The Plant Cytoskeleton: Methods and Protocols 1071628666, 9781071628669

Table of contents :
Preface
Contents
Contributors
Chapter 1: Quantification of Microtubule-Bundling Activity of MAPs Using TIRF Microscopy
1 Introduction
2 Materials
2.1 Tubulin Preparation
2.2 Microtubule-Associated Proteins (MAPs)
2.3 Special Chemicals
2.4 Buffers
2.5 Materials for Imaging Microtubules
2.6 Special Equipment
3 Methods
3.1 Preparation of Tubulin
3.2 Tubulin Polymerization Mixtures
3.3 Production of Microtubules
3.4 Optimization Steps for the Bundling Assays
3.5 Bundling Assay with MAPs
3.6 Quantification of Microtubule-Microtubule Interactions
4 Notes
References
Chapter 2: Actin: Static and Dynamic Studies
1 Introduction
2 Materials
2.1 Microfilament Binding and Bundling Assays
2.1.1 Co-sedimentation Assay for Determining Microfilament Binding and Bundling
2.1.2 Microfilament Visualization by Staining with Fluorescent Phalloidin
2.2 Single-Molecule TIRF Imaging Assays for Visualizing Actin Polymerization
2.2.1 Preparation of Biotinylated Actin
2.2.2 Preparation of Oregon Green-Labeled Actin
2.2.3 Preparation of SNAP-549-AtFH14-FH1-FH2
2.2.4 Slide Treatment
2.2.5 Preparation of Flow Chamber
2.2.6 Flow Chamber Treatment before Imaging
2.2.7 Microscopy
3 Methods
3.1 High-Speed Co-sedimentation Assays for Determining Microfilament Binding
3.2 Low-Speed Co-sedimentation Assay for Determining Microfilament Bundling
3.3 Microfilament Visualization by Staining with Fluorescent Phalloidin
3.3.1 Cover Glass Surface Treatment
3.3.2 Sample Preparation Before Imaging
3.3.3 Fluorescence Microscopic Analysis of Microfilaments
3.4 Single-Molecule TIRF for Visualizing Actin Polymerization
3.4.1 Preparation of Biotinylated Actin
3.4.2 Preparation of Oregon Green Actin
3.4.3 SNAP-AtFH14-FH1-FH2 Labeling
3.4.4 Slide Surface Treatment
3.4.5 Preparation of Flow Chamber
3.4.6 Flow Chamber Treatment Before Imaging
3.4.7 TIRF Microscopy and Kymograph Analysis
4 Notes
References
Chapter 3: 3D Visualization of Microtubules in Epidermal Pavement Cells
1 Introduction
2 Materials
2.1 Plant and Plant Growth Material
2.2 Fluorescence Labeling and Sample Mounting
2.3 Microscopy and Image Processing
3 Methods
3.1 Seed Sterilization and Stratification
3.2 Preparation of MS Growth Medium
3.3 Seed Germination and Growth
3.4 Preparation of Secondary Fluorescent Labeling Solution
3.5 Secondary Fluorescent Labeling of Arabidopsis Seedlings
3.6 Mounting Samples on Slides for Microscopical Observation
3.7 Confocal Microscopy for Imaging Microtubules in Epidermal Pavement Cells
3.8 2D Image Analysis by Projecting the 3D Data on a Plane
3.9 3D Image Analysis on the Spatial Distribution of Microtubules
3.10 Potential for Artificial Intelligence-Based Post-Acquisition Image Enhancement in the Interpretation of the Subcellular O...
4 Notes
References
Chapter 4: Quantitative Analysis of Microtubule Organization in Leaf Epidermis Pavement Cells
1 Introduction
2 Materials
2.1 Plant and Microscopy Material
2.2 Software Tools
3 Methods
3.1 Imaging of Cell Contours and Microtubules in Pavement Cells
3.2 Cell Segmentation and Quantification of Pavement Cell Shape Features
3.3 Quantification of Microtubule Organization Within Defined Regions of Interest
3.4 Quantification of Microtubule Enrichment at Necks and Lobes
4 Notes
References
Chapter 5: Single-Cell Confinement Methods to Study Plant Cytoskeleton
1 Introduction
2 Materials
2.1 Plant Material
2.2 Media
2.3 Callus and Protoplast Generation
2.4 Microwell Fabrication
2.5 Confinement in the Microwell
2.6 Microscopy
2.7 For Image Analysis
3 Methods
3.1 Seedling Growth
3.2 Protoplast Preparation from Seedlings
3.3 Protoplast Preparation from Calli
3.4 Microwell Mold Fabrication
3.5 Agarose Microwell Fabrication and Protoplast Confinement
3.6 NOA73 Microwell Preparation and Protoplast Confinement
3.7 Imaging
3.8 Quantification of Cytoskeleton Organization: Average Cytoskeleton Organization
3.9 Quantification of Cytoskeleton Organization: Local Cytoskeleton Organization
4 Notes
References
Chapter 6: Documentation of Microtubule Collisions with Myosin VIII ATM1 Containing Membrane-Associated Structures
1 Introduction
2 Materials
2.1 Seeds and Plant Growth Materials
2.2 Plasmids and Agrobacterium Strains
2.3 Media, Buffers, and Solutions
2.4 Microscopy and Image Analysis
3 Methods
3.1 Nicotiana benthamiana Plant Growth
3.2 Culturing and Preparation of the A. tumefaciens Suspension
3.3 Leaf Selection, Agroinfiltration, and Plant Incubation
3.4 In Planta Localization of Proteins
3.5 Image Analysis with Imaris
3.6 The Simulation Tool
3.7 Using the Simulation Tool
3.8 The Simulation Measurements File
4 Notes
References
Chapter 7: Imaging the Plant Cytoskeleton by High-Pressure Freezing and Electron Tomography
1 Introduction
2 Materials
2.1 Plant Material
2.2 High-Pressure Freezing and Freeze Substitution
2.3 Preparation of Grids and Sections for Electron Tomography
2.4 Image Acquisition and Software
3 Methods
3.1 High-Pressure Freezing
3.2 Freeze Substitution and Resin Embedding
3.3 Grid Preparation
3.4 Section Staining for Electron Tomography
3.5 Gold Particle Application
3.6 Data Collection: Acquiring Data for a Double-Axis Tomogram
3.7 Data Collection: Acquiring Data for a Montaged, Double-Axis Tomogram
3.8 Calculation and Segmentation of Tomograms
3.9 Tomogram Segmentation
4 Notes
References
Chapter 8: Confocal Microscopic Assays of Mitotically Active Proteins in an Agrobacterial Infiltration-Based, Cell Division-En...
1 Introduction
2 Materials
2.1 Plant Materials
2.2 Plasmids and Agrobacterium Strain
2.3 Agrobacterial Infiltration
2.4 Microscopy Materials
3 Methods
3.1 Growth and Preparation of an Agrobacterial Suspension for Infiltration
3.2 Live-Cell Imaging with Confocal Fluorescence Microscopy
4 Notes
References
Chapter 9: Assessment of Spindle Shape Control by Spindle Poleward Flux Measurements and FRAP Bulk Analysis
1 Introduction
2 Materials
2.1 Plant Material
2.2 Plant Growth
2.3 Sample Preparation
2.4 Microscope and Software
3 Methods
3.1 Plant Material and Growth Conditions
3.2 Sample Preparation
3.3 FRAP Imaging
3.4 Kymograph Analysis of Spindle Flux
3.5 Fluorescence Recovery After Photobleaching Analysis
4 Notes
References
Chapter 10: Expansion Microscopy of Plant Cells (PlantExM)
1 Introduction
2 Materials
2.1 For Tobacco Bright Yellow 2 (BY2) Cell Maintenance
2.2 For Fixation and Immunofluorescence of BY2 Cells
2.3 For Gelation and Expansion
2.4 Microscopy
2.5 Fixation of Arabidopsis Seedlings
3 Methods
3.1 Maintenance of BY2 Tobacco Cell Culture
3.2 Fixation of BY2 Cells and Immunofluorescence
3.3 Gelation, Digestion, and Expansion of Cultured Cells
3.4 Arabidopsis Seedling Sterilization and Growth
3.5 Fixation of Arabidopsis Seedlings and Immunofluorescence
3.6 Gelation, Digestion, and Expansion of Seedlings
3.7 Confocal Microscopy and Analysis
3.8 Lightsheet Microscopy and Analysis
4 Notes
References
Chapter 11: Microfluidic Device for High-Resolution Cytoskeleton Imaging and Washout Assays in Physcomitrium (Physcomitrella) ...
1 Introduction
2 Materials
2.1 Microdevice Fabrication and Bonding to the Glass-Bottom Dish
2.2 Moss Culture in the Microdevice
2.3 Microscopy
2.4 Washout Assays
3 Methods
3.1 Microdevice Fabrication
3.2 Culture of P. patens Protonema Cells in the Microdevice
3.3 Introducing Gametophore Leaf Cells into the Microdevice
3.4 Oblique Illumination Fluorescent Imaging of Microtubules
3.5 Performing Washout Assays During Imaging Using the Microdevice
4 Notes
References
Chapter 12: Using Spinning Disk Microscopy to Observe the Mitotic and Cytokinetic Apparatus in Physcomitrium patens
1 Introduction
2 Materials
2.1 Plants
2.2 Culture Media
2.2.1 Stock Solutions for Culture Medium
2.2.2 Preparation of Culture Media
2.2.3 Reagents and Instruments for Routine Subculture
2.3 Light and Temperature for Moss Culture
2.4 Microscope and Optics
3 Methods
3.1 Routine Subculture of Protonemata
3.2 Preparation of Protonemata Using a Glass-Bottom Dish for Imaging
3.3 Application of Chemicals to Protonemata Grown in a Thin Layer of Medium
3.4 Time-Lapse Observation of Mitosis and Cytokinesis
3.5 Analysis of Images
4 Notes
References
Chapter 13: Gaining Insight into Large Gene Families with the Aid of Bioinformatic Tools
1 Introduction
2 Materials
2.1 Hardware and OS
2.2 Software
2.3 Data Sources
3 Methods
3.1 Identifying Protein Sequences of Interest
3.2 Predicting and Revising Protein Sequences
3.3 Domain and Motif Identification
3.4 Alignment Construction and Processing
3.5 Calculation of a Maximum Likelihood Phylogenetic Tree
3.6 Data Presentation
3.7 Some Possible Follow-Up Analyses
4 Notes
References
Untitled
Chapter 14: Cell-to-Cell Connectivity Assays for the Analysis of Cytoskeletal and Other Regulators of Plasmodesmata
1 Introduction
2 Materials
2.1 Cell Connectivity Assay by Microprojectile Bombardment
2.2 Cell Connectivity Assay by Agrobacterium Infiltration
2.3 Imaging and Software
3 Methods
3.1 Cell Connectivity Assay by Microprojectile Bombardment
3.2 Cell Connectivity Assay by Agrobacterium Infiltration
3.3 Imaging and Cell Counting
4 Notes
References
Chapter 15: Studying Nuclear Dynamics in Response to Actin Disruption in Planta
1 Introduction
2 Materials
2.1 Equipment
2.2 Microscopy Consumables
2.3 Image Analysis Tools
2.4 Plant Samples
2.5 Chemicals and Constructs
3 Methods
3.1 Mounting of Samples for Imaging
3.2 Analysis of Circularity Index (CI) to Determine Nuclear Shape
3.3 Quantifying Nuclear Envelope Deformations
3.4 Tracking Nuclear Movement Using Temporal Color-Coded Projections
3.5 Tracking Nuclear Movement Using Kymographs
3.6 Manual Particle Tracking of Nuclear Movement
4 Notes
References
Chapter 16: Cytoskeleton Remodeling in Arabidopsis Stigmatic Cells Following Pollination
1 Introduction
2 Materials
2.1 Plant Material
2.2 Media
2.3 Microscopes
2.4 Other Materials
2.5 Computer Programs Required
3 Methods
3.1 Static View of the Stigmatic Cytoskeleton: Analysis of Cytoskeleton Fiber Orientation
3.2 Static View of the Stigmatic Cytoskeleton: Cytoskeleton Destabilization
3.3 Dynamic View of the Stigmatic Cytoskeleton Remodeling upon Pollination, Using an Inverted Microscope
3.4 Dynamic View of the Stigmatic Cytoskeleton Remodeling upon Pollination, Using an Upright Microscope
4 Notes
References
Chapter 17: Investigation of ROP GTPase Activity and Cytoskeleton Dynamics During Tip Growth in Root Hairs and Pollen Tubes
1 Introduction
2 Materials
2.1 Plant Materials and Transgenic Reporters
2.2 Solutions and Antibodies
2.3 Microscopy and Image Analysis Software
3 Methods
3.1 Pollen Tube Germination In Vitro
3.2 Immunofluorescent Staining of Microtubules in Arabidopsis Pollen Tubes
3.3 Observation of F-Actin Dynamics in Tip Growth Cells
3.4 Preparation of Root Hair Growth Medium
3.5 Measurement of Root Hair Growth Rate
3.6 Quantitative Analysis of the Fluorescence Intensity of Active ROPs (PM/Cytosol Ratio) in Root Hair Growth
4 Notes
References
Chapter 18: Functional Analysis of Phospholipid Signaling and Actin Dynamics: The Use of Apical Growing Tobacco Pollen Tubes i...
1 Introduction
2 Materials
2.1 Plant Material and Microscopy
2.2 Plasmids
2.3 Biolistic Transient Transformation
2.4 Image Analysis Software Required
3 Methods
3.1 Transient Expression of Constructs in Nicotiana Tabacum Pollen Tubes
3.2 Imaging Lipid Kinases Fusion Constructs and Phosphoinositide Biosensors
3.3 Co-expression and Imaging of Actin and Lipid Kinases Fusion Constructs
4 Notes
References
Chapter 19: Microtubule Reorganization During ABA-Induced Stomatal Closure in Arabidopsis
1 Introduction
2 Materials
2.1 Plant Materials and Sample Preparation
2.2 Microscopy
2.3 Software
3 Methods
3.1 Sample Preparation and Live-Cell Imaging
3.2 Quantitative Analysis
3.3 Measurement of Stomatal Apertures
4 Notes
References
Chapter 20: Imaging of Cortical Microtubules in Plants Under Salt Stress
1 Introduction
2 Materials
2.1 Plant Materials and Medium
2.2 Microscopy
3 Methods
3.1 Seedling Treatment
3.2 Image Processing
4 Notes
References
Chapter 21: Analysis of Actin Array Rearrangement During the Plant Response to Bacterial Stimuli
1 Introduction
2 Materials
2.1 Materials
2.2 Reagents
3 Methods
3.1 Preparation of MAMP and Bacterial Working Solutions
3.2 Bacteria or MAMP Treatment of Epidermal Pavement Cells
3.3 MAMP Treatment of Dark-Grown Hypocotyls
3.4 MAMP Treatment of Guard Cells
3.5 Live-Cell Imaging
3.6 Image Process and Analysis of Epidermal Pavement Cells
3.7 Analyzing Images from Epidermal Cells of Dark-Grown Hypocotyls
3.8 Analyzing Images of Guard Cells
4 Notes
References
Chapter 22: Live-Cell Imaging of Cytoskeletal Responses and Trafficking During Fungal Elicitation
1 Introduction
2 Materials
2.1 Bulk Growth of Bgh Powdery Mildew
2.2 Bgh Inoculation of Arabidopsis Leaves
2.3 Mounting Bgh-Infected Arabidopsis Leaves
2.4 Confocal Observation of Appressorium Interaction Sites
2.5 Growing Elongated Hypocotyls for Elicitation with PAMPs
2.6 Exposing Dark-Grown Hypocotyls to PAMP Elicitors
2.7 Mounting Hypocotyls for Live-Cell Imaging
3 Methods
3.1 Bulk Growth of Bgh Powdery Mildew
3.2 Bgh Inoculation of Arabidopsis Leaves
3.3 Mounting Bgh-Infected Arabidopsis Leaves
3.4 Confocal Observation of Appressorium Interaction Sites
3.5 Growing Elongated Hypocotyls for Elicitation with PAMPs
3.6 Exposing Dark-Grown Hypocotyls to PAMP Elicitors
3.7 Mounting Hypocotyls for Live-Cell Imaging
4 Notes
References
Chapter 23: Visualization and Quantification of the Dynamics of Actin Filaments in Arabidopsis Pollen Tubes
1 Introduction
2 Materials
2.1 Plant Materials
2.2 Materials and Chemical Reagents for Pollen Germination
2.3 Microscopy
2.4 Data Analysis Software
3 Methods
3.1 Preparation of Arabidopsis Pollen Germination Medium (PGM)
3.2 Pollen Germination and Pollen Tube Growth
3.3 Image Collection Using Spinning Disk Confocal Microscopy
3.4 Analysis of the Overall Polymerization of Apical Actin Filaments from Plasma Membrane at the Pollen Tube Tip Using Kymogra...
3.5 Analysis of Parameters Associated with the Dynamics of Individual Actin Filaments
4 Notes
References
Chapter 24: Noninvasive Long-Term Imaging of the Cytoskeleton in Arabidopsis Seedlings
1 Introduction
2 Materials
3 Methods
3.1 Setting Up Arabidopsis Seedlings for Growth Inside Imaging Chambers
3.2 Long-Term Imaging of Secondary Wall Formation
3.3 Long-Term Imaging of Seedlings During Salt Stress
3.4 Post-processing of Long-Term Recordings (Drift Correction and Maximum Projections)
4 Notes
References
Chapter 25: Visualization of Cytoskeleton Organization and Dynamics in Elongating Cotton Fibers by Live-Cell Imaging
1 Introduction
2 Materials
2.1 Plant Materials
2.2 Solutions and Microscopy Materials
2.3 Microscopy and Computer Program
3 Methods
3.1 Preparation of Cotton Fibers for Live-Cell Imaging
3.2 Image Acquisition
3.3 Image Processing
4 Notes
References
Chapter 26: Methods to Visualize and Quantify Cortical Microtubule Arrays in Arabidopsis Conical Cells
1 Introduction
2 Materials
2.1 Plant Materials and Growth Condition
2.2 Special Chemicals
2.3 Materials for Imaging Conical Cells and Microtubules
2.4 Special Equipment
2.5 Software
3 Methods
3.1 Imaging Conical Cells Using a Table-Top Scanning Electron Microscope
3.2 Imaging Conical Cells Using a Confocal Microscope
3.3 Live Confocal Imaging of Cortical Microtubules in Conical Cells
3.4 Quantification of Cortical Microtubules in Conical Cells
4 Notes
References
Chapter 27: Studying the Organization of the Actin Cytoskeleton in the Multicellular Trichomes of Tomato
1 Introduction
2 Materials
2.1 Plant Materials and Sample Preparation
2.2 Fixation and Actin Staining
2.3 Microscopy and Computer Programs
3 Methods
3.1 Treatment of Cytoskeleton Depolymerizing Drugs
3.2 Preparation of Tomato Trichomes for Live-Cell Imaging
3.3 Tomato Trichome Fixation and Imaging
3.4 Anisotropy Analysis Using FibrilTool
3.5 Cytoskeleton Dynamics Analysis Using Kymograph
4 Notes
References
Chapter 28: Light Microscopy Technologies and the Plant Cytoskeleton
1 Introduction
2 Imaging Modalities
2.1 Laser Scanning Confocal Microscopy (LSCM)
2.2 Super-Resolution Microscopy (SRM)
2.3 Three-Dimensional Structured Illumination Microscopy (3D-SIM)
2.4 Airyscan Laser Scanning Confocal Microscopy
2.5 Lightsheet Fluorescence Microscopy (LSFM)
2.6 Total Internal Reflection Fluorescence Microscopy (TIRFM)
References
Chapter 29: Investigating Plant Protein-Protein Interactions Using FRET-FLIM with a Focus on the Actin Cytoskeleton
1 Introduction
2 The Principles of FRET-FLIM
3 The Use of FRET-FLIM to Study Protein Interactions in Plants
4 Recent Developments in FRET-FLIM and Their Application to Plant Systems
5 Conclusion
References
Index

