 2. Pull a seedling from its culture plate (see Note 34). Lay the seedling sideways on a slide under a stereo-microscope. Using a sharp single-edge razor blade trim the cotyledons and oldest leaf primordia, and then cut midway along the hypocotyls to remove the roots (see Fig. 4). 3. Immediately place the dissected tissue in the fixative and swirl. The dissecting process should be prompt to preserve tissue integrity. If the sample begins to lose moisture during dissection, add a few drops of fixative while trimming to avoid drying of the sample (work under a chemical hood!). Up to 50 dissected seedlings can be placed in a single scintillation vial. 4. Loosen the caps of the scintillation vials and place them on ice in the vacuum apparatus. 5. Pull the vacuum slowly to 25 psi and let the samples sit for 10 min. This step removes air bubbles from the samples to allow the penetration of the fixative into the tissue. The samples will begin to sink. Slowly release the vacuum to return the samples to air. 6. Repeat the above step once to permit all the samples to sink to the bottom of the vials. 7. Pipet the fixation solution into a hazardous waste bottle and replace with fresh cold fixation solution. 8. Place tissue in vials at 4 °C for 16 h under constant gentle rotation.

Analysis of Meristem Development 3.4.2

Tissue Dehydration

177

Ethanol solutions used for dehydration should be kept at 4 °C. 1. Pipet the fixation solution into a hazardous waste bottle and replace it with the same volume of cold 15% ethanol. Incubate at 4 °C with rotation for 30 min. 2. Dehydrate the tissues through an ethanol series (30%, 50%, 75%, 95%, 100%), incubating at 4 °C with rotation for 30 min each. 3. Pour off the 100% ethanol solution and replace with 100% ethanol containing 0.1% Eosin Y to stain the tissue (see Note 35). Let the vials rotate overnight at 4 °C.

3.4.3 Tissue Embedding in Paraffin

1. Remove the vials from 4 °C and allow the solution to come to room temperature. 2. Remove the 100% ethanol solution, replace with room temperature 100% ethanol and incubate for 30 min. 3. Remove 1/4 of the volume of ethanol and replace it with the same volume of Histoclear, incubate for 30 min (see Note 36). Repeat this step four times. 4. Pour off the remaining solution and replace with 100% Histoclear. Repeat this step once. 5. Pour off the solution and fill the vial 1/3 full with 100% Histoclear. Add an equal volume of paraplast X-tra paraffin chips. Incubate overnight at room temperature. 6. Prepare a 500 mL beaker of paraplast chips and leave it in a 56 ° C oven for the chips to melt overnight. 7. In the morning, place vials at 42 °C to melt the paraplast chips. Keep adding paraplast chips and allow them to melt at 42 °C. Repeat until the vials are full. The remainder of the chips should melt within an hour or two. 8. When all chips have melted, transfer the vials to a 56 °C oven. 9. Remove the Histoclear/paraffin mixture, replace with pure molten paraplast from the 500 mL beaker and incubate at 56 °C. 10. Replace the molten paraplast every 8–10 h. Make at least 6 changes of molten paraplast. 11. On a bench-top microscope slide warming table, pour the paraplast solution containing the tissues into weighing boats. Split the contents of each vial into two weighing boats. Align and orient the samples using a pipet or needle (see Note 37). The embedded tissues can be stored at 4 °C for several months.

3.4.4 Tissue Sectioning and Mounting

1. Cut into the paraffin bed with a sharp single-edge razor blade to isolate each embedded seedling in a block, leaving at least 4 mm of paraffin on each side of the tissue.

178

Mona M. Monfared et al.

2. Mount the block on the microtome sample holder (see Note 38). 3. Trim the block further to a rectangular or trapezoid shape, leaving about 2 mm of paraffin around the tissue. The long edges must be parallel to one another and to the hypocotyl of the sample. 4. Insert the sample holder into the microtome. 5. Align the hypocotyl of the tissue sample parallel to the knife blade, by moving the sample holder as needed (see Note 39). 6. Section the tissue at an 8° knife inclination angle for 8 μm-thick sections. Manipulate the paraffin ribbons containing the sectioned tissue onto a clean microscope slide using fine paintbrushes. 7. Screen the ribbons under a dissecting microscope for the serial sections containing SAM tissue. 8. Trim the ribbons of interest with a razor blade to remove unwanted sections. 9. Transfer the ribbons, shiny side down, onto an adhesion slide using a fine paintbrush. Orient them on the slide in parallel rows. Place the ribbons such that each is a few millimeters away from the edge of the slide. 10. Add RNase-free water beneath the ribbons and transfer the slide onto the slide warmer set at 42 °C. 11. Add more water so as to cover nearly all of the slide surface and let the ribbons expand for 10 min. The ribbon should expand approximately 50% of its initial length. 12. Once the ribbons are fully expanded, slightly incline each slide and wick away the water using a kimwipe, always avoiding touching the ribbons. 13. Leave on a slide warmer overnight at 42 °C to dry. Cover with a lid to prevent dust falling on the slides. Once fully dry, sections can be stored in a clean microscope slide box at 4 °C for a couple of months. 3.4.5 Riboprobe Preparation (See Note 40)

In situ hybridization techniques were initially established using radioactively labeled probes. Improvements in non-radioactive labeling methods have led to equivalent sensitivity in signal detection. The most commonly used DIG indirect labeling approach uses a hapten coupled to UTP. Riboprobe can be made by run-off transcription from a linearized plasmid template. This requires cloning of the cDNA of interest into a suitable vector (such as pBSKII) that allows linearization with restriction enzymes leaving 5′ overhangs, because 3′ overhangs or blunt ends can lead to RNA synthesis artifacts. Alternatively, riboprobe template can be obtained by PCR amplification of plasmid DNA corresponding to

Analysis of Meristem Development

179

the sequence of interest, using flanking primers annealing to the M13 forward and M13 reverse sequences that are contained within most conventional vectors. Plasmid digestion and purification method: 1. Linearize 10 μg of plasmid DNA in a restriction digest with a total volume of 100 μL. Incubate the digestion reaction at 37 ° C for at least 2 h. Check for completion of the digestion by running a few μL of the digestion on a 1% agarose minigel. 2. Add 100 μl of phenol:chloroform:isoamyl alcohol (25:24:1) and vortex (see Note 41). 3. Incubate for 5 min at room temperature and spin at 17,900 × g for 5 min. 4. Transfer the aqueous phase containing the DNA to a new tube and add 100 μL chloroform:isoamyl alcohol (24:1). Vortex briefly. 5. Incubate for 5 min at room temperature and spin at 17,900 × g for 5 min. 6. Transfer the aqueous phase containing the DNA to a new tube and precipitate the linearized DNA by adding 1/10 volume (10 μL) of 3 M NaOAc and 3 volumes (300 μL) of 95% ethanol. Vortex briefly. 7. Incubate for 20 min at -20 °C and spin at 17,900 × g for 10 min. 8. Discard the supernatant and wash the pellet with 80% ethanol. Spin at 17,900 × g for 5 min. 9. Discard the supernatant, air dry the pellet, and resuspend in 50 μL of RNase-free Milli-Q water. PCR Amplification method: 1. Perform PCR on plasmid DNA using M13 forward and M13 reverse primer sequences. 2. Purify DNA template using a commercial PCR-purification kit. 3. Resuspend purified DNA in 50 μL of RNase-free Milli-Q water. 3.4.6 Riboprobe Synthesis

The riboprobe synthesis protocol is adapted from the Roche DIG RNA labeling mix manual. RNA is labeled to a density of 1 digoxogenin for every 20–25 nucleotides. 1. Prepare the transcription mix by combining in a 1.5 mL eppendorf tube 10 μL linearized template (2 μg), 2 μL DIG RNA labeling mix, 4 μL 5 × transcription buffer, 1 μL RNase-out and add RNase-free Milli-Q distilled water to a total volume of 19 μL. Add 1 μL of RNA polymerase (T7, T3, or SP6). Incubate at 37 °C for a minimum of 1 h and up to 4 h if a high concentration of probe is required.

180

Mona M. Monfared et al.

2. Load 1 μL of the transcription reaction on a 1% agarose minigel to check the RNA synthesis. 3. Add 1 μL DNaseI and incubate at 37 °C for 15 min. 4. Precipitate the riboprobe by adding 1/10 volume (2 μL) of 3 M NaOAc and 3 volumes (60 μL) of 95% ethanol. Vortex briefly. 5. Incubate at -20 °C for 20 min and spin at 17,900 × g for 10 min. 6. Wash the pellet with 80% ethanol and air dry. 7. Resuspend the pellet in 50 μL RNase-free Milli-Q distilled water. 8. Run 3 μL of the riboprobe on a 1% agarose minigel to check probe synthesis and verify DNA absence. Multiple RNA bands can be observed because the riboprobe may adopt secondary structures. Save 3 μL of the probe for the gel in the next step (see Note 42). If DNA is still present in the sample, return to step 3 and repeat the DNase I treatment. 9. If the riboprobe is longer than 1000 nucleotides, it may be beneficial to hydrolyze it for better tissue penetration (see Note 43). Long probes tend to stick randomly to the sample and give unspecific background. Add 50 μL of 2 × Probe hydrolysis solution to the 50 μl of riboprobe. Incubate at 60 °C for time t where t = (Li - Lf)/(k × Li × Lf), where Li is the initial length of the RNA probe in bp, Lf is the final length of the RNA probe in bp (optimal at 200–250 bp), K is the rate constant (K = 110 bp/min). 10. Stop the hydrolysis with 1/20 volume (5 μL) of 10% glacial acetic acid. Load 6 μL on a 1% agarose minigel, along with the 3 μL retained from the pre-hydrolysis riboprobe. The posthydrolysis probe should give a fuzzy band of smaller molecular weight. 11. Precipitate the riboprobe by adding 1 μL of 10 mg/mL yeast tRNA, 1/10 volume (10 μL) of 3 M NaOAc, and 3 volumes (300 μL) of 95% ethanol. 12. Incubate at -20 °C for 2 h and spin at 17,900 × g at 4 °C for 30 min. 13. Wash the pellet with 80% ethanol and air dry. 14. Resuspend the probe in 40 μL of 50% formamide and freeze at -20 °C. This should yield enough probe for 40 slides (see Note 44). 3.4.7 Slide PreHybridization and Hybridization Treatments

Slides can be placed back-to-back in the rack slots to maximize space. Make sure to pull them apart at each wash/incubation step to remove any solution trapped between them from the prior step.

Analysis of Meristem Development

181

An 18-slide experiment, corresponding to one rack, will require about 6 L of RNase-free Milli-Q water. After rehydration of the sectioned tissue samples, incubation with a protease is conducted as a pretreatment to permeabilize the tissue and thereby facilitate penetration of the probe. Proteinase K is an effective protease, the concentration of which is critical because too extensive protein degradation results in loss of tissue morphology while too limited protein degradation reduces probe penetration and produces a weaker signal. To enhance tissue permeabilization a gentle acid hydrolysis step is used prior to the proteinase K treatment, which helps break down the cell walls and partially solubilizes highly crosslinked basic nuclear proteins. Temperature, buffer composition and pH, blocking reagents and probe concentration are interdependent parameters that affect the ability of the riboprobe and tissue RNA to form duplexes during hybridization. Because the target RNA is embedded in the tissue sections, classical hybridization kinetics and Tm (melting point temperature, the temperature at which 50% of the probe is dissociated) cannot be calculated but must be determined empirically. Optimal hybridization temperatures vary from 45 to 60 °C. Formamide, which destabilizes the hydrogen bonds between probe and target sequences, is the organic solvent of choice to reduce the melting temperature of RNA-riboprobe duplexes, thereby permitting the hybridization to take place at lower temperatures. Hybridization stringency is also determined by the concentration of monovalent cations: under high Na+ concentrations (high stringency), only sequences with high degree of homology form stable duplexes. Yeast tRNA is used as blocking reagent to reduce non-specific binding of the riboprobe. Increased probe concentration leads to increased signal until a saturation point is reached. Further concentration increases result in non-specific binding that masks the true RNA localization. 1. Dewax the tissue sections in 100% histoclear twice for 10 min each with gentle shaking. Wash twice with 100% ethanol for 2 min each. 2. Rehydrate by dipping the slides sequentially into a freshly made graded series of ethanol dilutions (95%, 90%, 80%, 70%, 50%, 30%) with gentle shaking, for 2 min each step. Wash twice in RNase-free Milli-Q distilled water for 2 min each. 3. Incubate the slides in 0.2 M HCl for 20 min. 4. Prepare 50 mL of proteinase K buffer and preheat to 37 °C in the oven. 5. Wash the slides in RNase-free Milli-Q distilled water for 5 min. 6. Wash the slides in 1 × PBS for 5 min to neutralize the remaining acid.

182

Mona M. Monfared et al.

7. Wash the slides in RNase-free Milli-Q distilled water for 5 min. 8. Add 10 μL of proteinase K stock to 50 mL of proteinase K buffer, mix and incubate the slides at 37 °C for 30 min. 9. Wash the slides in 1× PBS for 5 min. 10. Stop the digestion by washing the slides in 2 mg/mL glycine in 1× PBS for 2 min. 11. Wash the slides in 1× PBS for 30 s. 12. Incubate the slides in 1× PBS containing 3.7% formaldehyde (mix 5 mL of 37% formaldehyde with 45 mL of 1× PBS) for 20 min (see Note 45). 13. Wash the slides in 1× PBS for 5 min. 14. Dehydrate the tissues in graded ethanol series in the reverse of step two (from RNase-free Milli-Q water to 100% ethanol). 15. Pour off all ethanol and air dry the slides. Under humid conditions, a vacuum treatment may be needed to eliminate any remaining ethanol. 16. Prehybridize the tissues by placing the slides on 55 °C slide warmer while adding 200 μL of prewarmed hybridization solution per slide, and cover with a cover glass. Alternatively, to conserve hybridization solution two slides may be sandwiched together with the tissue sides facing each other. Place the long edge of the sandwich on a clean surface and pipet 250 μL of prewarmed hybridization solution across the edge. The sandwich will spread the solution using capillary action. 17. The amount of riboprobe to use per slide ranges from 10 to 50 ng per kb. For riboprobe hybridization to an RNA target with an expected identical sequence, it is recommended to start the hybridization with a 50% formamide solution containing 1 M Na+ and 40 ng of probe (which should correspond to 0.5–5 μL of the 40 μL probe), at 55 °C. Use a total volume of 100 μL hybridization buffer per slide. 18. Incubate slides at 55 °C for at least 1 h in a microscope slide box humidified with paper towels soaked in water. 19. Per slide (multiply by the number of slides to be hybridized with the same riboprobe): aliquot 1 μL of DIG-labelled riboprobe, add 5 μL of 50% formamide and heat to 75 °C for 2 min to disrupt the secondary structure. 20. Add 100 μL of warm hybridization solution per slide (multiply by the number of slides) and pipet onto the slide from one edge. Overlay tissue sections by gently applying a cover glass from one edge of the slide to the other, avoiding air bubbles. The capillary action method described in step 16 may also be used here if the same probe will be hybridized to multiple slides.

Analysis of Meristem Development

183

21. Incubate slides overnight at 55 °C in a microscope slide box humidified with paper towels soaked in water. 22. Prepare the wash solutions for the next day. Prewarm 2 L of 0.2× SSC, 0.1% SDS washing solution at 55 °C overnight. 3.4.8 Post-Hybridization Washes

Post-hybridization washes with a high stringency buffer remove unbound riboprobe and separate mismatched duplexes. An RNAse treatment between the washes degrades the non-hybridized or washed ssRNA probes. Riboprobe degradation products will then be eliminated with subsequent washes. For efficient washes, it is crucial to lay the slides flat in a glass dish and immerse them in 200 mL of washing solution under gentle agitation. 1. Dip the slides in 0.2 × SSC, 0.1% SDS that has been prewarmed to 55 °C. The slide/cover glass duplexes will separate. Remove the cover glass. 2. Lay the slides flat at the bottom of a glass dish and wash with 0.2 × SSC, 0.1% SDS that has been prewarmed to 55 °C on a shaker for 10 min at 55 °C. Repeat this step once. 3. During the washes, prepare 400 mL of 2 × SSC and warm it at 37 °C for the RNase treatment in step 5. 4. Incubate the slides in 2 × SSC for 2 min at room temperature. 5. Prepare 10 μg/mL RNase solution in 2 × SSC that has been prewarmed to 37 °C and incubate the slides for 30 min at 37 ° C, under gentle agitation (see Note 46). 6. Wash the slides in 2 × SSC for 2 min at room temperature. 7. Perform a high stringency wash with 0.2 × SSC, 0.1% SDS that has been prewarmed to 55 °C on a shaker, for 10 min at 55 °C. Repeat this step once.

3.4.9

Detection

The detection process starts with immunological detection of the DIG-labeled riboprobes. Best results are obtained using an antiDIG antibody coupled to alkaline phosphatase (AP). The AP reaction with NBT and BCIP produces formazan, a stable blue/purple dye with bright reflective properties. Formazan does not precipitate, allowing long incubation times of greater than 24 h. Because formazan is soluble in organic solvents, it is crucial to embed the slides in aqueous mounting medium after detection. Mount the slides in a 50% glycerol solution, which gives good resolution for microscope observation and imaging as well as allowing unlimited storage of the stained sections. Process all washes and buffer incubations in glass dishes, with the slides lying flat at the bottom. Gentle rotation should be applied during the wash steps.

184

Mona M. Monfared et al.

1. Rinse the slides in 1 × TBS for 5 min at room temperature. 2. Incubate the slides in 0.5% Boehringer Blocking reagent in 1 × TBS for 1 h at room temperature. 3. Rinse the slides in 1 × TBS containing 0.5% Bovine Serum Albumin (BSA) and 0.1% Triton-X100 for 30 min at room temperature. 4. Replace the solution in step 3 with 1 × TBS containing 0.5% BSA and incubate at room temperature for 5 min. 5. Remove the slides from the glass dish and add anti-DIG AP conjugate (see Note 47) in 1 × TBS containing 0.5% BSA. Use 100 μL per slide and gently apply a cover glass, avoiding air bubbles. 6. Place the slides flat in a microscope slide box humidified with paper towels soaked in water, and incubate 2 h at room temperature. 7. Dip the slide/cover glass duplexes in 1 × TBS, 0.5% BSA, 0.1% Triton-X100. Remove the cover glass. 8. Lay the slides flat in the glass dish and rinse them in 1 × TBS, 0.5% BSA, 0.1% Triton X-100 for 10 min at room temperature on a shaker. Repeat this step three times. 9. Rinse slides in detection buffer for 15 min at room temperature with shaking. Save a few mL of this buffer for the next step. 10. Apply 100 μL of detection buffer per slide, add cover glass and place slides in a microscope slide box humidified with paper towels soaked in water. Incubate in the dark for 4–36 h (see Note 48). 11. When a satisfactory signal is observed, stop the reaction by dipping the slides in Milli-Q water. 12. Mount the slides by adding 80 μL of 50% glycerol and overlay with a cover glass, being careful that glycerol is evenly distributed across the slide. 13. Visualize the slides using a stereo-microscope and troubleshoot as necessary (see Note 49). 3.5 Confocal Laser Scanning Microscopy of the Inflorescence Meristem 3.5.1

Tissue Fixation

Most plant cells, particularly those in the aerial parts of the plant, produce a variety of substances that protect them from excess light radiation and as a result also prevent laser light and florescent emission from penetrating the tissue. Therefore, in contrast to mature embryo meristems, which are semi-transparent, inflorescence meristem tissues require fixation prior to staining and clearing. 1. Prepare fresh MAA fixation solution (see Note 13) and dispense ~10 mL fixation solution into each glass scintillation vial or similar container (see Note 18).

Analysis of Meristem Development

185

2. Clip off the young inflorescences when the bolting stem is a few cm tall. Remove the older flowers but leave the 5–7 smallest visible flower buds to prevent inadvertently damaging the meristem. Retain several cm of stem because it will facilitate later manipulations. 3. Immediately place the tissues into the MAA fixation solution. Multiple samples can be placed into a single vial but do not tightly pack the tissues. The tissues will float on the surface so gently swirl the vial to completely cover the tissue with fixation solution. 4. Place the vials at 4 °C and incubate overnight (see Note 50). 3.5.2 Tissue Staining and Rinsing

1. Remove the fixation solution and replace it with 80% ethanol (see Note 19). 2. Use fine forceps to transfer the tissues to 1.5 mL microcentrifuge tubes containing ~500 μL 80% ethanol and incubate at 80 °C for 5 min. 3. Use fine forceps to transfer the tissues back into scintillation vials containing fresh MAA fixation solution and incubate at 4 ° C for at least 1 h to complete the fixation process. 4. Rinse tissues twice with distilled water at room temperature. 5. Incubate tissues in 1% periodic acid at room temperature for 40 min. 6. Rinse tissues twice with distilled water. 7. Replace the water with 2–3 mL of fresh 10 μg/mL Propidium Iodide staining solution, such that the tissues are completely submerged in the staining solution. Incubate overnight at room temperature (see Note 20). After the staining incubation the tissue appears pale orange. 8. Replace the staining solution with 2–3 mL of distilled water. 9. Rinse tissues twice with distilled water.

3.5.3 Tissue Dehydration and Clearing

1. Incubate the tissues in 2–3 mL of 50% ethanol for 30 min to 1 h. 2. Incubate in 100% ethanol for 30 min to 1 h. 3. Incubate in 50% ethanol: 50% methyl salicylate for 30 min to 1 h. 4. Incubate in 100% methyl salicylate for at least 10 min (see Note 51).

3.5.4 Mounting and Imaging

1. Pipet a drop of 100% methyl salicylate in the center of a cover glass. 2. Use fine forceps to transfer the sample into the center of a cover glass. Ensure there is enough liquid to keep the sample in place.

186

Mona M. Monfared et al.

Fig. 5 Confocal laser scanning micrographs of an Arabidopsis inflorescence meristem and a floral meristem. (a) Optical longitudinal section of a wild-type Col-0 inflorescence meristem (IFM) producing floral meristems (FM) on the flanks. (b) Optical longitudinal section of a wild-type Col-0 stage 3 flower with sepal primordia (sp) arising from the flanks of the floral meristem (FM). Scale bars: 25 μm

View the specimen under the stereo-microscope. The inflorescence stem should lay flat against the cover glass (see Note 52). If the samples are sufficiently small, several can be transferred to a single cover glass leaving space between them to obtain clear images. 3. Flip the cover glass over atop a depression slide and seal the four corners of the slide with nail polish (see Note 22). 4. Visualize each slide using a Confocal Laser Scanning Microscope (see Fig. 5). Propidium iodide can be excited by a 514 nm argon laser beam and emits between 580 and 610 nm. 3.6 Live Imaging Confocal Laser Scanning Microscopy of the Inflorescence Meristem

1. Melt 1% agarose and pour into a petri dish to a depth of 0.5 cm.

3.6.1

4. View an intact inflorescence under the stereo-microscope and use fine forceps to dissect as many flower buds from the inflorescence as possible until the inflorescence meristem is visible.

Tissue Harvesting

2. Let the agarose harden and then use fine forceps to dig several vertical holes towards the center of the petri dish. Cover the dish until needed. 3. Pour sterile distilled water into a second petri dish.

5. Cut the inflorescence 0.5 cm below the apex. 3.6.2 Tissue Staining and Mounting

1. Use fine forceps to transfer the sample into the petri dish of water and immerse for 2 min. 2. Pipette 20 μL of 1 mg/mL Propidium Iodide staining solution into a 1.5 mL Eppendorf tube. 3. Transfer the tip of the inflorescence into the drop of PI staining solution and stain for 2 min.

Analysis of Meristem Development

187

4. Use fine forceps to place the base of the inflorescence stem vertically into a hole in the agarose dish. Multiple samples may be placed in the same agarose dish for imaging. 5. Fill the agarose dish with sterile distilled water to a depth of 2.5–5 cm so that the inflorescence apices are completely immersed. 6. Visualize each inflorescence meristem using a Confocal Laser Scanning Microscope with a water immersion objective. Propidium iodide can be excited by a 514 nm argon laser beam and emits between 580 and 610 nm. Imaging of fluorophores can be achieved using the appropriate excitation and emission spectra. 3.7 Scanning Electron Microscopy of the Inflorescence Meristem 3.7.1

Tissue Fixation

1. Aliquot ~8 mL of Glutaraldehyde fixation solution into each glass scintillation vial or similar container (see Note 18). 2. Using forceps or sharp scissors, gently clip off each inflorescence apex or single flower with 1 cm of the stem remaining and immediately place it into a scintillation vial. The tissue will float on the surface so gently swirl the vial to completely cover the tissue with fixation solution. 3. Incubate overnight at room temperature under constant rotation. 4. Remove the fixation solution into a hazardous waste bottle using a Pasteur pipet. 5. Optional: perform a secondary osmium tetroxide (OsO4) coating step (see Note 53).

3.7.2 Tissue Rinsing and Dehydration

1. Rinse the tissues three times with 25 mM PB wash solution each. Empty the first two washes into a hazardous waste bottle using a Pasteur pipet. 2. Dehydrate the samples through a graded ethanol series (30%, 50%, 65%, 75%, 89%, 95%, 100%), leaving the tissues in each solution for 15–30 min. 3. Wash the tissues three times with 100% ethanol for 15–30 min each and leave them in the third change overnight at room temperature. 4. The next day, repeat the 100% ethanol wash twice for 15–30 min each. 5. Store the samples in 100% ethanol until ready to dry them (see Note 54).

3.7.3 Critical Point Drying

1. Choose the appropriate size specimen basket to fit the samples. 2. Remove the basket lid and place the basket in a Petri dish.

188

Mona M. Monfared et al.

3. Cut small pieces of paper to fit in the specimen basket and write the genotype or sample name with a pencil. Using the forceps transfer each piece of paper into a separate chamber of the basket. 4. Fill the bottom of the Petri dish with 100% ethanol. 5. Quickly pour the samples from one scintillation vial into the Petri dish. Use forceps to gently transfer the samples into the corresponding chamber(s), minimizing their exposure to air. Repeat this step for each of the scintillation vials. 6. Once all the samples have been transferred close the lid over the specimen basket. 7. Dry the samples in the critical point dryer, following the manufacturer’s instructions (see Note 55). 8. Use forceps to transfer the dried samples to clean, dry scintillation vials for storage. 3.7.4

Tissue Mounting

1. Place two mounting bases on the stereo-microscope base. 2. On top of one mounting base, place a white index card that has been folded in the middle. On the other mounting base, place a mounting stub. 3. Use forceps to lift a conductive sticker by the edge and place it over the mounting stub (see Note 56). Place the two mounting bases side by side. 4. Carefully tip a specimen from the first vial onto the folded white index card. 5. View the specimen under the microscope. For an inflorescence meristem, grip it by the stem with one pair of forceps. With the other pair of forceps gently break off each older flower by pushing it very carefully down away from the stem (see Note 57), until the inflorescence apex and floral primordia are exposed. For a single flower, hold it by the stem and break off two sepals and petals from the tip downward to view the internal floral organs. 6. Once the dissection is finished, use forceps to transfer the specimen onto the mounting stub covered by the sticker. For an inflorescence, carefully affix it by the base of the stem such that it sits perpendicular to the mounting stub. For a flower, affix the side that still contains the sepals and petals to the mounting stub (see Note 58). 7. Coat the samples using a sputter coating apparatus (see Note 59) and visualize them with a scanning electron microscope (see Fig. 6), following instructions specific to the apparatus.

Analysis of Meristem Development

189

Fig. 6 Scanning electron micrographs of an Arabidopsis inflorescence apex and a developing flower. (a) Wildtype Landsberg erecta (Ler) inflorescence meristem producing floral meristems in a spiral phyllotaxy. Stage 1 through stage 5 floral meristems are shown. (b) Wild-type Ler developing flower with all four sepals removed to reveal the petal, stamen and carpel morphology. Scale bars: 50 μm 3.8 Floral Organ Number Counting

3.8.1 Flower Dissection and Counting

The first 10 flowers from each of 10 plants per genotype are sequentially removed over a period of days and all organs are counted and recorded. 1. Use small scissors or fine forceps to remove the first 10 flowers on the first plant (see Note 60). 2. Grip the first flower by the pedicel with forceps and move it under a stereo-microscope. 3. With the other hand, use another set of forceps to remove each sepal sequentially by drawing the organ down and away from the pedicel and then pulling gently. Score and record the total number of sepals (see Note 61). 4. Repeat step 3 with the petals and then the stamens. 5. For the carpels that remain, make a cross-section through the intact gynoecium using dissecting scissors or a sharp blade. Count and record the number of individual carpels revealed in the cross section (see Note 62). 6. Repeat steps 1–5 with the first 10 flowers on the next plant.

3.8.2 Data Analysis

1. Enter the data into a statistical program (e.g., Microsoft Excel or OpenOffice Spreadsheet) for analysis (see Fig. 7a). Arrange the data according to each genotype. For each genotype, the first ten flowers of ten plants are used for analysis, for a total of 100 flowers per genotype. 2. Calculate the mean and standard error values for each data set. A chi-square test can be performed to determine the statistical significance of values that differ between two genotypes.

190

Mona M. Monfared et al.

Fig. 7 Arabidopsis floral organ counts. (a) Sample tabulation of floral organ counting raw data in a Microsoft Excel file format. The total number of sepals, petals, stamens and carpels counted in each of flower on each plant is tabulated under the appropriate column. (b) Sample bar graph generated in Excel showing mean floral organ number for the sepals, petals, stamens and carpels from wild-type plants and those of three additional genotypes (A, B and C). The error bars represent standard error (S.E)

3. Generate a bar graph showing the mean number of sepals, petals, stamen, and carpels for each genotype, including wild type (see Fig. 7b). Mean floral organ number is placed along the Y axis and each floral organ type is placed along the X axis. Include error bars representing the standard error of measurement (S.E.) for each data set.

Analysis of Meristem Development

4

191

Notes 1. Propidium Iodide is a nucleic acid intercalating agent and should be handled wearing gloves and suitable protecting clothing. Treat all materials (pipets, tubes, etc.) that touch the solution as hazardous waste. 2. Methyl salicylate is an oral, dermal and inhalation irritant and should be handled wearing gloves and suitable protective clothing, and with adequate ventilation. 3. Formaldehyde and formaldehyde-containing solutions are toxic and should be handled wearing gloves under a fume hood. Treat all materials (pipets, tubes, etc.) that touch the solutions as hazardous waste. 4. Technovit 7100 resin is useful for working with plastic sections at room temperature. This resin hardens easily at room temperature, so it should be stored at 4 °C. 5. Toluidine blue is a general-purpose stain/dye. This dye stains certain cellular components with different colors: i.e., Lignin/ phenol is stained green/blue-green, pectin’s stain pink/reddish purple, and DNA stains green/blue-purplish. 6. The free software can be downloaded at https://imagej.nih. gov. Linux, Windows or MacOSX versions are available. A commonly used distribution of ImageJ called Fiji that bundles the core ImageJ application with a pre-installed, curated plugin package is available at https://imagej.net/software/fiji/ downloads. Various text and video tutorials for using the program are available online. 7. The ImageJ program can open, process, and save images in any format (TIFF, JPEG, PNG, GIF, BMP and raw data). For a complete list of supported data types, refer to the software documentation web page. 8. Glassware (cylinders, beakers, flasks, bottles, slide racks or Copeland jars) must be baked for 4 h at 180 °C to inactivate any trace of RNAse. It is not necessary to treat the water with DEPC, but simply use freshly autoclaved Milli-Q water stored in clean, baked glassware. 9. The Roche RNA DIG labeling and detection kits are excellent for preparation and in situ detection of riboprobes. The DIG RNA Labeling kit contains SP6/T7 RNA Polymerases, Protector RNase Inhibitor, DNaseI, and NTP labeling mixture. The DIG Nucleic Acid Detection kit includes Anti-Digoxigenin-AP Conjugate antibody, NBT, BCIP and Blocking reagent. Reagents in the system, such as Blocking reagent, NBT and BCIP, are also available separately from Roche. Alternative reagents are available from other commercial sources and give

192

Mona M. Monfared et al.

good results (e.g., RNA Polymerases and RNAsin from Promega). To prepare the 5 × blocking reagent solution, the 1 × TBS buffer must be warmed before adding the blocking reagent powder. The solution will look cloudy. 10. Paraformaldehyde powder and solution is toxic and should be handled wearing gloves under a fume hood. Treat all materials (pipets, tubes, etc.) that touch the powder or solution as hazardous waste. Because paraformaldehyde vapors are also toxic, verify that the vacuum system used for tissue infiltration does not exhaust into the laboratory. 11. Plant tissues have a cuticle and thus float on the surface of the fixative, preventing proper infiltration. The combination of the Triton X-100 detergent and DMSO solvent enhances the penetration of the fixative while reducing the disrupting effect of the vacuum on the tissue morphology. 12. Formamide is highly corrosive in contact with skin and eyes, so the hybridization buffer should be handled wearing gloves under a fume hood. Treat all materials (pipets, tubes, etc.) that touch the solution as hazardous waste. 13. Methanol and acetic acid should be handled wearing gloves and suitable protecting clothing. Treat all materials (pipets, tubes, etc.) that touch the solution as hazardous waste. 14. For inflorescences and single flowers, use a 10 × 5 mm specimen mount and corresponding mounting bases and conductive stickers. 15. Glutaraldehyde is highly toxic and should be handled wearing gloves under a fume hood. Treat all materials (pipets, tubes, etc.) that touch the solution as hazardous waste. 16. Imbibing seeds at cold temperature prevents embryo development during this step. Fixation is not necessary to prepare embryos for confocal laser scanning microscopy. 17. Isolating intact embryos is a key step for high quality imaging and morphology studies. The technique can be difficult to master at the beginning; it requires patience and practice so include extra seeds when trying to dissect embryos for the first time because inevitably some samples will be damaged. Sit comfortably with your forearms resting on the bench so as to have steady hands. Temporarily curtailing caffeine consumption may also be beneficial. 18. One scintillation vial should be used per genotype or experimental batch. Take care not to pack the samples tightly into the vial; there should be room between them as they float on the surface of the solution. Write the name of the sample on the outside of the glass using a marker and cover it with a piece of clear tape so it does not wash off during the ethanol steps.

Analysis of Meristem Development

193

19. Remove solutions by carefully pipetting off the liquid with Pasteur pipets or a micropipette, taking care not to touch the samples to avoid tissue loss or damage. 20. Alternatively, the staining process may be accelerated by adding Propidium Iodide to pseudo-Schiff reagent at a concentration of 100 μg/mL and incubating at room temperature for 1–2 h. 21. Cut off the very ends of the pipet tips with a razor blade. This will widen the tip end to avoid any damage to the sample during the pipetting process. 22. Sealing with nail polish is preferable to other methods because it can easily be dissolved with acetone in the event that samples are dropped or need to be re-oriented for imaging. 23. The fixed samples should remain at the bottom of the vials once the vacuum has been released. If the samples rise to the surface of the fixation solution, reapply the vacuum for another 5–10 min. 24. Resin-infiltrated samples can be kept for at least 6 months at 4 ° C. 25. This step will make the sample handling easier at later steps. 26. The resin will take 10–15 min to harden, so it is important to maintain the correct position of the sample continuously until the hardening process is complete. 27. The SAM should appear as a dome between the leaf primordia for a longitudinal section (see Fig. 2a), or as a circle in the center of the primordia for a transverse section (see Fig. 2b). After reaching the SAM tissue, section very carefully so as not to lose the ribbons that contain the central meristematic sections. 28. Perform this step quickly so that the section does not have time to roll up. 29. Use sharp tweezers to unfold the ribbon in distilled water. After putting the ribbon on the cover glass, check under the microscope again to determine whether the ribbon is completely unfolded and make any needed corrections using sharp forceps. 30. For optimal viewing the tissues should be stained light to medium blue. If the staining is too light the tissues will not be visible against the background, and if the staining is too dark the individual cells and layers will not distinguishable. Note that the addition of ImmunoHistoMount during the mounting step tends to lighten the tissue stain a few shades. 31. Before placing the cover glass carrying the ribbon onto the ImmunoHistoMount, carefully blow the dust off the ribbon to avoid permanently fixing debris onto the slide.

194

Mona M. Monfared et al.

32. Include a scale bar in each raw image to serve as a reference for the ImageJ measurements. 33. Measure at least 10 individual meristems to obtain statistically significant data. 34. For observation of the true vegetative shoot apex the seedling should not be more than 7 days old if grown in continuous light, to ensure that the SAM has not gone through reproductive transition. 35. The Eosin Y staining step is crucial for later tissue sectioning as it helps visualize and orient the tissue. 36. These solution replacements avoid the need to prepare a series of multiple dilutions because pure Histoclear is slowly added to the previous mixed solution, gradually bringing the content of the solution to 100% Histoclear. 37. The subsequent step of mounting the embedded tissue is made easier by aligning the samples as the paraffin hardens around them. Orient the inflorescence apices on their sides with the stems pointing in the same direction, and align them in straight rows of 12–14 inflorescences each. Leave ~1 cm of paraffin between each sample. 38. Scoop a small piece of paraffin onto the tip of a metal spatula and melt it in a flame. Transfer the melted paraffin onto the top of the sample holder and affix it to the bottom of the paraffin block. Hold the two together for ~30 s until the melted paraffin seals around the block. The paraffin block should be mounted such that the microtome knife will strike the longest side. 39. This step will ensure a clean longitudinal section through the shoot apical meristem in the majority of samples. 40. For expression analysis of a gene of unknown pattern, it is important to prepare a sense probe that will serve as negative control for specific hybridization of the antisense probe. 41. Plasmid DNA purification (steps 2 through 8) can also be performed using a commercial plasmid DNA purification kit. 42. The yield of the transcription reaction can be estimated by running an aliquot of the probe on a 1.5% mini agarose gel, next to an RNA standard of known concentration (e.g., RNA Molecular Weight Marker III, 0.3–1.5 kb, Roche). For testing the DIG labeling reaction yield, 1 μL of the probe can be deposited on a piece of filter, UV-crosslinked and incubated with 5 mL of detection buffer containing 5.5 μL of NBT and 4 μL of BCIP. A dark blue spot should become visible in the place where the probe was pipetted.

Analysis of Meristem Development

195

43. The necessity for riboprobe hydrolysis is controversial. Some researchers hydrolyze any riboprobe greater than 1 kb in length, while others find that it is not required to obtain a good signal. If a riboprobe greater than 1 kb in length gives a weak signal, then hydrolysis is recommended. 44. When using side-by-side sense and antisense probes, it is crucial to first load them on a 1.5% mini agarose gel to compare their concentrations in order to use the same amount of riboprobe per slide. 45. This step helps to re-fix the tissues after the destabilizing proteinase K treatment. 46. This step can be essential to eliminate any background signals. The washing temperature can also be raised to 65 °C to help reduce background. 47. Dilute 1:1000 to 1:500 for low abundance transcripts. Alternatively, the antibody can be diluted to 1:3000 for an overnight incubation at 4 °C. 48. Most probes require an overnight incubation. Signal from very rare transcripts can be better observed after 48 h; in this case, add fresh detection buffer plus substrate after ~24 h and continue the incubation. 49. If the signal is brown, the pH of the detection solution is probably incorrect. Make sure that the pH is 9.6 and increase the washing time in the detection solution before adding the NBT and BCIP. If the signal appears as a purple haze of ubiquitous staining, there may be unspecific hybridization or antibody recognition problems. Several modifications to the protocol can be tried, such as decreasing the amount of probe or antibody, increasing the hybridization temperature, and/or increasing the duration and temperature of the posthybridization washes. If this does not solve the problem, try a probe designed from another region of the gene of interest. The absence of signal may be due to a variety of causes, such as transcript abundance below the threshold of detection, poorly labeled probe, excessively stringent post-hybridization washes, deficient anti-AP antibody, and/or old NBT/BCIP substrates. Add positive controls to troubleshoot this situation. When using a probe for the first time, it is very informative to run the hybridization experiment side by side with a DIG-labeled riboprobe that is already known to work. 50. Tissues may be stored in fixative solution for several days. 51. Tissues may be stored in methyl salicylate for several months, although the propidium iodide signal intensity will degrade over time.

196

Mona M. Monfared et al.

52. The inflorescence stem should lie as flat as possible along the surface of the cover glass for best imaging of the inflorescence and flower meristems. This may require dissecting away the older flower buds in the drop of methyl salicylate using fine forceps. Grip the sample by the stem with one pair of forceps. With the other pair of forceps gently break off each flower bud by pulling it carefully down away from the stem, until the inflorescence apex is exposed. 53. Osmium tetroxide (OsO4) may be used as a secondary fixative if necessary to add additional density and contrast to the tissue [27]. Prepare 4 mL of a 1% OsO4 solution: 1 mL 4% OsO4, 1 mL 0.1 M PB, 2 mL distilled water for each sample vial and add it to the vial using a Pasteur pipet. Note that OsO4 is highly poisonous, even at low exposure levels, so it should always be handled under a fume hood while wearing gloves. Samples are incubated from overnight to several days, and the tissue should turn black. Usually, overnight is sufficient for both inflorescences and single flowers. If the samples are left too long in the solution, the OsO4 may begin to sediment and leave a grainy black residue on the sample surfaces. After incubation, empty the 1% OsO4 fixation solution into a hazardous waste bottle using a pasteur pipet and replace it with 25 mM PB. Treat all materials (pipets, tubes, etc.) that touch the OsO4 solution as hazardous waste. 54. For storage longer than 1 week, samples should be placed in 70% ethanol rather than in 100% ethanol. 55. Use of safety glasses is advised while operating the critical point dryer. 56. When placing the adhesive sticker on the stub, take care to lay the sticker flat, so that the surface of the stub is as smooth as possible. This will minimize topographical background when viewing the mounted sample in the electron microscope. 57. The dried tissues are brittle and prone to damage unless handled extremely carefully [27]. 58. Place 3–4 inflorescences, or as many as five single flowers horizontally aligned, on a single mounting stub. 59. This step coats the samples with a conductive metal to prevent the buildup of high-voltage charges on the surface during the microscopy process [27]. Generally, a 15–40 nm coating thickness is adequate, but use the minimum coating thickness possible. Over-coating obscures the surface detail and prevents high-resolution imaging, while under-coating can lead to charge buildup in the electron microscope. Samples may be re-coated if charging occurs. When first performing this protocol, use a few wild-type samples to empirically determine the optimal coating thickness needed to obtain satisfactory data.

Analysis of Meristem Development

197

60. This procedure generally takes several days to complete. Begin when the first arising flowers on the plant(s) of interest have opened, but before the floral organs begin to senesce and drop off. Remove each open flower for analysis. Because organ number in unopened flower buds is difficult to accurately quantify, when all the open flowers have been analyzed stop and continue the analysis the next day, once more buds have opened. 61. Some genotypes may produce flowers with fused or mosaic organs consisting of two types of tissue. If an abnormal floral organ is observed, add a new category to the results table and note the frequency of its occurrence [28]. The cellular composition of mosaic floral organs may be investigated using scanning electron microscopy. 62. To prevent excessive damage to the gynoecium, use a sharp cutting motion rather than a sawing motion.

Acknowledgements We thank George Chuck, Harley Smith, Sabine Zachgo, Elliot Meyerowitz, Beth Krizek, Joshua Levin, and Mark Running for sharing protocols and suggesting improvements, and Elisa Fiume, Helena Pires, Jinsun Kim, Cristel Carles, Wassim Hage, Chanman Ha, and JiHyung Jun for helpful comments. This work is supported by grants from the NSF (IOS-1052050) and USDA (CRIS 203021000-048-00D). References 1. Steeves TA, Sussex IM (1989) Patterns in Plant Development. Cambridge University Press, New York 2. Williams LE (2021) Genetics of shoot meristem and shoot regeneration. Ann Rev Genet 55:661–681 3. Smyth DR, Bowman JL, Meyerowitz EM (1990) Early flower development in Arabidopsis. Plant Cell 2(8):755–767 4. Barton MK (2010) Twenty years on: the inner workings of the shoot apical meristem, a developmental dynamo. Dev Biol 341:95–113 5. Somssich M, Je BI, Simon R, Jackson D (2016) CLAVATA-WUSCHEL signalling in the shoot meristem. Development 143:3238–3248 6. Uchida N, Torii KU (2019) Stem cells within the shoot apical meristem: identity, arrangement and communication. Cell Mol Life Sci 76:1067–1080

7. Prunet N (2017) Live confocal imaging of developing Arabidopsis flowers. J Vis Exp 122:55156 8. Running MP, Clark SE, Meyerowitz EM (1995) Confocal microscopy of the shoot apex. Methods Cell Biol 49:217–229 9. Truernit E, Bauby H, Dubreucq B et al (2008) High-resolution whole-mount imaging of three-dimensional tissue organization and gene expression enables the study of phloem development and structure in Arabidopsis. Plant Cell 20:1494–1503 10. Gomez-Felipe A, de Folter S (2019) A simple protocol for imaging floral tissues of Arabidopsis with confocal microscopy. In: de Folter S (ed) Plant MicroRNAs: methods and protocols, vol 1932. Springer Nature, New York, pp 187–195

198

Mona M. Monfared et al.

11. Clark SE, Running MP, Meyerowitz EM (1995) CLAVATA3 is a specific regulator of shoot and floral meristem development affecting the same processes as CLAVATA1. Development 121:2057–2067 12. Fletcher JC (2001) The ULTRAPETALA gene controls shoot and floral meristem size in Arabidopsis. Development 128:1323–1333 13. Xu C, Liberatore KL, MacAlister CA et al (2015) A cascade of arabinosyltransferases controls shoot meristem size in tomaro. Nat Genet 47:784–792 14. Chu H, Qian Q, Liang W et al (2006) The FLORAL ORGAN NUMBER4 gene encoding a putative ortholog of Arabidopsis CLAVATA3 regulates apical meristem size in rice. Plant Physiol 142:1039–1052 15. Pautler M, Eveland AL, LaRue T et al (2015) FASCIATED EAR4 encodes a bZIP transcription factor that regulates shoot meristem size in maize. Plant Cell 27:104–120 16. Jin L, Lloyd RV (1997) In situ hybridization: methods and applications. J Clin Lab Anal 11: 2–9 17. Gall JG, Pardue ML (1969) Formation and detection of RNA-DNA hybrid molecules in cytological preparations. Proc Natl Acad Sci U S A 63:378–383 18. Houben A, Orford SJ, Timmis JN (2006) In situ hybridization to plant tissues and chromosomes. Meth Mol Biol 326:203–218 19. Jackson D (1992) In situ hybridization in plants. In: Bowles DJ, Gurr SJ, McPherson R (eds) Molecular plant pathology: a practical approach. Oxford University Press, Oxford, pp 163–174 20. Suzaki T, Sato M, Ashikari M et al (2004) The gene FLORAL ORGAN NUMBER1 regulates floral meristem size in rice and encodes a

leucine-rich repeat receptor kinase orthologous to Arabidopsis CLAVATA1. Development 131: 5649–5657 21. Ambrose BA, Lerner DR, Ciceri P et al (2000) Molecular and genetic analyses of the Silky1 gene reveal conservation in floral organ specification between eudicots and monocots. Mol Cell 5:569–579 22. Carles CC, Lertpiriyapong K, Reville K, Fletcher JC (2004) The ULTRAPETALA1 gene functions early in Arabidopsis development to restrict shoot apical meristem activity, and acts through WUSCHEL to regulate floral meristem determinacy. Genetics 167:1893– 1903 23. Chuck G, Muszynski M, Kellogg EA et al (2002) The control of splikelet meristem identity by the branched silkless1 gene in maize. Science 298:1238–1241 24. Bowman JL, Smyth DR, Meyerowitz EM (1991) Genetic interactions among floral homeotic genes of Arabidopsis. Development 112:1–20 25. Zhao Y, Medrano L, Ohashi K et al (2004) HANABA TARANU is a GATA transcription factor that regulates shoot apical meristem and flower development in Arabidopsis. Plant Cell 16:2586–2600 26. Murashige T, Skoog F (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol Plant 15:473– 497 27. Bozzola JJ, Russell LD (1999) Electron microscopy: principles and techniques for biologists, 2nd edn. Jones and Bartlett, Sudbury 28. Levin JZ, Meyerowitz EM (1995) UFO: an Arabidopsis gene involved in both floral meristem and floral organ development. Plant Cell 7:529–548

Chapter 8 Cell Biological Analyses of Anther Morphogenesis and Pollen Viability in Arabidopsis and Rice Fang Chang, Shuangshuang Wang, Zesen Lai, Zaibao Zhang, Yue Jin, and Hong Ma Abstract Major advances have been made in our understanding of anther developmental processes in flowering plants through a combination of genetic studies, cell biological technologies, biochemical analyses, microarray and high-throughput sequencing-based approaches. In this chapter, we summarize widely used protocols for pollen viability staining, investigation of anther morphogenesis by scanning electron microscopy (SEM), light microscopy of semi-thin sections, ultrathin section-based transmission electron microscopy (TEM), TUNEL (terminal deoxynucleotidyl transferase-mediated 2′-deoxyuridine 5′-triphosphate (dUTP) nick end labeling) assay for tapetum programmed cell death, and laser microdissection procedures to obtain specific cells or cell layers for transcriptome analysis. Key words Anther anatomy, Semi-thin section, Pollen viability, Programed cell death, TUNEL assay, Laser microdissection, Pollen starch test, Callose staining, Ultrathin section, SEM, TEM.

1

Introduction Male reproductive development of flowering plants begins with the initiation of stamens in the flower and involves the differentiation of five main specialized cell layers, pollen mother cells surrounded with four well-organized somatic cell layers, the tapetum, the middle cell layer, the endothecium, and epidermal cell layer from the interior to the surface. Afterward pollen mother cells undergo meiosis to produce microspores, which then develop into mature pollen grains containing the reproductive cells, two sperm cells in each pollen grain [1, 2]. One of the most important phenotypes associated with male fertility is pollen viability. Three methods widely used to assess pollen viability are Alexander Red staining, FDA-PI staining, and I2/-KI staining. All these protocols are very useful, simple to handle, and do not need expensive equipment. After staining,

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_8, © Springer Science+Business Media, LLC, part of Springer Nature 2023

199

200

Fang Chang et al.

normal pollen can be easily distinguished from the abnormal ones by color. Specifically, the Alexander staining method assesses pollen viability by staining normal pollen grains magenta-red and the aborted pollen grains blue-green [3], the combination of fluorescein diacetate (FDA) and propidium iodide (PI) stains living cells green and dead cells orange [4–6], whereas the iodine chemical test detects the presence of starch in normal pollen grains by staining black [7]. The I2/-KI staining detectable starch is also accumulated in the vascular tissue and the endothecium of stage 5 to stage 8 anthers, and such starch accumulation is enhanced due to drought, cold, and high or low temperature stresses [7–9]. Anther morphology is often an important aspect of molecular genetic studies of reproductive gene functions and can be characterized using light microscopy of semi-thin sections of paraffinembedded samples. In addition, ultrathin sections of both anther somatic cells and the pollen grain can be examined using electron microscopy to observe subcellular structures. One unique feature of male meiosis in flowering plants is the formation and dissolution of the callose cell wall surrounding the meiocytes during and after meiosis [10–12]. The callose wall is composed mainly of ß-1,3-glucan, which has long been recognized and localized through aniline blue staining in plants. ß-1,3-glucan reacts with aniline blue in solution and is eventually labeled, whereas the unlabeled dye is colorless, such that the callose exhibits bluish green fluorescence after staining [13, 14]. One extremely important process required for normal pollen development is the programmed cell death of the tapetum [15– 19]. In Arabidopsis and rice, the programmed tapetum cell death likely commences at the tetrad stage and is complete after the second pollen mitosis [19, 20]. Because programmed cell death is characterized by DNA breaks, TUNEL (terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling) assay is increasingly applied to rapidly identify and quantify cell death in many tissues and cell types by catalytically incorporating at the 3’-OH DNA ends labeled-16-dUTP, which can then be visualized by microscopy [12, 17, 21–23]. In this chapter, we introduce a modified biotinlabeled 16-dUTP-based TUNEL method. In addition to methods used for anther phenotypic studies, a powerful technique for the isolation of individual cells from tissue sections of plants and animals is laser capture microdissection (LCM). The harvested cells can provide DNA, RNA, and protein for the profiling of genomic characteristics, gene expression, and protein spectra from specific cell types. LCM combined with highthroughput technologies including microarray, next-generation sequencing, proteomics, and metabolomics profiling makes it an extremely powerful tool for understanding the molecular events in different cell and tissue types [24–29]. In this chapter we describe a protocol for preparing samples for performing LCM of anther and other floral tissues for high-throughput transcriptomics.

Anther Morphogenesis and Pollen Viability

2

201

Materials

2.1 Alexander Red Staining

1. A dissecting stereo-microscope with a digital camera. 2. Microscope glass slides, coverslips, nail polish. 3. Carnoy’s fixative: 6:3:1 (v/v/v) ethanol-chloroform-glacial acetic acid. 4. Alexander’s Staining solution; mix the components in the following order to obtain a total volume of 100 mL of solution (see Note 1): 10 mL of 95% ethanol, 1 mL of Malachite green (1% solution in 95% ethanol), 52.5 mL of distilled water, 25 mL of glycerol, 5 mL of phenol, 0.5 g of chloral hydrate, 5 mL of acid fuchsin [1% (w/v) solution in water], 0.5 mL of orange G [1% (w/v) solution in water], 1 mL of glacial acetic acid.

2.2 Iodine Pollen Starch Detection

1. I2/KI solution: 0.2% (w/v) potassium iodide and 1% iodine in distilled water. Prepare by dissolving 0.2 g of potassium iodide in a small amount of distilled water, add 1 g of iodine crystal while stirring and, once dissolved, add the remainder of the 100 mL of water (see Note 2). 2. 70% Ethanol. 3. Microscope glass slides and coverslips. 4. Compound microscope.

2.3 FDA Staining and Imaging

1. FDA-PI staining solution: 10 μg/mL propidium iodide (PI) and 0.5 μg/mL fluorescein diacetate (FDA) in distilled water. 2. Microscope slides and coverslips. 3. Compound microscope.

2.4 Scanning Electron Microscopy for Anther Structure, Anther Dehiscence, and Pollen Wall Structure

1. Methanol. 2. Ethanol. 3. Tertiary-butanol (tert-butanol). 4. Tert-butanol dilution series: prepare 30%, 50%, 70%, 80%, and 90% (v/v) tert-butanol in ethanol. 5. Vacuum freeze dryer. 6. Au [99.99% (φ51 mm)]. 7. A scanning electron microscope (SEM).

2.5 Examination of Anther Anatomy Using Semi-Thin Sections

1. A glass scintillation vial. 2. Positively (+) charged slides (e.g., Thermo Fisher Scientific) and glass covers. 3. 42 °C heat plate.

202

Fang Chang et al.

4. A rotary microtome. 5. A fluorescence microscope with CCD camera. 6. Vacuum bell or vacuum oven. 7. Oven. 8. Heat plate or 42 °C slide-warmer. 9. Single-edge razor blades. 10. A dissecting needle. 11. Two Coplin jars. 12. FAA fixative: 50% (v/v) ethanol, 5.0% glacial acetic acid, 3.7% formaldehyde. 13. 100% Ethanol and graded ethanol series of 70% (v/v), 85%, and 95% ethanol in distilled water. 14. Eosin Y solution: 0.1% (w/v) Eosin Y in 100% ethanol. 15. Technovit 7100 kit including Technovit 7100 liquid, Hardener I, and Hardener II (Heraeus Kulzer Technik, Wehrheim, Germany). 16. Histoform S/Q. 17. 10× Phosphate Buffered Saline (PBS, pH 7.4): dissolve 80 g of NaCl, 2 g of KCl, 14.2 g of Na2HPO4, and 2.7 g of KH2PO4 in 900 mL distilled water. Adjust the PH of the buffer to 7.4 using 10 N HCl. Then add distilled water to the buffer to obtain a total volume of 1 L. 18. Toluidine blue O staining solution: 0.005% (w/v) Toluidine Blue O in 0.1 M PBS, pH 7.4 (see Note 3). 19. Microscope. 2.6

Callose Staining

1. Items 1 through 14 of Subheading 2.5 are also required for the callose staining method in order to prepare resinembedded inflorescence tissues. 2. 0.067 M phosphate buffer (pH 6.8): dissolve 11.81 g of Na2HPO4·12H2O and 4.5 g of KH2PO4 in 1 L distilled water. Adjust the PH of the buffer to 6.8 using 10 N HCl. 3. Aniline blue staining solution:0.05% (w/v) aniline blue in 0.067 M phosphate buffer (pH 6.8).

2.7 Staining of the Pollen Exine and Intine

1. Items 1 through 14 of Subheading 2.5 are also required for the pollen exine and intine staining method in order to prepare resin-embedded inflorescence tissues. 2. De-resins solution: dissolve 2 g of KOH in 5 mL of epoxy propane and 10 mL of methanol. 3. 1% acetic acid-methanol solution: 1:99 (v/v) acetic acidmethanol.

Anther Morphogenesis and Pollen Viability

203

4. 50% methanol buffer: 1:1 (v/v) methanol-0.067 M phosphate buffer. 5. Calcofluor white solution: mix, in 1:1 (v/v) ratio, 1 g/L of calcofluor white and 10% KOH (see Note 4). 6. Diethyloxadicarbocyanine iodide (DiOC2) stock solution (100 mM): dissolve 20 mg of DiOC2 in 435 μL of DMSO to prepare 100 mM stock (see Note 5). 7. A fluorescence microscope with CCD camera. 2.8 Detection of Programmed Cell Death Using TUNEL Assay

1. FAA fixative: add 1 mL of 38% formaldehyde and 1 mL of glacial acetic acid to 18 mL of 50% ethanol. 2. Vacuum bell or vacuum oven. 3. Oven. 4. Glass scintillation vials. 5. Graded ethanol series of 50% (v/v), 60%, 70%, 80%, and 90% ethanol in distilled water. 6. Eosin Y solution: 0.1% (w/v) Eosin Y in 95% ethanol. 7. Histo-Clear. 8. Histo-Clear dilution series: 25% (v/v), 50% and 75% HistoClear in ethanol. 9. Paraplast X-tra. 10. Heat plate or slide warmer. 11. Microtome. 12. 4% methanol-free formaldehyde solution in 1× PBS: For 1 L of 4% formaldehyde, first add 800 mL of 1× PBS to a glass beaker on a stir plate in a ventilated hood, heat while stirring to approximately 60 °C. Then, add 40 g of paraformaldehyde powder to the heated PBS solution and slowly raise the pH by adding 1 N NaOH dropwise from a pipette until the solution clears. Finally, cool and filter the solution, and adjust the volume of the solution to 1 L with 1× PBS. 13. Xylenes. 14. 96%, 90%, and 80% (v/v) ethanol in distilled water. 15. 10 mM Tris-HCl pH 8.0. 16. Proteinase K solution: dissolve proteinase K (PK) in 10 mM Tris-HCl (pH 8.0) to a final concentration of 20 mg/mL, with 1 mM CaCl2+ and 50% glycerol as stabilizers. 17. Coplin jars. 18. 1 M Tris-HCl stock buffer (pH 7.2): dissolve 121.1 g of Tris in 800 mL deionized water. Adjust the pH of the buffer to 7.2 using 10 N HCl. Then add deionized water to the buffer to obtain a total volume of 1 L.

204

Fang Chang et al.

19. 10× terminal deoxynucleotidyl transferase (TdT) buffer: mix 300 mL of 1 M Tris-HCl stock buffer (pH 7.2) and 700 mL of deionized water to achieve 0.3 M Tris-HCl (pH 7.2). Add sodium cacodylate (C2H6AsNaO2) and DTT (DL-Dithiothreitol, C4H10O2S2) to the 0.3 M Tris-HCl buffer (pH 7.2) to obtain a final concentration of 1.4 M sodium cacodylate and 1 mM DTT, respectively. 20. TdT incubation buffer: mix 41.5 μL of deionized water, 5 μL of 10× TdT buffer, 2 μL of 25 mM CoCl2, 1 μL of 1 mM Bio-16-dUTP, and 0.5 μL of 25 U/μL TdT in order to get a total volume of 50 μL of TdT incubation buffer. 21. 10× phosphate buffered saline (PBS, pH 7.4): dissolve 80 g of NaCl, 2 g of KCl, 14.2 g of Na2HPO4, and 2.7 g of KH2PO4 in 900 mL of distilled water. Adjust the pH of the buffer to 7.4 using 10 N HCl. Then add distilled water to the buffer to obtain a total volume of 1 L. 22. ExtrAvidin-peroxidase (available from, e.g., Sigma-Aldrich): 1: 50 diluted in PBS buffer (pH 7.4), 1% (w/v) Bovine Serum Albumin (BSA), 0.5% (v/v) Tween-20. 23. Terminating buffer: 0.3 M NaCl, 30 mM sodium citrate (C6H5Na3O7). 24. AEC: 3-Amino-9-EthylCarbazole. 25. AEC solution: dissolve 300 mg of AEC in 10 mL of dimethylformamide (DMF) and mix thoroughly to a final concentration of 30 mg/mL. 2.9 Ultrathin Section and Transmission Electron Microscopy (TEM) for Observation of Tapetum and Pollen Morphology

1. Items 1 through 14 of Subheading 2.5 are also required in order to prepare resin-embedded inflorescence tissues. 2. 0.1 M phosphate buffer (pH 7.2): dissolve 80 g of NaCl, 32.3 g of Na2HPO4·12H2O and 4.5 g of NaH2PO4·2H2O in 1 L distilled H2O. 3. 2.5% glutaraldehyde fixative in 0.1 M phosphate buffer: dissolve 10 g of paraformaldehyde in 300 mL of distilled water and add 2 mL of 1 M NaOH (make fresh each time by dissolving 5 pellets of NaOH in approximately 5 mL of distilled water), heat the solution at 65 °C in a water bath until the paraformaldehyde dissolves, then let the solution cool down and dissolve 12.248 g of Na2HPO4·12H2O and 2.464 g of NaH2PO4·2H2O, then add 50 mL of EM-grade glutaraldehyde and make up a total volume of 500 mL with distilled water. 4. 1% (w/v) OsO4 in 0.1 M phosphate buffer. 5. 100% ethanol and graded ethanol series of 50% (v/v), 70%, 80%, 90%, and 95% ethanol in distilled water.

Anther Morphogenesis and Pollen Viability

205

6. Acetone. 7. Propylene oxide. 8. Uranyl acetate: 2% (w/v) in 70% methanol. 9. Reynolds lead citrate: dissolve 0.66 g of Pb(NO3)2 in 15 mL of deionized water; then add 0.88 g of Na3(C6H5O7)·2H2O and shake vigorously for 1 min, cover and set for 30 min; add 10 N NaOH dropwise from a pipette until the solution clears, increase the volume to 25 mL with deionized water. 10. Microtome (see Note 6). 11. TEM microscopy with CCD camera. 2.10 Sample Preparation for Laser Capture Microdissection

1. A laboratory microwave oven (see Note 7). 2. A rotary microtome (see Note 8). 3. Positively (+) charged slides (e.g., Thermo Fisher Scientific). 4. Acetone. 5. Xylene. 6. Paraplast-X. 7. Vacuum bell or vacuum oven. 8. Diethyl pyrocarbonate (DEPC)-treated water.

3

Methods

3.1 Alexander Staining and Photography

This Alexander staining procedure is used to distinguish normal pollen grains from aborted pollen grains by color (see Note 9). 1. Collect Arabidopsis stage 12 flower buds (see Note 10), fix them in a microcentrifuge tube with 1 mL of Carnoy’s fixative for at least 2 h. The fixed samples can be stored in 70% ethanol at 4 °C for up to 1 year. 2. Transfer the flower buds to a glass slide. 3. Remove the fixative thoroughly from the plant material with absorbent paper. 4. Dissect the buds under a dissecting microscope to obtain individual anthers. 5. Place individual anthers into a microtube with an appropriate volume of Alexander staining solution to submerge the anthers completely. 6. Incubate at 25 °C for 30 min. 7. Place the anthers on a slide, cover it with a coverslip, and apply even pressure on the cover-slip to ensure that all anthers converge to one plane. The coverslip can be sealed using nail polish or wax.

206

Fang Chang et al.

Fig. 1 Detection of pollen viability and anther starch accumulation. (a, b) anthers and pollen grains stained with the Alexander staining solution. (a), an Arabidopsis stage-12 flower with six anthers. The area marked with a green box is shown in higher magnification in (b). (c) starch accumulation in rice pollen by I2/KI staining. (d) green florescence from FDA-stained Arabidopsis pollen grains. Red arrowheads indicate aborted pollen grains

8. Visualize and photograph the anthers using a microscope. Viable pollen grains should be stained magenta-red and aborted pollen grains stained blue-green (see Fig. 1a, b). 3.2 Iodine Pollen Starch Detection

1. Collect stage 12–13 flowers (see Note 10). 2. Dissect out anthers and collect in 70% ethanol. 3. Place the anthers on a glass slide with one drop of water. 4. Slice anthers open with forceps until they release pollen grains. 5. Add one or two drops (~5–10 μL) of 1% I2/KI solution. 6. Incubate for 5 min to stain the pollen grains. 7. Cover the pollen with a coverslip, observe and photograph. The normal mature pollen grains containing starch granules stain black; however, the immature pollen grains appear orange to red (see Fig. 1c).

3.3 FDA Staining and Imaging

1. Pollen grains of fully fertile or male-sterile plants are collected from dehiscing anthers. 2. Place the pollen grains into a 1.5 mL centrifuge tube containing 100 μL of FDA-PI staining solution for 5 min at room temperature. 3. Wash the pollen grain once with distilled water. 4. Transfer the pollen grains onto a glass slide with one drop of water by gentle pipetting. 5. Cover the pollen with a coverslip, observe using a fluorescence compound microscope and photograph using a digital camera with a fluorescence filter block for blue excitation and a longpass emission filter to detect the red fluorescence from PI and the green fluorescence from fluorescein. This method stains living cells green and dead cells orange (see Fig. 1d).

Anther Morphogenesis and Pollen Viability

207

Fig. 2 Scanning electron microscopy (SEM) images of Arabidopsis anther and pollen wall. (a) an Arabidopsis stage-12 anther. (b) a pollen grain. (c) the well-organized pollen exine structure. Scale bar = 100 μm, 5 μm, and 1 μm in (a–c) 3.4 Scanning Electron Microscopy for Anther Structure, Anther Dehiscence, and Pollen Wall Structure

1. Collect fresh floral buds and immediately fix them in pure methanol in a 1.5 mL centrifuge tube for 10 min. 2. Decant the liquid and wash the samples once with ethanol. 3. Treat the samples for 5 min in each of the following tertbutanol concentrations: 30%, 50%, 70%, 80%, and 90% (v/v) tert-butanol in ethanol, by transferring the samples gently using a pair of tweezers. Subsequently, dehydrate three times in 100% tert-butanol for 5 min each. Transfer the samples to 4 °C. 4. Place the sample into a 4 °C vacuum freeze dryer, dry completely under low vacuum condition. 5. Remove floral sepals and petals to show the stamens. 6. Place the anthers or pollen grains on specimen stubs, and sputter-coat with gold. 7. Observe under a scanning electron microscope. Figure 2 shows images of the Arabidopsis flower bud, anther, and pollen treated according to this method.

3.5 Anther Anatomy Using Semi-Thin Sections

1. Collect the entire flower inflorescence and quickly place it in a glass vial with FAA fixative. Make sure to use at least three times as much FAA as the volume of the tissue. 2. Keep the container with tissues under vacuum for about 15 min. 3. Release the vacuum slowly and the tissue is likely to sink down. 4. Apply vacuum a second time for another 15 min to improve tissue preservation. 5. Change the fixative solution with fresh fixative and leave it overnight.

208

Fang Chang et al.

6. Dehydrate the samples for 30 min in each of the following ethanol concentrations: 70%, 85%, 95% ethanol. Subsequently, dehydrate in 100% ethanol for 2 h, and in 0.1% (w/v) Eosin Y in 100% ethanol for another 2 h. 7. After dehydration, discard the 100% ethanol and add a 1:1 mixture of 100% ethanol and base liquid Technovit 7100 resin. Mix gently and incubate for 1–2 h at room temperature. 8. Prepare the infiltration solution: add 1 g of Hardener I to 100 mL of base liquid Technovit 7100 and mix. It takes approximately 5 min to dissolve the hardener thoroughly. The prepared solution remains stable for approximately 4 weeks at 4 °C. We usually prepare fresh before use. 9. Transfer the inflorescence to the infiltration solution and incubate for 3 h to overnight at room temperature. 10. Add 1/15 volume of Hardener II with a pipette, mix gently and thoroughly. 11. Add 200 μL of the solution into the Histoform S/Q, then place the infiltrated inflorescence into the Histoform and adjust its position as required. We suggest aligning the vertical axis of the inflorescence to the Histoform if you want to prepare transverse sections. 12. Incubate at 40 °C in an oven for more than 1 day for solidification. Then perform the following sectioning procedure. 13. Adjust a heat plate to 42 °C. 14. Trim the embedded tissue blocks (molds) to remove the embedding material without the tissue, to form a rectangular block with surfaces as close to the tissue as possible. 15. Prepare sections with a rotary microtome. We usually use sections of 0.5 μm in thickness. 16. Add several drops of water on the glass slides and transfer the sections with a fine dissecting needle to the top of the water to remove wrinkles. 17. Label the slides and place them on the 42 °C heat plate until the sections are completely dry. It usually takes 30 min to dry the sections and to affix them onto the glass slides. 18. Transfer the slides into a Coplin jar with 100 mL of Toluidine Blue O staining solution, stain the sections for 1 min. 19. Wash the sections in a Coplin jar with 100 mL of water, repeat the wash step twice. 20. Dry the sections. 21. Observe and photograph. Figure 3 shows images of Arabidopsis anthers treated according to this method.

Anther Morphogenesis and Pollen Viability

209

Fig. 3 The transverse view of Arabidopsis anther development stages 5–7. Stage 5 (a), stage 6 (b), stage 7 (c). Sections were generated and stained using the protocol described in Subheading 3.5. Epidermis (E); Endothecium (En); Middle layer (ML); Tapetum (T); Microsporocytes (Ms). Scale bar = 50 μm 3.6

Callose Staining

1. Collect floral buds, fix them, and then embed them with Technovit 7100 resin following Steps 1–12 of Subheading 3.5. 2. Prepare sections and dry the sections following Steps 13–17 of Subheading 3.5. 3. Stain the sections with 0.05% (w/v) aniline blue staining solution for 5–10 min. 4. Wash three times with 0.067 M phosphate buffer (pH 6.8). 5. Observe the fluorescence under the microscope with a 390–440 nm excitation filter and a 478 nm blocking filter (see Fig. 4a, b).

3.7 Staining of the Pollen Exine and Intine

1. Collect fresh stage-12 floral buds, quickly fix them into FAA solution and then embed them with Technovit 7100 resin following Steps 1–12 of Subheading 3.5. 2. Prepare sections and dry the sections following Steps 13–17 of Subheading 3.5. 3. Prepare fresh de-resins solution and incubate the sections for 5 min in the solution to remove the resin. 4. Wash the sections 1 min in 1% acetic acid-methanol solution. 5. Subsequently, wash the sections three times in 50% methanol buffer for 5 min each. 6. Mix the Calcofluor white solution and DiOC2 solution with a 2:1 (v/v) ratio. 7. Stain the sections with freshly mixed dye for 10 min at 25 °C in darkness. 8. Perform UV microscopy at the upright microscope and observe the fluorescence (see Fig. 4c, d) (see Note 11).

3.8 Detection of Programmed Cell Death Using TUNEL Assay

See Fig. 5 for a protocol overview of this assay. 1. Collect inflorescences with floral buds and fix them in glass vials following Steps 1–5 of Subheading 3.5.

210

Fang Chang et al.

Fig. 4 The callose wall stained with protocols in Subheading 3.6 and the pollen walls (exine and intine) stained with protocols in Subheading 3.7. (a) the callose in two stage-7 anthers. (b) enlarged images show the callose wall surrounding tetrads. The area marked with a white box in (a) is shown in higher magnification in (b). (c–e) stained pollen grains in a stage-12 anther of the WT (c) and bhlh010 bhlh089 double mutants (d–e). Red arrows indicate the orange-red fluorescence from pollen exine, and magenta arrows indicate the blue fluorescence from pollen intine. The staining results clearly showed that the exine and intine if WT pollen are relatively uniform (c); whereas the pollen exine of bhlh010 bhlh089 is uneven and partially missing, and the pollen intine of bhlh010 bhlh089 is relatively thin and discontinuous (d). Some bhlh010 bhlh089 pollen even has no intine (e). The scale bar = 50 μm for (a–b), and 10 μm for (c–e)

2. Keep Paraplast X-tra at 55 °C. 3. Dehydrate sample through 45 min washes in each of the following solutions: 50% ethanol, 60% ethanol, 70% ethanol, 80% ethanol, 90% ethanol, and 95% ethanol with 0.1% (w/v) EosinY. 4. Incubate the floral buds in 95% ethanol with 0.1% Eosin Y for 2 h. 5. Replace the 95% ethanol with 0.1% Eosin Y by Histo-Clear by incubating samples for 30 min in each of the following solutions: 25% Histo-Clear 75% ethanol; 50% Histo-Clear 50% ethanol; 75% Histo-Clear 25% ethanol; and 100% HistoClear. Repeat with fresh Histo-Clear. 6. Add about 20 chips of Paraplast into each glass vial (1/4 volume) and incubate overnight at room temperature.

Anther Morphogenesis and Pollen Viability

211

Fig. 5 Protocol overview for the TUNEL assay

7. Place the vials at 42 °C until the chips melt into the solution. 8. Move the vials to 55 °C and incubate for 12 h. 9. Change the melted Paraplast every 12 h for a total of at least six changes. 10. Turn on the heat plate and adjust to 70 °C. Place pre-labeled molds, a pair of forceps, a needle, and a 5 mL tip in the heat plate to preheat them. 11. Bring the vials of floral buds and quickly transfer the material into the molds. 12. Use the preheated forceps and needle to orient the tissue quickly, then pour more wax to fill up the mold. 13. Keep it undisturbed to allow solidification of the wax. Remove the molds and store the blocks at 4 °C. 14. Adjust the heat plate to 42 °C. 15. Label the slides, put several drops of water onto each one and put them on the heat plate. 16. Trim the molds to remove excess embedding material. Cut as close to the tissue as possible and generate rectangular blocks. 17. Place the mold into a microtome and cut 8–12 μm sections.

212

Fang Chang et al.

18. Put several drops of water on the glass slides and transfer the sections onto the water surface. 19. Incubate for several minutes to smoothen the sections. 20. Carefully remove the water with absorbent paper and avoid air bubbles. 21. Place the slides back on the heat plate and bake them overnight at 42 °C. The sections will adhere to the slides after this step. 22. Fix the inflorescence sections by immersing the slides in 4% methanol-free formaldehyde solution in 1× PBS buffer in a Coplin jar for 15 min at room temperature. 23. Wash the slides by immersing the slides into 100 mL of PBS buffer, and incubate at 70 °C for 10 min. 24. Transfer the slides into fresh xylene for 5 min at room temperature. Repeat once for a total of two xylene washes. 25. Wash the sample by immersing the slides in 96% ethanol for 3 min at room temperature. Repeat once for a total of two 96% ethanol washes. 26. Rehydrate the samples by sequentially immersing the slides through graded ethanol washes: 96%, 90%, 80% ethanol for 3 min and twice each at room temperature. 27. Wash the sample by immersing the slides in deionized water for 3 min at room temperature. 28. Wash the sample by immersing the slides in 10 mM Tris∙HCl (pH 8.0) for 5 min at room temperature. 29. Drain off the excess water from the sections and add 100 μL of the 20 μg/mL proteinase K to each slide so that the sections are covered by solution. Place the slides on a flat surface and incubate for 15 min at room temperature。 30. Rinse the slides in deionized water for 2 min at room temperature. Repeat three times for a total of four washes. 31. Fix the inflorescence sections as described in Step 22. 32. Wash the samples by immersing the slides in deionized water for 2 min at room temperature. Repeat three times for a total of four washes. 33. Remove excess liquid and add 100 μL of fresh prepared TdT incubation buffer to each slide to cover the sections. 34. Put the slides in a dark humidified chamber, incubate the box in 37 °C for 60 min. Do not let the sections to dry out and avoid exposure to light. 35. Wash the slides three times in deionized water in a Coplin jar at room temperature.

Anther Morphogenesis and Pollen Viability

213

36. Terminate the reaction by immersing the slides in the terminating buffer for 15 min at room temperature. 37. Wash the sections by immersing the slides in 1× PBS buffer (pH 7.4) for 5 min at room temperature. 38. Transfer the sections into 1× PBS (pH 7.4) buffer containing 2% (w/v) BSA and incubate for 10 min at room temperature. 39. Wash the sections by immersing the slides in 1× PBS buffer (pH 7.4) for 5 min at room temperature. 40. Incubate the samples in ExtrAvidin-peroxidase solution for 15 min. 41. Incubate the samples in 1× PBS buffer (pH 7.4) containing 2% (w/v) BSA for 10 min. 42. Wash the sections by immersing the slides in 1× PBS buffer (pH 7.4) for 5 min. Repeat three times for a total of four washes. 43. Stain with AEC staining solution for 30 min in 37 °C. 44. Rinse in deionized water three times. 45. Drain off excess water. 46. Observation and photograph under microscope. Red color indicates signals for broken DNA ends. 3.9 Ultrathin Section and Transmission Electron Microscopy (TEM) for Observation of Tapetum and Pollen Morphology

1. Collect floral buds and fix them in 0.1 M phosphate buffer (pH 7.2) with 2.5% glutaraldehyde for 1 h at room temperature under vacuum. 2. Transfer the sample to 4 °C for overnight. 3. Wash three times with 0.1 M phosphate buffer (pH 7.2), 15 min for each time. 4. Incubate on ice for 1–2 h in 1% (w/v) OsO4 in 0.1 M phosphate buffer (pH 7.2). 5. Wash the sample 3 times with 0.1 M phosphate buffer (pH 7.2), 15 min for each time. 6. Dehydrate for 30 min in each of the following ethanol concentrations: 50% ethanol; 70% ethanol; 80% ethanol; 90% ethanol; 95% ethanol; and 100% ethanol. Subsequently, dehydrate twice in 100% acetone, 30 min in each. 7. Stop the dehydration by incubating the sample in propylene oxide for 30 min. 8. Transfer to fresh propylene oxide and incubate for another 30 min. 9. Penetration (according to the variety of epoxy resin, mix strictly according to the formula ratio and stir evenly). The proportion and time of each step are as follows:

214

Fang Chang et al.

Acetone:Epoxy = 3:1, 3 h (room temperature) Acetone:Epoxy = 1:1, 3 h (at room temperature) Acetone:Epoxy = 1:3, 3 h—overnight (room temperature) Epoxy 100%, 24–72 h (at room temperature) In each of the preceding steps, samples are shaken on a rotary mixer to accelerate penetration. 10. Add 1.5–2% catalyst to 100% epoxy resin, mix and stir for 20–30 min. 11. Inject the fresh mixed epoxy resin into the embedded mold, and then place the sample in the resin of the embedded mold. The sample must be placed at the bottom to prevent movement during the polymerization process. 12. Polymerization: prepolymerize at 45 °C for 12 h, then polymerize at 60 °C for 24 h. 13. Prepare sections using an appropriate microtome (see Note 6). We usually use sections of 50–70 nm in thickness. 14. Collect sections. 15. Stain the sections first with uranyl acetate for 15 min, then with Reynolds lead citrate for 6–8 min. 16. Observe the ultrathin sections (50–70 nm thick) with TEM microscopy. Figure 6 shows images from Arabidopsis anthers treated according to this method. 3.10 Sample Preparation for Laser Capture Microdissection

See Fig. 7 for a flow diagram of this assay. 1. Collect inflorescences with floral buds and immediately place them in pure ice-cold acetone. 2. Heat for 15 min in a microwave oven (microwave oven settings: 400 W, 37 °C). 3. Vacuum (400 mm Hg) at room temperature for 30 min to help the infiltration.

Fig. 6 Ultrathin section and transmission electron microscopy of Arabidopsis tapetum and pollen grains. (a–c) TEM images for an anther lobe (a), a pollen grain (b), and a part of one tapetal cell (c) from (a). The areas marked with white boxes are shown in higher magnification in (b) and (c)

Anther Morphogenesis and Pollen Viability

215

Fig. 7 Flow diagram of sample preparation for LCM

4. Transfer floral buds to fresh acetone and microwave the sample for 15 min (400 W, 37 °C). Repeat twice. 5. Transfer floral buds to an acetone/xylene (1:1) solution. Microwave for 1 min 15 s (microwave settings: 500 W, 60 °C). 6. Transfer floral buds to pure xylene. Microwave for 1 min 15 s (500 W, 60 °C). 7. Replace the xylene with fresh Paraplast-X, microwave for 10 min (400 W, 70 °C). 8. Replace with fresh Paraplast-X and microwave for 30 min (400 W, 70 °C). Repeat this step four times. 9. Pour Paraplast with the tissue into a metal weighing dish on the hot side of the warming plate. 10. Scoop out the tissues with a weighing spatula, place them into the assembled base mold/ embedding ring combination, and orient tissue for sectioning on a rotary microtome later. 11. Cool down to RT and then transfer onto ice for easy un-molding. 12. Store the prepared tissue blocks in plastic bags at 4 °C. 13. Trim the Paraplast blocks into a narrow trapezoidal shape with parallel horizontal (5–10 μm thick) cuts at top and bottom, and slanted cuts on the sides. 14. Section using a rotary microtome. 15. Place ribbons of sections gently onto the surface of DEPC treated water on positively (+) charged slides, which were prewarmed at 40 °C on the heat plate; incubate for approximately 5 min or more.

216

Fang Chang et al.

16. Remove the water by tipping the slide onto absorptive paper towels while holding one end of the ribbon with a fine paintbrush. Wick off the residual water with tissue paper. 17. Store the dry slides at 4 °C until PALM procedure. 18. Warm up the slides to room temperature. 19. Deparaffinize the sections in two changes of xylene (5 min each xylene change). 20. Wash the sections three times with DEPC water. 21. Place on the microscope stage to mark tissue regions to be collected. 22. Catapult tissues into inverted tube-caps (500 μL microcentrifuge tubes), which were filled with about 50 μL of RNA extraction buffer (e.g., XB from the PicoPure RNA Isolation kit). After collecting a sufficient amount of tissue, extract RNA according to an RNA extraction protocol.

4

Notes 1. The final stain solution should be prepared by adding the constituents in the order shown. Store solution in the dark. 2. Keep the solution in a lightly stopped brown glass bottle. 3. Some have reported to use 0.01% (w/v) Toluidine Blue O, but in our experiments 0.0025–0.005% (w/v) works better, especially for early flower tissues. The prepared Toluidine Blue O staining solution should be stored in dark. 4. Store solution at 4 °C in darkness. 5. Before each experiment, the storage liquid should be thoroughly vortexed and mixed. 6. For example, a Ultracut E, Leica-Reichert microtome with a Histo Jumbo Diamond knife (Ultra 458, Diatome). 7. For instance, microwave BP-111RS (Microwave Research & Application, Inc, Laurel MD, USA). 8. For example, rotary microtome RM 2135 (Leica Microsystems Ltd). 9. This method is used to analyze whether the pollen is well developed or not, but not for the detection of the real time effects caused by a certain treatment to pollen vitality. 10. Arabidopsis flower development stages as defined in [30]. 11. We suggest taking the photos as soon as possible, because prolonged UV light irradiation usually causes fluorescence quenching in the intine part of pollen wall.

Anther Morphogenesis and Pollen Viability

217

References 1. Ma H (2005) Molecular genetic analyses of microsporogenesis and microgametogenesis in flowering plants. Annu Rev Plant Biol 56:393– 434 2. Singh MB, Bhalla P (2007) Control of male germ-cell development in flowering plants. BioEssays 29:1124–1132 3. Alexander MP (1969) Differential staining of aborted and nonaborted pollen. Stain Technol 44:117–122 4. Jones KH, Senft JA (1985) An improved method to determine cell viability by simultaneous staining with fluorescein diacetatepropidium iodide. J Histochem Cytochem 33: 77–79 5. Smith DJ, Oliver CE, Caton JS, Anderson RC (2005) Effect of sodium [36Cl]chlorate dose on total radioactive residues and residues of parent chlorate in beef cattle. J Agric Food Chem 53:7352–7360 6. Zhang H, Zdolsek JM, Brunk UT (1991) Effects of alloxan and reducing agents on macrophages in culture. APMIS 99:1038– 1048 7. Zhu QH, Ramm K, Shivakkumar R et al (2004) The ANTHER INDEHISCENCE1 gene encoding a single MYB domain protein is involved in anther development in rice. Plant Physiol 135:1514–1525 8. Mamun EA, Alfred S, Cantrill LC et al (2006) Effects of chilling on male gametophyte development in rice. Cell Biol Int 30:583–591 9. Pressman E, Peet MM, Pharr DM (2002) The effect of heat stress on tomato pollen characteristics is associated with changes in carbohydrate concentration in the developing anthers. Ann Bot 90:631–636 10. Rhee SY, Somerville CR (1998) Tetrad pollen formation in quartet mutants of Arabidopsis thaliana is associated with persistence of pectic polysaccharides of the pollen mother cell wall. Plant J 15:79–88 11. Worrall D, Hird DL, Hodge R et al (1992) Premature dissolution of the microsporocyte callose wall causes male sterility in transgenic tobacco. Plant Cell 4:759–771 12. Yeung EC, Oinam GS, Yeung SS, Harry I (2011) Anther, pollen and tapetum development in safflower, Carthamus tinctorius L. Sex Plant Reprod 24:307–317 13. Dong X, Hong Z, Sivaramakrishnan M et al (2005) Callose synthase (CalS5) is required for exine formation during microgametogenesis and for pollen viability in Arabidopsis. Plant J 42:315–328

14. Zhang ZB, Zhu J, Gao JF et al (2007) Transcription factor AtMYB103 is required for anther development by regulating tapetum development, callose dissolution and exine formation in Arabidopsis. Plant J 52:528–538 15. Li H, Yuan Z, Vizcay-Barrena G et al (2011) PERSISTENT TAPETAL CELL1 encodes a PHD-finger protein that is required for tapetal cell death and pollen development in rice. Plant Physiol 156:615–630 16. Li X, Gao X, Wei Y et al (2011) Rice APOPTOSIS INHIBITOR5 coupled with two DEADbox adenosine 5′-triphosphate-dependent RNA helicases regulates tapetum degeneration. Plant Cell 23:1416–1434 17. Phan HA, Iacuone S, Li SF, Parish RW (2011) The MYB80 transcription factor is required for pollen development and the regulation of tapetal programmed cell death in Arabidopsis thaliana. Plant Cell 23:2209–2224 18. Rogers HJ (2006) Programmed cell death in floral organs: how and why do flowers die? Ann Bot 97:309–315 19. Varnier AL, Mazeyrat-Gourbeyre F, Sangwan RS, Clement C (2005) Programmed cell death progressively models the development of anther sporophytic tissues from the tapetum and is triggered in pollen grains during maturation. J Struct Biol 152:118–128 20. Zhang C, Guinel FC, Moffatt BA (2002) A comparative ultrastructural study of pollen development in Arabidopsis thaliana ecotype Columbia and male-sterile mutant apt1-3. Protoplasma 219:59–71 21. Shen CX, Zhang QF, Li J et al (2010) Induction of programmed cell death in Arabidopsis and rice by single-wall carbon nanotubes. Am J Bot 97:1602–1609 22. Song L, Zhou Z, Tang S et al (2016) Ectopic expression of BnaC.CP20.1 results in premature Tapetal programmed cell death in Arabidopsis. Plant Cell Physiol 57:1972–1984 23. Vizcay-Barrena G, Wilson ZA (2006) Altered tapetal PCD and pollen wall development in the Arabidopsis ms1 mutant. J Exp Bot 57: 2709–2717 24. Barbazuk WB, Emrich SJ, Chen HD et al (2007) SNP discovery via 454 transcriptome sequencing. Plant J 51:910–918 25. Murphy SJ, Cheville JC, Zarei S et al (2012) Mate pair sequencing of whole-genome-amplified DNA following laser capture microdissection of prostate cancer. DNA Res 19:395–406 26. Ohtsu K, Takahashi H, Schnable PS, Nakazono M (2007) Cell type-specific gene expression

218

Fang Chang et al.

profiling in plants by using a combination of laser microdissection and high-throughput technologies. Plant Cell Physiol 48:3–7 27. Takahashi H, Kamakura H, Sato Y et al (2010) A method for obtaining high quality RNA from paraffin sections of plant tissues by laser microdissection. J Plant Res 123:807–813 28. Taylor TB, Nambiar PR, Raja R et al (2004) Microgenomics: identification of new

expression profiles via small and single-cell sample analyses. Cytometry A 59:254–261 29. Zheng Z, Andersson AF, Ye W et al (2011) A method for metagenomics of Helicobacter pylori from archived formalin-fixed gastric biopsies permitting longitudinal studies of carcinogenic risk. PLoS One 6:e26442 30. Smyth DR, Bowman JL, Meyerowitz EM (1990) Early flower development in Arabidopsis. Plant Cell 2(8):755–767

Chapter 9 Isolation of Meiocytes and Cytological Analyses of Male Meiotic Chromosomes in Soybean, Lettuce, and Maize Cong Wang, Xiang Li, Jiyue Huang, Hong Ma, Chung-Ju Rachel Wang, and Yingxiang Wang Abstract Meiosis is a specialized cell division that halves the number of chromosomes following a single round of DNA replication, thus leading to the generation of haploid gametes. It is essential for sexual reproduction in eukaryotes. Over the past several decades, with the well-developed molecular and cytogenetic methods, there have been great advances in understanding meiosis in plants such as Arabidopsis thaliana and maize, providing excellent references to study meiosis in other plants. A chapter in the previous edition described molecular cytological methods for studying Arabidopsis meiosis in detail. In this chapter, we focus on methods for studying meiosis in soybean (Glycine max), lettuce (Lactuca sativa), and maize (Zea mays). Moreover, we include the method that was recently developed for examination of epigenetic modifications, such as DNA methylation and histone modifications on meiotic chromosomes in plants. Key words Soybean, Lettuce, Maize, Meiosis, Immunofluorescence

1

Introduction Meiosis is a cell division essential for sexual reproduction and involves homologous chromosome interactions such as paring, synapsis, recombination, and segregation. Recombination results in genetic information exchange between homologs, with great impact on biodiversity, and is also the theoretical basis for crop breeding [1]. Over the last two decades, molecular genetic studies have achieved great progress in understanding plant meiosis, mainly using model plants such as Arabidopsis, rice, and maize [1–4]. With rapidly advancing technologies for whole-genome sequencing, research on natural variations and non-model plants has mushroomed dramatically, and increased the demand for observation and analyses of meiotic chromosome behavior in multiple species. Soybean (Glycine max) belongs to the legume family (Fabaceae) and is one of the most important oil crops and a major source

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_9, © Springer Science+Business Media, LLC, part of Springer Nature 2023

219

220

Cong Wang et al.

of plant proteins. Lettuce (Lactuca sativa), a member of the composite (Asteraceae) family, is an important vegetable crop. Together with Arabidopsis, they are members of the eudicot group (with about two thirds of all flowering plants). Maize (Zea mays, 2n = 20) is a member of the Poaceae family (grasses; within the monocot group) and is an important crop globally. The maize meiotic chromosomes are large and well defined cytologically, and a single maize tassel harbors thousands of anthers that develop gradually corresponding to their positions on a tassel, such that sequential stages of meiosis are easily observed [5, 6]. These advantages have made maize one of the premier model organisms for studying meiosis using a suite of cytogenetic and molecular tools. Thus, maize offers an excellent model complementary to studies in Arabidopsis and other eudicots [3]. A crucial aspect of meiotic prophase I is homologous chromosome association, which ensures the accurate segregation of chromosomes during meiosis I. Thus, examination of chromosome behavior and structures is often required. Electron microscopy (EM) has been used to visualize chromosomal structures related to synapsis and homologous recombination, including synaptonemal complexes (SCs) and recombination nodules (RNs) [7], especially in Arabidopsis and maize [8–13]. To examine chromosome behavior, FISH (fluorescence in situ hybridization) is a powerful tool using specific DNA sequences as probes to detect corresponding chromosomal regions and analyze chromosome interactions, including pairing and synapsis [14]. Arabidopsis and maize provide attractive systems for such studies; in particular, a maize meiotic chromosome spreading method commonly used for FISH is included [15]. The advances of high throughput DNA sequencing technologies have made it possible to sequence tissue- or cell-specific transcriptomes in many species [16]. However, plant male or female meiocytes are enclosed within reproductive organs and make up a small fraction of the cells in these organs [17, 18], thus it is difficult to obtain many pure meiocytes. Consequently, the ability to harvest meiocytes for transcriptional profiling can greatly facilitate the identification of meiotic genes. Here, we describe a relatively simple method to collect male meiocytes from suitably sized flower buds of soybean or capitula of lettuce, respectively, using a micromanipulator system [19], for high-throughput sequencing or cytological analysis of chromosome behavior. Epigenetic regulation of meiosis has gradually become a hotspot [1, 20]. Hence, identifying chromatin modifications can expand our understanding of their roles in meiosis. Traditional immunofluorescence method utilizing paraformaldehyde fixation commonly produces squashed chromosomes with residual cytoplasm or cytoskeleton proteins, thus causing relatively low resolution of images observed under microscope [14, 21]. In contrast,

Analyses of Meiosis in Plants

221

immunofluorescence method using Carnoy’s fixation produces well spread chromosome for imaging [22]. However, this method is unable to detect the signals of antibodies binding to modified chromatin, especially for methylated DNA (5-mC). We described an efficient method to observe the chromatin modification in Arabidopsis meiocytes. Since epigenetic modifications such as DNA methylation and histone modification are conserved among eudicots, we believe that the developed method can be used for other eudicot species.

2 2.1

Materials Plants

1. Plants of the Glycine max variety Huaxia3 [23] are grown in illuminated incubators, under 12 h day and 12 h night at 25 °C with 75% relative humidity. 2. Lactuca sativa (Japanese Butterhead) plants are grown in illuminated incubators, under 16 h light at 25 °C and 8 h dark at 17 °C with 75% relative humidity. 3. Arabidopsis thaliana plants are grown in greenhouse under 16 h light and 8 h dark at 22 °C with 75% relative humidity. 4. Seeds of maize inbred lines can be obtained from the North Central Regional Plant Introduction Station (http:// maizecoop.cropsci.uiuc.edu/ncrpis.php) or the International Maize and Wheat Improvement Center (http://maizecoop. cropsci.uiuc.edu/cimmyt.php). Mutants or other genetic stocks can be requested from the Stock Center of Maize Genetics Cooperation (https://maizecoop.cropsci.uiuc.edu/ request/). After germination, seedlings are transplanted into 10 L pots and grown under 16 h light at 28 °C and 8 h dark at 22 °C with supplemented lighting of at least 500 μE/m2/s.

2.2 Isolation of Male Meiocytes from Soybean and Lettuce

1. Dulbecco’s phosphate-buffered saline (DPBS): for a final volume of 1 L, dissolve 200 mg of potassium chloride (KCl), 200 mg of potassium phosphate monobasic (KH2PO4), 2.16 g of sodium phosphate dibasic (Na2HPO4–7H2O), and 8 g of sodium chloride (NaCl) in RNase-free water (see Note 1) and autoclave. 2. Collecting buffer: dilute DPBS 1:1 with RNase-free water, and add RNase Inhibitor to a final diluted concentration of 1 U/μL (see Note 1). 3. Syringes (BD Tuberculin 1 mL) and needles (BD 27 G × 1=2). 4. Watchmaker forceps (sharp tip). 5. Microscope slides (25 × 75 mm, 1.0 mm thick) and two-chamber depression slides.

222

Cong Wang et al.

6. RNase-free 2 mL microcentrifuge tubes. 7. 2 mm diameter zirconium beads. 8. Dissection microscope (Nikon or similar). 9. Inverted microscope (e.g., Zeiss). 10. Micromanipulator system: consists of an inverted microscope and a micro-manipulation platform. The micro-manipulation platform contains a capillary tube (glass, 0.86 mm inner diameter, 10 cm length; e.g., Sutter), a glass capillary pipette puller (two-step; e.g., Narishige), and a manipulator (three-axis coarse and fine micromanipulator; e.g., Narishige). 2.3 Chromosome Spread and Immunostaining in Arabidopsis, Soybean, and Lettuce

1. Sharp needles and watchmaker forceps. 2. Microscope slides. 3. 20 × 20, 24 × 24, and 24 × 32 mm coverslips. 4. Glass rod. 5. Heat-block. 6. Vacuum bell. 7. Small glass dish. 8. Razor blade. 9. Slide moisture chamber. 10. Carnoy’s fixative solution: 3 volumes of 100% ethanol, 1 volume of glacial acetic acid. 11. Paraformaldehyde fixative solution: 2% paraformaldehyde, 0.1% (v/v) Triton X-100 in 1 × PBS buffer pH 7.4. 12. 10× PBS: prepare 1 L of 10× PBS by dissolving 80 g of NaCl, 2 g of KCl, 14.2 g of Na2HPO4, and 2.7 g of KH2PO4 in 900 mL of deionized water. Use 10 N HCl to adjust the pH to 7.4 and adjust final volume to 1 L with deionized water. 13. 10 mM citrate buffer (pH 4.5): prepare 1.351 g of sodium citrate dihydrate and 1.038 g of citric acid in a suitable container. Add 800 mL of distilled water. Adjust solution to final desired pH using HCl or NaOH and add deionized water until the volume is 1 L. 14. 10 mM citrate buffer (pH 6.0): prepare 2.427 g of sodium citrate dihydrate and 335.8 mg of citric acid in a suitable container. Add 800 mL of distilled water. Adjust solution to final desired pH using HCl or NaOH and add distilled water until the volume is 1 L. 15. Digestion mixture for immunolocalization: 5% (w/v) cytohelicase, 3% (w/v) cellulase, and 3% (w/v) macerozyme in 10 mM citrate buffer (pH 4.5) (see Note 2).

Analyses of Meiosis in Plants

223

16. Digestion mixture for chromosome spread: dilute 1 volume of the digestion mixture for immunolocalization with 1 volume of 10 mM citrate buffer (pH 4.5). 17. 60% acetic acid. 18. Washing buffer I: 1 × PBS buffer, 1% (v/v) Triton X-100. 19. Washing buffer II: 1 × PBS buffer, 0.1% (v/v) Tween 20. 20. Blocking buffer (antibody diluent): Non-sterile goat serum (e.g., Sigma) (see Note 3). 21. Primary and secondary antibodies (see Note 8). 22. Labeled centromere probes: Synthetic oligonucleotides encompassing three fragments of centromere sequence conjugated with CY5. They are mixed for FISH experiments. Cen180_oligo1: CY5- GGTGTAGCCAAAGTCCRTAT GAGTCTTTGK; Cen180_oligo2: CY5-TCTTATACTCAAT CATACACATGACATCW ; Cen180_oligo3: CY5-AGTCATATTYGACTCCAA AACACTAACC. 23. DAPI: VECTASHIELD Antifade Mounting Medium with DAPI (e.g., Vector). 24. Fluorescent Microscopes (e.g., Zeiss). 2.4 Light Microscopy of Maize Meiosis

1. Fine needles. 2. Microscope slides and coverslips. 3. Iron nail. 4. Carnoy’s fixative solution: Please see Subheading 2.3. 5. 70% ethanol. 6. 2% Acetocarmine: dissolve 10 g of carmine (e.g., Fisher) in 500 mL of 45% glacial acetic acid. Boil in a reflux condenser for 4–5 h, and filter into a dark container after cooling. 7. An alcohol burner. 8. Microscope.

2.5 Fluorescent In Situ Hybridization (FISH) of Maize Chromosomes

1. Needles.

2.5.1 Maize Meiotic Chromosome Spreading

5. Razor blade.

2. Microscope slides. 3. Coverslips. 4. Petri dishes. 6. 10× enzyme buffer: Mix 40 mL of 100 mM citric acid and 60 mL of 100 mM sodium citrate, adjust to pH 4.8 using HCl. Store at 4 °C. Dilute 10× in water for use. 7. Digestion enzyme cocktail: 2% (w/v) cellulase, 1% (w/v) macerozyme, and 1% (w/v) cytohelicase in 1× enzyme buffer (see Note 2). Store in aliquots at -20 °C.

224

Cong Wang et al.

8. 45% acetic acid. 9. 50%, 75%, 95% ethanol. 10. DAPI: VECTASHIELD Antifade Mounting Medium with DAPI (e.g., Vector). 2.5.2

Labeled Probes

1. 5s rDNA: synthetic oligonucleotide TCAGAACTCCGAAGTTAAGCGTGCTTGGGCGAGAGTAGTAC modified with a fluorescence dye (e.g., CY3 or CY5) at the 5′ terminus. 2. 180 bp knob-1: GAAGGCTAACACCTACGGATTTTT GACC and 180 bp knob-2: AAGAAATGGTCTCCACCAGAAATCCAAA modified with a fluorescence dye at the 5′ terminus, respectively. Use both oligonucleotides for labeling heterochromatic knobs. 3. CentC repeat-1: CCCAATCCACTACTTTAGGTCCAAAAC GCACT and repeat-2: CATGTTTGGGGTGGTTTCGCGC AATTTCGTT modified with a fluorescence dye at the 5′ terminus, respectively. Use both oligonucleotides for labeling centromeres. 4. TR-1 repeat-1: AAATAAAATAAAGACTATGGAAAATTTAGCC and repeat-2: AAATTGCGTGAGTGAACTGTCCAAACAT modified with a fluorescence dye at the 5′ terminus, respectively.

2.5.3 Pretreatment of Chromosome Spreads

1. Slide moisture chamber. 2. 20× SSC (saline sodium citrate): Prepare 1 L of 20× SSC buffer by dissolving 175.3 g of NaCl and 88.2 g of sodium citrate (C6H5O7Na3.2H2O) in 900 mL of deionized water. Use 10 N HCl to adjust the pH to 7.0, and adjust final volume to 1 L. 3. 10 mM HCl. 4. RNase stock: 10 mg/mL DNase-free ribonuclease A in 10 mM Tris-HCl (pH 8.0). Store in aliquots at -20 °C. Dilute 100× prior to use. 5. Pepsin stock: 100 μg/mL in 10 mM HCl. Store in aliquots at 20 °C. Dilute 100× prior to use. 6. Formaldehyde fixative: 8% or 16% paraformaldehyde aqueous solution in ampoules. Prepare 4% paraformaldehyde in 2× SSC prior to use.

2.5.4

Hybridization

1. Hybridization cocktail (final concentration): 50% formamide, 2× SSC, 10% dextran sulfate, 200 μg/mL sheared salmon sperm DNA, 0.2% SDS, 1–5 ng/μL labelled DNA probe. 2. Denaturation solution: 70% formamide in 2× SSC. 3. Ethanol: 75%, 95%, and 100% ethanol. Prepare in Coplin jars and store at -20 °C.

Analyses of Meiosis in Plants 2.5.5 Stringent Wash and Mounting

225

1. Coplin jars. 2. Stringent washing solution: 20% formamide in 2× SSC. 3. 2× SSC: dilute from 20× SSC stock. 4. 1× PBS: dilute from 10× PBS stock.

3

Methods

3.1 Isolation of Meiocytes from Soybean and Lettuce

1. Soybean and lettuce plants are grown in the illuminated incubator for over 55 (Fig. 1a) and 75 days (Fig. 2a), respectively (see Note 4). Then the inflorescences are harvested from the apices of stalks and are placed onto an ethanol-cleaned slide with two chambers. 2. Collect the floral buds with innermost meiocytes undergoing meiosis on the slide under a dissection microscope, and use two needles to gently separate the petals in order to cut the anthers. Stage 6–8 flower buds of soybean are 3–4 mm in length (Fig. 1b), and the stage 4–5 capitula of lettuce are 1.5–2 mm in diameter (Fig. 2b) (see Note 4). 3. Using the needles, move the dissected anthers to a chamber of the slide with 10 μL of collecting buffer.

Fig. 1 Isolation of soybean male meiocytes. (a) The inflorescences of soybean. (b) Different stages of soybean flower development. The pollen mother cells undergo meiosis at stage 6–8 of flower development. (c, d) Male meiocyte clusters (masses pointed by green arrows) isolated from the stage 6–8 flowers. Scale bars: (a) 1 cm; (b) 2 mm; (c, d) 50 μm. (Figure adapted from Huang et al. [26] with permission from publisher)

226

Cong Wang et al.

Fig. 2 Isolation of lettuce male meiocytes. (a) The capitula of lettuce. (b) Different stages of lettuce capitulum development. The pollen mother cells undergo meiosis at stage 4–5 of capitulum development. (c, d) Male meiocyte clusters (masses pointed by green arrows) isolated from the stage 4–5 capitula. Scale bars: (a) 1 cm; (b) 2 mm; (c, d) 50 μm

4. Collect this way a total of 30–50 anthers in a chamber and slightly extrude the anthers with a pair of sharp forceps to release masses of clustered meiocytes. 5. Add 50 μL of collecting buffer into the chambers and rinse the forceps with buffer. 6. Masses of clustered meiocyte are collected in the microchamber (a capillary tube, see Subheading 2.2) by a micromanipulator system [21]. 7. Transfer the meiocytes to a RNase free 2.0 mL tube containing 2 mm beads (Zymo Research), and the tube with collected meiocyte masses (Figs. 1c and 2c) is quickly frozen in liquid nitrogen and stored in the -80 °C freezer (see Note 5). Normally, 200–300 meiocyte masses are sufficient to extract 1–2 μg RNA. 3.2 Chromosome Spread for Meiotic Chromosomes of Soybean and Lettuce

1. Collect inflorescences from healthy plants in 2 mL tubes with 1 mL of fresh Carnoy’s fixative solution. 2. Fix the collected inflorescences for 4 h at room temperature (see Note 6).

Analyses of Meiosis in Plants

227

3. Rinse the fixed inflorescences with citrate buffer (pH 4.5) three times for 5 min each. 4. Add enough digestion mixture to immerse the inflorescences and digest them for 15 min (Arabidopsis), 20 min (lettuce), and 22 min (soybean) at 37 °C (see Note 7). 5. Discard the digestion mixture and rinse the digested inflorescences with 10 mM citrate buffer (pH 6.0). Alternatively, the treated inflorescences can be stored at 4 °C for several days. 6. Place the inflorescences on one slide and add a drop (5–10 μL) of ddH2O. 7. Dissect floral buds by using needles to collect anthers, while maintaining moisture of samples during the entire process. 8. Collect about 40–50 anthers undergoing meiosis, and crush them thoroughly using forceps. 9. Transfer the slide to a heat block at 45 °C for 20–30 s, add 30 μL of 60% acetic acid, and let it stand on the heat block for 1 min. 10. Add two drops (~20 μL) of precooled Carnoy’s fixative (-20 ° C) to the center of the droplet on the slide to spread the meiotic chromosomes on the heat block. 11. Observe under microscope. Chromosome behavior at each meiotic stage for soybean and lettuce is shown in Figs. 3 and 4, respectively. 3.3 Immunostaining of Meiotic Proteins in Lettuce

1. Collect the capitula (around 1.5–2 mm) in a 2 mL tube with 1 mL paraformaldehyde fixative solution. 2. Put the tube in an ice box with the lid off and apply vacuum three times for 10 min each with the pressure of 0.01 MPa. 3. Rinse the fixed capitula with citrate buffer (pH 4.5) three times for 5 min each. 4. After fixation, digest the capitula using digestion mixture for 30 min at 37 °C (see Note 7). 5. Rinse the digested capitula with ddH2O three times and keep them on ice (see Note 8). 6. Immerse one capitulum under water in a small glass dish and, using sharp needles, dissect the florets to collect anthers under a dissection microscope. 7. Use a dropper pipette to transfer the anthers from the glass dish onto a slide. 8. Compress the anthers by pinching with a pair of sharp forceps to release meiocytes, then cover with a 20 × 20 mm coverslip. 9. Squash chromosomes by knocking the coverslip on the slide gently using a glass rod.

228

Cong Wang et al.

Fig. 3 Chromosome morphology of soybean male meiosis by chromosome spread. (a) Leptotene. (b) Zygotene. (c) Pachytene. (d) Diplotene. (e) Diakinesis. (f) Metaphase I. (g) Anaphase I. (h) Telophase I. (i) Prophase II. (j) Metaphase II. (k) Anaphase II. (l) Telophase II. Scale bars: 5 μm. (Figure adapted from Huang et al. [26] with permission from publisher)

10. Immerse the slide in liquid nitrogen for at least 1 min to freeze the squashed chromosomes. 11. Take the slide out from liquid nitrogen and quickly remove the coverslip from the slide with a razor blade. Leave the slide with chromosomes on the bench to dry. The slide can be used for immunostaining right away or stored in a sealed plastic case at -80 °C for up to 3 months. 12. Immerse the slide with washing buffer I for 60 min in a Coplin jar at room temperature. 13. Block the slide with goat serum for 30 min at 37 °C. 14. Incubate the slide with the diluted primary antibody in a moist chamber for 12–16 h at 4 °C (see Note 9). 15. Rinse the slide with washing buffer II three times for 15 min each and block it with goat serum for 30 min. 16. Incubate the slide with the diluted secondary antibody for 60 min at 37 °C in dark (see Note 9). 17. Rinse the slide with washing buffer II three times for 15 min each. 18. Add 8 μL of DAPI solution per slide and cover with a 24 × 24 mm coverslip for imaging using a fluorescence microscope (Fig. 5).

Analyses of Meiosis in Plants

229

Fig. 4 Chromosome morphology of lettuce male meiosis by chromosome spread. (a) Leptotene. (b) Zygotene. (c) Pachytene. (d) Diplotene. (e) Diakinesis. (f) Metaphase I. (g) Anaphase I. (h) Telophase I. (i) Prophase II. (j) Metaphase II. (k) Anaphase II. (l) Telophase II. Scale bars: 5 μm

Fig. 5 Immunolocalization of LsZYP1 in lettuce male meiosis. DAPI (blue), LsZYP1 (red), and merged signals are shown at zygotene (a–c) and pachytene (d–f) of lettuce meiosis. LsZYP1 foci start to appear on synapsed regions of homologs at zygotene and become continuous along chromosomal axes at pachytene. Scale bars: 5 μm

3.4 Immunolocalization of DNA Methylation and Histone Modifications in Arabidopsis Meiocytes

1. Fix inflorescences using paraformaldehyde fixative solution and apply vacuum twice for 10 min each with the pressure of 0.01 MPa at 4 °C. 2. Collect over 50 masses of clustered Arabidopsis meiocytes from anthers near stage 5 or 6 as described in Subheading 3.1, steps 2–6. Transfer these meiocytes to chambers of a new depression

230

Cong Wang et al.

slide with 20 μL of digestion mixture for chromosome spread and incubate for 15 min at room temperature. 3. Carefully collect masses of digested meiocytes using a micromanipulator system and transfer them to a new depression slide with 50 μL of ddH2O; wait for 5 min to allow the solution to equilibrate. 4. Re-collect the meiocytes and transfer them to a slide. Place the slide onto a heat block at 45 °C for 20–30 s, add 5 μL of 60% acetic acid, and gently shake for 1 min on the heat block. 5. Cover with a 20 × 20 mm coverslip. Squash the chromosomes by knocking the coverslip gently using a glass rod to fix the chromosomes on the slide. 6. Dry the slides at 45 °C for 3–5 min and immerse the slides in washing buffer I for 10 min at room temperature. The following steps are modified from a method described previously [22]. 7. Transfer them into a Coplin jar with 10 mM citrate buffer (pH 6.0) and microwave for 45 s at 700–900 W. 8. Transfer the slides immediately to the precooled 10 mM citrate buffer (pH 6.0) in another Coplin jar for 2 min at 4 °C (see Note 10), then to washing buffer II for 3 min. 9. Transfer the slides to a moist chamber, add 50 μL of blocking buffer per slide, cover each slide with a 24 × 32 mm coverslips and block for 30 min at room temperature. 10. Discard the blocking buffer on the slides and dilute the primary antibody with blocking buffer to a suitable concentration. 11. Add 25–30 μL to each slide covered with 24 × 32 mm coverslips and incubate for 24–28 h in the moist chamber (see Note 11). 12. Wash the slides with washing buffer II three times for 10 min each at room temperature, then add 50 μL of blocking buffer onto the slides and cover with a 24 × 32 mm coverslip; incubate the slides for 30 min at 37 °C. 13. Discard blocking buffer, and add 25–30 μL of working solution of secondary antibodies (1:200–1:1000) to each slide covered with 24 × 32 mm coverslips and incubate for 2 h in the moist chamber (see Note 12). 14. Wash the slides with washing buffer II three times for 10 min each at room temperature. 15. Add 8 μL of DAPI solution per slide and cover with a 24 × 24 mm coverslip. 16. Observe the slides using a fluorescence microscope (Figs. 6 and 7).

Analyses of Meiosis in Plants

231

Fig. 6 Immunolocalization of heterochromatic markers (5-mC and H3.1) in meiosis of Arabidopsis. Immunostaining with 5-mC and H3.1 at pachytene (a– e) and metaphase I (f–j). The 5-mC signals show co-localization with pericentromeric regions at pachytene and metaphase I. H3.1 signals are located on both centromeric and pericentromeric regions at pachytene and tend to overlap with highly condensed chromosomes at metaphase I. Yellow arrow points to the NORs (nucleolar organizer regions). Scale bars: 5 μm

Fig. 7 Immunolocalization of 5-mC with centromere FISH in meiosis of Arabidopsis. DAPI (blue), 5-mC (green), centromere (red), and merged signals are shown at pachytene (a–e) and metaphase I (f–j). Centromeric signals pointed by red arrows appear to be separated from the 5-mC signals. Yellow arrow points to the NORs. Scale bars: 5 μm

3.5 Isolation and Observation of Maize Meiosis by Light Microscopy

1. Male tassels containing meiotic anthers are confined within the internal stalk of plants, and usually can be found during V7– V10 stages (see Note 13). 2. Immature tassels or a tassel branch are collected and fixed in freshly mixed Carnoy’s fixative solution at room temperature for 8–12 h (see Note 14). 3. Replace Carnoy’s fixative with 70% ethanol and store in -20 °C for cytological observation or FISH experiments. 4. Dissect anthers with appropriate sizes using dissecting needles (a pair of fine needles with a bent tip) (see Note 15).

232

Cong Wang et al.

5. Place one anther from a single floret on microscope slide and save the remaining two anthers in a small petri dish containing 70% ethanol (see Note 16). 6. Add a small drop of 2% acetocarmine on the slide and cut the anther in half. Gently squeeze out meiocytes with dissecting needles. The anther wall debris is then removed by pulling away from the drop of stain using a needle. 7. Use an iron nail to mix the stain for about 1 min and apply a coverslip. Gently heat the slide over an alcohol flame. Do not boil the stain (see Note 17). 8. Carefully turn the slide upside down and place it on a stack of Whatman filters. 9. Use thumb to apply a steady pressure to squash meiocytes. Avoid shifting the coverslip by pressing both ends of the slide using two fingers of the other hand. If needed, gently heat the slide repeatedly until the stain gets intense enough. 10. Check meiotic stages under a light microscope. If it is the desired stage, store the other two anthers from the same floret for further study (Fig. 8a). 3.6 Fluorescent In Situ Hybridization (FISH) for Maize 3.6.1 Maize Meiotic Chromosome Spreading

1. When enough anthers at the desired meiotic stage are collected, thaw an aliquot of enzyme cocktail and keep it on ice. 2. Wash two anthers in 1× enzyme buffer for three times in a petri dish, 5 min each. 3. Transfer anthers into a big drop of enzyme cocktail in a clean petri dish and digest for 15–30 min at room temperature (see Note 18). 4. Carefully lift one anther using two dissecting needles and transfer it to a drop of enzyme buffer placed on a microscope slide. Let it sit for 1–2 min (see Note 19). 5. Carefully suck up the enzyme buffer using a small piece of filter paper. Do not touch or disturb the softened anther. 6. Add 10 μL of 45% acetic acid, cut the anther in half and gently swirl material in the drop of acetic acid with a needle. If the digestion is just right, meiocytes should easily be released. 7. Remove anther wall debris by collecting all visible bits of tissue and pulling away using a needle, while tilting the slide so the liquid portion can separate from the undigested tissue. Wipe out the debris using a Kimwipe paper. 8. Apply an 20 × 20 mm coverslip without trapping bubbles. To do so, using a right amount of liquid without visible debris remaining is important (see Note 20).

Analyses of Meiosis in Plants

233

Fig. 8 Maize pachytene chromosomes. (a) A maize meiocyte at the pachytene stage stained with acetocarmine. Note the dark stain of nucleolus (upper center) and dense cytoplasm (background). (b) A maize meiocyte at the pachytene stage spread and squashed after enzyme digestion. Note the bright staining of two somatic nuclei (upper right and lower center-right). (c) Maize pachytene chromosomes (stained with DAPI and shown in white, with the chromosome arms indicated) hybridized with several probes for the 5s rDNA (labeled with cy5, shown in blue), 180 bp knob (labeled with cy3 and FITC, shown in yellow-green), CentC centromere repeat (labeled with cy3, shown in red), TR-1 repeat (labeled with cy3 and cy5, shown in purple), or 45s rDNA (labeled with cy5, shown in blue). (d) The FISH signals (without the chromosomes) with corresponding probes as labeled. Note that maize centromeres often fuse together during prophase I, so only five centromere signals are detected in this cell

9. Carefully turn the slide upside down and place it on a stack of Whatman filters. Stabilize the slide by pressing two fingers on both ends of the slide. Do not shift the coverslip when turning the slide over. 10. Spread and squash meiocytes by tapping across the entire area covered by the coverslip using a pencil with vertical motions. Do not shift the coverslip when tapping.

234

Cong Wang et al.

11. Check the slide under a phase contrast or Differential Interference Contrast (DIC) microscope. If the preparation is not sufficiently flat and spread, squash the slide again. 12. Immediately dip the slide in liquid nitrogen using a pair of long forceps until liquid nitrogen stops boiling. Hold the forceps with one hand, and get a razor blade ready in the other hand. Pull out the slide quickly, and flick off the coverslip with a razor blade in a single swift motion. Allow the slide to air-dry. 13. Check the slides to choose suitable chromosome preparations for FISH. 14. If only DAPI-stained images are required, add DAPI and cover with a coverslip. Seal the coverslip with nail polish. The slide is now ready for microscopic observation (Fig. 8b). 3.6.2 Pretreatment of Chromosome Spreads

1. Wash slides twice in 95% ethanol for 5 min each and air-dry. 2. Add 200 μL of 100 μg/mL RNase, cover with a large plastic coverslip and incubate for 1 h at 37 °C in a humid chamber (see Note 21). 3. Remove coverslips carefully without scraping the surface of the slide and wash the slides in a Coplin jar with 2× SSC for 5 min, twice. 4. Incubate slides in 10 mM HCl in a Coplin jar for 5 min. Shake off excess fluid and wipe both ends of the slide. Quickly add 200 μL of 1 μg/mL pepsin, cover with a plastic coverslip, and incubate for 10–20 min at 37 °C in a humid chamber (see Note 22). 5. Stop the digestion by placing slides in distilled water for 1 min. Wash slides in 2× SSC for 5 min, twice. 6. Shake off excess fluid and wipe both ends of slides before adding 200 μL of 4% paraformaldehyde fixative. Re-fix chromosome spreads at room temperature for 10 min (see Note 23). 7. Wash slides in 2× SSC for 5 min, twice. 8. Dehydrate slides using a series of 75%, 90%, and 95% ethanol in Coplin jars for 5 min each and then air-dry slides. 9. Check slides under a phase-contrast or DIC microscope. Cytoplasm should be removed and chromosomes should be observed easily. Thus, choose good preparations for the next step.

3.6.3

Hybridization

1. Denaturation of chromosomes can be done with hybridization cocktail using a thermal cycler machine. To do so, first denature the hybridization cocktail at 85 °C for 5 min, and then immediately place the tube on ice (see Note 24).

Analyses of Meiosis in Plants

235

2. Add 20 μL of denatured hybridization cocktail on the chromosome preparation from each slide. Cover with a 22 × 22 mm coverslip and seal well with rubber cement. 3. Denature chromosomal DNA together with hybridization cocktail using a thermal cycler machine at 75 °C for 5 min and cool down slowly by turning off the thermal cycler. Leave slides at 37 °C in a humid chamber overnight. 4. Alternatively, chromosomal DNA on slides can be denatured in a Coplin jar alone by denaturation solution. To do so, heat the denaturation solution slowly in a Coplin jar in a water bath with lid to 72 °C in a chemical hood (see Note 25). 5. Denature the hybridization cocktail at 85 °C for 5 min, and immediately place the tube on ice. 6. Place two slides each time in the pre-heated denaturation solution in a water bath to denature chromosomal DNA for 2.5 min. 7. Immediately transfer slides into 75% pre-chilled ethanol (at 20 °C) and incubate for 2 min, followed by pre-chilled 95% and 100% ethanol for 2 min each with gentle agitation by hand. 8. Shake off ethanol and let slides air-dry. 9. Carefully pipette 20 μL of denatured hybridization cocktail on each slide. Cover with a 22 × 22 mm coverslip, seal well with rubber cement, and leave the slides at 37 °C in a humid chamber overnight. 3.6.4 Stringent Wash and Mounting

1. Prepare a Coplin jar containing the stringent washing solution and another jar containing 2× SSC. Heat them to 42 °C in a water bath inside a chemical hood (see Note 25). 2. Take slides from the hybridization chamber, carefully remove the rubber cement using a pair of forceps and submerge the slides in 2× SSC in a Coplin jar at room temperature. 3. Gently shake the jar to allow the coverslips to slide off and transfer the slides into another Coplin jar containing 2× SSC for 5 min. 4. Place slides in the pre-heated stringent wash solution at 42 °C for 10 min with gentle agitation by hand every minute. 5. Transfer slides to the pre-heated 2× SSC and take the jar to room temperature. Cool down with gentle shaking for 10 min. 6. Transfer slides to another Coplin jar containing 2× SSC and incubate for 5 min, twice with gentle shaking, at room temperature. 7. Wash slides with 1 × PBS buffer twice for 5 min each.

236

Cong Wang et al.

8. Let slides air-dry and add DAPI. Apply a 22 × 22 mm coverslip and seal the coverslip with nail polish. The slide is now ready for microscopic observation (Fig. 8c).

4

Notes 1. RNA extracted from male meiocytes using conventional methods is prone to degradation. Thus, meiocytes are incubated in 0.5 × DPBS to maintain vitality. DPBS is a buffer used for a variety of cell culture applications, and that can also be obtained ready-made from several providers (e.g., Life Technologies). RNase Inhibitor (for instance, from Takara Bio) is used to suppress RNA degradation. 2. We currently use macerozyme R-10 instead of the previously used pectolyase Y-23, since the former is more effective and needs less time for digestion. These enzymes can be obtained from several providers, but we routinely use cytohelicase from Sigma and cellulase and macerozyme from Yakult. 3. The goat serum fluid is a preferable substitution for previous blocking buffer (PBST-BSA) for blocking and antibody binding, with less background signals. 4. The developmental stages and the floral bud sizes are specific for soybean Huaxia3 and lettuce Japanese Butterhead. They might be different for other cultivars of soybean and lettuce. 5. Isolated meiocytes can be stored in -80 °C for a month without degradation. 6. After fixation, the inflorescences can be stored in Carnoy’s fixative at -20 °C for at least 2 months before use. 7. The digestion mixture (new recipe) works more effectively, so the amount of time for digestion should be strictly observed. Otherwise, over-digestion not only makes it difficult to separate anthers from other floral organs, but also potentially causes abnormal chromosomal morphology. Digestion time may need to be optimized for other species or backgrounds. 8. The digested inflorescences fixed by paraformaldehyde can be stored at 4 °C for only a few hours; longer storage will lead to protein degradation. 9. In our hands, anti-LsZYP1 antibody, generated by injecting peptide QNLNERKEKENTD into rabbits, is diluted 1:200 in blocking buffer. Goat anti-rabbit IgG (H + L) secondary antibody conjugated with Alexa Fluor® 555 (e.g., Thermo Fisher, A-21428) is used with 1:1000 dilution. 10. DNA is denatured after microwaving, and rapid cooling can make it easier to bind to the centromere probes.

Analyses of Meiosis in Plants

237

11. Overly long incubation time might reduce the DAPI signal when observing meiotic chromosomes. 12. At this step, it is also feasible to add repetitive centromere DNA probes into the working solution to examine the co-localization of centromere and heterochromatin markers. Two-hour incubation with secondary antibodies can also be performed together with hybridization using centromere probes (Fig. 7). 13. Maize developmental stage can be determined by the total number of leaves with visible collars, which is the white band at the base of the lead blade. Depending on different genetic backgrounds, meiotic anthers can be found in V7–V10 stages, about 4–6 weeks after germination [24]. 14. Fixation can be up to 24 h, or longer for an extended period of storage of fixed material. However, the material becomes harder and may need longer digestion after a longer fixation. 15. Depending on different genetic backgrounds, male meiosis is initiated in 1–1.5 mm anthers and complete in 2.5–3 mm anthers [25]. 16. Each male spikelet contains two florets, and the upper floret develops 1 day ahead of the lower floret. Usually, only the upper florets are used, for consistency in developmental progress. Each floret contains three anthers, which are synchronous with each other. So one can stage one anther and harvest two others for other experiments [6]. 17. Ferric hydroxide can enhance carmine staining, but results in precipitation. An iron nail is usually good enough to darken the stain. 18. The digestion time can vary, depending on previous fixation times, storage times, and different developmental stages. Material with a longer fixation or stored for several months usually becomes harder and needs longer digestions. 19. In this setting, two anthers in the enzyme cocktail could have different digestion times, allowing the optimization of digestion time. 20. Examine meiocytes under a microscope for evaluating the digestion results. Meiocytes should be released from anthers and flattened by coverslips. 21. Plastic coverslips are made from sterile hybridization bags, by cutting them into 24 × 45 mm rectangles. 22. Pepsin is used to remove cytoplasm covering chromosome preparations, resulting in better probe accessibility and less background. However, the digestion time needs to be adjusted. If there is little cytoplasm, skip steps 4–5.

238

Cong Wang et al.

23. Re-fixation with formaldehyde can preserve chromosomal morphology better during denaturation. 24. For detection of chromosomal landmarks, denaturation of chromosomal DNA in hybridization cocktail is usually sufficient. This method is quick and avoids the hazards of using a large amount of formamide. However, for detection of low-copy or single copy sequences, 50% formamide in hybridization cocktail is not rigorous enough. Thus, incomplete denaturation of chromosomal DNA causes weak or no signal. 25. In steps 4–9, chromosomal DNA is denatured in the denaturation solution containing 70% formamide, allowing all target sequences to become single-stranded. This method usually produces a better detection rate for low-copy or single-copy sequences. Stringent wash step can remove imperfect hybridization between labeled probes and chromosomal DNA that share a lower homology with probes. This step is more important for FISH in large genomes with higher numbers of gene families and duplicated genes. The stringent washing solution (20% formamide in 2× SSC) described here is not harsh, which allows some hybridization with 70–80% homologous sequences in the genome. Thus, it may cause some undesired signals. To raise stringency, reduction of sodium ion (e.g., 0.1× SSC) can remove non-specific hybridization and reduce background.

Acknowledgments This work has benefited from the National Science Foundation of China (31925005, 31900445), the Laboratory of Lingnan Modern Agriculture Project (NT2022002), the China Postdoctoral Science Foundation, and National Postdoctoral Program for Innovative Talents. References 1. Wang Y, Copenhaver GP (2018) Meiotic recombination: mixing it up in plants. Annu Rev Plant Biol 69:577–609 2. Luo Q, Li Y, Shen Y et al (2014) Ten years of gene discovery for meiotic event control in rice. J Genet Genomics 41(3):125–137 3. Lambing C, Franklin FC, Wang CR (2017) Understanding and manipulating meiotic recombination in plants. Plant Physiol 173(3): 1530–1542

4. Mercier R, Me´zard C, Jenczewski E et al (2015) The molecular biology of meiosis in plants. Annu Rev Plant Biol 66:297–327 5. Beadle GW (1932) Genes in maize for pollen sterility. Genetics 17(4):413–431 6. Cande WZ, Golubovskaya I, Wang CJR et al (2009) Meiotic genes and meiosis in maize. In: Bennetzen JL, Hake SC (eds) Handbook of maize. Springer, New York

Analyses of Meiosis in Plants 7. Zickler D, Kleckner N (1999) Meiotic chromosomes: integrating structure and function. Annu Rev Genet 33:603–754 8. Anderson LK, Salameh N, Bass HW et al (2004) Integrating genetic linkage maps with pachytene chromosome structure in maize. Genetics 166(4):1923–1933 9. Lee DH, Kao Y-H, Ku J-C et al (2015) The axial element protein DESYNAPTIC2 mediates meiotic double-strand break formation and synaptonemal complex assembly in maize. Plant Cell 27(9):2516–2529 10. Chen C, Zhang W, Timofejeva L et al (2005) The Arabidopsis ROCK-N-ROLLERS gene encodes a homolog of the yeast ATP-dependent DNA helicase MER3 and is required for normal meiotic crossover formation. Plant J 43(3):321–334 11. Li W, Chen C, Markmann-Mulisch U et al (2004) The Arabidopsis AtRAD51 gene is dispensable for vegetative development but required for meiosis. Proc Natl Acad Sci U S A 101(29):10596–10601 12. Li W, Yang X, Lin Z et al (2005) The AtRAD51C gene is required for normal meiotic chromosome synapsis and doublestranded break repair in Arabidopsis. Plant Physiol 138(2):965–976 13. Wijeratne AJ, Chen C, Zhang W et al (2006) The Arabidopsis thaliana PARTING DANCERS gene encoding a novel protein is required for normal meiotic homologous recombination. Mol Biol Cell 17(3): 1331–1343 14. Armstrong SJ, Sanchez-Moran E, Franklin FC (2009) Cytological analysis of Arabidopsis thaliana meiotic chromosomes. Methods Mol Biol 558:131–145 15. Wang CJ, Harper L, Cande WZ (2006) Highresolution single-copy gene fluorescence in situ hybridization and its use in the construction of a cytogenetic map of maize chromosome 9. Plant Cell 18(3):529–544

239

16. Martin JA, Wang Z (2011) Next-generation transcriptome assembly. Nat Rev Genet 12(10):671–682 17. Yang WC, Shi DQ, Chen YH (2010) Female gametophyte development in flowering plants. Annu Rev Plant Biol 61:89–108 18. Scott RJ, Spielman M, Dickinson HG (2004) Stamen structure and function. Plant Cell 16 Suppl:S46–S60 19. Yang HX, Lu P, Wang Y et al (2011) The transcriptome landscape of Arabidopsis male meiocytes from high-throughput sequencing: the complexity and evolution of the meiotic process. Plant J 65(4):503–516 20. Yelina N, Diaz P, Lambing C et al (2015) Epigenetic control of meiotic recombination in plants. Sci China Life Sci 58(3):223–231 21. Wang Y, Cheng Z, Lu P et al (2014) Molecular cell biology of male meiotic chromosomes and isolation of male meiocytes in Arabidopsis thaliana. Methods Mol Biol 1110:217–230 22. Chelysheva LA, Grandont L, Grelon M (2013) Immunolocalization of meiotic proteins in Brassicaceae: method 1. Methods Mol Biol 990:93–101 23. Wang L, Cao C, Ma Q et al (2014) RNA-seq analyses of multiple meristems of soybean: novel and alternative transcripts, evolutionary and functional implications. BMC Plant Biol 14:169 24. Begcy K, Dresselhaus T (2017) Tracking maize pollen development by the Leaf Collar Method. Plant Reprod 30(4):171–178 25. Nan G-L, Ronceret A, Wang RC et al (2011) Global transcriptome analysis of two ameiotic1 alleles in maize anthers: defining steps in meiotic entry and progression through prophase I. BMC Plant Biol 11:120 26. Huang J, Wang C, Li X et al (2020) Conservation and divergence in the meiocyte sRNAomes of Arabidopsis, soybean, and cucumber. Plant Physiol 182(1):301–317

Chapter 10 Genetic and Phenotypic Analyses of Carpel Development in Arabidopsis Vicente Balanza`, Patricia Ballester, Monica Colombo, Chloe´ Fourquin, Irene Martı´nez-Ferna´ndez, Clara I. Ortiz-Ramı´rez, and Cristina Ferra´ndiz Abstract Carpels are the female reproductive organs of the flower, organized in a gynoecium, which is likely the most complex organ of the plant. The gynoecium provides protection for the ovules, helps to discriminate between male gametophytes, and facilitates successful pollination. After fertilization, it develops into a fruit, a specialized organ for seed protection and dispersal. To carry out all these functions, coordinated patterning and tissue specification within the developing gynoecium has to be achieved. In this chapter, we provide different methods to characterize defects in carpel morphogenesis and patterning associated with developmental mutations, as well as a list of reporter lines that can be used to facilitate genetic analyses. Key words Carpel, Gynoecium, Chloral hydrate, Vascular cleaning, Pollination, Lignin, Transmitting tract, Carpel tissue reporters

1

Introduction Carpels are the female reproductive organs of the flower. In Arabidopsis, two congenitally fused carpels form a single pistil or syncarpous gynoecium. At maturity (anthesis), several parts are readily visible in an Arabidopsis pistil. At the base of the gynoecium, a short internode called the gynophore connects the pistil to the base of the flower. Above the gynophore is the ovary, which comprises most of the length of the gynoecium and contains between 50 and 80 ovules. The ovary is divided longitudinally by a septum, which is formed postgenitally forming two chambers or locules. The two ovary wall regions of the pistil are termed the valves, and the external part of the septum is termed the replum. At the apical end of the ovary are the style and the stigma. The stigma consists

Vicente Balanza`, Patricia Ballester, Monica Colombo, Chloe´ Fourquin, Irene Martı´nez-Ferna´ndez and Clara I. Ortiz-Ramı´rez have contributed equally to this work and are listed alphabetically. Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_10, © Springer Science+Business Media, LLC, part of Springer Nature 2023

241

242

Vicente Balanza` et al.

Fig. 1 The Arabidopsis mature gynoecium. Scanning electron micrograph of an Arabidopsis gynoecium at anthesis (stage 13). The different parts are indicated

of a single layer of specialized epidermal cells termed stigmatic papillae, and it is in charge of receiving and germinating pollen grains. The stigma constitutes the initial portion of the transmitting tract, a polysaccharide-rich tissue specialized in directing pollen tube growth. Basal to the stigma, the style appears as a short, solid cylinder with distinctive epidermal morphology. The central core of the style is composed of transmitting tract tissue, surrounded by vascular tissue. The tract continues through the central part of the septum, guiding the pollen tubes toward the ovules. After fertilization, the ovules develop into seeds and the Arabidopsis gynoecium is transformed into an elongated bilocular fruit called silique. This structure opens at maturity to release its seeds along four dehiscence zones defined by longitudinal furrows of smaller cells on either side of the replum. The lignification of specific cells in these zones contributes to the dehiscence process (see the following text) (Fig. 1) [1, 2]. Developmental mutants affect the distribution and identity of the different parts of the pistil [1, 2] and these morphological defects are best observed with standard scanning electron microscopy procedures (see Chapters 7 and 8). In addition to the determination of changes in overall morphology, a number of other specific techniques can help visualize frequent defects associated with mutations affecting carpel and fruit development. In this chapter we describe methods for studying these defects:

Genetic and Phenotypic Analyses of Carpel Development

243

1. First, we describe how to observe pollen tube growth by aniline blue staining. Aniline blue reveals callose deposits associated to pollen tube formation giving a brilliant yellow fluorescence with essentially no background fluorescence. Mutants affected in transmitting tissue development show poor pollen tube growth [3]. Transmitting tissue is also easily revealed by alcian blue staining, although this protocol is better described in another section of this same chapter (see Subheading 3.5). 2. Second, a method to optically clear plant tissues is provided. We use this method to observe vascular development in the gynoecium using darkfield microscopy, where lignified xylem elements appear white. Vascular strands are frequently affected by mutations in genes involved in gynoecium morphogenesis, and especially by altered auxin signaling [4, 5]. Main differences are observed in the position of the medial bundle bifurcation and the extent of vascular fans in the style. Tissue clearing is also a good option to visualize ovule development (see Chapter 11) and to obtain detailed GUS reporter activity when combined with differential interference contrast (DIC) microscopy (see Chapter 18). 3. Wildtype Arabidopsis gynoecium development is sensitive to auxin transport inhibition by 1-N-naphtylphtalamic acid (NPA), which affects overall fruit morphology and vascular development. Several classical Arabidopsis gynoecium mutants are hyper or hyposensitive to this treatment [5–7] and, thus, the characterization of NPA sensitivity in a new mutant can provide clues on potential genetic interactions with other factors of known pathways directing gynoecium development. 4. Finally, we describe different alternative methods to reveal lignin deposition patterns. Lignin is one of the most abundant organic polymers and an integral part of the secondary wall of plants, to which it confers mechanical strength. Lignin is a secondary metabolite which is synthesized from the phenylalanine/tyrosine metabolic pathway in plant cells [8]. In general, the polymerization of the lignin is from three types of monolignols (sinapyl alcohol, S unit; coniferyl alcohol, G unit, and pcoumaryl alcohol, H unit) by peroxidase (POD) and laccase (LAC) in secondary cell wall [8]. Lignin is produced in high quantities in the secondary walls of vascular tissue cells. In Arabidopsis fruits, lignification is associated to vascular bundles, the endocarpb (endb, the subepidermal internal cell layer) and two or three cell rows adjacent to the separation layer of the dehiscence zone, where it contributes to the establishment of tensions which help to pod shatter [9–11]. In a temporal sequence, lignin becomes easily detected first in the vascular tissues around developmental stage 15–16 [12, 13], while the lignification of the endb and the dehiscence zone appear later,

244

Vicente Balanza` et al.

between late stage 17A and stage 17B [12]. Although lignin staining is not quantitative, it can reveal visual cues informing about the integrity of tissues or abnormal cell wall deposition and lignification, that is very important for phenotypical characterization. The methods described in this chapter can easily be applied to the characterization of lignin deposition in fruits from other species with only minor optimization. 5. In addition to these methods for phenotypic analyses, we include a table of published reporter lines that can be useful for genetic analyses. They correspond to markers for specific tissues within the pistil or for hormone distribution, as well as to expression reporters for key factors involved in carpel morphogenesis. For details on the role of these factors, see [1, 2].

2

Materials

2.1 Aniline Blue Staining of Arabidopsis Pollen Tubes

1. Fine forceps (tweezers; 5s is a suitable grade). 2. Binocular microscope (or magnifying glasses). 3. Microfuge tubes. 4. Pasteur pipettes or micropipettes. 5. Microscope slides. 6. Coverslips. 7. Microscope with UV light and DAPI filter. 8. Fixing solution: absolute ethanol:acetic acid (3:1). 9. Softening solution: 8 M NaOH. 10. Aniline blue solution: Prepare 0.1% (w/v) aniline blue in 0.1 M K2HPO4-KOH buffer, pH 11 (see Note 1). 11. 50% glycerol.

2.2 Cleared Tissue for Observation of Vascular Development

1. Fine forceps (tweezers; 5s is a suitable grade). 2. Glass scintillation vials or similar containers. 3. Pasteur pipettes or micropipettes. 4. Microscope concavity slides (see Note 2). 5. Coverslips (20 × 20 mm). 6. Surgical blades and needles (see Note 3). 7. Nail polish (see Note 4). 8. White paper towels. 9. Microscope with dark field optics. 10. Fixing solution: absolute ethanol:acetic acid (6:1). chloral 11. Clearing solution: (8 g:1 mL:2 mL) (see Note 5).

hydrate:glycerol:H2O

Genetic and Phenotypic Analyses of Carpel Development

2.3

NPA Treatment

245

1. Forceps. 2. Plastic spray bottles (0.5–1 L volume). 3. Clear humidity domes for trays (see Note 6). 4. Binocular dissecting scope. 5. NPA solution: 100 μM 1-N-naphtylphtalamic acid (NPA), 0.01% Silwet L77. 6. Mock solution: 0.01% Silwet L77 (see Note 7).

2.4

Lignin Staining

1. Fine forceps and needles. 2. Glass scintillation vials or similar containers. 3. Vacuum bell. 4. Oven at 60 °C. 5. Microscope polylysine pretreated slides. 6. Coverslips (24 × 60 mm). 7. Histoprep metal base molds and plastic embedding cassettes. 8. 37–55 °C microscope slide warming table. 9. FAA fixation solution: 3.7% formaldehyde, 50% ethanol, 5% acetic acid. Prepare a fresh solution (see Note 8). 10. Eosin solution: 2% Eosin in 95% ethanol. 11. Histoclear (see Note 9). 12. Graded ethanol series in distilled water. 13. Paraplast X-tra paraffin chips. 14. Microtome. 15. Tissue flotation bath. 16. Fine brushes. 17. Phloroglucinol solution (Wiesner stain): Prepare 3% phloroglucinol solution in absolute ethanol (see Note 10). Add concentrated HCl in a 2:1 v/v proportion (see Note 11). 18. Stock alcian blue: Prepare 1% alcian blue 8G (also called Ingrain blue 1) solution in 50% ethanol (see Note 12). 19. Stock Safranin: Prepare 1% Safranin-O solution in 50% ethanol (see Note 13). 20. Acetate buffer: 0.1 M NaOAc—HOAc, pH 5.0. 21. Toluidine blue solution: Prepare 0.02% Toluidine blue solution in distilled water (see Note 14). 22. Mounting medium (Entellan, Merkoglass).

2.5

Genetic Analyses

See Table 1 for seed stocks.

246

Vicente Balanza` et al.

Table 1 Reporter lines can be useful to characterize a carpel mutant

Ref.

ABRC/ NASC

pDR5rev::3xVENUS Synthetic promoter responsive to auxin accumulation

[14]

N799364

TCS::GFP

[15, 16] CS23900

Reporter line

Use

Hormone reporter lines

Synthetic promoter responsive to cytokinin accumulation

Key developmental gene reporter lines SHP1::GUS

3.5 kb of SHP1 genomic sequence upstream of the translational start to GUS

[17]

SHP2::GUS

2.1 kb of SHP2 genomic sequence upstream of the translational start fused to GUS

[18]

IND::GUS

2.7 kb of IND genomic sequence upstream of the [19, 20] translational start and 0.5 kb of the coding region fused to GUS

FUL::GUS

[21] Ful-1 bears an enhancer trap inserted in the 5′UTR, rendering a null allele. GUS activity mimics faithfully FUL expression patterns and can be used as reporter for FUL in ful-1 heterozygous backgrounds 2.3 kb of FUL promoter sequence fused to GUS [22]

STY1::GUS

2 kb of STY1 promoter sequence fused to GUS

[23]

STY2::GUS

2.1 kb of STY2 promoter sequence fused to GUS

[24]

NGA3::GUS

2.7 kb of NGA3 genomic sequence upstream the translational start codon fused to GUS

[25]

HEC1::GUS

2972 bp of HEC1 genomic sequence upstream of the translational start and the coding region fused to GUS

[26]

HEC2::GUS

3058 bp of HEC2 genomic sequence upstream of the translational start codon fused to GUS

HEC3::GUS

2979 bp of HEC3 genomic sequence upstream of the translational start codon fused to GUS

CRC::GUS

8 kb of CRC genomic sequence upstream of the translational start and the coding region fused to GUS

SPT::GUS

[28] 6.5 kb of SPT genomic sequence upstream of the translational start and 313 bp of the coding region fused to GUS

N3759

N8847

[27]

ETT::GUS Reporter line with a translational fusion of ETTIN/ARF3 [29] (ARF3::ARF3:GUS) to GUS driven by ETT promoter

CS66480

Tissue-specific reporter lines SLG::GUS

[30, 31] Stigma Promoter region of the S. locus glycoprotein (SLG) gene from Brassica oleracea fused to GUS. GUS expression in the gynoecium is limited to stigmatic papillary cells (continued)

Genetic and Phenotypic Analyses of Carpel Development

247

Table 1 (continued)

Reporter line

Use

ASA1::GUS

Style [32] Promoter for the anthranilate synthase 1 (ASA1) gene fused to GUS. GUS expression in the gynoecium is limited to inner style tissues in pollinated wild-type gynoecia

GT140

Dehiscence zone Enhancer trap inserted in the promoter region of IND. GUS is detected in the dehiscence zone during fruit development

3

ABRC/ NASC

[33, 34] N26931

Methods

3.1 Aniline Blue Staining of Arabidopsis Pollen Tubes 3.1.1

Ref.

Material Collection

To compare the general performance and extent of pollen tube growth, Arabidopsis pistils can be collected from flowers 1 or 2 days after anthesis (flowers are wide open). Remove sepals, petals, and stamens from around the pistils. For in vivo pollen tube guidance experiments, flowers at developmental stage 12 must be emasculated and pollinated manually. 1. Remove any siliques, open flowers, open buds (it is possible to see the stigma poking out through the top of the bud), meristem, and smaller buds from the inflorescence. 2. While working at the binocular microscope it is best to immobilize the inflorescence, for instance using a restraint from a thin strip of paper held with sticky tape. 3. Emasculate stage 12 flowers: first carefully remove (or displace) sepals and petals and then remove the immature anthers. Flowers can be protected from undesired pollination by covering the inflorescence. 4. Perform pollination after 24 h by touching the stigma with a suitable anther from a mature flower (maximal pollination). Minimal pollination (pollination using minimal amount of pollen) can be used to observe the growth morphology and growth pattern of individual pollen tubes. In this case squash a mature anther on a microscope slide and, under the microscope, use a fine hair to pick up 1–5 pollen grains and transfer them to the stigma. Flowers can be labeled with the female and male accession and the date of pollination. 5. Pollinations are allowed to progress for 24 h, unless specified otherwise (e.g., to observe pollen tube progression, time series of 2, 4, 6, 12, 24, and 48 h are suggested). After the desired amount of time post pollination, the remaining sepals and petals are removed from around the pistil (see Note 15).

248

Vicente Balanza` et al.

3.1.2

Tissue Fixation

1. Place the excised pistils in a 1.5 mL microfuge tube containing the fixing solution for at least 1 h at room temperature. If necessary, tissue can be left in the fixing solution overnight or indefinitely.

3.1.3

Pistil Softening

1. Replace the fixing solution with the softening solution. Leave the pistils in the softening solution overnight at room temperature. 2. Carefully remove the softening solution by pipetting. The tissue is now extremely fragile and can be easily damaged. 3. Gently wash the pistils three to five times with distilled water.

3.1.4

Pistil Staining

3.1.5 Pistil Mounting and Visualization

1. Replace the water with the aniline blue solution. Leave for at least 2 h at room temperature under dark condition (for instance using a piece of aluminum foil). 1. Place the stained pistils in a drop of 50% glycerol on a microscope slide and cover with a cover slip. Put the cover slip on it carefully, starting from the end of the pistils, in order to avoid bubbles. 2. Observe with a standard fluorescence microscope under UV irradiation condition and DAPI emission filters to view the fluorescent signal from the tissue. See Fig. 2.

3.2 Cleared Tissue for Observation of Vascular Development

1. Anthesis (stage 13, [13]) is the preferred reference stage of flower development to compare vascular development in the gynoecium. This stage is easily recognized because the flower is fully open and anthers are dehiscent and leveled with the stigma of the pistil. When collecting the flowers, leave a long pedicel to facilitate manipulation without harming the pistil.

3.2.1

Material Collection

3.2.2

Tissue Fixation

1. Place anthesis flowers in vials containing enough fixing solution as to completely cover plant tissue (see Note 16). Fix at least 6 h or leave overnight at room temperature.

3.2.3

Tissue Clearing

1. Replace fixative with absolute ethanol, and incubate for 30 min at room temperature. Repeat twice. 2. Replace ethanol with clearing solution. Allow tissue to clear for at least 48 h at room temperature in darkness. Longer clearing times (up to 1 week) can improve transparency, but the tissue becomes very fragile and it is more difficult to manipulate.

3.2.4 Pistil Mounting and Visualization

1. Mount the samples under a dissection scope. We like to use concavity slides to avoid squashing the pistil. To properly orient the pistil, place the flower on a drop of clearing solution in the flat surface of the concavity slide, close to the depression, with the replum facing upward. Use fine forceps or needles to remove

Genetic and Phenotypic Analyses of Carpel Development

249

Fig. 2 Pollen tube growth staining. Pollen tube growth stained with aniline blue of 1-day postanthesis pistils from wildtype Columbia (left) and ngatha quadruple mutant (right). ngatha mutants are strongly defective in stigma and style development [6, 21] and therefore show poor pollen tube growth when compared to wild type

additional floral organs that may interfere with pistil visualization. Then, gently push the pistil to the center of the depression filled with clearing solution trying to maintain the proper orientation. 2. If needed, pipet enough clearing solution to completely cover the sample. Be careful not to touch the sample with the tip. Wipe out any excess with a white paper towel. 3. Place the coverslip over the depression in the concavity slide, avoiding making bubbles as much as possible. 4. Seal the coverslip with nail polish and let it dry in the chemical fume hood. 5. Observe with a standard microscope under darkfield illumination (see Fig. 3). 3.3

NPA Treatment

3.3.1 Plant Growth and Preparation

1. Arabidopsis plants are grown on soil under standard greenhouse conditions (see Note 17). Prepare two duplicate batches of plants in separate trays including wildtype controls and mutants of interest. Usually, 10–20 individual plants for each genotype are sufficient. One batch will be treated with NPA and another with mock solution. 2. Grow plants until bolting is apparent and first open flowers are visible (see Fig. 4a).

250

Vicente Balanza` et al.

Fig. 3 Cleared pistils observed under darkfield to reveal vascular development. Vascular patterning in wildtype (left), ngatha quadruple mutant (center), and ettin mutant (right) pistils at anthesis. Medial vein bifurcation in the wild-type is visible at the boundary of style and ovary (arrow), while lateral veins do not reach the apical end of the ovary (asterisk). In the ngatha mutant, vascular development is very reduced, and medial veins stop prematurely and do not bifurcate. In the ettin mutant, veins bifurcate and overproliferate extensively 3.3.2

Plant Treatment

1. Spray NPA solution plentifully to one batch of plants and mock solution to the duplicate. 2. Cover the trays with clear domes or clear plastic bags to maintain humidity (see Note 6). Leave plants covered for 8–12 h. 3. Repeat NPA treatments two more times, spaced again by 8–12 h. 4. After the three treatments have been performed, wash plants by generously spraying them with water. Grow plants uncovered under standard greenhouse conditions for 3–4 more weeks.

3.3.3 Phenotype Visualization

1. 3–4 weeks post-treatment, developing fruits should show visible phenotypes. Collect all fruits from the main inflorescence of each plant. Fruits of stage 16 and onward [12] can be directly

Genetic and Phenotypic Analyses of Carpel Development

251

Fig. 4 Typical fruit phenotypes observed after NPA treatment. (a) Optimal stage for plant treatment. (b) On the left, typical morphology of fruits of wild-type, ettin mutant, and ngatha quadruple mutant when untreated or treated with mock solution. On the right, examples of fruit morphologies for these three genotypes observed after NPA treatment. Wildtype fruits frequently show elongated styles and gynophores and shorter ovaries. ettin mutants are hypersensitive to NPA treatment, and most fruits are extremely reduced, valveless, and with extensive stigma proliferation. ngatha mutants are insensitive to treatment, displaying the same morphology to that observed in mock-treated plants

observed under the dissecting scope, while young fruits (stages 13–15) should be fixed for scanning electron microscopy analyses (see Chapters 7 and 8). 2. Score the number of fruits affected by NPA treatment and the severity of their phenotypes, comparing among the different genotypes and with the mock-treated plants. Establish phenotypic categories and assign fruits to each one. Fruit phenotypes usually range within the same plant. Typically, in wildtype Columbia ecotype, NPA treatment causes a reduction in the length of the valves, with a concomitant increase in the length of the style and the gynophore. A useful set of categories has been described for young fruits [35]: mild, when valves cover more than half of the length of the fruit; medium, when valves are very reduced and cover less than half of the length of the fruit; strong, for valveless fruits. By scoring the proportion of affected fruits in each category for wildtype and the mutant of interest, it can be determined if the mutation causes hyper or hyposensitivity to auxin transport inhibition (see Fig. 4b). 3.4 Lignin Staining: Whole Mount Phloroglucinol Staining (Wiesner Stain)

1. Collect whole fruits at late stage 17B or beginning of stage 18 (fruit yellowing but closed).

3.4.1

4. Pull the vacuum slowly for 20 min. This step removes trapped air bubbles from the fruits and improves penetration of the fixative (see Note 18).

Tissue Fixation

2. Place fruits in glass vials filled with FAA solution. 3. Loosen the caps of the scintillation vials and place the vials in a vacuum chamber.

5. Release the vacuum slowly until vacuum bell can be opened. Keep the samples 2 h at room temperature and then overnight at 4 °C.

252

Vicente Balanza` et al.

6. The next day, remove the FAA fixation solution and replace with 70% ethanol. Leave in 70% ethanol for at least 30 min (see Note 19). 3.4.2

Fruit Staining

3.4.3 Fruit Lignin Visualization

3.5 Lignin Staining: Tissue Section 3.5.1

1. Remove 70% ethanol from the vial and replace with phloroglucinol solution. Incubate for 2–5 min (time can be increased up to 30 min if staining is weak). Gently move the solution to assure that all the sections are stained. 1. Immediately examine and photograph the fruits since the staining is only clearly visible for less than 30 min. Observe the sections under bright-field illumination. For this, place the fruits on a Petri dish or a slide, uncovered, and preferably over a dark background (see Fig. 4; see Note 20). Follow the steps described in Subheading 3.4.1.

Tissue Fixation

3.5.2 Tissue Dehydration and Paraplast Embedding

1. Discard the 70% ethanol and immerse the samples in ascending ethanol series (EtOH 70%, 80%, 96%) for 30 min each. 2. Pour off the 96% ethanol and replace with Eosin solution (see Note 21). Incubate 2 h at room temperature and then overnight at 4 °C. 3. Incubate the samples in the following solutions for 1 h each: 100% ethanol (twice), 25% Histoclear/75% ethanol, 50% Histoclear/50% ethanol, 75% Histoclear/25% ethanol, 100% Histoclear (three times). 4. Pour off half of the final Histoclear wash and add an approximately equal volume of paraplast X-tra paraffin chips. Incubate overnight at 60 °C in the oven. 5. Prepare a beaker full of paraplast chips and leave it also at 60 °C to melt overnight. 6. The next morning, remove the Histoclear/paraffin mixture, and replace with pure melted paraplast from the beaker. Incubate at 60 °C for 3–4 h. Repeat the step replacing molten paraplast at least three times (see Note 22). 7. On a warming table (55 °C) pour the paraplast solution containing the tissue into metal molds. Place one fruit per mold orienting the sample so the length of the fruit remains perpendicular to the bottom of the mold. This is usually easier if the fruit is cut in two halves. Align and orient the sample using a needle or fine forceps. If half fruit is being oriented, make sure that the section plane is facing the bottom. Once in this position, carefully slide the mold to a cooler area of the warming

Genetic and Phenotypic Analyses of Carpel Development

253

table. This will make the paraplast in the bottom of the mold to harden slightly, trapping the fruit in the right position. Before the upper paraplast sets, place the embedding cassette on top, adding some more molten paraplast if needed. 8. Move the block off the warming table and let the paraplast fully set. Once it is solid, it can be moved to 4 °C. This makes the metal mold to pop off easily. The embedded tissue can be stored at 4 °C for several months. 3.5.3

Tissue Sectioning

1. Trim excess paraplast with a razor blade forming a rectangle around the tissue. Make sure that edges are parallel. 2. Mount the block on the microtome sample holder. 3. Section the tissue for 7–10 μm thick sections. It will form long ribbons that can be carefully moved to a cardboard and cut into 1.5 cm pieces with a surgical blade. Use a fine brush to move the ribbons (see Note 23). 4. Float the ribbons on 42 °C water for more than 1 min. This expands the tissue and reduces wrinkles. An alternative to this may be to put warm water on the slide and expand the tissue with a brush. 5. Fish out the ribbons from the bath with a polylysine-pretreated microscope slide. Use a blunt wooden stick to help orient the ribbons on the slide. 6. Place the slides on a slide set at 40–45 °C. Let them dry overnight (see Note 24).

3.5.4

Tissue Staining

Three different staining can be used to reveal lignin deposition. Phloroglucinol Stain for Lignin Phloroglucinol stain can provide clues to the extent of cinnamaldehyde present in the xylem, fiber, and tracheal tissues. It stains the tissues in pink or fuchsia color; the intensity of the color correlates with the level of lignification. 1. Dewax the sections by placing the slides for 10 min in Histoclear. Repeat this step with clear Histoclear. Then, wash the slides twice in 100% ethanol for 2 min. 2. Stain slides with phloroglucinol-HCl solution for 10–15 min (up to 30 min in highly lignified tissues). 3. Place a coverslip on each slide and wipe the edges of the slide to prevent acid from damaging the microscope. 4. Observe with a standard microscope preferentially equipped with DIC optics. Examine and photograph the slide immediately as the staining lasts only around 30 min (see Fig. 5b).