Citation preview

Methods in Molecular Biology 2604

Patrick J. Hussey Pengwei Wang Editors

The Plant Cytoskeleton Methods and Protocols

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.

The Plant Cytoskeleton Methods and Protocols

Edited by

Patrick J. Hussey Department of Biosciences, University of Durham, Durham, UK

Pengwei Wang College of Horticulture and Forestry Sciences, Huazhong Agricultural University, Wuhan, China

Editors Patrick J. Hussey Department of Biosciences University of Durham Durham, UK

Pengwei Wang College of Horticulture and Forestry Sciences Huazhong Agricultural University Wuhan, China

ISSN 1064-3745 ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-0716-2866-9 ISBN 978-1-0716-2867-6 (eBook) https://doi.org/10.1007/978-1-0716-2867-6 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023 This work is subject to copyright. All rights are solely and exclusively licensed 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 The plant cytoskeleton is composed of actin filaments and microtubules and a plethora of associated proteins that serve to anchor, crosslink, or otherwise regulate the network with the cell. The cytoskeleton is highly dynamic and functions in many cellular processes including cell division, cell expansion, and organelle movement. As a consequence, the cytoskeleton plays fundamental roles in the coordination of many biological events including fertilization—the passage of pollen tubes down the style; light perception—the movement of chloroplasts in the cell; hormone transport—the recycling of hormone receptors at cell membranes; and transpiration—the opening and closing of stomata, to name but a few. To be able to perform these roles, the cytoskeleton must be able to remodel and reorganize, and this is coordinated through signaling pathways, often specific to plant cells, that can respond to internal cues and external biotic and abiotic stresses. The plant cytoskeleton is quantitatively and qualitatively different from its animal and fungal counterparts, although there are significant areas of overlap and these differences are likely because plants have had over 600 million years to evolve their own unique processes for responding to their own developmental and environmental cues. Over the 29 chapters in this book, the reader is guided to the extensive literature on the plant cytoskeleton and, in particular, to the development of technologies and protocols that are currently being used to understand the nature and activities of the plant cytoskeleton. In this volume, we have chapters on technical protocols, case studies, and a few minireviews. A focus for many of the chapters is on sample preparation, and this is because the quality of your plant organ/tissue preparation (from single to multicellular samples) determines the quality of the data you will obtain. So very often, published research papers miss out the tiny details or the “lab tips” in the materials and methods sections that will ultimately lead to quality data. The volume starts with in vitro experiments using purified cytoskeletal components (Chaps. 1 and 2), followed by live cell imaging of the interphase array (Chap. 3), its quantitation (Chap. 4), the effects on its self-organization in single cells without walls that are geometrically constrained in microwells (Chap. 5), and the analysis of microtubule-membrane protein collisions (Chap. 6). Electron tomography to study finer details of these structural elements is described in the next chapter (Chap. 7). Mitotic and cytokinetic arrays are the focus of Chaps. 8, 9, and 10 with the development of a transiently induced cell-division-enabled leaf system (Chap. 8), analysis of spindle microtubule flux using FRAP (Chap. 9), and the application of expansion microscopy to visualize the different microtubule arrays (Chap. 10). Physcomitrella is studied in Chaps. 11 and 12, with the development of a microfluidic device for short- and long-term oblique illumination fluorescence microscopy imaging of the cytoskeleton, and spinning disk microscopy of mitotic and cytokinetic arrays, respectively. In the near middle of the book, we have a break from the analysis of cytoskeletal arrays and enter the complex world of bioinformatics and the analysis of large cytoskeletal gene families (Chap. 13). The following two chapters detail experiments that can be used to study the effects of disruption of the cytoskeleton on plasmodesmata, the cell junctions of plants (Chap. 14), and the nucleus, in particular, its shape organization and dynamics (Chap. 15). In Chaps. 16, 17, 18, 19, 20, 21, and 22, the techniques used to study the effects of internal and external signals on the remodeling of the cytoskeleton are detailed: pollination on stigmatic cells (Chap. 16), ROP GTPase activity

v

vi

Preface

and phospholipid signaling in tip growing cells (Chaps. 17 and 18, respectively), effects of ABA on microtubule organization in stomatal closure (Chap. 19), salt stress (Chap. 20), bacterial stimuli (Chap. 21), and fungal elicitation (Chap. 22). Chaps. 23, 24, 25, 26, and 27 detail protocols for visualizing the cytoskeleton in different cells, tissues, and different plants: pollen (Chap. 23), non-invasive long-term imaging in seedlings (Chap. 24), cotton fibers (Chap. 25), petal conical cells (Chap. 26), and the multicellular trichomes of tomato (Chap. 27). The remaining two chapters are mini reviews focusing on some of the frequently used microscopy modalities used to image the plant cytoskeleton (Chap. 28) and then a short review on the developing technologies in studying protein-protein interactions in living cells (Chap. 29). Finally, we would like to thank all the authors in this volume. Thank you for your hard work and timely submissions. When we first decided to edit such a volume, we had in mind that this would be particularly useful for early career scientists, but, upon reading all of these protocols, case studies, and reviews, it is clear to us that this is a volume that should be in all the laboratories of those interested or starting to be interested in plant cell and molecular biology research. Durham, UK Wuhan, China

Patrick J. Hussey Pengwei Wang

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

1 Quantification of Microtubule-Bundling Activity of MAPs Using TIRF Microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sharol Schmidt-Marcec, Austin Ross, and Andrei Smertenko 2 Actin: Static and Dynamic Studies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Huaqiang Ruan, Sha Zhang, Yi Zhang, and Haiyun Ren 3 3D Visualization of Microtubules in Epidermal Pavement Cells. . . . . . . . . . . . . . . Amir J. Bidhendi, Bara Altartouri, and Anja Geitmann 4 Quantitative Analysis of Microtubule Organization in Leaf Epidermis Pavement Cells. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sandra Klemm, Jonas Buhl, Birgit Mo¨ller, ¨ rstenbinder and Katharina Bu 5 Single-Cell Confinement Methods to Study Plant Cytoskeleton . . . . . . . . . . . . . . Pauline Durand-Smet, Antoine Chevallier, Le´ia Colin, Alice Malivert, Isaty Melogno, and Olivier Hamant 6 Documentation of Microtubule Collisions with Myosin VIII ATM1 Containing Membrane-Associated Structures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Eduard Belausov, Vikas Dwivedi, Sela Yechezkel, Sefi Bar-Sinai, and Einat Sadot 7 Imaging the Plant Cytoskeleton by High-Pressure Freezing and Electron Tomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Janice Pennington and Marisa S. Otegui 8 Confocal Microscopic Assays of Mitotically Active Proteins in an Agrobacterial Infiltration-Based, Cell Division-Enabled Leaf System of Tobacco (Nicotiana benthamiana) . . . . . . . . . . . . . . . . . . . . . . . . . . Yuh-Ru Julie Lee, Calvin H. Huang, and Bo Liu 9 Assessment of Spindle Shape Control by Spindle Poleward Flux Measurements and FRAP Bulk Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ¨ ller Sabine Mu 10 Expansion Microscopy of Plant Cells (PlantExM) . . . . . . . . . . . . . . . . . . . . . . . . . . . Timothy J. Hawkins, Joanne L. Robson, Bethany Cole, and Simon J. Bush 11 Microfluidic Device for High-Resolution Cytoskeleton Imaging and Washout Assays in Physcomitrium (Physcomitrella) patens . . . . . . . . . . . . . . . . Mari W. Yoshida and Elena Kozgunova 12 Using Spinning Disk Microscopy to Observe the Mitotic and Cytokinetic Apparatus in Physcomitrium patens . . . . . . . . . . . . . . . . . . . . . . . . . Yuji Hiwatashi and Takashi Murata

vii

v xi

1 13 25

43

63

77

89

103

113 127

143

159

viii

13

14

15

16

17

18

19

20

21

22

23

24

25

26

Contents

Gaining Insight into Large Gene Families with the Aid of Bioinformatic Tools. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fatima Cvrcˇkova´ and Radek Bezvoda Cell-to-Cell Connectivity Assays for the Analysis of Cytoskeletal and Other Regulators of Plasmodesmata. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Zoe Barr and Jens Tilsner Studying Nuclear Dynamics in Response to Actin Disruption in Planta. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Joseph F. McKenna and Katja Graumann Cytoskeleton Remodeling in Arabidopsis Stigmatic Cells Following Pollination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lucie Riglet and Isabelle Fobis-Loisy Investigation of ROP GTPase Activity and Cytoskeleton Dynamics During Tip Growth in Root Hairs and Pollen Tubes. . . . . . . . . . . . . . . Lei Zhu and Ying Fu Functional Analysis of Phospholipid Signaling and Actin Dynamics: The Use of Apical Growing Tobacco Pollen Tubes in a Case Study . . . . . . . . . . . Teresa Braga, Fernando Vaz Dias, Marta Fratini, Susana Serrazina, Ingo Heilmann, and Rui Malho Microtubule Reorganization During ABA-Induced Stomatal Closure in Arabidopsis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Liru Dou, Xiangfeng Wang, and Tonglin Mao Imaging of Cortical Microtubules in Plants Under Salt Stress . . . . . . . . . . . . . . . . Shuwei Wang, Liyuan Xu, Changjiang Li, Lei Zhu, Ying Fu, and Yan Guo Analysis of Actin Array Rearrangement During the Plant Response to Bacterial Stimuli. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bingxiao Wang, Minxia Zou, Qing Pan, and Jiejie Li Live-Cell Imaging of Cytoskeletal Responses and Trafficking During Fungal Elicitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Amber J. Connerton, Stefan Sassmann, and Michael J. Deeks Visualization and Quantification of the Dynamics of Actin Filaments in Arabidopsis Pollen Tubes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Qiaonan Lu, Xiaonan Liu, Xiaolu Qu, and Shanjin Huang Noninvasive Long-Term Imaging of the Cytoskeleton in Arabidopsis Seedlings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Felix Ruhnow, Staffan Persson, and Rene´ Schneider Visualization of Cytoskeleton Organization and Dynamics in Elongating Cotton Fibers by Live-Cell Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . Guangda Wang, Yanjun Yu, and Zhaosheng Kong Methods to Visualize and Quantify Cortical Microtubule Arrays in Arabidopsis Conical Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Xie Dang, Lilan Zhu, Huibo Ren, and Deshu Lin