254

Vicente Balanza` et al.

Fig. 5 Examples of fruit lignin staining obtained with the different described protocols. (a) Whole mount phloroglucinol staining of wild-type (left) and fruitful (right), a mutant with overlignified fruits. (b) Transversal section of an Arabidopsis mature fruit stained with phloroglucinol. The region corresponding to the valvereplum boundaries is shown. Lignin appears bright pink in the medial vascular vein, the subepidermal inner cell layer, and two small patches on the dehiscence zones (arrows). (c) Transversal section of an Arabidopsis mature fruit stained with alcian blue and Safranin-O. Lignin appears in red. Mucilage in the seed is also stained in bright red. (d) Transversal section of an Arabidopsis mature fruit stained with Toluidine blue. Lignin appears turquoise

Safranin and Alcian Blue Staining for Lignin Safranin and alcian blue are dyes most commonly used for staining plant tissues. The combination of both dyes allows to differentiate the cellulose in blue color (alcian blue) and the lignin in pink color (Safranin). 1. Prepare the staining solution: Add 5 mL of the alcian blue stock solution and 2 mL of the Safranin stock solution to 200 mL of the acetate buffer. The color should be dark purple. 2. Dewax the tissue sections in Histoclear twice for 10 min each. Then, wash twice in 100% ethanol for 2 min each. 3. Rehydrate the tissue by sequential washes of graded series of ethanol (90%, 70%, 50%, and 30%) for 2 min each. Wash the slides in distilled water for 2 min. 4. Place the slides in the staining solution for 30 min. 5. Wash with distilled water. 6. Let the slides dry for 1 h. 7. Apply a mounting medium and protect the slides with a coverslip (see Note 25).

Genetic and Phenotypic Analyses of Carpel Development

255

8. Observe with a standard microscope under bright field (see Fig. 5c). 9. Alternatively, lignin stained with safranin-O can also be observed with a fluorescence microscope, exciting at 492 nm and using a B-2A filter at 520 nm. Lignin will then fluoresce yellow to red (see Note 26). Toluidine Blue Staining Toluidine blue O is a cationic dye that binds to negatively charged groups [36]. An aqueous solution of this dye generates several colors when it reacts with different anionic groups in the cell [37, 38]. When the dye reacts with carboxylated polysaccharides such as pectic acid, the tissue stains a pinkish/purple color; when it binds to poly-aromatic substances such as lignin and tannins, green/greenish blue or bright blue; with nucleic acids, it turns to purplish/greenish blue [38]. 1. Dewax the tissue sections in Histoclear twice for 10 min each. Then, wash twice in 100% ethanol for 2 min each. 2. Rehydrate the tissue by sequential washes of graded series of ethanol (90%, 70%, 50%, and 30%) for 2 min each. Wash the slides in distilled water for 2 min. 3. Place the slides in a solution of 0.02% of Toluidine blue for 5 min. 4. Wash generously with distilled water, until it comes out almost clear. 5. Let dry the slides for 1 h. 6. Apply a mounting medium and protect the slides with a coverslip. 7. Observe with a standard microscope under bright field (see Fig. 5d). 3.6

4

Genetic Analyses

For genetic analyses, combination of mutants, overexpression lines, and/or reporter lines should be generated by crossing and genotyping of segregating populations following standard procedures. Several reporter lines can be useful to characterize a carpel mutant (see Table 1, with some of the most widely used reporter lines).

Notes 1. Aniline blue contains a fluorochrome (Sirofluor) that specifically binds to ß-1,3-glucan, a major component of the pollen tube wall. The water-soluble aniline blue (Sigma) is decolorized in aqueous phosphate (0.067–0.1 M K3PO4, or K2HPO4, or a mixture of both). The solution, initially dark blue or

256

Vicente Balanza` et al.

purple, will finally turn pale yellow as it decolorizes. The pH of the phosphate buffer varies in literature, but in general the higher the pH (up to 11.5), the more intense the fluorescence [29]. After preparation, store the aniline blue solution in the dark at 4 °C. 2. Normal microscope slides can also be used instead of concavity slides. For this, place transparent sticky tape on both sides of the sample, to avoid squashing it with the coverslip. 3. It is important to use disposable materials as much as possible, because the chloral hydrate can damage them. 4. Sealing of coverslips with nail polish prevents damage to the microscope lens. 5. Chloral hydrate is toxic to tissues of the mucous membranes and upper respiratory tract. It may be harmful by inhalation, ingestion, or skin absorption. Wear appropriate gloves, safety glasses, and mask, and use it in a chemical fume hood. Once prepared, clearing solution can be kept in darkness at room temperature for several months. 6. Clear domes fitting your standard growth trays can be used to keep humidity if they are tall enough to accommodate the bolting plants. Alternatively, use large plastic bags, placing the trays inside and sealing loosely with tape. 7. For a standard growth tray fitting 48 cell inserts or aracons, 200 mL of NPA and mock solution are sufficient. 8. Formaldehyde is toxic and volatile. It should be handled wearing gloves and under a fume hood. 9. Histoclear is moderately toxic and it has a strong lemon scent. Use preferably with gloves and under a fume hood. 10. Phloroglucinol solution in ethanol works better if prepared fresh. If needed, it can be stored in darkness for 3–4 days at room temperature. 11. Hydrochloric acid (HCl) is toxic and volatile. It should be used with great care under a fume hood, with gloves and safety glasses. 12. Alcian blue 8G is one of the most widely used cationic dyes for the demonstration of glycosaminoglycans and mucopolysaccharides. It will stain in blue the non-lignified cell walls. In addition, because transmitting tract cells secrete a complex extracellular matrix (ECM) very rich in acidic glycoproteins such as arabinogalactans, alcian blue is also widely used to reveal ECM by very intense blue staining. When used to stain ECM, counterstaining with neutral red or a similar dye is recommended.

Genetic and Phenotypic Analyses of Carpel Development

257

13. Safranin-O stains in red the lignified walls, the nuclei, and the chloroplasts. 14. Toluidine blue is a general-purpose dye, it stains the cell walls in blue and lignin in turquoise blue. Toluidine blue solution can be stored up to 1 month in a dark bottle at room temperature. 15. In wild-type carpels, pollen tubes are able to pass through the style in 4 h after pollination (hap) and reach the base of the ovary after 10–12 hap [3, 39]. 16. We prefer not to use microfuge tubes, since it is much easier to damage the pistil, which remains soft, when changing solutions. 17. In our case, plants are usually grown in cabinets at 21 °C under long-day (16 h light) conditions, illuminated by cool-white fluorescent lamps (150 μE/m2/s), in a 1:1:1 mixture of sphagnum:perlite:vermiculite. 18. Be careful not to pull too strong vacuum. You should be able to observe tiny air bubbles forming, but it is important to avoid boiling of the fixation solution. If you do not have a vacuum pump, keep the samples in fixative submerged in ice for 40 min. 19. At this point, the samples can be stored at 4 °C for weeks or months. 20. The phloroglucinol solution deteriorates the specimens. Minimize as much as possible the time required for taking pictures. 21. The eosin staining will help to visualize and orient the tissue when sectioning, but it will not give any color after sectioning, dewaxing, and specific staining are performed. 22. If needed, you can leave the samples overnight in molten paraplast in the oven and continue with the reminder paraplast changes the next morning. 23. If the tissue sections are ripped, try cooling the paraffin block on ice and then make thicker sections (up to 12 μm). 24. Once dry, the slides can be stored in a dry place, covered to avoid dust, for several months. 25. The mounted slides can be stored for a long time. The mounting medium binds the sample, slide, and coverslip with a clear durable film. 26. Due to changes in fluorescence emission, Safranin can differentiate regions of high and low lignin content more accurately than phloroglucinol; regions of high lignin fluoresce red/orange, and regions with low lignin fluoresce yellow [39].

258

Vicente Balanza` et al.

References 1. Balanza´ V, Navarrete M, Trigueros M, Ferra´ndiz C (2006) Patterning the female side of Arabidopsis: the importance of hormones. J Exp Bot 57(13):3457–3469 2. Ferrandiz C, Fourquin C, Prunet N, Scutt C, Sundberg E, Trehin C, Vialette-Guiraud A (2010) Carpel development. Adv Bot Res 55: 1–74 3. Crawford BCW, Ditta G, Yanofsky MF (2007) The NTT gene is required for transmittingtract development in carpels of Arabidopsis thaliana. Curr Biol 17(13):1101–1108 4. Alvarez J, Smyth D (2002) CRABS CLAW and SPATULA genes regulate growth and pattern formation during gynoecium development in Arabidopsis thaliana. Int J Plant Sci 163:17–41 5. Nemhauser JL, Feldman LJ, Zambryski PC (2000) Auxin and ETTIN in Arabidopsis gynoecium morphogenesis. Development 127:3877–3888 6. Alvarez JP, Goldshmidt A, Efroni I, Bowman JL, Eshed Y (2009) The NGATHA distal organ development genes are essential for style specification in Arabidopsis. Plant Cell 21:1373– 1393 7. Staldal V, Sohlberg JJ, Eklund DM, Ljung K, Sundberg E (2008) Auxin can act independently of CRC, LUG, SEU, SPT and STY1 in style development but not apical-basal patterning of the Arabidopsis gynoecium. New Phytol 180:798–808 8. Liu Q, Luo L, Zheng L (2018) Lignins: biosynthesis and biological functions in plants. Int J Mol Sci 19(2):335 9. Ferra´ndiz C, Liljegren SJ, Yanofsky MF (2000) Negative regulation of the SHATTERPROOF genes by FRUITFULL during Arabidopsis fruit development. Science 289:436–438 10. Liljegren SJ, Roeder AHK, Kempin SA, Gremski K, Østergaard L, Guimil S, Reyes DK, Yanofsky MF (2004) Control of fruit patterning in Arabidopsis by INDEHISCENT. Cell 116:843–853 11. Spence J, Vercher Y, Gates P, Harris N (1996) “Pod shatter” in Arabidopsis thaliana, Brassica napus and B. juncea. J Microsc 181:195–203 12. Roeder AHK, Yanofsky MF (2005) Fruit development in Arabidopsis. Arabidopsis Book 52(1):1–50 13. Smyth DR, Bowman JL, Meyerowitz EM (1990) Early flower development in Arabidopsis. Plant Cell 2:755–767 14. Heisler MG, Ohno C, Das P, Sieber P, Reddy GV, Long JA, Meyerowitz EM (2005) Patterns of auxin transport and gene expression during

primordium development revealed by live imaging of the Arabidopsis inflorescence meristem. Curr Biol 15(21):1899–1911 15. Marsch-Martinez N, Ramos-Cruz D, Irepan Reyes-Olalde J, Lozano-Sotomayor P, Zuniga-Mayo VM, de Folter S (2012) The role of cytokinin during Arabidopsis gynoecia and fruit morphogenesis and patterning. Plant J 72(2):222–234 16. Mu¨ller B, Sheen J (2008) Cytokinin and auxin interaction in root stem-cell specification during early embryogenesis. Nature 453:1094– 1097 17. Baum SF, Eshed Y, Bowman JL (2001) The Arabidopsis nectary is an ABC-independent floral structure. Development 128(22): 4657–4667 18. Savidge B, Rounsley SD, Yanofsky MF (1995) Temporal relationship between the transcription of two Arabidopsis MADS box genes and the floral organ identity genes. Plant Cell 7: 721–733 19. Girin T, Paicu T, Stephenson P, Fuentes S, Ko¨rner E, O’Brien M, Sorefan K, Wood TA, Balanza´ V, Ferra´ndiz C, Smyth DR, Ostergaard L (2011) INDEHISCENT and SPATULA interact to specify carpel and valve margin tissue and thus promote seed dispersal in Arabidopsis. Plant Cell 23(10):3641–3653 20. Sorefan K, Girin T, Liljegren SJ, Ljung K, Robles P, Galva´n-Ampudia CS, Offringa R, Friml J, Yanofsky MF, Ostergaard L (2009) A regulated auxin minimum is required for seed dispersal in Arabidopsis. Nature 459(7246): 583–586 21. Gu Q, Ferrandiz C, Yanofsky MF, Martienssen R (1998) The FRUITFULL MADS-box gene mediates cell differentiation during Arabidopsis fruit development. Development 125(8): 1509–1517 22. Hempel FD, Weigel D, Mandel MA, Ditta G, Zambryski PC, Feldman LJ, Yanofsky MF (1997) Floral determination and expression of floral regulatory genes in Arabidopsis. Development 124(19):3845–3853 23. Kuusk S, Sohlberg JJ, Long JA, Fridborg I, Sundberg E (2002) STY1 and STY2 promote the formation of apical tissues during Arabidopsis gynoecium development. Development 129(20):4707–4717 24. Trigueros M, Navarrete-Gomez M, Sato S, Christensen SK, Pelaz S, Weigel D, Yanofsky MF, Ferrandiz C (2009) The NGATHA genes direct style development in the Arabidopsis gynoecium. Plant Cell 21(5):1394–1409

Genetic and Phenotypic Analyses of Carpel Development 25. Gremski K, Ditta G, Yanofsky MF (2007) The HECATE genes regulate female reproductive tract development in Arabidopsis thaliana. Development 134(20):3593–3601 26. Groszmann M, Bylstra Y, Lampugnani ER, Smyth DR (2010) Regulation of tissue-specific expression of SPATULA, a bHLH gene involved in carpel development, seedling germination, and lateral organ growth in Arabidopsis. J Exp Bot 61:1495–1508 27. Fahlgren N, Montgomery TA, Howell MD, Allen E, Dvorak SK, Alexander AL, Carrington JC (2006) Regulation of AUXIN RESPONSE FACTOR3 by TAS3 ta-siRNA affects developmental timing and patterning in Arabidopsis. Curr Biol 16(9):939–944 28. Roe JL, Nemhauser JL, Zambryski PC (1997) TOUSLED participates in apical tissue formation during gynoecium development in Arabidopsis. Plant Cell 9(3):335–353 29. Toriyama K, Thorsness MK, Nasrallah JB, Nasrallah ME (1991) A Brassica S locus gene promoter directs sporophytic expression in the anther tapetum of transgenic Arabidopsis. Dev Biol 143(2):427–431 30. Sessions RA, Zambryski PC (1995) Arabidopsis gynoecium structure in the wild and in ettin mutants. Development 121(5):1519–1532 31. Liljegren SJ, Ditta GS, Eshed Y, Savidge B, Bowman JL, Yanofsky MF (2000) SHATTERPROOF MADS-box genes control seed dispersal in Arabidopsis. Nature 404(6779): 766–770 32. Sundaresan V, Springer P, Volpe T, Haward S, Jones JD, Dean C, Ma H, Martienssen R

259

(1995) Patterns of gene action in plant development revealed by enhancer trap and gene trap transposable elements. Genes Dev 9(14): 1797–1810 33. Johnson-Brousseau SA, McCormick S (2004) A compendium of methods useful for characterizing Arabidopsis pollen mutants and gametophytically-expressed genes. Plant J 39(5):761–775 34. Jiang L, Yang SL, Xie LF, Puah CS, Zhang XQ, Yang WC, Sundaresan V, Ye D (2005) VANGUARD1 encodes a pectin methylesterase that enhances pollen tube growth in the Arabidopsis style and transmitting tract. Plant Cell 17(2):584–596 35. Sohlberg JJ, Myrenas M, Kuusk S, Lagercrantz U, Kowalczyk M, Sandberg G, Sundberg E (2006) STY1 regulates auxin homeostasis and affects apical-basal patterning of the Arabidopsis gynoecium. Plant J 47(1): 112–123 36. Brien TP, Feder N, McCully ME (1964) Polychromatic staining of plant cell walls by toluidine blue O. Protoplasma 59:368–373 37. Mori B, Bellani LM (1996) Differential staining for cellulosic and modified plant cell walls. Biotech Histochem 71:71–72 38. Pradhan-Mitra P, Loque D (2014) Histochemical staining of Arabidopsis thaliana secondary wall elements. J Vis Exp 87:51381 39. Bond J, Donaldson L, Hill S, Hitchcock K (2008) Safranine fluorescent staining of wood cell walls. Biotech Histochem 83(3):161–171

Chapter 11 Genetic and Phenotypic Analysis of Ovule Development in Arabidopsis Dayton C. Bird, Chao Ma, Sara Pinto, Weng Herng Leong, and Matthew R. Tucker Abstract The plant seed is a remarkable structure that represents the single most important energy source in global diets. The stages of reproductive growth preceding seed formation are particularly important since they influence the number, size, and quality of seed produced. The progenitor of the seed is the ovule, a multicellular organ that produces a female gametophyte while maintaining a range of somatic ovule cells to protect the seed and ensure it receives maternal nourishment. Ovule development has been well characterized in Arabidopsis using a range of molecular, genetic, and cytological assays. These can provide insight into the mechanistic basis for ovule development, and opportunities to explore its evolutionary conservation. In this chapter, we describe some of these methods and tools that can be used to investigate early ovule development and cell differentiation. Key words Ovule, Germline, Clearing, Fluorescence, Microdissection

1

Introduction In flowering plants, the ovule is the site of female gametogenesis and the progenitor of the seed. In essence, the most important function of the ovule is to establish a generative and nutritive ampule that can sustain early seed development. The ovule initially supports the production of a single female germline cell, and second, it facilitates development of a single female gametophyte containing the egg cell and central cell [1]. These gametes are subsequently fertilized by two sperm cells delivered from the pollen tube, which cues the downstream events of seed formation. In most angiosperms, the germline is embedded deep within the ovule, surrounded by somatic tissues. In Arabidopsis, overall ovule structure forms as a result of cell divisions and cell expansion in three different domains; the nucellus, chalaza, and funiculus.

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_11, © Springer Science+Business Media, LLC, part of Springer Nature 2023

261

262

Dayton C. Bird et al.

Arabidopsis ovules are separated into these domains along a distalproximal axis [2]. The most proximal domain, the funiculus, develops into a stalk-like organ that facilitates transport of nutrients to the chalaza and the embryo from the maternal plant, providing seed support and orientation [3]. The chalaza, which is positioned centrally in the ovule, functions to produce the maternal integuments and seed coat which protects the embryo [4]. Finally, the nucellus is the site of germline formation and female gamete production [5]. The sexual reproductive pathway in Arabidopsis thaliana ovules (i.e., female germline development) is a highly structured and complex process. Megasporogenesis occurs early in development when the megaspore mother cell (MMC) develops from a sub-epidermal (L2) somatic cell, the archesporial cell, in the nucellus [6]. After MMC meiosis, three of the megaspores degenerate and the surviving cell is always positioned closest to the chalazal domain, that is, it is the most proximal megaspore [7]. The surviving cell, known as the functional megaspore, subsequently initiates megagametogenesis. This involves three rounds of mitosis to ultimately give rise to the mature seven-celled eight-nucleate female gametophyte [8]. These stages are well described in different plant species through the analysis of morphological changes, gene expression patterns, and mutant phenotypes [9–11]. In addition, immunocytological studies have shown that changes in cell wall composition accompany the different stages of germline development [12, 13]. For example, one key marker for megasporogenesis in angiosperms is callose, a β-1,3-glucan polymer, which accumulates in the cell wall of the MMC and megaspores during megasporogenesis [14]. Previous studies have shown that defects in somatic ovule tissues can influence early stages of female gamete development [15–17]. One example is the Arabidopsis WUSCHEL (WUS) gene that is expressed in the nucellar epidermis of the ovule. Despite this restricted expression pattern, wus mutant ovules show defects in adjoining cell types; they fail to produce integuments and abort development at the MMC or one-nucleate female gametophyte stage [18]. Another example is the ARGONAUTE9 (AGO9) gene, which encodes a small RNA-dependent silencing protein that restricts germline potential to only one sub-epidermal cell [8]. These genes confirm that transcription factors and small RNAs are involved in ovule development, but the downstream regulatory targets in ovule tissues remain largely unknown. Various methods have been developed to examine the details of Arabidopsis ovule development, the formation of a germline, and the interaction between different cell types at different stages of development. In the following sections we discuss methods that

Analysis of Ovule Development

263

might be used to examine morphological, compositional, and transcriptomic signatures of ovule development to aid in the characterization of ovule genes and their function.

2

Materials

2.1 Clearing of Ovules for Wholemount Analysis

1. Microfuge tubes, 1.5 mL. 2. Microscope slides, uncoated, 26 × 76 × 1.0 mm. 3. Hoyer’s Clearing Solution: 160 g of chloral hydrate, 100 mL of water, and 50 mL of glycerol. 4. Pointed tweezers, style 5, 114 mm, stainless steel. 5. Sterile hypodermic disposable needles, 17 Gauge × 1.5″, 38 mm. 6. Dissection stereo microscope. 7. Cover glass, 0.13–0.17 mm (No.1 thickness), 18 × 18 mm. 8. Compound microscope, with differential interference contrast (DIC) and Nomarski filter for a ×10, ×20 and/or ×40 objective, with black and white and color imaging cameras, with metal halide or LED fluorescence light source. For example, Axio Imager M2 with HXP 120 metal halide light source (Carl Zeiss Microscopy). 9. Preferred imaging software compatible with compound microscope of choice. For example, Zeiss Efficient Navigation (ZEN) 2 pro imaging software (Carl Zeiss Microscopy). 10. Fixing solution 1: ethanol:acetic acid (3:1).

2.2 Staining of Ovules for Confocal Analysis

1. Microfuge tubes, 1.5 mL.

2.2.1 Aniline Blue Staining for Callose

4. Phosphate Buffered Saline (PBS): 0.9% (w/v) NaCl, 0.0795% (w/v) Na2HPO4, and 0.0144% (w/v) NaH2PO4 in sterile water, pH = 7.4.

2. Microscope slides, uncoated, 26 × 76 × 1.0 mm. 3. Pointed tweezers, style 5, 114 mm, stainless steel.

5. Aniline Blue solution: 0.005% (w/v) Aniline Blue diammonium salt in PBS. 6. Decolorized Aniline Blue (DAB) solution: 0.1% (w/v) in 108 mM K3PO4 solution, pH = 11, filtered through active charcoal powder on filter paper. 7. Compound microscope as per 2.1. 8. Preferred imaging software compatible with compound microscope of choice. For example, Zeiss Efficient Navigation (ZEN) 2 pro imaging software (Carl Zeiss Microscopy). 9. Fixing solution 2: 10% (v/v) glacial acetic acid in ethanol. 10. 8 M NaOH.

264

Dayton C. Bird et al.

2.2.2 Renaissance Staining for Analysis of Cell Morphology

1. SR2200 fixing and staining working solution: 0.1% (w/v) SR2200 Cell Wall Stain [19, 20], 1% (v/v) DMSO, 0.05% (w/v) Triton-X 100, 3.95% (w/v) paraformaldehyde in PBS buffer. 2. Microfuge tubes, 1.5 mL. 3. Microscope slides, uncoated, 26 × 76 × 1.0 mm. 4. Pointed tweezers, style 5, 114 mm, stainless steel. 5. Sterile hypodermic disposable needles, 17 Gauge × 1.5″, 38 mm. 6. Vacuum desiccator and access to running tap or compatible air pump. 7. TDE solution: 20% (v/v) 2,2′-thiodiethanol (TDE) water solution specifically for ovules staged between 1-II and 3-VI [21] (see Note 1). 8. Slow rotary turner or low-speed horizontal shaker. 9. Laser scanning confocal microscope with 405 nm laser and compatible imaging software. For example, Nikon A1R laser scanning confocal microscope using a 405 nm solid state laser for illumination; and NIS-Elements imaging software.

2.3 Embedding and Sectioning Ovule Tissues for Immunolabelling

1. Microfuge tubes, 1.5 mL. 2. Microscope slides, uncoated, 26 × 76 × 1.0 mm. 3. Pointed tweezers, style 5, 114 mm, stainless steel. 4. Phosphate Buffered Saline (PBS): 0.9% (w/v) NaCl, 0.0795% (w/v) Na2HPO4, and 0.0144% (w/v) NaH2PO4 in sterile water, pH = 7.4. 5. 16% EM Grade paraformaldehyde aqueous solution. 6. 25% EM Grade glutaraldehyde aqueous solution. 7. Fixing solution 3: 4% (w/v) paraformaldehyde, 0.25% (w/v) glutaraldehyde, and 4% (w/v) sucrose in PBS. 8. Graded ethanol series in distilled water (70%, 95%, 100%). 9. Hard grade LR White resin, catalyzed. 10. Gelatin capsules, various sizes. 11. Histology Diamond Knife, size 8. 12. Ultramicrotome. For example, Leica EM UC6/UC7 (Leica Microsystems). 13. Polysine-Microscope Adhesion Slides, 25 × 75 × 1 mm. 14. Liquid blocker super PAP pen. 15. L-Glycine solution: 25 mM glycine in PBS. 16. Incubation buffer: 1% (w/v) Bovine Serum Albumin in PBS.

Analysis of Ovule Development

265

17. Primary and secondary antibodies for immunolabelling. For example, JIM13 anti-arabinogalactan-protein (AGP) rat monoclonal primary antibody. LM20 anti-homogalacturonan rat monoclonal secondary antibody (PlantProbes U.K., [22, 23]). 18. Calcofluor White Stain Solution (Sigma-Aldrich, Catalogue No. 18909). 19. AlexaFluor® 488 conjugated secondary antibody. 20. 90% (v/v) glycerol. 21. Compound microscope, as per 2.1. Ensure the appropriate filter set is used for fluorescence of the secondary antibody. 2.4 Crosses to Marker Lines for Assessment of Ovule Cell Identity

1. Plant marker lines for the megaspore mother cell (pKNU: nlsYFP, [10, 24]), the functional megaspore (pFM1:GUS, [25]; pLC2:nlsYFP, [24]), and the nucellar epidermis (pWUS:3xnlsGFP). Plant marker lines can be obtained by contacting the authors. 2. Microscope slides, uncoated, 26 × 76 × 1.0 mm. 3. Pointed tweezers, style 5, 114 mm, stainless steel. 4. Sterile hypodermic disposable needles, 17 Gauge × 1.5″, 38 mm. 5. 10% (v/v) glycerol. 6. 90% (v/v) acetone. 7. Cover glass, 0.13–0.17 mm (No.1 thickness), 18 × 18 mm. 8. Phosphate Buffer: 0.1 M NaH2PO4, 0.1 M Na2HPO4. 9. X-Gluc solution: 50 mM Na2HPO4, 50 mM NaH2PO4, 0.2% (v/v) Triton X-100, 2 mM potassium hexacyanoferrate(II) trihydrate, 2 mM potassium hexacyanoferrate(III), 1 mg/mL X-gluc (5-bromo-4-chloro-3-indolyl β-D-glucuronic cyclohexylammonium salt). 10. Clearing solution: 20% (v/v) lactic acid, 20% (v/v) glycerol in 1× PBS. 11. 90% (v/v) and 70% (v/v) ethanol. 12. Stereo microscope. For example, Stemi 2000-C (Zeiss). 13. Compound microscope as per 2.1 and filter sets 500/535 nm for yellow fluorescent protein (YFP) detection, 470/525 nm for green fluorescent protein (GFP), and 436/480 nm for cyan fluorescent protein (CFP). For example, Axio Imager M2 with HXP 120 metal halide light source (Carl Zeiss Microscopy).

266

Dayton C. Bird et al.

2.5 Laser Dissection of Ovule Tissues for Transcriptomic Analysis

1. 3:1 ethanol:acetic acid with 1 mM 1,4-Dithiothreitol (DTT). 2. Vacuum desiccator and access to running tap or compatible air pump. 3. Ice cold graded ethanol series in distilled water ethanol series (70%, 80%, 90%, 95%, 100%) with 1 mM DTT. 4. Butyl methyl methacrylate (BMM) embedding medium: 80% (v/v) n-butyl methacrylate, 20% (v/v) methyl methacrylate, 0.05% (w/v) benzoin methyl ether, and 1 mM dithiothreitol. Before use, bubble with N2 gas for 15 min, sit for 10 min, and then bubble with N2 again for another 15 min. Can be stored at 4 °C for up to 2 weeks when sealed. 5. Embedding capsules or molds suitable for the size of the material. For example, size 3 BEEM® polyethylene capsules (ProSciTech Pty Ltd., Catalogue No. EMS69910-05). 6. 90 × 14 mm Petri dish. 7. Cryochamber with UV light. 8. Cleaning agent to remove RNase. For example, RNase-Zap. 9. Polyethylene naphthalate (PEN) membrane glass slides. 10. Acetone. 11. Laser Microdissection Microscope. For example, Leica LMD6/LMD7 (Leica Microsystems). 12. Maxwell’s solution: 2 g KOH, 10 mL absolute methanol, and 5 mL propylene oxide [26]. 13. PicoPure™ RNA Isolation Kit. 14. NanoDrop Spectrophotometer or Bioanalyzer. 15. MessageAmp™ II aRNA Amplification Kit. 16. Speed-Vac.

3

Methods

3.1 Clearing of Ovules for Wholemount Analysis

Quick Method For a quick examination of ovule structure, samples are analyzed by clearing with Hoyer’s Solution [26, 27]. 1. Remove whole flowers or inflorescences from plants in growth rooms and transfer to the lab in microfuge tubes. 2. Place individual flowers at the desired stage directly into 30 μL of Hoyer’s Solution on a flat glass slide. 3. Using fine tweezers and a thin needle, separate the carpels from all other floral organs (discard) under a dissection microscope.

Analysis of Ovule Development

267

4. For early stages of ovule development (i.e., Stage 2-I to 2-V [21], around MMC expansion and meiosis) gently slice open the carpels (along both sides) using fine needles, to reveal the ovules and placenta. Carefully tease the ovules away from the placenta or leave them attached depending on the stage. 5. For later stages (i.e., anthesis), gently drag the needle along the placenta to snap the funiculus and release the ovules into the clearing solution. 6. Cover the samples, still in Hoyer’s Solution, with a coverslip and transfer immediately to a microscope. 7. Image samples using differential interference contrast (DIC) with Normarksi optics. 8. Capture images using a compound microscope camera (Fig. 1). Standard Method For a more detailed characterization, ovules are fixed, cleared, and analyzed as previously described [28]. This method preserves nuclear morphology and allows samples to be stored longer before processing. 1. Fix whole Arabidopsis inflorescences in a solution of 3:1 ethanol:acetic acid, overnight at 4 °C. 2. Wash 2× in 70% (v/v) ethanol for 10 min each (samples can be stored indefinitely in 70% ethanol at 4 °C). 3. For subsequent processing, place samples in Hoyer’s solution. 4. Dissect flowers under a stereo microscope and image as described in the preceding text (Fig. 1). 3.2 Staining of Ovules for Confocal Analysis 3.2.1 Aniline Blue Staining for Callose

Quick Callose Staining Method Aniline Blue binds callose, which is heavily deposited in the walls of the MMC before and during meiotic divisions [14, 29]. Hence this stain can be used as a morphological marker for the young germline cells. 1. Collect flowers as described in Subheading 3.1. 2. Using tweezers, place flowers on a glass slide in 20 μL of Aniline Blue solution (50 μg/mL), in PBS. 3. Dissect flowers to reveal the ovules as described in Subheading 3.1. The ovules will become detached and release into the Aniline blue solution, take care not to remove the ovules when removing the needle. 4. Pipette a further 20 μL of Aniline Blue solution onto the sample. 5. Immediately cover the sample with a coverslip and transfer to a fluorescence microscope.

Fig. 1 Clearing of Arabidopsis ovules using Hoyer’s solution. (a) Columbia WT ovule showing a single megaspore mother cell, quick method. (b) ago9–1/ovule showing an extra enlarged nucellus cell, quick method. (c) Columbia WT ovule showing a large megaspore mother cell and integument primordia, standard method. The nuclei are more obvious compared to the quick methods. (d) Columbia WT ovule showing a functional megaspore after megaspore selection, standard method. (e) Columbia WT ovule at anthesis containing a mature female gametophyte, quick method. (f) pAGO5:AGO5dn ovule at anthesis containing an aborted female gametophyte, quick method. afg, aborted female gametophyte; ccn, central cell nucleus; dm, degenerating megaspores; enc, enlarged nucellus cell; fm, functional megaspore; ii, inner integument, mmc, megaspore mother cell, oi, outer integument. Scale bar = 10 μm

Analysis of Ovule Development

269

6. Observe ovules under UV light with a CFP filter (Zeiss Filter set 47: 436/480 nm). Auto-fluorescence can also be used to highlight the ovule outline in the dsRED channel (Zeiss Filter set 43: 545/605 nm) and also capture the DIC channel with Normarksi optics. 7. Capture images using a compound microscope imaging camera and process with appropriate software (Fig. 2). Standard Callose Staining Method Ovules can also be stained with decolorized aniline blue (DAB) following the protocol previously described [30]. 1. Prepare DAB solution, store it in the fridge at 4 °C overnight. 2. Collect Arabidopsis flowers at different developmental stages and fix in 10% (v/v) glacial acetic acid in ethanol, overnight at 4 °C. 3. Place material in 8 M NaOH, overnight at 4 °C, wash three times with water and stain overnight in DAB at 4 °C. 4. Dissect flowers in the DAB solution using a stereo microscope and prepare slides with coverslips as described in Subheading 3.2.1.

3.2.2 Renaissance Staining for Analysis of Cell Morphology

5. Observe ovules as described in Subheading 3.2.1. The protocol is used for the observation and analysis of cell morphology using renaissance staining and confocal microscopy, and has been adapted from the detailed methods described previously [19, 20]. Fixation 1. Prepare enough SR2200 fix and stain working solution in the fume-cupboard, allowing 520 μL for each fixation. 2. Collect individual flowers at desired developmental stages in microfuge tubes. 3. Using tweezers, place flowers on a flat glass slide, and add 20 μL of SR2200 fixing and staining working solution. 4. Dissect the flowers to reveal the ovules by gently making an incision with a needle along the side of the carpels, and remove both ends of the pistil. 5. Transfer the open pistil with ovules to a tube with 500 μL of SR2200 fixing and staining working solution. Multiple pistils or siliques can be collected in a single tube. Alternatively, a 10 mL Falcon tube and microporous capsules can be used for material transfer, staining, and clearing. Ensure the capsules are sufficiently submerged. 6. Incubate the tubes at room temperature in a vacuum desiccator under gentle water vacuum for 30 min.

270

Dayton C. Bird et al.

Fig. 2 Aniline blue staining of Arabidopsis ovules to identify callose deposition in cell walls. (a, b) Columbia WT ovule showing a single megaspore mother cell surrounded by punctate callose deposits, quick method. (c, d) Columbia WT ovule showing a megaspore tetrad with callose deposition between megaspores, quick method. (e, f) Columbia WT ovule after megaspore selection callose in the vicinity of degenerated megaspores, quick method. (g) Columbia WT ovule containing a megaspore tetrad, standard method. (h) Columbia WT ovule after megaspore selection, standard method. fm, functional megaspore; mmc, megaspore mother cell; mt, megaspore tetrad. Callose fluorescence is shown in green while autofluorescence in shown in red. Differential contrast (DIC) microscopy was used to observe the cellular structure. Scale bar = 20 μm

Analysis of Ovule Development

271

7. Gently release the vacuum and fix material overnight at 4 °C. 8. Remove the fixing and staining solution and briefly wash the sample with 1 mL of deionized water. Clearing 1. Prepare a 20% (v/v) TDE solution (see Note 1). 2. Work in the fume hood and transfer samples carefully to tubes containing TDE solution. Ensure careful handling to retain each sample’s structural integrity. Ovules can be transferred via pipetting with cut-off tips. Make sure to transfer ovules in as little volume of the SR2200 fixing and staining solution as possible. Whole siliques can be transferred with tweezers or needles. 3. Agitate tissues in the TDE solution on a slow rotary turner or on a low-speed horizontal shaker for 10 min. 4. Once cleared, transfer tissues to a microscope slide and cover with a cover glass. Assembled slides can be stored at room temperature temporarily before microscopic analysis. Confocal Microscopy 1. Renaissance stained ovules are imaged using a laser scanning confocal microscope using a 405 nm solid state laser for illumination. Images were captured using NIS-Elements software and may be processed with compatible imaging software [13, 31] (Figs. 3 and 4) (see Note 2). 3.3 Embedding and Sectioning Ovule Tissues for Immunolabelling

Immunolabelling can be used to locate proteins and/or cell wall components depending on the availability of antibodies and the embedding medium. The following LR-white protocol is designed for the location of cell wall components in thin sections and is not

Fig. 3 Renaissance staining of pre-meiotic Arabidopsis ovules to assess cell morphology. (a) 3D projection of a Columbia WT ovule showing a single megaspore mother cell. (I–VI) Sequential optical sections of the ovule shown in (a) to reveal details of cell shape, size, and morphology. Scale bar = 20 μm in (a) and 10 μm in I. Panels I–VI are shown to scale

272

Dayton C. Bird et al.

Fig. 4 Renaissance staining of post-meiotic Arabidopsis ovules to assess cell morphology. (a) 3D projection of a Columbia WT ovule at stage FG4. (I–VI) Sequential optical sections of the ovule shown in (a) to reveal details of integument and gametophyte cell shape, size, and morphology. Scale bar = 20 μm in (a) and 20 μm in I. Panels I–VI are shown to scale

suitable for protein localization. If this is required, the BMM protocol (described in the following text for laser microdissection) is suitable. Fixation of A. thaliana Flowers 1. Collect individual flowers at desired developmental stages using tweezers and place in Eppendorf tubes with Fixing solution 3. 2. Apply vacuum treatment as per Subheading 3.2.2 for 20 min. 3. Remove from the vacuum and let flowers fix for 4–8 h or overnight at 4 °C. 4. Wash flowers with PBS for 10 min, and repeat twice more. 5. Dehydrate in an ethanol series from 70% (v/v) to 100% dehydrated ethanol, with three washes at each ethanol step, 20 min each. 6. Incubate samples in 50% (v/v) LR White resin in dehydrated ethanol, for 8 h or overnight at 4 °C [32]. 7. Wash flowers with LR White resin for 8 h, and repeat twice more. 8. Transfer samples to gelatin capsules and fill capsules with extra resin to the brim. Cover capsules with lids to exclude air. 9. Polymerize resin capsules at 60 °C for 3 days (see Note 3). 10. Store samples at room temperature until required for sectioning. Immunolabelling with Cell Wall Antibodies and Fluorophores 1. Section samples to 1 μm using a diamond knife and an Ultramicrotome.

Analysis of Ovule Development

273

2. Place sections on poly-lysine coated glass slides (see Note 4), 5–7 on each slide, on water droplets and incubate at 70 °C from 1 h to overnight for adhesion. 3. Circle individual sections with a PAP pen on slides. Work via pipetting to dispense and remove liquids onto circled samples. 4. Rehydrate samples in PBS for 7 min. 5. Incubate with 25 mM glycine in PBS for 20 min. 6. Wash samples with Incubation Buffer (IB) for 15 min, and repeat twice more. 7. Incubate samples with the appropriate primary antibody (1: 100 dilution in IB) on the slides in a humid chamber, at room temperature for 1 h. Be sure to include negative control slides to be incubated with the IB only. 8. Wash samples with IB every 15 min (three times) and incubate samples with the secondary antibody (1:200 dilution in IB), in a humid dark chamber, at room temperature for 2 h. 9. Wash samples with IB for 15 min (once) and then with sterile water for 10 min (twice). 10. Stain samples with 0.1% (w/v) calcofluor white stain for 90 s and wash with sterile water three times. 11. Mount slides in 90% (v/v) glycerol and observe under a fluorescent light microscope with appropriate filters for the secondary antibody, for example, 365/445 nm for calcofluor, 545/605 nm for AlexaFluor® 555, and 470/525 nm for AlexaFluor® 488 [33]. 12. Capture images using a compound microscope camera and process with compatible imaging software (Fig. 5). 3.4 Crosses to Marker Lines for Assessment of Ovule Cell Identity

Robust marker lines for the megaspore mother cell (pKNU: nlsYFP; pKNU:erCFP, [10, 24]), the functional megaspore (pFM1:GUS, [34]; pLC2:nlsYFP, [24]), and the nucellar epidermis (pWUS:3xnlsGFP; pWUS:nlsYFP, [25]) are useful when assessing changes in identity of ovule cells (see Note 5). Plant Material and Growth Conditions 1. Sow seeds into moist soil and incubate at 4 °C (in the dark) for 2 days to stratify. 2. Transfer trays to a growth chamber under long-day conditions: 16 h light at 23 °C and 8 h dark at 18 °C. The light intensity was set at 180 μmol. Controlled Crosses 1. Emasculate closed flowers at stages 11–12 [35]. In our growth conditions this can involve removing all floral organs (other than the carpel), or simply by removing the anthers.

274

Dayton C. Bird et al.

Fig. 5 Immunolabelling of Arabidopsis flower tissue sections with cell wall-related antibodies. (a–d) Serial sections of an Arabidopsis stage 12 flower immunolabelled with (a) secondary antibody (negative control), (b) JIM13 (2.31) to label arabinogalactan protein epitopes, (c) LM19 to label homogalacturonan epitopes, (d) LM20 (2.31) to identify methylesterified homogalacturonan epitopes. (e) Post-meiotic ovule labelled with anticallose antibody, highlighting several megaspores in the megaspore tetrad. (f) Post-meiotic ovule containing a developing functional megaspore/female gametophyte labelled with JIM13. an, anthers; ov, ovules. Scale bar = 200 μm in (a)–(d), 15 μm in (e) and 20 μm in (f)

2. Immediately hand pollinate with dehiscent anthers from the donor plant (see Notes 6 and 7). 3. Retain seed from both parents for future comparisons. Preparation of Plant Material for Fluorescence Microscopy (Quick Method) 1. To check fluorescent signal of the marker lines and marker line crosses, collect individual flowers at an appropriate developmental stage (e.g., Stage 2-II for pKNU markers, [21]). 2. Place flowers on a glass slide under a stereomicroscope and carefully remove the carpel using needles and tweezers as described in Subheading 3.2. 3. Mount the carpels in 10% (v/v) glycerol.

Analysis of Ovule Development

275

Fig. 6 Marker genes for identification of ovule cells before and after meiosis using fluorescence microscopy. (a) Ovule expressing pKNU:erCFP, highlighting a megaspore mother cell prior to meiosis. (b) Ovule expressing pWUS:nlsYFP marking the nucellus after tetrad formation. (c) Ovule expressing pAGO5: erYFP, marking the nucellus and inner integument. (d) pLC2:nlsYFP ovule showing signal in the nucleus of the one-celled-gametophyte. mmc, megaspore mother cell. Scale bar = 20 μm

4. Gently slice open the carpels and scrape out the ovules before applying a 20 × 20 mm coverslip. Gently squashing can help to release more ovules from the carpels. 5. For fluorescent markers, immediately observe the ovules using a fluorescent light microscope equipped with a differential interference contrast (DIC) prism for bright field images, and UV light for fluorescence detection (500/535 nm for YFP detection, 470/525 for GFP, and 436/480 for CFP). 6. Capture images using a compound microscope camera and process with compatible imaging software (Fig. 6).

276

Dayton C. Bird et al.

GUS Staining Assay 1. The GUS assay is performed as described by Liljegren et al. [36]. Arabidopsis inflorescences are fixed in 90% (v/v) acetone for 2 h, at -20 °C. 2. The inflorescences are washed two times with phosphate buffer. 3. Incubate inflorescences in X-gluc solution overnight (see Note 8), at 37 °C. 4. Wash inflorescences in 90% (v/v) ethanol for 10 min and place in 70% (v/v) ethanol until visualization. 5. The flowers are placed in clearing solution and dissected under a stereo microscope using hypodermic needles. 6. Image samples using differential interference contrast (DIC) with Normarksi optics. 7. Capture images using a compound microscope camera. 3.5 Laser Dissection of Ovule Tissues for Transcriptomic Analysis

Laser capture microdissection (LCM) allows specific tissue types to be harvested and analyzed by quantitative PCR or global transcriptional profiling. Before attempting, it appears to be a somewhat daunting technique that requires specialized infrastructure and skill. However, results clearly demonstrate that LCM provides a robust method to generate global spatial information regarding mRNA abundance. Fixation, Embedding, and Sectioning For laser microdissection, the protocol described in Okada et al. [37] and Tucker et al. [24] was adapted. 1. Collect staged flowers into vials containing an ice-cold mixture of 3:1 ethanol:acetic acid with 1 mM 1,4-Dithiothreitol (DTT). 2. Apply a vacuum of -70 kPa for 3 × 5 min, with a slow release in between each step. 3. Store samples at 4 °C overnight, then transfer into 70% ethanol and store at -20 °C until infiltration/embedding. 4. Dehydrate through an ethanol series (70%, 80%, 90%, 95%, 100%) with gentle agitation for at least 30 min per solution. Prepare each ethanol solution with 1 mM DTT and keep ice-cold. 5. In a fume cupboard, infiltrate samples with BMM (see Note 9) resin through three BMM:ethanol stages: (1) 2–4 h in 1:3, (2) 2–4 h in 1:1, and (3) 2–4 h in 3:1 ratios. 6. Transfer samples to pure BMM and infiltrate overnight. Repeat once.

Analysis of Ovule Development

277

7. Still in a fume cupboard, transfer infiltrated samples into BEEM capsules containing BMM using tweezers. 8. Fill the capsule to the brim and quickly close the lid to exclude any air bubbles. 9. Place BEEM capsules upside-down on a petri dish such that the flowers rest on the flat lid and are easier to section. 10. Place petri dish in a CryoChamber under UV light at -20 °C for 5 days to allow polymerization. Up to 1/3 of samples may be lost due to bubbles inhibiting polymerization near the tissue of interest if air is not excluded correctly. 11. Serially section samples at 4–5 μm, using an ultramicrotome with an 8 mm glass knife. 12. Using RNase-Zap treated tweezers, place sections onto diethyl pyrocarbonate (DEPC)-treated water droplets on PEN-membrane slides. 13. Evaporate water using a 42 °C slide warmer. Laser Microdissection 1. Prior to laser capture, remove BMM resin from tissue sections by gently rinsing slides in 100% acetone for 10 min (two to three times). 2. Using a Laser Microdissection Microscope, collect selected cells/tissues into specified 0.5 mL tubes. For small cells such as the meiotic tetrad or MMC, we aim to collect at least 100 cells (individual sections) for each replicate (see Note 10) (Fig. 7). 3. Add 8 μL of DEPC H2O and store at -80 °C until RNA extraction. RNA Extraction, Amplification, and cDNA Preparation from Laser-Dissected Samples 1. Extract total RNA from each laser-dissected tissue sample using PicoPure™ RNA Isolation Kit. RNA is eluted in a final volume of 11 μL. 2. Assess the concentration and integrity 1 μL of the resulting RNA using a NanoDrop One or a bioanalyzer. 3. Amplify the RNA twice using an RNA amplification kit. The kit utilizes 10 μL of RNA for two rounds of linear amplification, each of which involves: First Strand cDNA synthesis, Second Strand cDNA synthesis, cDNA purification, in vitro transcription, and aRNA purification. 4. The first round of amplification results in 100 μL of aRNA. The concentration and integrity of 1 μL can be assessed using a NanoDrop One (see Note 11).

278

Dayton C. Bird et al.

Fig. 7 Laser microdissection of Arabidopsis ovule tissues and assessment of amplified RNA. (a) Stage 12 flower on PEN slide. (b) Empty collection cap of 0.5 mL tube. (c) Collection cap after collection of >100 nucellus samples. (d) Collection cap after collection of >100 whole ovules. (e) Quality of Total RNA compared to amplified RNA (aRNA) after assessment on a bioanalyzer. (f) Relative expression of control genes in laser captured and fresh RNA, as assessed by semi-quantitative RT-PCR. Genes include INO (INNER NO OUTER), ANT (AINTEGUMENTA), WUS (WUSCHEL), AGO10 (ARGONAUTE10), PDF2 (PROTODERMAL GROWTH FACTOR 2). FG, female gametophyte; OV, whole ovule; NUC, nucellus; YB, young buds; OF, open flower; SI, siliques

5. For most samples, the remaining 99 μL should be concentrated by speed-vac to a volume less than or equal to 10 μL and used in a second round of RNA amplification following a similar protocol to the first. 6. The resulting 100 μL of aRNA should be assessed by bioanalyzer before assessing by qRT-PCR or RNA sequencing.

4

Notes 1. Concentration varies depending on tissue types. Use 20% TDE for young ovules, increase the concentration for mature ovules. 2. This method provides images that are ideally suited for cell volume and dimension analysis using the LithoGraphX software package [13, 31], or a compatible preferred software package. 3. We have also used samples embedded in (1) Epon812, (2) paraffin, or (3) BMM for cell wall immunolabelling. The resin for these samples needs to be removed before starting the protocol using either (1) Maxwell’s solution, (2) Histochoice®, or (3) 100% acetone, respectively.

Analysis of Ovule Development

279

4. Regular glass slides can be used, however extra care should be taken during solution changes. 5. We have used multiple fluorophores and localization tags in the developing ovule. By far the most stable and reliable marker has been the triple nuclear-localized VENUS marker, while endoplasmic reticulum localized CFP is the most labile. Recent unpublished data suggest that mStrawberry-based markers are also an excellent choice. 6. Successful pollination will be apparent within 2 days through elongation of the silique. 7. All primary lines are maintained as homozygous stocks but staining or fluorescent protein expression should be confirmed before crossing. 8. When expression is low, incubation in the X-Gluc solution can be extended further to allow for visualization of the stain. 9. We have also successfully used Technovit 9100 for embedding, sectioning, and laser microdissection. This resin has the advantage that a UV chamber is not required for polymerization. We have also successfully tested the method on a range of plant species. 10. It is typical to have a minimum of three replicates for laserdissected tissue samples for RNA analysis. 11. For some samples, one round of amplification may be sufficient for quantitative real-time PCR (qRT-PCR) or RNA sequencing. References 1. Pinto SC, Mendes MA, Coimbra S, Tucker MR (2019) Revisiting the female germline and its expanding toolbox. Trends Plant Sci 24:455– 467. https://doi.org/10.1016/j.tplants. 2019.02.003 2. Galbiati F, Sinha Roy D, Simonini S et al (2013) An integrative model of the control of ovule primordia formation. Plant J 76:446– 455. https://doi.org/10.1111/tpj.12309 3. Khan D, Millar JL, Girard IJ et al (2015) Transcriptome atlas of the Arabidopsis funiculus--a study of maternal seed subregions. Plant J 82: 41–53. https://doi.org/10.1111/tpj.12790 4. Creff A, Brocard L, Ingram G (2015) A mechanically sensitive cell layer regulates the physical properties of the Arabidopsis seed coat. Nat Commun 6:6382. https://doi.org/ 10.1038/ncomms7382 5. Wilkinson LG, Tucker MR (2017) An optimised clearing protocol for the quantitative assessment of sub-epidermal ovule tissues

within whole cereal pistils. Plant Methods 13: 67. https://doi.org/10.1186/s13007-0170217-z 6. Schmidt A, Wuest SE, Vijverberg K et al (2011) Transcriptome analysis of the Arabidopsis megaspore mother cell uncovers the importance of RNA helicases for plant germline development. PLoS Biol 9:e1001155. https://doi.org/10.1371/journal.pbio. 1001155 7. Drews GN, Koltunow AM (2011) The female gametophyte. Arabidopsis Book 9:e0155. https://doi.org/10.1199/tab.0155 8. Olmedo-Monfil V, Duran-Figueroa N, Arteaga-Vazquez M et al (2010) Control of female gamete formation by a small RNA pathway in Arabidopsis. Nature 464:628–632. https://doi.org/10.1038/nature08828 9. Musial K, Koscinska-Pajak M, Antolec R, Joachimiak AJ (2015) Deposition of callose in young ovules of two Taraxacum species varying

280

Dayton C. Bird et al.

in the mode of reproduction. Protoplasma 252:135–144. https://doi.org/10.1007/ s00709-014-0654-8 10. Payne T, Johnson SD, Koltunow AM (2004) KNUCKLES (KNU) encodes a C2H2 zincfinger protein that regulates development of basal pattern elements of the Arabidopsis gynoecium. Development 131:3737–3749. https://doi.org/10.1242/dev.01216 11. Rinne PL, van der Schoot C (1998) Symplasmic fields in the tunica of the shoot apical meristem coordinate morphogenetic events. Development 125:1477–1485 12. Coimbra S, Almeida J, Junqueira V et al (2007) Arabinogalactan proteins as molecular markers in Arabidopsis thaliana sexual reproduction. J Exp Bot 58:4027–4035. https://doi.org/10. 1093/jxb/erm259 13. Lora J, Herrero M, Tucker MR, Hormaza JI (2017) The transition from somatic to germline identity shows conserved and specialized features during angiosperm evolution. New Phytol 216:495–509. https://doi.org/10. 1111/nph.14330 14. Reiser L, Fischer RL (1993) The ovule and the embryo sac. Plant Cell 5:1291–1301. https:// doi.org/10.1105/tpc.5.10.1291 15. Bencivenga S, Colombo L, Masiero S (2011) Cross talk between the sporophyte and the megagametophyte during ovule development. Sex Plant Reprod 24:113–121. https://doi. org/10.1007/s00497-011-0162-3 16. Chevalier E, Loubert-Hudon A, Zimmerman EL, Matton DP (2011) Cell-cell communication and signalling pathways within the ovule: from its inception to fertilization. New Phytol 192:13–28. https://doi.org/10.1111/j. 1469-8137.2011.03836.x 17. Robinson-Beers K, Pruitt RE, Gasser CS (1992) Ovule development in wild-type Arabidopsis and two female-sterile mutants. Plant Cell 4:1237–1249. https://doi.org/10. 1105/tpc.4.10.1237 18. Gross-Hardt R, Lenhard M, Laux T (2002) WUSCHEL signaling functions in interregional communication during Arabidopsis ovule development. Genes Dev 16:1129– 1138. https://doi.org/10.1101/gad.225202 19. Musielak TJ, Schenkel L, Kolb M et al (2015) A simple and versatile cell wall staining protocol to study plant reproduction. Plant Reprod 28:161–169. https://doi.org/10.1007/ s00497-015-0267-1 20. Slane D, Bu¨rgel P, Bayer M (2017) Staining and clearing of Arabidopsis reproductive tissue for imaging of fluorescent proteins. In: Schmidt A (ed) Plant germline development:

methods and protocols. Springer, New York, pp 87–94 21. Schneitz K, Hu¨lskamp M, Pruitt RE (1995) Wild-type ovule development in Arabidopsis thaliana: a light microscope study of cleared whole-mount tissue. Plant J 7:731–749. https://doi.org/10.1046/j.1365-313X. 1995.07050731.x 22. Knox JP, Linstead PJ, Cooper JPC, Roberts K (1991) Developmentally regulated epitopes of cell surface arabinogalactan proteins and their relation to root tissue pattern formation. Plant J 1:317–326. https://doi.org/10.1046/j. 1365-313X.1991.t01-9-00999.x 23. Mayer KFX, Rogers J, Dolezel J et al (2014) A chromosome-based draft sequence of the hexaploid bread wheat (Triticum aestivum) genome. Science 345:1251788. https://doi. org/10.1126/science.1251788 24. Tucker MR, Okada T, Hu Y et al (2012) Somatic small RNA pathways promote the mitotic events of megagametogenesis during female reproductive development in Arabidopsis. Development 139:1399–1404. https:// doi.org/10.1242/dev.075390 25. Tucker MR, Hinze A, Tucker EJ et al (2008) Vascular signalling mediated by ZWILLE potentiates WUSCHEL function during shoot meristem stem cell development in the Arabidopsis embryo. Development 135:2839– 2843. https://doi.org/10.1242/dev.023648 26. Vidal S, Lombardero M, Sanchez P et al (1995) An easy method for the removal of Epon resin from semi-thin sections. Application of the avidin-biotin technique. Histochem J 27:204– 209 27. Werner D, Gerlitz N, Stadler R (2011) A dual switch in phloem unloading during ovule development in Arabidopsis. Protoplasma 248:225–235. https://doi.org/10.1007/ s00709-010-0223-8 28. Brambilla V, Battaglia R, Colombo M et al (2007) Genetic and molecular interactions between BELL1 and MADS box factors support ovule development in Arabidopsis. Plant Cell 19:2544–2556. https://doi.org/10. 1105/tpc.107.051797 29. Rodkiewicz B (1970) Callose in cell walls during megasporogenesis in angiosperms. Planta 93:39–47 30. Pereira AM, Lopes AL, Coimbra S (2016) JAGGER, an AGP essential for persistent synergid degeneration and polytubey block in Arabidopsis. Plant Signal Behav 11:e1209616. https://doi.org/10.1080/15592324.2016. 1209616

Analysis of Ovule Development 31. Burton RA, Collins HM, Kibble NAJ et al (2011) Over-expression of specific HvCslF cellulose synthase-like genes in transgenic barley increases the levels of cell wall (1,3;1,4)-β-dglucans and alters their fine structure. Plant Biotechnol J 9:117–135. https://doi.org/10. 1111/j.1467-7652.2010.00532.x 32. Barbier de Reuille P, Routier-Kierzkowska A-L, Kierzkowski D et al (2015) MorphoGraphX: a platform for quantifying morphogenesis in 4D. eLife 4:5864. https://doi.org/10.7554/eLife. 05864 33. Tucker MR, Lou H, Aubert MK et al (2018) Exploring the role of cell wall-related genes and polysaccharides during plant development. Plants (Basel) 7:42. https://doi.org/10. 3390/plants7020042 34. Acosta-Garcı´a G, Vielle-Calzada J-P (2004) A classical arabinogalactan protein is essential for

281

the initiation of female gametogenesis in Arabidopsis. Plant Cell 16:2614–2628. https:// doi.org/10.1105/tpc.104.024588 35. Smyth DR, Bowman JL, Meyerowitz EM (1990) Early flower development in Arabidopsis. Plant Cell 2:755–767. https://doi.org/10. 1105/tpc.2.8.755 36. Liljegren SJ, Ditta GS, Eshed Y et al (2000) SHATTERPROOF MADS-box genes control seed dispersal in Arabidopsis. Nature 404:766– 770. https://doi.org/10.1038/35008089 37. Okada T, Hu Y, Tucker MR et al (2013) Enlarging cells initiating apomixis in Hieracium praealtum transition to an embryo sac program prior to entering mitosis. Plant Physiol 163:216–231. https://doi.org/10. 1104/pp.113.219485

Part III Experimental Systems

Chapter 12 Floral Induction Systems for the Study of Arabidopsis Flower Development Diarmuid O´’Maoile´idigh, Bennett Thomson, and Frank Wellmer Abstract Assessing the molecular changes that occur over the course of flower development is hampered by difficulties in isolating sufficient amounts of floral tissue at specific developmental stages. This is especially problematic when investigating molecular events at early stages of Arabidopsis flower development, as floral buds are minute and are initiated sequentially so that a single flower on an inflorescence is at a given developmental stage. Moreover, young floral buds are hidden by older flowers, which presents an additional challenge for dissection. To circumvent these issues, floral induction systems that allow the simultaneous induction of a large number of flowers on the inflorescence of a single plant were developed. To allow the plant community to avail of the full benefits of these systems, we address some common problems that can be encountered when growing these plants and collecting floral buds for analysis. Key words Floral induction system, Synchronous flowering, Stage-specific flower development, Tissue collection

1

Introduction Over the past two decades, a wealth of information has been gathered on the composition and topology of the gene regulatory networks underlying Arabidopsis flower development (reviewed in: [1, 2]). Much of this progress was made possible by the development of floral induction systems which help to overcome the difficulty in isolating sufficient amounts of floral material at distinct developmental stages for analysis. To establish these systems, we polymerase chain reaction (PCR)-amplified the entire genomic locus of the APETALA1 (AP1) gene, which controls the onset of flower development, and translationally fused the final exon of this gene to the hormone binding domain of the rat glucocorticoid receptor (noted GR) and to the mouse androgen receptor ligandbinding domain (noted AR), respectively [3, 4]. The resulting AP1PRO:AP1-GR and AP1PRO:AP1-AR transgenes mediate the

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_12, © Springer Science+Business Media, LLC, part of Springer Nature 2023

285

286

Diarmuid O´’Maoile´idigh et al.

Fig. 1 The response of AP1PRO:AP1-GR ap1-1 cal-1 plants to treatment with a dexamethasone-containing solution. (a, b) Inflorescence-like meristems 6 d after treatment with (a) a “mock” solution and (b) a solution containing 10 μM dexamethasone. (c, d) Flowers that developed after treatment of inflorescencelike meristems with (c) a “mock” solution and (d) a solution containing 10 μM dexamethasone

expression of the AP1-GR and AP1-AR fusion proteins in a domain that resembles that of the endogenous AP1 gene. The fusion proteins, when expressed in plants, localize to the cytoplasm. Treatment of plants with a steroid hormone leads to the nuclear import and the activation of the AP1 transcription factor. For the construction of flower induction systems, we combined the AP1PRO:AP1GR and AP1PRO:AP1-AR transgenes with plants that are doubly mutant for AP1 and its paralog CAULIFLOWER (CAL). Flower formation in ap1-1 cal-1 plants is blocked for a prolonged period of time and, as a result, these plants undergo a massive overproliferation of inflorescence-like meristems [5] (Fig. 1a). Treatment of AP1PRO:AP1-GR ap1-1 cal-1 plants with a dexamethasone-containing solution, or of AP1PRO:AP1-AR ap1-1 cal-1 plants with a dihydrotestosterone-containing solution, results in the transformation of these meristems into floral buds, which develop in a relatively synchronized manner (Fig. 1b), as was observed for a similar transgenic line that expressed the AP1-GR fusion under the control of the constitutive Cauliflower Mosaic Virus 35S promoter in the ap1-1 cal-1 background [6]. Examination of the flowers from AP1PRO:AP1-GR ap1-1 cal-1 plants after treatment with a dexamethasone-containing solution revealed that

Floral Induction Systems

287

they closely resemble those of wild-type plants (Fig. 1d). In the absence of treatment, plants are almost indistinguishable from non-transgenic ap1-1 cal-1 plants (Fig. 1c). Similar results have been obtained for the AP1PRO:AP1-AR ap1-1 cal-1 line. The development of the floral induction systems has provided the plant community with valuable tools to study the molecular events that bring about flower formation. The lines have been successfully used not only to elucidate the roles of specific floral regulators but also to characterize flower development on a global scale using genome-wide methods (e.g., [6–16]). They can be combined, for example, with inducible promoter systems, mediating the specific expression or perturbation of floral regulators, to determine the composition and architecture of the gene regulatory network underlying flower development [3, 17]. Below, we present a detailed description of how to grow and maintain these plants and how to collect tissue, which can be used for a variety of downstream applications.

2

Materials

2.1 Plant Lines and Growth

Both floral induction systems (i.e., AP1PRO:AP1-GR ap1-1 cal-1 and AP1PRO:AP1-AR ap1-1 cal-1) are available in two versions: the first expresses a phosphinothricin acetyl transferase, which confers resistance to the herbicide glufosinate (“Basta”); the second expresses an aminoglycoside 3′-phosphotransferase, which confers resistance to the antibiotic kanamycin. The availability of lines with two different resistance markers facilitates researchers to transform the floral induction system plants with constructs of their choice, as long as these constructs have resistance markers that are different to the one of the floral induction system used. 1. Autoclaved Note 1).

soil:vermiculite:perlite

(3:1:1)

mixture

(see

2. Pots. 2.2 Reagents for Induction of Flower Formation

1. Dexamethasone. 2. Dihydrotestosterone. 3. Silwet L-77 surfactant. 4. 2-Hydroxypropyl-β-cyclodextrin. 5. Dexamethasone stock solution: 10 mM dexamethasone in 100% ethanol. Store at -20 °C for up to several months. 6. Dihydrotestosterone stock solution: 50 mM dihydrotestosterone in 100% ethanol. Store at -20 °C for up to several months.

288

Diarmuid O´’Maoile´idigh et al.

7. Dexamethasone treatment solution: 10 μM dexamethasone, 0.015% (v/v) Silwet L-77. Add 10 μL of dexamethasone stock solution and 1.5 μL of Silwet L-77 to 10 mL of distilled water (see Note 2). 8. Dihydrotestosterone treatment solution: 500 μM dihydrotestosterone, 2.5 mg/mL 2-Hydroxypropyl-β-cyclodextrin, 0.015% (v/v) Silwet L-77. Add 100 μL of dihydrotestosterone stock solution, 25 mg Hydroxypropyl-β-cyclodextrin, and 1.5 μL of Silwet L-77 to 10 mL of distilled water (see Note 3). 2.3 Reagents for AgrobacteriumMediated Transformation Using the Floral Dip Method

1. Vacuum pump. 2. Vacuum desiccator. 3. Silwet L-77 surfactant. 4. Sucrose. 5. Liquid Luria-Bertani (LB) medium. 6. Transformation solution: 5% (w/v) sucrose, 0.025% (v/v) Silwet L-77.

3

Methods

3.1 Plant Growth and Treatment

1. Sow AP1PRO:AP1-GR ap1-1 cal-1 or AP1PRO:AP1-AR ap1-1 cal-1 seeds on soil (see Note 4) and grow at a temperature between 16 and 20 °C until the plant stems are between 1 and 3 cm in length (see Note 5) (Fig. 2a). 2. Liberally apply the dexamethasone or dihydrotestosterone treatment solutions (see Notes 2 and 3) onto the inflorescence-like meristems (Fig. 2b) using a Pasteur pipette, until the inflorescence-like meristems are drenched and have turned dark green (Fig. 2c). 3. Allow flowers to develop up to the desired floral stage (see Notes 6 and 7) (e.g., Fig. 2d). 4. Collect floral tissue with a fine jeweler’s forceps. Ensure that only floral tissue, and not the underlying stem tissue, is collected. To do this, scrape the top layer of tissue from the inflorescence-like meristem with the forceps as shown in Fig. 2e (see Note 8). 5. If the tissue will be processed for chromatin immunoprecipitation experiments for a transcription factor specifically expressed in flowers, it can be harvested more liberally. However, the harvest of non-meristematic or non-floral tissues may lead to a higher background in such assays.

Floral Induction Systems

289

Fig. 2 Procedure for treating the inflorescence-like meristems with the induction solution. (a) The stems of the plants should have grown to between 1 and 3 cm before treatment with the induction solution. (b, c) The inflorescence-like meristems should be directly treated with the induction solution until they turn dark green as in (c), compared to (b) in the absence of treatment. (d, e) Removal of flower buds from a synchronously flowering inflorescence. (d) Inflorescence meristem 6 d after induction before tissue collection. (e) Floral tissue has been removed from the inflorescence shown in (d) by scraping the surface using jeweler’s forceps 3.2 AgrobacteriumMediated Transformation Using the Floral Dip Method

1. Grow the plants until they make flowers independently of hormone-treatment (see Note 9). 2. Grow a liquid LB culture of Agrobacterium tumefaciens containing the plant transformation vector of choice for approximately 24 h at 28 °C. Pellet the bacteria by centrifugation at ~4500 g. Decant the supernatant and resuspend in the transformation solution.

290

Diarmuid O´’Maoile´idigh et al.

3. Place the transformation solution into the vacuum desiccator and dip the plants into the Agrobacterium-containing solution. Seal the vacuum desiccator with the lid and apply the vacuum (500 mbar for 5 min). 4. Remove the plants and shake off excess Agrobacterium-containing transformation solution. Lay the plants down in a tray by putting the pots on their side, cover trays with plastic wrap, and incubate at 4 °C overnight in the dark (see Note 10). 5. Remove the plastic wrap, straighten up the pots, and move the plants back to the growth room. Wait for the plants to produce seeds and isolate transformants using the appropriate selection protocols.

4

Notes 1. Growing plants on a mixture of sterile compost, perlite, and vermiculite in a ratio of 3:1:1 (or similar) is recommended, as plants of the floral induction system sown on other growth media often suffer from stunted growth. The addition of vermiculite and perlite aerates the compost and prevents soil compaction. Plants of the floral induction system appear to be more susceptible to pathogen infection, so sterilization of the growth medium is required to decrease the risk of disease. To sterilize the medium, we seal a moistened mixture of compost, perlite, and vermiculite in an autoclave bag and autoclave for 1 h at 121 °C. It is also important to let the soil dry slightly before watering, as over-watering impairs growth of these lines and favors fungus growth. 2. Preparation of the treatment solution immediately before use is essential. 3. Preparation of the treatment solution immediately before use is essential. The addition of 2-Hydroxypropyl-β-cyclodextrin helps to keep dihydrotestosterone in solution. 4. Do not overcrowd the plants. A minimum distance of ~2.5 cm should be kept between each seedling. We find it helpful to remove surplus seedlings on two separate occasions: the first round to thin the population of seedlings shortly after germination and the second round after identifying plants that are experiencing some growth difficulties (e.g., stunted growth). Increase the distance between plants if growth appears stunted or plants transition to flowering earlier than expected. 5. Temperature fluctuations can lead to early transitioning of the inflorescence-like meristems to flowering. If you encounter this problem, grow the plants between 16 and 18 °C and avoid temperature fluctuations. Furthermore, rotating pots and trays

Floral Induction Systems

291

to homogenize airflows, light, and temperatures experienced by individual plants will decrease the frequency of asynchronous bolting. 6. After treatment with the induction solution, young floral buds can sometimes grow in an irregular manner (i.e., there can be areas of induced synchronous flower growth and areas of unresponsive inflorescence-like meristematic tissue on a single inflorescence-like meristem). To improve the response of the inflorescence-like meristems to dexamethasone treatment, grow the plants between 16 and 18 °C. When the plants have bolted to between 1 and 3 cm, transfer the plants to 20–22 °C and after approximately 24 h, treat the inflorescence-like meristems with the induction solution. After treatment with a solution containing either dexamethasone or dihydrotestosterone solution, continue growing the plants between 20 and 22 ° C. If the inflorescence-like meristems turn to a pale green or a yellowish color, ensure the Silwet L-77 is completely mixed into the induction solution. Otherwise, decrease the amount of Silwet L-77 used in the induction solution. 7. The number of days after hormone treatment can be correlated with specific stages of flower development (see Table 1) [6, 8, 15]. These correlations were made based on AP1PRO:AP1-GR ap1-1 cal-1 plants being grown in continuous cool-white light at 20 °C. They may vary depending on the exact growth conditions. 8. Tissue collection is routinely performed jeweler’s forceps. When collecting floral processed and used for gene expression take care to remove only the top layer inflorescence-like meristems.

using a sharpened tissue that will be analysis, one must of tissue from the

Table 1 Correlation of days after dexamethasone treatment with approximate stages of flower development. Correlations are based on plants of the floral induction system being grown at 20 °C in continuous light Day

Stage

1

2

2

3

3

4

4

5–6

5

6–7

6

7–9

8

>9

292

Diarmuid O´’Maoile´idigh et al.

9. Do not induce synchronous flowering if the plants are to be transformed with another transgene by floral dip; allow the plants to transition to flower formation independently of hormone treatment. In this case, grow the plants at a temperature between 20 and 22 °C. Transformation should occur when the first flowers have reached maturity—this is best determined by the formation of stigmatic tissue on the tip of gynoecia and/or by the appearance of pollen grains on anthers. 10. After transforming plants, do not incubate at 4 °C for longer than 20 h as this can negatively affect fertility levels.

Acknowledgements Work in the F.W. laboratory is funded by grants from Science Foundation Ireland and the Environmental Protection Agency. References 1. Wils CR, Kaufmann K (2017) Gene-regulatory networks controlling inflorescence and flower development in Arabidopsis thaliana. Biochim Biophys Acta 1860(1):95–105 2. Thomson B, Wellmer F (2019) Molecular regulation of flower development. Curr Top Dev Biol 131:185–210 3. O’Maoileidigh DS, Thomson B, Raganelli A et al (2015) Gene network analysis of Arabidopsis thaliana flower development through dynamic gene perturbations. Plant J 83:344–358 ´ Maoile´idigh DS, Wuest SE, Rae L et al 4. O (2013) Control of reproductive floral organ identity specification in Arabidopsis by the C function regulator AGAMOUS. Plant Cell 25(7):2482–2503 5. Bowman JL, Alvarez J, Weigel D et al (1993) Control of flower development in Arabidopsis thaliana by APETALA1 and interacting genes. Development 119:721–743 6. Wellmer F, Alves-Ferreira M, Dubois A et al (2006) Genome-wide analysis of gene expression during early Arabidopsis flower development. PLoS Genet 2(7):e117 7. Kaufmann K, Wellmer F, Muino JM et al (2010) Orchestration of floral initiation by APETALA1. Science 328(5974):85–89 8. Wuest SE, O’Maoileidigh DS, Rae L et al (2012) Molecular basis for the specification of floral organs by APETALA3 and PISTILLATA. Proc Natl Acad Sci U S A 109(33): 13452–13457 9. Smaczniak C, Immink RG, Muino JM et al (2012) Characterization of MADS-domain transcription factor complexes in Arabidopsis

flower development. Proc Natl Acad Sci U S A 109(5):1560–1565 10. Sun B, Xu Y, Ng KH et al (2009) A timing mechanism for stem cell maintenance and differentiation in the Arabidopsis floral meristem. Genes Dev 23(15):1791–1804 11. Das P, Ito T, Wellmer F et al (2009) Floral stem cell termination involves the direct regulation of AGAMOUS by PERIANTHIA. Development 136(10):1605–1611 12. Jiao Y, Meyerowitz EM (2010) Cell-type specific analysis of translating RNAs in developing flowers reveals new levels of control. Mol Syst Biol 6:419 13. Yan W, Chen D, Schumacher J et al (2019) Dynamic control of enhancer activity drives stage-specific gene expression during flower morphogenesis. Nat Commun 10(1):1705 14. Chen D, Yan W, Fu LY et al (2018) Architecture of gene regulatory networks controlling flower development in Arabidopsis thaliana. Nat Commun 9(1):4534 ´ ’Maoile´idigh DS, Drost H-G et al 15. Ryan PT, O (2015) Patterns of gene expression during Arabidopsis flower development from the time of initiation to maturation. BMC Genomics 16:488 16. Neumann M, Xu X, Smaczniak C et al (2022) A 3D gene expression atlas of the floral meristem based on spatial reconstruction of single nucleus RNA sequencing data. Nat Commun 13(1):2838 17. Zheng B, Thomson B, Wellmer F (2018) A specific knockdown of transcription factor activities in Arabidopsis. Methods Mol Biol 1830:81–92

Chapter 13 Protoplasting and Fluorescence-Activated Cell Sorting of the Shoot Apical Meristem Cell Types G. Venugopala Reddy Abstract The shoot apical meristems (SAMs) are located at the tip of the shoot apex. The SAM harbors stem cells that divide continually to provide cells for developing above-ground organs. Several important developmental events occur in SAMs, such as stem cell maintenance, organ differentiation, and flowering commitment which are under genetic control. The SAM is a collection of specialized cells organized in specific spatial domains. Deciphering the gene regulatory networks, guided by the developmental and environmental signals, in these discrete cell types is essential to decoding the SAM function. Here, I provide updates to the previously published protocols for the protoplasting and subsequent purification through fluorescenceactivated cell sorting (FACS) of SAM cell types (Reddy, Fluorescence activated cell sorting of shoot apical meristem cell types. In: Riechmann JL, Wellmer F (eds) Flower development. Methods in molecular biology, vol 1110. Humana, New York, pp 315–321, 2014), which has provided genome-wide gene expression patterns at a single cell-type resolution. Key words Stem cells, FACS, Protoplasts, Fluorescent reporter, Arabidopsis, Central Zone, Peripheral Zone, Rib-meristem, CLAVATA3, WUSCHEL

1

Introduction The shoot apical meristem (SAM) harbors a set of 35–40 stem cells, as suggested by the gene expression patterns, and they proliferate to support the development of all the above-ground organ systems [1]. The critical developmental events such as maintenance of stemcell identity, organ primordia specification and differentiation, and the temporal control of flowering time are regulated in SAMs [2]. The SAM represents a dynamic and interacting network of functionally distinct cell types. Traditionally, SAMs of higher plants have been divided into distinct domains of cells primarily based on their location within the SAMs and cytological criteria [2]. The central zone (CZ) is at the tip and harbors a set of stem cells. The progeny of stem cells enter into differentiation pathways when they enter the surrounding regions—the flanking peripheral zone (PZ).

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_13, © Springer Science+Business Media, LLC, part of Springer Nature 2023

293

294

G. Venugopala Reddy

Within the PZ, cells differentiate as leaves or flowers in a specified spatiotemporal sequence. Specific cellular behaviors and differentiation events lead to the development of boundary regions and the separation of organs from the SAMs. Thus, stem cell daughters progress through a complex and sequential differentiation process to acquire distinct cellular states/identities [2]. Besides this radial organization, the SAM of higher plants is a complex, multilayered structure. For example, in dicots, the tunica consists of the outer epidermal (L1 layer) and an inner sub-epidermal (L2 layer), whereas the corpus forms a multilayered structure located beneath the L2 layer [3, 4]. Cells of the corpus are also referred to as the Rib-meristem (RM), which differentiates to provide cells for the development of the stem and provide cues to the overlying CZ to specify them as stem cells [5, 6]. Thus, the SAM stem cell niche is a collection of distinct cell types that express different genes and exhibit distinct cell behaviors both along the radial domain and across cell layers that are clonally distinct. Therefore, precise regulation of gene expression dynamics by the cell–cell communication machinery is critical to ensure the timely transition from one cell type to another. Various signals regulate the SAM structure and function. Genetic analysis has implicated signaling systems, transcription factors, hormones, and chromatin regulators in SAM maintenance, organ differentiation, and flowering time [7]. However, how different pathways interact to generate and maintain gene expression and growth dynamics has yet to be understood. The more significant challenge here is understanding how the interconnected network of cells interprets complex spatiotemporal signals to regulate gene expression patterns. This more significant endeavor requires developing and applying gene expression profiling methods at a higher spatial and temporal resolution. In Arabidopsis, the Fluorescence-activated cell sorting (FACS) of protoplasts derived from three distinct cell types, followed by genome-wide expression profiling (microarray analyses), yielded gene expression profiles in different cell types and 15 different zones along the proximo-distal axis of the developmental root gradient [8]. This study revealed that the gene expression patterns are continuous and traverse across the traditionally demarcated anatomical features. A further refinement of this map using additional markers revealed a complex spatiotemporal organization of gene expression patterns, including expression fluctuations along the proximo-distal developmental gradient [9]. Similar to the studies on the Arabidopsis root, SAM cell types’ expression profiling has been developed. The protoplasting and the FACS-mediated cell sorting have been optimized to generate gene expression profiles of cells located both in the radial spatial domains, the CZ, the PZ, organ boundaries, and also in different cell layers—the L1, the L2, and the L3/corpus/the RM

FACS Sorting of Plant Stem Cell Protoplasts

295

[10, 11]. The description of FACS-based isolation of protoplasts from specific cell types of the Arabidopsis SAMs, along with slight modifications in tissue processing by recent studies in Arabidopsis [12] and maize [13], are provided in the following text.