173

193

203

215

227

237

249 257

263

271

285

297

311

317

Contents

27

28 29

ix

Studying the Organization of the Actin Cytoskeleton in the Multicellular Trichomes of Tomato. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327 Zhijing Xu, Xiaolu Qu, Shuang Wu, and Pengwei Wang Light Microscopy Technologies and the Plant Cytoskeleton. . . . . . . . . . . . . . . . . . 337 Timothy J. Hawkins Investigating Plant Protein–Protein Interactions Using FRET-FLIM with a Focus on the Actin Cytoskeleton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353 Patrick Duckney and Patrick J. Hussey

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

367

Contributors BARA ALTARTOURI • Department of Plant Science, McGill University, Ste-Anne-deBellevue, QC, Canada ZOE BARR • Biomedical Sciences Research Complex, University of St Andrews, Fife, UK SEFI BAR-SINAI • The Institute of Plant Sciences, Volcani Institute ARO, Rishon LeZion, Israel EDUARD BELAUSOV • The Institute of Plant Sciences, Volcani Institute ARO, Rishon LeZion, Israel RADEK BEZVODA • Department of Experimental Plant Biology, Faculty of Science, Charles University, CZPrague, Czechia AMIR J. BIDHENDI • Department of Plant Science, McGill University, Ste-Anne-deBellevue, QC, Canada TERESA BRAGA • Universidade de Lisboa, Faculdade de Cieˆncias de Lisboa, BioISI, Lisbon, Portugal JONAS BUHL • Department of Molecular Signal Processing, Leibniz Institute of Plant Biochemistry (IPB), Halle (Saale), Germany KATHARINA BU¨RSTENBINDER • Department of Molecular Signal Processing, Leibniz Institute of Plant Biochemistry (IPB), Halle (Saale), Germany SIMON J. BUSH • Department of Biosciences, Durham University, Durham, UK ANTOINE CHEVALLIER • Laboratoire de Reproduction et De´veloppement des Plantes, Universite´ de Lyon, ENS de Lyon, UCBL, INRAE, CNRS, Lyon Cedex 07, France BETHANY COLE • Department of Biosciences, Durham University, Durham, UK LE´IA COLIN • Laboratoire de Reproduction et De´veloppement des Plantes, Universite´ de Lyon, ENS de Lyon, UCBL, INRAE, CNRS, Lyon Cedex 07, France AMBER J. CONNERTON • Biosciences, University of Exeter, Exeter, UK FATIMA CVRCˇKOVA´ • Department of Experimental Plant Biology, Faculty of Science, Charles University, CZPrague, Czechia XIE DANG • Plant Proteomic Research Center, Fujian Agriculture and Forestry University, Fuzhou, China MICHAEL J. DEEKS • Biosciences, University of Exeter, Exeter, UK FERNANDO VAZ DIAS • Universidade de Lisboa, Faculdade de Cieˆncias de Lisboa, BioISI, Lisbon, Portugal LIRU DOU • State Key Laboratory of Plant Physiology and Biochemistry; Department of Plant Sciences, College of Biological Sciences, China Agricultural University, Beijing, China PATRICK DUCKNEY • Department of Biosciences, Durham University, Durham, UK PAULINE DURAND-SMET • Laboratoire Matie`re et Syste`mes Complexes, Unite´ Mixte de Recherche 7057, CNRS and Universite´ Paris Cite´, Paris cedex 13, France VIKAS DWIVEDI • The Institute of Plant Sciences, Volcani Institute ARO, Rishon LeZion, Israel ISABELLE FOBIS-LOISY • Laboratoire Reproduction et Developpement des Plantes, Universite de Lyon, ENS de Lyon, CNRS, INRAE, Lyon, France MARTA FRATINI • Institute of Biochemistry and Biotechnology/Plant Biochemistry, MartinLuther-University Halle-Wittenberg, Halle (Saale), Germany

xi

xii

Contributors

YING FU • State Key Laboratory of Plant Physiology and Biochemistry, College of Biological Sciences, China Agricultural University, Beijing, China ANJA GEITMANN • Department of Plant Science, McGill University, Ste-Anne-deBellevue, QC, Canada; ECP3-Multi-Scale Imaging Facility, McGill University, SainteAnne-de-Bellevue, QC, Canada KATJA GRAUMANN • Department of Biological and Medical Sciences, Oxford Brookes University, Oxford, UK YAN GUO • State Key Laboratory of Plant Physiology and Biochemistry, College of Biological Sciences, China Agricultural University, Beijing, China OLIVIER HAMANT • Laboratoire de Reproduction et De´veloppement des Plantes, Universite´ de Lyon, ENS de Lyon, UCBL, INRAE, CNRS, Lyon Cedex 07, France TIMOTHY J. HAWKINS • Department of Biosciences, Durham University, Durham, UK INGO HEILMANN • Institute of Biochemistry and Biotechnology/Plant Biochemistry, MartinLuther-University Halle-Wittenberg, Halle (Saale), Germany YUJI HIWATASHI • School of Food Industrial Sciences, Miyagi University, Sendai, Japan CALVIN H. HUANG • Department of Plant Biology, College of Biological Sciences, University of California, Davis, CA, USA SHANJIN HUANG • Center for Plant Biology, School of Life Sciences, Tsinghua University, Beijing, China PATRICK J. HUSSEY • Department of Biosciences, Durham University, Durham, UK SANDRA KLEMM • Department of Molecular Signal Processing, Leibniz Institute of Plant Biochemistry (IPB), Halle (Saale), Germany ZHAOSHENG KONG • State Key Laboratory of Plant Genomics, Institute of Microbiology, Innovative Academy of Seed Design, Chinese Academy of Sciences, Beijing, China; Shanxi Agricultural University, Taigu, China ELENA KOZGUNOVA • Division of Biological Science, Graduate School of Science, Nagoya University, Nagoya, Aichi, Japan; Institute for Advanced Research, Nagoya University, Nagoya, Japan YUH-RU JULIE LEE • Department of Plant Biology, College of Biological Sciences, University of California, Davis, CA, USA CHANGJIANG LI • State Key Laboratory of Plant Physiology and Biochemistry, College of Biological Sciences, China Agricultural University, Beijing, China JIEJIE LI • Beijing Key Laboratory of Gene Resource and Molecular Development, College of Life Science, Beijing Normal University, Beijing, China; Key Laboratory of Cell Proliferation and Regulation of Ministry of Education, College of Life Science, Beijing Normal University, Beijing, China DESHU LIN • Plant Proteomic Research Center, Fujian Agriculture and Forestry University, Fuzhou, China BO LIU • Department of Plant Biology, College of Biological Sciences, University of California, Davis, CA, USA XIAONAN LIU • School of Life Sciences, Qilu Normal University, Jinan, China QIAONAN LU • Center for Plant Biology, School of Life Sciences, Tsinghua University, Beijing, China RUI MALHO´ • Universidade de Lisboa, Faculdade de Cieˆncias de Lisboa, BioISI, Lisbon, Portugal ALICE MALIVERT • Laboratoire de Reproduction et De´veloppement des Plantes, Universite´ de Lyon, ENS de Lyon, UCBL, INRAE, CNRS, Lyon Cedex 07, France

Contributors

xiii

TONGLIN MAO • State Key Laboratory of Plant Physiology and Biochemistry; Department of Plant Sciences, College of Biological Sciences, China Agricultural University, Beijing, China JOSEPH F. MCKENNA • School of Life Sciences, University of Warwick, Coventry, UK ISATY MELOGNO • Laboratoire de Reproduction et De´veloppement des Plantes, Universite´ de Lyon, ENS de Lyon, UCBL, INRAE, CNRS, Lyon Cedex 07, France BIRGIT MO¨LLER • Martin Luther University Halle-Wittenberg, Institute of Computer Science, Halle (Saale), Germany SABINE MU¨LLER • Department of Biology, Friedrich-Alexander University ErlangenNuremberg, Erlangen, Germany TAKASHI MURATA • Department of Applied Bioscience, Kanagawa Institute of Technology, Atsugi, Japan MARISA S. OTEGUI • Center for Quantitative Cell Imaging, Bock Laboratories, Madison, WI, USA QING PAN • Beijing Key Laboratory of Gene Resource and Molecular Development, College of Life Science, Beijing Normal University, Beijing, China JANICE PENNINGTON • Department of Botany, University of Wisconsin, Madison, WI, USA STAFFAN PERSSON • Copenhagen Plant Science Center (CPSC), Department of Plant and Environmental Sciences, University of Copenhagen, Frederiksberg, Denmark; Joint International Research Laboratory of Metabolic and Developmental Sciences, School of Life Sciences and Biotechnology, Shanghai Jiao Tong University, Shanghai, China XIAOLU QU • Key Laboratory of Horticultural Plant Biology (Ministry of Education), College of Horticulture and Forestry Sciences, Huazhong Agricultural University, Wuhan, China; Hubei Hongshan Laboratory, Wuhan, China HAIYUN REN • Key Laboratory of Cell Proliferation and Regulation Biology of Ministry of Education, College of Life Science, Beijing Normal University, Beijing, China; Center for Biological Science and Technology, Guangdong Zhuhai-Macao Joint Biotech Laboratory, Advanced Institute of Natural Science, Beijing Normal University, Zhuhai, China HUIBO REN • Plant Proteomic Research Center, Fujian Agriculture and Forestry University, Fuzhou, China LUCIE RIGLET • The Sainsbury Laboratory, Bateman Street, CB2 1LR, University of Cambridge, Cambridge, UK JOANNE L. ROBSON • Department of Biosciences, Durham University, Durham, UK AUSTIN ROSS • Institute of Biological Chemistry, Washington State University, Pullman, WA, USA HUAQIANG RUAN • Key Laboratory of Cell Proliferation and Regulation Biology of Ministry of Education, College of Life Science, Beijing Normal University, Beijing, China FELIX RUHNOW • Copenhagen Plant Science Center (CPSC), Department of Plant and Environmental Sciences, University of Copenhagen, Frederiksberg, Denmark EINAT SADOT • The Institute of Plant Sciences, Volcani Institute ARO, Rishon LeZion, Israel STEFAN SASSMANN • Biosciences, University of Exeter, Exeter, UK SHAROL SCHMIDT-MARCEC • Institute of Biological Chemistry, Washington State University, Pullman, WA, USA RENE´ SCHNEIDER • Institute of Biochemistry and Biology, Plant Physiology Department, University of Potsdam, Potsdam, Germany SUSANA SERRAZINA • Universidade de Lisboa, Faculdade de Cieˆncias de Lisboa, BioISI, Lisbon, Portugal

xiv

Contributors

ANDREI SMERTENKO • Institute of Biological Chemistry, Washington State University, Pullman, WA, USA JENS TILSNER • Cell & Molecular Sciences, The James Hutton Institute, Dundee, UK BINGXIAO WANG • Beijing Key Laboratory of Gene Resource and Molecular Development, College of Life Science, Beijing Normal University, Beijing, China GUANGDA WANG • State Key Laboratory of Plant Genomics, Institute of Microbiology, Innovative Academy of Seed Design, Chinese Academy of Sciences, Beijing, China PENGWEI WANG • Key Laboratory of Horticultural Plant Biology (Ministry of Education), College of Horticulture and Forestry Sciences, Huazhong Agricultural University, Wuhan, China; Hubei Hongshan Laboratory, Wuhan, China SHUWEI WANG • State Key Laboratory of Plant Physiology and Biochemistry, College of Biological Sciences, China Agricultural University, Beijing, China XIANGFENG WANG • State Key Laboratory of Plant Physiology and Biochemistry; Department of Plant Sciences, College of Biological Sciences, China Agricultural University, Beijing, China SHUANG WU • College of Horticulture, School of Future Technology, Fujian Agriculture and Forestry University, Fuzhou, China LIYUAN XU • State Key Laboratory of Plant Physiology and Biochemistry, College of Biological Sciences, China Agricultural University, Beijing, China ZHIJING XU • Key Laboratory of Horticultural Plant Biology (Ministry of Education), College of Horticulture and Forestry Sciences, Huazhong Agricultural University, Wuhan, China; Hubei Hongshan Laboratory, Wuhan, China SELA YECHEZKEL • The Institute of Plant Sciences, Volcani Institute ARO, Rishon LeZion, Israel MARI W. YOSHIDA • Division of Biological Science, Graduate School of Science, Nagoya University, Nagoya, Aichi, Japan YANJUN YU • State Key Laboratory of Plant Genomics, Institute of Microbiology, Innovative Academy of Seed Design, Chinese Academy of Sciences, Beijing, China SHA ZHANG • Key Laboratory of Cell Proliferation and Regulation Biology of Ministry of Education, College of Life Science, Beijing Normal University, Beijing, China YI ZHANG • Key Laboratory of Cell Proliferation and Regulation Biology of Ministry of Education, College of Life Science, Beijing Normal University, Beijing, China LEI ZHU • State Key Laboratory of Plant Physiology and Biochemistry, College of Biological Sciences, China Agricultural University, Beijing, China LILAN ZHU • Plant Proteomic Research Center, Fujian Agriculture and Forestry University, Fuzhou, China MINXIA ZOU • Beijing Key Laboratory of Gene Resource and Molecular Development, College of Life Science, Beijing Normal University, Beijing, China

Chapter 1 Quantification of Microtubule-Bundling Activity of MAPs Using TIRF Microscopy Sharol Schmidt-Marcec, Austin Ross, and Andrei Smertenko Abstract Cross-linking of microtubules by microtubule-associated proteins (MAPs) results in the formation of microtubule bundles. It has been shown that a majority of microtubules in interphase plant cells are bundled. Bundling can contribute to maintaining structural stability and sustaining spatial organization of microtubule arrays. While bundling can be readily detected by an electron or fluorescent microscope, quantifying this activity remains technically challenging. Here we describe a method for quantifying microtubule-bundling in vitro using green and red stable microtubules. Furthermore, this method distinguishes between different types of microtubule-microtubule interactions: bundling, annealing, and branching. Our technique can be used to compare bundling activity of different MAPs and generate parameters for modeling their contribution to organization and dynamics of microtubule arrays. Key words Microtubule-associated protein, Bundling, Quantification, TIRF microscopy

1

Introduction Microtubules provide rigid yet dynamic scaffolding for all cellular processes and serve as tracks for intracellular trafficking. Microtubule organization changes in response to developmental and environmental cues. One of the most spectacular examples is the formation of structurally distinct microtubule assemblies (known as arrays) that materialize within the several hours of cell division [1]. The remarkable flexibility of array rearrangement is facilitated by microtubule dynamic instability which is the ability to switch between polymerization and depolymerization phases [2]. During a polymerization phase, subunits of tubulin protein dimers are added onto the microtubule tips, and during a depolymerization phase, tubulin dissociates from the tips. Transition between the phases depends on tubulin availability and is governed by

Authors: Sharol Schmidt-Marcec and Austin Ross have equally contributed to this chapter. Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_1, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

1

2

Sharol Schmidt-Marcec et al.

microtubule-associated proteins (MAPs) that can promote polymerization, depolymerization, or stabilize microtubules by modulating tubulin association or dissociation rates [3]. Another essential role of MAPs is linking adjacent microtubules to form bundles. Electron microscopy analysis demonstrates that microtubule-bundling is common during interphase [4]. Bundling is thought to increase the mechanical resilience of microtubules and reinforce their spatial organization; loss of even several microtubules in the bundles will not affect the overall organization of microtubule arrays. A majority of known plant MAPs, including MAP65, WDL3, WVD2, and MOR1/GEM1, possess bundling activity [5–9]. In vitro analysis suggests that microtubule-bundling can stabilize microtubules by reducing tubulin dissociation from the shrinking tips [10]. However, this conclusion was challenged by in vivo observations showing a similar depolymerization rate of single or bundled microtubules [11]. Each cell produces a number of MAPs with distinct microtubule-bundling activity. For example, MAP65 forms 25 nm bridges between microtubules in vitro [5, 12], whereas WVD2 glues microtubules together without gaps [8]. Both types of bundles are readily observed in cells using electron microscopy [13]. This fact suggests the existence of cooperative or antagonistic relationships between different microtubule-bundling proteins. Another microtubule protein MACET4 facilitates both annealing of microtubules and bundling [14]. Our understanding about why plant cells need different types of bundles and how different bundling activities contribute to constructing microtubule arrays remains limited. The progress in this direction is hindered by the lack of quantitative data for the in vitro bundling activity of different MAPs. Without this information, it is not possible to assess the outcome of in vitro assays with proteins that promote microtubulebundling as well as determine the physiological relevance of the bundling activity. Ultimately, a detailed assessment of microtubulebundling in vitro will contribute to understanding the phenotypes of the corresponding mutants. Here we describe a sensitive method for measuring microtubule-bundling using stable red and green microtubules. We demonstrate the suitability of this method for determining bundling as well as different types of microtubule-microtubule interactions. Our approach can be used to determine domains of MAPs responsible for microtubule-bundling by testing mutants and truncated versions. Furthermore, this method will facilitate both a functional understanding of microtubule-bundling and also for generating parameters for modeling the assembly of microtubule arrays.

Quantification of Microtubule-Bundling

2

3

Materials

2.1 Tubulin Preparation

1. Tubulin: Purified from bovine brain using the high-molarity PIPES buffer method [15] and stored at -80 °C. 2. ATTO 488 NHS-Ester (ATTO-TEC #AD 488–31). 3. ATTO 565 NHS-Ester (ATTO-TEC #AD 565–31). 4. NHS-Rhodamine (Thermo Fisher Scientific #46406).

2.2 MicrotubuleAssociated Proteins (MAPs)

All MAPs were stored at the back of the -80 °C freezer in 10–50 μL aliquots: 1. MAP65–1: Arabidopsis thaliana microtubule-associated protein 65–1, purified as recombinant proteins in E. coli and used as a positive control for microtubule-bundling [12]. 2. MACET4: Purified as recombinant proteins in E. coli and used as an example of a protein of interest [14].

2.3 Special Chemicals

1. Bradford Reagent Bio-Rad Protein Assay Dye (Bio-Rad Laboratories, #5000006). 2. Taxol: 10 mM stock solution in DMSO and stored in 10 μL aliquots at -80 °C.

2.4

Buffers

1. All reagents for preparing buffers should be molecular biology grade or higher. Microtubule-stabilizing buffer (MTSB): 50 mM PIPES, pH 6.8 with KOH, 2 mM EGTA, 2 mM MgSO4 (see Notes 1 and 2). 2. BRB80 buffer: 80 mM PIPES, pH 6.8 with KOH, 1 mM MgCl2, 1 mM EGTA, 1 mM DTT (see Notes 3 and 4). 3. Low pH cushion: 60% (v/v) glycerol in 1× BRB80 buffer (see Note 5). 4. Buffered glycerol: 0.1 M PIPES-KOH, pH 6.8, 4 mM MgSO4, 2 mM EGTA, 90% (v/v) glycerol, 2 mM β-mercaptoethanol (see Note 6). 5. Dialysis buffer: 20% (w/v) glycerol, 50 mM PIPES, pH 6.8, 2 mM EGTA, 2 mM MgSO4, 50 mM KCl. DTT is added as necessary. 6. GTP: 100 mM GTP stock solution in 100 mM PIPES pH 6.8. Store in10 μL aliquots at -80 °C. 7. 0.5% glutaraldehyde in BRB80.

2.5 Materials for Imaging Microtubules

1. Standard microscope slides. 2. 22 × 22 mm, #1 coverslips (VWR Cat. No. 16004-094).

4

Sharol Schmidt-Marcec et al.

2.6 Special Equipment

1. Hitachi Micro Ultracentrifuge CS150NX equipped with a S80AT2 rotor and centrifuge tubes (#S300533A). 2. Total Internal Reflection Fluorescence (TIRF) microscope Leica DMI6000 SD equipped with 100× immersion oil objective lens (HCX 1.47NA oil immersion), 10× ocular lens, and temperature control enclosure. 3. MetaMorph 7.8.6.0 Software for imaging.

3

Methods

3.1 Preparation of Tubulin

1. Purified tubulin from bovine brain was labeled using ATTO 488 NHS-Ester, ATTO 565 NHS-Ester, or NHS-Rhodamine using the published procedure [16] and stored at -80 °C (see Note 7). 2. Thaw non-labeled tubulin on ice in pre-chilled microfuge tubes (see Note 8). 3. Thaw two different colors of labeled tubulin on ice in pre-chilled microfuge tubes (see Note 7). 4. Transfer tubulin to pre-chilled centrifuge tubes (see Note 9). 5. Clarify tubulin and MAP proteins by centrifuging in a pre-chilled S80AT2 rotor for 10 min at 250,000 × g, 4 °C (see Notes 10 and 11). 6. Aspirate tubulin or MAP supernatant to pre-chilled microfuge tubes without disturbing the pellet. Incubate protein on ice for 5 min before measuring concentration. 7. Measure protein concentration using the Bio-Rad Protein Assay Dye following manufacturer recommendations (see Note 12).

3.2 Tubulin Polymerization Mixtures

1. Use a stoichiometric molar ratio of 1:6 ATTO 488 tubulin: Unlabeled tubulin (see Note 13). 2. Determine the total volume of ATTO 488 tubulin, and then calculate the total amount of labeled tubulin in moles. 3. Calculate the volume of unlabeled tubulin required for the desired molar ratio with the labeled tubulin. 4. Add unlabeled tubulin to ATTO 488 tubulin and gently mix via pipetting. Dilute mixture with 1× BRB80 supplemented with DTT to a final tubulin concentration 40 μM and add GTP to a final concentration 1 mM. 5. Incubate reaction on ice for 5 min, transfer to pre-chilled centrifuge tubes, and centrifuge in pre-chilled S80AT2 rotor at 250,000 × g, 2 °C, 5 min.

Quantification of Microtubule-Bundling

5

6. Transfer tubulin supernatant to pre-chilled microfuge tube on ice, make 10-μL aliquots, and snap-freeze in liquid N2. Store at -80 °C. 7. Use the same procedures to make the red tubulin polymerization mixture using stoichiometric molar ratio of 1:4 ATTO 565 tubulin/unlabeled tubulin or 1:4 rhodamine/unlabeled tubulin. 3.3 Production of Microtubules

1. Take one 10-μL aliquot of each green and red tubulin polymerization mixture from -80°, and place on ice. Immediately transfer to 37 °C water bath. Thoroughly mix in 5.0 μL of buffered glycerol pre-warmed in 37 °C water bath by gentle pipetting and mixing with a pipette tip (see Note 14). 2. Incubate reaction in the 37 °C water bath for 20 min to polymerize microtubules. 3. Pre-warm a S80AT2 rotor at 37 °C, add 150 μL of low pH glycerol cushion supplemented with 10 μM Taxol to the centrifuge tubes, and pre-warm in 37 °C water bath for 5 min. 4. Gently dilute the microtubule reaction by adding 75 μL of pre-warmed BRB80 supplemented with 20 μM Taxol (see Note 15). Layer the microtubule suspension over the pre-warmed glycerol cushion from step 3 (see Note 16). 5. Centrifuge reaction for 5 min at 85,000 x g, 35 °C. 6. Aspirate supernatant, and then aspirate glycerol cushion from pelleted microtubules (see Note 17). 7. Gently resuspend the microtubule pellet in 20 μL of warm BRB80 supplemented with 20 μM Taxol, and keep at 37 °C. 8. Determine optimal dilution factor for each stock of microtubules. Dilute 1 μL of microtubules in 49 μL of warm BRB80 supplemented with 20 μM Taxol (Table 1), then place 8 μL of this mix on a #1, 22 × 22 cover slip, and examine under a TIRF microscope. The desired microtubule length is 1 to 5 μm at a density of ca. 50–75 microtubules per field of view (Fig. 1).

Table 1 Example of microtubule dilutions

a

Dilutiona

1:50

1:25

1:75

1:50

1:25

1:75

BRB80 + 20 μM Taxol, (μL)

49

24

74

49

24

74

Red microtubules (μL)

1

1

1

0

0

0

Green microtubules (μL)

0

0

0

1

1

1

Begin with 1:50 dilution, and then based on the imaging results, perform additional dilutions to find optimal density of seeds

6

Sharol Schmidt-Marcec et al.

Fig. 1 Examples of seed dilutions. Individual microtubules can not be resolved at high density. Scale bars are 5 μm

Table 2 Example of reactions for optimizing the ratio between red and green microtubules Reaction

A

B

C

BRB80 + 20 μM Taxol

10 μL

10 μL

10 μL

Dialysis buffer

6 μL

6 μL

6 μL

Green microtubulesa

2 μL of 2:15 stock (1:75 final dilution)

2 μL of 2:18 of stock (1:90 final dilution)

2 μL of 2:20 of stock (1:80 final dilution)

Red microtubulesa

2 μL of 2:10 stock (1:50 final dilution)

2 μL of 2:8 dilution (1:40 final dilution)

2 μL of 2:8 dilution (1:40 final dilution)

a

All ratios shown as microtubules:buffer

Further dilutions of microtubules in BRB80 with 20 μM Taxol may be required to get the desired density (Table 1; see Notes 18 and 19). 3.4 Optimization Steps for the Bundling Assays

1. Make a 20 μL reaction by mixing green and red microtubules using the optimal dilution factors found in the previous subheading (Table 2; see Note 20). 2. Incubate reaction for 5 min at 37 °C. Fix the reaction by adding 10 μL of 0.5% glutaraldehyde in BRB80. Pipet 8 μL of reaction onto the center of a warmed (37 °C) 22 × 22 coverslip, place the coverslip reaction side down on a warmed (37 °C) microscope slide, and image (see Note 21). 3. Acquire red and green channels sequentially for a total of 20 to 30 images at different regions on the coverslip. Sequential images can be acquired by clicking the Configure Acquisition panel in MetaMorph. On the Configure Acquisition pop-up, click the 488 tab to show the green microtubule channel, and then click capture. Repeat with the 561 tab. There will be two images on the screen.

Quantification of Microtubule-Bundling

7

4. To overlay the channel images of the same frame, click on the Display drop-down menu in MetaMorph, and select Color Combine. A new pop-up will appear. For the red component drop-down menu, select the recently acquired 565 image. On the green component drop-down menu, select the recently acquired 488 image. Click Color Combine at the bottom. Save the combined image. Move the slide to a new frame, and take the next sequence of images (see Note 22). 5. Verify that there are approximately 50–75 microtubules per a field of view, and there are similar numbers of red and greed microtubules (see Note 23). 6. If microtubule density and ratio between red and green microtubules are not satisfactory, then adjust the dilution factor of red and/or green microtubules as exemplified in Table 2 (see Note 24). 7. Use the microtubule dilution factor from the optimized control reaction for the remaining experimental reactions in the next subheading. 3.5 Bundling Assay with MAPs

1. Add each component (see Subheading 3.4, step 1) of the bundling assay, and add the MAP last at a final concentration of 0.1 μM, 0.5 μM, 1.0 μM, 1.5 μM, and 2.0 μM (Table 3; see Note 25). 2. Set up a positive control reaction with the classical bundling protein, MAP65–1, at a final concentration of 0.1 μM [5, 12]. 3. Process and image the reactions as described in steps 2–4 of Subheading 3.4. 4. Repeat assays three times using fresh preparations of microtubules and MAPs.

Table 3 Examples of reactions for the bundling assay (see Note 20) 0.1 μM MAP

0.5 μM MAP

1.0 μM MAP

BRB80 + 20 μM Taxol

8 μL

8 μL

8 μL

Dialysis buffer

6 μL

6 μL

6 μL

Green microtubules

2 μL

2 μL

2 μL

Red microtubules

2 μL

2 μL

2 μL

MAP

2 μL of 1 μM stock

2 μL of 5 μM stock

2 μL of 10 μM stock

8

Sharol Schmidt-Marcec et al.