2

Materials

2.1 Protoplasting and Cell Sorting

1. Plant material. 50–60 mg of ap1-1;cal-1 SAM tissue (from about 200 shoots) expressing a fluorescent marker [14]. In a later study [12] ap1-1;cal-1 SAM tissue from about 50 shoots was used. 2. Solution A: 10 mM KCl, 2 mM MgCl2, 2 mM CaCl2, 0.1% (w/v) Bovine Serum Albumin (BSA), 2 mM 2-(N-morpholino Ethane Sulfonic acid) [MES], 600 mM Mannitol. To prepare 25 mL of solution A, mix 250 mL of 1 M KCl, 50 mL of 1 M MgCl2, 50 mL of 1 M CaCl2, add 0.025 g of BSA, 0.0097 g of MES, and 2.75 g of Mannitol, and make up the volume to 25 mL with sterile water (see Note 1). 3. Solution B: 1.5% (w/v) Cellulase, 1% (w/v) Pectolyase, and 1% (w/v) Hemicellulase in solution A. To prepare solution B, dissolve 300 mg of Cellulase, 200 mg of Pectolyase, and 200 mg of Hemicellulase in 20 mL of solution A (see Note 1). 4. Incubator shaker (e.g., New Brunswick Excella E24). 5. 70 mm nylon cell strainer (e.g., BD Falcon). 6. 35 × 10 mm Petri Dish. 7. Fine forceps and scalpels. 8. Temperature Controlled Centrifuge (e.g., Beckman Coulter Allegra X-15R). 9. FACS system (e.g., FACS-Aria Beckton Dickinson).

2.2 Isolation of Total RNA from Sorted Cells

1. RNA extraction kit (e.g., RNeasy Plant Mini Kit from QIAGEN). 2. Glycoblue. 3. 7.5 M ammonium acetate 4. RNase-free water. 5. Agilent 6000 RNA Nano kit. 6. Tabletop microcentrifuge (e.g., Eppendorf Centrifuge 5415 D). 7. Agilent 2100 Bioanalyzer. 8. Nanodrop ND-1000 Spectrophotometer.

296

3

G. Venugopala Reddy

Methods

3.1 Protoplasting and Cell Sorting

This method has been developed based on two previous studies [8, 15]. 1. Grow ap1-1;cal-1 plants expressing a fluorescent cell type marker (see Note 2 for promoters and fluorescent tags) for 4 weeks. 2. Prepare 25 mL of solution A. Adjust the pH to 5.5 with 1 M Tris using HCl. 3. To prepare solution B, dissolve cell wall digesting enzymes cellulase (300 mg), Pectolyase (200 mg), and Hemicellulase (200 mg) in 20 mL of freshly prepared buffer A. Mix vigorously by pipette until the solution becomes transparent/clear (see Note 3). 4. In the meantime, bring the temperature of the shaker to 22 °C. 5. Harvest 50–60 mg of ap1;cal SAMs as quickly as possible (see Note 4 for instructions on tissue harvest). 6. Transfer the tissue to a 70 mm nylon cell strainer and place the strainer in a 35 × 10 mm Petri dish. 7. Add approximately 6–7 mL of solution B and place the petri dish on a shaker set at 120 rpm maintained at 22 °C. At every 10 min interval, gently rinse the SAM surface with a jet of solution B using a 1 mL pipette (see Note 5). 8. Set the centrifuge at 4 °C. 9. If protoplasting is efficient, the protoplasting solution should turn turbid 45 min into the procedure. 10. Pipetting and mixing of the solution may result in partial loss of solution B. Supplement solution B at regular intervals (see Note 5). 11. Upon 1 h and 15 min of treatment, transfer the contents of the Petri dish to a 15 mL Falcon tube and centrifuge at 500 g at 4 ° C for 10 min. Remove the supernatant without disturbing the pellet. 12. Dissolve the pellet in 500–600 μL of solution A by pipetting gently. Once the pellet is dissolved completely, place the tube on ice until the protoplasts are loaded onto the FACS. 13. Separate mGFP5-ER or Ds-Red-N7 (see Note 2 for promoters and fluorescent tags) expressing protoplasts by passing them through a FACS-Aria (Becton Dickinson) fitted with a 100 μM nozzle at a flow rate of 5000–7000 events per second and with a fluid pressure of 35 psi. Select GFP-positive cells by their emission intensity in the green channel (~530 nm) and Ds-Red cells by their emission intensity in the red channel (~610 nm).

FACS Sorting of Plant Stem Cell Protoplasts

297

Collect sorted protoplasts directly into the lysis buffer (Qiagen RLT buffer), mix, and freeze immediately at -80 °C. Fortyfive minutes of sorting should yield about 125,000–150,000 CLV3 (see Note 2 for promoters and fluorescent tags) positive protoplasts. 3.2 Isolation of Total RNA from Sorted Cells

1. Collect fluorescently labeled protoplasts from FACS flow in 500 μL of RLT buffer (QIAGEN RNeasy kit) and adjust the volume of cell suspension up to (3× of original) 3 mL by adding RLT buffer. 2. Add 30 μL of ß-mercaptoethanol to store it at -80 °C or proceed without freezing for RNA isolation. 3. Add 0.5 volume of chilled ethanol and mix gently. 4. Apply 700 μL of the solution to an RNeasy column and centrifuge at 10,600 g in a microcentrifuge for 15 s. Add the solution to the same sample column and repeat the same step until the remaining precipitated solution is loaded onto the column. 5. Add 350 μL of buffer RW1 to the spin column and centrifuge at 10,600 g in a microcentrifuge for 15 s. 6. To perform on-column DNA digestion, add 80 μL of DNA digesting solution to the sample column at RT for 15 min (10 μL of DNase plus 70 μL of RDD buffer). 7. Stop DNA digestion by adding 350 μL of buffer RW1 and centrifuge at 10,600 g in a microcentrifuge for 15 s. 8. Apply 500 μL of RPE buffer provided with the RNeasy kit after supplementing it with ethanol. 9. Centrifuge at 10,600 g in a microcentrifuge for 2 min. 10. Air dry the sample column for 2 min. 11. Elute RNA from the column by applying 50 μL of RNase-free water, centrifuge at 17,900 g for 1 min, and repeat the same step. 12. Precipitate RNA by adding 0.5× volumes of 7.5 M ammonium acetate and 2.5 volumes of ethanol, along with 0.5 μL of Glycoblue (15 mg/mL) overnight at -20 °C. 13. Centrifuge at 17,900 g for 30 min, remove supernatant and wash the RNA pellet with cold 70% ethanol. 14. Air-dry the pellet for 5 min and re-suspend the RNA pellet in 4–6 μL of RNase-free water. 15. Determine RNA integrity using the Agilent 6000 RNA Nano kit on an Agilent 2100 Bioanalyzer according to the manufacturer’s protocol (see Note 6 for expected RNA yield).

298

4

G. Venugopala Reddy

Notes 1. Manufacturer information of chemicals, kits, and enzymes that we used in this protocol: BSA and MES (Fisher), Mannitol (Sigma, USA), Hemicellulase (Sigma, USA Cat # H2125), Cellulase (Yakult, Japan Cat #203039), Pectolyase (Yakult, Japan Cat #202047), RNeasy Plant Mini Kit (QIAGEN). 2. Promoters and fluorescent tags for sorting SAM cell types have been described in earlier studies [10, 11]. In brief, the cell types in the radial domain were isolated using the following reporters: pCLV3::mGFP5-ER (CLAVATA3 promoter driving the expression of endoplasmic reticulum localized mGFP5) is a marker for stem cells/the CZ; pFIL::dsRED-N7 (FILAMENTOUSFLOWER promoter driving the expression of nuclearlocalized dsRED) is a marker for differentiating cells of the organ primordia; the pKAN1::KAN1-GFP (KANADI1 promoter driving the expression of KAN1-GFP translational fusion) was used to label cells in the outer edges of the PZ, referred to as KAN1/outer PZ; the pLAS::LAS-GFP (LATERAL SUPPRESSOR promoter driving the expression of LAS-GFP translational fusion) was used to isolate cells in the organ boundaries, referred to as organ boundaries. The cell types in different cell layers were isolated using the following reporters: pHMG::H2b-mYFP (High Mobility Group gene promoter driving the expression of Histone2B-mYFP translational fusion) marks the centrally located L1 layer cells in SAMs; pATML1::mGFP5-ER (ATML1 promoter driving the expression of endoplasmic reticulum localized mGFP5) marks all L1 layer cells extending into the differentiated organs; pHDG4::H2b-mYFP (HOMEODOMAIN GLABROUS 4 gene promoter driving the expression of Histone2B-mYFP translational fusion) marks the centrally located L2 layer of SAMs; pWUS::mGFP5-ER (WUSCHEL promoter driving the expression of endoplasmic reticulum localized mGFP5) is a marker for the L3 layers/corpus/RM. The vascular cells: the shoot phloem cells are labeled with pS17s::H2b-mYFP (S17s gene promoter driving the expression of Histone2B-mYFP translational fusion), and the shoot xylem cells are labeled with pATHB8::H2b-mYFP (ATHB8 gene promoter driving the expression of Histone2B-mYFP translational fusion). Generally, reporter lines with a higher signal to noise ratio will reduce contamination during sorting and improve the purity of cell types. Therefore, fluorescent proteins with tags that direct them to specific intracellular compartments is recommended.

FACS Sorting of Plant Stem Cell Protoplasts

299

3. Protoplasting solution (solution B) should be clear and without any sediments to obtain optimum protoplast yield. 4. Avoid chopping SAM tissue into smaller bits, as that would increase the debris, interfering with the flow sorting process. 5. Applying a jet of protoplasting solution onto the SAM tissue at regular intervals (approximately every 10 min) by using a 1 mL pipette during the protoplasting process increases the protoplast yield. However, Pipetting and mixing of the solution may result in partial loss of solution B. Supplement solution B at regular intervals. Alternately, the tissue could be dispersed further by lifting the cell strainer and pressing the SAMs gently against the cell strainer using an autoclaved pestle [12]. 6. 100,000 cell type-specific protoplasts usually yield between 200 and 300 ng of total RNA, sufficient for microarray analysis with a two-step amplification. However, with the current improvements in RNA sequencing methods, one may only need a few protoplasts. However, improving the protoplasts yield is still a constant endeavor for downstream metabolomics and proteomics processing.

Acknowledgments My laboratory is currently funded by a National Science Foundation grant (IOS-2055690). References 1. Reddy GV (2008) Live-imaging stem-cell homeostasis in the Arabidopsis shoot apex. Curr Opin Plant Biol 11:88–93 2. Steeves TA, Sussex IM (1989) Patterns in plant development. Shoot apical meristem mutants of Arabidopsis thaliana. Cambridge University Press, New York 3. Satina S, Blakeslee AF, Avery AG (1940) Demonstration of the three germ layers in the shoot apex of Datura by means of induced polyploidy in periclinal chimeras. Am J Bot 27:895–905 4. Tilney-Bassett RAE (1986) Plant chimeras. Edward Arnold, London 5. Baurle I, Laux T (2003) Apical meristems: the plant’s fountain of youth. BioEssays 25:961– 970 6. Yadav RK, Perales M, Gruel J, Girke T, Jo¨nsson H, Reddy GV (2011) WUSCHEL protein movement mediates stem cell homeostasis in the Arabidopsis shoot apex. Genes Dev 25:2025–2030

7. Barton MK (2010) Twenty years on: the inner workings of the shoot apical meristem, developmental dynamo. Dev Biol 341:95–113 8. Birnbaum K, Shasha DE, Wang JY, Jung JW, Lambert GM, Galbraith DW, Benfey PN (2003) A gene expression map of the Arabidopsis root. Science 302:1956–1960 9. Brady SM, Orlando DA, Lee JY, Wang JY, Koch J, Dinneny JR, Mace D, Ohler U, Benfey PN (2007) A high-resolution root spatiotemporal map reveals dominant expression patterns. Science 318:801–806 10. Yadav RK, Girke T, Pasala S, Xie M, Reddy GV (2009) Gene expression map of the Arabidopsis shoot apical meristem stem cell niche. Proc Natl Acad Sci U S A 106:4941–4946 11. Yadav RK, Tevakkoli M, Xie M, Girke G, Reddy GV (2014) A high-resolution gene expression map of the Arabidopsis shoot meristem stem cell niche. Development 141:2735– 2744

300

G. Venugopala Reddy

12. Ram H, Sahadevan S, Gale N, Caggiano MP, Yu X, Ohno C, Heisler MG (2020) An integrated analysis of cell-type specific gene expression reveals genes regulated by REVOLUTA and KANADI1 in the Arabidopsis shoot apical meristem. PLoS Genet 16:e1008661 13. Satterlee JW, Scanlon MJ (2022) Protoplast isolation from undifferentiated maize seedling shoot tissue. In: Wang K, Zhang F (eds) Protoplast technology. Methods in molecular biology, vol 2464. Humana, New York

14. Reddy GV (2014) Fluorescence activated cell sorting of shoot apical meristem cell types. In: Riechmann JL, Wellmer F (eds) Flower development, Methods in molecular biology, vol 1110. Humana, New York, pp 315–321 15. Guangyu C, Conner AJ, Christey MC, Fautrier AG, Field RJ (1997) Protoplast isolation from shoots of asparagus cultures. Int J Plant Sci 158:537–542

Chapter 14 Protoplast Isolation for Plant Single-Cell RNA-seq Shulin Ren and Ying Wang Abstract The growth and development of plants depends on diversified gene expression in different cell types. Compared to traditional bulk RNA sequencing, droplet-based single-cell RNA sequencing (scRNA-seq) allows for transcriptome profiling of individual cells within heterogeneous tissues. scRNA-seq provides a high-resolution atlas of cellular characterization and vastly improves our understandings of the interactions between individual cells and the microenvironment. However, the difficulty in protoplast isolation has limited the application of single-cell sequencing technology in plant research. Here we describe a highefficiency protoplast isolation protocol for scRNA-seq. Key words Single-cell RNA sequencing, Protoplast isolation, Transcriptome

1

Introduction Over the past decade, RNA sequencing (RNA-seq) has become a powerful tool for quantifying gene expression across the transcriptome and played an irreplaceable role in enhancing our understanding of plant developmental processes and responses to environmental changes [1, 2]. Traditional RNA-seq is performed in “bulk”, and the resulting data reflect an average of gene expression patterns across thousands to millions of cells, therefore losing cell heterogeneity information [3]. However, the functionality of complex organisms is the result of an interplay of different cell types and their specific functions [4]. To comprehensively understand and identify the functions of different cell types in complex tissues, a technology that captures gene expression at the level of single cell type or even single cell is required. Prior to the emergence of droplet-based single-cell RNA sequencing (scRNA-seq), single cell analyses mainly relied on laser capture microdissection (LCM) or fluorescence-activated cell sorting (FACS) to collect individual cells [5, 6]. Nevertheless, LCM presents some inevitable limitations in that it is low throughput with only a limited number of cells retrieved, and it is technically challenging due to the difficulty in

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_14, © Springer Science+Business Media, LLC, part of Springer Nature 2023

301

302

Shulin Ren and Ying Wang

identifying specific cell types merely based on their morphologies [7, 8]. Flow cytometry based-FACS enables high-throughput isolation of specific cell populations expressing particular fluorescent markers, but its reliance on appropriate fluorescent marker lineages makes it labor-intensive and limits its application to cell-type lineage studies [6, 9]. More importantly, these techniques profile single cell types, rather than individual single cells. By contrast, dropletbased scRNA-seq has been developed to overcome these drawbacks. scRNA-seq can be used to obtain transcripts at single-cell resolution by isolating single cells in large quantities, capturing transcripts from individual cells, and generating libraries for highthroughput sequencing. Recently, scRNA-seq has been used to profile heterogeneity among distinct tissue types and cell types [10–15]. For instance, scRNA-seq of Arabidopsis root cells has revealed a high-resolution atlas that captures all major cell types in roots at various developmental stages, and even very small cell populations, such as the quiescent center (QC), have been captured and identified [12]. Similarly, scRNA-seq of the vegetative shoot apex also revealed the highly heterogeneous cellular composition and further delineated cell-cycle continuums in the shoot apex at the tissue level [13]. scRNA-seq has also been employed to understand de novo root regeneration at the spatiotemporal resolution [14] and to unveil the roles of stochastic expression in protoplast regeneration [15]. However, protoplast isolation, as the first and rate-limiting step in scRNA-seq practice, due to its relatively long duration and low efficiency, still greatly limits the application of scRNA-seq in plant tissue. Here, we provide a fast and high-efficiency protocol for protoplast isolation for scRNA-seq. The entire procedure can be completed within 3 h. This protocol can also be used in conjugation with assay for transposase-accessible chromatin with highthroughput sequencing (ATAC-seq) to map the chromatin accessibility at the genome-scale [15].

2

Materials

2.1 Plant Growth and Tissue Collection

1. Plant material: Arabidopsis thaliana seeds were grown on 1/2 Murashige and Skoog (MS) medium for 2 weeks with 16 h light and 8 h dark conditions at 22 °C and 75% relative humidity. 2. 1/2 MS medium: 0.22% Murashige and Skoog (MS) basal salt mixture, 1% sucrose, 0.8% agar, adjusted to pH 5.7 with 0.1 M KOH. 3. Forceps (RNase-free). 4. Razor blade (RNase-free). 5. Petri dishes (RNase-free).

Protoplast Isolation for Plant Single-Cell RNA-seq

2.2 Protoplast Isolation

303

1. DiEthyl PyroCarbonate (DEPC)-treated RNase-free water. 2. Wash buffer: 20 mM KCl, 8% mannitol, 0.5% polyvinylpyrrolidone, 20 mM MES (pH = 5.5–5.7). 3. Enzyme solution: 20 mM KCl, 8% Mannitol, 0.5% polyvinylpyrrolidone, 20 mM MES (pH 5.5–5.7). Add driselase to solution (0.1% w/v) and heat at 55 °C for 10 min. Cool to room temperature and add cellulase RS (0.6% w/v) and macerozyme (0.2% w/v). Filter the final enzyme solution through a 0.22 μM syringe filter into a petri dish. 4. W5 solution: 2 mM MES (pH 5.7), 5 mM KCl, 125 mM CaCl2, and 154 mM NaCl. 5. Mannitol solution: 9% mannitol and 0.02–0.1% bovine serum albumin (BSA) (see Note 1). 6. RNase-free tubes and petri dishes. 7. 40 μM and 70 μM cell strainers. 8. 0.22 μM syringe filter. 9. Shaker. 10. Swing-out rotor refrigerated centrifuge. 11. Hemocytometer. 12. Optical microscope.

3

Methods 1. Collect and cut plant tissues into small pieces using a sharp razor blade without tissue crushing at the cutting site (see Note 2). 2. Suspend the minced tissue in wash buffer and filter through 70 μM cell strainers, rinse the tissue gently at least twice with wash buffer. 3. Transfer the strainer into enzyme solution and make sure tissue are submerged nicely (see Note 3). 4. Incubate at 25 °C in the dark on an orbital shaker set at 40 rpm for 1–2 h (see Note 4). 5. Dilute the enzyme solution with an equal volume of W5 solution. 6. Place a 40 μM cell strainer on top of a 50 mL centrifuge tube, filter the enzyme suspension, and keep the tube tilted so that the filtrate flows along the tube wall (see Note 5). 7. Centrifuge the filtered suspension at 100 × g for 4 min at 14 °C (see Note 6).

304

Shulin Ren and Ying Wang

8. Remove the supernatant, add W5 solution into the tube, and re-suspend protoplasts by gentle swirling until protoplasts are completely suspended (see Note 7). 9. Centrifuge at 100 × g for 4 min at 14 °C (see Note 6). 10. Repeat steps 8 and 9 for a total of two washes. 11. Carefully remove the supernatant (see Note 8). 12. Quantify the number of protoplasts with a hemocytometer and resuspend the protoplasts in mannitol solution to a final concentration of approximately 2000 protoplasts/μL (see Note 9). 13. The protoplast suspension is now ready for droplet-based scRNA library preparation.

4

Notes 1. BSA is necessary to stabilize protoplasts and reduce clumping in single-cell suspensions. Normally we recommend to start with the concentration of 0.02% and gradually increase BSA concentration in case cell clumps are still formed at low concentrations [16]. 2. Use a thin, sharp blade to cut tissues quickly and avoid crushing large numbers of cells. 3. For different plant samples, the enzyme concentrations can be appropriately adjusted. Although driselase is generally beneficial for protoplast isolation, be cautious when increasing driselase concentration as the risk of protoplast rupture increases with increasing driselase concentration. 4. Slow and regular shaking is beneficial for the separation of protoplasts from the tissue, and rapid or violent shaking should be avoided from this step. 5. The strainer needs to be rinsed with W5 solution. 6. For Eppendorf® centrifuge 5804R, set acceleration/deceleration (ACC/DEC) rates as 1. 7. Remove the supernatant as much as possible to reduce impurities and do not disturb protoplasts at the bottom. 8. Remove the supernatant as much as possible, because the supernatant contains calcium ions, which may affect sequencing quality in future steps. 9. If necessary, filter the protoplast suspension with a 40 μm cell strainer to remove undigested cell clumps and large debris particles.

Protoplast Isolation for Plant Single-Cell RNA-seq

305

Acknowledgments This work was supported by the grants 2019YFA0903902 and 2022YFE0101100 from National Key R&D Program of China (NKP), the grant 32270345 from National Natural Science Foundation of China (NSFC), the grant 110202001021 (JY-04) from Bureau of National Tobacco, and the Fundamental Research Funds for the Central Universities to YW. References 1. Wang J, Qin Q, Pan JJ, Sun LJ, Sun YF, Xue Y, Song K (2019) Transcriptome analysis in roots and leaves of wheat seedlings in response to low-phosphorus stress. Sci Rep 9:19802 2. Premathilake AT, Ni JB, Shen JQ, Bai SL, Teng YW (2020) Transcriptome analysis provides new insights into the transcriptional regulation of methyl jasmonate-induced flavonoid biosynthesis in pear calli. BMC Plant Biol 20:388 3. Takehisa H, Sato Y, Igarashi M, Abiko T, Antonio BA, Kamatsuki K, Minami H, Namiki N, Inukai Y, Nakazono M, Nagamura Y (2012) Genome-wide transcriptome dissection of the rice root system: implications for developmental and physiological functions. Plant J 69:126– 140 4. Brady SM, Orlando DA, Lee JY, Wang JY, Koch J, Dinneny JR, Mace D, Ohler U, Benfey PN (2007) A high-resolution root spatiotemporal map reveals dominant expression patterns. Science 318:801–806 5. DeCarlo K, Emley A, Dadzie OE, Mahalingam M (2011) Laser capture microdissection: methods and applications. Methods Mol Biol 755:1–15 6. Carter AD, Bonyadi R, Gifford ML (2013) The use of fluorescence-activated cell sorting in studying plant development and environmental responses. Int J Dev Biol 57:545–552 7. Zhan JP, Thakare D, Ma C, Lloyd A, Nixon NM, Arakaki AM, Burnett WJ, Logan KO, Wang DF, Wang XF, Drews GN, Yadegaria R (2015) RNA sequencing of laser-capture microdissected compartments of the maize kernel identifies regulatory modules associated with endosperm cell differentiation. Plant Cell 27:513–531 8. Datta S, Malhotra L, Dickerson R, Chaffee S, Sen CK, Roy S (2015) Laser capture

microdissection: big data from small samples. Histol Histopathol 30:1255–1269 9. Gifford ML, Dean A, Gutierrez RA, Coruzzi GM, Birnbaum KD (2008) Cell-specific nitrogen responses mediate developmental plasticity. PNAS 105:803–808 10. Zhang TQ, Xu ZG, Shang GD, Wang JW (2019) A single-cell RNA sequencing profiles the developmental landscape of Arabidopsis root. Mol Plant 12:648–660 11. Hou ZM, Liu YH, Zhang M, Zhao LH, Jin XY, Liu LP, Su ZX, Cai HY, Qin Y (2021) Highthroughput single-cell transcriptomics reveals the female germline differentiation trajectory in Arabidopsis thaliana. Commun Biol 4:1149 12. Denyer T, Ma XL, Klesen S, Scacchi E, Nieselt K, Timmermans MCP (2019) Spatiotemporal developmental trajectories in the Arabidopsis root revealed using highthroughput single-cell RNA sequencing. Dev Cell 48:840–852 13. Zhang TQ, Chen Y, Wang JW (2021) A singlecell analysis of the Arabidopsis vegetative shoot apex. Dev Cell 56:1056–1074 14. Liu W, Zhang Y, Fang X, Tran S, Zhai N, Yang Z, Guo F, Chen L, Yu J, Ison MS, Zhang T, Sun L, Bian H, Zhang Y, Yang L, Xu L (2022) Transcriptional landscapes of de novo root regeneration from detached Arabidopsis leaves revealed by time-lapse and singlecell RNA sequencing analyses. Plant Commun 3:100306 15. Xu M, Du Q, Tian C, Wang Y, Jiao Y (2021) Stochastic gene expression drives mesophyll protoplast regeneration. Sci Adv 7:eabg8466 16. Yoo SD, Cho YH, Sheen J (2007) Arabidopsis mesophyll protoplasts: a versatile cell system for transient gene expression analysis. Nat Protoc 2:1565–1572

Chapter 15 Plant Nuclei Isolation for Single-Nucleus RNA Sequencing Xu Xin, Fei Du, and Yuling Jiao Abstract Transcriptome profiling has been significantly hampered by the heterogeneity among individual cells within a tissue or an organ. Recent advances in single cell transcriptome profiling have significantly advanced our understanding of the transcriptome. However, plant single-cell RNA sequencing (scRNA-seq) relies on the isolation of protoplasts, which is not only impossible for many cell types but also induces acute wounding responses. To solve these problems, single-nucleus RNA sequencing (snRNA-seq) has been applied to plant research, in which nuclei are isolated and subject to encapsulation and profiling. Compared with scRNAseq, snRNA-seq can be applied to a wider range of tissue types and plant species. Nevertheless, fewer transcripts can be obtained from each nucleus than each protoplast. In this chapter, we describe a detailed and general protocol to prepare nuclei from plant tissues that are ready for subsequent library construction and high-throughput sequencing. Key words Shoot apices, snRNA-seq, Nuclei, Plant

1

Introduction High-throughput transcriptome analyses, such as RNA sequencing (RNA-seq), have broadened our understanding of plant development and response to the environment. However, RNA-seq is traditionally performed using tissues containing different cell types, and therefore generates averaged gene expression profiles from a mixed population of cells. The specificity of individual cells in this population is inevitably obscured [1]. In recent years, advances in single cell omics technologies enabled the transcriptome-wide study of cell heterogeneity [2]. Single-cell omics requires isolation of either individual cells or nuclei as the starting materials for subsequent separation and sequencing, known as single-cell RNA sequencing (scRNA-seq) and singlenucleus RNA sequencing (snRNA-seq), respectively. For plant cells, due to the presence of the cell wall, it is challenging to obtain intact protoplasts of desired cell types for scRNA-seq. To release protoplasts, enzymatic cell wall digestion is employed in plant

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_15, © Springer Science+Business Media, LLC, part of Springer Nature 2023

307

308

Xu Xin et al.

tissues, which may also introduce artifactual stress responses. Due to the differences in cell wall properties, the procedure of protoplast isolation needs to be optimized for distinct tissues and plant species. Moreover, cell encapsulation platforms, such as 10× Genomics, often require cells below a certain diameter (typically less than 40 μm) and larger plant cells cannot be recovered. By contrast, although fewer transcripts are detected by snRNA-seq [3], nucleus acquisition is simpler and the procedure can be adapted to a wider range of tissue types. Compared with isolated live protoplasts, tissues frozen by liquid nitrogen can be used in snRNA-seq to prevent the artifactual transcriptional stress responses [4– 6]. Because of these advantages, snRNA-seq may identify more cell types than scRNA-seq in certain tissue types [6]. In plants, snRNA-seq has been applied to many species and tissue types [7–10]. In this chapter, we present a pipeline for isolating plant nuclei for snRNA-seq. This method has been used for snRNA-seq of tomato shoot apices and is applicable to other plant species and tissue types [10].

2

Materials Prepare all reagents with RNase-free water to prevent RNA degradation (see Note 1). 1. 10 mM MES-KOH: 10 mM MES, adjust pH to 5.4 with 1 M KOH. 2. Nuclei isolation buffer (NIB): 10 mM MES-KOH (pH 5.4), 10 mM NaCl, 10 mM KCl, 2.5 mM Ethylene Diamine Tetraacetic Acid (EDTA), 250 mM sucrose, 0.1 mM spermine, 0.5 mM spermidine, 1 mM dithiothreitol (DTT). The solution is freshly prepared just before use. 3. Protease inhibitor cocktail (e.g., Sigma-Aldrich) (see Note 2). 4. 10% Triton X-100 (see Note 3). 5. 0.4% trypan blue. Filter with a 0.22 μm filter before use. 6. DAPI (4′,6-diamidino-2-phenylindole): prepare a stock at a concentration of 2.5 mg/mL, and make a 1:1000 dilution before use. 7. Dissecting plates: fill 10 cm petri dishes with autoclaved 3% (w/v) agar in water. 8. 0.22 μm polyethersulfone membrane sterile syringe filter (e.g., Merck-Millipore). 9. 1 mL sterile syringe with sharp needle: The sharp syringe needle tip is used as a blade to cut leaves from seedlings. 10. Fine tweezers (sterile). 11. Tissue homogenizer (optional; see Note 4).

Plant Nuclei Isolation for Single-Nucleus RNA Sequencing

309

12. 40 μm cell strainer (e.g., Falcon). 13. Centrifuge with a cooling system and a horizontal rotor. 14. RNase-free 1.5 mL centrifuge tubes. 15. RNase-free 50 mL centrifuge tubes. 16. Stereomicroscope. 17. Orthotopic fluorescence microscope: It is used for examining the number and status of isolated nuclei. 18. Cell counting plate. 19. RNase-free pipette tips. 20. Laminar flow hood. 21. Plant material: 2-week-old tomato (Solanum lycopersicum) cv. M82 seedlings. The seeds are sterilized with 40% (v/v) bleach and geminated on 1/2 MS medium with 3% sucrose and 0.3% (w/v) phytagel (pH 5.8) in culture vessels at 23 °C in long-day conditions (16 h light/8 h dark).

3

Method 1. Precool the centrifuge to 4 °C. Steps 2 and 3 are performed in a laminar flow hood to prevent contamination. 2. Dig a small hole on the dissecting plate using tweezers. 3. Remove large leaves from a tomato seedling with tweezers and cut off the shoot apex together with part of the hypocotyl. Using a stereomicroscope, insert the hypocotyl part into the hole on the dissecting plate with the shoot apex outside the hole. Using a syringe needle tip, sequentially cut and remove smaller leaves until only the first three leaf primordia are left. Harvest tomato shoot apices (shoot apical meristem together with the first three leaf primordia) and freeze immediately in liquid nitrogen before storing at -80 °C until use (see Note 5). 4. Resuspend frozen shoot apices in 10 mL of nuclei isolation buffer (NIB) with protease inhibitor cocktail (0.1%, v/v) in a 50 mL centrifuge tube, and homogenize using a homogenizer at 5000 rpm on ice for 1 min (see Note 4). 5. Lyse cells on ice for 30 min. 6. Filter the homogenate gently throughout a 40 μm cell strainer into a new 50 mL centrifuge tube. Tilt the centrifuge tube to allow the liquid to trickle down the wall. 7. Repeat filtering once. 8. To eliminate chloroplasts, add 10% Triton X-100 dropwise to the solution to a final Triton X-100 concentration of 0.1% (v/v). Check with a light microscope every minute after mixing well, until most chloroplasts are degraded (see Note 3).

310

Xu Xin et al.

9. Centrifuge the nucleus suspension at 1000 g for 5 min at 4 °C (see Note 6). 10. Resuspend the pelleted nuclei in 10 mL of NIB. Centrifuge at 1000 g for 5 min at 4 °C (see Note 6). Gently remove the supernatant. 11. Repeat washing once. 12. Suspend the pelleted nuclei in appropriate amount of NIB. 13. Determine the variability, integrity, and concentration of the nuclei by 0.4% trypan blue staining (1:10 dilution staining) under a light microscope, or by DAPI staining under a fluorescence microscope. The nuclei preparation should have a clumping rate of less than 10% and the fragmentation rate should be less than 20% (see Notes 7–10). 14. Adjust the concentration of nuclei to ~1000 nuclei/μL with NIB, and subject it for encapsulation with the 10× Genomics single cell cassette according to the manufacture’s instruction. Approximately 20,000 nuclei are loaded for encapsulation for one replicate.

4

Notes 1. RNase-free water can be purchased from venders or homemade through diethyl pyrocarbonate (DEPC) treatment. To prepare DEPC-treated water, 0.1% (v/v) DEPC is added into MiliQ water. Mix thoroughly by shaking at 160 rpm overnight at 37 ° C. Autoclave at 121 °C for 20 min to remove DEPC and cool to room temperature before use. 2. The Protease Inhibitor Cocktail from Sigma-Aldrich in DMSO solution for use in tissue culture media (identified with Cat# P1860) is a mixture of protease inhibitors with broad specificity and contains aprotinin, bestatin, leupeptin, E-64, and pepstatin A. 3. As a detergent, Triton X-100 dissolves the lipid in chloroplast membranes and thus makes chloroplasts to degrade. In contrast, nuclear envelopes are not affected by TritonX-100. 4. It is optional to use a tissue homogenizer (e. g., Omni Tissue Homogenizer, PerkinElmer). Homogenized tissue may lead to broken nuclei. Manual cutting could maintain more intact nuclei. Nevertheless, with manual cutting it is difficult to release nuclei evenly and thoroughly from all cells. 5. It takes time to collect enough shoot apices (typically 400 apices for one replicate). Dissected samples can be firstly put in a small boat made by clean tin foil paper that floats on the surface of liquid nitrogen before reaching the adequate amount and storing at -80 °C.

Plant Nuclei Isolation for Single-Nucleus RNA Sequencing

311

6. Use reduced acceleration and deceleration rates. For instance, if using a Sorvall ST1 R Plus centrifuge set both the acceleration and deceleration rates as 2. If using an Eppendorf 5810R or 5804R centrifuge, both acceleration and deceleration rates should be set to 0. 7. To detect DAPI staining, a 405 nm laser is used for excitation and emission is collected at 425–475 nm. 8. Nuclear clumping rate = the number of clumped nuclei/total nuclei (including intact nuclei and clumped nuclei). A nuclei clump is scored as 1 nucleus. 9. Fragmentation rate = the number of fragments/the number of (fragments + nuclei + clumped nuclei). A nuclei clump is scored as 1 nucleus. 10. If there are clumps or debris larger than 50 μm, filtering throughout a 40 μm cell strainer is recommended.

Acknowledgments This work was supported by the CAS Strategic Priority Research Program (grant XDA24020203) and the National Key R&D Program of China (grant 2023YFE0101100). References 1. de Souza N (2010) Single-cell methods. Nat Methods 7:35 2. Mo Y, Jiao Y (2022) Advances and applications of single-cell omics technologies in plant research. Plant J 110:1551–1563 3. Bakken TE, Hodge RD, Miller JA, Yao ZZ, Nguyen TN, Aevermann B, Barkan E, Bertagnolli D, Casper T, Dee N, Garren E, Goldy J, Graybuck LT, Kroll M, Lasken RS, Lathia K, Parry S, Rimorin C, Scheuermann RH, Schork NJ, Shehata SI, Tieu M, Phillips JW, Bernard A, Smith KA, Zeng HK, Lein ES, Tasic B (2018) Single-nucleus and single-cell transcriptomes compared in matched cortical cell types. PLoS One 13:e0209648 4. Denisenko E, Guo BB, Jones M, Hou R, de Kock L, Lassmann T, Poppe D, Clement O, Simmons RK, Lister R, Forrest ARR (2020) Systematic assessment of tissue dissociation and storage biases in single-cell and single-nucleus RNA-seq workflows. Genome Biol 21:130 5. van den Brink SC, Sage F, Vertesy A, Spanjaard B, Peterson-Maduro J, Baron CS, Robin C, van Oudenaarden A (2017) Singlecell sequencing reveals dissociation-induced gene expression in tissue subpopulations. Nat Methods 14:935–936

6. Wu HJ, Kirita Y, Donnelly EL, Humphreys BD (2019) Advantages of single-nucleus over single-cell RNA sequencing of adult kidney: rare cell types and novel cell states revealed in fibrosis. J Am Soc Nephrol 30:23–32 7. Farmer A, Thibivilliers S, Ryu KH, Schiefelbein J, Libault M (2021) Singlenucleus RNA and ATAC sequencing reveals the impact of chromatin accessibility on gene expression in Arabidopsis roots at the singlecell level. Mol Plant 14:372–383 8. Marand AP, Chen Z, Gallavotti A, Schmitz RJ (2021) A cis-regulatory atlas in maize at singlecell resolution. Cell 184:3041–3055.e21 9. Neumann M, Xu X, Smaczniak C, Schumacher J, Yan W, Bluthgen N, Greb T, Jonsson H, Traas J, Kaufmann K, Muino JM (2022) A 3D gene expression atlas of the floral meristem based on spatial reconstruction of single nucleus RNA sequencing data. Nat Commun 13:2838 10. Tian C, Du Q, Xu M, Du F, Jiao Y (2020) Single-nucleus RNA-seq resolves spatiotemporal developmental trajectories in the tomato shoot apex. bioRxiv:2020.09.20.305029. https://doi.org/10.1101/2020.09.20. 305029

Chapter 16 Isolation of Nuclei Tagged in Specific Cell Types (INTACT) in Arabidopsis Ruben M. Benstein, Markus Schmid, and Yuan You Abstract Many functionally distinct plant tissues have relatively low numbers of cells that are embedded within complex tissues. For example, the shoot apical meristem (SAM) consists of a small population of pluripotent stem cells surrounded by developing leaves and/or flowers at the growing tip of the plant. It is technically challenging to collect enough high-quality SAM samples for molecular analyses. Isolation of Nuclei Tagged in specific Cell Types (INTACT) is an easily reproducible method that allows the enrichment of biotintagged cell-type-specific nuclei from the total nuclei pool using biotin-streptavidin affinity purification. Here, we provide a detailed INTACT protocol for isolating nuclei from the Arabidopsis SAM. One can also adapt this protocol to isolate nuclei from other tissues and cell types for investigating tissue/cell-typespecific transcriptome and epigenome and their changes during developmental programs at a high spatiotemporal resolution. Furthermore, due to its low cost and simple procedures, INTACT can be conducted in any standard molecular laboratory. Key words Arabidopsis, Shoot apical meristem, Cell-type-specific promoter, Biotinylation, INTACT, in situ streptavidin histochemistry, low-input RT-qPCR

1

Introduction Over the past decades, numerous studies on the molecular mechanisms that control flowering time in Arabidopsis thaliana have identified many individual genes involved in this critical developmental phase transition. However, most gene expression and epigenetic analyses have been performed using complex tissues or whole seedlings. Conceivably, molecular signals from small specialized tissues or underrepresented cell populations might have been diluted to a level where they become undetectable by standard experimental approaches. At the time of flowering, the shoot apical meristem (SAM) changes its identity from a vegetative meristem (VM) that produces leaves into an inflorescence meristem (IM) that gives rise to flower meristems where floral organs will derive. Despite its central role in

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_16, © Springer Science+Business Media, LLC, part of Springer Nature 2023

313

314

Ruben M. Benstein et al.

organizing the floral transition, molecular analyses of the processes within the SAM remain challenging due to its small size and inaccessibility. Recent technological advances, such as Laser Microdissection (LM), Fluorescence Activated Cell Sorting (FACS), and isolation of nuclei tagged in specific cell types (INTACT), have made it possible to isolate specific cells or tissues that were previously inaccessible to molecular analyses. While both LM and FACS have been successfully used to obtain samples from the SAM for cellspecific transcriptomic studies [1, 2], these two methods involve harsh sample manipulation and require specialized and expensive equipment, restricting their use to a few laboratories. By contrast, INTACT, which allows the isolation of nuclei from a tissue of interest for subsequent transcriptomic and (epi-)genomic analyses without specialized facilities, can be performed in any molecular biology laboratory. Thus, the main advantages of INTACT are its simplicity, speed, cost-effectiveness, and, finally, the high purity of the isolated nuclei [3]. INTACT relies on the co-expression of an engineered nuclear targeting fusion protein (NTF), which consists of a nuclear envelope association domain (WPP-domain), a fluorescence protein for visualization, and a protein ligase acceptor peptide (BLRP), with the Escherichia coli biotin ligase (BirA). Expression of either one or both transgenes from a tissue- or cell-type-specific promoter results in the in vivo biotinylation of the envelope of selected nuclei, facilitating affinity purification of said nuclei using streptavidincoated magnetic beads (Fig. 1a). Under continuous adaption and optimization, INTACT has been successfully used for the isolation of nuclei from Arabidopsis root hair and non-hair cells, endosperm, early embryo, SAM, mesophyll cells, and phloem companion cells [3–9]. Nucleic acids and chromatin derived from INTACT-purified nuclei have been used as starting materials in a wide range of “-omics” studies, including transcriptome analyses such as microarrays and RNA-sequencing (RNA-seq), and chromatin-based assays such as chromatin immunoprecipitation followed by nextgeneration sequencing (ChIP-seq), high-throughput chromatin conformation capture (Hi-C), and assay for transposase accessible chromatin coupled with high-throughput sequencing (ATAC-seq) [3–11]. Here we provide the most recent version of our protocol for the isolation of nuclei from the SAM by INTACT. The protocol is based on the original description of INTACT [3] and our previously published method describing the isolation of nuclei from the SAM, which provided evidence of tissue-specific chromatin states at the SAM [6]. In addition to the INTACT protocol, which includes an optimized procedure for the extraction of crude nuclei from plant tissues and the purification of the SAM nuclei using streptavidin-coated magnetic beads, we also describe the steps for

INTACT in Arabidopsis

315

Fig. 1 The INTACT system at the Arabidopsis SAM. (a) Illustration of the INTACT procedure. (b, c) Microscopy images of an inflorescence of the SAM-specific INTACT line in (b) brightfield and under (c) mCherry filter. The red fluorescence indicates the expression of RedNTF in the whole meristem. (d) A confocal 3D reconstruction of the whole inflorescence meristem using ex 587 nm; em 610 nm. The SAM nuclei show as red fluorescent dots. Scale bar = 20 μm. (e, f) Histological detection of biotinylated proteins in (e) the VM and (f) the IM. Scale bar = 50 μm. (These figures are reproduced from Fig. 1a, b of You et al. (2017) [6] under a Creative Commons Attribution 4.0 International License). (g-i) Microscopic images of isolates after purification by INTACT from (g) the ProUBQ10:BirA line and from (h, i) the SAM-specific INTACT line. The SAM nuclei stained with DAPI are shown in (g, h) blue under the DAPI filter with low transmission light and shown in (i) red under the mCherry filter. The lines are grids of a hemocytometer

verifying the expression, efficiency, and consistency of the biotinylation of the NTF protein by in situ histochemistry to obtain reliable INTACT-compatible transgenic plants, and provide details on assessing the enrichment of isolated nuclei using low-input, Smart-seq2 qPCR [12].

2

Materials

2.1 Generation of INTACT Reporter Lines

1. Homozygous ProUBQ10:BirA line in the background of Arabidopsis accession Col-0 [6] (see Notes 1 and 2). This line is Basta resistant. 2. Coding sequence of RedNTF amplified from published RedNTF-expressing plasmids or INTACT lines [6] (see Note 3). 3. DNA fragments of promoters and terminators for expressing RedNTF in cell/tissue types of interest (see Note 4).

316

Ruben M. Benstein et al.

4. In planta expression Ti vector for Agrobacterium-mediated plant transformation (e.g., pGreen-based plasmid vectors [13]). 5. Molecular cloning system of user’s choice. 6. Competent cells of E. coli (e.g., strain DH5α). 7. Competent cells of Agrobacterium tumefaciens (e.g., strain GV3101:PMP90:pSOUP). 8. Growth medium containing appropriate concentration of selective chemical reagent other than Basta (e.g., kanamycin or hygromycin) to select the ProUBQ10:BirA lines transformed with the NTF-expressing construct. 2.2 Verification of INTACT Reporter Lines by Microscopy and In Situ Histochemistry

1. Upright confocal microscope or epifluorescence microscope with a mCherry-compatible filter (excitation 587 nm; emission 610 nm). 2. FAA: 3.7% formaldehyde, 5% acetic acid, 50% ethanol. 3. Paraffin embedded in situ histological sections of the tissue of interest (e.g., SAM) of INTACT reporter lines on glass slides (detailed protocols available in Chapters 7, 8, and 17). 4. Histological clearing solution. 5. Ethanol dilution series:100%, 95%, 90%, 80% (v/v). 6. 60% (v/v) EtOH+0.75% (w/v) NaCl. 7. 30% (v/v) EtOH+0.75% (w/v) NaCl. 8. 0.75% (w/v) NaCl. 9. PBS: 10 mM NaH2PO4/Na2HPO4 pH 7.0, 130 mM NaCl. 10. Staining dish with removable rack for slides. 11. Large plastic box. 12. Paper towels. 13. Metal holders. 14. Glass slides and cover slips. 15. Nail polish. 16. TBS-T: 50 mM Tris pH 7.5, 0.9% (w/v) NaCl, 0.1% (v/v) Triton-X-100. 17. TBS-TB: TBS-T + 1% (w/v) blocking reagent (Roche, cat num 11096176001). 18. Streptavidin-Alkaline Phosphatase (AP). 19. TMN-50: 100 mM Tris pH 9.5, 100 mM NaCl, 50 mM MgCl2. 20. NBT-BCIP solution: 18.75 mg/mL nitro blue tetrazolium chloride (NBT) and 9.4 mg/mL 5-bromo-4-chloro-3-indolyl-phosphate, toluidine-salt (BCIP) in 67% DMSO (v/v).

INTACT in Arabidopsis

317

21. TE: 10 mM Tris-HCl, 1 mM ethylenediaminetetraacetic acid (EDTA), pH 8.0. 22. 50% glycerol. 23. Bright field microscope. 24. Shaker. 2.3

INTACT

1. Liquid nitrogen container. 2. Sharp forceps and/or micro dissecting scissors. 3. Mortars and pestles. 4. Small metal spoon. 5. 40 μm cell strainers. 6. Barrier pipette tips. 7. 1.5 mL protein low binding microcentrifuge tubes. 8. 2.0 mL protein low binding microcentrifuge tubes. 9. 15/50 mL centrifuge tubes. 10. Cooling centrifuge for microcentrifuge tubes and for 15/50 mL centrifuge tubes. 11. Tube-rotator capable of 20 RPM. 12. Magnetic-separation rack for 1.5- and 2 mL microcentrifuge tubes or 1 × 1 × 0.5 cm neodymium magnets (see Note 5). 13. 1 mL syringes with needles. 14. Vacuum desiccator and vacuum pump (for X-ChIP). 15. Neubauer improved bright line hemocytometer. 16. Fluorescence microscope with DAPI and mCherry filters. 17. Liquid nitrogen. 18. Nuclear Preparation Buffer (NPB): 20 mM MOPS (pH 7.0), 40 mM NaCl, 90 mM KCl, 2 mM EDTA, 0.5 mM ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA), 0.5 mM Spermidine, 0.2 mM Spermine, 1× cOmplete Protease Inhibitor (Roche) (see Note 6). 19. NPB+T: NPB containing 0.1% (v/v) Triton X-100. 20. NPB+D: NPB containing 2 μg/μL 4′,6-diamidino-2-phenylindole (DAPI). 21. NPB+F: NPB containing 1% (v/v) formaldehyde (for Crosslinking X-ChIP). 22. 2 M Glycine (for X-ChIP). 23. Dynabeads M-280 Streptavidin (Invitrogen). 24. cOmplete protease inhibitor (Roche).

318

Ruben M. Benstein et al.

2.4 Assessment of the Enrichment of SAM-Specific Nuclei by Low Input SmartqPCR

1. Kit for extracting RNA at the nanogram levels (e.g., RNeasy micro kit from Qiagen). 2. SuperScript II reverse transcriptase including Superscript II first strand buffer (Thermo Fischer). 3. 10 mM dNTP. 4. RNase inhibitor. 5. 100 mM Dithiothreitol (DTT). 6. 5 M Betaine. 7. 100 mM MgCl2. 8. 10 μM SmartSeq2_TemplateSwitchingOligo (TSO; CAGTGGTATCAACGCAGAGTACrGrG+G) [12].

AAG

9. 10 μM SmartSeq2_dt oligo (AAGCAGTGGTATCAACGCA GAGTACTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTVN) [12]. 10. 10 μM SmartSeq2_ISPCR oligo (AAGCAGTGGTATCAACG CAGAGT) [12]. 11. RNase-free H2O. 12. High-fidelity PCR enzymes and buffers (available from various suppliers, e.g., New England Biolabs (NEB). 13. Solid-phase reversible immobilization (SPRI) DNA purification beads. 14. 2× SYBR green I quantitative PCR (qPCR) master mix. 15. Gene-specific qPCR oligos (Table 1). 16. PCR thermocycler with heated lid. 17. Real-time PCR system.

Table 1 Oligos for the quantification of enrichment of SAM-specific nuclei by the INTACT method Oligo

Sequence

TUB2_fwd

ACGCTACCTCACAGCCTCTGC

TUB2_rev

CACGTTGTTGGGGATCCACTCCA

At3g59270_fwd

CGAGTGCCTGTGCAAACCTTGG

At3g59270_rev

CCATGTCCTCATCTTCATCGATATCTCC

CLV3_fwd

GATGAAAATGGAAAGTGAATGG

CLV3_rev

GGGAGCTGAAAGTTGTTTCTTG

INTACT in Arabidopsis

3

319

Methods

3.1 Establishment of INTACT Reporter Plant Lines 3.1.1 Generation of INTACT Reporter Lines

1. Grow the homozygous ProUBQ10:BirA line [6] to the flowering stage. 2. Clone selected gene promoter fragment in front of the RedNTF fragment, followed by the selected terminator fragment in an in planta expression Ti vector which facilitates the expression of a resistant gene to a selective chemical reagent other than Basta (e.g., kanamycin or hygromycin). 3. Transform the plasmid construct into the Basta-resistant ProUBQ10:BirA plants by the Agrobacterium-mediated floral dip method [14]. 4. Identify independent homozygous tissue/cell-type-specific RedNTF-expressing INTACT lines by selective germination on a growth medium containing the appropriate selective chemical reagent.

3.1.2 Verification of the INTACT Reporter Lines

1. Confirm the tissue/cell-type-specific expression of RedNTF in the individual INTACT reporter lines by confocal or epifluorescence microscopy with an mCherry-compatible filter (Fig. 1b–d). 2. To verify the efficiency of the spatiotemporal biotinylation of RedNTF protein by in situ histochemistry, collect tissues containing cell types/tissues of interest (e.g., collect apices for examining RedNTF expression in the SAM) of the INTACT reporter line into tubes containing ice-cold FAA. 3. Embed the tissue in paraffin and section the tissue to 8 μm thickness (detailed protocols available in Chapters 7, 8, and 17). 4. Label the glass slides with the histological sections in pencil and place them vertically in a staining dish rack, so the following steps can be done in batch. 5. Dewax and rehydrate the histological sections by following the wash series in Table 2 (histological-clearing steps must be carried out in a fume hood). Keep the rehydrated slides submerged in PBS. 6. Transfer the rack with slides into blocking reagent TBS-TB and incubate for 30–60 min at room temperature (RT) with gentle agitation. 7. Just before applying, prepare a 1:2000 dilution of the streptavidin-AP in TBS-TB. 8. After blocking, dry the slides underneath quickly and apply 120 μL of the streptavidin-AP solution per slide (see Note 7). 9. Carefully put on a coverslip, avoiding creating bubbles, and incubate at RT for 1.5 h.

320

Ruben M. Benstein et al.

Table 2 Dewaxing and rehydration steps [6] Step Nr.

Solution

Time

1.

Histological-clearing solution

10 min

2.

Histological-clearing solution

10 min

3.

100% EtOH

2 min

4.

100% EtOH

2 min

5.

95% EtOH

1 min

6.

90% EtOH

1 min

7.

80% EtOH

1 min

8.

60% EtOH+0.75% NaCl

1 min

9.

30% EtOH+0.75% NaCl

1 min

10.

0.75% NaCl

1 min

11.

PBS

2 min

10. Remove the coverslip using forceps and immediately put the slide vertically in a metal rack submerged in TBS-TB in a staining dish. 11. Wash the sections by incubating the slides in TBS-TB for 30 min at RT with gentle agitation. 12. Repeat the wash step 11 two times with fresh TBS-TB. 13. To remove the detergent in the wash solution and to equilibrate the sections, incubate the slides two times in TMN-50 for 5 min each at RT. 14. During equilibration, prepare the staining solution by mixing 20 μL of NBT-BCIP solution with 1 mL of TMN-50 (see Note 8). 15. Prepare a humid chamber by putting two water-soaked sheets of paper towels on the bottom of a large plastic box. 16. Apply 150 μL of staining solution on the sections and cover the slides with a coverslip. Place slides in metal holders and keep the slides elevated above the wet paper. 17. To develop the staining, incubate at RT overnight in the dark. 18. After incubation, check whether the colouring reaction is complete by observing a few slides under a microscope. If the reaction is insufficient, apply fresh staining solution to the sections and incubate the slides for additional time in the humid chamber at RT in the dark. If the reaction is sufficient, proceed to the next step.

INTACT in Arabidopsis

321

19. Remove the coverslips from the slides and immediately put them vertically in a metal rack submerged in TE in a staining dish to stop the staining reaction and prevent drying of the sections. 20. Wash the sections 3 times in total by incubation for 5 min in fresh TE at RT. 21. After washing, apply 75 μL of 50% glycerol to the sections and cover the slides with coverslips. 22. Seal the slides with nail polish and store them in a slide box until needed. 23. To examine the localization and the efficiency of biotinylation in the tissue, analyze the patterns of the colour reaction on the histological sections under a microscope (Fig. 1e, f). 3.2 INTACT Procedure 3.2.1 Harvesting Starting Material

1. Our optimized protocol enables a yield of approximately 10,000 nuclei from the meristems of 100 plants (21 days old, SD grown). The number of plants can be adjusted according to the number of nuclei needed for the downstream applications. 2. Manually dissect the shoot apices (approx. 2 mm in diameter) by removing the leaves and the hypocotyl with sharp tweezers and/or micro dissecting scissors (see Note 9). 3. Freeze the apices immediately in 15 mL centrifuge tubes suspended in liquid nitrogen. 4. Store the frozen apices at -80 °C until use.

3.2.2

INTACT

1. For one INTACT purification, prepare 50 mL of NPB and place it on ice (see Notes 6 and 10). 2. Transfer the frozen tissue (Subheading 3.2.1) to a mortar pre-chilled with liquid nitrogen and grind the material to a fine powder. Do not let the sample thaw and avoid direct contact of the sample with liquid nitrogen (see Note 11). 3. Fill 15 mL of the ice-cold NPB into a fresh 50 mL centrifuge tube (or NPB+F if crosslinking of the chromatin is desired; appropriate safety measures should be taken when working with formaldehyde). 4. While agitating the 15 mL of cold NPB in one hand, use a small liquid nitrogen-chilled metal spoon to transfer a small portion of the sample into the buffer (see Note 12). 5. Gently agitate the buffer until the sample portion has thawed and resuspended. 6. Repeat steps 4 and 5 until the whole sample is resuspended and keep the solution on ice until step 7 or proceed with crosslinking if desired. For optimal crosslinking, apply a weak vacuum to the lysate in NPB+F buffer for 10 min at RT in a desiccator; stop the crosslinking and neutralize the

322

Ruben M. Benstein et al.

formaldehyde by adding 0.94 mL of 2 M glycine to the lysate (final concentration of 125 mM). Mix by slight agitation and incubate for another 5 min at RT in a vacuum desiccator. 7. Insert a 40 μm cell strainer into a 50 mL centrifuge tube on ice and wet the strainer with 1 mL of NPB. 8. Pass the lysate through the cell strainer to remove tissue debris and collect the flow-through. 9. Wash the debris by running 1 mL of ice-cold NPB dropwise through the strainer. Combine the flow-through from steps 8 and 9. 10. To remove soluble contaminants such as cytoplasmic proteins, metabolites, and transcripts, centrifuge the filtered lysate for 10 min at 1000 × g at 4 °C and discard the supernatant. 11. Resuspend the pellet in 1.5 mL of ice-cold NPB stepwise: gently apply 200 μL of ice-cold NPB to the pellet and resuspend the pellet carefully by agitation. When the buffer turns visibly green, collect the supernatant to a new 2 mL microcentrifuge tube. Repeat the procedure until the whole pellet is resuspended in 1.5 mL of NPB and collected in one tube. 12. Centrifuge the crude nuclei extract (CNE) for 5 min at 1000 × g at 4 °C and discard the supernatant. 13. Resuspend the pellet stepwise in 1.5 mL of ice-cold NPB as in step 11 and collect the resuspended CNE in a new 2 mL microcentrifuge tube. 14. Incubate the CNE suspension in a tube rotator for 30 min at 20 RPM and 4 °C (see Note 13). 15. During the incubation, wash 15 μL of Dynabeads (M-280 streptavidin) two times with 1 mL of ice-cold NPB. After each wash, collect the beads with a magnetic rack or a neodymium magnet and discard the supernatant. 16. Resuspend the beads in 50 μL of ice-cold NPB and keep them on ice until needed. 17. Insert a 40 μm cell strainer into a 50 mL centrifuge tube on ice and wet the strainer with 250 μL of ice-cold NPB. 18. Carefully pass the incubated CNE suspension from step 14 through the cell strainer and wash the retained debris on the strainer dropwise with 250 μL of ice-cold NPB (see Note 14). Transfer the flow-through to a new 2 mL microcentrifuge tube. 19. Add the washed Dynabeads (step 16) to the pre-cleared CNE and incubate the mixture in a tube rotator for 30 min at 20 RPM and 4 °C. 20. After the incubation, take 10 μL of the input sample (beadCNE mixture) for later analysis (see Subheading 3.2.3). Flash freeze the input sample in liquid nitrogen and store it at -80 °C until use (Subheading 3.2.3).

INTACT in Arabidopsis

323

21. Dilute the rest of the input samples with 15 mL of ice-cold NPB+T in a 50 mL centrifuge tube and mix the suspension by slowly inverting the tube. Avoid forming bubbles and keep the suspension on ice. 22. Insert a 1.5 mL microcentrifuge tube into a magnetic rack or attach it to a magnet (see Note 5) and place it on ice. 23. Fill the tube with 1 mL of the diluted input sample and incubate it for 1 min to allow nuclei-bead clusters to be captured by the magnet (see Note 15). 24. Remove the liquid containing unbound nuclei with a needle and syringe, without disturbing the nuclei-beads cluster magnetically collected on the inner wall of the tube. 25. Keep the tube in the magnet rack or attached to the magnet, and quickly repeat steps 23 and 24 until all the diluted suspension (from step 21) has been processed. 26. In order to wash the bead-bound nuclei, remove the tube from the magnetic rack or the magnet and resuspend the beads and bead-bound nuclei in 1 mL of ice-cold NPB stepwise: gently run 250 μL of ice-cold NPB over the pellet and transfer the resuspended beads and bead-bound nuclei to a new 1.5 mL tube with a 1000 μL pipette. Repeat another 3 times until the whole pellet is resuspended in 1 mL total and collected in the same tube. If the beads and nuclei form visible aggregates, pipette carefully up and down to resuspend them. 27. Place the tube in the magnetic rack or attach it to a magnet for 1 min, then remove the liquid using a syringe with a fresh needle. 28. Repeat steps 26–27 three more times. 29. Resuspend the bead-bound nuclei in 500 μL of ice-cold NPB and transfer to a new 1.5 mL microcentrifuge tube and store the sample on ice. 30. To determine the purity and yield of the purified nuclei, mix 10 μL of bead-nuclei suspension with 10 μL of NPB+D and incubate for 3 min on ice in the dark. 31. Load a hemocytometer with DAPI-stained bead-nuclei suspension and analyze the sample using a fluorescence microscope (Fig. 1g–i). 32. Bead-bound nuclei appear as conspicuous clusters of beads in brightfield (Fig. 1h), showing blue fluorescence in the DAPI channel (Fig. 1h) and red fluorescence in the mCherry channel (Fig. 1i). 33. Use the DAPI channel with low transmission light to count the DAPI-stained bead-bound nuclei and DAPI-stained unbound nuclei in the hemocytometer grid (see Note 16). A successful

324

Ruben M. Benstein et al.

INTACT experiment will yield approx. 10,000 SAM-specific nuclei per 100 plants with low (10,000 SAM nuclei isolated by the INTACT method, respectively, using the kit for extracting RNA at the nanogram levels following the manufacturer’s instructions, including the on-column DNaseI treatment. 3. Elute the RNA in 10 μL of RNase-free H2O. 4. Transfer 1 μL of isolated RNA (see Note 18) to a PCR tube on ice and mix with 1 μL of 10 mM dNTP and 1 μL of 10 μM SmartSeq2_dt oligo. 5. Incubate the RNA-oligo mixture at 72 °C for 3 min. 6. For oligo annealing, transfer the sample immediately to ice. 7. Prepare Smart-RT master mix [12] containing 2 μL of Superscript II first strand buffer, 2 μL of 5 M Betaine, 0.15 μL of RNase-free H2O, 0.6 μL of 100 mM MgCl2, 0.25 μL of RNase Inhibitor (40 U/μL), 0.5 μL of Superscript II Reverse Transcriptase, and 1 μL of 10 μM SmartSeq2 TSO. 8. Mix the 3 μL of oligo annealed RNA with 7 μL of SmartRT master mix. 9. Transfer the reaction to a PCR thermocycler with a heated lid (>90 °C) and incubate at 42 °C for 90 min, followed by 10 cycles of 50 °C for 2 min and 42 °C for 2 min. Stop the reaction by heating to 72 °C for 15 min. 10. Transfer the reaction mix containing the synthesised Smart cDNA to ice and add 14 μL of H2O, 1 μL of 10 μM Smart_ISPCR oligo, and 25 μL of High-fidelity 2× PCR master mix. 11. Run the reaction in a PCR thermocycler with the program: 3 min at 98 °C followed by 12 cycles of 98 °C for 15 sec, 67 °C for 20 sec, and 72 °C for 6 min (see Note 19). Finish the amplification by a final extension for 5 min at 72 °C. 12. Isolate the amplified ds-cDNA using SPRI beads according to the manufacturer’s instructions and elute the ds-cDNA in 40 μL of elution buffer (EB).

INTACT in Arabidopsis

325

13. The samples can be stored at -20 °C until needed. 14. To verify the enrichment of tissue/cell-type-specific nuclei by the INTACT method, carry out a qPCR analysis of marker genes, for example, CLV3 and At3G59270 for the SAM sample, on input and INTACT samples using gene-specific primers (see Table 1). One reaction contains: 0.5 μL of ds-cDNA, 0.25 μL of 10 μM fwd-oligo, 0.25 μL of 10 μM rev-oligo, 4 μL of H2O, and 5 μL of 2× SYBR Green I qPCR master mix. 15. Analyze the enriched expression of the SAM-marker genes in the purified nuclei samples relative to the input samples by the ΔΔCt or equivalent method [15], using the expression of a house-keeping gene, for example, TUB2, as the control (see Table 1).

4

Notes 1. The original INTACT system made use of the ACT2 promoter to drive the expression of BirA in Arabidopsis roots [3]. However, ProACT2 expression is relatively low in the SAM [16, 17] and subjected to circadian oscillation [18]. Thus, in our experiments, we replaced ProACT2 with the Arabidopsis UBQ10 (At4G05320) promoter (ProUBQ10) for the strong, even, and ubiquitous expression of BirA [6]. 2. In our system we have established a robust ProUBQ10:BirA line. New INTACT reporter lines can be easily generated by transforming an NTF-expression construct into this background. It should be noted that, while our system uses wildtype E. coli BirA [6], a plant-codon-optimized BirA version, mBirA, has been used in other INTACT systems, for example, for Arabidopsis embryos [5]. 3. The original INTACT system employs an NTF protein containing Green Fluorescent Protein (GFP). In our system, we replaced GFP with mCherry to make a variant to which we refer as “RedNTF.” This strategy makes it possible to combine the INTACT line with the existing reporter lines that express GFP-tagged transcription factors involved in the regulation of flowering time and flower development, for cell-type-specific ChIP-seq using an anti-GFP antibody. 4. To establish INTACT, it is crucial to choose a specific promoter that drives the expression of RedNTF precisely in the cell/ tissue types of interest, as this is what provides specificity for the purification by INTACT. For example, for establishing INTACT for the whole meristem, we combined the promoter and terminator of CLV3 (At2g27250) to generate a ProCLV3: RedNTF:TerCLV3 construct that drives the expression of

326

Ruben M. Benstein et al.

RedNTF in the stem cell population at the SAM, and the promoter and terminator of At3g59270 to generate ProAt3g59270:RedNTF:TerAt3g59270 that drives RedNTF expression in the entire meristem except for the stem cells [6]. 5. Instead of a magnetic rack, neodymium magnets can be used as an effective and more flexible alternative to collect Dynabeads. For the isolation of bead-bound nuclei, we use single tube magnets, consisting of two neodymium magnets of 1 × 1 × 0.5 cm each, which are attached to a longitudinally cut 2 mL microcentrifuge tube. Such a magnet is small, extremely cost efficient, and easy to handle. 6. Sterile filtered NPB stock, omitting spermidine, spermine, and protease inhibitor, can be prepared and stored at 4 °C for at least 3 months. For the INTACT experiment, prepare the working NPB buffer by adding 250 μL of 100 mM spermidine, 100 μL of 100 mM spermine, and 1 tablet of cOmplete protease inhibitor per 50 mL NPB. Keep the buffer on ice or at 4 °C. 7. Process one slide at a time to prevent drying of hydrated sample sections. 8. For higher sensitivity, the staining solution can be prepared with TNM-50 containing 10% polyvinyl alcohol (70–100 kDa). 9. To maximize the recovery of nuclei it is important to grind the tissue to a very fine powder. If a small sample of tissue needs to be processed, the size of the mortar and pestle needs to be adjusted accordingly, and one might add a small amount of wild-type tissue to increase the tissue volume for easy handling. 10. To prevent potential contamination during the procedure, it is recommended to divide the freshly prepared buffer as follows: two times 1.5 mL aliquots for wash and incubation steps, five times 1 mL aliquots for final wash steps, 15 mL for the preparation of NPB+T, 1 mL for the preparation of NPB+D, and two times 1 mL to wash the Dynabeads. 11. Avoid direct contact between the sample and liquid nitrogen. The nuclei yield decreases when the ground sample comes into direct contact with boiling liquid nitrogen. For that reason, we place the mortar in a Styrofoam box filled with some liquid nitrogen (ca. 1–2 cm in depth) to keep the mortar cold and the sample frozen. Do not add liquid nitrogen directly to the sample throughout the procedure. 12. Avoid freezing the NPB buffer with the sample. Adding too much of the ice-cold freshly ground sample to the cold NPB at once will lead to sample-buffer freezing, resulting in reduced efficiency of the nuclei preparation and ultimately yield loss. It is advisable to add the freshly ground samples only in small portions to the NPB at a time to prevent freezing of the buffer.

INTACT in Arabidopsis

327

13. Clear the CNE before adding Dynabeads. Depending on the tissue type, flocculation can occur, which leads to the carryover of debris and plastids. Therefore, incubation of the CNE for 30 min at 4 °C and subsequent removal of flocculated particles by passing the samples through a 40 μm cell strainer is very helpful. We found that clearing the CNE before nucleibead binding prevented the nuclei and beads from sticking to the inner walls of tubes and pipette tips and from non-specific binding. When using our protocol, it is therefore unnecessary to treat the tubes and pipette tips with bovine serum albumin (BSA) or include BSA in the nuclei-bead binding step, as in other INTACT protocols [3–11]. This is beneficial because BSA is a potential source of biotin and nucleases, which might reduce the quality of chromatin and RNA samples extracted from the purified nuclei. 14. Keep the volume 5 kb, FLOWERING LOCUS T [7, 8]), in introns (e.g., AGAMOUS [9–11]; TARGET OF FLC AND SVP1 [12]), and 3′ of the coding sequence (e.g., FLOWERING LOCUS C [13]). RNA in situ hybridization directly visualizes cells expressing an transcript of interest by the means of a specific complementary, labelled RNA probe, which can be detected with antibodies coupled to alkaline phosphatase or a fluorophore on thin tissue sections at cellular resolution. All available RNA in situ hybridization protocols include numerous steps, such as tissue embedding and sectioning, probe synthesis and hybridization, washing steps, signal detection, and microscopy. Each step in these laborious protocols has the potential to affect the outcome, that is, the signal strength, presence or absence of background, and visibility of individual cells. We have successfully used RNA in situ hybridization to visualize the distinct gene expression of floral marker genes, such as LFY on Arabidopsis thaliana inflorescence tissue sections (Fig. 1a), or closely related gene family members on tissue harvested from Arabidopsis thaliana and Arabis alpina plants (TPS1 [14], and Fig. 1b; SPL [15]; NIA [16]), but also tissue samples harvested from other species, for example, Solanum tuberosum, potato ([17]; Fig. 1c) and Chrysanthemum morifolium plants (Fig. 1d). The protocol described here is based on a version published by Weigel and Glazebrook [18] with major modifications. Within the method text and notes we are providing the means to avoid all the above-mentioned problems through hands-on advice regarding those steps that are critical for a distinct visualization of gene expression without any background.

RNA In Situ Hybridization

333

Fig. 1 Examples of RNA in situ hybridization. Specific probes for (a) LEAFY (AtLFY) hybridized to sections through Arabidopsis thaliana, and (b) TREHALOSE PHOSPHATE SYNTHASE1 (AaTPS1) hybridized to sections through Arabis alpina shoot apices, as well as (c) MACROCALYX (StMC) and SUCROSE TRANSPORTER2 (StSUC2) probes hybridized to sections through Solanum tuberosum shoot apex and tuber, respectively. (d) An APETALA1 (CmAP1) specific probe hybridized to sections through a developing Chrysanthemum morifolium flower

334

2

Vladislav Gramma and Vanessa Wahl

Materials Prepare all solutions with RNase-free water and in baked glassware to prevent RNA degradation (see Note 1). Work with gloves and spray working surfaces with RNaseZap (Invitrogen).

2.1 Probe Synthesis and Dot Blot

1. SP6 and T7 or T3 oligos (see Note 2). 2. DIG RNA labelling kit (SP6/T7, Roche). 3. 0.5 M ethylenediaminetetraacetic acid (EDTA) (pH 8.0). 4. 4 M LiCl. 5. Absolute ethanol. 6. Diethyl pyrocarbonate (DEPC)-water (see Note 1). 7. 80% ethanol (ice-cold): Prepare from absolute ethanol. 8. Carbonate buffer: 80 mM NaHCO3, 120 mM Na2CO3. Autoclave and aliquot. Store at -20 °C). 9. 10× neutralization buffer: 10% acetic acid (v/v) in DEPCwater. Prepare fresh before use, mix well. 10. Glycogen: 20 mg/mL. 11. 1 M MgCl2. 12. Deionized formamide (store at 4 °C). 13. 50% (w/v) dextran sulfate: Prepare 2 mL aliquots in 15 mL falcon tubes, store at -20 °C (see Note 3). 14. 1 M Tris-HCl (pH 8.0). 15. Na-phosphate buffer (pH 6.8): mix 46.3 mL of 1 M Na2HPO4 and 53.7 mL of 1 M NaH2PO4 for a final volume of 100 mL. 16. In situ salts (for 10 mL): 3 M NaCl (6 mL of 5 M stock), 100 mM Tris-HCl (pH 8.0, 1 mL of 1 M stock), 100 mM Na-phosphate (pH 6.8, 1 mL of 1 M stock), 50 mM EDTA (pH 8.0, 1 mL of 0.5 M stock) (1 mL of DEPC-water, for a final volume of 10 mL). Aliquot and store at -20 °C (see Note 3). 17. 50× Denhardt’s solution: Prepare 0.2 mL aliquots, store at 20 °C (see Note 3). 18. 50 mg/mL yeast tRNA: Prepare 0.1 mL aliquots, store at 20 °C (see Note 3). 19. Hybridization buffer (for 10 mL): mix 5 mL of deionized formamide, 2 mL of 50% dextran sulfate, 1 mL of 10× in situ salts, 0.2 mL of 50× Denhardt’s solution, 0.1 mL of tRNA (50 mg/mL), and 1.7 mL DEPC-water (see Note 3). 20. Nylon N+ membrane.

RNA In Situ Hybridization

335

21. Stratalinker or similar (bake the membrane for 2 h at 80 °C as an alternative to UV-crosslinking with a Stratalinker). 22. Nitro blue tetrazolium (NBT)/5-bromo-4-chloro-3-indolyl phosphate (BCIP) solution for dot blot (5 mL): 100 μL in TNM-50. 23. 1× TE. 2.2 Tissue Embedding and Sectioning

1. FAA fixative: 50% Ethanol (v/v), 5% Glacial Acetic Acid (v/v), 3.7% Formaldehyde (v/v), 41.3% Water (v/v). 2. Tissue processor (e.g., ASP300S, Leica). 3. Biopsy cassettes or active flow biopsy cassettes (Leica): choose pore size according to size of tissue sample. 4. For loading the ASP300S you will need the following: FAA-fixative, 70% technical ethanol, 90% technical ethanol, 99% technical ethanol + eosin (1 spatula is enough), absolute ethanol, xylene, Paraplast (Leica). 5. Surgipath Paraplast plus (Leica). 6. Metal base molds (Leica). 7. Processing cassette (Leica). 8. PFA solution: 4% PFA in 1× PBS with 0.1% Tween-20. Prepare fresh, stir about 1 h at 60 °C, cool down to 4 °C. 9. Ethanol series for manual protocol: 30% technical ethanol + 0.75% NaCl; 40% technical ethanol + 0.75% NaCl; 50% technical ethanol + 0.75% NaCl; 60% technical ethanol + 0.75% NaCl; 70% technical ethanol + 0.75% NaCl; 85% technical ethanol + 0.75% NaCl; absolute ethanol; absolute ethanol + a spatula of eosin. 10. Transfer into Paraplast in manual protocol: 25% histoclear in ethanol; 50% histoclear in ethanol; 75% histoclear in ethanol; 100% histoclear; Paraplast (Leica). 11. Embedding station (e.g., EG1160, Leica). 12. Rotary microtome (e.g., RM2265, Leica) equipped with low-profile disposable blades (819, Leica).

2.3 Probe Hybridization

All solutions should be prepared with DEPC-treated water (see Note 1) in baked glassware (see Note 4). 1. Shaking water bath. 2. Shaker. 3. Glass container (e.g., glass staining trough H554.1, Carl Roth, Germany; Fig. 2a).

336

Vladislav Gramma and Vanessa Wahl

Fig. 2 Outline of the Probe hybridization protocol. All steps are indicated with numbers and letters according to the protocol described in Subheading 3.3 of the text. Subheading 3.4 refers to the following section on microscopy. Abbreviations: RT room temperature

4. Rack for glass container (e.g., H552.1, Carl Roth, Germany; Fig. 2a). 5. Wire hanger for glass staining rack (e.g., H553.1, Carl Roth, Germany). 6. Histoclear. 7. Absolute ethanol (see Note 5). 8. 95% ethanol (v/v) (see Note 5). 9. 90% ethanol (v/v) (see Note 5). 10. 80% ethanol (v/v) (see Note 5). 11. 60% ethanol (v/v) + 0.75% NaCl (w/v) (see Note 5). 12. 30% ethanol (v/v) + 0.75% NaCl (w/v) (see Note 5).

RNA In Situ Hybridization

337

13. 0.75% NaCl (w/v) (see Note 5). 14. 10× PBS: 1.3 M NaCl, 30 mM NaH2PO4, 70 mM Na2PO4 (pH 7.0 with HCl). 15. 1× PBS. 16. 1 M Tris-HCl (pH 7.5). 17. Proteinase K (PCR grade). 18. Proteinase K dilution buffer: 100 mL of 1 M Tris-HCl (pH 7.5) and 100 mL of 0.5 M EDTA (pH 8.0). 19. Hybridization buffer (see Note 3). 20. Polysine-coated slides and cover slips (24 mm × 60 mm). 21. Soaking solution (35 mL): prepare fresh from 17.5 mL of formamide, 3.5 mL of 20× SSC, and 14 mL of DEPC-water (see Note 6). 22. Humidified boxes/preparation slide folder (plastic) (e.g., Preparation Folder ROTILABO® Plastic Set). 23. 20× SSC: 3 M NaCl, 0.3 M Na-citrate (pH 7.0, with HCl). 24. 2× and 0.2× SSC: Prepare from 20× SSC (see Note 7). 25. 10× TBS: 60.57 g/L Tris, 0.9% NaCl (w/v). 26. TPS-T: Prepare 1× TBS from 10× TBS. Prepare fresh by adding 0.1% Triton-X-100 (v/v). 27. Blocking reagent solution: 1% (w/v) Roche Blocking reagent (Cat Num 11096176001) in TBS-T (see Note 8). 28. Anti-DIG AP antibody Fab fragments (150 U) (e.g., Roche 11093274910). 29. Anti-Dig solution: 1:1250 Anti-DIG AP antibody in blocking reagent solution. 30. 5 M NaCl. 31. 1 M MgCl2. 32. TNM-50 (1 L): 100 mM Tris-HCl (pH 9.5, 100 mL of 1 M stock), 100 mM NaCl (20 mL of 5 M stock), 50 mM MgCl2 (50 mL of 1 M stock). Adjust the final volume to 1 L with DEPC-water. 33. PVA-TNM-50: 10% Polyvinyl alcohol (PVA—m. w. app. 115,000) (w/v) in TNM-50 (see Note 9). 34. NBT-BCIP stock solution (8 mL). 35. NBT-BCIP solution for hybridization: Prepare fresh by diluting NBT-BCIP 1:50 in PVA-TNM-50 (see Note 9). 2.4

Microscopy

Histological sections are imaged with an upright microscope (e.g., Olympus BX-61) equipped with a digital camera (e.g., DP74, Olympus).

338

3

Vladislav Gramma and Vanessa Wahl

Method Use filter tips and autoclave tubes fresh from the bag before use. Work with gloves and spray working surfaces with RNaseZap.

3.1

Probe Synthesis

1st Day 1. Amplify full-length cDNA of interest cloned into the pGEM®T Easy vector by PCR using SP6 and T7 or T3 oligos and elute the band from a gel (see Note 2). 2. Take 200 ng of template cDNA and add DEPC-water to a final volume of 13 μL. 3. Assemble the probe synthesis reaction on ice. For this, mix the following DIG RNA labelling kit reagents (refer to # on tube in kit) from top to bottom by vortexing: 10× transcription buffer (#8) RNAse inhibitor (#10) 10× NTP mix (#7) Enzyme (#11or #12) Template cDNA

2.0 μL 1.0 μL 2.0 μL 2.0 μL (40 U) 13 μL

4. Incubate at 37 °C for 2 h. 5. Add 0.25 μL of RNase-free DNase (tube #9 of kit, equals 2 U). 6. Incubate at 37 °C for 15 min. 7. Place on ice and add 1 μL of 0.5 M EDTA, mix well (vortex!). 8. Add the following to precipitate the RNA: 4 M LiCl Ethanol

2.5 μL 75.0 μL

9. Mix well and incubate at -70 °C for at least 30 min or at -20 °C for at least 2 h. 10. Centrifuge at 4 °C for 30 min at top speed: you will see a pellet! 11. Discard the supernatant. 12. Wash the pellet in 200 μL of ice-cold 80% ethanol and centrifuge at top speed for 10 min. 13. Air-dry and redissolve the pellet in 100 μL of DEPC-water. 14. Mix 100 μL of RNA with 100 μL of carbonate buffer and incubate at 60 °C for cDNA length-dependent time (see Note 10). 15. After incubation neutralize the reaction with 20 μL of 10× neutralization buffer (mix well!).

RNA In Situ Hybridization

339

16. Add and mix the following components: Glycogen (20 mg/mL) 1 M MgCl2 Ethanol

1 μL 1 μL 600 μL

17. Incubate at -20 °C overnight. 2nd Day 18. Centrifuge at 4 °C and top speed for 30 min. 19. Discard the supernatant. 20. Wash the pellet with 200 μL of ice-cold 80% ethanol and centrifuge at top speed for 10 min. 21. Air dry and redissolve in 50 μL of DEPC-water. 22. Remove 5 μL of sample to check on a gel: Run the samples on a 1.5% Tris-borate-EDTA (TBE) gel with DNA size marker (no DEPC-water necessary for buffer). The fragmented RNA should smear between 100–400 bp. 23. Dilute the remaining RNA in 450 μL of hybridization buffer. Store the probe at -20 °C. 24. For a dot blot, spot the probe on a Nylon N+ membrane (e.g., 1:100, 1:1000, 1:10,000) and compare with the labelled (do not use undiluted!) and unlabeled control RNA included in the DIG RNA labelling kit (note: 1:100 spot equals 1 ng of labelled RNA). 25. Dry the membrane and UV-crosslink (2 times, each 30 seconds) or temperature-crosslink (2 h at 80 °C) the RNA. 26. Incubate in approximately 20 mL of blocking reagent solution (see Note 8) for 30 min while gently shaking at room temperature. 27. Replace with anti-DIG AP antibody 1:5000 in approximately 5 mL of fresh blocking reagent solution (see Note 8) and shake the membrane for about 1 h. 28. Wash 5 times with approximately 20 mL of TPS at room temperature for 2 min each. 29. Replace with NBT-BCIP solution: cover/submerge the membrane with NBT-BCIP solution and keep in the dark until signal is visible (do not shake). 30. Wash the membrane with water and stop the reaction with 1× TE. 31. Make a photocopy of the membrane, as the signal will fade over time.

340

Vladislav Gramma and Vanessa Wahl

3.2 Tissue Embedding and Sectioning

Automated embedding protocol (for tissue processor ASP300S and tissue embedding system EG1160) 1. Harvest the tissue in freshly prepared FAA-fixative in a falcon tube (keep dark and cold). 2. Transfer the material into a biopsy cassette or an active flow biopsy cassette and label it with a pencil (note: all other markers will come off during infiltration!). 3. Transfer the cassette into the appropriate metal tray and the tray into the retort of the ASP300S. 4. Start a program with the following specifications: Station

Reagent

Duration

Temp.

Drain (s)

1

FAA-fixative

4h

140

2

Ethanol, 70% techn.

1h

80

3

Ethanol, 90% techn.

1h

80

4

Ethanol, 90% techn.

1h

80

5

Ethanol, 99% techn. + eosin

1h

89

6

Ethanol, 99% techn.

1h

80

7

Ethanol, absolute

1h

140

8

Xylene

1h

80

9

Xylene

1h

80

10

Xylene

1 h, 15 min

140

Wax 1

Paraplast

1h

62 °C

140

Wax 2

Paraplast

1h

62 °C

140

Wax 3

Paraplast

3h

62 °C

140

5. Make sure that the functions “stirrer,” “recirculation,” and “pressure-vacuum” are switched on. The machine stops when the samples are in the last Paraplast step until you give the command “drain retort.” After taking the samples out, immediately store at 65 °C in the EG1160 station and clean the ASP300S with a “routine cleaning cycle.” 6. Pour the molds with the EG1160. For this purpose, preheat metal base molds, processing cassettes (label with pencil before), and tools (e.g., forceps) in the preheating chamber. 7. Place warm molds on prewarmed working area and fill with Paraplast. Place tissue sample in the middle position using prewarmed forceps and place cassette on top. If necessary, pour a little more Paraplast on top.

RNA In Situ Hybridization

341

8. Carefully move the mold to the 4 °C working area and leave to solidify. Then remove the metal mold and store the Paraplast blocks at 4 °C (see Note 11). Manual Embedding Protocol Steps 2 to 5 from the automated protocol can be processed manually; however, the tissue samples are fixed in different fixative: 1. Harvest tissue samples in paraformaldehyde (PFA) solution in a vial (e.g., Eppendorf tube, Falcon). 2. Vacuum-infiltrate until tissue stays at the bottom of the vial. Keep on ice. 3. Keep in PFA-solution without Tween-20 at 4 °C overnight. 4. Dehydrate at 4 °C on a shaker: PBS PBS 30% ethanol + 0.75% NaCl 40% ethanol + 0.75% NaCl 50% ethanol + 0.75% NaCl 60% ethanol + 0.75% NaCl 70% ethanol + 0.75% NaCl 85% ethanol + 0.75% NaCl

30 min 30 min 1h 1h 1h 1h 1h 1h

5. Keep in absolute ethanol without shaking overnight. 6. Shake at room temperature in Absolute ethanol + eosin Absolute ethanol

2× 30 min 2× 30 min

7. Transfer into wax by shaking in 25% histoclear 50% histoclear 75% histoclear 100% histoclear 100% histoclear

1h 1h 1h 2× 1 h 2× 1 h

8. Add one Paraplast chip and keep without shaking at room temperature overnight. 9. Transfer to 42 °C and add 1 Paraplast chip every hour until it no longer dissolves. 10. Transfer to 60 °C and add 1 Paraplast chip every hour until about double the previous volume or there is no more space in the tube. 11. Exchange the Paraplast with fresh dissolved Paraplast and keep at 60 °C overnight.

342

Vladislav Gramma and Vanessa Wahl

12. Over the following 3 days, replace with fresh dissolved Paraplast three times a day and keep at 60 °C. 13. Continue with step 6 of the automated protocol. Sectioning 1. For sectioning, trim the Paraplast block in order to get as close to the tissue as possible. Try to trim it as squarely as possible. This helps to get straight ribbons while sectioning. 2. Place the Paraplast block into the microtome and cut sections: 20 μm to get closer to the tissue, 8 μm for the sections. 3. Place ribbons in a 40 °C water bath to remove compressions using a wet wooden tool (e.g., toothpick) or a brush, then transfer them to a slide leaving as little room as possible between the ribbons. 4. Keep slides on a 42 °C heating plate for some hours or overnight. 5. Label slides with pencil only! Every other marker will come off during the hybridization process. 6. Slides can be stored in a box at 4 °C for some time (usually several months). 3.3 Probe Hybridization

Use filter tips and autoclave tubes twice before use. Work with gloves and spray working area with RNaseZap. 1st Day (Fig. 2) 1. Set water bath to 37 °C. 2. Place a glass container with 200 mL proteinase K dilution buffer in water bath (shaking). 3. Prepare humidified box using a plastic preparation slide folder and soaking solution (see Note 6). 4. Immerse the slides and dip them up and down (Fig. 3a) in the following solutions. (Note: The slides should be completely covered. Depending on the size of the glass container, e.g., container H554.1: approximately 200 mL) (a) Histoclear I 10 min (b) Histoclear II 10 min (c) 100% ethanol 2 min (discard afterward) (d) 100% ethanol 2 min (keep) (e) 95% ethanol 1 min (f) 80% ethanol 1 min (g) 60% ethanol + 0.75% NaCl 1 min (h) 30% ethanol + 0.75% NaCl 1 min (i) 075% NaCl 2 min (j) PBS 2 min

RNA In Situ Hybridization

343

Fig. 3 Handling tips for crucial steps of the Probe hybridization protocol. (a) During all steps of the ethanol series the samples have to be rigorously dipped up and down manually (steps 4 and 6). (b) Do not pull! Instead lift the coverslip off the slide to prevent damaging the tissue sections (steps 15 and 20)

5. Add 10.4 μL of proteinase K to 200 mL of prewarmed Proteinase K dilution buffer shortly before use (see Note 12). (k) Proteinase K (l) PBS (m) PBS

30 min (no longer!), 55 °C, shaking 5 min 5 min

6. Immerse the slides in the reversed order solutions (i.) to (d.) followed by fresh 100% ethanol for 30 seconds each (Fig. 3a, see Note 5). 7. Dry slides at room temperature for 30 min to 1 h in a container with a few sheets of tissue paper soaked with ethanol (see Note 13). 8. About 150 μL of hybridization buffer is mixed with the appropriate probe per slide (see Note 14). 9. Incubate at 80 °C for 2 min, chill on ice and keep on ice until use. 10. Distribute on the polysine-coated slide with the tissue sections and cover with a cover slip (see Note 15). 11. Incubate in a humidified box (preparation folder and soaking solution) at 55 °C overnight (see Note 6). 12. Place 2× SSC and 0.2× SSC in a 55 °C incubator ready for the next day. Keep one bottle 0.2× SSC at room temperature (see Note 7).

344

Vladislav Gramma and Vanessa Wahl

2nd Day (Fig. 2) 13. Set shaking water bath to 55 °C. 14. While the slides are being washed in Step 15, prepare about 1.0 L of blocking reagent solution (see Note 8). 15. Carefully remove the cover slips from the slides one by one (Fig. 3b, see Notes 16 and 17) immediately immersing them in 2× SSC at 55 °C (water bath). Dip them up and down a few times before shaking them in: 4 × 0.2× SSC (55 °C) 2 × 0.2× SSC (1st at 37 °C, 2nd > RT) 0.2× SSC (RT)

each 30 min each 5 min 5 min

16. Immerse the slides in 1× PBS. The slides can be stored overnight in 1× PBS, but it is advised to proceed with detection on the same day if possible. 17. Immerse and incubate slides in blocking reagent solution for 1 h (appr. 200 mL, gently shaking). 18. Prepare antibody (anti-Dig) solution 10 min before use. 19. Dry the underside of the slides and apply 100 μL of anti-DIG solution (see Note 16). Cover with a cover slip and incubate at room temperature for 90 min in a humidified box (preparation folder and water) (see Note 6). 20. Remove the cover slips (Fig. 3b, see Notes 16 and 17) and immediately wash four times with blocking reagent solution (approximately 200 mL each wash). 21. Remove the blocking reagent solution and wash twice with TNM50, 5 min each wash, to completely remove the detergent. 22. Prepare NBT-BCIP solution (see Note 9). 23. Apply approximately 200 μL NBT-BCIP solution per slide with a plastic pipette, cover with a cover slip, and place the slides in a box (e.g., plastic preparation slide folder) humidified with water (see Note 6). 24. Wrap the box in aluminum foil and incubate at room temperature overnight or until staining developed (see Notes 18 and 19). 3.4 Microscopy: The Secret of Good PictureTaking

Make use of an upright microscope (e.g., Olympus BX61). The 10× ocular will suffice in most cases. Use differential interference contrast (DIC), as this will result in contrasted images of tissue sections which are normally not contrasted when analyzed with brightfield microscopy. Make sure to adjust DIC optics properly. The image should have a 3-D appearance, resembling an object illuminated by a unidirectional light source. Use the “white balance” to generate a

RNA In Situ Hybridization

345

background that does not distract from the main message of the picture (usually bright grey). Once this is obtained for the 10× ocular, the settings should be fixed (in the software) and not changed between slides of one experiment (same probe). Before taking pictures, the focus should be adjusted properly for each section. Other than that, the secret of taking good pictures is fairly simple: focus well (cells should be visible) and keep an eye on the message you want to convey. In some cases, a single picture will be enough (e.g., a middle section through a shoot apical meristem), in others several pictures will have to be taken, for example, when the object is very large (e.g., a young, developing potato tuber), for which several individual pictures have to be “stitched” to generate an overview or when a three-dimensional visualization is to be demonstrated using pictures from consecutive sections. For larger sections, take several overlapping (30% overlap) individual pictures covering the section and make use of the Adobe Photoshop Photomerge plugin for picture assembly or use the software of the microscope. Small gradients in the background of the tissue can also be adjusted using Adobe Photoshop. However, utmost care should be taken to adhere to the common standards of preventing scientific misconduct.

4

Notes 1. RNase-free water can be purchased or homemade through diethyl pyrocarbonate (DEPC) treatment. For this purpose, 0.1% (v/v) DEPC is added into Milli-Q water. Mix the DEPC-water thoroughly by stirring at 160 rpm overnight at room temperature. Autoclave at 180 °C to remove DEPC and let cool to room temperature before use. 2. Template production. Oligos are common SP6 and T7 primers to generate templates from pGEM®-T Easy (Promega) vector constructs. We use a simple Taq polymerase-based protocol in a Bio-Rad thermocycler. In order to generate clean template of sufficient concentration (>100 ng/μL), it might make sense to run several PCR reactions, run them over an agarose gel in individual lanes, and extract them from the gel using a single column of a clean-up kit (e.g., Promega Wizard® Gel and PCR Clean-Up system). Wash well and elute with DEPC-water. Elute with as little DEPC-water as possible to achieve a high concentration of the PCR product (>100 ng/μL). Check concentration and the level of purity using a Nanodrop or similar. Run 1 μL on a gel to confirm that the amount of DNA fits with the Nanodrop results.

346

Vladislav Gramma and Vanessa Wahl

3. The Hybridization Buffer will be in direct contact with the probe (during storage and hybridization) and is therefore supposed to be prepared with special care. Dextran sulfate (50%) does not easily solubilize at that concentration. It is therefore recommended to prepare 2 mL aliquots in 15 mL falcon tubes, to which the DEPC-water component (+1.7 mL = 3.7 mL) of the Hybridization buffer is added (add all other components prior to use). All components should be stored at -20 °C until use, except for deionized formamide, which does not necessarily need to be kept at -20 °C (though recommended by the manufacturer), but can be stored at 4 °C instead. However, if you wish to do so, prepare 5 mL aliquots to prevent losing time over melting the entire bottle each time you are preparing the Hybridization Buffer. 4. Baking: overnight at 160 °C or 6 h at 180 °C. Use baked capped bottles for (stock) solutions. Wrap all other glassware and tools (glass staining trough, rack for glass container with wire hanger, measuring cylinders, magnetic stir bars, tools such as spatula, etc.) in aluminum foil before baking. In our experience baking is the most convenient and robust way to inactivate RNases on glassware and tools. However, when baking is impossible, all equipment could potentially also be decontaminated with alternative methods (e.g., soaking in freshly prepared 0.1% aqueous solution of DEPC followed by autoclaving, or 3% hydrogen peroxidase treatment). 5. Preparation and use of the ethanol series: For 200 mL of 30/60% ethanol + 0.75% NaCl, mix 60/120 mL of ethanol with 136/72 mL of DEPC-water and 6 mL of 5 M NaCl. Except for the 100% ethanol, the rest of the ethanol series can be reused several times when carry-over is avoided and treated with care to avoid contamination. During the experiment, it is handy to place the well-labelled bottles with the remaining ethanol solutions behind each glass container to avoid confusion. 6. Humidified boxes: Two kinds of humidified boxes are used to incubate slides during individual steps of probe hybridization. The first is prepared by placing several layers of cut tissue paper soaked with freshly mixed soaking solution into a plastic preparation slide folder (e.g., Preparation Folder ROTILABO® Plastic Set); for the second, the tissue paper is soaked with water. If the same box is used for both, the tissue paper soaked with soaking solution should be removed and the preparation folder carefully rinsed with water before the second box is prepared with water to avoid being exposed to formamide during microscopy.

RNA In Situ Hybridization

347

7. SSC washes for probe hybridization: for the experiment 2× SSC and 0.2× SSC are kept in a 55 °C incubator, a bottle of 0.2× SSC is kept at room temperature. A remainder of a 0.2× SSC bottle at 55 °C is transferred to room temperature during the last washing step at 55 °C, generating the 0.2× SSC for the “1st at 37 °C” and the “2nd > RT” washing steps. 8. The blocking reagent solution for both the dot blot and probe hybridization is prepared by dissolving Roche blocking reagent (1% w/v) in warm TBS with a magnetic stir bar on a magnetic stirrer. TritonX-100 for the blocking solution that is used during probe hybridization should only be added after the Roche blocking reagent is fully dissolved. Test the final volume needed for the dot blot or the blocking and washing steps of probe hybridization (depending on the size of plastic and glassware used) to avoid unnecessary costs. 9. The NBT-BCIP solution for hybridization is prepared from a PVA solution in TNM-50. For that purpose, PVA (for a final 50% (w/v) solution) is weighed in a 50 mL Falcon tube, mixed with a fraction of the required volume of TNM-50, and carefully heated up using a microwave avoiding overboiling. When all PVA is solubilized, add TNM-50 up to the final volume. Be patient, as this will take quite a while. Also, consider that bubbles will form when heating and mixing. These will have to clear/rise and the solution cool down before it can be mixed with NBT-BCIP and applied to the tissue. Centrifugation will help to speed up the process. 10. The duration of the incubation time required for fragmentation is calculated by the following formula: Time =

Li - Lf K Li Lf

Li: initial length of the probe (in kb) Lf: Final length of the probe (in kb; about 0.1–0.2) K: 0.11 kb/minute 11. When kept continuously at 4 °C, tissue embedded in Paraplast can be stored “indefinitely.” We have generated beautiful pictures, hybridizing sections prepared from 10-year-old samples. 12. Proteinase K digestion is performed to allow better penetration of the probe into the tissue. The concentration of proteinase K is critical in that it should be sufficient but not high enough to destroy the morphology of the tissue. We use proteinase K at a final concentration of about 1 μg/mL for 30 min for 8 μm through tissue of Arabidopsis thaliana,

348

Vladislav Gramma and Vanessa Wahl

Arabis alpina, Solanum tuberosum, Solanum lycopersicum, Chrysanthemum morifolium, and Manihot esculenta. Tissue from all other species should be tested beforehand. 13. A “dry box” is necessary to remove any residual water from the tissue. Use a plastic box and a few layers of tissue paper soaked with 100% ethanol. Transfer slides (in glass rack) to the box and cover tightly with lid. 14. In order to estimate how much of probe is to be used, pilot experiments are necessary. We usually start with 2 μL of the antisense probe on sections through a tissue which is known to express the gene of interest (e.g., determined by RT-qPCR). Depending on the quality of the probe and the expression level of the gene analyzed, staining might be rapidly oversaturated early on the next day. In this case it is necessary to dilute the probe (in rare cases up to 100-fold, e.g., Arabidopsis thaliana Histone H4). Some reviewers still request to see the hybridization of a sense probe as negative control. This might however be problematic for genes that naturally express an antisense transcript (e.g., FLC, Tian et al., 2019). For this reason, hybridizing the antisense probe to an RNA null mutant (CRISPR/Cas9 deletion) is more appropriate. 15. Take 150 μL of hybridization buffer mixed with probe or 100 μL of the blocking reagent solution with Anti-Dig antibody and apply to one end of the slide and use the coverslip to slowly and carefully cover the entire slide without any air bubbles. 16. Process one slide after the other. Do not allow the tissue sections on the slides to fall dry! Slides should always be submerged in a solution, or covered with coverslips. 17. It is absolutely essential to remove the coverslip with care to prevent the tissue from being damaged. Use your finger nail to gently push it a tiny bit from the grid side of the slide over the rim of the other end of the slide. Then, make sure to not slide/pull it over the tissue but instead lift it off in one go (Fig. 3b). 18. Good probes for genes that are expressed well yield staining after only one night of incubation. However, some probes need to be diluted (see Note 14) for others the incubation time will have to be extended (up to several days). 19. Mounting of the slides is not necessary unless they are to be kept for some reason. In this case, remove excess liquid from the slides before checking them under the microscope. Check whether the color reaction is complete by observing a few slides under the microscope. If the reaction is too weak, you can remove the cover slip and apply fresh NBT-BCIP

RNA In Situ Hybridization

349

solution. If the color development is satisfactory, remove cover slip and immediately stop the reaction by washing twice with 1× TE for 5 min. Mount in glycerol/TE (50% v/v, about 75 μL/slide). Seal the slides with nail polish and store the slides in a slide box at room temperature.

Acknowledgments VW would like to thank Rebecca Schwab for introducing her to this amazing method so many years ago. The authors would like to thank former Wahl group members for providing in situ example pictures (i.e., Tanja Seibert—Arabidopsis thaliana, Arabis alpina, Solanum tuberosum; Malleshaiah SharathKumar—Chrysanthemum morifolium), and Jan Lohmann for finding the PVA “solution.” Special thanks go to the many students VW was allowed to teach over the years that have also contributed to improving and testing the individual steps of the protocol, in particular Tanja Seibert, Pedro Garcia, Annika Franke, and Luise Hecker. Research in the Wahl group was funded by the BMBF (SolaMI, 031B0191), the DFG (SPP1530: WA3639/1-2, 2-1), and the Max Planck Society. References 1. Asano T, Masumura T, Kusano H, Kikuchi S, Kurita A, Shimada H, Kadowaki K (2002) Construction of a specialized cDNA library from plant cells isolated by laser capture microdissection: toward comprehensive analysis of the genes expressed in the rice phloem. Plant J 32:401–408 2. Birnbaum K, Shasha DE, Wang JY, Jung JW, Lambert GM, Galbraith DW, Benfey PN (2003) A gene expression map of the Arabidopsis root. Science 302:1956–1960 3. Deal RB, Henikoff S (2011) The INTACT method for cell type-specific gene expression and chromatin profiling in Arabidopsis thaliana. Nat Protoc 6:56–68 4. Tang F, Barbacioru C, Wang Y, Nordman E, Lee C, Xu N, Wang X et al (2009) mRNA-Seq whole-transcriptome analysis of a single cell. Nat Methods 6:377–382 5. Brennecke P, Anders S, Kim JK, Kolodziejczyk AA, Zhang X, Proserpio V, Baying B et al (2013) Accounting for technical noise in single-cell RNA-seq experiments. Nat Methods 10:1093–1095 6. Efroni I, Ip PL, Nawy T, Mello A, Birnbaum KD (2015) Quantification of cell identity from single-cell gene expression profiles. Genome Biol 16:9

7. Adrian J, Farrona S, Reimer JJ, Albani MC, Coupland G, Turck F (2010) cis-Regulatory elements and chromatin state coordinately control temporal and spatial expression of FLOWERING LOCUS T in Arabidopsis. Plant Cell 22:1425–1440 8. Liu L, Adrian J, Pankin A, Hu J, Dong X, von Korff M, Turck F (2014) Induced and natural variation of promoter length modulates the photoperiodic response of FLOWERING LOCUS T. Nat Commun 5:4558 9. Sieburth LE, Meyerowitz EM (1997) Molecular dissection of the AGAMOUS control region shows that cis elements for spatial regulation are located intragenically. Plant Cell 9: 355–365 10. Deyholos MK, Sieburth LE (2000) Separable whorl-specific expression and negative regulation by enhancer elements within the AGAMOUS second intron. Plant Cell 12:1799– 1810 11. Back G, Walther D (2021) Identification of cis-regulatory motifs in first introns and the prediction of intron-mediated enhancement of gene expression in Arabidopsis thaliana. BMC Genomics 22:390 12. Richter R, Kinoshita A, Vincent C, MartinezGallegos R, Gao H, van Driel AD, Hyun Y et al

350

Vladislav Gramma and Vanessa Wahl

(2019) Floral regulators FLC and SOC1 directly regulate expression of the B3-type transcription factor TARGET OF FLC AND SVP 1 at the Arabidopsis shoot apex via antagonistic chromatin modifications. PLoS Genet 15:e1008065 13. Tian Y, Zheng H, Zhang F, Wang S, Ji X, Xu C, He Y, Ding Y (2019) PRC2 recruitment and H3K27me3 deposition at FLC require FCA binding of COOLAIR. Sci Adv 5:eaau7246 14. Wahl V, Ponnu J, Schlereth A, Arrivault S, Langenecker T, Franke A, Feil R, Lunn JE, Stitt M, Schmid M (2013) Regulation of flowering by trehalose-6-phosphate signaling in Arabidopsis thaliana. Science 839:704–707

15. Olas JJ, Van Dingenen J, Abel C, Dzialo MA, Feil R, Krapp A, Schlereth A, Wahl V (2019) Nitrate acts at the Arabidopsis thaliana shoot apical meristem to regulate flowering time. New Phytol 223:814–827 16. Olas JJ, Wahl V (2019) Tissue-specific NIA1 and NIA2 expression in Arabidopsis thaliana. Plant Signal Behav 14:1656035 17. Seibert T, Abel C, Wahl V (2020) Flowering time and the identification of floral marker genes in Solanum tuberosum ssp. andigena. J Exp Bot 71:986–996 18. Weigel D, Glazebrook J (2002) Arabidopsis: a laboratory manual. Cold Spring Harbor Laboratory Press

Chapter 18 The GUS Reporter System in Flower Development Studies Janaki S. Mudunkothge, C. Nathan Hancock, and Beth A. Krizek Abstract The β-glucuronidase (GUS) reporter gene system is an important technique with versatile uses in the study of flower development in a broad range of species. Transcriptional and translational GUS fusions are used to characterize gene and protein expression patterns, respectively, during reproductive development. Additionally, GUS reporters can be used to map cis-regulatory elements within promoter sequences and to investigate whether genes are regulated post-transcriptionally. Gene trap/enhancer trap GUS constructs can be used to identify novel genes involved in flower development and marker lines useful in mutant characterization. Flower development studies primarily have used the histochemical assay in which inflorescence tissue from transgenic plants containing GUS reporter genes are stained for GUS activity and examined as whole-mounts or subsequently embedded into wax and examined as tissue sections. In addition, quantitative GUS activity assays can be performed on either floral extracts or intact flowers using a fluorogenic GUS substrate. Another use of GUS reporters is as a screenable marker for plant transformation. A simplified histochemical GUS assay can be used to quickly identify transgenic tissues. Key words GUS, Reporter gene, Transcriptional reporter, Histochemical staining, Sections, MUG, Fluorometric assay

1

Translational reporter,

X-Gluc,

Introduction A key step in investigating gene function is to know the precise spatial and temporal expression of a gene. Tissue-specific gene expression patterns can be determined directly at the mRNA level by in situ hybridization or at the protein level by immunolocalization. However, many plant genes are members of gene families and it may be difficult to obtain RNA probes or antibodies that are specific for a particular gene or protein. In addition, these procedures are complicated and time consuming. An alternative strategy is to use reporter gene constructs for such studies. Reporter genes encode proteins whose presence is easily assayed. The two most commonly used reporters in plants are β-glucuronidase (GUS) encoded by the E. coli uidA gene and green fluorescent protein (GFP) from the jellyfish Aequorea victoria. There are many

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_18, © Springer Science+Business Media, LLC, part of Springer Nature 2023

351

352

Janaki S. Mudunkothge et al.

different GFP derivatives now available for use in flower development studies. This chapter will focus on the use of the GUS reporter system. To report on gene expression patterns, two types of GUS gene fusion constructs are typically used. Transcriptional reporters consist of the promoter region of a gene of interest driving GUS expression and thus report on promoter activity. Translational reporters consist of an in-frame fusion of a gene of interest to GUS under the control of the promoter region. Translational reporters are often in a genomic context with 5′ upstream sequence, exons and introns of the gene of interest, and 3′ UTR sequence. Both traditional and Gateway compatible GUS plasmids are available from the Arabidopsis Biological Resource Center (ABRC) (https://abrc.osu.edu). In addition, a synthetic GUS reporter (GUSPlus), based on the sequence from Staphylococcus sp., is expressed specifically in transgenic plant cells and not bacterial cells due to the insertion of an intron within the coding sequence [1]. GUSPlus exhibits higher sensitivity compared with E. coli GUS. Plasmids containing GUSPlus are available from the ABRC or Addgene (https://addgene.org/Claudia_Vickers/). For GUS reporter fusions to reflect endogenous gene expression patterns, the constructs must contain all of the required gene regulatory regions. Since in many cases this is not known, if possible one should confirm that the GUS expression pattern matches the gene expression pattern that is determined using a direct method (in situ hybridization or immunolocalization). Alternatively, one can determine whether a translational reporter fusion complements the corresponding mutant. The GUS enzyme is a hydrolase that can cleave many different β-glucuronides. Different GUS substrates are commercially available for both qualitative histochemical assays and quantitative fluorometric assays. Flower development studies typically involve histochemical assays utilizing the substrate 5-bromo-chloro-3indolyl glucuronide (X-gluc), which produces a blue precipitate at the site of GUS activity. Cleavage of X-gluc does not directly result in the indigo pigment; instead the hydrolysis product must undergo an oxidative dimerization to form the precipitate. This blue precipitate becomes visible in the plant tissue after bleaching of chlorophyll from the tissue using ethanol. Tissue can either be viewed as a whole mount (Fig. 1a) or can be embedded into Paraplast and sectioned for higher resolution of the GUS signal. Under light field illumination, the GUS signal appears blue (Fig. 1b) while under dark-field illumination, the GUS signal appears pink (Fig. 1c). Quantitative GUS assays utilize 4-methyl umbelliferyl glucuronide (MUG), a fluorogenic substrate that can be used to detect GUS activity in tissue extracts as well as intact tissue. Cleavage of MUG by GUS produces the fluorescent compound 4-methyl umbelliferone (MU), which is easily quantified

GUS Reporter System

353

Fig. 1 GUS staining of plant tissues. (a) GUS stained inflorescence of a translational GUS reporter AIL6m:gAIL6-GUS-3′. Young flowers in center of inflorescence show GUS activity. (b, c) GUS stained and sectioned AIL6m:gAIL6-GUS-3′ inflorescence with GUS activity in young stage 2–4 flowers under bright field (b) and dark field illumination (c). (d) GUS staining indicates the presence of a transgene in soybean hairy root transformations. (e) Different gene expression levels are indicated by intensity of the GUS staining in rice callus 24 h after gene gun bombardment. (f) Transposition of the mPing transposable element is marked by the GUS stained blue segment in a wheat leaf. Abbreviations: st 2, stage 2 flower; st 3, stage 3 flower; st 4, stage 4 flower

using a fluorometer. The protocols described here are adapted from previously published protocols [2–5] and have been used for Arabidopsis and rice inflorescences [6, 7]. Similar protocols have been used for other plants including maize, soybean, and sorghum [8– 10]; these protocols differ in their use of a staining solution that includes 20% methanol and Triton X-100 (0.06–0.2%). Methanol at a concentration of 20% volume has been shown to eliminate endogenous GUS activity in plant tissues [11]. Besides providing information on gene expression patterns, GUS reporter constructs can also be used to investigate gene regulation and to identify new genes involved in flower development. Cis-acting regulatory elements can be identified by generating a series of transcriptional GUS fusions containing different lengths of 5′ sequence and determining which fusions reproduce the endogenous gene expression pattern [12, 13]. Comparison of GUS expression in transcriptional and translational fusions can be used to determine if a gene is regulated primarily at the

354

Janaki S. Mudunkothge et al.

transcriptional level and whether cis-regulatory elements are present in introns, coding region, or 3′ UTR sequence [14]. GUS reporter lines can also be used in gene identification strategies either as gene traps/enhancer traps to identify genes that are expressed in a particular tissue or at a particular developmental stage [15, 16] or as material for forward genetic screens to identify genes required for expression of the reporter [17]. Gene trap strategies have identified cell and tissue-specific marker lines useful in the characterization of flower development in mutant plants [18]. In addition to the purposes described in the preceding text, GUS reporter constructs are also useful as screenable markers to indicate the presence or state of a transgene. This use is important when developing transformation protocols, monitoring transposon activity, and performing cell lineage analyses. As flower development studies extend beyond traditional models such as Arabidopsis, snapdragon, petunia, tomato, maize, and rice, the development of robust transformation protocols for other plants will be essential for understanding gene function. In such cases, a strong constitutive promoter (i.e., 35S CaMV) is typically used to drive GUS expression. A small sample of tissue can be quickly and inexpensively stained for GUS activity to detect the presence of the transgene (Fig. 1d, e). Using our simplified GUS screening assay (Subheading 3.1.4), a large number of progeny can be screened for the transgene, allowing for selection of transgenic lines and estimation of transgene copy number. This tool is also important to determine whether germ line tissues are transgenic [8, 19, 20]. A modified version of the 35S:GUS construct with the addition of a transposable element between the promoter and GUS gene has been used to monitor transposable element behavior (Fig. 1f) [21–23]. Furthermore, these sorts of constructs have been used in sector boundary analyses to identify the number of cells giving rise to a flower and each floral organ [24] and clonal analysis of tomato lateral roots to study the pattern of cell division and organization in the root meristem [25]. Thus, the GUS system can be adapted to a variety of purposes in a broad range of species.

2

Materials

2.1 Histochemical GUS Assay 2.1.1

GUS Staining

Use ultrapure water to prepare all solutions. Dispose of acetone, formaldehyde, and xylenes according to waste disposal regulations. 1. 1 M Na2HPO4. 2. 1 M NaH2PO4. 3. 100 mM Potassium ferricyanide (K3Fe(CN)6) (see Note 1). 4. 100 mM Potassium ferrocyanide (K4Fe(CN)6) (see Note 1).

GUS Reporter System

355

5. 100 mM X-Gluc (see Note 2). 6. 90% acetone. 7. Rinse solution: 50 mM sodium phosphate buffer pH 7.2, 0.5 mM K3Fe(CN)6, 0.5 mM K4Fe(CN)6. For 100 mL, add 3.42 mL of 1 M Na2HPO4, 1.58 mL of 1 M NaH2PO4, 500 μL of 100 mM K3Fe(CN)6, and 500 μL of 100 mM K4Fe(CN)6 to a graduated cylinder and add water to 100 mL. Prepare fresh before use. 8. Stain solution: 50 mM sodium phosphate buffer pH 7.2, 0.5 mM K3Fe(CN)6, 0.5 mM K4Fe(CN)6, 2 mM X-gluc. To 10 mL of Rinse solution, add 200 μl of 100 mM X-Gluc stock. Prepare fresh before use (see Note 3). 9. 20 mL glass scintillation vials. 10. Wire mesh screen. 11. Vacuum oven. 12. Ethanol. 2.1.2 Embedding/ Sectioning/Mounting

1. Xylenes 2. Paraplast tissue embedding medium 3. Microscope slide warming table (Eberbach Corporation) 4. Microtome 5. Tissue floatation bath 6. Slide warmer 7. Superfrost plus glass slides 8. Permount mounting medium

2.2 Fluorometric GUS Assay 2.2.1

Protein Extracts

Use ultrapure water to prepare all solutions.

1. Micropestles that fit microcentrifuge tubes 2. 1 M Na2HPO4. 3. 1 M NaH2PO4. 4. β-mercaptoethanol. 5. 0.5 M ethylenediaminetetraacetic acid (EDTA) pH 8.0. 6. 10% (w/v) sodium dodecyl sulfate (SDS). 7. 10% (v/v) Triton X-100. 8. GUS extraction buffer: 50 mM sodium phosphate buffer pH 7, 10 mM β-mercaptoethanol, 10 mM EDTA, 0.1% SDS, 0.1% Triton X-100. For 10 mL, add 288.5 μL 1 M Na2HPO4, 211.5 μL 1 M NaH2PO4, 7 μL β-mercaptoethanol, 200 μL 0.5 M EDTA, 100 μL SDS, and 100 μL Triton X-100 to a 10 mL graduated cylinder and add water to 10 mL (see Note 4).

356 2.2.2

Janaki S. Mudunkothge et al. Fluorometric Assay

1. 25 mM MUG (see Note 5). 2. GUS assay buffer: 50 mM sodium phosphate buffer pH 7, 10 mM β-mercaptoethanol, 10 mM EDTA, 0.1% SDS, 0.1% Triton X-100, 1 mM MUG. To 5 mL of GUS extraction buffer, add 200 μL of 25 mM MUG stock. Prepare fresh before use. 3. 1 mM MU (see Note 6). 4. Stop buffer: 0.2 M Na2CO3. 5. Fluorometer (see Note 7). 6. Fluorometer cuvettes. 7. 1 mg/mL bovine serum albumin (BSA). 8. Bradford solution for protein concentration determination. 9. Spectrophotometer.

3

Methods

3.1 Histochemical GUS Assay

1. Add 10 mL of cold 90% acetone to scintillation vials and store vials on ice.

3.1.1

2. Harvest inflorescence tissue into acetone vials on ice. Incubate vials on ice for 15 min.

GUS Staining

3. Remove acetone completely (see Note 8) and replace with approximately 10 mL of rinse solution. Swirl vials gently and incubate at room temperature for 5 min. 4. Remove Rinse solution completely and replace with Stain solution. Use 2 mL of Stain solution for every 10 Arabidopsis inflorescences collected. The tissue should be completely covered by the Stain solution. 5. Vacuum infiltrate the samples by putting the vials (with loose lids) into a vacuum oven and bring the pressure to between 10 and 15 inHg. After 10 min, slowly release the vacuum and swirl the tissue gently to get rid of air bubbles. Repeat the vacuum infiltration for a total of four times, 10 min each. 6. Close the lids of the vials and incubate at 37 °C for several hours to several days (see Note 9). 7. Remove the stain solution completely and process the samples through an ethanol series: 15%, 30%, 50%, 70%, 85%, 95%, and 100% for 30 min each at room temperature (see Note 10). At the end of the ethanol series, there may still be some chlorophyll in the tissue. By the next day, all of the chlorophyll should be gone and the blue precipitate visible. 8. Examine the tissue under a dissecting microscope (see Note 11). Fill a depression slide with 100% ethanol, place the

GUS Reporter System

357

inflorescence tissue in the ethanol, orient and dissect the tissue as needed using a dissecting microscope. Take pictures with a digital camera connected to the dissecting microscope. 9. Better resolution of the GUS staining pattern can be achieved by examining tissue sections of the material. The procedure for embedding and sectioning tissue is described in the following text. 3.1.2

Embedding

1. Remove the 100% ethanol and process the tissue samples through an ethanol/xylenes series: 75% ethanol/25% xylenes, 50% ethanol/50% xylenes, 25% ethanol/75% xylenes for 30 min each at room temperature (see Note 12). 2. Replace the last ethanol/xylenes mixture with 100% xylenes and leave for 1 h at room temperature. After 1 h, replace the 100% xylenes with fresh 100% xylenes and incubate for 1 more hour. 3. Remove the last xylenes and fill the vials halfway with 100% xylenes. Fill the remaining space in the vial with Paraplast chips. Incubate the vials at 42 °C until all chips are in solution. Pour out the xylenes/Paraplast mix into a waste container and replace with 100% molten Paraplast. Incubate the vials in an oven at 60 °C (see Note 13). 4. Incubate the vials overnight at 60 °C and replace the molten Paraplast a total of six times within 2 days (3–4 times per day with 3–4 h between Paraplast changes) (see Note 14). 5. Heat up microscope slide warming table. Place plastic molds, weigh boats, or paper boats made using 3 × 5 cards coated with transparent tape on the hot side of the slide warming table. 6. Remove vial from 60 °C, swirl briefly, and quickly pour tissue into boat on slide warming table. Add more molten Paraplast as needed to cover all of the tissue. Distribute and orient inflorescences in boats using a wooden handled teasing needle and carefully slide the boat to a cooler region of slide warming table (see Note 15). Repeat tissue orientation and continue sliding boat to coolest side of slide warming table. Leave here for several minutes until Paraplast has solidified a bit around the tissue. Then move the boat to ambient temperature surface and leave for several hours until fully solidified. Store tissue blocks at 4 °C until ready to section.

3.1.3 Sectioning and Mounting

1. Cut individual inflorescences out of the tissue block with a razor blade. Trim away excess Paraplast and mount tissue on tissue holder with either inflorescence straight up and down (for longitudinal sections) or pointing directly at you (for transverse sections).

358

Janaki S. Mudunkothge et al.

2. Cut sections through the entire inflorescence taking sections at 8–10 μm and forming a long ribbon. Transfer the ribbon to a piece of black paper using two paintbrushes, one at either end of the ribbon. Using a razor blade, cut the ribbon into pieces that will fit widthways on a microscope slide. Transfer each ribbon piece to a water bath using a wood applicator, maintaining the pieces in order (see Note 16). 3. Put a Superfrost Plus microscope slide into the waterbath at a 45° angle and push the batch of ribbon pieces onto the slide using a wood applicator. 4. Bake the slides on a slide warmer at 42 °C overnight. Store slides at 4 °C until ready to mount. 5. To remove the Paraplast, incubate slides in 100% xylenes for 10 min. Repeat with a second batch of 100% xylenes for 10 min. 6. Remove one slide at a time from xylenes and blot briefly on paper towels. Add two drops of Permount and cover with a coverslip, carefully pressing out air bubbles. Let the slides dry overnight in the hood. 7. Once slides are dry, wipe them clean with a tissue wipe soaked in xylenes to remove any excess mounting material. 8. Examine slides on compound microscope and take pictures. If GUS staining is intense, a blue color is apparent under brightfield illumination. Weaker GUS staining can be observed under dark-field illumination and appears pink. 3.1.4 Alternative Simplified GUS Screening Assay

1. Harvest plant tissues and place in labeled, appropriately sized tubes. 2. Add enough Stain solution to completely cover the tissue. 3. Close the tubes and incubate 12–24 h at 37 °C. 4. Remove the Stain solution completely and cover the tissue with 70% ethanol. 5. Incubate for up to 24 h, replacing the 70% ethanol solution every few hours as needed to completely remove chlorophyll.

3.2 Fluorometric GUS Assay 3.2.1 Preparation of Protein Extracts

1. Fill 1.5 mL microcentrifuge tube on ice with liquid N2. Harvest tissue into tube and grind tissue with a chilled micropestle until it is a fine powder. 2. Add 150 μL of GUS extraction buffer to tube while tissue is still frozen. Leave sample on ice. Grind tissue further when buffer has thawed. 3. Spin tube in microcentrifuge at 4 °C. 4. Transfer supernatant to a new microcentrifuge tube and keep on ice (see Note 17).

GUS Reporter System 3.2.2

MU Standard Curve

359

1. The fluorometer needs to have the following filter set: excitation 365 nm, emission 455 nm. Turn on fluorometer and let it warm up for about 30 min. 2. Prepare fresh MU dilutions in Stop buffer from 1 mM stock of MU. Typical concentrations are 1 μM, 500 nM, 150 nM, 50 nM, and 20 nM MU. 3. Calibrate the fluorometer using the highest MU standard and measure the fluorescence of the other MU standards to generate a MU standard curve.

3.2.3 MUG Assay on Plant Extracts

1. Aliquot 600 μL of GUS assay buffer to a micro centrifuge tube for each sample. Preheat tubes to 37 °C. 2. For each sample, prepare five 1.5 mL microcentrifuge tubes with 900 μL of Stop buffer. 3. Add 10 μL of plant protein extract to preheated Assay buffer. Vortex to mix. 4. Immediately remove 100 μL from the reaction and add to 900 μL of Stop buffer. 5. After 1 min, remove another 100 μL from the reaction and add to 900 μL of Stop buffer. 6. Repeat removals of 100 μL of reaction at 5 min, 10 min, and 15 min (see Note 18). 7. Measure fluorescence of each sample in a fluorometer. 8. Plot the fluorescence of each sample versus minutes and fit the data to a line. The unit of slope is fluorescent units/minute. 9. Divide the fluorescent units/minute slope of each plant extract by the slope of the MU standard curve to calculate the amount of MU generated in each sample per unit of time (nmol MU/min). 10. To calculate GUS activity in nmol MU/min/mg, divide the nmol MU/min values by the total mg of protein used in each assay as determined in the Bradford Assay in the following text.

3.2.4

Bradford Assay

1. Prepare dilutions of BSA stock solution: 0.2 mg/mL, 0.4 mg/ mL, 0.6 mg/mL, 0.8 mg/mL, and 1.0 mg/mL. 2. Prepare diluted Bradford solution (1 part dye reagent and 4 parts ultrapure water) and transfer 1 mL to microcentrifuge tubes. You need one tube for each BSA stock and each plant sample. 3. Add 20 μL of BSA stock solution or 20 μL of plant protein extract to a microcentrifuge tube and vortex to mix. 4. Incubate for 10 min at room temperature.

360

Janaki S. Mudunkothge et al.

5. Measure absorbance at 595 nm in plastic cuvettes. 6. Calculate the total amount of protein in the plant samples using the BSA standard curve. 3.2.5 MUG Assay on Intact Inflorescences or Flowers

1. Prepare 96-well microtiter plates containing 100 μL GUS extraction buffer in each well. 2. Harvest tissue into microtiter plate and incubate at 37 °C (see Note 19). 3. Stop reaction by the addition of 50 μL Stop buffer. 4. Transfer 100 μL of each well to a fresh microtiter plate and measure fluorescence. 5. The GUS activity in the tissue can be normalized per tissue weight or per mg of protein if the tissue is subsequently assayed for total protein.

4

Notes 1. Make up 10 mL of these solutions at a time and store in the dark at 4 °C. The potassium ferrocyanide stock will oxidize quickly, do not keep it more than 1–2 months. Potassium ferricyanide and potassium ferrocyanide are used to oxidize the soluble hydrolysis product so that it does not diffuse away from the site of production prior to undergoing the oxidative dimerization. High concentrations of potassium ferricyanide and potassium ferrocyanide can inhibit GUS activity. 2. The X-Gluc stock solution is made up in dimethyl formamide. Work in a hood when using dimethyl formamide. The X-Gluc solution can be stored at -20 °C for several weeks but should be colorless. Do not use the solution if it has turned red. 3. The staining solution may be supplemented with 0.06–0.2% Triton X-100 and 20% Methanol as described in protocols for maize, soybean, and sorghum [8–10]. 4. Add β-mercaptoethanol to GUS extraction buffer right before use. Work in a hood when using β-mercaptoethanol. 5. The MUG stock solution is made up in GUS extraction buffer. Prepare fresh before use. 6. Dissolve 19.8 mg of MU in 100 mL of ultrapure H2O and store in a dark bottle at 4 °C for up to a month. 7. If assay will be performed on intact tissue in microtiter well plates, then a fluorometer with a microplate reader is required. 8. Remove the acetone by pouring it out of the vial while pressing a wire mesh screen against the vial lip to retain the tissue. Remove the last bit of acetone with a pipetor. For all

GUS Reporter System

361

subsequent steps involving removal of solutions from scintillation vials (except those involving Paraplast), use a wire mesh screen to keep the inflorescence tissue in the vial. 9. The staining period includes the vacuum infiltration time and the incubation period at 37 °C. The exact time needed for staining depends on the individual line. Typically samples are incubated from several hours to overnight. 10. For better tissue preservation, samples can be fixed in FAA (50% ethanol, 10% glacial acetic acid, 3.7% formaldehyde) in place of the 50% ethanol. Then continue with the ethanol series. 11. There have been reports of GUS staining in pollen grains due to diffusion of the primary GUS reaction product from anthers [26]. To confirm GUS staining in pollen, isolate pollen grains from the anther before staining the tissue with GUS and stain them separately for GUS activity. In addition, some SAIL T-DNA insertion lines carry a LAT52:GUS marker which will result in GUS staining in pollen grains [27]. This construct could potentially result in gene silencing of other GUS reporters that might be introduced into the T-DNA line. 12. All work with xylenes should be performed in a fume hood. Exposure of tissues to xylenes must be minimized since xylenes can solubilize the blue precipitate. 13. The transition of tissue from 100% ethanol to 100% Paraplast should take place in a single day. 14. After removing the vial from the 60 °C oven, place it on your hand for a short time to cool the bottom before pouring out the molten Paraplast. A thin layer of partially solidified Paraplast will keep the tissue from coming out of the vial. 15. Do not leave tissue too long on hot part of warming table as bubbles will form around the tissue and interfere with getting good sections. 16. Place the ribbon on the paper such that the shiny side faces up. Wet the tip of the wood applicator in the water bath, then touch it to the ribbon piece to pick up the piece. Flip the wood applicator over so that the piece is on top and slowly put the wood applicator into the water bath. The ribbon piece should come off the wood applicator and float in the bath. Carefully push the ribbon piece up against the side of the water bath. It should stick there. Then transfer the second piece and push it up against the first piece and repeat with each additional piece. They will gently stick to each other and the order can be preserved. 17. Tissue extracts can be stored at -70 °C.

362

Janaki S. Mudunkothge et al.

18. More than one sample can be processed at a time but the different samples will need to be staggered by appropriate periods of time. 19. The exact time needed for incubation in the MUG solution depends on the individual line. Weaker reporter lines may require up to 16 h [28]. References 1. Broothaerts W, Mitchell HJ, Weir B et al (2005) Gene transfer to plants by diverse species of bacteria. Nature 433:629–633 2. Blazquez M (2002) Quantitative GUS activity assays. In: Weigel D, Glazebrook J (eds) Arabidopsis: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, pp 249–252 3. Bomblies K (2002) Whole-Mount GUS staining. In: Weigel D, Glazebrook J (eds) Arabidopsis: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, pp 243–248 4. Jefferson RA (1987) Assaying chimeric genes in plants: the GUS gene fusion system. Plant Mol Biol Rep 5:387–405 5. Jefferson RA, Kavanagh TA, Bevan MW (1987) GUS fusions: β-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO J 6:3901–3907 6. Krizek BA (2015) Intronic sequences are required for AINTEGUMENTA-LIKE6 expression in Arabidopsis flowers. BMC Res Notes 8:556 7. Prasad K, Kushalappa K, Vijayraghavan U (2003) Mechanism underlying regulated expression of RFL, a conserved transcription factor, in the developing rice inflorescence. Mech Dev 120:491–502 8. Zhang Z, Xing A, Staswick P, Clemente TE (1999) The use of glufosinate as a selective agent in Agrobacterium-mediated transformation of soybean. Plant Cell Tissue Organ Cult 56:37–46 9. Dwivedi KK, Roche DJ, Clemente TE et al (2014) The OCL3 promoter from Sorghum bicolor directs gene expressio to abscission and nutrient-transfer zones at the base of floral organs. Ann Bot 114:489–498 10. Wang H, Fan M, Wang G et al (2017) Isolation and characterization of a novel pollen-specific promoter in maize (Zea mays L.). Genome 60: 485–495 11. Kosugi S, Ohashi Y, Nakajima K, Arai Y (1990) An improved assay for β-glucuronidase in transformed cells: methanol almost completely

suppresses a putative endogenous β-glucuronidase activity. Plant Sci 70:133–140 12. Hill TA, Day CD, Zondlo SC et al (1998) Discrete spatial and temporal cis-acting elements regulate transcription of the Arabidopsis floral homeotic gene APETALA3. Development 125:1711–1721 13. Tilly JJ, Allen DW, Jack T (1998) The CArG boxes in the promoter of the Arabidopsis floral organ identity gene APETALA3 mediate diverse regulatory effects. Development 125: 1647–1657 14. Lee J-Y, Colinas J, Wang JY et al (2006) Transcriptional and posttranscriptional regulation of transcription factor expression in Arabidopsis roots. Proc Natl Acad Sci U S A 103:6055– 6060 15. Springer PS (2000) Gene traps: tools for plant development and genomics. Plant Cell 12: 1007–1020 16. Sundaresan V, Springer P, Volpe T et al (1995) Patterns of gene action in plant development revealed by enhancer trap and gene trap transposable elements. Genes Dev 9:1797–1810 17. Chiu W-H, Chandler JW, Cnops G et al (2007) Mutations in the TORNADO2 gene affect cellular decisions in the peripheral zone of the shoot apical meristem of Arabidopsis thaliana. Plant Mol Biol 63:731–744 18. Liljegren SJ, Ditta GS, Eshed Y et al (2000) SHATTERPROOF MADS-box genes control seed dispersal in Arabidopsis. Nature 404:766– 770 19. Clemente TE, LaVallee BJ, Howe AR et al (2000) Progeny analysis of glyphosate selected transgenic soybeans derived from Agrobacterium-mediated transformation. Crop Sci 40: 797–803 20. Murray F, Brettell R, Matthews P et al (2004) Comparison of Agrobacterium-mediated transformation of four barley cultivars using the GFP and GUS reporter genes. Plant Cell Rep 22:397–402 21. Fitzmaurice WP, Lehman LJ, Nguyen LV et al (1992) Development and characterization of a generalized gene tagging system for higher

GUS Reporter System plants using an engineered maize transposon Ac. Plant Mol Biol 20:177–198 22. Fitzmaurice WP, Nguyen LV, Wernsman EA et al (1999) Transposon tagging of the Sulfur gene of tobacco using engineered maize Ac/Ds elements. Genetics 153:1919–1928 23. Qu S, Jeon JS, Ouwerkerk PB, Bellizzi M et al (2009) Construction and application of efficient Ac-Ds transposon tagging vectors in rice. J Integr Plant Biol 51:982–992 24. Bossinger G, Smyth DR (1996) Initiation patterns of flower and floral organ development in Arabidopsis thaliana. Development 122: 1093–1102

363

25. Lee G, Rodgers L, Taylor BH (1995) β-Glucuronidase as a marker for clonal analysis of tomato lateral roots. Transgenic Res 4:123– 131 26. Mascarenhas JP, Hamilton DA (1992) Artifacts in the localization of GUS activity in anthers of petunia transformed with a CaMV 35S-GUS construct. Plant J 2:405–408 27. Sessions A, Burke E, Presting G et al (2002) A high throughput Arabidopsis reverse genetics system. Plant Cell 14:2985–2994 28. Blazquez MA, Soowal LN, Lee I, Weigel D (1997) LEAFY expression and flower initiation in Arabidopsis. Development 124:3835–3844

Chapter 19 Expression and Functional Studies of Leaf, Floral, and Fruit Developmental Genes in Non-model Species Natalia Pabo´n-Mora, Harold Sua´rez-Baron, Yesenia Madrigal, Juan F. Alzate, and Favio Gonza´lez Abstract Researchers working on evolutionary developmental plant biology are inclined to choose non-model taxa to address how specific features have been acquired during ontogeny and fixed during phylogeny. In this chapter we describe methods to extract RNA, to assemble de-novo transcriptomes, to isolate orthologous genes within gene families, and to evaluate expression and function of target genes. We have successfully optimized these protocols for non-model plant species including ferns, gymnosperms, and a large assortment of angiosperms. In the latter, we have ranged a large number of families including Aristolochiaceae, Apodanthaceae, Chloranthaceae, Orchidaceae, Papaveraceae, Rubiaceae, Solanaceae, and Tropaeolaceae. Key words Expression analyses, Functional analyses, Non-model plants, Orthologous genes identification, Transcriptomics

1

Introduction Most protocols for RNA extraction, small- and large-scale gene expression analyses, and functional studies have been standardized in the model angiosperm Arabidopsis thaliana. Nevertheless, the current questions and approaches in developmental plant biology require to go beyond the model species to assess gene functional evolution across plant lineages. The current phylogenetic framework of seed plants has laid the foundation of key questions in evolution and development including, among others, the origin and homology of the ovule integument(s) in gymnosperms and angiosperms [1]; the identity of floral parts and in particular the homology of the perianth across flowering plants [2–4]; the genetic bases underlying floral symmetry [5, 6]; the diversity of fruit types and seed dispersal mechanisms [7, 8], and the fusion of floral organs resulting in synorganized flowers [9].

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_19, © Springer Science+Business Media, LLC, part of Springer Nature 2023

365

366

Natalia Pabo´n-Mora et al.

Frequently, researchers aim to isolate and identify orthologs of previously characterized genes in non-model plants that often lack reference genomes or transcriptomes. Such comparative analyses require standardization of RNA extraction methods, large-scale sequencing, contig assemblage, homologous gene searches, and methods to evaluate targeted gene expression and function. Here we compile several protocols successfully standardized in non-model plants from initial tissue collection to transcriptome generation and assemblage, targeted gene searches, expression (small and large scale), and functional analyses. We explain in detail RNA-seq transcriptomic analyses in which high-throughput sequencing and computational methods are applied to identify candidate genes and quantify their relative abundance in a specific tissue or floral organ. Developments in RNA-seq methods have provided a complete characterization of RNA transcripts, including improvements in transcription site mapping, strand-specific measurements, small RNA characterization, detection of gene fusion and alternative splicing events, and identification of orthologous genes from model and non-model species [10, 11]. RNA-seq provides the best cost-effective strategy for developing a reference set of orthologous genes in non-model plant species lacking a sequenced genome [12]. Finally, we carefully describe two protocols for functional analyses using the virus-induced gene silencing (VIGS) technique. VIGS relies on the post-transcriptional downregulation occurring as a result of the plant infection with a recombinant virus [13, 14]. Plants with intact RNAi Silencing Complex (RISC) machineries respond to viral infection in three major steps: (1) generating complementary RNA to the virus with the RNA-dependent RNA polymerase, (2) promoting the activation of DICER-like enzymes in response to the double-stranded RNA, and (3) downstream activation of RISC where siRNAs recruited by the complex can bind to complementary RNA molecules triggering cleavage or blocking translation in the ribosome [15]. This reversegenetics technique has been extensively implemented in model and non-model plants [16, 17].

2

Materials Prepare all solutions with diethyl pyrocarbonate (DEPC)-treated water or autoclaved milliQ water to avoid RNA degradation. All glassware and plasticware should be RNase-free and autoclaved. Always use filter tips and manipulate the material with sterile gloves. Carefully follow all waste disposal regulations when disposing waste material.

Expression and Function of Genes in Non-model Species

2.1 Small-Scale Expression Analyses

1. Sterile tubes (1.5 mL, 5 mL, 15 mL, 50 mL).

2.1.1 Plant Material Collection

3. Scissors and autoclaved dissecting forceps.

2.1.2

1. Liquid nitrogen.

Total RNA Isolation

367

2. Liquid nitrogen. 4. Permanent markers.

2. RNase-free and autoclaved centrifuge 1.5 mL tubes. 3. Total RNA reagent Trizol. 4. Chloroform. 5. 70% ethanol. 6. 100% ethanol. 7. Isopropyl alcohol. 8. 5 M NaCl. 9. DEPC-treated water or nuclease-free water. 10. RQ1 RNase-Free DNase, 10× Reaction Buffer and stop solution. 11. Refrigerated centrifuge. 12. Micropipettes. 13. Sterile filter tips. 14. Vortex. 15. Centrifuge sterile 1.5 mL tubes. 16. Nuclease-free mortar and pestle. 2.1.3

RNA Quality Test

1. Sterile filter tips. 2. Micropipettes. 3. NanoDrop (e.g., NanoDrop™ 2000/2000c; Thermo Fisher Scientific) (Quantus or Qubit Fluorometers could also be used). 4. Bioanalyzer (Agilent Technologies).

2.1.4 Sample Submission to the Sequencing Facilities

1. Dry ice. 2. Centrifuge sterile 1.5 mL tubes. 3. DEPC treated water. 4. Absolute ethanol. 5. Parafilm.

2.1.5

Primer Design

Primers can be designed manually or using online servers and programs (see below). 1. https://www.genscript.com/tools/real-time-pcr-tagmanprimer-design-tool

368

Natalia Pabo´n-Mora et al.

2. http://bioinfo.ut.ee/primer3-0.4.0/ 3. https://www.thermofisher.com/co/en/home/life-science/ oligonucleotides-primers-probes-gnes/custom-dna-oligos/ oligo-design-tools/oligoperfect.html 2.1.6

cDNA Synthesis

1. SuperScript™ III First-Strand Synthesis System (see Note 1). 2. RNase H (5 U/μL). 3. Sterile 0.2 mL tubes. 4. Micropipettes. 5. Sterile filter tips.

2.1.7

RT-PCR

1. EconoTaq® PLUS GREEN 2× Master Mix or a similar system (see Note 2). 2. Nuclease-free water. 3. Bovine Serum Albumin (BSA) (2.5 μg/μL). 4. Q solution-Betaine (1.2 M). 5. Forward and reverse specific primers (10 mM each). 6. Template cDNA. 7. Sterile 0.2 mL tubes. 8. Sterile filter tips.

2.1.8 Visualization and Result Readings

1. Agarose molecular biology grade. 2. TAE 1× buffer: can be prepared from 50× TAE concentrate. To prepare 50× TAE add 242 g of Tris base, 57.1 mL of Glacial Acetic Acid, 18.6 g of Ethylenediaminetetraacetic acid (EDTA), and adjust volume to 1 L with milliQ H2O and adjust pH to 8.0. 1× TAE is a 10 mL of concentrate 50× TAE in 490 mL of distilled water. 3. 100 bp and 1 kb DNA ladder. 4. Ethidium bromide solution, 2 μg/mL, or GelRed® Nucleic Acid Gel Stain. 5. BioDoc Analyzer (other similar equipment can be used). 6. Electrophoresis chamber. 7. Micropipettes. 8. Sterile filter tips.

2.1.9

qRT-PCR

1. Fresh cDNA. 2. Forward and reverse primers (1 mM). 3. White PCR plates. 4. MicroAmp™ Optical Adhesive Film. 5. Micropipettes.

Expression and Function of Genes in Non-model Species

369

6. Sterile filter tips. 7. Maxima SYBR Green/ROX qPCR Master Mix or a similar system. 8. Diluted cDNA template. 9. Nuclease-free water. 10. qPCR qTower3.0 system and the qPCRsoft software (other similar equipment can be used). 2.2 Virus-Induced Gene Silencing (VIGS)

1. TRV2-LIC vector (https://abrc.osu.edu/stocks/number/ CD3-1867).

2.2.1

2. Gene-specific primers (see Subheading 3.6.1, step 1).

Vector Construction

3. TaKaRa Taq™ system or a similar PCR system with proofreading Taq (having endonuclease activity). 4. PCR product clean-up system. 5. PstI restriction enzyme 20,000 U/mL and buffer. 6. T4 DNA polymerase 3000 U/mL and buffer. 7. BSA 10% w/v. 8. dTTP 100 mM. 9. dATP 100 mM. 10. Electrocompetent E. coli cells. 11. 50 mg/mL Kanamycin in water (stock solution). 12. QIAprep plasmid purification system. 2.2.2 Plant Growth and Agroinfiltration

1. Seedlings or juvenile plants. 2. Agrobacterium strain GV3101. 3. SOC medium: 2% tryptone, 0.5% yeast extract, 10 mM NaCl, 2.5 mM KCl, 10 mM MgCl2, 10 mM MgSO4, and 20 mM glucose. 4. 50 mg/mL Gentamycin in water (stock solution). 5. 25 mg/mL Rifampicin in methanol (stock solution). 6. LB plates with Kananamycin (50 μg/mL), Rifampicin (25 μg/ mL), and Gentamycin (50 μg/mL) (Kan50-Rif25-Gen50). 7. 15 mL Falcon tubes. 8. 60% glycerol. 9. LB liquid medium with Kananamycin (50 μg/mL), Rifampicin (25 μg/mL), and Gentamycin (50 μg/mL) (Kan50-Rif25Gen50). 10. Cryogenic tubes. 11. 1 M MES: add 0.1952 gr of 2-(N-morpholino) ethane sulfonic acid (MES) per mL of water, making sure it is fully dissolved prior to use.

370

Natalia Pabo´n-Mora et al.

12. 1 M MgCl2: add 0.2033 gr of magnesium chloride hydrate per mL of water, making sure it is fully dissolved prior to use. 13. 0.1 M Acetosyringone: add 0.0196 gr of 4′-Hydroxy-3′,5′-dimethoxyacetophenone (Acetosyringone) per mL of methanol, making sure it is fully dissolved prior to use. 14. Agrobacterium resuspension medium: prepare by mixing 1 mL of 1 M MES, 1 mL of 1 M MgCl2, and 200 μL of 0.1 M Acetosyringone stocks, and deionized water up to 100 mL. 15. 5% (w/v) sucrose (depending on the plant species to infiltrate). 16. Electroporator system and cuvettes. 17. 0.3 mL insulin syringes and ɸ8.15 or ɸ 10.85 and a ½ inch needles. 18. Light carts. 2.3 RNA-seq Data Analyses

1. PRINSEQ-LITE (v0.20.4) software; it can be downloaded from http://printseq.sourceforge.net. 2. Trimmomatic software; available at http://www.usadellab. org/cms/?page=trimmomatic. 3. Trinity software; the complete package can be downloaded from https://github.com/trinityrnaseq/trinityrnaseq/ releases.

2.4 Differentially Expressed Genes (DEGs) Identification

1. Kallisto software; available at https://pachterlab.github.io/ kallisto/ and https://github.com/griffithlab/rnaseq_tuto rial/wiki/Kallisto. 2. DESeq2 software and additional information can be found at: http://bioconductor.org/packages/release/bioc/html/ DESeq2.html.

3

Methods

3.1 Experimental Design

Selective sampling of plant tissue (i.e., meristems, floral structures, fruits, or leaves) is essential to succeed at de novo sequencing projects. If the aim is the identification of candidate orthologous genes, the RNA from pooled plant tissue samples (i.e., meristems, leaves, floral buds, and fruits in a single extraction) might be the best option for sequencing of expressed transcripts. Our results from a broad range of neotropical plant species [18–27] have shown that a high-quality assembled transcriptome from pooled tissues and organs and the posterior isolation of orthologous candidate genes can be achieved without technical replicates. On the other hand, if the aim of the study is differential gene expression (DEGs) analyses, developmental stages must be defined by the

Expression and Function of Genes in Non-model Species

371

Fig. 1 Technical and biological replicates in RNA-seq experiments with Aristolochia fimbriata (Aristolochiaceae) flowers (a, b) and Epidendrum fimbriatum (Orchidaceae) meristems (c) as an example. (a) Technical replicates correspond to three (or four) aliquots of the same RNA sample processed in identical experimental conditions (b) Biological replicates correspond to aliquots from different individuals, for instance, flowers from different plants at the same developmental stage, or different plants under the same treatment; in this case, each individual is processed separately. (c) Because of the small size of plant material (50–200 μm), each biological replicate corresponds to a pool from three different individuals, for instance, meristems from different plants at the same developmental stage; in this case, each individual is dissected separately to ensure removal of lateral organs. Some experiments will require a round of biological replicates followed by technical replicates

specific research question and the number of replicates for each sample should be established prior to tissue collection. Scaled, photographic records of each stage are highly recommended. It is also critical to define the number of replicates with the awareness of the differences between technical and biological replicates (Fig. 1). Usually, three biological replicates are recommended. Each RNA extraction will require testing different protocols and will result in a separately assembled library, as will be explained in the following

372

Natalia Pabo´n-Mora et al.

text. In either case, high-quality RNA is required with a good balance between total RNA quality and quantity (A260/280 ratio of 1.8 or above; RNA integrity number (RIN) > 7; amount of total RNA between 2.0 and 4.0 μg). Importantly, low concentrations of total RNA with good RIN coming from small sized samples (e.g., 50–100 μm meristems from the miniature orchid Epidendrum fimbriatum) also yield successful RNA sequencing (Table 1). 3.2 Purification of Total RNA 3.2.1 Plant Material Collection

3.2.2

Total RNA Isolation

The collection of material for expression analyses depends upon the experimental design. If different developmental stages are of interest, collections must include sizes and developmental features desired (Figs. 2 and 3). For example, gene expression studies of target genes possibly involved in fruit patterning can be evaluated in carpels-to-fruits at different developmental stages [18, 25]. Plant dissections can also consider comparing vegetative and reproductive stages during processes like floral transition [23, 24] or sporangia development [27]. In the former scenario, careful dissections of vegetative meristem from leaf primordia and inflorescence meristem from floral bud must be implemented. Dissections from different floral whorls or different portions of a single floral organ at different development stages can also be made [19–22, 25, 26]. If different organ portions are needed, so are appropriate dissections [26]. For example, the expression of candidate genes involved in floral symmetry can be evaluated in RNA extracted from the dorsal and ventral portions of the flower [20]. The material requires rapid manipulation from the living plant to the liquid nitrogen. The tissue needs to be placed in separate properly labelled, pre-chilled Eppendorf tubes. Make sure no liquid nitrogen remains inside the Eppendorf tube, then cap tubes securely and immediately flash freeze in liquid nitrogen. Tubes can be stored at -80 °C (see Note 3). Keep samples on liquid nitrogen to minimize RNA degradation. Perform all centrifugation steps at 4 °C unless indicated otherwise. Wear gloves and use sterile techniques when working with RNA. Use a vertical laminar flow hood to avoid breathing vapors, especially when using phenol-chloroform solutions or 2-mercaptoethanol dilutions (see Note 4). 1. Macerate tissue samples using liquid nitrogen on a nucleasefree mortar and pestle. Add 1 mL of TRIzol Reagent per 50–100 mg of tissue. The sample volume should not exceed 10% of the volume of TRIzol Reagent used for homogenization (see Note 5). 2. Homogenize with brief vortex and incubate the sample for 5 min at room temperature to allow complete dissociation of nucleoprotein complexes (see Note 6).

793.94

402.851

793.94

1494.99

Anemia villosa

Equisetum bogotensis

Equisetum giganteum

RNA concentration (ng/μL)

Adiantum raddianum

Monilophytes (ferns)

Sample

46.34

23.81

12.89

23.81

RNA amount (μg)

1.53

1.43

1.6

1.4

A260/280

7.2

8.2

7.6

8.2

RIN

(continued)

Total length of sequence: 60182177 bp Total number of sequences: 68633 Average contig length is: 876 bp Largest contig: 8004 bp Shor test contig: 201 bp Total GC count: 27381900 bp GC %: 45.50%

Total length of sequence: 60075475 bp Total number of sequences: 57810 Average contig length is: 1039 bp Largest contig: 10098 bp Shor test contig: 201 bp Total GC count: 26877826 bp GC %: 44.74%

Total length of sequence: 91852863 bp Total number of sequences: 87092 Average contig length is: 1054 bp Largest contig: 9315 bp Shor test contig: 201 bp Total GC count: 43644331 bp GC %: 47.52%

Total length of sequence: 76504695 bp Total number of sequences: 64168 Average contig length is: 1192 bp Largest contig: 8085 bp Shor test contig: 201 bp Total GC count: 35045170 bp GC %: 45.81%

Transcriptome general stats

Table 1 Examples of quality control (QC) values from non-model plant transcriptomes resulting in reproducible libraries

Expression and Function of Genes in Non-model Species 373

1466.47

RNA concentration (ng/μL)

79.19

RNA amount (μg)

2.01

A260/280

376.830

45.138

Elleanthus aurantiacus

Epidendrum fimbriatum 1.128

11.305

1.78

2.02

Orchidaceae (organs ground together in a single extraction for a reference transcriptome)

Angiosperms

Welwitschia mirabilis

Gymnosperms

Sample

Table 1 (continued)

8.7

9.7

9.2

RIN

Total length of sequence: 142256180 bp Total number of sequences: 185489 Average contig length is: 766 bp Largest contig: 16666 bp Shor test contig: 201 bp Total GC count: 61194092 bp GC %: 43.02%

Total length of sequence: 100537418 bp Total number of sequences: 91814 Average contig length is: 1095 bp Largest contig: 12295 bp Shor test contig: 201 bp Total GC count: 43698850 bp GC %: 43.47%

Total length of sequence: 136551623 bp Total number of sequences: 108641 Average contig length is: 1256 bp Largest contig: 16255 bp Shor test contig: 201 bp Total GC count: 56385191 bp GC %: 41.29%

Transcriptome general stats

374 Natalia Pabo´n-Mora et al.

267.722

61.870

278.389

104.585

260.292

Gomphichis scaposa

Miltonia roezli

Masdevalia wendlandiana

Maxilaria aurea

Oncidium “Gower Ramsey”

7.809

3.660

13.919

3.094

8.032

1.92

1.86

1.92

1.84

1.91

8.5

9.3

9.2

9

7.5

(continued)

Total length of sequence: 69431269 bp Total number of sequences: 84942 Average contig length is: 817 bp Largest contig: 11821 bp Shortest contig: 201 bp Total GC count: 30942496 bp GC %: 44.57%

Total length of sequence: 72925370 bp Total number of sequences: 68647 Average contig length is: 1062 bp Largest contig: 12136 bp Shortest contig: 201 bp Total GC count: 32517740 bp GC %: 44.59%

Total length of sequence: 78531621 bp Total number of sequences: 75953 Average contig length is: 1033 bp Largest contig: 15559 bp Shortest contig: 201 bp Total GC count: 32591379 bp GC %: 41.50%

Total length of sequence: 64965610 bp Total number of sequences: 59091 Average contig length is: 1099 bp Largest contig: 11852 bp Shortest contig: 201 bp Total GC count: 28249545 bp GC %: 43.48%

Total length of sequence: 85109234 bp Total number of sequences: 78810 Average contig length is: 1079 bp Largest contig: 18109 bp Shortest contig: 201 bp Total GC count: 37794560 bp GC %: 44.41%

Expression and Function of Genes in Non-model Species 375

307.751

Stelis pusilla

31.04

Epidendrum fimbriatum (Shoot Apical Meristem) 1.552

2.220

12.002

RNA amount (μg)

Borojoa patinoi

381.08

14.10

Rubiaceae (organs ground together in a single extraction for a reference transcriptome)

44.394

Epidendrum fimbriatum (Inflorescence Meristem)

Orchidaceae (dissected meristems)

RNA concentration (ng/μL)

Sample

Table 1 (continued)

1.48

1.61

1.51

1.86

A260/280

6.2

9.2

9.1

9.1

RIN

Total length of sequence: 52310506 bp Total number of sequences: 56872 Average contig length is: 919 bp Largest contig: 8186 bp Shor test contig: 201 bp Total GC count: 22424549 bp GC %: 42.87%

Total length of sequence: 126079117 bp Total number of sequences: 110484 Average contig length is: 1141 bp Largest contig: 16986 bp Shor test contig: 284 bp Total GC count: 53296016 bp GC %: 42.27%

Total length of sequence: 148828377 bp Total number of sequences: 128450 Average contig length is: 1158 bp Largest contig: 17775 bp Shor test contig: 290 bp Total GC count: 62269075 bp GC %: 41.84%

Total length of sequence: 90879205 bp Total number of sequences: 109206 Average contig length is: 832 bp Largest contig: 11960 bp Shor test contig: 201 bp Total GC count: 39504321 bp GC %: 43.47%

Transcriptome general stats

376 Natalia Pabo´n-Mora et al.

34.53

986.338

189.551

356.741

Condaminea cor ymbosa

Cof fea arabica

Morinda citrifolia

Palicourea angustifolia 10.7

5.68

35.58

1.968

1.65

1.55

1.62

1.5

9.7

9.6

6.9

7

(continued)

Total length of sequence: 174935973 bp Total number of sequences: 154399 Average contig length is: 1133 bp Largest contig: 12282 bp Shor test contig: 201 bp Total GC count: 73216981 bp GC %: 41.85%

Total length of sequence: 161582814 bp Total number of sequences: 130936 Average contig length is: 1234 bp Largest contig: 15484 bp Shor test contig: 201 bp Total GC count: 65622136 bp GC %: 40.61%

Total length of sequence: 124406074 bp Total number of sequences: 103789 Average contig length is: 1198 bp Largest contig: 8143 bp Shor test contig: 201 bp Total GC count: 52281377 bp GC %: 42.02%

Total length of sequence: 124406074 bp Total number of sequences: 103789 Average contig length is: 1198 bp Largest contig: 8143 bp Shor test contig: 201 bp Total GC count: 52281377 bp GC %: 42.02%

Expression and Function of Genes in Non-model Species 377

RNA concentration (ng/μL)

RNA amount (μg)

986.15

512.576

Brugmansia suaveolens

Cestrum nocturnum 21.52

28.59

Solanaceae (organs ground together in a single extraction for a reference transcriptome)

Sample

Table 1 (continued)

1.82

1.95

A260/280

8.9

8.9

RIN

Total length of sequence: 121964260 bp Total number of sequences: 142459 Average contig length is: 856 bp Largest contig: 11926 bp Shor test contig: 200 bp GC %: 41.13%

Total length of sequence: 109351485 bp Total number of sequences: 94825 Average contig length is: 1153 bp Largest contig: 15820 bp Shor test contig: 234 bp GC %: 41.13%

Transcriptome general stats

378 Natalia Pabo´n-Mora et al.

Expression and Function of Genes in Non-model Species

379

Fig. 2 Gene expression studies in Bocconia frutescens (Papaveraceae) as an example. (a) Floral developmental stages (S0 to S3 and anthesis) as well as fruits (Fr1 and Fr2) and leaves chosen for expression analyses of target MADS-box genes. (b) RT-PCR results for 13 selected floral MADS box genes isolated from the transcriptome; ACTIN (lower row) used as positive control for cDNA synthesis and amplification in all samples. (c) qRT-PCR results for six selected floral MADS box genes in some of the floral organs at specific developmental stages; stamens at developmental stage S3 are used for reference as fold change 1.0 in BofrAP3 and BofrPI2/4, while carpels at developmental stage S3 are used as reference in all other genes tested. Glyceraldehyde 3-phosphate dehydrogenase (GADPH) was used as endogenous control for qRT-PCR. (Figure modified from [19])

380

Natalia Pabo´n-Mora et al.