Fig. 2 Impact of MAP concentrations on microtubule-bundling. (a) Microtubule-microtubule interactions at different concentrations of MACET4. Higher concentrations of MACET4 induce artificial microtubule aggregates. Scale bar is 5 μm. (b) Quantification of microtubule alignment at different concentrations of MACET4 using Pearson’s correlation coefficient as the readout. The drop in the values at 2 μM MACET is the consequence of artificial aggregation of microtubules at high concentration of MACET4 3.6 Quantification of MicrotubuleMicrotubule Interactions

1. Bundling of microtubules results in colocalization between green and red channels (Fig. 2a). The degree of colocalization is proportional to the microtubule-bundling and can be quantified as Pearson’s correlation coefficient (PCC) in ImageJ using Colocalization Threshold plugin (Fig. 2b). 2. Although, PCC informs of the degree of bundling, it does not provide any information about the nature of interactions between microtubules. This can be assessed by scoring the frequency of the three major types of microtubule-microtubule interactions, bundling, annealing, and branching, across all acquired fields of view (Fig. 3). 3. For scoring different types of interactions between microtubules, select the lowest concentration of MAP that exhibits a statistically significant increase of colocalization between red and green channels in step 1 (see Note 26). For example, according to Fig. 2, this will be 0.1 μM. These conditions will reveal the most physiologically relevant interactions and avoid artificial microtubule aggregation.

Quantification of Microtubule-Bundling

9

Fig. 3 Different types of microtubule-microtubule interactions. MACET4 causes three types of microtubule interaction, whereas MAP65 induces only microtubule alignment. Scale bars are 5 μm

4. Calculate frequency of each interaction in the sample (see Note 27). 5. Interpretation: A true microtubule-bundling protein is expected to cause significant overlap of green and red channels and higher Pearson’s correlation coefficient relative to the negative control at concentration below 1 μM. Bundling at higher concentrations is most likely not physiological. This outcome must be verified by analysis of the microtubule-microtubule interactions. A protein that induces annealing or branching may not affect Pearson’s correlation coefficient. A greater frequency of a particular microtubule-microtubule interaction event relative to the negative control informs on the specific activity of a given MAP.

4 Notes 1. Always use KOH rather than NaOH to adjust pH. 2. Make the following stock solutions, and store at -20 °C: 1 M PIPES pH 6.8, 1 M MgSO4 7 H20 0.1 M EGTA pH 6.8, 1 M DTT, 0.1 M MgCl2. Avoid multiple few freeze/thaw cycles. 3. Make BRB80 buffer as 5× or 10× stock ahead of time, aliquote, and store at -20 °C. 4. Defrost 1 M DTT stock stored at -20 °C, and add to the 1× BRB80 buffer on the day of the experiment. Do not refreeze BRB80 + DTT solution. We make DTT stock in 0.5 mL aliquots. 5. Make a 60% glycerol mix ahead of time. Warm the glycerol in a 37 °C water bath prior to adding concentrated BRB80 stock for easier handling. Store the final solution at -20 °C. 6. Dissolve PIPES powder in double distilled H2O by adding solid KOH pellets. The solution will clear up as the pH nears 6.8. Add stock solution of MgSO4 and EGTA. Store buffer at -20 °C. Add nine parts of fresh glycerol to one part of the buffer. Add fresh 1.4 μL of β-mercaptoethanol for each 10 mL of buffered glycerol prior to using.

10

Sharol Schmidt-Marcec et al.

7. Store tubulin in 25–50 μL aliquots at a concentration ca. 100 μM. We make droplets by pipetting tubulin solution directly into liquid N2. Droplets are fully frozen once sunk. Use clean tweezers to transfer frozen pellets to microfuge tubes or cryotubes, and store at -80 °C. Tubulin is best stored in a less frequently accessed part of the freezer. Avoid freeze/thaw cycles of tubulin solutions. Aim to use tubulin within 4 h after thawing. 8. When thawing, warm the tube with your hands, and gently flick. When tubulin is about half thawed, return to ice. 9. Avoid warming up of the tubulin solution after thawing as at high concentrations tubulin can polymerize at room temperature. 10. For tubulin volumes lower than 30 μL, add ice-cold MTSB buffer to increase the volume by ca. 25%. This will reduce tubulin losses during centrifugation. Avoid high dilutions if the initial tubulin concentration is low. 11. The pellet can be invisible. Before centrifugation, mark the rim of the tube on the centrifugal side where the pellet will aggregate. 12. We add 2 μL of tubulin or 5 μL of MAP in 1 mL of the protein assay solution. The calibration curve is made with bovine serum albumin. 13. The stoichiometric ratio of labeled to unlabeled tubulin may need optimization for each tubulin batch in the range between 1:3 and 1:6. Low ratio generally results in fast bleaching. 14. Due to viscosity of the buffered glycerol, pay attention to mix the glycerol with the tubulin polymerization solution. Avoid production of bubbles. 15. Avoid vigorous pipetting, and gently mix with a pipette tip. 16. Layer the microtubule suspension over the cushion gently to avoid pipetting the microtubules into the cushion. 17. The goal of this procedure is to avoid contaminating microtubules with free tubulin. Hold the pipette tip on the meniscus while aspirating the top layer of BRB80. Use the same technique to aspirate the glycerol cushion and remove any cushion near the microtubule pellet. Be careful not to disturb the pellet with the pipette tip as it may be invisible. 18. Too high concentration of microtubules will result in poor resolution of individual microtubules under the TIRF microscope. Optimally diluted stock will produce crisp images of individual microtubules under the TIRF microscope.

Quantification of Microtubule-Bundling

11

19. It may require multiple dilutions to achieve the optimal microtubule density. If microtubules are too long, then reduce polymerization time. If microtubules are too dense, then decrease tubulin concentration or increase dilution factors. 20. Add components of the reaction in the following order: BRB80 with 20 μM Taxol, dialysis buffer, and red and green microtubules. 21. Set the exposure time on red and green channel between 200 and 500 ms. Each channel may require adjustment to the exposure time. We image each reaction within 15 min of mounting. 22. Adjust the brightness and contrast on the single-channel images before the color combine step if necessary. 23. High density of microtubules in the reaction may result in stochastic interactions. If necessary, dilution of each of the red and green microtubules might require tweaking to minimize these interactions and achieve an equal proportion of red and green microtubules. 24. As an example in Table 2, reaction A contains green microtubules at a final dilution of 1:75 and red microtubules at a final dilution of 1:50. If green microtubules outnumber red microtubules, then optimization of reaction B was made with more diluted green microtubules (1:90) and less diluted red microtubules (1:40). If this reaction contains an optimal number of red microtubules per field of view but too low density of green microtubules, then reaction C contained the same dilution of red microtubules (1:40) and dilution of green microtubules (1: 80). Repeat adjustments until the optimal outcome is achieved. 25. Make serial dilutions of the MAP working stock so that 2 μL leads to the desired final concentration. For example, to accomplish a 0.1 μM final concentration of MAP in a 20 μL reaction, add 2 μL of 1 μM working stock in a 20 μL reaction. Similarly, a bundling assay with a 0.5 μM final concentration of MAP requires the addition of 2 μL of 5 μM working stock. 26. Higher concentrations of MAPs can induce aggregation of microtubules where individual interactions cannot be resolved (Fig. 2a). 27. Count at least 250 events for each experimental condition. If the number of events is below 250, acquire more images. Individual (free) microtubules regardless of the color are counted as an event. Calculate the frequency of each event as exemplified by the following equation for annealing events: Annealing = 100


annealing events f ree þ bundling þ annealing þ branching

12

Sharol Schmidt-Marcec et al.

Acknowledgments Research reported in this publication was supported by the National Science Foundation CAREER award #1751204 and National Institute of Health NIGMS award #T32GM008336. References 1. Wasteneys GO (2002) Microtubule organization in the green kingdom: chaos or self-order? J Cell Sci 115:1345–1354 2. Mitchison T, Kirschner M (1984) Dynamic instability of microtubule growth. Nature 312:237–242 3. Alfaro-Aco R, Petry S (2015) Building the microtubule cytoskeleton piece by piece. J Biol Chem 290:17154–17162 4. Barton DA, Vantard M, Overall RL (2008) Analysis of cortical arrays from Tradescantia virginiana at high resolution reveals discrete microtubule subpopulations and demonstrates that confocal images of arrays can be misleading. Plant Cell 20:982–994. https://doi.org/ 10.1105/tpc.108.058503 5. Chan J, Jensen CG, Jensen LCW, Bush M, Lloyd CW (1999) The 65-kDa carrot microtubule-associated protein forms regularly arranged filamentous cross-bridges between microtubules. PNAS of USA 96:14931–14936 6. Jiang C-J, Sonobe S (1993) Identification and preliminary characterization of a 65 kDa higher-plant microtubule-associated protein. J Cell Sci 105:891–901 7. Liu X, Qin T, Ma Q, Sun J, Liu Z, Yuan M, Mao T (2013) Light-regulated hypocotyl elongation involves proteasome-dependent degradation of the microtubule regulatory protein WDL3 in Arabidopsis. Plant Cell 25:1740– 1755 8. Perrin RM, Wang Y, Yuen CY, Will J, Masson PH (2007) WVD2 is a novel microtubuleassociated protein in Arabidopsis thaliana. Plant J 49:961–971

9. Yasuhara H, Muraoka M, Shogaki H, Mori H, Sonobe S (2002) TMBP200, a microtubulebundling polypeptide isolated from telophase tobacco BY-2 cells is a MOR1 homologue. Plant Cell Phys 43:595–603 10. Fache VJ, Gaillard J, Van Damme D, Geelen D, Neumann D, Stoppin-Mellet V, Vantard M (2010) Arabidopsis kinetochore fiberassociated MAP65-4 cross-links microtubules and promotes microtubule bundle elongation. Plant Cell 22:3804–3815 11. Shaw SL, Lucas J (2011) Intrabundle microtubule dynamics in the Arabidopsis cortical array. Cytoskeleton 68:56–67 12. Smertenko AP, Chang HY, Wagner V, Kaloriti D, Fenyk S, Sonobe S, Lloyd C, Hauser MT, Hussey PJ (2004) The Arabidopsis microtubule-associated protein AtMAP65-1: molecular analysis of its microtubule-bundling activity. Plant Cell 16:2035–2047 13. Sonobe S, Yamamoto S, Motomura M, Shimmen T (2001) Isolation of cortical MTs from tobacco BY-2 cells. Plant Cell Phys 42:162– 169 14. Schmidt S, Smertenko A (2019) Identification and characterization of the land-plant-specific microtubule nucleation factor MACET4. J Cell Sci 132:jcs232819 15. Castoldi M, Popov AV (2003) Purification of brain tubulin through two cycles of polymerization-depolymerization in a highmolarity buffer. Protein Expr Purif 32:83–88 16. Peloquin J, Komarova Y, Borisy G (2005) Conjugation of fluorophores to tubulin. Nat Methods 2:299–303

Chapter 2 Actin: Static and Dynamic Studies Huaqiang Ruan, Sha Zhang, Yi Zhang, and Haiyun Ren Abstract The actin cytoskeleton is a highly dynamic network in plant cells, which is precisely regulated by numerous actin-binding proteins. Hence, characterizing the biochemical activities of actin-binding proteins is of great importance. Here we describe methods for determining the binding and bundling of microfilaments as well as methods for visualizing microfilaments using fluorescent phalloidin and single-molecule TIRF imaging. Key words Actin, Actin-binding proteins, AtFH14, Microfilament binding and bundling assays, Single-molecule imaging, TIRF

1

Introduction The actin cytoskeleton is critical for numerous physiological processes in plant cells, including organelle movement, cytoplasmic streaming, and intracellular transport [1–3]. Multiple actinbinding proteins regulate the nucleation, elongation, cross-linking, and disassembly of actin filaments, providing spatial and temporal control of actin dynamics in vivo [4–6]. Understanding how this integrative protein system works requires methods that allow functions of individual proteins, and different proteins working in combination, to be the assessed and quantified. Conventional biochemical studies provide great tools for identifying actin interactions and quantifying the thermodynamics during actin polymerization and depolymerization. Actin labeling at the Cys374 position of single monomers has led to major advances in our understanding of microfilament growth, branching, and treadmilling [7]. However, data from the above methods are often insufficient, and we still need to infer the underlying mechanisms and the kinetics of the molecular interactions indirectly. The advent of single-molecule fluorescence imaging methods has

Authors Huaqiang Ruan and Sha Zhang have equally contributed to this chapter. Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_2, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

13

14

Huaqiang Ruan et al.

overcome these limitations by monitoring the individual molecules and filaments simultaneously in real time [8, 9]. These techniques provide a powerful new approach to directly visualizing the order of molecular events, quantifying the rate constants, and investigating the coordination of multiple factors [10]. Here we took the study of Arabidopsis formin AtFH14, in particular domains FH2 and FH1-FH2, as examples to describe protocols of high- and low-speed co-sedimentation assays, fluorescent phalloidin staining assays, and single-molecule total internal reflection fluorescence (TIRF) microscopy imaging assays. These studies can be applied to investigate the interaction of actin-binding proteins and microfilaments in vitro, which will provide new insight into how actin dynamics are regulated in plant cells. In this chapter, we describe the following methods using the FH2 and FH1-FH2 domains of Arabidopsis formin, AtFH14, as examples: (a) high-speed co-sedimentation assay for determining microfilament binding, (b) low-speed co-sedimentation assay for determining microfilament bundling, (c) microfilament visualization by staining with fluorescent phalloidin, and (d) single-molecule TIRF imaging assays for visualizing actin polymerization.

2

Materials

2.1 Microfilament Binding and Bundling Assays 2.1.1 Co-sedimentation Assay for Determining Microfilament Binding and Bundling

1. Optima™ TLX Ultracentrifuge (Beckman Coulter). 2. Eppendorf Centrifuge 5424 R. 3. TLA-55 Rotor (Beckman Coulter). 4. Formin dialysis buffer: 1
PBS with 1 mM DTT, pH 7.4. 5. G-buffer: 5 mM Tris–HCl, 0.2 mM ATP, 0.5 mM DTT, 0.1 mM CaCl2, and 1 mM NaN3, pH 7.0. 6. 10
KMEI polymerization buffer: 500 mM KCl, 10 mM MgCl2, 10 mM EGTA, 100 mM imidazole, pH 7.0. 7. Actin and AtFH14-FH2 used here as described in [8]. 8. Coomassie Brilliant Blue R-250. 9. ImageJ (https://imagej.nih.gov/ij/), Quantity One (Bio-Rad), and GraphPad Prism 8 (https://www.graphpad. com).