Fig. 3 Gene expression studies in Cattleya trianae (Orchidaceae) during flower transition as an example. (a) Plant developmental stages including vegetative stages like pseudobulb (PB), axillary bud (AB), leaves (L), and shoot apical meristem (SAM) and reproductive stages like inflorescence meristem (IM), floral bud (FB), and flower in anthesis (F), all chosen for expression analyses of target PEBP (FT and TFL1), SOC1, and SVP genes. (b) RT-PCR results for four selected flowering transition PEBP (FT and TFL1), SOC1, and SVP genes isolated from the transcriptome; ACTIN (lower row) used as positive control for cDNA synthesis and amplification in all samples. (c) qRT-PCR results for four selected PEBP genes in different plant developmental stages. IM is used for reference as fold change 1.0 in all genes tested. 18S was used as endogenous control for qRT-PCR. (Figure modified from [23])

Expression and Function of Genes in Non-model Species

381

3. Add 0.2 mL of chloroform per 1 mL of TRIzol Reagent. Cap tubes securely. Mix by inversion, incubate tubes at room temperature for 2–3 min. Centrifuge the samples at 12,000 g for 15 min at 4 °C. 4. Transfer upper aqueous phase carefully without disturbing the interphase into a fresh tube with 0.5 mL of cold isopropyl alcohol per 1 mL of TRIzol Reagent used for the initial homogenization. This step will precipitate the RNA from the aqueous phase (see Note 7). 5. Incubate samples at 15–30 °C for 10 min and centrifuge at 12,000 g for 15 min at 4 °C. The RNA, often invisible before centrifugation, precipitates as a pellet on the side and bottom of the tube (see Note 8). 6. Remove the supernatant completely and wash the RNA pellet once with 1 mL of cold 70% ethanol per 1 mL of TRIzol Reagent used for the initial homogenization. Flick until the pellet is no longer at the bottom of the tube and gets cleaned by the ethanol. Centrifuge at 7500 g for 5 min at 4 °C. Remove all leftover ethanol and save the pellet. Air-dry RNA pellet for 5–10 min on ice. If ethanol does not fully evaporate take off residual ethanol by pipetting it out of the tube taking care in not disturbing the pellet (see Note 9). 7. Resuspend RNA on 20–40 μL of DEPC-treated water or nuclease-free water by flicking. 8. Set up the DNase digestion reaction using 1 μL of 10× Reaction Buffer and 0.1 μL of the RNase-Free DNase enzyme per 10 μL of RNA. Incubate at 37 °C for 30 min. 9. Add 1 μL of RQ1 DNase Stop Solution per reaction to terminate it and incubate at 65 °C for 10 min to inactivate the DNase. Store the RNA at -80 °C. 3.2.3

RNA Quality Test

An accurate evaluation of the integrity and the amount of RNA is necessary prior to library construction for RNA-seq (Table 1). Degraded or low-quality RNA for next generation sequencing applications can result in failure of the library generation, low yield of reads after sequencing, poor read quality, assembly issues, and finally incomplete and poor-quality transcripts. The integrity of RNA can be determined by the presence of two intact ribosomal RNA bands at approximately 4500 nt and 1900 nt after a gel electrophoresis by loading 8–10 μL of the concentrated sample with 6 μL loading buffer. Indirect measurements of RNA quantity and purity can also be obtained by spectrometry, generally using a NanoDrop device. Here, OD 260/280 is expected to be between 1.8 and 2.0, and OD 260/230 must be as close to 2.0 as possible in RNA samples of optimal quality (see Note 10). More accurate measurements can be obtained directly with fluorescence-

382

Natalia Pabo´n-Mora et al.

based nucleic acid quantification methods as they are more sensitive than absorbance-based methods (i.e., 200× more sensitive than NanoDrop™). For instance, the Quantus™ (Promega Corporation) or the Qubit 4 Fluorometer (Thermo Fisher Scientific) can be used. Sequencing facilities however will require the use of capillary electrophoresis systems, such as the Bioanalyzer (Agilent Technologies), to check the RNA integrity. The RNA Integrity Number (RIN) is commonly used as a reference measure of integrity with minimum recommended threshold of at least 7. This procedure separates RNA molecules according to their size, and a typical pattern with a 28S rRNA band at 4.5 k, which is twice the intensity of the 18S rRNA band at 1.9 kb, is observed when the total RNA is in optimal conditions. 3.2.4 Sample Submission to the Sequencing Facilities

Each sequencing core, laboratory, or facility has its own protocol for sample submission that needs to be followed. However, usually RNA samples are submitted frozen in DEPC treated water or precipitated in absolute ethanol (see Note 9) in clearly labeled 1.5- or 2.0-mL microcentrifuge tubes sealed and tightly wrapped in parafilm. Each tube can further be placed in individual 5 × 5 cm Ziploc bags together with a label written in pencil in case there are any undesired leaks. Several tubes in individual bags can be accommodated safely in a 50 mL Falcon tube. Dry ice is mandatory for shipping; 3–5 kg of dry ice should last for 24–48 h. The use of ice packs is not recommended for shipping RNA.

3.3 RNA-seq Data Analyses

The analysis begins with raw sequence (paired end) reads, usually in fastq format or its gunzip compressed version fastq.gz (sample_name_1.fastq.gz / sample_name_2.fastq.gz) (see Note 11). In our standardized pipeline, the RNA-seq experiment is conducted using the TruSeq stranded mRNA library construction kit (Illumina) and sequenced on a production-scale sequencer from the NGS Illumina platforms (see Note 12).

3.3.1 Read Quality Control and Preprocessing

3.3.2

Quality Trimming

The goal of pre-processing is to remove low-quality bases at the end of the reads, short reads ( ddsHTSeq ddsHTSeq 10,]

Expression and Function of Genes in Non-model Species

387

> dds rld # Plot PCA plot >

plotPCA(rld,

intgroup="condition",

ntop=nrow(counts

(ddsHTSeq)))

The source can be found by typing DESeq2:::plotPCA.DESeqTransform or getMethod("plotPCA", "DESeqTransform"). Extracting results from the (DESeq) differential gene expression analysis (two-group comparison) > dds dds res norm.counts all write.table(all, file="DESeq2_all_rm.txt", sep="\t")

3.4.5 Figures on Differentially Expressed Genes (plotPCA)

This function is essentially two lines of code: building a data.frame and passing this to the plotMA method for data.frame from the geneplotter package. The code of this function can be seen with: getMethod ("plotMA", "DESeqDataSet"). > dds dds plotMA(dds) > res 15 Kb following manufacturer instructions. DNA quantification should be performed with the Qubit dsDNA BR (broad range) assay kit. DNA purity is evaluated with the ratio 260/280 nm and 260/230 nm measured using the Nanodrop device. Finally, the DNA fragment size is estimated by DNA electrophoresis in a 0.8% agarose gel with 0.001% (v/v) SYBR safe dye in 1× TBE buffer. DNA from λ phage digested with HindIII can be used as DNA fragment size marker. The upper band has a size of 23.13 Kb while the second upper band has a size of 9.41 Kb. A good-quality preparation of HMW DNA should show the following: • Nanodrop 260/280 ratio: 1.8–2.0 • Nanodrop 260/230 ratio: 2.0–2.2 • Ratio of concentration measured by Nanodrop/Qubit: 1.0–1.5 • DNA fragment sizes peak >10 Kb

3.1.4 DNA Library Preparation

The DNA library is prepared using the 1D genomic DNA sequencing kit (Oxford Nanopore Technologies), following the manufacturer instructions. The manufacturer recommends using 0.2 pmol of HMW DNA to prepare the library. The amount (in μg) needed

464

Tomas Hasing and Aureliano Bombarely

depends on the average fragment size and can be estimated with the following formula: DNA amount ðμgÞ = 0:00066 × ½DNA average fragment size in bp × ½molarity ðpmolÞ So, for example, if the average fragment size is 20 Kb, 2.64 μg of DNA are necessary for 0.2 pmol. 3.1.5 MinION DNA Sequencing and Base Calling

The DNA library is sequenced using the MinION platform following the manufacturer instructions. Some points to be considered: • The pore efficiency decreases with the age of the flow cell, so it is recommended to use the cells as soon as they are received. • A high pore occupancy (>70%) indicates a successful library preparation. • If the number of available pores decreases quickly in the first 2 h, this may indicate that the DNA sample has some contaminations that block the pores. • A flow cell can be reused one time if it is washed following the manufacturer instructions. However, the number of available pores and their efficiency are reduced. • A standard plant DNA sequencing yields 4 Gb/flow cell. • For a decent genome assembly at least 25× ON is needed and 50× is recommended. For example, to obtain a decent assembly for a 1 Gb genome, at least 25 Gb of reads should be obtained from the sequencing. The bioinformatic work will involve different steps and process. To maintain certain structure, it is recommended to create different directories and run each of the steps (or move each of the different outputs) to these directories. Additionally, all these examples are performed with a computer with 8 threads (e.g., -t 8 or -p 8 with some programs). See Note 3. > mkdir genome_assembly > cd genome_assembly > mkdir 00_raw > mkdir 01_processed > mkdir 02_assembly > mkdir 03_annotation

The base calling is performed using the Albacore software (supplied by Oxford Nanopore Technologies) and the FAST5 files obtained from the DNA sequencing and the MinKNOW program. Once the FAST5 files have been copied to the input

From Genomes to Genes in Flowering Species

465

directory (e.g., 00_raw), the typical command line to generate the . fastq in the 01_processed directory is something like the following: > full_1dsq_basecaller.py -f -k

-t

-s

01_processed -o fastq -q 100000 -i 00_raw >

cat

01_processed/*fastq

>

|

gzip

-1

>

ON_reads.fastq.gz

The quality of the reads can be assessed using different programs. Three of them are recommended: • Fastq-stats (from the Ea-utils package), to get a quick summary of the read dataset. A typical command is as follows: > cd 01_processed > mkdir seqstats >

fastq-stats

ON_reads.fastq.gz

>

seqstats/

ON_reads.stats.txt

• MinIONQC, to get a more complete report of the sequencing and base calling process. More information at https://github. com/roblanf/minion_qc. Copy the ON sequencing summary file to the 00_raw directory and then run a command as follows. It will create different outputs including some figures summarizing the sequencing process. > mkdir ONseq_process > Rscript MinIONQC.R -i ../00_raw/ -o ONseq_process

• FastQC, to get a HTML report of the Fastq files. FastQC has a graphical interface (GUI) or it can be run as a command line as follows: > mkdir fastqc_summaries >

fastqc

-t

8

-o

fastqc_summaries

ON_reads.

fastq.gz

3.1.6 Genome Sequence Assembly

There are different programs to perform the genome sequence assembly such as FALCON [43], Canu [44], Miniasm [25], and Flye [45]. For this protocol, the Minimap2, Miniasm, and Racon pipeline [25] is used. It is also recommended to assemble and filter the chloroplast/mitochondrial reads with tools such as Organelle_PBA [46].

466

Tomas Hasing and Aureliano Bombarely

1. Run Minimap2 on the raw ON reads with a command such as the following: > cd ../02_assembly/ >

minimap2

-x

ava-pb

-t

8

../01_processed/

ON_reads.fastq.gz ../01_processed/ON_reads.fastq.gz | gzip -1 > ON.paf.gz

2. Run Miniasm on the .paf and .fastq files with a command such as the following: > miniasm -f ../01_processed/ON_reads.fastq.gz ON.paf.gz > ON_assembly.gfa

3. Convert the .gfa file to .fasta with the command: > awk ’$1 ~/S/ {print ">"$2"\n"$3}’ ON_assembly.gfa > ON_assembly.fasta

4. Run FastqSeqStats on the .fasta file with a command such as the following: > FastaSeqStats -i ON_assembly.fasta

5. Map the ON reads back to the assembly with Minimap2 with a command such as the following: > minimap2 -x map-pb -t 8 ON_assembly.fasta ../ 01_processed/ON_reads.fastq.gz

|

gzip

-1

>

ON4assembly.paf.gz

6. Call the consensus sequence with Racon with a command such as the following: > racon -t 8 ../01_processed/ON_reads.fastq.gz ON_assembly.fasta ON4assembly.paf.gz > ON_consensus1.fasta

From Genomes to Genes in Flowering Species

467

7. Run FastqSeqStats on the .fasta file with a command such as the following: > FastaSeqStats -I ON_consensus1.fasta

8. Map again the ON reads on the consensus produced by Racon with: > minimap2 -x map-pb -t 8 ON_consensus1.fasta ../01_processed/ON_reads.fastq.gz | gzip -1 > ON4consensus1.paf.gz

9. Call a new consensus over the previous one with a command such as the following: > racon -t 8 ../01_processed/ON_reads.fastq.gz ON_consensus1.fasta ON4consensus1.paf.gz > ON_consensus2.fasta

10. Run FastqSeqStats on the .fasta file with a command such as the following: > FastaSeqStats -i ON_consensus2.fasta

11. Map again the ON reads on the consensus produced by Racon with: > minimap2 -x map-pb -t 8 ON_consensus2.fasta ../01_processed/ON_reads.fastq.gz | gzip -1 > ON4consensus2.paf.gz

12. Call a new consensus over the previous one with a command such as the following: > racon -t 8 ../01_processed/ON_reads.fastq.gz ON_consensus2.fasta ON4consensus2.paf.gz > ON_consensus3.fasta

Once the genome assembly has been produced, its quality is evaluated by different methods:

468

Tomas Hasing and Aureliano Bombarely

• Assembly stats. The program FastaSeqStats (or any alternative such as Quast: http://quast.bioinf.spbau.ru/manual.html). There are different parameters such as total assembly size, total number of contigs/scaffolds, longest sequence, and N50/L50 that help to decide which assembly is better in terms of assembly stats (it does not mean biological accuracy). An assembly with a size closest to the estimated genome, with a lower number of contigs, a lower N50, and a higher L50 is considered the best. • Mapping stats. An assembly that has the highest percentage of reads mapped back to it is considered as the one that captures the highest proportion of the genome. There are several tools that can be used to map reads back to the genome. In the previous example, Minimap2 is used to map reads back to the assembly. To calculate how many reads have been mapped to a sequence dataset using a .paf file, use the following command: > gunzip -c ON4consensus2.paf.gz | cut -f1 | sort -u | wc -l

• Consensus accuracy estimation. The consensus accuracy can be estimated by mapping short reads back to the assembly and calling variants. Homozygous variants are considered possible errors in the consensus call. Some homozygous variants may come from errors in the read mapping and variant calling, for which it is recommended to use different read mapping tools (e.g., Bowtie2 and BWA) and different variant calling tools (e.g., GATK and FreeBayes). • Gene space captured. Other possible way to evaluate a genome assembly is to estimate the proportion of the gene space that has been captured. There are two popular options: (1) compare the genome assembly with a well-known group of proteins present in a taxonomic group (e.g., viridiplantae) using tools such as CEGMA and BUSCO and (2) estimate the proportion of mapped reads from a transcriptome of the same species. An example of a BUSCO command line is as follows: > python run_BUSCO.py --in ON_consensus2.fasta --cpu

8

--mode

genome

--lineage_path

busco/

datasets/embryophyta_odb10/ --out BUSCO_emb10

3.1.7 Genome Annotation

There are two types of annotations that it can be performed in a genomic sequence: (1) structural, that means the characterization of which parts of the genome will encode genetic elements such as genes, exons, UTRs, and transposons and (2) functional, where such elements are associated to possible functions. There are three types of datasets that are needed to perform a basic structural

From Genomes to Genes in Flowering Species

469

annotation: (a) repetitive sequences, (b) transcripts of the same species, and (c) proteins of closely related species. Supposing that it was possible to access to some transcriptomic data (e.g., files with the names “RNASeq_1st_pair.fq” and “RNASeq_2nd_pair.fq”), a simple genome structural annotation will involve the following steps: 1. Repetitive element pre-annotation with RepeatModeler will produce a “consensus.fasta” file that it will contain the repetitive sequences. To obtain that file, run the following commands: > cd ../03_annotation > cp ../02_assembly/ON_consensus2.fasta ./ > BuildDatabase -name my_species -engine ncbi ON_consensus2.fasta > RepeatModeler -engine ncbi -pa 8 -database my_species

2. Transcriptomic pre-annotation with Hisat2 and StringTie with command lines such as follows: > hisat2-build ON_consensus2.fasta ON_consensus>

hisat2

-p

Seq_1st_pair.fq

8

-x

-2

ON_consensus

-1

RNASeq_2nd_pair.fq

RNA|

samtools view -F4 -Sb -o RNASeq.bam –> samtools sort -@ 8 -o RNASeq.bam RNASeq.bam > stringtie RNASeq.bam -p 8 -o RNASeq.gtf > gffread RNASeq.gtf -o RNASeq.gff3

3. The full genome annotation with Maker-P will also require a protein dataset, which can be downloaded from UniProt (https://www.uniprot.org/), GenBank (https://www.ncbi. nlm.nih.gov/genbank/), or any specific plant genomic database such as Phytozome (https://phytozome.jgi.doe.gov/pz/ portal.html). It will also require the training of SNAP (http:// gmod.org/wiki/MAKER_Tutorial_2011) and Augustus Once (https://github.com/Gaius-Augustus/BRAKER). these requirements are met, the next step is to edit the “maker_opts.ctl” file produced by Maker and then run the annotation pipeline from the directory where maker_opts.ctl file is with a command such as follows:

470

Tomas Hasing and Aureliano Bombarely

> mpiexec -n 50 /maker> fasta_merge -d ON_consensus2.maker.output/ master_datastore_index.log gff3_merge

-n

-d

ON_consensus2_-o

ON_consensus2>

ON_consensus2.maker.output/

ON_consensus2_master_datastore_index.log

-o

ON_consensus2.maker.gff

4. The stats of the annotation can be obtained running the following command on the ON_consensus2.maker.gff file. > grep -v “#” ON_consensus2.maker.gff | cut -f2,3 | sort | uniq -c

5. Additionally, BUSCO can be run on the protein datasets created by Maker to evaluate the completeness of the annotated gene space. For example: > python run_BUSCO.py --in ON_consensus2.all. maker.proteins.fasta

--cpu

--mode

protein --lineage_path busco/datasets/embryophyta_odb10/ --out BUSCO_emb10

6. Functional annotation can be performed mainly running BLASTP with several protein datasets such as NR (GenBank: ftp://ftp.ncbi.nlm.nih.gov/blast/db/), SwissProt (Uniprot: ftp://ftp.uniprot.org/pub/databases/uniprot/current_ release/knowledgebase/complete/uniprot_sprot.fasta.gz), and/or Arabidopsis proteome (TAIR: https://www.ara bidopsis.org/download_files/Sequences/TAIR10_blastsets/ TAIR10_pep_20110103_representative_gene_model_ updated). The results can be integrated in the GFF file using AHRD (https://github.com/groupschoof/AHRD) and AddAHRD2Gff from the GenoToolBox package. An example could be as follows: > wget ftp://ftp.uniprot.org/pub/databases/ uniprot/current_release/knowledgebase/ complete/uniprot_sprot.fasta.gz>

gunzip

uni-

prot_sprot.fasta.gz > makeblastdb -in uniprot_sprot.fasta -dbtype prot -parse_seqids > blastp -query ON_consensus2.all.maker.proteins.fasta -db uniprot_sprot.fasta -out ON_-

From Genomes to Genes in Flowering Species

471

consensus2.prot.vs.swp.txt -outfmt 6 -num_threads -num_descriptions 10## Create and modify your_use_case_file.yml accordingly> java -Xmx2g -jar ./dist/ahrd.jar your_use_case_file.yml> AddAHRD2Gff -g ON_consensus2.maker.gff

-a

-o

ON_consensus2.maker.annotated.gff

3.2

3.2.1

QTL Analysis

DNA Extraction

The QTL analysis for a specific trait involves three important parts: (1) development of the population with the segregation of the phenotype of interest, (2) phenotyping and genotyping of such population, and (3) development of a genetic map and QTL analysis. It is far from the goal of this protocol to explain in detail the development and the phenotyping of the population because they are specific of the phenotype and the model/organism selected for the study. Nevertheless, in the following sections it is explained the genotyping of the population using genotyping by sequencing and a basic QTL data analysis retrieving a list of possible candidate genes. DNA extraction is performed using a modified version of the CTAB method [47] that is optimized for floral crops such as gesneriads [11] and begonias, where commercial kits such as the DNeasy Plant Mini Kit (Qiagen) delivers a low yield. DNA quality is evaluated as described above (Subheading 3.1.3) for the Qubit and the Nanodrop analysis. 1. Chop 100 mg of fresh tissue (a young leaf if it is possible) with a razor blade. 2. Transfer to a 1.5 mL Eppendorf tube, frozen with liquid nitrogen, and then grind using a polypropylene pellet pestle. 3. Add 400 μL of 2× CTAB buffer, 4 μL of RNase A, and 12 μL of 2-mercaptoethanol. 4. Incubate at 65 °C for 1 h in a water bath or a heating block to deactivate the endogenous DNases. 5. Incubate in ice 5 minutes. 6. Add 400 μL of chloroform and vortex thoroughly. 7. Spin at 15,000 g in a microcentrifuge for 2 min. 8. Transfer the upper, aqueous phase to a fresh reaction tube. 9. Add 400 μL of 2-propanol and mix well. 10. Spin in centrifuge for 5 min at 20,000 g to pellet the DNA. 11. Remove supernatant and wash pellet with 500 μL of 70% ethanol.

472

Tomas Hasing and Aureliano Bombarely

12. Remove ethanol by inverting the tube (be careful not to lose the pellet) and spin briefly (~5 s). 13. Remove leftover ethanol with pipette and air-dry pellet for 10 min. 14. Add 100 μL of TE buffer and allow the pellet to dissolve. 15. Clean DNA with “Monarch PCR and DNA cleanup kit” following the manufacturer instructions (use 20–30 μL of elution buffer). 16. Quantify the DNA samples as in Subheading 3.1.3. 17. Dilute each of the samples to a concentration of 10 ng/μL. 3.2.2 Genotyping-BySequencing Library Preparation

The genotyping-by- sequencing library preparation methodology is based on the protocol published by Elshire [19]. Although this is a cost-efficient methodology ( mkdir genotyping > cd genotyping > mkdir reference > mkdir 00_raw > mkdir 01_demultiplexed > mkdir 02_processed

474

Tomas Hasing and Aureliano Bombarely

> mkdir 03_mapped > mkdir 04_variants

2. Demultiplex the samples (a “barcodes.txt” file with three columns, “SampleID,” “barcode_sequence,” and “restriction_enzyme,” is required). GBSX will create one file (two if the reads are paired ends) per sample (e.g., if the barcode.txt has a line with a sample named “F2_001,” a barcode “ATCG” and the restriction enzyme “ApeKI,” it will investigate the “GBSseq. fastq.gz” file and it will create a file named “F2_001.fastq.gz” with all the reads with the barcode “ATCG” plus the restriction site for ApeKI). > java -jar GBSX_v1.2.jar --Demultiplexer -f1 00_raw/GBSseq.fastq.gz

-i

barcodes.txt

-o

01_demultiplexed/ -gzip

3. Get the read stats for each of the files using fastq-stats (an alternative program could be FastQC). > cd 01_demultiplexed > mkdir stats## Run with each of the samples > fastq-stats F2_001.fastq.gz > stats/F2_001. stats.txt > fastq-stats F2_002.fastq.gz > stats/F2_002. stats.txt ### . . .

A bash script could be written to run the fastq-stats on each of the sample files. To do it, one option could be the use of a command that list each of the fastq files (e.g., ls), concatenate the output to a command able to add some text (e.g., awk ‘{print “blabla”}’), and then redirect the output into a file that later on is executed as script. An example could be something like the following: > cd 01_demultiplexed> ls | grep fastq | sed -r s/.fastq.gz// | awk BEGIN{ print “#!/bin/bash”} { print “fastq-stats ”$1”.fastq.gz > stats/”$1”. stats.txt” } > run_stats.sh> chmod 755 ./run_stats.sh > ./run_stats.sh

From Genomes to Genes in Flowering Species

475

4. Trim the adapter sequences and low-quality bases and remove the reads shorter than 50 bp with Fastq-mcf. A fasta file with the sequence of the Illumina adapter needs to be created to run the program (e.g., “adapters.fasta”). More information about the adapter sequences can be found at https://support. illumina.com/downloads/illumina-adapter-sequences-docu ment-1000000002694.html > fastq-mcf -q 30 -l 50 -o ../02_processed/ F2_001.q30l50.fastq.gz adapters.fasta F2_001. fastq.gz >

fastq-mcf

-q

30

-l

50

-o

../02_processed/

F2_002.q30l50.fastq.gz adapters.fasta F2_002. fastq.gz ## . . .

A bash script could be used instead to run the same command with each of the files. To create a bash script, an example could be as follows: ## To write a bash script will all the files replace previous commands by:> ls | grep fastq | sed -r s/.fastq.gz// | awk BEGIN{ print “#!/ bin/bash”}{

print

“fastq-mcf

-q

02_processed/”$1”.q30l50.fastq.gz

30

-l

50

-o

Illumina_a-

dapters.fasta ”$1”.fastq.gz” } > run_process.sh > chmod 755 ./run_process.sh > ./run_process.sh

5. Get the read stats for each of the files using fastq-stats (an alternative program could be FastQC) as it was described in point 2. 6. Generate the reference indexes for the read mapping using Bowtie2-build. > cd ../reference > bowtie2-build ON_consensus2.fasta ON_consensus2 > samtools faidx ON_consensus2.fasta

7. Map the reads of each sample to the reference with Bowtie2 and select only the reads that have been mapped with Samtools.

476

Tomas Hasing and Aureliano Bombarely

> bowtie2 -p -x ON_consensus2 -U 02_processed F2_001.q30l50.fastq.gz | samtools view -F 4 -Sb -o 03_mapped/F2_001.bam >

bowtie2

-p

-x

ON_consensus2

-U

02_processed F2_001.q30l50.fastq.gz | samtools view -F 4 -Sb -o 03_mapped/F2_001.bam ## . . . ## Alternatively, to write a bash script:> ls 02_processed | grep fastq | sed -r s/.fastq. gz// | awk BEGIN{ print “#!/bin/bash”}{ print “bowtie2 -p 8 -x ON_consensus2 -U 02_processed/ ”$1”.q30l50.fastq.gz | samtools view -F 4 -Sb -o 03_mapped/”$1”.bam -” } > run_map.sh > chmod 755 ./run_map.sh > ./run_map.sh

8. Sort the read mapping files with Samtools and index the mapping file. > cd .. >

samtools

sort

-@8

-o

03_mapped/F2_001.bam

-o

03_mapped/F2_002.bam

03_mapped/F2_001.bam >

samtools

sort

-@8

03_mapped/F2_002.bam ## . . .

## Alternatively, to write a bash script: > ls 03_mapped | grep bam | awk BEGIN{ print “#!/ bin/bash”}{

print

“samtools

sort

-@8

-o

03_mapped/”$1” 03_mapped/”$1 } > run_sort.sh > chmod 755 ./run_sort.sh > ./run_sort.sh

9. Retrieve some mapping stats for the mapped reads using Picard-tools. > mkdir 02_mapped/stats > picard-tools CollectAlignmentSummaryMetrics INPUT=03_mapped/F2_001.bam stats/F2_001.picardstats.txt

OUTPUT=03_mapped/ REFERENCE_SE-

QUENCE=reference/ON_consensus2.fasta > picard-tools CollectAlignmentSummaryMetrics

From Genomes to Genes in Flowering Species

INPUT=03_mapped/F2_002.bam

477

OUTPUT=03_mapped/

stats/F2_002.picardstats.txt

REFERENCE_SE-

QUENCE=reference/ON_consensus2.fasta

## Alternatively, to write a bash script and run it:> ls 03_mapped | grep bam | sed s/.bam// | awk BEGIN{ print “#!/bin/bash”}{ print “picard-tools CollectAlignmentSummaryMetrics INPUT=03_mapped/”$1”.bam OUTPUT=03_mapped/stats/ ”$1”.picardstats.txt REFERENCE_SEQUENCE=reference/ON_consensus2.fasta” } > run_bamstats.sh > chmod 755 ./run_bamstats.sh > ./run_bamstats.sh

10. Calculate the genome coverage of each sample using Bedtools. > bedtools genomecov -ibam 03_mapped/F2_001. bam

-g

reference/ON_consensus2.fasta.fai

>

03_mapped/stats/F2_001.genomcov.txt >

bedtools

bam

-g

genomecov

-ibam

03_mapped/F2_002.

reference/ON_consensus2.fasta.fai

>

03_mapped/stats/F2_002.genomcov.txt ##... ## Alternatively, to write a bash script and run it: > ls 03_mapped | grep bam | sed s/.bam// | awk BEGIN{ print “#!/bin/bash”}{ print “bedtools genomecov -ibam 03_mapped/”$1”.bam -g reference/ON_consensus2.fasta.fai

>

03_mapped/

stats/”$1”.genomcov.txt”} > run_covbedstats.sh > chmod 755 ./run_covbedstats.sh > ./run_covbedstats.sh

11. Merge all the mapping files using Bamaddrg. > bamaddrg -s F2_001 -b 03_mapped/F2_001.bam -s F2_002

-b

03_mapped/F2_002.bam

-s

P1

-b

03_mapped/P1.bam -s P2 -b 03_mapped/P2.bam > 03_mapped/MyF2Pop.bam##

Alternatively,

to

write a script avoiding the writing of all the bam files:

478

Tomas Hasing and Aureliano Bombarely

ls 03_mapped/ | grep bam | sed ’s/.bam//’ | perl -ne ’BEGIN{ @a=(); } chomp($_); push(@a, "-s ". $_." -b ".$_.".bam "); END{ $l=join(" ", @a); print "#!/bin/bash\n\nbamaddrg $l > 03_mapped/ MyF2Pop.bam \n"} ’ > run_bamaddrg.sh > chmod 755 run_bamaddrg.sh > ./run_bamaddrg.sh

12. Calculate the genome coverage of all the samples with Bedtools. > bedtools genomecov -ibam 03_mapped/MyF2Pop. bam

-g

reference/ON_consensus2.fasta.fai

>

03_mapped/stats/MyF2Pop.genomecov.txt

13. Perform the variant calling with FreeBayes. > freebayes --bam 03_mapped/MyF2Pop.bam --vcf 04_variants/MyF2Pop.vcf erence/

–fasta-reference

ON_consensus2.fasta

ref-

--min-mapping-

quality 20 --min-coverage 10 --no-mnps --nocomplex## With the newest version of freebayes is recommended also the use of ## --limit-coverage 100000 (1000 * number of samples).

14. Calculate the variant calling stats Vcf-stats. > vcf-stats 04_variants/MyF2Pop.vcf -p 04_variants/MyF2Pop_stats

15. Filter out InDels, non-biallelic and SNP missing in more of the 80% of the samples. > vcftools --vcf 04_variants/MyF2Pop.vcf -remove-indels --max-missing 0.8 --min-alleles 2

--max-alleles

--out

2

--recode

04_variants/

lic.80M.vcf

--recode-INFO-all

MyF2Pop.OnlySNP.Bialle-

From Genomes to Genes in Flowering Species

479

16. Variants (or SNP like in this case) that are in the same linkage group have redundant information to develop a genetic map, so they can be removed from the variant dataset before the genetic map construction. In this case, variants with a distance smaller than 1 Kb can be removed. > vcftools --vcf 04_variants/MyF2Pop.OnlySNP. Biallelic.80M.vcf recode-INFO-all

--thin --out

1000

--recode

--

04_variants/MyF2Pop.

OnlySNP.Biallelic.80M.1Kb.vcf

17. The final step is to convert the VCF file to a TXT file that is the input for the R/QTL program. For other formats such the one used by Structure there are several options, such as PGDSpider, but in this case the GenoToolBox Vcf2Mapmaker script is recommended. Basically, this script compares each of the SNPs of the population with the parents defined as -a and -b, converting the SNP information to A (genotype like parent A), B (genotype similar to parent B), and H (heterozygous). > Vcf2Mapmaker -i 04_variants/MyF2Pop.OnlySNP. Biallelic.80M.vcf

-o

04_variants/MyF2Pop.

OnlySNP.Biallelic.80M.mk.csv -f csv -a P1 -b P2 -s ’(a,h,b)’ -p 0.001-H

3.2.4 Development of the Genetic Map

The rise of the NGS technologies has pushed the number of markers used to generate genetic maps from dozens to thousands. Traditional popular programs such as Mapmaker [49] are being replaced by new programs with new algorithms capable to handle these larger amounts of data. Two R packages are described in this protocol: R/QTL [40] and OneMap [41]. R/QTL is used to produce a draft of a genetic map that is later refined with OneMap. Once the map is produced, the phenotypic data are added, and the QTL analyzed using the R/QTL package again. To describe the development of a genetic map in detail is out of the scope of this protocol, but here we describe the main steps. For more information about this topic, we suggest to follow the R/QTL tutorial (https://rqtl.org/tutorials/geneticmaps.pdf) and the book “A Guide to QTL Mapping with R/qtl” [50]. 1. Open R Studio and, using the “File” tab, navigate to the directory where it is the “MyF2Pop.OnlySNP.Biallelic.80 M. mk.csv” file. Click in the same tab, “More” and then “Set as a working dir.” See Note 4.

480

Tomas Hasing and Aureliano Bombarely

2. Load the QTL library and the .csv file with the SNPs using the function read.cross(). > library (qtl) > draft_gmap = read.cross("csv", "", "MyF2Pop. OnlySNP.Biallelic.80M.mk.csv", estimate.map = FALSE) > summary(draft_gmap)

3. Check that the histogram with the genotype distribution has a bell shape. > comp_genotypes = comparegeno(draft_gmap) >

hist(comp_genotypes[lower.tri

(comp_genotypes)], breaks=seq(0, 1, len=101), xlab="N. of matching genotypes", main="Matching Genotypes Distribution")

4. Check the deviation of the expected segregation and remove those markers with a p value below a cutoff value (0.0001). > genotable = geno.table(draft_gmap) >

segreg_dev

=

genotable[genotable$P.value




todrop

=

rownames(genotable[genotable$P.

value < 1e-5, ]) >

draft_gmapf

=

drop.markers

(draft_gmap,

todrop) > summary(draft_gmapf)

5. Estimate the recombination frequency (RF) and the LOD score for each marker. Plot the RF versus LOD for each marker to see their distribution. The LOD score should be higher for RF = 0 and RF = 1, having a minimum for RF = 0.5. > draft_gmapf = est.rf(draft_gmapf) > draft_gmapf_rf = pull.rf(draft_gmapf) > draft_gmapf_lod = pull.rf(draft_gmapf, what="lod") > plot(as.numeric(draft_gmapf_rf), as.numeric (draft_gmapf_lod),

xlab="Recombination

quency", ylab="LOD Score")

Fre-

From Genomes to Genes in Flowering Species

481

6. Create the linkage groups using a minimum LOD score of 6 and a maximum RF of 0.35. Higher values for the minimum LOD can be used to test if more strict values will produce the same number of groups. > draft_gmap_lg = formLinkageGroups(draft_gmapf, max.rf=0.35, min.lod = 6) > table(draft_gmap_lg[,2])

7. If several linkage groups close to the expected number is obtained, rerun the command adding “reogMarkers=TRUE” and then plot the heatmap for the RFs. > draft_gmap_lg = formLinkageGroups(draft_gmapf, max.rf=0.35, min.lod = 6, reorgMarkers = TRUE) >

plotRF(draft_gmap_lg,

alternate.chrid

=

TRUE))

8. The draft genetic map produced by the previous steps may have some errors caused by the switch of genetic markers. There are two options to fix them. The first is reordering a pair of them one by one and checking if the LOD score of the new order is lower than that of the previous one. If it is, it is a good indication that the new order may be better. Because this process can be time consuming for maps with thousands of markers, the second option is to use OneMap to automatically sort the markers. > write.cross(draft_gmap_lg, format=”mm”, filestem=”my_rqtl_draftmap.mm”) > library (onemap) > draft_omap = read.mapmaker("./", " my_rqtl_draftmap.mm")

9. Recalculate the distances and rebuild the map. As with the R/QTL, different LOD score values can be explored to adjust the number of linkage groups to what is expected. As example, LOD = 7 and RF = 0.4 is used. > twopts_omap = rf.2pts(draft_omap) > markmap_omap = make.seq(twopts_omap, "all") > onemap_lg = group(markmap_omap, LOD=7, max.

482

Tomas Hasing and Aureliano Bombarely

rf=0.4) > set.map.fun(type="kosambi")

10. A loop is used for sorting each of the linkage groups. > ## Set up N to the number of LGs > idx = seq(1,n); > lg = list(); > lg.ord = list(); > lg.saf = list(); > lg.frc = list();> for (i in rev(idx)) { print (paste("Processing LG=", idx[i])); lg[[i]] = make.seq(onemap_lg, i); lg.ord[[i]]

=

order.seq(lg[[i]],

n.init=6,

touchdown=TRUE); lg.saf[[i]] = make.seq(lg.ord[[i]], "safe"); lg.frc[[i]] = make.seq(lg.ord[[i]], "force"); }

11. Finally, export the map > write.map(lg.frc, "my_onemap_draftmap.txt")

3.2.5

QTL Analysis

One of the requirements to run a quantitative trait analysis (QTL) is to have the phenotypic information in the same file format that was used to develop the map (e.g., “csv” or “mapmaker”). In step 8 of Subheading 3.2.4, a Mapmaker file format (.mm) containing the genotypic information was produced. This file can be modified to add the phenotypic information. The Mapmaker format is composed of two header lines (L1 and L2), where L1 contains the population information (e.g., “data type f2 intercross”) and L2 a summary of the data contained in the file. This summary is composed by three numbers (x y z) in which the first one, x, summarizes the number of individuals, the second one, y, summarizes the number of markers and the third one, z, the number of lines containing the phenotypes (e.g., “801811 2” means 80 individuals, 1811 markers and 2 lines containing phenotypes). After these two lines, the file will have the lines containing the genotypic information, one line per marker with the genotype for each of the individuals (in the previous example, 1811 lines in which each line has 81 columns, one with the marker name and then one per genotype for each individual coded as A, B, or H in the case of an F2 cross). Finally, the file will contain the phenotypic information (as many lines as z). Supposing that the phenotypic information has been coded in a text file (e.g., “phenotypes.txt”) in which the first

From Genomes to Genes in Flowering Species

483

column is the individual name and the other columns have the phenotypic information (e.g., line 1 “#IND ZygomorphicFlowers,” line 2 “F1_001 0 8,” and line 3 “F2_001 1 5”), and the individuals are in the same order as in the VCF file used in step 17 of Subheading 3.2.3, the steps to add this information to a new file could be the following: 1. Load the genetic map and the phenotype data in R (or R Studio). > geno = read.table("my_rqtl_draftmap.mm") > pheno = read.table("phenotypes.txt", header = F) >

geno_pheno

=

na.omit(rbind(geno_test,as.

data.frame(t(pheno_test))

[2:nrow(pheno_t-

est),]))> write.table(geno_pheno,"my_rqtl_draftmap_pheno.mm",

quote

=

FALSE,

row.

names=FALSE, col.names=FALSE)

2. The new file containing both genotypic and phenotypic information can be read into R/qtl: > final_data.qtl = read.cross("mm", file=”my_rqtl_draftmap_pheno.mm”,

mapfile="my_one-

map_draftmap.map")

3. Once the genetic information has been combined with the phenotypic data, we can run the QTL analyses for each of the traits in the file. First, we need to calculate QTL genotype probabilities using Hidden Markov Models: > final_data.qtl = calc.genoprob(final_data. qtl, step=1, error.prob = 0.001)

4. Now we can start with the simplest case by conducting a single QTL scan with a normal model. For example, to analyze and visualize the results on the zygomorphic flowers (column 3): > out.mr_zygomorphic summary(out.mr_zygomorphic, threshold = 3) > plot(out.mr_zygomorphic, ylab = "LOD score") >

plot(out.mr_zygomorphic,

"LOD score")

chr

=

10,

ylab

=

484

Tomas Hasing and Aureliano Bombarely

This returns a list of markers with LOD values higher than a threshold (in this case 3), and a figure displaying the LOD values across all chromosomes (first plot statement) and within chromosome 10 (second plot statement). 5. To perform a two-dimensional genome scan with a two QTL model, we can use the function scantwo. The calculations are intensive, and this task may take some time to complete as this function also tests for interactions between loci. > out.mr_zygomorphic summary(out.mr_zygomorphic, threshold = 3) > plot(out.mr_zygomorphic, ylab = "LOD score") >

plot(out.mr_zygomorphic,

chr

=

10,

ylab

=

"LOD score")

The significance threshold for a QTL analysis depends on several factors such as the length of the chromosome where the signal was detected, if the chromosome is an autosomal chromosome, the model selected for the scan, the number of permutations, etc. A good starting point is to consider a threshold of 3 as it is described in the previous box, although strong signals often have values greater than 8. Once the markers with a significant signal have been obtained, there are two ways to estimate the intervals: (1) LOD supported and (2) Bayes credible intervals. In the first approach, the markers that are usually retrieved are close to the marker that presented the maximum and with a LOD score within 1.5 units from this maximum. In the second approach, 10LOD are calculated for each marker and then a formula is applied to estimate the interval with a 95% of confidence. Both approaches can be applied with R/QTL with the following commands: # 1- LOD scores (e.g. chromosome 10) > lodint(out.mr_zygomorphic, 10, 1.5)# 2- Bayes credible intervals (e.g. chromosome 10) > bayesint(out.mr_zygomorphic, 10, 0.95)

Finally, if the markers were named as “ReferenceSequenceID_Position” (e.g., “Sispe038Scf0186_10451”) and the markers obtained in the interval are physically contiguous on the same reference sequence (e.g., “Sispe038Scf0186_10451” and “Sispe038Scf0186_450498”), the list of candidate genes can be obtained using the GFF annotation file obtained in Subheading 3.1.7 using Bedtools.

From Genomes to Genes in Flowering Species

485

## First, create a Bed file with the location >

echo

-e

“SeqID\tstart_position\tend_posi-

tion” > QTL_PhysicalLocation.bed## Second, get the overlap with the GFF file and obtain the list of genes > bedtools intersect -a QTL_PhysicalLocation. bed -b ON_consensus2.maker.annotated.gff -wb## To redirect the output into a file selecting only the rows ## where the type is “gene” > bedtools intersect -a QTL_PhysicalLocation. bed -b ON_consensus2.maker.annotated.gff -wb | awk { if ($3 ~ /gene/) print $0 } > QTL_Genes. gff## To get the list of GeneIDs > bedtools intersect -a QTL_PhysicalLocation. bed -b ON_consensus2.maker.annotated.gff -wb | awk { if ($3 ~ /gene/) print $9 } | cut -d ; -f1 | sed s/ID=// > QTL_GeneID_list.txt

3.3 Transcriptomic Analysis

3.3.1 RNA Extraction and Quality Evaluation

One of the most common approaches used to reduce the number of candidate genes is to complement the QTL analysis with a transcriptomic analysis of the specific trait that it is being analyzed. For example, if the focus is the analysis of the zygomorphic flowers, a transcriptomic analysis of the floral meristems could help to reduce the number of candidate genes. It is recommended to perform the RNA extraction using a commercial kit such as the RNeasy Plant Mini Kit (Qiagen). RNA quality evaluation is performed with the Nanodrop, Qubit, and a fragment analyzer system such as a Bioanalyzer or a Tapestation. Specifically, the parameters that can be assessed to evaluate the quality of the RNA are as follows: • Nanodrop 260/280 ratio: 1.8–2.0 • Nanodrop 260/230 ratio: 2.0–2.2 • Ratio of concentration measured by Nanodrop/Qubit: 1.0–1.5 • For a fragment analyzer (Bioanalyzer), a RIN > 6.0 (RIN > 7.0 is recommended).

3.3.2 RNA-Seq Library Preparation

There are three different options for the RNA-Seq library preparation: Preparation and sequencing at a genomic facility, use of an RNA-Seq library kit, or library preparation following an in-house protocol. The reduction of the cost of the library preparation at the sequencing facilities makes it the recommended option for few samples ( mkdir transcriptomics > cd transcriptomics > mkdir reference > mkdir 00_raw > mkdir 01_processed > mkdir 02_mapped > mkdir 03_quantified

2. Get the read stats for each of the files using fastq-stats (an alternative program could be FastQC). > cd 00_raw > mkdir stats ## Run with each of the samples >

fastq-stats

Zyg_S1_R1.fastq.gz

>

stats/

>

stats/

>

stats/

>

stats/

Zyg_S1_R1.stats.txt >

fastq-stats

Zyg_S1_R2.fastq.gz

Zyg_S1_R2.stats.txt >

fastq-stats

Zyg_S2_R1.fastq.gz

Zyg_S2_R1.stats.txt >

fastq-stats

Zyg_S2_R2.fastq.gz

Zyg_S2_R2.stats.txt ### . . .

Alternatively, a bash script could be used to get the stats for all the samples: > cd 00_raw> ls | grep fastq | sed -r s/.fastq. gz// | awk BEGIN{ print “#!/bin/bash”}{ print “fastq-stats ”$1”.fastq.gz > stats/”$1”.stats. txt” } > run_stats.sh > chmod 755 ./run_stats.sh > ./run_stats.sh

From Genomes to Genes in Flowering Species

487

3. Remove the adapter sequences and low-quality bases. If the read falls below certain length (e.g., 50 bp), the whole pair should be removed. Both files of the same pair (e.g., Zyg_S1_R1.fastq.gz and Zyg_S1_R2.fastq.gz) should be processed at the same time. As in the case of the GBS reads, a fasta file with the sequence of the Illumina adapter needs to be created to run the program (e.g., “adapters.fasta”). More information about the adapter sequences can be found at https:// suppor t.illumina.com/downloads/illumina-adaptersequences-document-1000000002694.html > fastq-mcf -q 30 -l 50 -o ../01_processed/ Zyg_S1_q30l50_R1.fastq.gz -o ../01_processed/ Zyg_S1_q30l50_R2.fastq.gz

adapters.fasta

Zyg_S1_R1.fastq.gz Zyg_S1_R2.fastq.gz >

fastq-mcf

-q

30

-l

50

-o

../01_processed/

Zyg_S2_q30l50_R1.fastq.gz -o ../01_processed/ Zyg_S2_q30l50_R2.fastq.gz

adapters.fasta

Zyg_S2_R1.fastq.gz Zyg_S2_R2.fastq.gz ## . . .

Alternatively, a bash script could be used instead to run the same command with each of the files. An example could be as follows: ## To write a bash script will all the files replace previous commands by: > ls | grep fastq | sed -r s/_R[1-2].fastq.gz// | sort -u | awk BEGIN{ print “#!/bin/bash”}{ print “fastq-mcf -q 30 -l 50 -o 01_processed/ ”$1”.q30l50_R1.fastq.gz -o 01_processed/”$1”. q30l50_R2.fastq.gz

Illumina_adapters.fasta

”$1”_R1.fastq.gz ”$1”_R1.fastq.gz” } > run_process.sh > chmod 755 ./run_process.sh > ./run_process.sh

4. Get the read stats for each of the files using fastq-stats (an alternative program could be FastQC) as it was described in the point 2. 5. Generate the reference indexes for the read mapping using Hisat2-build.

488

Tomas Hasing and Aureliano Bombarely

> cd ../reference >

hisat2-build

ON_consensus2.fasta

ON_consensus2

6. Map the reads of each sample to the reference with Hisat2 and convert to bam with Samtools. > cd ../01_processed > hisat2 --dta -p 4 -x ../reference/ON_consensus2

-1

Zyg_S1_q30l50_R1.fastq.gz

-2

Zyg_S1_q30l50_R1.fastq.gz | samtools view -Sb -o ../02_mapped/Zyg_S1.bam > hisat2 --dta -p 4 -x ../reference/ON_consensus2

-1

Zyg_S2_q30l50_R1.fastq.gz

-2

Zyg_S2_q30l50_R1.fastq.gz | samtools view -Sb -o Zyg_S2.bam ## . . . ## Alternatively, to write a bash script: > ls 01_processed | grep fastq | sed -r s/_R [1-2].fastq.gz// | awk BEGIN{ print “#!/bin/ bash”}{ print “hisat2 --dta -p 4 -x ../reference/ON_consensus2 -1 ”$1”_R1.q30l50.fastq.gz -2 ”$1”_R2.q30l50.fastq.gz | samtools view -F 4 -Sb -o 03_mapped/”$1”.bam -” } > 01_processed/ run_map.sh > cd 01_processed > chmod 755 ./run_map.sh > ./run_map.sh

7. Sort the reads of each sample to the reference with Hisat2. > cd ../mapped > samtools sort -@4 -o Zyg_S1.bam Zyg_S1.bam > samtools sort -@4 -o Zyg_S2.bam Zyg_S2.bam ## . . .

## Alternatively, to write a bash script: > ls 02_mapped | grep bam | awk BEGIN{ print “#!/ bin/bash”}{ print “samtools sort -@4 -o ”$1” ”$1”\n” } > 02_mapped/run_sort.sh > cd 02_mapped

From Genomes to Genes in Flowering Species

489

> chmod 755 ./run_sort.sh > ./run_sort.sh

8. Build the transcript models with Stringtie (you can use the gene model file obtained in the annotation if you want to run a guided transcript model method). > stringtie --dta -p 4 -G ../reference/ON_consensus2.maker.gff

-o

Zyg_S1.gtf

-l

Zyg_S1

Zyg_S1.bam > stringtie --dta -p 4 -G ../reference/ON_consensus2.maker.gff

-o

Zyg_S2.gtf

-l

Zyg_S2

Zyg_S2.bam ## . . .

## Alternatively, to write a bash script: > ls 02_mapped | grep bam | sed -r s/.bam// | awk BEGIN{ print “#!/bin/bash”}{ print “stringtie --dta -p 4 -G ../reference ON_consensus2.maker.gff -o ”$1”.gtf -l “$1” “$1”.bam \n” } > 02_mapped/run_stringtie.sh > cd 02_mapped > chmod 755 ./run_stringtie.sh > ./run_stringtie.sh

9. Merge the transcript models for each of the samples with Stringtie. > ls 02_mapped | grep gtf > 02_mapped/transcript_models_filelist.txt > cd 02_mapped > stringtie merge -p 4 -G ../reference/ON_consensus2.maker.gff

-o

transcript_models.gtf

transcript_models_filelist.txt

10. Quantify the transcript abundance with Stringtie. > stringtie -e -B -p 4 -G transcript_models.gtf -o

../03_quantified/ballgown/Zyg_S1/Zyg_S1.

gtf Zyg_S1.bam > stringtie -e -B -p 4 -G transcript_models.gtf -o

../03_quantified/ballgown/Zyg_S2/Zyg_S2.

490

Tomas Hasing and Aureliano Bombarely

gtf Zyg_S2.bam ## . . .

## Alternatively, to write a bash script: > ls 02_mapped | grep bam | sed -r s/.bam// | awk BEGIN{ print “#!/bin/bash”}{ print “stringtie -e -B -p 4 -G transcript_models.gtf -o ../ 03_quantified/ballgown/”$1”/”$1”.gtf “$1”.bam \n” } > 02_mapped/run_stringtie_quant.sh > cd 02_mapped > chmod 755 ./run_stringtie_quant.sh > ./run_stringtie_quant.sh

11. Finally, the quantified transcripts are uploaded in R using the Ballgown program and the differentially expressed genes are calculated. Open R Studio and, using the “File” tab, navigate to the directory “03_quantified” (or where the Ballgown directory is). Click in the same tab, “More” and then “Set as a working dir.” 12. Create a sample table file with the phenotypes that are being analyzed in the RNA-Seq. The columns can be separated by tabular (.txt) or by commas (.csv). An example of such type of file could be as follows: “ids”,”type”,”tissue”,”replicate” “ZYG_ME01”,”Zygomorphic”,”Flower_meristem”, ”Replicate01” “ZYG_ME02”,”Zygomorphic”,”Flower_meristem”, ”Replicate02” “ZYG_ME03”,”Zygomorphic”,”Flower_meristem”, ”Replicate03” “ZYG_BU01”,”Zygomorphic”,”Flower_bud”, ”Replicate01” “ZYG_BU02”,”Zygomorphic”,”Flower_bud”, ”Replicate02” “ZYG_BU03”,”Zygomorphic”,”Flower_bud”, ”Replicate03” “ACT_ME01”,”Actinomorphic”,”Flower_meristem”, ”Replicate01” “ACT_ME02”,”Actinomorphic”,”Flower_meristem”, ”Replicate02” “ACT_ME03”,”Actinomorphic”,”Flower_meristem”, ”Replicate03”

From Genomes to Genes in Flowering Species

491

“ACT_BU01”,”Actinomorphic”,”Flower_bud”, ”Replicate01” “ACT_BU02”,”Actinomorphic”, ”Flower_bud”,”Replicate02” “ACT_BU03”,”Actinomorphic”, ”Flower_bud”,”Replicate03”

To upload the file (e.g., “PhenoData.txt”), use the commands “read.delim()” (for tabular delimited files) or “read.csv()” (for comma delimited files). > pheno_data = read.csv("PhenoData.txt")

13. In R, upload the Ballgown data and filter the expression data where the sum of all the conditions is below 1 FPKM. ## Upload the expression data and filter the transcripts with FPKM > 1 > bg = ballgown(dataDir = "ballgown", samplePattern = "0", pData = pheno_data) > bg_filt = subset(bg, "rowVars(texpr(bg)) > 1", genomesubset=TRUE)

There are different steps that can be taken in order to assess the quality of the expression data such as the distribution of FPKM for each of the samples and the relation between samples and replicates. Please check the Ballgown manual for more details (https://www. bioconductor.org/packages/release/bioc/vignettes/ballgown/ inst/doc/ballgown.html). 14. Estimate the genes that are differentially expressed between the phenotypes of interest. In the example, the phenotype of interest is actinomorphic versus zygomorphic flowers, so “type” is the covariate variable and “tissue” is the variable for the adjustvars parameter. Once the test has been performed, sort the values based in their p value. > results_genes = stattest(bg_filt, feature="gene", covariate="type", adjustvars = c("tissue"), getFC=TRUE, meas="FPKM") > results_genes=arrange(results_genes,pval)

492

Tomas Hasing and Aureliano Bombarely

15. The expression values can be explored in R or can be exported to a .csv file to later on open it with a program like Microsoft Excel. To export the data, use the command “write.delim()” or “write.csv()” depending of the format. > write.csv(results_genes, "RNASeq_Zyg_vs_Act_Flowers.genesDE.csv", row.names = FALSE)

16. At this point, the list of candidate genes of the QTL and the differential expression analysis can be compared to reduce the number of candidates. To do it, the list of genes obtained during the QTL analysis can be uploaded in R and then compared with the list of DE genes obtained in the RNASeq analysis. ## Upload the QTL gene list > QTL_gene_list = read.table("QTL_GeneID_list. txt") ## Compare both list > intersect(QTL_gene_list$V1, results_genes $id)

4

Notes 1. The equipment needed for the phenotyping will depend on the phenotype of interest. The simplest flower morphology characterization will require a camera, a white and/or a black screen, a ruler, a light source, a computer, and a program to analyze the image (e.g., tpsDig2). More complicated analysis of the morphology, specially at the early stages of flower development, may require more sophisticated techniques of microscopy such as x, y, and z. The characterization of flower color may require the measure of anthocyanin and/or carotenoid content in a specific organ such as the petals or the stamens. 2. Alternatively a bead-mill homogenizer such as a TissueLyser could also be used. 3. Plain boxes are used to describe Linux commands 4. Curly boxes are used to describe R commands

From Genomes to Genes in Flowering Species

493

References 1. Luo D, Carpenter R, Vincent C, Copsey L, Coen E (1996) Origin of floral asymmetry in Antirrhinum. Nature 383:794–799. https:// doi.org/10.1038/383794a0 2. Provart NJ et al (2016) 50 years of Arabidopsis research: highlights and future directions. New Phytol 209:921–944. https://doi.org/10. 1111/nph.13687 3. Saint-Oyant LH et al (2018) A high-quality genome sequence of Rosa chinensis to elucidate ornamental traits. Nat Plants 4:473–484. https://doi.org/10.1038/s41477-0180166-1 4. Raymond O et al (2018) The Rosa genome provides new insights into the domestication of modern roses. Nat Genet 50:772–777. https://doi.org/10.1038/s41588-0180110-3 5. Bombarely A (2018) Roses for Darwin. Nat Plants 4:406–407. https://doi.org/10.1038/ s41477-018-0195-9 6. Cai J et al (2015) The genome sequence of the orchid Phalaenopsis equestris. Nat Genet 47: 65–72. https://doi.org/10.1038/ng.3149 7. Yan L et al (2015) The genome of dendrobium officinale illuminates the biology of the important traditional Chinese orchid herb. Mol Plant 8:922–934. https://doi.org/10.1016/j.molp. 2014.12.011 8. Zhang GQ et al (2017) The Apostasia genome and the evolution of orchids. Nature 549:379– 383. https://doi.org/10.1038/nature23897 9. Bombarely A et al (2016) Insight into the evolution of the Solanaceae from the parental genomes of Petunia hybrida. Nat Plants 2: 16074https://doi.org/10.1038/nplants. 2016.74 10. Li M et al (2019) Genome structure and evolution of Antirrhinum majus L. Nat Plants 5: 174–183. https://doi.org/10.1038/s41477018-0349-9 11. Hasing T et al (2019) Extensive phenotypic diversity in the cultivated Florist’s Gloxinia, Sinningia speciosa (Lodd.) Hiern, is derived from the domestication of a single founder population. Plants People Planet 1:363–374. https://doi.org/10.1002/ppp3.10065 12. Mackay I, Powell W (2007) Methods for linkage disequilibrium mapping in crops. Trends Plant Sci 12:57–63. https://doi.org/10. 1016/j.tplants.2006.12.001 13. Xu Y, Li P, Yang Z, Xu C (2017) Genetic mapping of quantitative trait loci in crops. Crop J 5:175–184. https://doi.org/10. 1016/j.cj.2016.06.003

14. Jung H, Winefield C, Bombarely A, Prentis P, Waterhouse P (2019) Tools and strategies for long-read sequencing and De Novo assembly of plant genomes. Trends Plant Sci 24:700– 724. https://doi.org/10.1016/j.tplants. 2019.05.003 15. Goodwin S et al (2015) Oxford Nanopore sequencing, hybrid error correction, and de novo assembly of a eukaryotic genome. Genome Res 25:1750–1756. https://doi. org/10.1101/gr.191395.115 16. Neely RK, Deen J, Hofkens J (2011) Optical mapping of DNA: single-molecule-based methods for mapping genomes. Biopolymers 95:298–311. https://doi.org/10.1002/bip. 21579 17. Burton JN et al (2013) Chromosome-scale scaffolding of de novo genome assemblies based on chromatin interactions. Nat Biotechnol 31:1119–1125. https://doi.org/10. 1038/nbt.2727 18. Hirsch CN et al (2013) Retrospective view of North American potato (Solanum tuberosum L.) breeding in the 20th and 21st centuries. G3 Genes, Genomes, Genet 3:1003–1013. https://doi.org/10.1534/g3.113.005595 19. Elshire RJ et al (2011) A robust, simple genotyping-by-sequencing (GBS) approach for high diversity species. PLoS One 6: e19379. https://doi.org/10.1371/journal. pone.0019379 20. Schalamun M et al (2019) Harnessing the MinION: an example of how to establish long-read sequencing in a laboratory using challenging plant tissue from Eucalyptus pauciflora. Mol Ecol Resour 19:77–89. https://doi.org/10. 1111/1755-0998.12938 21. Aronesty E (2011) ea-utils: “Command-line tools for processing biological sequencing data”. https://expressionanalysis.github.io/ ea-utils/ 22. Lanfear R, Schalamun M, Kainer D, Wang W, Schwessinger B (2019) MinIONQC: fast and simple quality control for MinION sequencing data. Bioinformatics 35:523–525. https://doi. org/10.1093/bioinformatics/bty654 23. Andrews S (2010) FastQC - A quality control tool for high throughput sequence data. http://www.bioinformatics.babraham.ac.uk/ projects/fastqc/. Babraham Bioinformatics. 24. Li H (2018) Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics 34: 3094–30100. https://doi.org/10.1093/bioin formatics/bty191 25. Vaser R, Sovic´ I, Nagarajan N, Sˇikic´ M (2017) Fast and accurate de novo genome assembly

494

Tomas Hasing and Aureliano Bombarely

from long uncorrected reads. Genome Res 27: 737–746. https://doi.org/10.1101/gr. 214270.116 26. Smit A, Hubley R, Green P RepeatModeler Open-1.0. https://www.repeatmasker.org/ RepeatModeler/ 27. Smit A, Hubley R, Green P RepeatMasker 4.0. https://www.repeatmasker.org 28. Kim D, Langmead B, Salzberg SL (2015) HISAT: a fast spliced aligner with low memory requirements. Nat Methods 12:357–360. https://doi.org/10.1038/nmeth.3317 29. Pertea M et al (2015) StringTie enables improved reconstruction of a transcriptome from RNA-seq reads. Nat Biotechnol 33:290– 295. https://doi.org/10.1038/nbt.3122 30. Slater GSC, Birney E (2005) Automated generation of heuristics for biological sequence comparison. BMC Bioinformatics 6:31. https://doi.org/10.1186/1471-2105-6-31 31. Korf I (2004) Gene finding in novel genomes. BMC Bioinformatics 5:59. https://doi.org/ 10.1186/1471-2105-5-59 32. Stanke M et al (2006) AUGUSTUS: ab initio prediction of alternative transcripts. Nucleic Acids Res 34:W435–W439. https://doi.org/ 10.1093/nar/gkl200 33. Campbell MS et al (2014) MAKER-P: a tool kit for the rapid creation, management, and quality control of plant genome annotations. Plant Physiol 164:513–524. https://doi.org/ 10.1104/pp.113.230144 34. Herten K, Hestand MS, Vermeesch JR, Van Houdt JKJ (2015) GBSX: a toolkit for experimental design and demultiplexing genotyping by sequencing experiments. BMC Bioinformatics 16:73. https://doi.org/10.1186/ s12859-015-0514-3 35. Langmead B, Salzberg SL (2012) Fast gappedread alignment with Bowtie 2. Nat Methods 9: 357–359. https://doi.org/10.1038/nmeth. 1923 36. Li H et al (2009) The sequence alignment/ map format and SAMtools. Bioinformatics 25: 2078–2079. https://doi.org/10.1093/bioin formatics/btp352 37. Garrison E, Marth G (2012) Haplotype-based variant detection from short-read sequencing. arXiv:1207.3907. https://doi.org/10.48550/ arXiv.1207.3907 38. Danecek P et al (2011) The variant call format and VCFtools. Bioinformatics 27:2156–2158. https://doi.org/10.1093%2Fbioinformatics% 2Fbtr330

39. Lischer HEL, Excoffier L (2012) PGDSpider: an automated data conversion tool for connecting population genetics and genomics programs. Bioinformatics 28:298–299. https:// doi.org/10.1093/bioinformatics/btr642 40. Broman KW, Wu H, Sen S´, Churchill GA (2003) R/qtl: QTL mapping in experimental crosses. Bioinformatics 19:889–890. https:// doi.org/10.1093/bioinformatics/btg112 41. Margarido GRA, Souza AP, Garcia AAF (2007) OneMap: software for genetic mapping in outcrossing species. Hereditas 144:78–79. https://doi.org/10.1111/j.2007.0018-0661. 02000.x 42. Frazee AC et al (2015) Ballgown bridges the gap between transcriptome assembly and expression analysis. Nat Biotechnol 33:243– 246. https://doi.org/10.1038/nbt.3172 43. Chin CS et al (2013) Nonhybrid, finished microbial genome assemblies from long-read SMRT sequencing data. Nat Methods 10: 563–569. https://doi.org/10.1038/nmeth. 2474 44. Koren S et al (2017) Canu: scalable and accurate long-read assembly via adaptive κ-mer weighting and repeat separation. Genome Res 27:722–736. https://doi.org/10.1101/gr. 215087.116 45. Kolmogorov M, Yuan J, Lin Y, Pevzner PA (2019) Assembly of long, error-prone reads using repeat graphs. Nat Biotechnol 37:540– 546. https://doi.org/10.1038/s41587-0190072-8 46. Soorni A, Haak D, Zaitlin D, Bombarely A (2017) Organelle_PBA, a pipeline for assembling chloroplast and mitochondrial genomes from PacBio DNA sequencing data. BMC Genomics 18:49 47. Doyle JJ, Doyle JL (1991) DNA isolation from small amount of plant tissue. Phytochem Bull. https://doi.org/10.2307/4119796 48. Wallace JG, Mitchell SE (2017) Genotypingby-sequencing. Curr Protoc Plant Biol. https://doi.org/10.1002/cppb.20042 49. Lander ES et al (1987) MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1:174–181. https://doi.org/10.1016/0888-7543(87) 90010-3 50. Zhou Q (2010) A guide to QTL mapping with R/qtl. J Stat Softw 32:1–3. https://doi.org/ 10.18637/jss.v032.b05

Chapter 23 Multi-Omics Methods Applied to Flower Development Raquel A´lvarez-Urdiola, Jose´ Toma´s Matus, and Jose´ Luis Riechmann Abstract Developmental processes in multicellular organisms depend on the proficiency of cells to orchestrate different gene expression programs. Over the past years, several studies of reproductive organ development have considered genomic analyses of transcription factors and global gene expression changes, modeling complex gene regulatory networks. Nevertheless, the dynamic view of developmental processes requires, as well, the study of the proteome in its expression, complexity, and relationship with the transcriptome. In this chapter, we describe a dual extraction method—for protein and RNA—for the characterization of genome expression at proteome level and its correlation to transcript expression data. We also present a shotgun proteomic procedure (LC-MS/MS) followed by a pipeline for the imputation of missing values in mass spectrometry results. Key words Protein extraction, RNA extraction, Proteomics, Transcriptomics, Flower development, LC-MS/MS, Arabidopsis

1

Introduction The capacity of cells to orchestrate different gene expression programs is crucial for developmental processes in multicellular organisms, and it is hardwired and encoded in the genome in the form of cis-regulatory sequences that interact with transcription factors, co-regulators, and other types of regulatory proteins or RNAs, as well as of epigenetic marks, altogether determining when, where, and how genes are expressed. For the past 20 years, the exponential advances in technologies and informatics tools for generating and processing large biological datasets (omics) have added new approaches to development studies in plants. Through the use of genomics and transcriptomics (in particular, RNA-Seq, ChIP-Seq, and other high-throughput sequencing-derived methods), the hierarchical levels of plant genetic and molecular organization are being described in detail. In particular, several studies of reproductive organ development have considered genome-wide analyses of transcription factor DNA-binding and global gene expression

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_23, © Springer Science+Business Media, LLC, part of Springer Nature 2023

495

496

Raquel A´lvarez-Urdiola et al.

changes (e.g., [1–5]) and modeled complex gene regulatory networks (reviewed in [6–9]). Even so, a global and comprehensive view of developmental processes would also benefit from the characterization of the corresponding proteome. The analysis of the proteome of eukaryotic cells is challenging due to the substantial diversity in the properties of the individual proteins that compose it (e.g., abundance, stability, molecular weight, structure, hydrophobicity, hydrophilicity, posttranslational modifications (PTMs), and so on). Nevertheless, along with an enhancement of throughput, sensitivity, and resolution of analytical technologies in MS, computational methods have been developed focusing on the identification and quantification of proteins in complex samples [10–13]. In plants, MS-based proteomics approaches have been applied for the measurement of differential protein expression or the detection of PTMs (e.g., [14, 15]) in different tissues and biological processes (reviewed in [13]). Deep proteome studies have led to the development of proteome atlases of the major plant organs for different plant species [16– 21]. Besides, cell type-specific proteome studies are crucial for a better understanding of the unique biological functions and properties of individual cell types in a tissue [22], as well as subcellular plant proteomics and predictions [23–25]. As the proteome is in constant flux, several proteome studies are based on temporal series during developmental processes or stress responses [26–29]. Furthermore, results from more than one type of omics can be matched in order to obtain deeper insights into biological processes [16, 30–33]. These integration studies are usually referred as multiomics, trans-omics, or integrated omics in current literature. Quantitative proteomics allows to study at a genome-wide level the correlation between mRNA expression levels and the abundance of the corresponding proteins (reviewed in [34, 35]), an issue that has been extensively studied in different species and processes during the past few years. For instance, in plants combined transcriptome-proteome analyses have already been used to study petal shape [36], carotenoid synthesis [37], photoperiodic control of the proteome [38], or leaf development [39], as well as reproductive development; in particular, embryogenesis [40], male reproductive development [41–43], and flower development either in general [44, 45] or focusing on the functions of specific proteins [46]. In these combined studies, the interpretation of the existence, or lack thereof, of correlation between the changes in transcript dynamics and protein abundance, and its biological meaning, is still a lingering issue: numerous studies conclude that there is not a strong correlation between the levels of these macromolecules [41, 43, 47–51], whereas in others such correlation is more apparent [38–40, 45]. The lack of correlation could be in part derived from the difficulties to obtain truly comparable datasets at the

Flower Development Multi-Omics

497

transcript and protein levels, and because the sensitivity of extraction and quantification techniques for mRNAs and proteins highly differ. However, the observed differences might also be caused by posttranslational regulation of protein levels [47], or by their different expression and degradation kinetics, as longer protein halflives buffer changes in mRNA levels [48–51]. Time-lapse studies could be an approach for addressing this gap, as successive analyses at different time points could allow the discovery of correlative behaviors of protein and mRNA levels through time [52, 53]. In addition, a major concern in label-free quantitative proteomics that hinders the subsequent data analysis and its comparison with other omics data is the high rate of missing values. Three types of missing values can be defined, depending on the nature of the missingness: (1) missing completely at random (MCAR) and (2) missing at random (MAR) values, which are due to minor errors or stochastic fluctuations and to conditional dependencies, respectively; and (3) missing not at random (MNAR) values, which have a targeted effect [54]. Depending on the nature of these “not assigned values” (NAs), different methods can be used to impute them. As there are many types of NAs that coexist in most quantitative datasets, hybrid strategies of imputation could be a better approach [54, 55]. In this chapter, we describe a protocol for common extraction of total proteins and RNA from the same Arabidopsis inflorescence samples to maximize comparability between the proteomic and transcriptomic data. We also present a shotgun proteomic procedure by liquid chromatography-tandem mass spectrometry (LC-MS/MS), and a pipeline for the imputation of missing values in the mass spectrometry results to distinguish the nature of the missingness and to treat NAs accordingly.

2

Materials 1. Mortar and pestle. 2. Liquid nitrogen. 3. Microcentrifuge tubes.

2.1 Protein Extraction

1. Protein low-binding tubes (2 mL). 2. Isopropanol. 3. 0.3 M guanidine in 95% ethanol. 4. 90% ethanol. 5. SDS-PAGE 5× buffer. 6. E buffer: 125 mM Tris–HCl pH 8.8, 1% (w/v) SDS, 10% (v/v) glycerol, 50 mM Na2S2O5 [56].

498

2.2

Raquel A´lvarez-Urdiola et al.

RNA Extraction

1. RNase free tubes (1.5 mL). 2. Trizol. 3. Chloroform. 4. Phenol:chloroform:isoamyl alcohol (25:24:1). 5. LiCl 3 M. 6. 85% and 100% (v/v) ethanol. 7. DEPC water.

2.3

LC-MS/MS

1. DL-dithiothreitol (DTT) (see Note 1). 2. Iodoacetamide. 3. Urea. 4. Ammonium bicarbonate. 5. Endoproteinase LysC. 6. Trypsin. 7. Formic acid. 8. MicroSpin C18 columns (The Nest Group, Inc). 9. Nano Trap C18 columns with an inner diameter of 100 μm packed with C18 particles of 5 μm particle size (Thermo Fisher Scientific) (optional, depending on the setup of each laboratory). 10. Reverse-phase chromatography 15–50 cm length) (see Note 2).

columns

(C18,

2

μm,

11. Buffer A: 0.1% formic acid in water. 12. Buffer B: 0.1% formic acid in acetonitrile. 13. Bovine serum albumin (New England Biolabs cat # P8108S). 14. Orbitrap Eclipse mass spectrometer (Thermo Fisher Scientific) (see Note 3). 15. EASY-nLC 1000 (Thermo Fisher Scientific).

3

Methods

3.1 Protein Extraction

1. With a different mortar and pestle for each sample, grind the tissue (i.e., inflorescences) with liquid nitrogen until obtaining a whitish fine powder (see Notes 4 and 5). 2. Place the powder in a microcentrifuge tube (~250 mg per sample). 3. Add 1 mL of Trizol, vortex for at least 15 s until it is completely homogenized, and incubate on ice for 5 min. This step must be done in an extraction hood.

Flower Development Multi-Omics

499

Fig. 1 Picture of the three phases formed in step 4 of the protein extraction method (see Subheading 3.1)

4. Add 200 μL of chloroform, vortex for 15 s, incubate on ice for 5 min, and centrifuge at 4 °C for 15 min at maximum speed (see Note 6) (Fig. 1). 5.a. Transfer 500–600 μL of the top, aqueous phase into a clean microcentrifuge tube (RNase free) and add the same volume of phenol:chloroform:isoamyl alcohol, vortex for 10 s, incubate on ice for 5 min, and centrifuge at 4 °C for 15 min at maximum speed (to continue with RNA extraction from the sample, see Subheading 3.2). 5.b. Add 300 μL of ethanol 100% to the organic phase in the original microcentrifuge tube to continue with protein extraction. Incubate on ice. 6. Centrifuge for 10 min at 2000 g. Place the supernatant in a clean 2 mL microcentrifuge tube (protein low bind). 7. Add 1 mL of isopropanol and incubate at room temperature for 10 min (see Note 7). 8. Centrifuge at 4 °C for 10 min at 12,000 g. Discard supernatant, which contains phenol, into a container adequate for its controlled elimination. 9. Wash by resuspending the pellet in 2 mL of a solution of 0.3 M guanidine in 95% ethanol (see Note 8). 10. Sonicate in a sonication bath for 5 min and centrifuge at 4 °C for 5 min at 8000 g.

500

Raquel A´lvarez-Urdiola et al.

11. Repeat the washing procedure (steps 9 and 10) twice. The obtained pellet can be stored at -20 °C for months. 12. Wash again by the same procedure (steps 9–11) with 90% ethanol. 13. Let the pellet dry for a few minutes and resuspend in an appropriate buffer (see Note 9). 14. Quantify by Bradford with 1 and 2 μL of sample. Add SDS-PAGE 5× buffer to obtain a final 1× concentration when loading the gel. 3.2

RNA Extraction

1. Transfer approximately 500 μL of the top, aqueous phase after the centrifugation in protein extraction step 5.a to a clean microcentrifuge tube (RNase free) and add 1 volume (500 μL) of pure isopropanol. Shake and mix. 2. Incubate on ice for 15 min, centrifuge at 4 °C for 10 min at maximum speed, and discard supernatant. 3. Resuspend the pellet in 750 μL of LiCl 3 M, incubate on ice for 10 min, and centrifuge at 4 °C for 10 min at maximum speed. 4. Discard supernatant and wash the pellet with 500 μL of ethanol 85% (v/v), vortexing gently for 10 s. 5. Centrifuge at 4 °C for 10 min at maximum speed and discard supernatant. 6. Let the pellet dry and resuspend in 21 μL of diethylpyrocarbonate (DEPC)-treated water (see Note 10). 7. Sample quantification with NanoDrop spectrophotometer.

3.3

LC-MS/MS

3.3.1 Sample Preparation

1. Prepare or dissolve protein samples (see Subheading 3.1, step 13) in 6 M urea 200 nM ammonium bicarbonate. 2. Reduce the samples (10 μg of protein) with 30 nmol DTT at 37 °C for 1 h. 3. Alkylate the samples (10 μg of protein) in the dark with 60 nmol of iodoacetamide at 25 °C for 30 min. 4. Dilute the protein extract to 2 M urea with 200 mM ammonium bicarbonate for digestion with endoproteinase LysC (1:10 w:w), and incubate 37 °C overnight. 5. Dilute twofold with 200 mM ammonium bicarbonate for trypsin digestion (1:10 w:w), and incubate at 37 °C for 8 h. 6. After digestion, add formic acid (10% of the final volume) to acidify the peptide mix. 7. Desalt the samples with MicroSpin C18 columns prior to LC-MS/MS analysis, following manufacturer’s instructions.

Flower Development Multi-Omics 3.3.2 Chromatographic and Mass Spectrometric Analysis

501

1. Load the peptides onto the analytical column (C18, 2 μm, 15–50 cm length). 2. Separation of the peptides by reverse-phase chromatography with the corresponding columns. 3. Chromatographic gradients start at 93% buffer A and 7% buffer B with a flow rate of 250 nL/min for 5 min and gradually increase 65% buffer A and 35% buffer B in 60 min. 4. After each analysis, wash the column for 15 min with 10% buffer A and 90% buffer B. 5. Peptide eluates are dried in a vacuum centrifuge, and resuspended with buffer A at a final concentration of 1 μg/μL prior to analysis by LC-MS/MS. 6. Operate the mass spectrometer to acquire peptide spectra (see Note 11).

3.3.3

Data Analysis

1. Search the acquired spectra against the desired peptide database (see Note 12), plus a list of common contaminants (suggested: [57]), and all the corresponding decoy entries. 2. Set the parameters accordingly to the experimental and mass spectrometric settings and, if appropriate, select variable posttranslational modifications to be detected (see Note 13). 3. Determine the protein abundance estimation [58, 59]. 4. Add the information to the appropriate repositories (see Note 14).

3.3.4 Treatment of Missing Values and Data Imputation

1. Missing values should first be classified as M(C)AR or MNAR depending on their nature. For instance, for a given protein, if the data from all replicates of the same condition or time point show NAs, probably they are MNAR missing values, whereas if there is only one missing value out of four replicates, it is probably a MAR. Other cases may be more difficult to classify as M(C)AR or MNAR, for instance if there are two NAs out of four replicates. In those instances, other parameters can be considered, for example, the presence or absence of NAs in the adjacent time points (in a time-course experiment) or in the most similar samples in the experiment. 2. Discard all proteins with MNARs or MARs in every sample. 3. Replace MNARs by the minimum of detection of the dataset (deterministic minimum imputation method [60]). 4. Estimate the remaining MARs and MCARs by other imputation method (e.g., k-nearest neighbor (kNN) imputation [61]).

502

Raquel A´lvarez-Urdiola et al.