2.1.2 Microfilament Visualization by Staining with Fluorescent Phalloidin

1. Microscope cover glass 24
50 mm and microscope cover glass 22
22 mm. 2. 1% Versa-Clean detergent. 3. Ethanol. 4. 10 μg/mL polylysine. 5. Formin dialysis buffer: 1
PBS with 1 mM DTT, pH 7.4.

Analyzing Microfilament Binding and Assembly

15

6. G-buffer: 5 mM Tris–HCl, 0.2 mM ATP, 0.5 mM DTT, 0.1 mM CaCl2, and 1 mM NaN3, pH 7.0. 7. 10
KMEI polymerization buffer: 500 mM KCl, 10 mM MgCl2, 10 mM EGTA, 100 mM imidazole, pH 7.0. 8. Alexa Fluor™ 488 Phalloidin(Invitrogen A12379)stock solution: In a dark room, add 150 μL methyl alcohol to the vial of Alexa Fluor™ 488 Phalloidin to yield a stock solution of 66 μM. Mix by vortexing for 10 min until all the dye is fully dissolved, aliquot at 5 μL per tube, and store at 20  C. 9. Alexa Fluor™ 488 Phalloidin working solution: 6.6 μM Alexa Fluor™ 488 Phalloidin, 50 mM KCl, 1 mM MgCl2, 1 mM EGTA, 10 mM imidazole, pH 7.0. 10. LSM880 equipped with PlanApo 63
/1.4 oil objectives (ZEISS). 11. Install Kymograph plugin [11] in the ImageJ (https://imagej. nih.gov/ij/). 2.2 Single-Molecule TIRF Imaging Assays for Visualizing Actin Polymerization

1. Optima XE-100 ultracentrifuge (Beckman Coulter).

2.2.1 Preparation of Biotinylated Actin

4. Biotin dialysis buffer: 5 mM HEPES, 1 mM ATP, and 0.5 mM DTT, pH 7.3.

2. Type 70 Ti rotor (Beckman Coulter). 3. SnakeSkin dialysis tubing: 10 K MWCO, 22 mm (Thermo Fisher).

5. 10
polymerization buffer: 10 mM MgCl2, 0.5 M KCl, and 10 mM EGTA, pH 7.0. 6. NHS-Biotin: 70 mM NHS-Biotin in DMF. 7. Blocking buffer: 1 M Tris–HCl, pH 8.0. 8. Glycerol. 9. G-buffer: 5 mM Tris–HCl, 0.2 mM ATP, 0.5 mM DTT, 0.1 mM CaCl2, and 1mM NaN3, pH 8.0. 10. 1 mL syringe with 26 G needle. ¨ KTA avant chromatography system (Cytiva) and Superdex 11. A 75 5/150 GL (GE Healthcare). 2.2.2 Preparation of Oregon Green-Labeled Actin

1. Optima XE-100 ultracentrifuge (Beckman Coulter). 2. Type 70 Ti rotor (Beckman Coulter). 3. SnakeSkin dialysis tubing: 10 K MWCO, 22 mm (Thermo Fisher). 4. Labeling buffer: 5 mM Tris–HCl, 0.2 mM ATP, 0.1 mM CaCl2, and 1 mM NaN3, pH 8.0. 5. G-buffer: 5 mM Tris–HCl, 0.2 mM ATP, 0.5 mM DTT, 0.1 mM CaCl2, and 1mM NaN3, pH 8.0.

16

Huaqiang Ruan et al.

6. 10
KMEI polymerization buffer: 10 mM MgCl2, 0.5 M KCl, 100 mM imidazole, and 10 mM EGTA, pH 7.0. 7. Oregon Green 488 iodoacetamide (Invitrogen O6034): 10 mM stock solution in DMF. Add 907 μL of DMF to 5 mg of the compound. 8. Tenbroeck tissue grinder. 9. 5 mL syringe with 26 G needle. ¨ KTA avant chromatography system (Cytiva) and Superdex 10. A 75 10/300 GL (GE Healthcare). 2.2.3 Preparation of SNAP-549-AtFH14-FH1FH2

1. SNAP-Surface™ 549 50 nmol (NEB): Add 50 μL of fresh DMSO to the vial of SNAP-Surface 549 (50 nmol) to yield a labeling stock solution of 1 mM in a dark room. Mix by vortexing for 10 min until all the dye is fully dissolved. 2. Formin dialysis buffer: 1
PBS with 1 mM DTT, pH 7.4.

2.2.4

Slide Treatment

1. Microscope glass slide 24
50 mm and coverslip 22
22 mm. 2. 1% Versa-Clean detergent. 3. Piranha solution: 3:1 concentrated sulfuric acid (H2SO4) to 30% hydrogen peroxide (aqueous H2O2) solution (this is a highly corrosive mixture. Please see Note 1, read all health and safety precautions, and discuss this procedure with your local health and safety officer prior to preparing this solution). 4. Ethanol. 5. Passivation buffer: 2 mg/mL mPEG-silane MW 2000 (Laysan Bio) and 2 μg/mL biotin-PEG-silane MW 3400 (Aladdin) in 80% ethanol, pH 2.0 with HCl.

2.2.5 Preparation of Flow Chamber 2.2.6 Flow Chamber Treatment before Imaging

1. Double-sided tape (2.5 cm
2 mm
120 μm). 2. Nail polish. 1. 1% BSA in PBS pH 7.4. 2. Filter paper. 3. Streptavidin buffer: 12.5 μg/mL streptavidin (Invitrogen S888) in 20 mM Tris–HCl, pH 8.0, 1 mM DTT, 100 mM KCl (see Note 2). 4. Catalase stock: 4 mg/mL catalase in 50 mM KH2PO4, pH 7.0, store at 4  C (see Note 3). 5. Glucose oxidase stock: 20 mg/mL glucose oxidase in 50 mM KH2PO4, pH 7.0, 100 mM sodium acetate, and 250 mM KCl, store at 20  C. 6. Glucose stock: 300 mM glucose in H2O.

Analyzing Microfilament Binding and Assembly

17

7. 2
TIRF buffer: 100 mM KCl, 2 mM MgCl2, 2 mM EGTA, 20 mM imidazole, 0.4 mM ATP, 100 mM DTT, 1% methylcellulose (4000 cP), 30 mM glucose, 20μg/ml catalase, 100 μg/ mL glucose oxidase (see Notes 4, 5, and 6). 2.2.7

3

Microscopy

1. Olympus IXplore TIRF microscope system. Install Kymograph plugin in the ImageJ (https://imagej.nih.gov/ij/).

Methods

3.1 High-Speed Cosedimentation Assays for Determining Microfilament Binding

1. Dialyze actin monomers in G-buffer pH 7.0 overnight at 4  C (see Note 7). Dialyze AtFH14-FH2 in formin dialysis buffer overnight at 4  C. 2. Clarify actin and AtFH14-FH2 by centrifugation at 164,000
g for 45 min at 4  C. 3. Add one tenth of 10
KMEI buffer to polymerize actin (5 μM) in a 30 μL reaction mixture at room temperature for 2 h. 4. Mix microfilaments (3 μM) with different concentrations of AtFH14-FH2 in a 100 μL reaction mixture, and incubate at room temperature for 30 min. 5. Centrifuge the reaction mixture at 164000
g for 1 h at 4  C. 6. Separate supernatant and pellet after centrifugation, and conduct polyacrylamide gel electrophoresis, respectively. 7. Stain protein bands in the gels with Coomassie Brilliant Blue R-250 (Fig. 1a). 8. Use ImageJ or Quantity One (Bio-Rad) to measure the gray value of protein bands (see Note 8). 9. Fit a hyperbolic function using the amount of AtFH14-FH2 in the pellet versus supernatant, and determine the binding affinity with GraphPad Prism 8 using the equation of one sitespecific binding (Fig. 1c).

3.2 Low-Speed Cosedimentation Assay for Determining Microfilament Bundling

1. The protocol for the low-speed co-sedimentation assay is almost the same as that for high-speed co-sedimentation assays with the exception that the reaction mixture is centrifuged at 13000
g for 30 min at 4  C. 2. Calculate the ratio of actin in the pellet to total actin in order to quantify the bundling effect (Fig. 1b, d) (see Note 8).

18

Huaqiang Ruan et al.

Fig. 1 High and low-speed co-sedimentation assays. (a) High-speed co-sedimentation assay for determining microfilament binding. Lane 1, actin alone; lanes 2–6, actin plus 0.3, 0.7, 1, 2, or 8 μM AtFH14-FH2, respectively; lane 7, 8 μM AtFH14-FH2 alone. (b) Low-speed co-sedimentation assays to determine the bundling activity of AtFH14-FH2. Lane 1, actin alone; lanes 2–6, actin plus 0.3, 0.7, 1, 2, or 8 μM AtFH14-FH2, respectively; lane 7, 8 μM AtFH14-FH2 alone. (c) Statistical analysis of (a). The concentration of bound AtFH14-FH2 was plotted against the concentration of free AtFH14-FH2 and fitted with a hyperbolic function. (d) Statistical analysis of (b). The proportion of actin in the pellet was calculated with respect to total actin. Error bars represent the standard error 3.3 Microfilament Visualization by Staining with Fluorescent Phalloidin 3.3.1 Cover Glass Surface Treatment

1. Put the slides in H2O with 1% Versa-Clean detergent in a glass container and sonicated in a bath sonicator for 1 h. 2. Rinse the slides with H2O. 3. Rinse the slides with ethanol. 4. Passivate the slides with 80 μL of polylysine for 5 min. 5. Rinse the slides with H2O and dry them in the air.

Analyzing Microfilament Binding and Assembly

19

Fig. 2 Microfilament visualization by staining with fluorescent phalloidin. (a) Microfilaments were distributed in a single and dispersed form (left). Microfilaments were induced to form actin bundles in the presence of 1 μM AtFH14-FH2 (right). Scale bar, 10 μm. (b) Statistical analysis of skewness in (a). Error bars represent the standard error 3.3.2 Sample Preparation Before Imaging

1. Dialyze actin monomers in G-buffer pH 7.0 overnight at 4  C. 2. Clarify actin and AtFH14-FH2 by centrifugation at 164000
g for 45 min at 4  C. 3. Add one tenth of 10
KMEI buffer to polymerize actin (3 μM) at room temperature for 30 min. 4. Mix microfilaments (1 μM) with different concentrations of AtFH14-FH2 in a 10 μL reaction mixture, and incubate at room temperature for 30 min. 5. Mix 1 μL of Alexa Fluor™ 488 Phalloidin working solution dye with 1 μL of the reaction mixture, and incubate at room temperature for 10 min. 6. Drop 1 μL of the reaction solution from the previous step onto the cover glass, and gently cover the coverslip.

3.3.3 Fluorescence Microscopic Analysis of Microfilaments

1. Image actin filaments or bundles with a Zeiss LSM880 microscope with PlanApo 63
/1.4 oil objectives and excitation of 488 nm. Analyze the acquired images with ZEN Blue software and ImageJ (Fig. 2a). 2. Calculate the skewness of fluorescent images with the highskewness plugin in ImageJ for quantifying microfilament bundling (Fig. 2b).

20

Huaqiang Ruan et al.

3.4 Single-Molecule TIRF for Visualizing Actin Polymerization 3.4.1 Preparation of Biotinylated Actin

1. Clarify 1 mL of 150 μM G-actin at 164000
g, and incubate at 4  C for 45 min. 2. Dialyze G-actin versus two changes of 1 L biotin dialysis buffer for 2 h each to remove all amines. 3. Polymerize actin by addition of 10
polymerization buffer to make the final concentration of 1 mM MgCl2, 50 mM KCl, and 1 mM EGTA. 4. Incubate on ice for 1 h. 5. Add NHS-Biotin to the actin solution at the ratio of 9:1 (i.e., 20 μL of 70 mM NHS-Biotin to 1 mL of 150 μM actin), and stir gently at 4  C for 2 h. 6. Quench reaction by adding Tris–HCl to 10 mM, pH 8.0. 7. Slowly add 25% (v/v) glycerol to the reaction mixture (down the side of a tube, do not mix), and centrifuge at 164,000
g, 4  C, for 2 h to pellet microfilaments in glycerol cushion. 8. Resuspend actin pellet in 1 mL of G-buffer. 9. Syringe actin through a 26 G needle 10–20 times. 10. Dialyze the actin against G-buffer at 4  C for 24 h, and centrifuge at 164,000
g, 4  C, for 45 min. 11. Free NHS-Biotin is removed by gel filtration through Superdex 75 5/150 GL. 12. Aliquot the actin at 0.1 mL per tube, freeze it in liquid nitrogen, and store at 80  C before use (see Note 9).

3.4.2 Preparation of Oregon Green Actin

1. Clarify ~5.5 mL of 150 μM G-actin by centrifugation at 164,000
g, 4  C, for 45 min, and dialyze G-actin versus two changes of 4 L labeling buffer for 2 h each to remove all sulphydryl. 2. Dilute the actin to 35 mL in G-buffer to give a final concentration of 1 mg/ml (23.8 μM). 3. Polymerize actin for 1 h at 4  C by adding one tenth of 10
KMEI polymerization buffer, and slowly add 12 moles Oregon Green (OG) iodoacetamide per mole of actin while stirring (e.g., 1 mL of OG dye stock per 35 mL actin). Cover the flask with foil and stir gently overnight at 4  C. 4. Centrifuge the mixture in a 70 Ti rotor at 164,000
g, at 4  C for 2 h, to pellet the actin filaments. The resulting pellet should be green. 5. In a dark room, wash the pellet surface once with G-buffer, and resuspend and transfer the pellets into a Tenbroeck tissue grinder on ice. 6. Grind to dissolve the pellets, and syringe the actin through a 26-G needle for 10–20 times.

Analyzing Microfilament Binding and Assembly

21

7. Dialyze the actin against G-buffer at 4  C for 24 h, and centrifuge at 164,000
g, 4  C for 2 h. 8. Calculate the labeling rate: measure the concentration of Oregon Green actin (OG-actin) and total actin using the equations: [OG-actin] ¼ OD491/77800 M1 cm1 and [Total actin] ¼ [OD290 – (OD491
0.16991)]/26,600 M1 cm1. The labeling rate equals [OG-actin]/[Total actin]. 9. Aliquot the actin at 0.5 mL per tube, and freeze the aliquots in liquid nitrogen. 3.4.3 SNAP-AtFH14FH1-FH2 Labeling

1. Dialyze SNAP-AtFH14-FH1-FH2 in 1 L of formin dialysis buffer for 2 h at 4  C. 2. Dilute SNAP-AtFH14-FH1-FH2 to the concentration of 20 μM. To this, incubate 500 μL of SNAP-AtFH14-FH1FH2 with 20 μL of 1 mM SNAP-Surface 549 dye (two moles SNAP-Surface 549 per mole of formin) at 4  C overnight. 3. Dialyze the reaction mixture against two changes of 1 L formin dialysis buffer for 2 h each to remove free dye (see Note 10). 4. Measure the concentrations of SNAP-tagged proteins and degree of labeling using the Bradford protein assay and BSA standards, and measure fluorophore absorbance in solution using the extinction coefficients: SNAP-Surface 549: 1 cm1, respectively. ε560 ¼ 140,300 M 5. Aliquot the protein at 0.1 mL per tube, and freeze it in liquid nitrogen.