Fig. 2 Stringent analysis to identify reliably undetected and detected fraction of a proteome. The analysis allows to impute values for MAR and MNAR considering their biological meaning. The figure illustrates results from a time-course experiment using the Arabidopsis floral induction system pAP1:AP1-GR ap1cal [1], in which samples were collected at 1-day intervals after floral induction (day 0), up to day 5. Log2 TOP3 abundances through time of two flower development regulators, APETALA 3 (AP3) (a) and TERMINAL FLOWER 1 (TFL1) (b), before and after the “reliability analysis” (RA), and after kNN imputation (from left to right) (n = 4 biological replicates) 3.3.5 Example: Treatment of Missing Values in a Time Series Experiment

This missing value classification and data imputation approach can be readily used in, for instance, time-course developmental studies [1, 62], as illustrated in Fig. 2 as an example. In this case, the data processing pipeline consisted on: 1. Classification of each time point (day) for each protein depending on its number of NAs (number of replicates with missing values at a certain time point) and the number of NAs of its immediately adjacent days (neighbors). (a) Neighbors are considered as: • Unreliable neighbor: Over 50% NAs. • Reliable neighbor: Up to 50% NAs (included). (b) Initial and final time points are considered as: • Reliably undetected: 100% NAs (MNARs). • Unreliably undetected: Over 50% NAs (included) (unclear MNARs) + unreliable neighbor.

Flower Development Multi-Omics

503

• Unreliably detected: Over 50% NAs (included) (unclear MARs) + reliable neighbor. • Reliably detected: Up to 35% NAs (MARs). (c) Intermediate time points are considered as: • Reliably undetected: 100% NAs + unreliable neighbors (MNARs). • Unreliably undetected: Over 50% NAs (included) + unreliable neighbors (probably MNARs). • Unreliably detected: Over 50% NAs (included) + reliable neighbors (probably MARs). • Reliably detected: Up to 35% NAs (MARs). 2. Replace reliably undetected time points by the minimum of detection of the dataset (deterministic minimum imputation method [60]). 3. Replace unreliably undetected time points by NAs in all replicates. 4. Discard all proteins which are reliably or unreliably undetected in every time point. 5. Estimate the remaining NAs by k-nearest neighbor (kNN) imputation (k = 10) [61].

4

Notes 1. Reagents for LC-MS/MS can be obtained from several suppliers. As an example, we list here the specific products we use: urea (GE Healthcare; Sigma-Aldrich, P/N 17-1319-01), ammonium bicarbonate (BioUltra, ≥99.5% (T); SigmaAldrich, P/N 09830), iodoacetamide (BioUltra; SigmaAldrich, P/N I1149), DL-dithiothreitol (for electrophoresis, ≥99%; Sigma-Aldrich, P/N D9163), formic acid for analysis EMSURE® (ACS Reag. Merck, P/N 1.00264.0100), sequencing grade modified trypsin (Promega, P/N V5111), and lysyl endopeptidase (Wako Chemicals GmbH, P/N 129-02541). 2. Suitable reverse-phase chromatography columns are, for instance, 25 cm columns with an inner diameter of 75 μm, packed with 1.9 μm C18 particles (Nikkyo Technos Co.); and 50 cm columns with an inner diameter of 75 μm, packed with 2 μm C18 particles (EASY-Column, Thermo Fisher Scientific, ES903). 3. This is just a concrete example of a “modern high-resolution mass spectrometer”; other instruments could be used. 4. For sample collection, to reduce sample contamination with human proteins (i.e., keratins and collagen), make sure to

504

Raquel A´lvarez-Urdiola et al.

always use nitrile gloves (instead of latex) and laboratory coats. Pipets, materials, and solutions exclusively used for proteomics. Take precaution to avoid hair contamination. If flower organs or tissues are going to be dissected, cool tweezers and any other sampling instrument with liquid nitrogen. 5. If samples are grown in petri dishes (e.g., Arabidopsis seedlings), discard white clots which correspond to agar. 6. Three phases are formed, the aqueous phase contains RNA (~550 μL, transparent), the interphase, DNA (white), and the organic phase, proteins and lipids (~450 μL, pink) (Fig. 1). 7. It is possible to stop the protocol here and store the samples at -20 °C for a few days. 8. Use a pipette crushing against the bottom of the tube and leave in a colloidal suspension as thin as possible. 9. Resuspend final proteins in acetonitrile, acetic, or formic acid, depending on the analysis protocol. For Western Blot, use E buffer [56]. The buffer volume should be chosen depending on the desired protein concentrations, varying from 20 to 50 μL. 10. Use high pure water, reagents, and products. 11. 1–2 μg of peptides are loaded onto an analytical column (25 cm, C18 2 μm particle size) using an autosampler device (e.g., EASY nLC 1000, Thermo Fisher Scientific) and the peptides are then separated by reverse-phase chromatography using a water-acetonitril chromatographic gradient. Modern high-resolution mass spectrometers are recommended for data acquisition (e.g., Orbitrap or qTOF). The mass spectrometer is operated in data-dependent acquisition (DDA) mode, in which a full MS scan is recorded in each cycle, followed by the fragmentation of the 10–30 most intense precursor ions to obtain the fragment ion spectra. 12. The results may vary significantly depending on the characteristics of the reference database for peptide identification. It is possible to use public repositories of proteins for the different organisms or to design a specific database. 13. Once the database has been constructed, the raw LC-MS/MS data needs to be interpreted using a database search engine (such as SEQUEST [63], Mascot [64], Phenyx [65], X! Tandem [66], OMSSA [67], pFind [68], InsPecT [69], ByOnic [70], Comet [71], MS-GF+ [72], MaxQuant [73], or MSTracer [74]). As example, the Mascot search engine (v2.6) can be used, using the search parameters accordingly to the experimental and mass spectrometry settings. For peptide identification a precursor ion mass tolerance below 10–20 ppm is recommended, whereas the fragment ion mass tolerance can go from 10 to 20 ppm for high-resolution mass analyzers

Flower Development Multi-Omics

505

(Orbitrap, TOF) to 0.5 Da if a linear ion trap is used for the analysis of the tandem mass spectra. Common peptide modifications such as oxidation of methionine and N-terminal protein acetylation are used as variable modifications. False discovery rate (FDR) in peptide identification is set to a maximum of 1%. 14. Share data and results in a public repository. Data sharing in the public domain is the standard for omics research and a requirement for publication. For proteomics, the Proteomics IDEntifications (PRIDE) database (https://www.ebi.ac.uk/pride/) at the European Bioinformatics Institute (EMBL-EBI, Hinxton, Cambridge, UK) has enabled public data deposition of MS data since 2004, and its archival component has become the largest repository for proteomics data sharing worldwide [75]. The PRIDE database provides access to most of the experimental proteomics data described in MS-related scientific publications.

Acknowledgments Work in the authors’ laboratory was supported by grant BFU201458289-P (funded by MICIN/AEI/10.13039/501100011033 and by “ERDF A way of making Europe”) and by grant 2017SGR718 (from the Agencia de Gestio´ d’Ajuts Universitaris I de Recerca) to JLR, and by institutional grant SEV-2015-0533 (funded by MCIN/AEI/10.13039/501100011033) and by the CERCA Programme/Generalitat de Catalunya. R.A. is supported by fellowship PRE2018-084278 funded by MCIN/AEI/ 10.13039/501100011033 and by “ESF Investing in your future.” We are grateful to Eva Borra`s and Eduard Sabido´ from the CRG/UPF Proteomics Unit for their advice and help in proteomics research. References ˜ o JM, Ferrier T, 1. Kaufmann K, Wellmer F, Muin Wuest SE, Kumar V et al (2010) Orchestration of floral initiation by APETALA1. Science 328(85):85–89 ´ ’Maoile´idigh DS, Thomson B, Raganelli A, 2. O Wuest SE, Ryan PT, Kwas¨niewska K et al (2015) Gene network analysis of Arabidopsis thaliana flower development through dynamic gene perturbations. Plant J 83(2):344–358 3. Chen D, Yan W, Fu LY, Kaufmann K (2018) Architecture of gene regulatory networks controlling flower development in Arabidopsis thaliana. Nat Commun 9:4534

˜ o JM, Matus JT, 4. Pajoro A, Madrigal P, Muin Jin J, Mecchia MA et al (2014) Dynamics of chromatin accessibility and gene regulation by MADS-domain transcription factors in flower development. Genome Biol 15:R41 5. Wuest SE, O’Maoileidigh DS, Rae L, Kwasniewska K, Raganelli A, Hanczaryk K et al (2012) Molecular basis for the specification of floral organs by APETALA3 and PISTILLATA. Proc Natl Acad Sci U S A 109(33): 13452–13457

506

Raquel A´lvarez-Urdiola et al.

6. Wellmer F, Riechmann JL (2010) Gene networks controlling the initiation of flower development. Trends Genet 26(12):519–527 7. Wils CR, Kaufmann K (2017) Gene-regulatory networks controlling inflorescence and flower development in Arabidopsis thaliana. Biochim Biophys Acta Gene Regul Mech 1860(1): 95–105 8. Pajoro A, Biewers S, Dougali E, Valentim FL, Mendes MA, Porri A et al (2014) The (r)evolution of gene regulatory networks controlling Arabidopsis plant reproduction: a two-decade history. J Exp Bot 65(17): 4731–4745 9. Heisler MG, Jo¨nsson H, Wenkel S, Kaufmann K (2022) Context-specific functions of transcription factors controlling plant development: from leaves to flowers. Curr Opin Plant Biol 69:102262 10. Taka´cˇ T, Sˇamajova´ O, Sˇamaj J (2017) Integrating cell biology and proteomic approaches in plants. J Proteome 169:165–175 11. Aslam B, Basit M, Nisar MA, Khurshid M, Rasool MH (2017) Proteomics: technologies and their applications. J Chromatogr Sci 55(2): 182–196 12. Grossmann J, Roschitzki B, Panse C, Fortes C, Barkow-Oesterreicher S, Rutishauser D et al (2010) Implementation and evaluation of relative and absolute quantification in shotgun proteomics with label-free methods. J Proteome 73(9):1740–1746 13. Mergner J, Kuster B (2022) Plant proteome dynamics. Annu Rev Plant Biol 73:67–92 14. Zhang Z, Hu M, Feng X, Gong A, Cheng L, Yuan H (2017) Proteomes and phosphoproteomes of anther and pollen: availability and progress. Proteomics 17(20). https://doi. org/10.1002/pmic.201600458 15. Navrot N, Finnie C, Svensson B, H€agglund P (2011) Plant redox proteomics. J Proteome 74(8):1450–1462 16. Mergner J, Frejno M, List M, Papacek M, Chen X, Chaudhary A et al (2020) Massspectrometry-based draft of the Arabidopsis proteome. Nature 579(7799):409–414 17. Kumar M, Carr P, Turner SR (2022) An atlas of Arabidopsis protein S-acylation reveals its widespread role in plant cell organization and function. Nat Plants 8(6):670–681 18. Abraham P, Gannone RJ, Adams RM, Kalluri U, Tuskan GA, Hettich RL (2013) Putting the pieces together: high-performance LC-MS/MS provides network-, pathway-, and protein-level perspectives in Populus. Mol Cell Proteomics 12(1):106–119

19. Szymanski J, Levin Y, Savidor A, Breitel D, Chappell-Maor L, Heinig U et al (2017) Label-free deep shotgun proteomics reveals protein dynamics during tomato fruit tissues development. Plant J 90(2):396–417 20. Duncan O, Tro¨sch J, Fenske R, Taylor NL, Millar AH (2017) Resource: mapping the Triticum aestivum proteome. Plant J 89(3): 601–616 21. Marx H, Minogue CE, Jayaraman D, Richards AL, Kwiecien NW, Siahpirani AF et al (2016) A proteomic atlas of the legume, M. truncatula, and its nitrogen fixing endosymbiont, S. meliloti. Nat Biotechnol 34(11):1198 22. Dai S, Chen S (2012) Single-cell-type proteomics: toward a holistic understanding of plant function. Mol Cell Proteomics 11(12): 1622–1630 23. Emanuelsson O, Von Heijne G, Schneider G (2001) Analysis and prediction of mitochondrial targeting peptides. Methods Cell Biol 65:175–187 24. Bruce BD (2000) Chloroplast transit peptides: structure, function and evolution. Trends Cell Biol 10(10):440–447 25. Bernhofer M, Goldberg T, Wolf S, Ahmed M, Zaugg J, Boden M et al (2018) NLSdb—major update for database of nuclear localization signals and nuclear export signals. Nucleic Acids Res 46(D1):D503–D508 26. Bassal M, Abukhalaf M, Majovsky P, Thieme D, Herr T, Ayash M et al (2020) Reshaping of the Arabidopsis thaliana proteome landscape and co-regulation of proteins in development and immunity. Mol Plant 13(12):1709–1732 27. Feng Z, Kong D, Kong Y, Zhang B, Yang X (2022) Coordination of root growth with root morphology, physiology and defense functions in response to root pruning in Platycladus orientalis. J Adv Res 36:187–199 28. Jain A, Singh HB, Das S (2021) Deciphering plant-microbe crosstalk through proteomics studies. Microbiol Res 242:126590 29. Niu Z, Liu L, Pu Y, Ma L, Wu J, Hu F et al (2021) iTRAQ-based quantitative proteome analysis insights into cold stress of winter rapeseed (Brassica rapa L.) grown in the field. Sci Rep 11:23434 30. Koehler G, Rohloff J, Wilson RC, Kopka J, Erban A, Winge P et al (2015) Integrative “omic” analysis reveals distinctive cold responses in leaves and roots of strawberry, fragaria × ananassa ‘Korona’. Front Plant Sci 6:826

Flower Development Multi-Omics 31. Le Signor C, Aime´ D, Bordat A, Belghazi M, Labas V, Gouzy J et al (2017) Genome-wide association studies with proteomics data reveal genes important for synthesis, transport and packaging of globulins in legume seeds. New Phytol 214(4):1597–1613 32. Zhu G, Wang S, Huang Z, Zhang S, Liao Q, Zhang C et al (2018) Rewiring of the fruit metabolome in tomato breeding. Cell 172(1–2):249–261.e12 33. Lehmann BD, Colaprico A, Silva TC, Chen J, An H, Ban Y et al (2021) Multi-omics analysis identifies therapeutic vulnerabilities in triplenegative breast cancer subtypes. Nat Commun 12:6276 34. Manzoni C, Kia DA, Vandrovcova J, Hardy J, Wood NW, Lewis PA et al (2018) Genome, transcriptome and proteome: the rise of omics data and their integration in biomedical sciences. Brief Bioinform 19(2):286–302 35. Kumar D, Bansal G, Narang A, Basak T, Abbas T, Dash D (2016) Integrating transcriptome and proteome profiling: strategies and applications. Proteomics 16(19):2533–2544 36. Wu Y, Tang Y, Jiang Y, Zhao D, Shang J, Tao J (2018) Combination of transcriptome sequencing and iTRAQ proteome reveals the molecular mechanisms determining petal shape in herbaceous peony (Paeonia lactiflora Pall.). Biosci Rep 38(6):BSR20181485 37. Decourcelle M, Perez-Fons L, Baulande S, Steiger S, Couvelard L, Hem S et al (2015) Combined transcript, proteome, and metabolite analysis of transgenic maize seeds engineered for enhanced carotenoid synthesis reveals pleotropic effects in core metabolism. J Exp Bot 66(11):3141–3150 38. Seaton DD, Graf A, Baerenfaller K, Stitt M, Millar AJ, Gruissem W (2018) Photoperiodic control of the Arabidopsis proteome reveals a translational coincidence mechanism. Mol Syst Biol 14(3):e7962 39. Omidbakhshfard MA, Sokolowska EM, Di Vittori V, Perez de Souza L, Kuhalskaya A, Brotman Y et al (2021) Multi-omics analysis of early leaf development in Arabidopsis thaliana. Patterns 2(4):100235 40. Huang Y, Zhou L, Hou C, Guo D (2022) The dynamic proteome in Arabidopsis thaliana early embryogenesis. Development 149(18): dev200715 41. Keller M, Simm S, Bokszczanin KL, Bostan H, Bovy A, Chaturvedi P et al (2018) The coupling of transcriptome and proteome adaptation during development and heat stress response of tomato pollen. BMC Genomics 19(1):447

507

42. Ji J, Yang L, Fang Z, Zhuang M, Zhang Y, Lv H et al (2018) Complementary transcriptome and proteome profiling in cabbage buds of a recessive male sterile mutant provides new insights into male reproductive development. J Proteome 179:80–91 43. Xing M, Sun C, Li H, Hu S, Lei L, Kang J (2018) Integrated analysis of transcriptome and proteome changes related to the ogura cytoplasmic male sterility in cabbage. PLoS One 13(3):e0193462 44. Jing D, Chen W, Hu R, Zhang Y, Xia Y, Wang S et al (2020) An integrative analysis of transcriptome, proteome and hormones reveals key differentially expressed genes and metabolic pathways involved in flower development in loquat. Int J Mol Sci 21(14):5107 45. Chen R, Chen G, Huang J (2017) Shot-gun proteome and transcriptome mapping of the jujube floral organ and identification of a pollen-specific S-locus F-box gene. PeerJ 5: e3588 46. Lu D, Ni W, Stanley BA, Ma H (2016) Proteomics and transcriptomics analyses of Arabidopsis floral buds uncover important functions of ARABIDOPSIS SKP1-LIKE1. BMC Plant Biol 16:61 47. Vogel C, Marcotte EM (2012) Insights into regulation of protein abundance from proteomics and transcriptomis analyses. Nat Rev Genet 13(4):227–232 48. Raj A, Peskin CS, Tranchina D, Vargas DY, Tyagi S (2006) Stochastic mRNA synthesis in mammalian cells. PLoS Biol 4(10):1707–1719 49. Taniguchi Y, Choi PJ, Li G, Chen H, Babu M, Hearn J et al (2011) Quantifying E. coli proteome and transcriptome with single-molecule sensitivity in single cells. Science 329(5991): 533–539 50. Csa´rdi G, Franks A, Choi DS, Airoldi EM, Drummond DA (2015) Accounting for experimental noise reveals that mRNA levels, amplified by post-transcriptional processes, largely determine steady-state protein levels in yeast. PLoS Genet 11(5):1005206 51. Oliva-Vilarnau N, Vorrink SU, IngelmanSundberg M, Lauschke VM (2020) A 3D cell culture model identifies Wnt/β-catenin mediated inhibition of p53 as a critical step during human hepatocyte regeneration. Adv Sci 7(15):2000248 ˜ es T, Novais SC, Natal-da-Luz T, 52. Simo Devreese B, de Boer T, Roelofs D et al (2019) Using time-lapse omics correlations to integrate toxicological pathways of a formulated fungicide in a soil invertebrate. Environ Pollut 246:845–854

508

Raquel A´lvarez-Urdiola et al.

53. Tarazona S, Balzano-Nogueira L, Conesa A (2018) Multiomics data integration in time series experiments. Compr Anal Chem 82: 505–532 54. Lazar C, Gatto L, Ferro M, Bruley C, Burger T (2016) Accounting for the multiple natures of missing values in label-free quantitative proteomics data sets to compare imputation strategies. J Proteome Res 15:1116–1125 55. Jin L, Bi Y, Hu C, Qu J, Shen S, Wang X et al (2021) A comparative study of evaluating missing value imputation methods in label-free proteomics. Sci Rep 11:1720 56. Martı´nez-Garcı´a JF, Monte E, Quail PH (1999) A simple, rapid and quantitative method for preparing Arabidopsis protein extracts for immunoblot analysis. Plant J 20: 251–257 57. Beer LA, Liu P, Ky B, Barnhart KT, Speicher DW (2017) Efficient quantitative comparisons of plasma proteomes using label-free analysis with MaxQuant. Methods Mol Biol 1619: 339–352 58. Gerster S, Kwon T, Ludwig C, Matondo M, Vogel C, Marcotte EM et al (2014) Statistical approach to protein quantification. Mol Cell Proteomics 13(2):666–677 59. Fabre B, Lambour T, Bouyssie´ D, Menneteau T, Monsarrat B, Burlet-Schiltz O et al (2014) Comparison of label-free quantification methods for the determination of protein complexes subunits stoichiometry. EuPA Open Proteom 4:82–86 60. Meleth S, Deshane J, Kim H (2005) The case for well-conducted experiments to validate statistical protocols for 2D gels: different pre-processing = different lists of significant proteins. BMC Biotechnol 5:7 61. Troyanskaya O, Cantor M, Sherlock G, Brown P, Hastie T, Tibshirani R et al (2001) Missing value estimation methods for DNA microarrays. Bioinformatics 17(6):520–525 62. Wellmer F, Alves-Ferreira M, Dubois A, Riechmann L, Meyerowitz EM (2006) Genome-wide analysis of gene expression during early Arabidopsis flower development. PLoS Genet 2(7):e117 63. Eng JK, Mccormack AL, Yates JR (1994) An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. J Am Soc Mass Spectrom 5: 977–989

64. Perkins DN, Pappin DJ, Creasy DM, Cottrell JS (1999) Probablity-based protein identification by searching sequence databases using mass spectrometry data. Electrophoresis 20: 3551–3557 65. Colinge J, Masselot A, Giron M, Dessingy T, Magnin J (2003) OLAV: towards highthroughput tandem mass spectrometry data identification. Proteomics 3(8):1454–1463 66. Craig R, Beavis RC (2004) TANDEM: matching proteins with tandem mass spectra. Bioinformatics 20(9):1466–1467 67. Geer LY, Markey SP, Kowalak JA, Wagner L, Xu M, Maynard DM et al (2004) Open mass spectrometry search algorithm. J Proteome Res 3:958–964 68. Fu Y, Yang Q, Sun R, Li D, Zeng R, Ling CX et al (2004) Exploiting the kernel trick to correlate fragment ions for peptide identification via tandem mass spectrometry. Bioinformatics 20(12):1948–1954 69. Tanner S, Shu H, Frank A, Wang LC, Zandi E, Mumby M et al (2005) InsPecT: identification of posttranslationally modified peptides from tandem mass spectra. Anal Chem 77(14): 4626–4639 70. Bern M, Cai Y, Goldberg D (2007) Lookup peaks: a hybrid of de novo sequencing and database search for protein identification by tandem mass spectrometry. Anal Chem 79(4): 1393–1400 71. Eng JK, Jahan TA, Hoopmann MR (2013) Comet: an open-source MS/MS sequence database search tool. Proteomics 13(1):22–24 72. Kim S, Pevzner PA (2014) MS-GF+ makes progress towards a universal database search tool for proteomics. Nat Commun 5(1):5277 73. Tyanova S, Temu T, Cox J (2016) The MaxQuant computational platform for mass spectrometry-based shotgun proteomics. Nat Protoc 11(12):2301–2319 74. Zeng X, Ma B (2021) MSTracer: a machine learning software tool for peptide feature detection from liquid chromatography-mass spectrometry data. J Proteome Res 20(7): 3455–3462 75. Perez-Riverol Y, Bai J, Bandla C, Garcı´aSeisdedos D, Hewapathirana S, Kamatchinathan S et al (2022) The PRIDE database resources in 2022: a hub for mass spectrometry-based proteomics evidences. Nucleic Acids Res 50(D1):D543–D552

Chapter 24 Peptidomics Methods Applied to the Study of Flower Development Raquel A´lvarez-Urdiola, Eva Borra`s, Federico Valverde, Jose´ Toma´s Matus, Eduard Sabido´, and Jose´ Luis Riechmann Abstract Understanding the global and dynamic nature of plant developmental processes requires not only the study of the transcriptome, but also of the proteome, including its largely uncharacterized peptidome fraction. Recent advances in proteomics and high-throughput analyses of translating RNAs (ribosome profiling) have begun to address this issue, evidencing the existence of novel, uncharacterized, and possibly functional peptides. To validate the accumulation in tissues of sORF-encoded polypeptides (SEPs), the basic setup of proteomic analyses (i.e., LC-MS/MS) can be followed. However, the detection of peptides that are small (up to ~100 aa, 6–7 kDa) and novel (i.e., not annotated in reference databases) presents specific challenges that need to be addressed both experimentally and with computational biology resources. Several methods have been developed in recent years to isolate and identify peptides from plant tissues. In this chapter, we outline two different peptide extraction protocols and the subsequent peptide identification by mass spectrometry using the database search or the de novo identification methods. Key words Peptidome, Ultrafiltration, Ammonium sulphate, Reverse-phase chromatography, C-18, Arabidopsis, Mass spectrometry, Database

1

Introduction Although a variety of peptides have been well documented in both animal and plant genomes, until recently the coding potential of eukaryotic short open reading frames (sORFs) at the genome-wide level had mostly been overlooked. One of the reasons behind this gap is the computational and experimental difficulties for their identification and functional characterization, and particularly for determining whether these sequences are in fact translated. However, it has become clear over the past few years that small peptides (usually defined as shorter than 100 amino acids in length) constitute an important part, largely still uncharacterized, of the eukaryotic proteome [1–13]. Moreover, the massive and widespread

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_24, © Springer Science+Business Media, LLC, part of Springer Nature 2023

509

510

Raquel A´lvarez-Urdiola et al.

transcription of the eukaryotic genome and the pervasive translation of long noncoding RNAs (lncRNAs) habilitate sORFs and the resulting small peptides as raw materials for de novo gene origin and evolution [14–19]. In plants, several peptides have been functionally characterized as key players in diverse signalling pathways of plant development, including flower formation and maturation, in Arabidopsis and other plant species (i.e., [20–22]). Moreover, the presence of novel, uncharacterized Arabidopsis small peptides has been inferred from transcriptome data, in particular ribosome profiling (PolyRibo-Seq) experiments [23–25], leaving the door open for their identification through proteomics and peptidomics approaches. In fact, in studies with human cells and for selected SEPs identified from lncRNAs, primarily by Poly-Ribo-Seq, it was experimentally estimated that SEPs can be present in the cell at concentrations that are within the range of typical cellular proteins [26], that SEPs can exhibit different and specific subcellular localizations [27, 28], and that they can carry out important biological functions (e.g., [29– 33]). Furthermore, in addition to transcriptomics, computational tools have also been used as a source of knowledge on new potentially coding sORFs, in plants as well as in other eukaryotic organisms and bacteria (e.g., [34–36]). The sources of peptides that altogether would constitute the peptidome of a plant are several and include the following: (1) processing from larger functional or nonfunctional precursors; (2) additional short open reading frames (sORFs) in known protein-coding genes (up- or downstream the main ORF, in introns, as short splice variants or in a different reading frame from that of the main ORF); and (3) sORFs in long noncoding RNAs (lncRNAs), transcripts of unknown function (TUFs), intergenic regions, junctions, and microRNA precursors [37–40]. For instance, computational analyses suggested that several thousands of novel, potentially coding sORFs could exist in the intergenic regions of the Arabidopsis genome [35]. In fact, it was found that when overexpressed, some of those novel sORFs could induce developmental alterations in plant size, leaf number and shape, fertility, or cause lethality, raising the possibility that (many) sORFs with coding potential but that are still uncharacterized in plant genomes might be associated with morphogenesis [37] and other developmental and physiological processes. RNA-based methods are a very powerful tool to detect potentially translating sORFs, and the analysis of ribosome profiling data obtained from a variety of eukaryotic organisms provided strong support to the idea that lncRNAs are an important source of new peptides [41, 42]. Ribosome profiling has also been used to demonstrate extensive translation of open reading frames, including novel sORFs, in plant species such as Arabidopsis [23, 25, 43], maize [44] and tomato [45]. The evaluation of the coding

Flower Development Peptidomics

511

Fig. 1 Workflow for peptide discovery and characterization based on mass spectrometry. Extraction method and MS data analysis

potential of the sequences identified through ribosome profiling is mostly computational but there are mass spectrometry (MS)-based methods able to detect peptides that are translated from novel sORFs, thereby directly validating the protein-coding potential of the transcripts [27, 38, 44, 46–53]. In parallel, the improvement of mass spectrometry and data interpretation bioinformatic algorithms have facilitated the analysis of complex protein mixtures. However, the detection of novel plant peptides derived from small ORFs that are not annotated in reference databases presents specific challenges that need to be addressed, both experimentally and with computational resources (Fig. 1). The first requirement is an efficient and high-quality extraction from abundant starting material, for which several methods have been developed and optimized. Most basic protocols used for protein extraction from plant tissue are trichloroacetic acid (TCA)-acetone and phenol-based methods. The optimal composition of the extraction buffer depends on the species and tissue of interest [54, 55], but other aspects must be considered, such as heat treatment of the sample to diminish nonspecific protease digestions [38, 56, 57] or the addition of protease inhibitors to avoid protein degradation [38, 44, 46, 49, 58, 59] (Table 1). Besides, the processing and degradation of cellular proteins can generate peptidic fragments that increase the complexity of the peptidome sample, deteriorating the signal-to-noise ratio in the experiments. Therefore, strategies to separate larger proteins from peptides prior to LC-MS/MS analyses are crucial to improve the

Sample type

Col-0 leaves (four-leaf stage)

Leaf tissue and leaf protoplast

Root cultures, xylem sap

Inbred line B73 seeds

Species

A. thaliana

A. thaliana

Medicago truncatula

Z. mays

Phenol extraction and ammonium sulphate precipitation

o-chlorophenol/acetone precipitation Size exclusion chromatography

SDS-PAGE (10% gel) 10 kDa MWCO filter

Heat treatment of the sample to diminish nonspecific protease digestions Trichloroacetic acid (TCA) precipitation 10 kDa MWCO filter

Highlights of the extraction method characteristics 1860 novel SEPs

Identified peptides

1. Identification of sORFs based 2695 small peptides (up to 100aa) on Ribo-seq data 2. LC-MS/MS (DIA, MaxQuant)

12 peptide hormones Nano-LC-ESI-MS/MS (DDA, Proteome Discoverer, SEQUEST) matching spectra against three databases: (1) 225 sequences from known members of different peptide families in M. truncatula identified using BLAST; (2) M. truncatula whole protein database; (3) ~940 C-terminally encoded peptide (CEP) sequences from different species

1. Canonical database (Araport11) 127 protein-derived auxin-responsive 2. LC-MS/MS (DDA, ProteinLynx, ProteinPilot) peptides

1. Database: six-frame translation of the complete A. thaliana genome 2. LC-MS/MS (DDA, Mascot)

Identification workflow

Table 1 Overview of peptide extraction methods applied in LC-MS/MS studies in different plant species in recent years

[44]

[122]

[54]

[38]

References

512 Raquel A´lvarez-Urdiola et al.

Recovery of analytes (peptides) using Sep-Pak C18 Cartridges

Methanol based extraction.

Aerial tissue (~12-week-old plants) 30 kDa MWCO filter

Panax ginseng Ginseng radix (dry root)

Capsicum chinense x frutescens

Solanum Tomato leaves lycopersicum

14 AMPs

308 peptides 1. Nano-LC-MS/MS (DIA, PEAKS) with canonical database search (Swissprot) 2. De novo nano-LC-MS/MS (DIA, SPIDER) peptide identification

1. Database generated using antimicrobial peptides (AMPs) prediction algorithms + canonical C-chinense reference proteome 2. LC-MS/MS (DDA, Mascot)

46 unique peptides 1. Tomato hypothetical peptide derived from database (TomHT database) 25 pre-proteins and randomized databases (Ran Databases) 2. LC-MS/MS (DDA, Mascot)

(continued)

[90]

[59]

[123]

[38]

1. Database: rice genome (MSU7) 236 annotated and [47] Leaves of O. sativa L. ssp. japonica Anion-exchange 52 unannotated novel considering and considering cv. Nipponbare and suspension chromatography and three different possible PTMs small secreted proteins cells derived from Nipponbare acetone precipitation (SSPs) SDS-PAGE using (16.5% 2. LC-MS/MS (DDA, Mascot) calli gel)

1. Database: six-frame translation 1993 novel SEPs of the complete Z. mays genome 2. LC-MS/MS (DDA, Mascot)

Oryza sativa

Heat treatment of the sample to diminish non-specific protease digestions Trichloroacetic acid (TCA) precipitation 10 kDa MWCO filter

Maize inbred line B73 leaves (three-leaf stage)

Z. mays

Flower Development Peptidomics 513

Plant stem material

Eucalyptus grandis

Physcomitrella Protonemata patens

Sample type

Species

Table 1 (continued)

Gel filtration

Organic solvent precipitation SDS-PAGE

Highlights of the extraction method characteristics

Identified peptides

1. sORF database generated using 828 peptide sequences sORFfinder at genome and transcriptome level 2. LC-MS/MS (DDA, MaxQuant)

41 novel peptides 1. Database: three-frame translation (Virtual Ribosome v.2.0) of mRNA, ncRNA, and transcribed RNA sequences publicly available combined with E. grandis protein database 2. Database-guided LC-MS/MS (DDA, PEAKS) + novel peptide mapping and genomic classification (BLAST)

Identification workflow

[49]

[48]

References

514 Raquel A´lvarez-Urdiola et al.

Flower Development Peptidomics

515

identification and sequence coverage of low-abundance peptides [55, 60]. Peptides can be separated and purified using different methods such as electrophoresis gels [27, 47, 48, 61] or molecular weight cut-off (MWCO) filters [38, 52, 54, 58, 59, 62] (Table 1). Moreover, the optimal polypeptide size for detection by LC-MS/ MS is approximately 10–20 amino acids, suggesting that trypsin (or trypsin + Lys-C) cleavage is crucial for high-sensitivity SEP detection. Nevertheless, smaller peptides that may be amenable to protease cleavage should be detectable as well [26]. An additional difficulty lies in undersampling (i.e., identification of only a subset of the peptides) by conventional data acquisition methods [63]. According to a study to optimize a SEP discovery MS workflow using human samples [52], SEP detection is stochastic due to their size and expression characteristics. Therefore, to avoid undersampling and thus identify more SEPs, it is often more efficient to perform multiple technical and/or biological replicates (multiple runs on the MS platform) than, for example, introduce extensive fractionation methods before LC-MS/MS analyses (as in [26]). For peptide identification from tandem mass spectra, there are two approaches that could be used: database search and de novo sequencing. In database search, all potential peptide sequences included in a specified database are retrieved for each spectrum, and each peptide-spectrum match is scored via a scoring function calculated by database search engines (such as SEQUEST [64], Mascot [65], Phenyx [66], X! Tandem [67], OMSSA [68], pFind [69], InsPecT [70], ByOnic [71], Comet [72], MS-GF+ [73], MaxQuant [74], or MSTracer [75]). This guided approach is widely used for peptidomics and proteomics, and can be based on canonical (well-annotated) protein databases (e.g., UniProt) or customized databases containing putative SEPs identified by bioinformatic (e.g., sORFinder) [76] or transcriptomic analyses (i.e., RNA-sequencing or ribosome profiling). The annotation of the genome of the organism under study is the first source for preparing the database for MS database search (i.e.,, all proteins and peptides that are already known or identified from that genome). However, for the identification of novel SEPs in MS data, it is necessary to design more specific, expanded databases that should also include the potential novel coding sORFs. Current integrated peptidomics pipelines include different database creation strategies, from the use of ribosome profiling data to the six-frame or three-frame identification of sORFs at the genome or transcriptome level, respectively (Tables 2 and 3; see also the Notes section). For instance, a recent MS-based study identified over 1000 novel human proteins derived from alternative ORFs identified by RNA-seq (mostly corresponding to SEPs, 57aa median length) [27]. In plants, approximately 70,000 transcribed sORFs were detected in Physcomitrella patents (moss) using “sORF

516

Raquel A´lvarez-Urdiola et al.

Table 2 Repositories for SEP-database generation Database

Description

Collected data

Organism

References

ARA-PEP

Putative peptides encoded by sORFs in the A. thaliana genome

A. thaliana Tiling arrays, RNA-seq data, and other publicly available datasets

[39]

PsORF

sORFs across different plant species

Genomic, transcriptomic, ribo-seq, and MS data

35 plant species

[104]

PlantPepDB

Manually curated database of plant-derived peptides

Experimentally validated peptides, peptides with evidence at transcript level, based on computational predictions or inferred by homology

Several plant species including algae, bryophyte, angiosperms, and gymnosperms

[124]

RPFdb v2.0

Genome-wide information of translated mRNA

Ribo-seq samples

Plants: A. thaliana Others: 28 different species

[103]

CANTATAdb lncRNA data from lncRNA identified computationally using 2.0 plant and algae publicly available RNAseq data

39 plant species [125] (including three algae)

AlnC

Angiosperm lncRNA Catalogue

682 angiosperm plant species (809 tissues)

[126]

GWIPS-viz

Online Ribo-seq samples visualization tool for ribo-seq data

Plants: A.thaliana, Z. mays Others: bacteria, animals, etc.

[102]

uORFlight

Database for the uORF identified in evaluation of genome and uORF frequency transcriptome among different annotations accessions

[127] Plants: A. thaliana, O. sativa, B. napus, G. max, G. raimondii, M. truncatula, S. lycopersicum, S. tuberosum, T. aestivum, Z. mays Others: fungus, metazoan, and vertebrate

uORFdb

Comprehensive literature database on eukaryotic uORFs

Plants: A. thaliana Others: human, mouse, rat, virus, yeast, etc.

lncRNA in angiosperms (1KP transcriptome data)

uORF-related references; manually curated from all uORF-related literature listed at the PubMed database

[105]

(continued)

Flower Development Peptidomics

517

Table 2 (continued) Database

Description

Collected data

C-PAmP

Computationally predicted plant antimicrobial peptides

2112 plant species in Selection of peptides UniProtKB/Swissincluded in the Prot Antimicrobial Peptide Database (APD) and the Collection of AntiMicrobial Peptides (CAMP)

StraPep

Structure database Structural data collected of bioactive from UniProtKB and peptides PDB

452 different species including bacteria, yeast, animals, humans, and plants

[129]

DRAMP 3.0

Manually curated data repository of antimicrobial peptides

Variety of organisms, including bacteria, archaea, protists, fungi, animals, and plants

[130]

Peptides retrieved from Pubmed, Swiss-prot, and Lens

Organism

References [128]

Finder” [76], from which 828 distinct peptide sequences were identified by LC-MS/MS [49]. Customized peptide databases can also be derived from the six-frame translation of genomic sequences, an approach that has been successfully used in microorganisms [62, 77], and recently also in both monocot and dicot plants, where a total of 1993 and 1860 SEPs were identified in maize and Arabidopsis, respectively [38]. Altogether, these and other studies illustrate the existence of a substantial, uncharted fraction of the eukaryotic proteome that is mainly composed of small proteins (peptidome) (Table 1). In contrast to database search, for de novo peptide sequencing, peptide sequences are extracted directly from tandem mass spectra using specific algorithms such as PEAKS [78], SPIDER [79], UniNovo [80], pNovo+ [81], Novor [82], DeepNovo [83], or DeepNovo-DIA [84]. The de novo method is less powerful than database search, as many spectra cannot be unambiguously sequenced due to incomplete fragmentation. In addition, the de novo method is relatively slow when compared with the databasesearch engines, and the large search space of all possible amino acid sequences for each spectrum often leads to higher false discovery rates. Moreover, the complexity of tandem mass spectra can be significantly increased when posttranslational modifications (PTMs) are considered as well [85]. Some algorithms have been used for solving the de novo identification problems involving dynamic programming, integer linear programming, machine learning or other methods, and advances in mass spectrometry

518

Raquel A´lvarez-Urdiola et al.

Table 3 Tools for database design Tool

Description

Method

Example

SPADA

Small peptide alignment discovery application. Free software tool that identifies and predicts the gene structure for short peptides with one or two exons

Sequence similarity

[112] Creation of an M. truncatula small secreted peptide database (MtSSPdb) using SPADA and sORF Finder [131]

sORF Finder Program package for the identification of sORFs with high coding potential

Codon pattern, codon substitution and crossspecies conservation

[35, 76] 51 new sORFs identified using sORF finder and the ARA-PEP repository (LC-MS/MS results) [46]

PhyloCSF

Phylogenetic Codon Substitution Frequencies: method to determine whether a multispecies nucleotide sequence alignment is likely to represent a protein-coding region

Codon pattern, codon substitution and crossspecies conservation

Identification of small peptide-coding “long noncoding” RNAs in soybean [132]

[34]

MiPepid

RNA-seq sORF annotation Machine in mammalian species learning

Identification of 82 novel species-specific translated sORFs (LC-MS/MS) from lncRNA (database generated using MiPepid) [19]

[113]

lncPepid

RNA-seq sORF annotation Machine learning in plants. A discovery tool trained using maize and Arabidopsis data that considers sequence composition and physicochemical properties

CPPredsORF

Predicts the coding Machine potential of sORFs based learning on non-AUG initiation of translation

DeepCPP

Optimization of CPPred

Deep learning

References

[115]

sORF finder, miPepid, CPPred, and DeepCPP used as control groups [115]

[114]

sORF finder, miPepid, CPPred, and DeepCPP used as control groups [115]

[116]

(continued)

Flower Development Peptidomics

519

Table 3 (continued) Tool

Description

RiboTaper

Method

Example

References

Ribo-seq Statistical approach that identifies translated regions based on the characteristic threenucleotide periodicity of ribo-seq data

Identification of uORFs, dORFs, and altORFs in A. thaliana [23]

[106]

PRICE

Ribo-seq Computational method that models experimental noise to resolve overlapping sORFs and noncanonical translation initiation in an accurate manner

[107] Validation of the method using major histocompatibility complex class I (MHC I) peptidomics [107]

RiboCode

Unbiased method to recover the signal of active translation from the ribo-seq data

Ribo-seq

[108] Identification of 9388 sORF encoding peptides (2-100aa) in maize, from which 2695 SEPs were verified by MS data [44]

Ribo-seq

[109]

RiboPlotR

Ribo-seq Visualization package written in R. Representation of RNA-seq coverage and Ribo-seq reads in genomic coordinates for all annotated transcript isoforms of a gene

[110] RiboPlotR combines transcriptome annotation files, standard RNA-seq bam files, and Ribo-seq P-site position/count files to plot RNA-seq and Riboseq data with genomic coordinates for each isoform. Tested in Arabidopsis and tomato [110]

RiboNT

Noise-tolerant sORF predictor that can use RPFs with poor periodicity

Identification of sORFs in Arabidopsis seedlings that are evolutionary conserved in diverse plant species [111]

RiboStreamR Quality control platform for Ribo-seq data in the form of an R shine web application

Ribo-seq

[111]

instruments have improved de novo sequencing results [86]. However, further optimizations of algorithms, particularly with respect to data confidence, are still necessary to turn the technique into an actual alternative to commonly used database search peptide identification methods. Despite the difficulties, de novo identification has been successfully implemented for SEP detection in several

520

Raquel A´lvarez-Urdiola et al.

plant species [48, 59, 87–89]. When combined with classic database search strategies, de novo approaches can help to provide more comprehensive results [59, 90, 91] (Table 1). In fact, several research groups have developed software that directly combines both, database search and de novo sequencing, for peptide identification from mass spectra [71, 92]. Despite the advances in mass spectrometry and data interpretation, however, a problem still not fully addressed is the (high) number of unassigned spectra. New mass spectrometry sampling methods, such as data-independent acquisition (DIA [93]), together with the development of new machine learning tools to predict peptide fragmentation [94–97] promise a bright and very exciting future in the peptidome field, in which a significant amount of information will be confidently recovered from the acquired data. In this chapter, we provide two plant peptide extraction methods based on different extraction buffers and precipitation techniques, and describe an example of an LC-MS/MS pipeline, also introducing some suggestions for database design.

2 2.1

Materials General

1. Protein low-binding microcentrifuge tubes (1.5 or 2 mL). 2. Mortar and pestle. 3. Liquid nitrogen.

2.2

Ultrafiltration

1. Extraction buffer: 1× phosphate-buffered saline (PBS), 1.5 M urea, 10 mM dithiothreitol (DTT), 2% v/v acetonitrile (ACN), 0.5% v/v trifluoroacetic acid (TFA), 10 μM MG-132 proteasome inhibitor, 1 tablet of Proteinase Inhibitor cocktail cOmplete (Roche) per each 50 mL of buffer, and 1 mM phenylmethylsulfonyl fluoride (PMSF) (see Note 1). Prepare fresh for each experiment (see Note 2). 2. 10× phosphate buffer saline (PBS): 1.37 M NaCl, 0.027 M KCl, 80 mM Na2HPO4, and 20 mM KH2PO4 pH 7 (NaOH). Prepare 1 L and autoclave it. 3. Ultra-0.5 mL 30-K centrifugal filter devices (Amicon®) (see Note 3).

2.3 Ammonium Sulphate Precipitation

1. Extraction buffer: 1× PBS, 2 M urea, 2% v/v acetonitrile, 10 mM DTT, 5% v/v trifluoroethanol (TFE), 50 mM Tris– HCl pH 7.6, 10 μM MG-132, 1 tablet of Proteinase Inhibitor cocktail cOmplete (Roche) per each 50 mL of buffer, and 1 mM PMSF. Prepare fresh for each experiment (see Note 2). 2. Ammonium sulphate (salt, EM/HPLC grade).

Flower Development Peptidomics

2.4 Reverse-Phase Chromatography Peptide Extraction

521

1. C18 spin columns, containing 8 mg of resin each (Pierce, Thermo Scientific). 2. Activation solution: 50% v/v ACN in distilled water (400 μL per sample). 3. Equilibration solution: 0.5% v/v TFA in 5% v/v ACN (400 μL per sample). 4. Sample buffer: 2% v/v TFA in 20% v/v ACN (1 μL for every 3 μL of sample) (see Note 4). 5. Wash solution: 0.5% v/v TFA in 5% v/v ACN (400–800 μL per sample) (see Note 5). 6. Elution buffer: 0.1% v/v formic acid in 70% v/v ACN (42 μL per sample) (see Note 6). 7. Qubit protein assay kit. 8. Qubit fluorometer.

2.5

LC-MS/MS

1. DL-dithiothreitol (DTT) (see Note 7). 2. Iodoacetamide. 3. Urea. 4. Ammonium bicarbonate. 5. Lysyl endopeptidase. 6. Trypsin. 7. Formic acid. 8. MicroSpin C18 columns (The Nest Group, Inc). 9. Nano Trap C18 columns with an inner diameter of 100 μm packed with C18 particles of 5 μm particle size (Thermo Fisher Scientific) (optional, depending on the setup of each laboratory). 10. Reverse-phase chromatography 15–50 cm length) (see Note 8).

columns

(C18,

2

μm,

11. Buffer A: 0.1% v/v formic acid in water. 12. Buffer B: 0.1% v/v formic acid in acetonitrile. 13. Bovine serum albumin (New England Biolabs cat # P8108S). 14. Orbitrap Eclipse mass spectrometer (Thermo Fisher Scientific) (see Note 9). 15. EASY-nLC 1000 (Thermo Fisher Scientific).

3

Methods Below we provide two peptide extraction methods based on different extraction buffers and precipitation techniques (see Subheadings 3.1 and 3.2), both of which are to be followed by a reversephase chromatography (see Subheading 3.3) (Fig. 2) and describe an example of an LC-MS/MS pipeline (see Subheading 3.4).

522

Raquel A´lvarez-Urdiola et al.

Fig. 2 Schematic representation of the two peptide extraction methods described in this chapter. AS, activation solution; ES, equilibration solution; WS, wash solution; EB, elution buffer 3.1

Ultrafiltration

1. Collect tissue of interest with clean material (see Note 10) and freeze directly in liquid nitrogen. Keep at -80 °C until required. 2. Using a different mortar and pestle for each sample, grind the tissue with liquid nitrogen until obtaining a whitish fine powder (see Note 11). 3. Collect 0.5 g of blended tissue distributed in two 2 mL Eppendorf tubes (see Note 12). 4. Add a total of 1.2 mL of extraction buffer to 0.5 g of tissue, vortex immediately, and transfer to ice while preparing the rest of the samples (see Note 13). 5. Incubate the samples with continuous shaking for 1 h at 4 °C. 6. Spin the samples for 1 min at 4 °C in a microcentrifuge (max speed, ≥14,000 × g) to precipitate cellular debris and solid particles in suspension. Repeat as many times as necessary (see Note 14). 7. Insert each Amicon filter device in one of the provided microcentrifuge tubes. 8. Add up to 500 μL of the clean supernatant in the Amicon filter device and centrifuge at 14,000 × g for 10 min at 4 °C as indicated by manufacturer (see Notes 15 and 16). Repeat until all sample has passed through the same Amicon filter.

Flower Development Peptidomics

523

9. Keep the filtrate in the provided microcentrifuge tubes (flowthrough) (see Note 17). Keep the samples on ice to immediately continue with the reverse-phase chromatography or store them at -80 °C until use. 3.2 Ammonium Sulphate Precipitation

1. Collect tissue of interest with clean material (see Note 10) and freeze directly in liquid nitrogen. Keep at -80 ° C until required. 2. Using a different mortar and pestle for each sample, grind the tissue with liquid nitrogen until obtaining a whitish fine powder (see Note 11). 3. Collect 0.5 g of blended tissue distributed in two 2 mL Eppendorf tubes (see Note 12). 4. Add a total of 1.2 mL of extraction buffer to 0.5 g of tissue, vortex immediately, and transfer to ice while preparing the rest of the samples (see Note 13). 5. Incubate the samples shaking for 30 min at 4 °C. 6. Spin the samples for 1 min at 4 °C in a microcentrifuge (max speed, ≥14,000 × g) to precipitate cellular debris and solid particles in suspension (see Note 14). 7. Add 75% (w/v) of ammonium sulphate to the supernatant to precipitate the proteins in solution at 4 °C. The salt must be added little by little pipetting slowly each time until proteins precipitate (see Note 18). 8. Centrifuge at maximum speed (≥14,000 - g) for 25 min at 4 °C. 9. Place the supernatant in a new low-binding protein tube (smaller peptides will remain in the supernatant, whereas larger proteins precipitate). Keep the samples on ice to immediately continue with the reverse-phase chromatography or store them at -80 °C until use.

3.3 Reverse-Phase Chromatography Peptide Extraction

Prepare the reverse phase chromatography C18 columns as indicated by the manufacturer protocol. In brief: Sample preparation: 1. Mix 3:1 parts of sample:sample buffer. The final sample mix will contain approximately 0.5% TFA in 5% ACN (see Note 19). Column preparation: 2. Tap the column to settle the resin on the bottom of each column. Remove top and bottom caps (in that order). Place the column into a 2 mL receiver tube. 3. Add 200 μL of activation solution to wet the resin. Make sure to rinse the walls of the spin column (see Note 20).

524

Raquel A´lvarez-Urdiola et al.

4. Centrifuge at 1000 × g for 1 min. Discard the flow-through and repeat steps 3 and 4. 5. Add 200 μL of equilibration solution, centrifuge at 1500 × g for 1 min, and discard the flow-through. Repeat this step once. Sample binding: 6. Place the column into a receiver tube and load up to 150 μL of sample on top of resin bed (see Note 21). 7. Centrifuge at 1500 × g for 1 min. Repeat steps 6 and 7 as many times as needed to load all the sample in the same column (see Notes 22 and 23). Column wash: 8. Add 200 μL of wash solution to the column and centrifuge at 1500 × g for 1 min. Repeat this step once (see Note 5). Elution: 9. Place the column in a new protein low-binding receiver tube and add 21 μL of elution buffer to the top of the resin bed. 10. Centrifuge at 1500 × g for 1 min and repeat steps 9 and 10 with the same receiver tube. 11. Quantify the concentration and amount of total protein in each sample using a Qubit protein assay kit: Mix 199 μL of Qubit buffer with 1 μL of Qubit reagent for each sample. Add 2 μL of sample to 198 μL of the reaction mixture, vortex, and spin the tube. Incubate at room temperature for 15 min before measuring. 12. Store the samples at -80 °C until further analysis. 3.4

LC-MS/MS

3.4.1 Sample Preparation

1. Prepare or dissolve samples in 6 M urea, 200 mM ammonium bicarbonate. 2. Reduce the samples (10 μg of protein) with 30 nmols of dithiothreitol at 37 °C for 1 h. 3. Alkylate the samples (10 μg of protein) in the dark with 60 nmols of iodoacetamide at 25 °C for 30 min. 4. Dilute the sample extract to 2 M urea with 200 mM ammonium bicarbonate for digestion with endoproteinase LysC (1:10 w:v), and incubate at 37 °C overnight. 5. Dilute twofold with 200 mM ammonium bicarbonate for trypsin digestion (1:10 w:w), and incubate at 37 °C for 8 h. 6. After digestion, add formic acid (10% v/v of the final volume) to acidify the peptide mix. 7. Desalt the samples with MicroSpin C18 columns prior to LC-MS/MS analysis, following manufacturer’s instructions.

Flower Development Peptidomics 3.4.2 Chromatographic and Mass Spectrometric Analysis

525

1. Load the peptides onto the analytical column (C18, 2 μm, 15–50 cm length). 2. Separation of the peptides by reverse-phase chromatography with the corresponding columns. 3. Chromatographic gradients start at 93% buffer A and 7% buffer B with a flow rate of 250 nL/min for 5 min and gradually increase 65% buffer A and 35% buffer B in 60 min. 4. After each analysis, wash the column for 15 min with 10% buffer A and 90% buffer B. 5. Peptide eluates are dried in a vacuum centrifuge, and resuspended with buffer A at a final concentration of 1 μg/μL prior to analysis by LC-MS/MS. 6. Operate the mass spectrometer to acquire peptide spectra (see Note 24).

3.4.3 Data Analysis for Database-Search Peptide Identification

1. Search the acquired spectra against the desired peptide database (see Note 25), plus a list of common contaminants (suggested: [98]), and all the corresponding decoy entries. 2. Set the parameters accordingly to the experimental and mass spectrometric settings and, if appropriate, select variable posttranslational modifications to be detected (see Notes 26 and27). 3. Determine the peptide abundance estimation [99, 100]. 4. Add the information to the appropriate repositories (see Note 28).

4

Notes 1. Octyl-glucoside, a detergent, could be added (0.1% v/v) to the extraction buffer. The use of detergents is only necessary for the extraction and solubilization of hydrophobic peptides and proteins. However, the presence of detergents in peptide samples decreases chromatographic resolution in LC-MS/MS. Thus, they must be removed prior to MS analysis [101]. As a general rule for MS/MS experiments, keep laboratory wear and high-quality chemicals separated from the rest of the laboratory materials, always use gloves and, if possible, disposable plastic material of the highest quality. 2. Prepare a new extraction buffer on every extraction day as protease inhibitors could not work properly otherwise. MG-132 is available from several suppliers (we have routinely used MG-132 from Sigma-Aldrich). Proteinase Inhibitor cocktail cOmplete is from Roche. Different extraction buffers have been proposed in the recent years, and their final composition

526

Raquel A´lvarez-Urdiola et al.

needs to be selected considering the final objective of the study and the type of analytes of interest (e.g., phosphopeptides), because its formulation may affect the final state of the peptides and proteins in the samples (Table 1). 3. The Amicon® Ultra-0.5 product line includes five different cut-offs depending on its nominal molecular weight limit (NMWL); 30-K (30 kDa filter) devices are recommended, as peptides would normally be below the 30 kDa cut-off. 4. ACN can be substituted for methanol in all sample preparation buffers, depending on the desired composition of the final elution buffer. 5. The required washing volume will be dependent upon amount and type of contaminants present in the samples. Samples already containing large amounts of urea or >100 mM ammonium bicarbonate derived from the extraction buffer (Table 1) need to be washed one or two additional times. 6. The elution buffer used can be tailored to the downstream application. Acceptable buffers include 50–70% (v/v) ACN or methanol with or without 0.1% (v/v) TFA. For best results in LC-MS/MS analysis, TFA is replaced with 0.1% (v/v) formic acid. 7. Reagents for LC-MS/MS can be obtained from several suppliers. As an example, we list here the specific products we use: urea (GE Healthcare; Sigma-Aldrich, P/N 17-1319-01), ammonium bicarbonate (BioUltra, ≥99.5% (T); SigmaAldrich, P/N 09830), iodoacetamide (BioUltra; SigmaAldrich, P/N I1149), DL-dithiothreitol (for electrophoresis, ≥99%; Sigma-Aldrich, P/N D9163), formic acid for analysis EMSURE® (ACS Reag. Merck, P/N 1.00264.0100), sequencing grade modified trypsin (Promega, P/N V5111), and lysyl endopeptidase (Wako Chemicals GmbH, P/N 129-02541). 8. Suitable reverse-phase chromatography columns that we have used are, for instance, 25 cm columns with an inner diameter of 75 μm, packed with 1.9 μm C18 particles (Nikkyo Technos Co.); and 50 cm columns with an inner diameter of 75 μm, packed with 2 μm C18 particles (EASY-Column, Thermo Fisher Scientific, ES903). 9. This is just a concrete example of a “modern high-resolution mass spectrometer”; other instruments could be used. 10. To reduce sample contamination with human proteins (i.e., keratins and collagen) during sample collection, the use of nitrile gloves and laboratory coats is recommended. Take precaution to avoid hair contamination. If flower organs or tissues are going to be dissected, cool tweezers and any other sampling instrument with liquid nitrogen.

Flower Development Peptidomics

527

11. Keep samples (before and after grinding) always frozen by pouring liquid nitrogen in the mortar sporadically. Cool collection spatulas before using them to collect homogenized tissue. 12. The extraction yields are around 1 mg of total protein for each 0.5 g of tissue. Peptides might represent about 1% of the total protein, and therefore the expected yield for these extraction methods would be 10–15 μg of peptides. For Arabidopsis inflorescences, a volume of 1 mL of blended tissue in a 2 mL Eppendorf tube is equivalent to approximately 0.5 g of tissue. Dividing the sample in different tubes facilitates its dissolution in the extraction buffer, that is, using tubes with only 0.25 g (equivalent to 0.5 mL of volume) of blended tissue. After finishing the entire extraction protocol (including the reversephase chromatography with C18 columns), 10–30 μg of total peptides are obtained when using 30-K filters. The efficiency of the ammonium sulphate precipitation method may be lower (~6 μg of total peptides) (Fig. 2). 13. If total sample has been divided in two tubes, add approximately 0.6 mL of extraction buffer to each tube (with 0.25 g of blended tissue). 14. After each 1 min spin, transfer the supernatant to a new tube. Be careful to avoid both the pellet and remaining particles in suspension. Repeat the spin in a new tube as many times as needed until supernatant is clear (2 or 3 times should be enough). 15. When the sample has been divided in two tubes, the efficiency of using one single Amicon filter for all subsamples and the same collection tube is sufficient to achieve a suitable yield. 16. The required centrifugation time may vary according to the NMWL of the columns used. This protocol is defined for 30-K (or upper) devices, yet a higher centrifugation time is necessary for 10-K or 3-K devices (15 and 30 min, respectively). 17. The filtrate contains the smallest peptides depending on the weight limit of the filter device. However, if needed, it is possible to recover the concentrated solute by placing the filter device upside down in a clean microcentrifuge tube and centrifuging at 1000 × g for 2 min at 4 °C. For optimal recovery, it is important to perform the reverse spin immediately after filtrating. Besides, desalting, buffer exchange or diafiltration of this concentrated solute can be accomplished before eluting it by reconstituting the concentrate retained in the column to the original sample volume with the desired solvent and repeating the ultrafiltration process from the beginning to the concentrated solute elution.

528

Raquel A´lvarez-Urdiola et al.

18. Ammonium sulphate calculator from EnCor Biotechnology Inc. (http://www.encorbio.com/protocols/AM-SO4.htm) (selecting 4 °C temperature) can be used to calculate the needed amount of ammonium sulphate for each specific sample. The salt addition will increase the sample volume, which should be considered for the reverse-phase chromatography. For smallest peptides, ammonium sulphate could be added up to 80–85%. 19. The final exact concentrations of TFA and ACN will vary according to the extraction buffer, that is, ultrafiltration or ammonium sulphate precipitation protocol. In these examples, the concentration of the sample:sample buffer mix prior to reverse-phase chromatography would be 6.5% (v/v) of ACN for both extraction methods, 0.875% (v/v) TFA for ultrafiltration and 0.5% (v/v) TFA for ammonium sulphate precipitation. Nevertheless, these slight variations do not appear to result in significant differences in the efficiency of the reversephase chromatography process. 20. Add solutions carefully, especially in the activation step. Pour the solution through the walls of the column to avoid producing irregularities in the resin. 21. Each column can bind up to 30 μg of total peptide from 10 to 150 μL sample volumes. 22. In some cases, the extraction yield can be increased by recovering the flow-through and recentrifuging it after each step. 23. Flow-through may be retained to confirm sample binding. 24. 1–2 μg of peptides are loaded onto an analytical column (25 cm C18 2 μm particle size) using an autosampler device (e.g., EASY nLC 1000 and Thermo Fisher Scientific) and the peptides are then separated by reverse-phase chromatography using a water-acetonitrile chromatographic gradient. Modern high-resolution mass spectrometers are recommended for data acquisition (e.g., Orbitrap or qTOF). The mass spectrometer is operated in data-dependent acquisition (DDA) mode, in which a full MS scan is recorded in each cycle, followed by the fragmentation of the 10–30 most intense precursor ions to obtain the fragment ion spectra. 25. Obtained raw data are analyzed using a database search strategy. However, the results are susceptible to the characteristics of the reference database used for peptide identification. It is advisable to add the lists of putative SEPs to a database containing the canonical peptides and proteins of each organism (available in ENSEMBL, Uniprot, or other databases). The total number of sequences included in the database is also important, as an excessively large database (e.g., over 100,000 sequences) may lead to a higher false discovery rate in the

Flower Development Peptidomics

529

identifications. There are several different approaches for the identification of novel potential SEP sequences to be included in the reference database. One approach that has often been used is to make use of Ribo-seq data. There are multiple repositories that contain sORFs identified by Ribo-seq for plant species such as GWIPS-viz (Arabidopsis thaliana and Zea mays) [102], RPFdb v2.0 (A. thaliana) [103], PsORF (35 plant species) [104], and uORFdb (A. thaliana and others) [105] (Table 2); as well as several tools for the analysis of Ribo-seq data and sORF identification such as RiboTaper [106], PRICE [107], RiboCode [108], RiboStreamR [109], RiboPlotR [110], or RiboNT [111] (Table 3). An alternative (and complementary) approach for the identification of putative novel SEPs is to make use of the genome or lncRNA transcriptome sequences through sORFprediction tools such as SPADA [112], sORF Finder [35, 76], PhyloCSF (Phylogenetic Codon Substitution Frequencies) [34], MiPepid [113], CPPred [114], lncPepid [115], or DeepCPP [116] (Table 3). In addition, the putative peptide databases can also be derived from the six-frame translation of the corresponding genome sequence or from the three-frame translation of transcriptomic datasets (e.g., RNA-seq data and lncRNA), an approach referred to as peptidogenomics [117]. It is a strategy that has been successfully implemented in microorganisms [62, 77], and plants [38]. An additional consideration for the generation of the putative SEP database is whether the presence of translation initiation codons in the ORFs (the standard ATG or noncanonical codons such as CTG or ACG; see [118–120]) is a requirement or not, as both approaches have been used (e.g., [35, 38]). 26. Once the database has been constructed, the raw LC-MS/MS data needs to be interpreted using a database search engine (such as SEQUEST [64], Mascot [65], Phenyx [66], X! Tandem [67], OMSSA [68], pFind [69], InsPecT [70], ByOnic [71], Comet [72], MS-GF+ [73], MaxQuant [74], or MSTracer [75]). As example, the Mascot search engine (v2.6) can be used, using the search parameters accordingly to the experimental and mass spectrometry settings. For peptide identification a precursor ion mass tolerance below 10–20 ppm is recommended, whereas the fragment ion mass tolerance can go from 10 to 20 ppm for high-resolution mass analyzers (Orbitrap and TOF) to 0.5 Da if a linear ion trap is used for the analysis of the tandem mass spectra. Common peptide modifications such as oxidation of methionine and N-terminal protein acetylation are used as variable modifications. False discovery rate (FDR) in peptide identification is set to a maximum of 1%.

530

Raquel A´lvarez-Urdiola et al.

27. Validation of (selected) identified peptides is highly recommended due to the intrinsic limitations of FDR estimation when working with large databases, although this cannot be done yet in a high-throughput manner. Peptide identifications that pass the FDR threshold can be further validated with the purchase and full LC-MS/MS characterization of synthetic peptides with the same identified sequence (e.g., [38]), and/or by comparison with the fragmentation patterns and retention time predicted by the new machine learning algorithms (e.g., Prosit and MS2PIP) [94, 95, 97]. 28. Share data and results in a public repository. Data sharing in the public domain is the standard for omics research and a requirement for publication. For proteomics, the Proteomics IDEntifications (PRIDE) database (https://www.ebi.ac.uk/pride/) at the European Bioinformatics Institute (EMBL-EBI, Hinxton, Cambridge, UK) has enabled public data deposition of MS data since 2004, and its archival component has become the largest repository for proteomics data sharing worldwide [121]. The PRIDE database provides access to most of the experimental proteomics data described in MS-related scientific publications. Moreover, several repositories for sORFs and SEPs in plants have been developed with different purposes and using information from multiple in silico and experimental approaches (Table 2).

Acknowledgments Our work on peptidomics was supported by grant BFU201458289-P (funded by MICIN/AEI/ 10.13039/501100011033 and by “ERDF A way of making Europe”) and by grant 2017SGR718 (from the Agencia de Gestio´ d’Ajuts Universitaris I de Recerca) to JLR, and by institutional grant SEV-2015-0533 (funded by MCIN/AEI/10.13039/501100011033) and by the CERCA Programme/Generalitat de Catalunya. R.A. is supported by fellowship PRE2018-084278 funded by MCIN/AEI/ 10.13039/501100011033 and by “ESF Investing in your future.” The CRG/UPF Proteomics Unit is part of the Spanish Infrastructure for Omics Technologies (ICTS OmicsTech). We also acknowledge “Secretaria d’Universitats i Recerca del Departament d’Economia i Coneixement de la Generalitat de Catalunya” (2017SGR595) and support of the Spanish Ministry of Science and Innovation to the EMBL partnership, the Centro de Excelencia Severo Ochoa, and the CERCA Programme/Generalitat de Catalunya.

Flower Development Peptidomics

531

References 1. Tavormina P, De Coninck B, Nikonorova N, De Smet I, Cammuea BPA (2015) The plant peptidome: an expanding repertoire of structural features and biological functions. Plant Cell 27(8):2095–2118 2. Hsu PY, Benfey PN (2018) Small but mighty: functional peptides encoded by small ORFs in plants. Proteomics 18:1700038 3. Brunet MA, Leblanc S, Roucou X (2020) Reconsidering proteomic diversity with functional investigation of small ORFs and alternative ORFs. Exp Cell Res 393(1):112057 4. Brunet MA, Levesque SA, Hunting DJ, Cohen AA, Roucou X (2018) Recognition of the polycistronic nature of human genes is critical to understanding the genotypephenotype relationship. Genome Res 28(5): 609–624 5. Mudge JM, Ruiz-Orera J, Prensner JR, Brunet MA, Calvet F, Jungreis I et al (2022) Standardized annotation of translated open reading frames. Nat Biotechnol 40(7): 994–999 6. Lyapina I, Ivanov V, Fesenko I (2021) Peptidome: chaos or inevitability. Int J Mol Sci 22: 13128 7. Hellens RP, Brown CM, Chisnall MAW, Waterhouse PM, Macknight RC (2016) The emerging world of small ORFs. Trends Plant Sci 21(4):317–328 8. Takahashi F, Hanada K, Kondo T, Shinozaki K (2019) Hormone-like peptides and small coding genes in plant stress signaling and development. Curr Opin Plant Biol 51:88–95 9. Andrews SJ, Rothnagel JA (2014) Emerging evidence for functional peptides encoded by short open reading frames. Nat Rev Genet 15(3):193–204 10. Couso JP, Patraquim P (2017) Classification and function of small open reading frames. Nat Rev Mol Cell Biol 18(9):575–589 11. Plaza S, Menschaert G, Payre F (2017) In search of lost small peptides. Annu Rev Cell Dev Biol 33:391–416 12. Wright BW, Yi Z, Weissman JS, Chen J (2022) The dark proteome: translation from noncanonical open reading frames. Trends Cell Biol 32(3):243–258 13. Orr MW, Mao Y, Storz G, Qian SB (2021) Alternative ORFs and small ORFs: shedding light on the dark proteome. Nucleic Acids Res 48(3):1029–1042 14. Ruiz-Orera J, Hernandez-Rodriguez J, Chiva C, Sabido´ E, Kondova I, Bontrop R et al (2015) Origins of de novo genes in

human and chimpanzee. PLoS Genet 11(12):e1005721 15. Ruiz-Orera J, Verdaguer-Grau P, Villanueva˜as JL, Messeguer X, Alba` MM (2018) Can Translation of neutrally evolving peptides provides a basis for de novo gene evolution. Nat Ecol Evol 2(5):890–896 16. Ruiz-Orera J, Alba` MM (2019) Translation of small open reading frames: roles in regulation and evolutionary innovation. Trends Genet 35(3):186–198 ˜ as JL, Alba` MM 17. Ruiz-Orera J, Villanueva-Can (2020) Evolution of new proteins from translated sORFs in long non-coding RNAs. Exp Cell Res 391(1):111940 18. Blevins WR, Ruiz-Orera J, Messeguer X, ˜as JL, EspiBlasco-Moreno B, Villanueva-Can nar L et al (2021) Uncovering de novo gene birth in yeast using deep transcriptomics. Nat Commun 12(1):604 19. Fesenko I, Shabalina SA, Mamaeva A, Knyazev A, Glushkevich A, Lyapina I et al (2021) A vast pool of lineage-specific microproteins encoded by long non-coding RNAs in plants. Nucleic Acids Res 49(18): 10328–10346 20. Goto H, Okuda S, Mizukami A, Mori H, Sasaki N, Kurihara D et al (2011) Chemical visualization of an attractant peptide, LURE. Plant Cell Physiol 52(1):49–58 21. Santiago J, Brandt B, Wildhagen M, Hohmann U, Hothorn LA, Butenko MA et al (2016) Mechanistic insight into a peptide hormone signaling complex mediating floral organ abscission. eLife 5:e15075 22. Covey PA, Subbaiah CC, Parsons RL, Pearce G, Lay FT, Anderson MA et al (2019) A pollen-specific RALF from tomato that regulates pollen tube elongation. Plant Physiol 153:703–715 23. Hsu PY, Calviello L, Wu HYL, Li FW, Rothfels CJ, Ohler U et al (2016) Super-resolution ribosome profiling reveals unannotated translation events in Arabidopsis. Proc Natl Acad Sci U S A 113(45):E7126–E7135 24. Juntawong P, Girke T, Bazin J, Bailey-Serres J (2014) Translational dynamics revealed by genome-wide profiling of ribosome footprints in Arabidopsis. Proc Natl Acad Sci U S A 111(1):E203–E212 25. Bazin J, Baerenfaller K, Gosai SJ, Gregory BD, Crespi M, Bailey-Serres J (2017) Global analysis of ribosome-associated noncoding RNAs unveils new modes of translational

532

Raquel A´lvarez-Urdiola et al.

regulation. Proc Natl Acad Sci U S A 114(46): E10018–E10027 26. Slavoff SA, Mitchell AJ, Schwaid AG, Cabili MN, Ma J, Levin JZ et al (2013) Peptidomic discovery of short open reading frameencoded peptides in human cells. Nat Chem Biol 9(1):59–64 27. Vanderperre B, Lucier JF, Bissonnette C, Motard J, Tremblay G, Vanderperre S et al (2013) Direct detection of alternative open reading frames translation products in human significantly expands the proteome. PLoS One 8(8):e70698 28. Aspden JL, Eyre-Walker YC, Phillips RJ, Amin U, Mumtaz MAS, Brocard M et al (2014) Extensive translation of small open reading frames revealed by poly-ribo-seq. eLife 3:e03528 29. Huang JZ, Chen M, Chen D, Gao XC, Zhu S, Huang H et al (2017) A peptide encoded by a putative lncRNA HOXB-AS3 suppresses colon cancer growth. Mol Cell 68(1): 171–184 30. Nelson BR, Makarewich CA, Anderson DM, Winders BR, Troupes CD, Wu F et al (2016) A peptide encoded by a transcript annotated as long noncoding RNA enhances SERCA activity in muscle. Science 351:271–275 31. Makarewich CA, Baskin KK, Munir AZ, Bezprozvannaya S, Sharma G, Khemtong C et al (2018) MOXI is a mitochondrial micropeptide that enhances fatty acid β-oxidation. Cell Rep 23(13):3701–3709 32. Prensner JR, Enache OM, Luria V, Krug K, Clauser KR, Dempster JM et al (2021) Noncanonical open reading frames encode functional proteins essential for cancer cell survival. Nat Biotechnol 39(6):697–704 33. Boix O, Martinez M, Vidal S, Gime´nezAlejandre M, Palenzuela L, Lorenzo-Sanz L et al (2022) pTINCR microprotein promotes epithelial differentiation and suppresses tumor growth through CDC42 SUMOylation and activation. Nat Commun 13(1): 6840 34. Lin MF, Jungreis I, Kellis M (2011) PhyloCSF: a comparative genomics method to distinguish protein coding and non-coding regions. Bioinformatics 27(13):i275–i282 35. Hanada K, Zhang X, Borevitz JO, Li WH, Shiu SH (2007) A large number of novel coding small open reading frames in the intergenic regions of the Arabidopsis thaliana genome are transcribed and/or under purifying selection. Genome Res 17(5):632–640 36. Miravet-Verde S, Ferrar T, Espadas-Garcı´a G, Mazzolini R, Gharrab A, Sabido E et al

(2019) Unraveling the hidden universe of small proteins in bacterial genomes. Mol Syst Biol 15(2):e8290 37. Hanada K, Higuchi-Takeuchi M, Okamoto M, Yoshizumi T, Shimizu M, Nakaminami K et al (2013) Small open reading frames associated with morphogenesis are hidden in plant genomes. Proc Natl Acad Sci U S A 110(6):2395–2400 38. Wang S, Tian L, Liu H, Li X, Zhang J, Chen X et al (2020) Large-scale discovery of non-conventional peptides in maize and Arabidopsis through an integrated peptidogenomic pipeline. Mol Plant 13(7):1078–1093 39. Hazarika RR, De Coninck B, Yamamoto LR, Martin LR, Cammue BPA, Van Noort V (2017) ARA-PEPs: a repository of putative SORF-encoded peptides in Arabidopsis thaliana. BMC Bioinformatics 18(1):37 40. Couzigou J-M, Lauressergues D, Be´card G, Combier J-P, Ecard GB (2015) miRNAencoded peptides (miPEPs): a new tool to analyze the roles of miRNAs in plant biology. RNA Biol 12:1178–1180 41. Ruiz-Orera J, Messeguer X, Subirana JA, Alba MM (2014) Long non-coding RNAs as a source of new peptides. eLife 3:e03523 42. Hartford CCR, Lal A (2020) When long noncoding becomes protein coding. Mol Cell Biol 40(6):e00528–e00519 43. Kurihara Y, Makita Y, Shimohira H, Fujita T, Iwasaki S, Matsui M (2020) Translational landscape of protein-coding and non-proteincoding RNAs upon light exposure in Arabidopsis. Plant Cell Physiol 61(3):536–545 44. Liang Y, Zhu W, Chen S, Qian J, Li L (2021) Genome-wide identification and characterization of small peptides in maize. Front Plant Sci 12:695439 45. Wu HYL, Song G, Walley JW, Hsu PY (2019) The tomato translational landscape revealed by transcriptome assembly and ribosome profiling. Plant Physiol 181(1):367–380 46. Mergner J, Frejno M, List M, Papacek M, Chen X, Chaudhary A et al (2020) Massspectrometry-based draft of the Arabidopsis proteome. Nature 579:409–414 47. Wang P, Yao S, Kosami K-I, Guo T, Li J, Zhang Y et al (2020) Identification of endogenous small peptides involved in rice immunity through transcriptomics-and proteomicsbased screening. Plant Biotechnol J 18:415– 428 48. Jorge GL, Balbuena TS (2021) Identification of novel protein-coding sequences in Eucalyptus grandis plants by high-resolution mass

Flower Development Peptidomics spectrometry. Biochim Biophys Acta Proteins Proteom 1869:140594 49. Fesenko I, Kirov I, Kniazev A, Khazigaleeva R, Lazarev V, Kharlampieva D et al (2019) Distinct types of short open reading frames are translated in plant cells. Genome Res 29(9):1464–1477 50. Ouspenskaia T, Law T, Clauser KR, Klaeger S, Sarkizova S, Aguet F et al (2021) Unannotated proteins expand the MHC-I-restricted immunopeptidome in cancer. Nat Biotechnol 40:209–217 ˜ ez JK, 51. Chen J, Brunner AD, Cogan JZ, Nun Fields AP, Adamson B et al (2020) Pervasive functional translation of noncanonical human open reading frames. Science 367:140–146 52. Ma J, Ward CC, Jungreis I, Slavoff SA, Schwaid AG, Neveu J et al (2014) Discovery of human sORF-encoded polypeptides (SEPs) in cell lines and tissue. J Proteome Res 13(3):1757–1765 53. Flower CT, Chen L, Jung HJ, Raghuram V, Knepper MA, Yang CR (2020) Genetic and genomics investigation of structure and function of the kidney: an integrative proteogenomics approach reveals peptides encoded by annotated lincRNA in the mouse kidney inner medulla. Physiol Genomics 52(10):485 54. Luo W, Xiao Y, Liang Q, Su Y, Xiao L (2019) Identification of potential auxin-responsive small signaling peptides through a peptidomics approach in arabidopsis thaliana. Molecules 24:3146 55. Barashkova AS, Rogozhin EA (2020) Isolation of antimicrobial peptides from different plant sources: does a general extraction method exist? Plant Methods 16:143 56. Damerval C, De Vienne D, Zivy M, Thiellement H (1986) Technical improvements in two-dimensional electrophoresis increase the level of genetic variation detected in wheatseedling proteins. Electrophoresis 7(1):52–54 57. Chatterjee M, Gupta S, Bhar A, Das S (2012) Optimization of an efficient protein extraction protocol compatible with two-dimensional electrophoresis and mass spectrometry from recalcitrant phenolic rich roots of chickpea (Cicer arietinum L.). Int J Proteomics 2012: 536963 58. Shi Y, Li J, Li L, Lin G, Bilal AM, Smagghe G et al (2021) Genomics, transcriptomics, and peptidomics of Spodoptera frugiperda (Lepidoptera, Noctuidae) neuropeptides. Arch Insect Biochem Physiol 106:e21740 59. Culver KD, Allen JL, Shaw LN, Hicks LM (2021) Too hot to handle: antibacterial

533

peptides identified in ghost pepper. J Nat Prod 84:2200–2208 60. Kuljanin M, Dieters-Castator DZ, Hess DA, Postovit L-M, Lajoie GA (2017) Comparison of sample preparation techniques for largescale proteomics. Proteomics 17(1–2): 1600337 61. Flower CT, Chen L, Jung HJ, Raghuram V, Knepper MA, Yang C-R (2020) An integrative proteogenomics approach reveals peptides encoded by annotated lincRNA in the mouse kidney inner medulla. Physiol Genomics 52:485–491 62. Cao S, Liu X, Huang Y, Yan Y, Zhou C, Shao C et al (2021) Proteogenomic discovery of sORF-encoded peptides associated with bacterial virulence in Yersinia pestis. Commun Biol 4:1248 63. Grossmann J, Roschitzki B, Panse C, Fortes C, Barkow-Oesterreicher S, Rutishauser D et al (2010) Implementation and evaluation of relative and absolute quantification in shotgun proteomics with label-free methods. J Proteome 73(9):1740–1746 64. Eng JK, Mccormack AL, Yates JR (1994) An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. J Am Soc Mass Spectrom 5:977–989 65. Perkins DN, Pappin DJ, Creasy DM, Cottrell JS (1999) Probablity-based protein identification by searching sequence databases using mass spectrometry data. Electrophoresis 20: 3551–3557 66. Colinge J, Masselot A, Giron M, Dessingy T, Magnin J (2003) OLAV: towards highthroughput tandem mass spectrometry data identification. Proteomics 3(8):1454–1463 67. Craig R, Beavis RC (2004) TANDEM: matching proteins with tandem mass spectra. Bioinformatics 20(9):1466–1467 68. Geer LY, Markey SP, Kowalak JA, Wagner L, Xu M, Maynard DM et al (2004) Open mass spectrometry search algorithm. J Proteome Res 3:958–964 69. Fu Y, Yang Q, Sun R, Li D, Zeng R, Ling CX et al (2004) Exploiting the kernel trick to correlate fragment ions for peptide identification via tandem mass spectrometry. Bioinformatics 20(12):1948–1954 70. Tanner S, Shu H, Frank A, Wang LC, Zandi E, Mumby M et al (2005) InsPecT: identification of posttranslationally modified peptides from tandem mass spectra. Anal Chem 77(14):4626–4639

534

Raquel A´lvarez-Urdiola et al.