3.4.4 Slide Surface Treatment

1. Put the slides in H2O with 1% Versa-Clean detergent in a glass container, and sonicate them in a bath sonicator for 1 h. 2. Rinse the slides with H2O. 3. Separate the slides in a glass container with piranha solution at room temperature overnight in a water bath (see Note 1, read all health and safety precautions, and discuss this procedure with your local health and safety officer prior to the experiment). 4. Rinse the slides with ethanol, and dry them in a N2 stream. 5. Passivate the slides using polyethylene glycol (PEG) to minimize nonspecific binding. Add 0.1% biotin-PEG to tetherbiotinylated protein after streptavidin treatment. Bake the slides at 70  C for 16 h, and layer each slide with 100 μL passivation buffer.

3.4.5 Preparation of Flow Chamber

1. Place two parallel strips of double-sided tape onto the PEG-coated slide (24
50 mm). Cover a PEG-coated coverslip over the strips of double-sided tape. 2. Press the coverslip with blue tip gently. 3. Seal the top and bottom sides with nail polish.

22

Huaqiang Ruan et al.

3.4.6 Flow Chamber Treatment Before Imaging

1. Prior to each reaction, incubate flow chambers with 1% BSA, 2 min. 2. Incubate 12.5 μg/mL streptavidin for 10 s. 3. Wash the chamber with 1% BSA. 4. Incubate the chamber with 1
TIRF buffer. 5. Incubate the chamber with 50 nM SNAP-549-AtFH14-FH1FH2 and 1μM actin (10% OG-labeled and 0.2% biotinylated) for 2 min. Add 20 μL of 2
TIRF buffer immediately before imaging.

3.4.7 TIRF Microscopy and Kymograph Analysis

1. Perform TIRF using, for example, an Olympus IXplore TIRF microscope equipped with solid lasers (emission 488 nm, 100 mW; emission 561 nm, 100 mW); Olympus ultrasonic stage (IX3-SSU); Objective: Apo 100XOHR (100X, N. A.1.65); CCD camera: iXon3 EMCCD camera (Andor); maintain the focus by the TruFocus Z-Drift compensator (Olympus). Image OG-actin and SNAP-549-AtFH14-FH1-FH2 with excitation of 488 nm and 561 nm, respectively. Use a separate and single-fluor filter cube for each color (Fig. 3). 2. Perform Kymograph analysis to display actin polymerization over time (Fig. 3).

4

Notes 1. Piranha solution is highly corrosive and may result in explosion or injury resulting from chemical and thermal burns if not handled with extreme caution. Moreover, piranha solutions may irritate the respiratory tract if vapor is inadvertently inhaled. Check and observe health and safety guidelines prior to initiating this procedure, e.g., https://ehs.princeton.edu/ book/export/html/513. Specifically, the piranha solution is to be prepared in a fume hood, and be certain that you are wearing protective equipment: gloves (butyl), an acid-resistant lab coat, and chemical splash safety goggles. Always use a clean glass container, and very slowly add hydrogen peroxide to the sulfuric acid (never in the reverse!) while gently stirring. The reaction will be exothermic and may boil. Handle with care to avoid thermal burns. 2. Store 250 μg/mL streptavidin stock solution (100 μL/tube) in 20 mM Tris–HCl, pH 8.0, 1 mM DTT, 100 mM KCl, 20% glycerol at 80  C. Mix 50 μL of the stock solution with 950 μL streptavidin buffer: 20 mM Tris–HCl (pH 8.0), 1 mM DTT, 100 mM KCl before usage. 3. Solutions of catalase should not be frozen. Freeze-thaw will cause a 50–70% loss of enzyme activity.

Analyzing Microfilament Binding and Assembly

23

Fig. 3 Processive microfilament elongation by single SNAP-549-AtFH14-FH1FH2 molecules in the presence of LIPRF1. (a) Microfilament elongation by SNAP549-AtFH14-FH1-FH2 in the presence of LlPRF1 (in seconds). Yellow arrowheads indicate SNAP-549-FH1-FH2-associated barbed ends (Scale bar, 2 μm.) Merged kymographs of filament length (x axis; scale bar, 2 μm) over time (y axis; scale bar, 120 s)

4. To make 2% methylcellulose, add 4 g of methylcellulose to 180 mL of boiling H2O, and stir at 80  C for 3 days. After dissolving, add H2O to 200 mL. 5. Add catalase, glucose oxidase, and glucose to make 2
TIRF buffer just before the experiment.

24

Huaqiang Ruan et al.

6. Make 2
TIRF buffer without the oxygen-scavenging system (catalase, glucose oxidase, and glucose) first. For example, add 1 mL of 10
KMEI buffer, 20 μL of 0.1 M ATP, 0.5 mL of 1 M DTT, 2.5 mL of 2% methylcellulose, and 380 μL of H2O to make 4400 μL of 2
TIRF buffer without catalase, glucose oxidase, and glucose. Aliquot them to 880 μL per tube, and store at 20  C before use. Before experiment, add 10 μL of 4mg/mL catalase stock, 10 μL of 20 mg/mL glucose oxidase stock, and 100 μL of 300 mM glucose stock to get 2
TIRF buffer. 7. Actin monomers should be dialyzed before the experiment. 8. In order to ensure accurate results, the gray values for protein bands should be subtracted from the background value. 9. The labeling rate of biotin-actin is quantified by using Pierce™ biotin quantitation kit (Thermo Fisher 28). 10. Along with dialysis, SNAP-549-AtFH14-FH1-FH2 can be purified again with Ni-NTA beads to completely remove free dye. References 1. Pollard TD, Cooper JA (2009) Actin, a central player in cell shape and movement. Science 326(5957):1208–1212 2. Ruan HQ, Li J, Wang T et al (2021) Secretory vesicles targeted to plasma membrane during pollen germination and tube growth. Front Cell Dev Biol 8:615447 3. Li JJ, Blanchoin L, Staiger CJ (2015) Signaling to actin stochastic dynamics. Annu Rev Plant Biol 66:415–440 4. Ren HY, Xiang Y (2007) The function of actinbinding proteins in pollen tube growth. Protoplasma 230(3–4):171–182 5. Lian N, Wang XW, Jing YP et al (2021) Regulation of cytoskeleton-associated protein activities: linking cellular signals to plant cytoskeletal function. J Integr Plant Biol 63(1):241–250 6. Hussey PJ, Ketelaar T, Deeks MJ (2006) Control of the actin cytoskeleton in plant cell growth. Annu Rev Plant Biol 57:109–125 7. Kovar DR, Harris ES, Mahaffy R et al (2006) Control of the assembly of ATP- and

ADP-actin by formins and profilin. Cell 124(2):423–435 8. Zhang S, Liu C, Wang JJ et al (2016) A processive arabidopsis formin modulates actin filament dynamics in association with profilin. Mol Plant 9(6):900–910 9. Janco M, Dedova I, Bryce NS et al (2020) Visualizing the in vitro assembly of tropomyosin/actin filaments using TIRF microscopy. Biophys Rev 12(4):879–885 10. Breitsprecher D, Jaiswal R, Bombardier JP et al (2012) Rocket launcher mechanism of collaborative actin assembly defined by singlemolecule imaging. Science 336(6085): 1164–1168 11. Higaki T, Kutsuna N, Sano T et al (2010) Quantification and cluster analysis of actin cytoskeletal structures in plant cells: role of actin bundling in stomatal movement during diurnal cycles in Arabidopsis guard cells. Plant J 61(1):156–165

Chapter 3 3D Visualization of Microtubules in Epidermal Pavement Cells Amir J. Bidhendi , Bara Altartouri , and Anja Geitmann Abstract The plant cytoskeleton is instrumental in cellular processes such as cell growth, differentiation, and immune response. Microtubules, in particular, play a crucial role in morphogenesis by governing the deposition of plant cell wall polysaccharides and, in consequence, the cell wall mechanics and cell shape. Scrutinizing the microtubule dynamics is therefore integral to understanding the spatiotemporal regulation of cellular activities. In this chapter, we outline steps to acquire 3D images of microtubules in epidermal pavement cells of Arabidopsis thaliana cotyledons using a confocal microscope. We introduce the steps to assess the microtubule distribution and organization using image processing software Bitplane Imaris and ImageJ. We also demonstrate how the interpretation of image material can be facilitated by post-processing with general-purpose image enhancement software using methods trained by artificial intelligence-based algorithms. Key words Cytoskeleton, Microtubules, Pavement cells, Confocal microscopy, 3D analysis, MAP4, Propidium iodide, Bitplane Imaris, ImageJ, Artificial intelligence, Image enhancement, Topaz Labs

1

Introduction The plant cytoskeleton is involved in a host of cellular activities including sensing and responding to intrinsic and extrinsic cues such as pathogen attack and wound response, cell wall deposition, cell growth, and differentiation [1]. These processes underlie feedback loops that operate at tissue, cell, and subcellular scales. Microtubules are a key element of many of these feedback loops through their involvement in both sensing and response [2–4]. A key response is mediated by the microtubules’ influence on cellulose deposition by way of marking the location of the insertion of cellulose synthase complexes into the plasma membrane and guiding the movement of the enzymes [1, 5]. Microtubules thus influence cell wall mechanical anisotropy and exert spatiotemporal control on cell growth [2, 6, 7]. Our understanding of the involvement of the cytoskeleton in complex developmental signaling and

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_3, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

25

26

Amir J. Bidhendi et al.

response mechanisms in plants is expanding, but it remains limited by our ability to monitor related processes in real time and in 3D. A deeper understanding necessitates improved tools combining highresolution microscopical techniques and live probes with complementary experimental approaches such as mechanical testing [8– 10]. Epidermal pavement cells at the leaf surface of many plant species form intriguing shapes that resemble interlocking pieces of a jigsaw puzzle. The generation of these shapes has been shown to involve intricate coordination of events implicating the cortical microtubules linking mechanical instability, cell geometry, mechanical stress, and cell wall chemistry [2, 11, 12]. In this chapter, we provide a step-by-step protocol to visualize and quantify the pavement cell microtubules in 3D.

2

Materials

2.1 Plant and Plant Growth Material

1. Arabidopsis fluorescent marker line for microtubules, e.g., GFP-MAP4 [13]. 2. 100% ethanol (EtOH). 3. Double-distilled autoclaved water (ddH2O). 4. Bleach. 5. Murashige and Skoog (MS) salt mixture. 6. Sucrose. 7. 1N KOH. 8. Agar. 9. Eppendorf tubes. 10. Pipette tips. 11. Glass bottle. 12. Petri plates. 13. Parafilm.

2.2 Fluorescence Labeling and Sample Mounting

1. Propidium iodide (PI). 2. Tweezers. 3. Microscope slides. 4. Coverslips. 5. Kimwipe paper.

2.3 Microscopy and Image Processing

1. Confocal laser scanning microscope or spinning disk microscope with high numerical aperture (NA) objective. 2. ImageJ or Fiji software or equivalent image processing software.

3D Visualization of Microtubules

27

3. Bitplane Imaris software or equivalent image analysis software. 4. Artificial intelligence (AI)-based image enhancement software, e.g., Topaz Labs’ Gigapixel AI (optional).

3

Methods

3.1 Seed Sterilization and Stratification

For demonstration purposes in this chapter, we use Arabidopsis thaliana because of the large inventory of fluorescent proteinexpressing lines (RFP, GFP, and YFP) for this species. In choosing the optimal fluorescent protein marker line for the experiment, one must carefully consider imaging parameters (see Notes 1–3). There are several methods for seed sterilization and stratification that can be suitable for different usage scenarios. The protocol presented here is time-efficient, and it is most suitable when seeds are to be used for sowing within a few days to 1–2 weeks after sterilization. The method is based on ethanol and bleach and was observed to yield a high rate of seed germination as well as a low rate of seed mortality and contamination pre- and postgermination. 1. Work in sterile conditions with autoclaved and sterile tools inside a laminar flow hood. 2. Transfer the required amount of Arabidopsis seeds to an Eppendorf tube, filling only the very bottom (e.g., 2–3 mm) to allow room for proper sterilization and rinsing in the next steps. If necessary, use multiple tubes. 3. Fill the tube with 100% ethanol (EtOH). For 15 s, gently agitate the tube to maximize flow and contact between the ethanol and seed surfaces. 4. Promptly remove the EtOH using sterilized pipette tips. 5. Fill the Eppendorf tube with double-distilled autoclaved water (ddH2O). Keep agitating the tube for 20 s. 6. Remove the ddH2O using sterilized pipette tips, and cover the seeds with 50% commercial bleach. Gently agitate the tube for 5 min. This can be performed either manually or using a tube rotator. Ensure that seeds do not make aggregates so that the bleach solution can efficiently reach the seed surfaces. 7. Remove the bleach and rinse the seeds three to five times by filling the Eppendorf tube with autoclaved ddH2O, gently shaking the tube each time for 20 s before replacing the ddH2O with fresh liquid. 8. Leave the seeds in the tube half-filled with autoclaved ddH2O. Wrap the Eppendorf tubes in aluminum foil, and leave them in the fridge in the dark at 4°C for 3–4 days for stratification. After this period, seeds can be used for seeding (see Note 4).

28

Amir J. Bidhendi et al.

3.2 Preparation of ½ MS Growth Medium

1. Pour 900 mL of ddH2O into a beaker, and add 2.2 g MS salt mixture [14]. Once the salt is dissolved, adjust the pH to 5.6–5.8 using 1N KOH. 2. Add 10 g sucrose. 3. Adjust the volume of the solution to 1 L with ddH2O. 4. Pour media into a glass bottle suitable for autoclaving. 5. Add 8 g plant agar, and gently stir the solution (agar will remain granular since it does not dissolve at room temperature). 6. Autoclave using the appropriate program for liquids. 7. Let the solution cool to a moderately warm temperature. This can be achieved slowly by leaving it under the hood or more rapidly by bathing the bottles in cold water in the sink. Do not let the solution become too cold as the medium will begin to solidify complicating pouring and distribution into Petri plates. 8. Pour the solution into sterilized Petri plates, and let the media solidify under the hood (see Note 5). 9. The plates can be stored for up to a month in the fridge at 4 °C.