71. Bern M, Cai Y, Goldberg D (2007) Lookup peaks: a hybrid of de novo sequencing and database search for protein identification by tandem mass spectrometry. Anal Chem 79(4):1393–1400 72. Eng JK, Jahan TA, Hoopmann MR (2013) Comet: an open-source MS/MS sequence database search tool. Proteomics 13(1):22–24 73. Kim S, Pevzner PA (2014) MS-GF+ makes progress towards a universal database search tool for proteomics. Nat Commun 5(1):5277 74. Tyanova S, Temu T, Cox J (2016) The MaxQuant computational platform for mass spectrometry-based shotgun proteomics. Nat Protoc 11(12):2301–2319 75. Zeng X, Ma B (2021) MSTracer: a machine learning software tool for peptide feature detection from liquid chromatography-mass spectrometry data. J Proteome Res 20(7): 3455–3462 76. Hanada K, Akiyama K, Sakurai T, Toyoda T, Shinozaki K, Shiu S-H (2010) sORF finder: a program package to identify small open reading frames with high coding potential. Bioinformatics 26(3):399–400 77. Yang X, Jensen SI, Wulff T, Harrison SJ, Long KS (2016) Identification and validation of novel small proteins in Pseudomonas putida. Environ Microbiol Rep 8(6):966–674 78. Ma B, Zhang K, Hendrie C, Liang C, Li M, Doherty-Kirby A et al (2003) PEAKS: powerful software for peptide de novo sequencing by tandem mass spectrometry. Rapid Commun Mass Spectrom 17:2337–2342 79. Han Y, Ma B, Zhang K (2005) Spider: software for protein identification from sequence tags with de novo sequencing error. J Bioinforma Comput Biol 3(3):697–716 80. Jeong K, Kim S, Pevzner PA (2013) UniNovo: a universal tool for de novo peptide sequencing. Bioinformatics 29(16): 1953–1962 81. Chi H, Chen H, He K, Wu L, Yang B, Sun R-X et al (2013) pNovo+: de novo peptide sequencing using complementary HCD and ETD tandem mass spectra. J Proteome Res 12:615–625 82. Ma B (2015) Novor: real-time peptide de novo sequencing software. J Am Soc Mass Spectrom 26:1885–1894 83. Tran NH, Zhang X, Xin L, Shan B, Li M (2017) De novo peptide sequencing by deep learning. Proc Natl Acad Sci U S A 114(31): 8247–8252 84. Tran NH, Qiao R, Xin L, Chen X, Liu C, Zhang X et al (2019) Deep learning enables de novo peptide sequencing from data-

independent-acquisition mass spectrometry. Nat Methods 16(1):63–66 85. Pathan M, Samuel M, Keerthikumar S, Mathivanan S (2017) Unassigned MS/MS spectra: who am I? In: Keerthikumar S, Mathivanan S (eds) Proteome bioinformatics. Methods in molecular biology, vol 1549. Humana Press, New York, pp 67–74 86. Muth T, Renard BY (2018) Evaluating de novo sequencing in proteomics: already an accurate alternative to database-driven peptide identification? Brief Bioinform 19(5): 954–970 87. Wu H, Johnson MC, Lu CH, Fritsche KL, Thomas AL, Lai Y et al (2015) Peptidomics study of anthocyanin-rich juice of elderberry. Talanta 131:640–644 88. Gemperline E, Keller C, Jayaraman D, Maeda J, Sussman MR, Ane´ J-MA et al (2016) Examination of endogenous peptides in Medicago truncatula using mass spectrometry imaging. J Proteome Res 15:4403–4411 89. Gemperline E, Keller C, Li L (2016) Mass spectrometry in plant-omics. Anal Chem 88(7):3422–3434 90. Ye X, Zhao N, Yu X, Han X, Gao H, Zhang X (2016) Extensive characterization of peptides from Panax ginseng C. A. Meyer using mass spectrometric approach. Proteomics 16: 2788–2791 91. Zhang K, Mckinlay C, Hocart CH, Djordjevic MA (2006) The Medicago truncatula small protein proteome and peptidome. J Proteome Res 12:3355–3367 92. Wang X, Li Y, Wu Z, Wang H, Tan H, Peng J (2014) JUMP: a tag-based database search tool for peptide identification with high sensitivity and accuracy. Mol Cell Proteomics 13(12):3663–3673 93. Ro¨st HL, Rosenberger G, Navarro P, Gillet L, Miladinovic´ SM, Schubert OT et al (2014) OpenSWATH enables automated, targeted analysis of data-independent acquisition MS data. Nat Biotechnol 32:219–223 94. Wilhelm M, Zolg DP, Graber M, Gessulat S, Schmidt T, Schnatbaum K et al (2021) Deep learning boosts sensitivity of mass spectrometry-based immunopeptidomics. Nat Commun 12:3346 95. Gessulat S, Schmidt T, Zolg DP, Samaras P, Schnatbaum K, Zerweck J et al (2019) Prosit: proteome-wide prediction of peptide tandem mass spectra by deep learning. Nat Methods 16:509–518 96. Ekvall M, Truong P, Gabriel W, Wilhelm M, K€all L (2022) Prosit transformer: a transformer for prediction of MS2 spectrum intensities. J Proteome Res 21(5):1359–1364

Flower Development Peptidomics 97. Gabriels R, Martens L, Degroeve S (2019) Updated MS2PIP web server delivers fast and accurate MS2 peak intensity prediction for multiple fragmentation methods, instruments and labeling techniques. Nucleic Acids Res 47(W1):W295–W299 98. Beer LA, Liu P, Ky B, Barnhart KT, Speicher DW (2017) Efficient quantitative comparisons of plasma proteomes using label-free analysis with MaxQuant. Methods Mol Biol 1619:339–352 99. Gerster S, Kwon T, Ludwig C, Matondo M, Vogel C, Marcotte EM et al (2014) Statistical approach to protein quantification. Mol Cell Proteomics 13(2):666–677 100. Fabre B, Lambour T, Bouyssie´ D, Menneteau T, Monsarrat B, Burlet-Schiltz O et al (2014) Comparison of label-free quantification methods for the determination of protein complexes subunits stoichiometry. EuPA Open Proteom 4:82–86 101. Yeung YG, Stanley ER (2010) Rapid detergent removal from peptide samples with ethyl acetate for mass spectrometry analysis. Curr Protoc Protein Sci 16(16):12 102. Michel AM, Fox G, Kiran A M, De Bo C, O’Connor PBF, Heaphy SM et al (2014) GWIPS-viz: development of a ribo-seq genome browser. Nucleic Acids Res 42: D859–D864 103. Wang H, Yang L, Wang Y, Chen L, Li H, Xie Z (2019) RPFdb v2.0: an updated database for genome-wide information of translated mRNA generated from ribosome profiling. Nucleic Acids Res 47:D230–D234 104. Chen Y, Li D, Fan W, Zheng X, Zhou Y, Ye H et al (2020) PsORF: a database of small ORFs in plants. Plant Biotechnol J 18:2158–2160 105. Wethmar K, Barbosa-Silva A, AndradeNavarro MA, Leutz A (2014) uORFdb--a comprehensive literature database on eukaryotic uORF biology. Nucleic Acids Res 42: D60–D67 106. Calviello L, Mukherjee N, Wyler E, Zauber H, Hirsekorn A, Selbach M et al (2016) Detecting actively translated open reading frames in ribosome profiling data. Nat Methods 13(2):165–170 107. Erhard F, Halenius A, Zimmermann C, L’Hernault A, Kowalewski DJ, Weekes MP et al (2018) Improved Ribo-seq enables identification of cryptic translation events. Nat Methods 15(5):363–366 108. Xiao Z, Huang R, Xing X, Chen Y, Deng H, Yang X (2018) De novo annotation and characterization of the translatome with ribosome profiling data. Nucleic Acids Res 46(10):e61

535

109. Perkins P, Mazzoni-Putman S, Stepanova A, Alonso J, Heber S (2019) RiboStreamR: a web application for quality control, analysis, and visualization of Ribo-seq data. BMC Genomics 20:422 110. Larry Wu H-Y, Yingshan Hsu P (2021) RiboPlotR: a visualization tool for periodic Riboseq reads. Plant Methods 17:124 111. Song B, Jiang M, Gao L (2021) RiboNT: a noise-tolerant predictor of open reading frames from ribosome-protected footprints. Life (Basel) 11(7):701 112. Zhou P, Silverstein KAT, Gao L, Walton JD, Nallu S, Guhlin J et al (2013) Detecting small plant peptides using SPADA (small peptide alignment discovery application). BMC Bioinformatics 14(1):335 113. Zhu M, Gribskov M (2019) MiPepid: MicroPeptide identification tool using machine learning. BMC Bioinformatics 20(1):559 114. Tong X, Hong X, Xie J, Liu S (2020) CPPredsORF: coding potential prediction of sORF based on non-AUG. bioRxiv. https://doi. org/10.1101/2020.03.31.017525 115. Zhao S, Meng J, Luan Y (2022) LncRNAencoded short peptides identification using feature subset recombination and ensemble learning. Interdiscip Sci 14(1):101–112 116. Zhang Y, Jia C, Fullwood MJ, Kwoh CK (2021) DeepCPP: a deep neural network based on nucleotide bias information and minimum distribution similarity feature selection for RNA coding potential prediction. Brief Bioinform 22(2):2073–2084 117. Kersten RD, Yang Y, Xu Y, Cimermancic P, Nam S-J, Fenical W et al (2011) A mass spectrometry-guided genome mining approach for natural product peptidogenomics. Nat Chem Biol 7(11):794–802 118. Cao X, Slavoff SA (2020) Non-AUG start codons: expanding and regulating the small and alternative ORFeome. Exp Cell Res 391(1):111973 119. Na CH, Barbhuiya MA, Kim MS, Verbruggen S, Eacker SM, Pletnikova O et al (2018) Discovery of noncanonical translation initiation sites through mass spectrometric analysis of protein N termini. Genome Res 28(1):25–36 120. Li YR, Liu MJ (2020) Prevalence of alternative AUG and non-AUG translation initiators and their regulatory effects across plants. Genome Res 30(10):1418–1433 121. Perez-Riverol Y, Bai J, Bandla C, Garcı´aSeisdedos D, Hewapathirana S, Kamatchinathan S et al (2022) The PRIDE database resources in 2022: a hub for mass

536

Raquel A´lvarez-Urdiola et al.

spectrometry-based proteomics evidences. Nucleic Acids Res 50(D1):D543–D552 122. Patel N, Mohd-Radzman NA, Corcilius L, Crossett B, Connolly A, Cordwell SJ et al (2018) Diverse peptide hormones affecting root growth identified in the Medicago truncatula secreted peptidome. Mol Cell Proteomics 17(1):160–174 123. Chen YL, Lee CY, Cheng KT, Chang WH, Huang RN, Nam HG et al (2014) Quantitative peptidomics study reveals that a woundinduced peptide from PR-1 regulates immune signaling in tomato. Plant Cell 26(10): 4135–4148 124. Das D, Jaiswal M, Khan FN, Ahamad S, Kumar S (2020) PlantPepDB: a manually curated plant peptide database. Sci Rep 10(1):2194 MW, Bryzghalov O, 125. Szczes´niak Ciomborowska-Basheer J, Makałowska I (2019) CANTATAdb 2.0: expanding the collection of plant long noncoding RNAs. In: Chekanova JA, Wang HLV (eds) Plant long non-coding RNAs, Methods in molecular biology, vol 1933. Humana Press, New York, pp 415–429 126. Singh A, Vivek AT, Kumar S (2021) AlnC: an extensive database of long non-coding RNAs in angiosperms. PLoS One 16(4):e0247215

127. Niu R, Zhou Y, Zhang Y, Mou R, Tang Z, Wang Z et al (2020) uORFlight: a vehicle toward uORF-mediated translational regulation mechanisms in eukaryotes. Database 2020:baaa007 128. Niarchou A, Alexandridou A, Athanasiadis E, Spyrou G (2013) C-PAmP: large scale analysis and database construction containing high scoring computationally predicted antimicrobial peptides for all the available plant species. PLoS One 8(11):e79728 129. Wang J, Yin T, Xiao X, He D, Xue Z, Jiang X et al (2018) StraPep: a structure database of bioactive peptides. Database 2018:bay038 130. Shi G, Kang X, Dong F, Liu Y, Zhu N, Hu Y et al (2022) DRAMP 3.0: an enhanced comprehensive data repository of antimicrobial peptides. Nucleic Acids Res 50(D1):D488– D496 131. Boschiero C, Dai X, Lundquist PK, Roy S, de Bang TC, Zhang S et al (2020) MtSSPdb: the Medicago truncatula small secreted peptide database. Plant Physiol 183(1):399–413 132. Lin X, Lin W, Ku YS, Wong FL, Li MW, Lam HM et al (2020) Analysis of soybean long non-coding RNAs reveals a subset of small peptide-coding transcripts. Plant Physiol 182(3):1359–1374

Chapter 25 Quantifying Gene Expression Domains in Plant Shoot Apical Meristems Pau Formosa-Jordan and Benoit Landrein Abstract The shoot apical meristem is the plant tissue that produces the plant aerial organs such as flowers and leaves. To better understand how the shoot apical meristem develops and adapts to the environment, imaging developing shoot meristems expressing fluorescence reporters through laser confocal microscopy is becoming increasingly important. Yet, there are not many computational pipelines enabling a systematic and highthroughput characterization of the produced microscopy images. This chapter provides a simple method to analyze 3D images obtained through laser scanning microscopy and quantitatively characterize radially or axially symmetric 3D fluorescence domains expressed in a tissue or organ by a reporter. Then, it presents different computational pipelines aiming at performing high-throughput quantitative image analysis of gene expression in plant inflorescence and floral meristems. This methodology has notably enabled the quantitative characterization of how stem cells respond to environmental perturbations in the Arabidopsis thaliana inflorescence meristem and will open new avenues in the use of quantitative analysis of gene expression in shoot apical meristems. Overall, the presented methodology provides a simple framework to analyze quantitatively gene expression domains from 3D confocal images at the tissue and organ level, which can be applied to shoot meristems and other organs and tissues. Key words Quantitative image analysis, Confocal microscopy, Expression domain, Floral meristem, Inflorescence meristem, Stem cells, Arabidopsis thaliana

1

Introduction Plant morphogenesis mainly relies on the maintenance of postembryonic stem cell activities located in specific regions referred as meristems. The maintenance of the pool of stem cells in the Arabidopsis thaliana shoot apical meristem (SAM) largely depends on a feedback loop between WUSCHEL (WUS), a mobile homeodomain transcription factor promoting stem cell proliferation, and CLAVATA3 (CLV3), a diffusing peptide whose expression is induced by WUS but which restricts its expression through binding to specific receptors [1–5]. The dynamic expression of WUS and CLV3 in specific domains, the organizing center and the central

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4_25, © Springer Science+Business Media, LLC, part of Springer Nature 2023

537

538

Pau Formosa-Jordan and Benoit Landrein

zone, respectively, are key for orchestrating stem cell homeostasis. Furthermore, stem cell homeostasis has been shown to depend on additional regulators such as cytokinins and transcription factors from the HAIRY MERISTEM (HAM) family, whose expression in the SAM is also tightly regulated [6–9]. Studying the spatiotemporal dynamics of the stem cell regulatory network is fundamental to understand its function. In recent years, live imaging techniques relying on the use of fluorescent reporters imaged by 3D laser scanning microscopy have been developed, allowing time-lapse imaging of gene expression in both inflorescence and floral meristems [10, 11]. Having a quantitative analysis of such expression domains is pivotal for a better understanding of stem cell behavior, and has notably been used to evaluate stem cell plasticity and robustness upon environmental perturbations [12]. As these expression domains span over dozens or hundreds of cells, it is convenient to have pipelines optimized to extract certain quantities at the tissue level (see Note 1). This chapter presents a simple method that enables a highthroughput extraction and quantification of either radially or axially symmetric expression domains in the shoot meristems from 3D laser scanning microscopy images. Taking advantage of the symmetry of the expression domains, the method that is proposed extracts the fluorescent signal of the domain but also a characteristic length of the domain—a proxy for the size of the studied domain—by fitting a function to a radial fluorescence profile. The size of the expression domains of key regulators acting in the plant shoot meristems has been shown to be very relevant for understanding how the plant develops and adapts to environmental cues [8, 12], and therefore, it is paramount to develop methods that facilitate its quantification. The presented method also provides, along with other features, the mean autofluorescence levels surrounding the expression domain, which facilitates the extraction of the fluorescence levels in the expression domains, as well as a radial profile shape. Along with the detailed presentation of the method, this chapter further outlines two main applications to characterize fluorescence domains in inflorescence and floral meristems in a highthroughput manner. These pipelines include preprocessing steps, domain detection, and analysis using the proposed fluorescence profile fitting method. Overall, this chapter provides the reader with different tools to quantitatively characterize fluorescence domains in the shoot meristems. This chapter will not address aspects in relation to live imaging of the shoot meristems, and we recommend the reader to refer to other sources for this purpose (e.g., see Refs. [10, 12]).

Quantifying Expression Domains in Shoot Meristems

Method

2.1 Characterization of a Fluorescence Domain

This method consists on extracting, from a 3D confocal microscopy stack, basic measurements of a fluorescence domain whose z-projection is radially symmetric (see Note 2) and exhibits higher fluorescence levels at the center of the domain. Specifically, this method takes advantage of the radially symmetric decay of the fluorescence in the expression domain to compute a characteristic length L, which can be used as a proxy for the expression domain size. To extract the radial fluorescence profile, the method performs an angular integration around the centroid of the fluorescence domain on the sum slices projected and Gaussian blurred image (Fig. 1). The angular integration consists of computing the mean fluorescence intensity of N concentric subregions within a circular region R around the centroid r0. The most central subregion is a circle of a certain radius dρ, and the other subregions are N - 1 concentric i-rings (i = 1,. . .N - 1) of width dρ filling the R region (Fig. 1b); the i-th ring is defined by those pixels whose (x, y) coordinates fulfill the relation ρi2 < (x - x0)2 + (y - y0)2 ≤ ρi + 12, x0 and y0 being the coordinates of the centroid r0 and where ρi = i dρ. Then, the fluorescence profile can be fitted to several possible functions, such that we can extract the domain features more straightforwardly. In particular, we fit the profile to a generalized exponential function [13] defined as

a

b

c

y

max

min

x

Fluorescense (A.U.)

2

539

600 400 200 CE 00

Fit Data DE LE DE /e

20 Radius (µm)

40

Fig. 1 Method for fluorescence domain characterization. (a) Maximal intensity projection of a 3D confocal microscopy stack of the shoot apical meristem, showing the pWUS::WUS-GFP expression pattern. (b) Illustrating the profile extraction in a zoomed image of the inflorescence domain shown in a. The outer white circle denotes the limits of the R region of interest (ROI) where the radial intensity profile will be computed. Within this region, the average intensity is computed within concentric rings like the blue one depicted. The blue circle at the center illustrates the discrete unit for computing the average intensity at the origin of the intensity profile. (c) Radial intensity profile with a generalized exponential fit. The fit allows to extract the characteristic size LE at which the maximal signal DE has dropped an e-factor, having subtracted the inferred mean autofluorescence CE. Our method proposes the extraction of this characteristic size as a proxy for the fluorescence domain size. Scale bars: 50 μm (a), 25 μm (b). (Figure panels are modified from Ref. [12])

540

Pau Formosa-Jordan and Benoit Landrein

F E ðr Þ = C E þ D E e

-

nE r LE

,

ð1Þ

,

ð2Þ

and to a decreasing Hill function of the form F H ðr Þ = C H þ

DH 1þ

r LH

nH

LX being the characteristic length of the domain [13], CX the mean autofluorescence levels of the sum slices projection, DX the height of the fluorescence peak or mean maximal levels, and nX an exponent accounting for the radial decay of the fluorescence domain, what we will refer as profile shape parameter; X denotes the use of either the generalized exponential (X = E) or the Hill (X = H) fitting function. Note that apart from LX, the fit also determines the parameters CX, DX, and nX, defined above (Fig. 1c). We expect that certain profiles will fit better to a Hill profile, for example, in the case of a profile presenting a sigmoidal shape, while other profiles having a smoother decay will have a better fit to a generalized exponential function (we discuss how to determine the best fit in the inflorescence pipeline subsection). From Eqs. (1) and (2), we can see that LE and LH provide the distance from the center of the fluorescence domain at which the maxima intensity decreases “e”-fold and twofold, respectively (Fig. 1c), having subtracted the autofluorescence. Hence, unless the fluorescence domain boundary is very well defined by a sharp decline of the fluorescence intensity, such characteristic lengths LE and LH should not be interpreted as an exact radius of the whole fluorescence domain, but rather as a characteristic size associated to it. Note that the concept of a characteristic size has been widely used for characterizing morphogen gradients [13–15]. The shape parameter of the generalized exponential fit, nE, will tell us whether the profile is wider (nE < 2) or narrower (nE > 2) than a Gaussian, while the shape parameter from the Hill fit, nH, will give an idea of how sigmoidal the profile is, by how large nH will be. The mean maximal levels DX might also be informative and show interesting behaviors upon certain perturbations. Apart from these signal profile features, one may want to find the total fluorescence in the defined region R of the studied domain, what we refer as MR. The fitting function method also helps in retrieving this parameter, as it infers the mean autofluorescence CX, which needs to be properly subtracted from the total fluorescence. To compute the total fluorescence of the selected domain, we perform the following operation: MR =γ

pi , i∈R

ð3Þ

Quantifying Expression Domains in Shoot Meristems

541

where pi are the pixel intensity levels in the i-th pixel of the sum slices projection, and γ is the resolution in square microns units of the z-projected image under study. The γ factor corrects the differences in intensity levels between images that have different resolution in the xy plane (see Note 3). Then, the total autofluorescence in the R region, derived by the Hill or generalized exponential fitting, denoted by BR,X, is subtracted from the summed fluorescence in the selected domain, namely, I R,X = M R - B R,X ,

ð4Þ

where BR,X = CXAR, being AR the area of the R region. Below, we provide applications of the presented method through different pipelines applied to analyze expression domains in Arabidopsis inflorescence and floral meristems. These applications can be straightforwardly adapted to study any 3D fluorescence profile whose 2D projection is radially symmetric. 2.2

Applications

The reader can put into practice the presented method through pipelines that have been implemented with the Matlab RegionsAnalysis code ( ht tps:// gitlab.com /slcu/ teamHJ/pau/ RegionsAnalysis, see Notes 4 and 5). The relevant pipelines can be executed by means of the different scripts located in the RegionsAnalysis/Applications/ShootMeristemsAnalysis subfolder, with the help of the Readme file found in the same folder. Such pipelines will enable to analyze all the multichannel microscopy z-stacks in either lsm or czi formats that are located in a specific folder at once. After executing the pipelines of interest, different outputs will be generated in the containing folder of the microscopy data, including a csv file with the main descriptors of the studied fluorescence domains (see Readme file). Available microscopy data of shoot meristems with fluorescence reporters can be found at [16]. As a start, we suggest the reader uses the script main_script_basicpipeline_fittings, which enables the user to define the fluorescence domains under study by manually selecting the different domains of interest (see Readme file). This script will compute the characteristic sizes and fluorescence intensities of the selected domains. The reader can use this script to have a sense of how large the R regions surrounding the fluorescence domains should be, and it can be useful to get familiarized with other parameters that will be further used in the other pipelines (see Note 6). The two other pipelines presented below automatically find the expression domains of interest in the z-stack, the first pipeline focusing on analyzing gene expression in the inflorescence meristem and the second one in floral meristems. Domain extraction is performed on a Gaussian-filtered maximal intensity slices projection of the fluorescence image under study, while the analysis of the

542

Pau Formosa-Jordan and Benoit Landrein

Inflorescence domain analysis Z-projection: Max intensity

Gaussian blurr (5 µm)

Otsu thresholding

intensity profile shape characteristic size autofluorescence intensity maxima total intensity

z-Stack

1

Z-projection: Sum slices

Gaussian blurr (5 µm)

Central domain analysis

Fig. 2 Sketch of the inflorescence meristems pipeline on images showing the pWUS::WUS-GFP reporter. (Adapted from Ref. [12]). Scale bars: 50 μm

domains (e.g., the size computation) is performed on a Gaussianfiltered sum slices projection (Fig. 2). In both tasks, the Gaussian filter is used to smooth out the fluorescence images and get rid of local variations in signal due to the intracellular localization (either nuclear or endoplasmic reticulum) of the fluorescent markers (see Note 7). The automatic identification of the expression domains is performed through an Otsu thresholding over the corresponding z-projected and Gaussian-filtered fluorescence images [17] (see Note 8). These pipelines assume that the images are acquired such that the z-stacks show transversal sections of the meristems. Note that to correctly detect the inflorescence and floral fluorescence domains, the inflorescence meristem should be as close as possible to the center of the image. 2.2.1 Inflorescence Meristem Pipeline

This pipeline can be executed by running the script main_script_inflorescense_meristem, which consists of the following main stages (Fig. 2): 1. Automatic detection of the expression domain in the inflorescence meristem. This pipeline executes some preprocessing steps as mentioned above, followed by an Otsu thresholding, producing a binarized image, with domains having a value of 1 when the pixel intensity is above the threshold, and 0 when it is below. As the floral primordia usually express the reporters present in the inflorescence meristem, the binarized image shows multiple domains. Then, the code assumes that the binarized domain corresponding to the inflorescence meristem is the one that is the closest to the geometric center of the ensemble of the domains’ centroids in each image (see Note 9). Then, the code finds the centroid of the inflorescence domain, r0, and defines a circular region R of radius ρR that should contain the fluorescence domain under study in the meristem. It is important to manually check whether the automatic detection of the domain at the inflorescence meristem has been correctly performed (see Notes 9, 10, and 11).

15

3

15

1.8

E

pWUS::GFP domain area ( m 2)

2

2.2

e

2000

0 -50

-200 -250 40

2.6

10

3.5

60

80

100

120

Mean autofluorescence CE (A.U)

6

3

2.5

1500 1000

2

1.5

500 0

2.4

Profile shape nE

Characteristic size L ( m)

d

50

-150

r =0.829 p-value: keep_genes=["AT2G17950", "AT2G27250"] > genes_remove_list=[] > pre_mode = "expr_pre" # or "cell_mapping" > ns=4 # num_neighbors_source > nt=4 # num_neighbors_target > ap=0.1 # alpha > ep=0.05 # epsilon > tr="none" # transform > md="euclidean" # method_dist > ts=50 # top_sccells > tc=100 # top_cor_genes > mi=5000 # max_iter > to=1e-9 # tol > ms=50 # min_sccells_gene_expressed > ce=False # cell_enrichment

3.2.3 Loading the Required Data

As the first step of the pipeline, we have to load the scRNA-seq DGE and the spatial reference expression dataset. To load the scRNA-seq dataset we issue the command > dge, gene_names, cell_ids = gep.load_scDGE(dge_path=dge_path)

This returns not only the expression values in “dge” but also the names of the genes (“gene_names”) as well as the name of the cells (“cell_ids”). We can load the spatial expression dataset in the following way: > sem = gep.load_spatial_expression_matrix(coord_path=coord_path)

3.2.4 Data

Preprocessing the

We start the preprocessing step by removing genes in the scRNAseq DGE that are expressed in less than “min_sccells_gene_expressed” cells. We have the option to force this preprocessing step to retain some genes via the argument “keep_genes”. The function returns the preprocessed DGE as well as the names of the genes that remain. > dge, genes_name_keep = gep.pre_process_scDGE(dge, gene_names=gene_names, min_sccells_gene_expressed=ms, keep_genes=keep_genes)

Next, we subset the genes in the spatial expression matrix to only contain the genes also present in the scRNA-seq DGE. The following function call returns the preprocessed spatial expression

3D Mapping of Single-Cell RNA-seq Data

573

matrix, the name of the reference genes, and another matrix containing the coordinates of the cells in the 3D meristem model. > insitu_matrix, coord, sel_genes = gep.pre_process_spatial_expression_matrix(sem=sem, genes_name_keep=genes_name_keep, genes_remove_list=genes_remove_list)

We next get the indices of the reference genes in the scRNA-seq DGE. This step is necessary for easy referencing of rows in the scRNA-seq DGE. > index_genes = gep.get_idx_ref_genes_in_scDGE(dge=dge, genes_name_keep=genes_name_keep, sel_genes=sel_genes)

Now we find cells in the scRNA-seq DGE that are most similar to the cells in the spatial expression dataset. We have to specify the distance measure that we want to use to calculate distances between cells and we have to specify how many most similar cells from the spatial expression matrix we want to retain for each cell in the scRNA-seq DGE via the argument “top_sccells”. This produces a set of potentially duplicated cells. If we want to keep this enriched set of cells, we can set the argument “enrich” to “TRUE” (see Note 1). The function returns the indices of the selected cells in the scRNA-seq DGE. > topcells = gep.get_best_cells(dge=dge, index_genes=index_genes, insitu_matrix=insitu_matrix, method_dist=md, top_sccells=ts, enrich=ce)

Another necessary step before we can start with the reconstruction is to generate a scRNA-seq DGE with genes that are most informative of the cell types in that dataset. In our approach, we select the genes whose expression profiles across cells are most highly correlated. > dge_hvg = gep.subset_dge_to_informative_features(dge=dge, topcells=topcells, sel_genes=sel_genes, genes_name_keep=genes_name_keep, top_cor_genes=tc, pre_mode=pre_mode)

3.2.5 Preparation for novoSpaRc-Based Expression Reconstruction

With all that information, we can now calculate cost matrices between cells in “expression space” (scRNA-seq DGE) and “physical space” (spatial expression dataset). In particular, these matrices contain the distances between cells within the scRNA-seq dataset (based on their expression of informative genes) and the distances between cells within the spatial expression dataset (based on their location in 3D space), respectively. More information on the

574

Manuel Neumann and Jose M. Muino

importance of which distance measure to choose for the calculation of cell-to-cell mappings is provided in Note 2. > cost_expression, cost_locations = gep.get_expression_location_cost(dge_hvg=dge_hvg, coord=coord, ns=ns, nt=nt)

Since we know the expression profile of a set of marker genes in the 3D meristem, we can also obtain another cost matrix which describes the distance between cells of the scRNA-seq and spatial expression dataset with respect to the expression of the (in this case 23) reference genes. This allows us to describe how cells in “expression space” and cells in “physical space” are connected based on their expression of reference genes. Note that novoSpaRc (the underlying tool that implements the OT-based expression reconstruction) can also be used without such knowledge. More on that in the “Notes” section. > cost_marker_genes = gep.get_marker_gene_cost(dge=dge, insitu_matrix=insitu_matrix,

pre_mode=pre_mode,

method_-

dist=md, tr=tr, index_genes=index_genes, topcells=topcells)

In the last step, before we can predict mappings between cells, we have to define uniform distributions over the cells in the scRNAseq DGE and the spatial expression matrix. > p_locations, p_expression = gep.get_distributions_over_expression_location(dge_hvg=dge_hvg, coord=coord)

3.2.6 Predict Cell-to-Cell Mappings and 3D Gene Expression Profiles

Now we can predict how cells in the two datasets are related. We have to provide the function with the cost matrices we calculated above, the distributions over cells, and we have to specify a few parameters influencing how cells are mapped to one another. The most important parameters are “alpha” and “epsilon”. “epsilon” is a regularization constant where large values result in more regularization. Values for “alpha” can be between 0 and 1, with values closer to 0 giving more importance to the information present in the spatial expression reference dataset while reconstructing cell-tocell mappings. > gw = gep.predict_cell_to_cell_mappings(cost_marker_genes=cost_marker_genes, cost_expression=cost_expression, cost_locations=cost_locations,

p_expression=p_expression,

p_locations=p_locations, alpha=ap, epsilon=ep, max_iter=mi, tol=to, verbose=verbose)

3D Mapping of Single-Cell RNA-seq Data

575

With the matrix of the previous step and the original preprocessed scRNA-seq DGE, we can now reconstruct the 3D gene expression patterns for all genes of interest. Note that the number of reference genes used to predict 3D gene expression profiles has a great impact on the prediction performance as described in Note 3. > sdge = gep.predict_spatial_gene_expression(gw=gw, dge=dge, topcells=topcells, pre_mode=pre_mode)

3.2.7 Format and Save the Results

To make the matrix of cell mappings more informative, we add back the original names of cells in the scRNA-seq DGE and transform it to a pandas.DataFrame. > cell_ids_sel = cell_ids[topcells] > gw = round(pd.DataFrame(gw, index=cell_ids_sel), 3)

Also, we transpose the reconstructed spatial DGE to have dimension (cells x genes), add coordinates of the cells, add gene names, and transform it to a pandas.DataFrame. > sdge = np.transpose(sdge) > sdge = np.concatenate((coord, sdge), axis=1) > col_names = np.concatenate((np.array(["x", "y", "z"]), genes_name_keep), axis=0) > sdge = round(pd.DataFrame(sdge, columns=col_names), 3)

Finally, we can save the two output matrices which will be used in the next sections. > gw.to_csv(out_base_path + "/cell-to-cell_mapping.csv", index=True, header=True, sep=’,’) > sdge.to_csv(out_base_path + "/sdge.csv", index=True, header=True, sep=’,’)

3.2.8 Visualizing 3D Gene Expression Profiles

After having predicted 3D gene expression profiles, we can visualize those predictions. With the predicted spatial expression matrix (spDGE) in the environment, we can use the plotting function “pl.plot_3d_meristem()” to visualize the 3D gene expression of any gene. Before we can do that, we have to define a dictionary (for convenience) that maps gene symbols to TAIR IDs. >

st

=

{"ETT":"AT2G33860",

"CO2":"AT1G62500"}

"STM":"AT1G62360",

576

Manuel Neumann and Jose M. Muino

Fig. 1 Visualization of 3D gene expression profiles. The predicted (a) and “true” (b) 3D gene expression profile of ETT

We can then execute the following code to plot the 3D gene expression profile of the gene ETT: > gene_symbol="ETT" > col_code = ["red","green"] > fig = pl.plot_3d_meristem(df=sdge, gene = st[gene_symbol], colorscale = col_code, plot_title = gene_symbol, show_legend=True, backgrnd_col="white", font_col="black", legend_title="expression level") > fig

Since we still have the spatial reference expression dataset loaded in our environment, we can also visualize the “true” binary 3D gene expression profile of ETT. > fig = pl.plot_3d_meristem(df=sem, gene = st[gene_symbol], colorscale = col_code, plot_title = gene_symbol, show_legend=True, backgrnd_col="white", font_col="black", legend_title="expression level") > fig

We can see that the prediction (Fig. 1a) matches very well with the “true” 3D expression profile (Fig. 1b). 3.3 Projecting scRNA-seq Clusters onto the 3D Meristem

Next to predicting 3D gene expression profiles, the results of the novoSpaRc tool can also be used to project single-cell UMAP clusters onto the 3D flower meristem model. In other words, we are able to infer from which location in the flower meristem cell clusters in UMAP space originated from. The code to perform this projection of UMAP clusters onto the 3D meristem model is written in the R programming language and can be followed in the JupyterLab notebook “3—UMAP_cluster_projection”. Alternatively, the user can also use the precomputed output file “cluster_assignment_3d.csv” that comes with the downloaded GitHub repository. In brief, first the cell-to-cell mapping matrix is normalized by the sum of columns. Then the probabilities of all cells in the spatial

3D Mapping of Single-Cell RNA-seq Data

577

reference dataset belonging to a particular cluster is obtained by summing the probabilities of all cells belonging to a particular cluster for this cell. This is repeated for all clusters and all cells in the spatial reference dataset. 3.4 Visualizing scRNA-seq Clusters in the 3D Flower Meristem

In the previous section, we described how clusters of cells in UMAP space can be projected onto the 3D meristem model. Now we will be visualizing those projections.

3.4.1

First, we have to load the data that contains the projections.

Load the Data

> path = base_dir + "/data/cluster_assignment_3d.csv" > m = pd.read_csv(path, sep=",", index_col=0, decimal=".") # (cells x genes)

We also load the dataset that contains the assignment of UMAP clusters numbers and the biological annotation of those cluster for all cells in the scRNA-seq dataset. The dataset for this scRNA-seq dataset is provided with the downloaded GitHub repository. > path = base_dir + "/data/seurat_cell_meta_data.csv" > umap = pd.read_csv(path, sep=",", index_col=0, decimal=".") # (cells x UMAP_coordinates)

3.4.2 Visualize the UMAP Plot of scRNA-seq Cells

In order to get an understanding of how cells in the scRNA-seq dataset relate to each other, we are using the function “pl.plot_umap()” to visualize those cells in lower dimensional (UMAP) space (Fig. 2a). The resulting plot also contains information about the biological annotation of the clusters. > pl.plot_umap(umap=umap, plot_size_x=9, plot_size_y=7, label_size=15)

Fig. 2 Representation of single-cell UMAP clusters. (a) Visualization of clustering of single cells in UMAP space and (b) predicted probabilistic distribution of UMAP clusters in 3D flower meristem

578

Manuel Neumann and Jose M. Muino

3.4.3 Plot the Projection of UMAP Cluster in the 3D Meristem

The position of clusters in UMAP space and the biological annotation gives us insight into the biological function of cells in our dataset. But this information is limiting since we do not know the location in the flower meristem where cells in a particular cluster originated from. By executing the code below, we can plot the projection of UMAP cell clusters onto the 3D flower meristem (Fig. 2b) and can resolve that problem. > col_code = ["yellow", "red"] > cluster_name = "cluster_9___epidermis" > fig = pl.plot_3d_meristem(df=m, gene = cluster_name, colorscale = col_code, plot_title = cluster_name, show_legend=True, backgrnd_col="white", font_col="black", legend_title="cluster probability") > fig

In the code shown here, we project cells in cluster 9 (cluster_9___epidermis) onto the 3D flower meristem. We can do this for any cluster. The color of the cells indicates the probability that a particular cell belongs to cluster 9. We can see that mainly cells in the L1 layer have a high probability of belonging to cluster 9. In other words, cells in cluster 9 probably originate from the L1 layer of the flower meristem, which fits with the annotation “epidermis” of this cluster. 3.5 Evaluating Prediction Performance (AUCROC and PEP-Score)

We do not know how accurate the predicted 3D gene expression profiles of genes that are not present in the spatial expression dataset are. We have no way of calculating, for example, an AUROC (Area Under the Receiver Operating characteristic Curve) score since we have, by definition of a de novo prediction, no expression vector to compare the prediction to. But we can use the information available in the scRNA-seq dataset to estimate the prediction performance of de novo predicted gene expression profiles. In particular, we can calculate how similar the expression profile of a particular “gene A” is to the expression profiles of the genes in the reference dataset that was used to “train” the model. The similarity is calculated based on the expression profiles in the scRNA-seq dataset. If the expression profiles of “gene A” is very similar to the genes in the reference expression dataset, we can assume that the model is able to reconstruct the 3D gene expression pattern for “gene A” successfully. The similarity measure is based on the Spearman rank correlation and expressed in the Predicted Expression Performance (PEP) score (see Note 4). The higher the PEP-score the more confident we are that the prediction we made is accurate.

3D Mapping of Single-Cell RNA-seq Data

579

The exact way to calculate PEP-scores is shown in the JupyterLab notebook “5—Evaluate Prediction Performance”. For other applications of the PEP-score, see Note 5.

4

Notes 1. Since the OT-based model implemented in novoSpaRc tries to match gene expression and physical space of cells, it proved helpful to ensure that the scRNA-seq and spatial expression dataset contain cells of similar type. We therefore recommend performing a selection of single-cell RNA-seq cells to better represent the cells in the spatial expression dataset. Even though not improving the prediction performance significantly in our case, it might also be helpful to additionally enrich rare types of cells in the scRNA-seq dataset to better represent the population of cells in the spatial gene expression dataset. 2. There are many ways to calculate distances between cells within or between datasets. It is recommended to test the performance of different distance measures in order to obtain the best possible reconstruction of spatial gene expression patterns for a particular dataset and organism. We recommend using the Euclidean distance as well as the Hamming and Jaccard distances for binary data. 3. As has been demonstrated in Neumann and Xu et al. [9] the more reference genes are present in the spatial expression dataset, the higher the performance of the 3D gene expression prediction. It is recommended to have at least 20 genes, better 50 or 100 genes in the spatial reference dataset to cover enough spatial gene expression space that can be leveraged to reconstruct de novo gene expression patterns in 3D. 4. The PEP-score is a way to have a rough estimation of the expected performance of the prediction for each gene. However, we advise to validate the expression predictions of the most interesting genes for a given study by reporter gene assays. 5. Since the PEP-score is calculated based on the scRNA-seq dataset, it is independent of the spatial expression dataset. That gives the opportunity to estimate which genes the model will be able to de novo predict with a high performance before even training and running the model. Moreover, this allows one to find which and how many genes are necessary to have in the spatial reference expression dataset to accurately predict as many genes as possible. In that way, the experimental planning and generation of spatial reference expression datasets can be informed by the PEP-score to minimize the experimental cost while maximizing the ability to accurately predict 3D gene expression patterns.

580

Manuel Neumann and Jose M. Muino

References 1. Chen KH, Boettiger AN, Moffitt JR, Wang S, Zhuang X (2015) Spatially resolved, highly multiplexed RNA profiling in single cells. Science 348:aaa6090 2. Rodriques SG et al (2019) Slide-seq: a scalable technology for measuring genome-wide expression at high spatial resolution. Science 363:1463–1467 3. Rao A, Barkley D, Franc¸a GS, Yanai I (2021) Exploring tissue architecture using spatial transcriptomics. Nature 596:211–220 4. Welch JD et al (2019) Single-cell multi-omic integration compares and contrasts features of brain cell identity. Cell 177:1873–1887.e17 5. Korsunsky I et al (2019) Fast, sensitive and accurate integration of single-cell data with harmony. Nat Methods 16:1289–1296 6. Lopez R et al (2019) A joint model of unpaired data from scRNA-seq and spatial transcriptomics for imputing missing gene expression measurements. arXiv:1905.02269. https:// doi.org/10.48550/arxiv.1905.02269

7. Abdelaal T, Mourragui S, Mahfouz A, Reinders MJT (2020) SpaGE: spatial gene enhancement using scRNA-seq. Nucleic Acids Res 48:e107 8. Nitzan M, Karaiskos N, Friedman N, Rajewsky N (2019) Gene expression cartography. Nature 576:132–137 9. Neumann M et al (2022) A 3D gene expression atlas of the floral meristem based on spatial reconstruction of single nucleus RNA sequencing data. Nat Commun 13:2838 10. Satija R, Farrell JA, Gennert D, Schier AF, Regev A (2015) Spatial reconstruction of single-cell gene expression data. Nat Biotechnol 33:495–502 11. Refahi Y et al (2021) A multiscale analysis of early flower development in Arabidopsis provides an integrated view of molecular regulation and growth control. Dev Cell 56:540– 556.e8 12. Leggio B et al (2019) MorphoNet: an interactive online morphological browser to explore complex multi-scale data. Nat Commun 10: 2812

INDEX A Abaxial ............................................... 11, 12, 65, 71, 115, 133, 393, 437, 557 ABC model ..................................................17–28, 41–49, 67, 68, 70, 75 ABCDE model ................................................... 17–25, 74 Abscisic acid (ABA)....................................................... 406 Activation tagging ......................430, 433, 440, 445, 446 Adaxial ................................. 12, 65, 66, 69, 70, 115, 393 AGAMOUS (AG) .......................................17–25, 27–29, 45, 47, 48, 52, 67–70, 120, 132–135, 332, 408, 409, 430, 433, 436, 437 Agrobacterium Agrobacterium-mediated transformation .............. 136 Agroinfiltration........................................... 369, 393–395, 554, 555, 557, 562 Alcian blue staining .............................................. 243, 254 AlcR/AlcA system................................................ 432, 438 Alexander staining....................................... 200, 205, 206 Amborella trichopoda.................................................89, 90 Ammonium sulphate .................512, 520, 523, 527, 528 ANA-grade ..................................... 85, 90, 92–94, 97, 98 Angiosperms origin..................................... 84, 85, 89, 99, 100, 365 Aniline blue staining .................................. 200, 202, 209, 243, 244, 247–248, 263, 267–270 Anther anatomy .......................................................... 201, 207 dehiscence......................................117, 118, 201, 207 development .....................22, 68, 118, 119, 209, 405 lobe .......................................................................... 214 morphogenesis ........................................................ 199 structure.......................................................... 201, 207 Anthesis ............................................. 112, 116, 122, 241, 242, 247, 248, 250, 267, 268, 379, 380 Antibodies ......................................... 167, 181, 191, 195, 221, 223, 228, 230, 236, 237, 265, 271–274, 325, 332, 337, 339, 344, 348, 351, 444 APETALA1 (AP1)......................... 42, 67, 285, 333, 437 APETALA2 (AP2) ................................... 9, 67, 133, 408 APETALA3 (AP3)............................................17, 43, 67, 398, 408, 433, 437 Arabidopsis accession .................................................................. 315

transformation ........................ 49, 135, 434, 441, 454 Arabidopsis Biological Resource Center (ABRC) ..................................................... 247, 352 Asterids ............................................................... 40, 45, 52 Austrobaileyales ........................................... 85, 93, 97, 98 Auxin.................................................9–13, 16, 25–27, 49, 51, 113–122, 133, 243, 246, 251

B Backcross .............................................140, 141, 145, 455 BASTA ........................................................................... 149 β-glucuronidase (GUS).............................. 243, 246, 247, 265, 272, 276, 332, 351–361 Biotinylation ........................................314, 315, 319, 321 Bisexual ....................................................... 83, 85, 90, 94, 95, 97, 102, 103 Boundary .............................................. 12–13, 16, 17, 19, 24–26, 28, 43, 69, 117, 174, 250, 254, 294, 298, 354, 424, 425, 437, 540 Bradford assay ............................................................... 359 Brassinosteroid ..................................................... 118, 121 5-Bromo-4-chloro-3-indolyl phosphate (BCIP) .......167, 169, 181, 191, 194, 195, 316 5-Bromo-chloro-3-indolyl glucuronide (X-gluc) ......265, 276, 279, 352

C Callose staining .................................................. 202, 209, 210 CAPS, see Cleaved amplified polymorphic sequences (CAPS) CArG-box..............................................14, 18, 22, 24, 25 Carpel development ......................... 21, 23, 47, 70, 241–257 Cauliflower Mosaic Virus (CaMV) .................... 286, 354, 430, 432, 433, 435 cDNA............................................................................. 388 Cell death, see Programmed cell death (PCD) Cell type specific promoter..................................................... 314 Cellular resolution................................................ 331–349 Cell wall .............................................................13, 19, 21, 181, 200, 243, 244, 256, 257, 262, 264, 270–272, 278, 296, 307, 308

Jose´ Luis Riechmann and Cristina Ferra´ndiz (eds.), Flower Development: Methods and Protocols, Methods in Molecular Biology, vol. 2686, https://doi.org/10.1007/978-1-0716-3299-4, © Springer Science+Business Media, LLC, part of Springer Nature 2023

581

FLOWER DEVELOPMENT: METHODS AND PROTOCOLS

582 Index

Central zone (CZ) .....................293, 294, 298, 437, 537 Cetyl-trimethyl-ammonium bromide (CTAB)...........142, 145, 146, 149, 152, 157, 460, 471 ChIP-seq................................................21, 314, 325, 495 Chloral hydrate....................................201, 244, 256, 263 Chromatin immunoprecipitation (ChIP) ........................ 288, 431 Chromosome meiotic ................................................... 219, 220, 227 spreads .................................. 221, 223, 228–230, 233 CLAVATA3 (CLV3)..............................13, 133, 298, 537 Cleaved amplified polymorphic sequences (CAPS) ...................................................... 137, 150 Clustered regularly interspaced short palindromic repeats (CRISPR)...................................52, 348, 432, 440 Colony PCR ...................... 148, 158, 390, 391, 398, 438 Computational ............................................. 24, 366, 496, 509–511, 516, 519, 568–571 Confocal laser scanning microscopy (CLSM)........................................... 164–166, 168, 170–171, 184–187, 192 Confocal microscope .................................. 264, 271, 316 Cycloheximide...................................................... 435, 445 Cytokinins .................................................... 9, 12, 17, 23, 113, 115, 118–122, 246, 538

D DAPI, see 4′-6-Diamidino-2-phenylindole (DAPI) Darkfield microscopy .................................................... 243 Database ........................................ 41, 88, 94–96, 98, 99, 383, 405, 469, 501, 504, 505, 511–520, 525, 528–530 Dehiscence............................................ 47, 117–119, 122, 201, 207, 242, 243, 247, 254 Derived CAPS (dCAPS) ...................................... 137, 150 Dexamethasone (DEX).............................. 286–288, 291, 431, 435, 436, 438, 444, 445, 447 4′-6-Diamidino-2-phenylindole (DAPI) ....................223, 224, 228–231, 233, 234, 236, 237, 244, 248, 308, 310, 311, 315, 317, 323 Differential interference contrast (DIC).....................233, 234, 243, 253, 263, 267, 269, 270, 275, 276, 343 Digoxigenin (DIG) .................................... 178, 179, 191, 194, 334, 338, 339 DNA extraction .............................................. 142, 146, 435, 438, 448, 458–460, 462, 471–472 purification ....................................194, 318, 438, 445 DR5:GFP......................................................................... 50

E Embedding ........................................ 166, 173, 175, 176, 208, 211, 215, 245, 252, 253, 264–266, 271, 274, 279, 332, 335, 340–342, 355, 357

Embryo dissection ........................................................ 170, 171 Endothecium ..................... 116, 118, 119, 199, 200, 209 Enhancer screen .............................................................. 132–134 Eosin ..........................................167, 176, 194, 202, 203, 208, 210, 245, 252, 257, 335, 340 Epitope tag ............................................................................ 444 Ethylene......................................................................... 116 Ethyl methanesulfonate (EMS).......................... 132–137, 140–142, 144, 145, 152, 156 Eudicot .................................................40, 43, 53, 59, 60, 65, 67, 69, 72, 74, 75, 84, 85, 87, 97, 220, 221 EvaGreen .............................................409, 414, 420, 424 Evo devo .......................................................................... 96 Evolution .................................................... 41, 51–53, 59, 62, 74, 85, 87, 89, 91, 95, 97, 102, 365, 510 Expression analysis ............................................20, 21, 25, 49, 62, 70, 74, 99, 135, 165, 194, 244, 262, 285, 291, 294, 301, 307, 313, 314, 331, 332, 365–371, 379, 380, 384, 386–389, 403, 423, 430, 492, 495–497, 515, 538, 540, 543–546, 568, 569 domain ........................................................... 6, 20, 22, 25–27, 42, 44, 46, 49, 74, 113, 114, 120, 285, 293, 294, 298, 314, 403, 431, 432, 537–546, 548 pattern.............................................12, 42, 45, 46, 49, 51, 68, 69, 75, 90, 99, 118, 164, 165, 194, 246, 262, 293, 294, 301, 331, 351–353, 403, 431, 436, 438, 539, 546, 568, 571, 575, 577, 579

F FDA, see Fluorescein diacetate (FDA) FISH, see Fluorescence in situ hybridization (FISH) Floral bud ........................................................ 10, 12, 65, 96, 207, 209–211, 214, 215, 225, 227, 236, 286, 291, 370, 371, 380 crops....................................................... 454, 456, 471 determinacy .......................... 23, 25, 45, 47, 132–135 development ................................................ 12, 29, 60, 132–135, 379, 384 dip method ....................................136, 288, 289, 319 induction system ....................................285–292, 502 meristem ................................................10, 11, 13–17, 19–21, 23, 25–28, 46, 49–51, 62, 65, 67, 69, 75, 111–113, 133, 163–197, 437, 446, 456, 485, 538, 540, 541, 543–546 mutant ............................................................ 131–160 organ ......................................................17–20, 22, 24, 41–50, 52–53, 59, 63–65, 67–69, 75, 83, 92, 95, 96, 111–114, 122, 135, 165, 188, 190, 197, 236, 249, 266, 272, 313, 354, 365, 366, 371, 379, 398, 437, 438

FLOWER DEVELOPMENT: METHODS

AND

PROTOCOLS Index 583

organ formation ...................................................... 163 organ growth............................................................. 18 organ identity .............................................. 18, 23, 29, 41–49, 67–76, 430 organ initiation...........................................49, 51, 113 organ number..........................................51, 135, 165, 170, 189, 190, 446 organ primordia ............................................... 49, 165 organ specification .................................................... 19 primordia ...............................................9, 13–17, 188, 430, 542, 544–546, 548 tissues ................................................ 12, 29, 200, 288, 289, 291, 403–425, 435 transition................................. 3–9, 41, 112, 314, 371 Floret .............................................. 62, 63, 227, 232, 237 Floriculture ........................................................... 453–492 Flower development ............................................ 3–29, 52–53, 59–76, 87, 91, 94, 99, 111–122, 163–197, 225, 248, 285–292, 325, 351–362, 365–399, 403, 404, 408, 429–448, 453–492, 495–505, 509–530 development stages ........................................ 215, 371 evolution....................................................... 41, 87, 97 origin..................................................................83–103 Fluorescein diacetate (FDA)....................... 200, 201, 206 Fluorescence activated cell sorting (FACS) ....................... 293–299, 301, 314, 331 domain ........................................... 538–540, 542–548 in situ hybridization (FISH) ........................ 220, 225, 231–236, 238 Fluorescent reporter ............................................ 164, 538 Fluorometer................................................ 353, 356, 359, 360, 367, 382, 458, 460–462, 521 Fluorometric assay ............................................... 352, 356 Founder cell....................................................69, 113–115 Fruit development ................................................ 46–48, 51, 122, 242, 243, 247, 365, 371

screen .........................................25, 51, 131–160, 354 suppressor screen............................................ 131–160 Genevestigator............................................. 405, 408, 409 Genome annotation ..............................................460, 468–470 assembly ....................... 454, 455, 463, 464, 467, 468 coverage .......................................................... 477, 478 sequence .........................................60, 90, 91, 93, 99, 384, 454, 456, 458–460, 462–471, 529 Genomic DNA .................................................... 141, 143, 152, 160, 423–425, 433, 438, 445, 448, 459, 463, 555, 560 Genomics ................................................6, 12, 16, 19, 29, 68, 88, 91, 92, 95, 139, 143, 200, 246, 285, 308, 310, 314, 352, 432, 434, 454, 456, 458, 468, 469, 485, 495, 514, 516, 517, 519, 554 Genotyping -by-sequencing (GBS) ......... 457, 461, 471–479, 487 Gentamycin .........................................369, 391, 398, 433 Germline ...................................................... 261, 262, 267 Gibberellins .................................................. 14, 112, 113, 116, 118, 119, 121, 122 Glucocorticoid receptor (GR)................... 285, 431–433, 436, 438, 444 Glufosinate, see BASTA Glume ...........................................................61–65, 73–75 Grass flower ......................................................60–62, 65, 74 Green fluorescent protein (GFP) ......................... 49, 246, 265, 275, 325, 332, 351–362 GUS, see B-glucuronidase (GUS) Gymnosperms ...................................... 29, 84, 85, 87–89, 92, 93, 99, 100, 102, 365, 374, 516 Gynoecium ................................................. 23, 24, 65, 67, 69–72, 85, 89, 93, 94, 96, 97, 111, 112, 119–122, 133, 189, 197, 241–243, 246–248 Gynophore................................................... 120, 241, 251

G

Histochemistry ............................................ 315, 316, 319 Histoclear .......................................... 167, 175, 176, 181, 194, 245, 252–256, 335, 336 Histology .............................................................. 164, 264 Homeotic conversion.................. 23, 44–46, 49, 64, 69, 73, 446 gene....................................................... 17, 19, 23, 43, 45, 46, 49, 69, 72, 73, 75, 99, 430 mutant .................................................. 17, 23, 43, 46, 49, 64, 69, 72, 99, 430 protein ...................................... 19, 23, 44, 49, 72, 99 transformation .................................................. 17, 430 Hormone ..........................................14, 49, 73, 111–122, 244, 246, 285, 286, 291, 292, 294, 423, 447, 512, 562

Gametogenesis .............................................................. 261 Gametophyte .............................................. 117, 118, 261, 262, 268, 272, 274, 278, 436 Gene duplication.............................. 46, 47, 50, 53, 68, 102 expression domain............................. 17, 72, 537–549 expression profile.................. 294, 307, 331, 567–579 regulatory network (GRN)................................15, 29, 51, 287, 538 Genetic enhancer screen .............................................. 131–160 modifier screen ........................................................ 132 redundancy ..................................................... 132, 429

H

FLOWER DEVELOPMENT: METHODS AND PROTOCOLS

584 Index

Hygromycin (Hyg) .............................316, 319, 433, 440 Hypocotyl ........................................................10, 25, 172, 176, 178, 309, 320, 392, 393, 399

I Image J software ........................................................... 174 Imaging....................................................... 164, 171, 181, 185–187, 192, 193, 196, 201, 206, 221, 228, 263, 264, 269, 271, 273, 275, 388, 538, 545 Immunofluorescence .................................................... 220 Immunolabelling........................264, 265, 271, 272, 274 Immunoprecipitation.................................................... 314 Immunostaining.........................222, 223, 227, 228, 231 Inducible promoter................................................ 287, 430, 431 system ................................... 287, 430–433, 436, 438 Inflorescence...................................................... 10, 13–15, 17, 20, 40, 41, 49, 60–62, 74, 95, 111, 112, 163, 168, 170–189, 192, 194, 196, 202, 207–209, 212, 214, 225–228, 236, 247, 250, 266, 267, 276, 289, 313, 315, 332, 353, 356–358, 360, 361, 371, 376, 380, 410, 413, 416, 437, 438, 441, 445, 447, 497, 498, 527, 538–548 Insertion lines ...........................................................52, 361, 431 mutant ........................................ 44, 52, 53, 137, 431 Insertional mutagenesis ............................. 135, 136, 139, 143, 147–149, 429 In situ hybridization .........................................87, 90, 94, 96, 97, 99, 164, 165, 167–169, 175–184, 220, 223–225, 232–236, 331–349, 351, 352, 403, 568, 569 Isolation of nuclei tagged in specific cell types (INTACT) ........................................313–327, 331

J Jasmonates ......................... 113, 115, 116, 118, 119, 122 Jasmonic acid (JA) ................................................. 29, 115

K Kanamycin (Kan)................................................. 287, 316, 319, 369, 390, 391, 433, 434, 440, 442, 443 Knockdown ..................................................................... 72 Knock-out....................................... 44, 47, 48, 52, 72, 74

L Laser capture microdissection (LCM) ............... 200, 205, 214–216, 274, 301, 331 Laser microdissection (LM) ............................... 266, 272, 274, 277–279, 314 Laser scanning microscopy (LSM)............................... 538 Leaf ......................................................... 3–5, 8, 9, 13, 20, 21, 23, 41, 51, 52, 65, 70, 90, 92, 94–96, 102,

112, 113, 133, 144–146, 157, 163, 164, 175, 176, 193, 237, 294, 308, 309, 313, 320, 353, 365–399, 462, 471, 496, 510, 512, 513, 554, 557, 558, 560–564 Left border (LB) ................................................. 136, 139, 143, 148, 288, 289, 369, 390, 391, 393, 398, 433–435, 438, 441, 442, 447 Lemma ................................................... 61–66, 70, 73–75 Lettuce ............................... 219–222, 225–227, 229, 236 LhG4..................................................................... 431, 436 Library ........................................................ 140, 141, 152, 156, 160, 302, 304, 371, 373, 377, 382, 384, 385, 458, 459, 461, 463, 464, 472, 473, 480, 485, 571 Light microscopy ........................................ 200, 223, 230 Lignin ................................ 191, 243–245, 251–255, 257 Liquid chromatography-tandem mass spectrometry (LC-MS/MS) .......................................... 497–504, 511–515, 517, 518, 520, 521, 524–526, 529, 530 Live imaging ............................... 164, 168, 186–187, 538 Lodicule ............................. 61, 62, 64–66, 68, 69, 73–75 Luminescence .............................554, 555, 558, 559, 563

M M1........................................................136, 142, 145, 156 M2...............................................142, 145, 156, 263, 265 MADS .................................................................. 6, 19, 20, 22, 24, 28, 29, 43, 47, 72, 379 Magnoliids .......................................................... 85, 97, 99 Maize .................................................... 60, 164, 219–221, 223–225, 230, 232–237, 295, 353, 354, 360, 405, 510, 513, 517–519 Map based cloning ...................................... 132, 137, 140 Mapping population ....................................140, 142, 145, 146, 150, 156, 455–457 Marker .................................................. 11–13, 15, 21, 62, 70, 73, 113, 114, 136, 137, 150, 159, 167, 192, 194, 231, 237, 244, 262, 265, 267, 272, 274, 275, 279, 287, 294–296, 298, 302, 325, 332, 339, 340, 342, 354, 361, 367, 433, 436, 440, 444, 463, 473, 479–482, 484, 542, 546, 562, 574 Mass spectrometry (MS).................................16, 76, 143, 497, 501, 511, 517, 520, 529 Mathematical model ..................................................... 422 Megagametogenesis ...................................................... 262 Meiocytes.................................................... 200, 219–221, 225–228, 230, 232, 233, 236, 237 Meiosis ............................................... 118, 199, 200, 219, 220, 223, 225–231, 237, 262, 267, 275, 405 Meristem determinacy ............................................................... 69 embryonic meristem ..................... 165–166, 170–171

FLOWER DEVELOPMENT: METHODS floral meristem (FM).............................10, 11, 13–17, 19–21, 23, 25–28, 46, 49–51, 62, 65, 67, 69, 75, 111–113, 133, 163–197, 437, 446, 456, 485, 538, 540, 541, 543–546 identity................................15, 17, 20, 28, 46, 49, 50 inflorescence meristem (IM) ...................... 15, 20, 41, 49, 112, 113, 163, 164, 168, 170–184, 186–189, 289, 313, 315, 371, 376, 380, 437, 438, 540, 542–548 shoot apical meristem (SAM) ....................... 3, 4, 8, 9, 11–13, 15–17, 25, 112, 163, 164, 166, 172–175, 178, 193, 194, 293–299, 309, 313–316, 318, 319, 322, 325, 326, 345, 376, 380, 438, 537 size ............................... 133, 164–166, 174, 175, 543 termination .......................................................... 26–28 4-Methyl umbelliferyl glucuronide (MUG)...............352, 356, 359, 360, 362 Microarrays ............................................... 89, 92, 93, 200, 294, 299, 314, 331, 405 MicroRNA (miRNA) .................20, 26, 43, 44, 132, 510 Microtome ......................................... 166, 167, 173, 178, 194, 202, 203, 205, 208, 211, 214, 215, 245, 253, 335, 342, 355 MIQE guidelines.................................................. 421, 425 Misexpression ....................................................... 429–448 Modified high‐efficiency thermal asymmetric interlaced PCR (mhiTAIL-PCR).............136, 139, 151–154 Monocot ........................59–76, 85, 87, 88, 97, 220, 517 Morphology.............................................. 8, 41, 164, 165, 175, 181, 189, 192, 200, 202, 204–210, 228, 229, 236, 238, 242, 243, 247, 251, 264, 267, 269–272, 302, 347, 455, 492, 547, 548 mRNAs ................................................................... 23, 497 MUG, see 4-Methyl umbelliferyl glucuronide (MUG) Mutagenesis chemical .......................................................... 132, 429 EMS ............................................... 132–137, 141, 144 insertional ............................................. 135, 136, 139, 143, 147–149, 429 Mutant gain-of-function ..................................................43, 63 loss-of-function ........................................... 25, 63, 73, 133–135, 137, 430

N NASC, see Nottingham Arabidopsis Stock Center (NASC) Nectary ....................................................... 18, 22, 23, 42, 47, 48, 133, 437, 456 Neutral red .................................................. 166, 173, 256 Next-generation sequencing (NGS) .................. 200, 314, 382, 457, 461, 479 Nicotiana benthamiana ................................................. 554 Nightshades ...............................................................40–54

AND

PROTOCOLS Index 585

Nitroblue Tetrazolium (NBT) .................. 167, 169, 181, 191, 194, 195, 316 1-N-naphtylphtalamic acid (NPA) ..................... 243, 245, 249–251, 256 Nomarski ....................................................................... 263 Non model plant .................................219, 366, 373, 383 Norflurazon (Norf)....................................................... 433 Nottingham Arabidopsis Stock Center (NASC) ..................................................... 246, 247 Nuclei................................ 172, 233, 257, 268, 307–311, 313–327, 331, 332 Nymphaea thermarum ..............................................90, 92

O Orthologous............................................... 44, 46, 49, 52, 72, 88, 99, 366, 370, 383, 405 Oryza sativa....................................................60, 383, 513 Osmium tetroxide (OsO4) .................................. 187, 196 Ovary ....................................................40, 47, 62, 65, 73, 97, 120, 241, 250, 251, 257, 438 Overexpression .....................................48, 255, 430, 433, 435, 436, 554–556, 561 Ovule clearing............................................................ 263, 266 development .....................21, 22, 121, 243, 261–279

P Palea ................................................. 61–66, 68–70, 73–75 Paraffin........................................................ 167, 175, 176, 178, 194, 245, 252, 257, 278, 316, 319 Peptidome ...........................................510, 511, 517, 520 Peptidomics .......................................................... 509–530 Perianth ................................................59, 65, 68, 83, 85, 89, 96, 97, 99, 102, 103, 365, 430 Peripheral zone (PZ) .................................. 293, 294, 298 Petal ................................................17–23, 25, 28, 40–46, 49, 51–53, 59, 60, 67, 69, 83, 92, 111, 112, 115–116, 132–134, 164, 188–190, 207, 225, 247, 437, 438, 455, 456, 492, 496 Petunia ...................... 40–53, 67, 75, 354, 440, 454, 456 Phenotype...............................................6, 10, 23, 25, 26, 43, 45, 51–53, 68, 70, 72, 73, 75, 119, 132–135, 137–141, 143–146, 149–152, 156, 157, 165, 199, 250, 251, 262, 430, 435, 436, 444–447, 453–458, 471, 482, 483, 490–492 Phloroglucinol............................ 245, 251–254, 256, 257 Phyllotaxy ............................................85, 89, 93, 97, 189 Phylogeny ................................................... 42, 59, 60, 84, 85, 88, 90, 99, 100, 102 Physalis ................................................................ 40, 47, 52 Pistil ......................................................18, 59, 62, 64, 65, 70, 74, 116, 119, 241, 242, 244, 247–250, 257, 269, 437, 455

FLOWER DEVELOPMENT: METHODS AND PROTOCOLS

586 Index

PISTILLATA (PI)...................................... 17–22, 28, 45, 46, 67–69, 75, 165, 168, 186, 200, 201, 206, 437 Pollen exine ...............................................202, 207, 209, 210 intine ...................................................... 202, 209, 210 morphology .................................................... 202, 210 mother cell............................................. 199, 225, 226 starch test........................................................ 201, 206 tube .................................................84, 85, 89, 90, 96, 97, 121, 242–244, 247–249, 255, 257, 261 viability............................................................ 199, 206 Pollination ................................................ 52, 65, 99, 112, 116, 247, 257, 279 Polymerase chain reaction (PCR) ...................... 136, 139, 142–145, 147, 149–153, 157–159, 178, 179, 279, 318, 322, 337, 338, 345, 368, 369, 387–390, 397, 398, 403–425, 434, 435, 438, 441, 444–446, 460, 461, 472, 473, 554, 555, 561 Positional cloning ...................................... 137, 138, 140, 142, 143, 146, 149–150, 157 Primer design .................... 367, 388, 404, 414, 415, 424 Primordia ................................................ 9, 10, 13–17, 25, 45, 49, 51, 62, 73–75, 112, 113, 115–117, 164, 165, 174–176, 186, 188, 193, 268, 293, 298, 309, 371, 430, 437, 542, 544–546, 548 Probe hybridization ........................................ 181, 182, 194, 195, 336, 339, 343, 346–348 synthesis ................................................. 180, 332, 338 Programmed cell death (PCD) .................. 200, 203, 209 Promoter .................................................. 6, 8, 22, 25, 49, 68, 246, 247, 286, 296–298, 314, 315, 319, 325, 326, 352, 354, 430–433, 435–440, 442, 447, 556, 557, 559–561 Propidium iodide ....................................... 165, 166, 168, 171, 172, 185–187, 191, 193, 195, 200, 201 Protein –DNA interaction ...........................89, 553, 554, 559 extraction .............................. 314, 497–499, 511, 527 Proteome .................................................... 470, 496, 502, 509, 512, 513, 517 Proteomics.........................................................16, 19, 20, 200, 299, 496, 497, 504, 505, 510, 515, 530 Protoplast ................................................... 294–297, 299, 302–304, 307, 308, 332, 512 Protoplasting ........................................................ 293–299 Pseudo-Schiff reagent .........................165, 166, 168, 193

Q Quantitative image analysis ........................ 538, 546, 558 Quantitative PCR (qPCR).................................. 274, 315, 318, 325, 369, 389, 397, 417, 418, 425 Quantitative real-time PCR (qRT-PCR) ..................... 404 Quantitative trait loci (QTL) ............................. 454, 456, 457, 460–461, 471–485, 492

R Real-time PCR (RT-PCR)..................318, 409, 421, 422 Real-time quantitative PCR (RT-qPCR) ............ 331, 348 Reference gene ...................................404–406, 408, 410, 413, 415, 422, 423, 568, 573, 574, 579 Replum .......................................119, 120, 241, 242, 248 Reporter gene......................................... 49, 351, 403, 568, 579 line ........................................................ 244, 246, 255, 298, 315, 316, 319, 325, 354, 362, 431 transcriptional reporter .................................. 332, 352 translational reporter .............................................. 352 Reverse genetics .............................................41, 366, 430 Reverse phase chromatography .......................... 498, 501, 503, 504, 521, 523–528 Reverse transcription (RT) ........................ 144–148, 154, 215, 297, 319–322, 336, 347, 387, 403, 404, 411, 416–418, 425 Rib meristem ................................................................. 294 Riboprobes ..........................................178–182, 191, 195 Rice ......................................................... 59–76, 164, 199, 200, 206, 219, 353, 354, 405, 440, 513 Right border (RB) .....................136, 139, 433, 435, 446 RNA extraction ....................................................... 216, 277, 295, 365, 366, 371, 393, 395, 411, 416, 462, 485, 498, 499 in situ hybridization.............. 87, 164, 165, 167–169, 175–184, 331–349 integrity ................................................. 297, 382, 404 isolation ......................................................... 297, 367, 371–381, 395, 411, 423 polymerase ................... 167, 179, 191, 192, 366, 440 quality ................................... 367, 372, 377, 416, 485 RNA-seq ................................................... 92, 93, 96, 301, 307, 314, 331, 332, 366, 370, 371, 377, 382–384, 395, 396, 405, 457, 458, 460, 462, 485–492, 495, 515, 516, 518, 519, 529, 569 Robustness................................12, 63, 69, 131, 165, 538

S Safranin ........................................................ 245, 254, 257 Scanning electron microscopy (SEM) ............... 164, 165, 170, 187–188, 197, 201, 207, 242, 251 Sectioning ...........................................164, 166, 172–178, 194, 208, 215, 253, 257, 264–265, 271, 272, 274, 279, 332, 335, 340–342, 355, 357–358 Seedling ................................................5, 8, 94, 149, 165, 172, 174–176, 194, 221, 290, 308, 309, 313, 369, 391, 393, 439, 444, 504, 519 Seed......................................................... 8, 48, 62, 87–92, 94, 95, 99, 100, 120, 131–134, 136, 141, 142, 144, 145, 149, 152, 156, 165, 170, 171, 192,

FLOWER DEVELOPMENT: METHODS 221, 242, 245, 254, 261, 262, 272, 274, 288, 290, 302, 309, 365, 435, 444, 445, 447, 455, 512 Semi-thin section ........................................ 200, 201, 207 SEPALLATA (SEP).................................... 18, 23, 43, 46, 49, 50, 63, 67–69, 72, 74, 76, 515, 519, 529 Sepal ............................................................ 17–21, 25, 26, 42–49, 51, 52, 59, 60, 65–67, 69, 83, 97, 111–115, 133, 135, 164, 186, 188–190, 207, 247, 437 Septum.................................................119, 121, 241, 242 Sequencing deep................................................................. 140, 141 Sexual reproduction ............................................. 119, 219 SHOREMAP................................................................. 140 Simple sequence length polymorphism (SSLP) ....................................................... 137, 150 Single-cell RNA-Seq (scRNA-seq)..................... 301–304, 307, 308, 332, 567–579 Single-nucleus RNA-Seq (snRNA-seq) .............. 307, 308 Solanaceae...........................40–41, 44–54, 378, 391–393 Soybean............................................... 219–222, 225–228, 236, 353, 360, 518 Spikelet ............................................................. 59–76, 237 Sputter coating .............................................................. 188 Stage-specific ................................................................... 94 Stamen ........................................................ 17, 18, 22, 23, 26, 28, 42, 43, 45–47, 64, 67–69, 115–118, 122, 132, 189, 190, 437 Stem cell ..................................................... 13, 25, 28, 51, 113, 133, 163, 164, 293, 294, 298, 326, 537, 538, 547 Stigma .............................................. 62, 65, 68, 120, 121, 241, 242, 246–249, 251 Streptavidin........................................................... 317, 322 Style.........................................................62, 97, 120, 121, 241–243, 247, 249–251, 257, 263–265 Sulfadiazine.................................................................... 433 Suppressor .................................... 3, 6, 12, 131–160, 298 SYBR Green ............................................... 318, 325, 369, 389, 409, 412 Synchronous ................................................ 237, 291, 292

T Tapetum...................................................... 116–118, 199, 200, 202, 204–210, 214, 437 TaqMan ....................................................... 409, 414, 419 T-DNA insertion................................................ 136, 137, 139, 141, 361, 430, 445 insertional mutagenesis................................. 136, 139, 143, 147–149 integration ............................................................... 136 Tetrad................................. 200, 210, 270, 274, 275, 277 The Arabidopsis Information Resource (TAIR) ...................................................... 470, 575

AND

PROTOCOLS Index 587

Thermal Asymmetric Interlaced Polymerase Chain Reaction (TAIL-PCR).............137, 139, 143, 151 Tissue clearing...........................................171, 185, 243, 248 collection ....................................................... 289, 291, 302, 366, 371, 410, 416 dehydration .......................................... 171, 173, 176, 185, 187, 252, 253 dissection ................................................................. 172 embedding............................................ 173, 175, 176, 274, 332, 335, 340–342, 355 fixation .................................................. 172, 175, 176, 184, 187, 252, 274 mounting .............................. 176–178, 186, 188, 355 sectioning...............................................173, 176–178, 194, 253, 274, 335, 340–342, 355 staining ................................................. 171, 173, 185, 186, 253–255 Toluidine blue ............................................ 166, 173, 191, 202, 208, 215, 245, 254, 255, 257 Tomato ............................................... 40–44, 46–52, 164, 308, 309, 354, 406, 407, 437, 440, 510, 513, 519 Transcript detection .................................................................. 195 Transcription factor (TF)..............................3, 42–44, 47, 67, 70, 73, 87, 115, 133, 262, 286, 288, 294, 325, 431–433, 436, 445, 495, 537, 538, 553, 559 Transcriptome ............................................ 21, 29, 88, 92, 93, 99, 220, 301, 307, 314, 366, 370, 373, 374, 376, 378–380, 382–384, 386, 396, 398, 468, 510, 514–516, 519, 529, 567 Transcriptomics ...................................... 5, 29, 88, 92–94, 97, 99, 200, 263, 266, 274–279, 314, 332, 366, 408, 454, 456, 457, 460–462, 469, 485–492, 495, 497, 510, 515, 516, 529 Transfer-DNA (T-DNA) ................... 135–137, 139–141, 143, 147–149, 361, 430, 441, 445, 446 Transformation.....................................21, 87, 91, 94, 96, 97, 99, 136, 137, 141, 143, 147, 158, 286, 288–290, 292, 316, 353, 354, 391, 398, 429, 436, 441, 447 Transmission electron microscopy (TEM) .................204, 205, 210, 214 Transmitting tract ...............................119, 121, 242, 256 Transposable element .......................................... 353, 354 Trithuria............................................................. 90, 92, 95 TRIzol.........................................367, 371, 381, 395, 498 TUNEL assay .............................................. 203, 204, 209 Two component system.............................. 430, 431, 436

U Ultrafiltration ............................. 520, 522–523, 527, 528 Ultrathin section ........................ 200, 204–205, 210, 214 Unisexual ................................................... 83, 97, 99, 102

FLOWER DEVELOPMENT: METHODS AND PROTOCOLS

588 Index V

W

Valve............................................119, 120, 122, 241, 251 Vascular clearing..................................................................... 243 development ........................... 65, 115, 243, 248, 250 Virus induced gene silencing (VIGS) ............47, 87, 366, 369–370, 392

Whole-mount ............................. 164, 251–252, 254, 352 WUSCHEL............................ 13, 51, 262, 278, 298, 537

X X-gluc .......................................................... 352, 355, 360 XVE system ................................................................... 432 Xylem ...................................................243, 253, 298, 512