3.3 Seed Germination and Growth

1. Working in sterile conditions under the hood, remove the seeds from the Eppendorf tubes, for example, by using autoclaved pipette tips, and place them on the surface of the solid MS medium (see Note 6). 2. Seal the plate lids using Micropore tape or Parafilm strips, and place them in the growth room under appropriate lighting conditions (e.g., 16-h-long lighting). 3. Check the samples sporadically for growth and/or contaminations. Collect the seedlings at a proper stage (e.g., 2 days) after germination for mounting and microscopical observation (see Note 7).

3.4 Preparation of Secondary Fluorescent Labeling Solution

To visualize the borders of pavement cells, here we use propidium iodide (PI) in parallel to the natively expressed GFP. PI binds cell wall polysaccharides, is easy to work with, and is compatible with live imaging [9, 11] (see Notes 8 and 9). 1. Dissolve the PI, if obtained in powder form, at the desired concentration in ddH2O in an Eppendorf tube. A concentration of 0.5–1 mg/ml was observed to produce a bright signal in pavement cell walls. Vortex the tube. For time-lapse studies, using considerably lower concentrations (e.g., 0.01 mg/ml) might be beneficial to minimize toxicity, stress, or altered growth (see Note 10). 2. PI stock solutions can be prepared and stored in aliquots. Keep the stock in the dark in the fridge (4°C). For longer-term storage, we did not observe any degradation in the quality of the stain resulting from freezing and thawing.

3D Visualization of Microtubules

3.5 Secondary Fluorescent Labeling of Arabidopsis Seedlings

29

1. Using tweezers, remove the seedlings from the Petri plates, and place them on a glass slide or an empty Petri plate. Care must be taken to place the tweezer tips beneath the cotyledons to gently lift the seedlings rather than squeezing them. 2. Cover each seedling with one to two drops of the PI staining solution. Ensure that the seedling is fully immersed in the staining solution. Cover the samples to protect them from light. 3. After 5–10 min, remove the PI solution using a pipette or a piece of Kimwipe paper, and rinse the seedlings three times using ddH2O. Rinsing can be performed by placing a few drops of ddH2O on each seedling for 15–20 s before removing and replacing the liquid with fresh rinsing solution. The seedlings are now ready to be mounted for microscopy.

3.6 Mounting Samples on Slides for Microscopical Observation

1. Gently place each seedling on one to two drops of ddH2O on a glass slide (see Note 11). Place one edge of a coverslip on the water, and slowly lower the other edge so that the cotyledons are gently flattened between the slide and the coverslip for a proper field of view and ease of focus during microscopy. If the seedlings are too large, the thickened stems can prevent a good seal or proper imaging of the cotyledons. Spacers can be used to adjust the distance between coverslip and slide (see Note 7). 2. Gently press on the edges of the coverslip using a piece of Kimwipe paper to squeeze out and absorb excess mounting liquid, if any, from below the coverslip. This will allow a better seal and focus and prevent displacements of the coverslip during imaging. 3. Perform the imaging as soon as possible. Keep the samples in dark conditions before and during imaging (see Note 12).

3.7 Confocal Microscopy for Imaging Microtubules in Epidermal Pavement Cells

We use a Zeiss LSM 510 META confocal microscope at basic settings for imaging in the present chapter (see Note 13). Other confocal microscopes provide similar functionality under similar settings. 1. Locate the region of interest on the cotyledon in brightfield mode at low light intensity settings. 2. Switch to the laser scanning mode by choosing the appropriate settings including the laser, excitation, and emission filters. 3. We used excitation wavelengths of 532 nm and 489 nm and emission windows of 590–625 nm and 500–530 nm for PI and GFP channels, respectively. 4. Choose an objective with proper magnification based on the region of interest and detail to be studied. For a given magnification, choose the objective that has the highest NA for the best results.

30

Amir J. Bidhendi et al.

5. Adjust the laser power to an acceptable minimum to avoid photobleaching or stressing the samples. Determine an optimal combination set of scanning parameters, including laser power, digital gain, pinhole diameter, scanning speed, and image size (see Notes 13–15). 6. After determining the optimal scanning parameters, the scanning depth must be determined. This is performed by setting the first and last slices of the z-stack and determining either the number of slices or the distance between slices. Generally, the microscope software allows choosing optimal z-step distance based on the used objective and the pinhole diameter (see Notes 16 and 17). 7. Choose the beginning and end levels of the z-stack to start the scanning well above the sample surface and end below the target structure. Note that the periclinal walls of pavement cells bulge outward and hence form highly curved surfaces. To fully capture the outer half of the epidermal cell layer, the starting z-level should therefore be above the highest bulge, and the lowest z-level should capture the anticlinal walls of all cells in the field of view without leaving any gaps. If in doubt, it is generally preferable to acquire more than less (see Note 18). 8. Save the z-stacks in proper format with all relevant experimental notes (e.g., time and date, treatments, etc.; see Note 19). 3.8 2D Image Analysis by Projecting the 3D Data on a Plane

Analyses of micrographs and z-stacks can be performed in 2D or 3D depending on the level of detail required and the type of information desired. 2D analysis of the z-stacks is the first stage in the image analysis routine because it requires fewer computational resources and can easily be performed using freely available platforms such as ImageJ (https://imagej.nih.gov/ij/) or Fiji (https:// imagej.net/software/fiji/). Here we provide a step-by-step procedure to obtain information on the distribution of microtubules along the wavy borders of pavement cells. For this demonstration, we rely on z-stacks acquired from the GFP channel only. The process for merging a second channel into the final projection is very similar to the one described below for adding a single optical section. The use of a second channel will be discussed in the next sections (3D visualization). 1. Launch the ImageJ (or Fiji) program, and open the corresponding micrograph (here in .lsm or .czi format). To open the file format, sometimes additional ImageJ plugins such as Bio-Formats are necessary which can be added to the ImageJ plugin library, if not already included. 2. Scroll down in the z-stack until the slice in which the anticlinal borders of all or most cells within the region of interest are visible (Fig. 1a). For this, flat cotyledons are particularly helpful at high magnification because curved organ surfaces lead to the

Fig. 1 2D analysis of confocal microscope z-stack of microtubules in GFP-MAP4 pavement cells. (a) A single optical slice showing epidermal cells of Arabidopsis GFP-MAP4 cotyledon. A slice to capture most of the cell borders is generally located at the mid-depth of the epidermal layer and can be duplicated, pseudo-colored, and used to mark cell borders in z-projections of the 3D stack. (b) Maximum intensity z-projection of the confocal microscope z-stack showing microtubules of epidermal pavement cells in 2D. (c) The slice in (a) is merged into (b) to mark cell borders using a different pseudo-color (red, single slice as in (a); yellow, same stack as in (b). (d) Orthogonal views are useful to assess structures oriented perpendicular to the xy plane. Yellow lines in xy view indicate the locations of the orthogonal yz and xz views. (e and f) ROIs can be determined using Selection tools (here oval). Microtubule density can be evaluated using Analyze → Measure. Here the value for “mean” is the mean pixel signal intensity in each oval ROI and indicates that the signal (and hence the microtubule density) is higher adjacent to the indentation side of a cell border undulation (mean = 24.3) compared to the opposing protrusion side (mean = 12.3). Scale bars = 10 μm

32

Amir J. Bidhendi et al.

border of some cells falling out of the focal plane requiring increasingly deeper scans. Duplicate the selected optical slice (Shift+D in windows or using the Image drop-down menu). Uncheck “duplicate stack.” 3. If needed, the contrast and brightness of the selected slice can be adjusted (Shift+C or from the Image drop-down menu → Adjust). You can leave this micrograph open or save it (.tiff). 4. To analyze the microtubule distribution in 2D, a z-projection can be performed. z-Projection can be performed using Image → Stacks → Z Project (Fig. 1b). Enter the number of first and last z-slices of interest. Choose the desired type of projection depending on the type of analysis, or if unsure start with max intensity. Duplicate the resulting projection to generate a separate working file for the next steps. 5. A slice from a different channel such as the one used for the PI signal can be merged into a composite micrograph to outline cell borders in microtubule-dense regions. 6. Alternatively, to facilitate distinguishing the pavement cell borders using microtubule data only, a composite figure can be produced with the pseudo-colored border slice obtained previously (Fig. 1a) merged into the result of the z-projection (Fig. 1b). For this, with both border and z-projected micrographs open in ImageJ, use Image → Color → Merge Channels. Choose the desired color channels for each item from the drop-down menus. Check “Create composite,” “Keep source images,” and “Ignore source LUTs” (Fig. 1c). Before this, images can be set at suitable types and bit depths (Image → Type → RGB, 16-bit, or 32-bit). The resulting micrograph can be saved in desirable type and format (e.g., RGB, .tiff). 7. Measurements such as average signal intensity can be performed within a region of interest (ROI) using Area Selection tools and Analyze → Measure (or M). Before this, set the desired parameters that are to be included in the report. For this, go to Analyze → Set Measurements. For example, here we checked “Area, Mean gray value, Standard deviation, Min & Max gray value, and Display label.” To perform repeated measurements and comparisons, such as comparing the microtubule density on two sides of a border undulation, the selected ROI marker can be moved to different locations on the image using the cursor (Fig. 1e, f). 8. To detect structures oriented in the z-direction, orthogonal projections can be performed in x or y directions similarly to z-projection. Orthogonal views (Image → Stacks → Orthogonal views) can also be very helpful as they allow observing the same region from different views (Fig. 1d).

3D Visualization of Microtubules

33

9. If a substantial and non-uniform background signal is present, the estimation of microtubule density based on signal intensity within a selected ROI may be less reliable. Moreover, this approach does not provide information on the directionality of microtubules and does not consider the microtubules aligned perpendicular to the plane of view. To analyze the distribution and orientation of the fibril-like structures, tools such as the FibrilTool can be used [15]. Microtubule orientation can be analyzed on 2D projections yielding useful information on the cortical microtubules along the periclinal or anticlinal walls, respectively. However, 2D projections do not allow for the assessment of microtubules that change orientation in space, e.g., when curving at a cell edge. To obtain full details on the spatial distribution of the cytoskeleton, 3D analysis may be the only option. While ImageJ allows for some level of 3D reconstruction and analysis (e.g., Image → Stacks → 3D Projection), here we will demonstrate the use of an alternative powerful image analysis tool, Bitplane Imaris. 3.9 3D Image Analysis on the Spatial Distribution of Microtubules

Cellular structures are inherently 3D objects. While 2D analysis techniques in many cases provide valuable insight, they can also obscure crucial 3D details. Bitplane Imaris is a powerful commercial software for 3D and 4D analysis of microscopical images. Both the free and full versions of the Imaris software offer attractive functionalities for visualization of confocal z-stacks. In this section, we use the full version Imaris 9.8 (https://imaris.oxinst.com) to visualize z-stacks obtained from GFP and PI channels. Steps to import the stacks are as follows. 1. Imaris “Arena” is the starting point for importing the z-stacks and converting files to Imaris file format (Fig. 2a). This module allows managing folders and files. Image files and stacks can be processed individually or in batch mode (e.g., File → Batch convert). Under “Arena,” use “Observe Folder” to add a folder path containing the z-stacks. Right-click on the z-stacks, and select “Convert to native Imaris file format” to convert them to .ims format (Fig. 2b). 2. In the “Surpass” view, open the z-stack by File → Open, and locate the .ims file of the converted z-stack. The z-stack can be viewed in either 3D or slice view. In slice view, the slice number can be adjusted using the slider (Fig. 2c). Image brightness and contrast can be adjusted in the Display Adjustment toolbar via setting the min. and max. values in the histogram (see Note 20). The pseudo-color assigned to the channel can be adjusted by clicking on the channel number (e.g., chL1) in Display adjustment (Fig. 2c–7).

34

Amir J. Bidhendi et al.

Fig. 2 Imaris software used for 3D visualization of confocal z-stacks of microtubules in pavement cells of the Arabidopsis GFP-MAP4 line. (a) Arena is the environment used for file management and conversion. (b) Input stack or images or image series are converted to native Imaris format (.ims) in Arena environment. (c) Some

3D Visualization of Microtubules

35

3. In 3D view mode, the basic MIP (maximum intensity projection) can be used to visualize the z-stack. The resulting 3D object can be translated and rotated using right and left mouse buttons. The scroll wheel can be used to zoom in and out of a region (Fig. 2d). 4. To visualize the cell borders using the z-stacks from the PI channel, additional images can be imported and added to the existing volume. This can be achieved by File → Add image and choosing the .ims file corresponding to the z-stack of interest. In the same fashion as with the previous channel, the intensity level of the second channel can be adjusted to allow revealing the appropriate level of details between the two overlapping stacks (Fig. 2e). 5. For presentation and publication purposes, animations can be made using the Animation tool in the Surpass environment. Animation can be created by adding “Key Frames” entailing two or several translations and rotations to create a desirable camera path and perspectives based on features of interest. The resulting animation can be played back, recorded, and saved under the options for the Animation tool (Fig. 2e). 3.10 Potential for Artificial IntelligenceBased PostAcquisition Image Enhancement in the Interpretation of the Subcellular Organization of Microtubules

It is crucial to aim for the best-quality images at the sample preparation and image acquisition steps to eliminate the need for excessive image post-processing (see Note 21). Here we introduce sample images of microtubules in pavement cells processed using Topaz Labs’ Gigapixel AI (v5.8)—a photo enlargement software that uses trained AI model to enlarge, deblur, and denoise photos (https://topazlabs.com/gigapixel-ai). 1. Save the z-stack as individual .tiff images using ImageJ, File → Save as → Image sequence → .tiff. 2. Import the Image sequence in Topaz Lab’s Gigapixel AI by File → Open images (select all). 3. In the imported images, select a slice that includes regions with in-plane microtubules. This slice will be used to adjust enhancement parameters for the whole batch.

ä Fig. 2 (continued) Surpass environment tools and options used to configure and visualize the images and 3D stacks: Display Adjustment module (1), Surpass environment button (2), 3D view (3), Slice view (4), Section or orthogonal views (5), Animation tool (6), Channel pseudo-color adjustment (7), Slice selection slider (8), and Histogram parameters (9). A single slice of a z-stack is shown in the figure. (d) MIP display mode is used to visualize the z-stack in 3D using the 3D view option. (e) A separate z-stack (propidium iodide channel, green) is added to the existing GFP-MAP4 stack (orange) to mark the cell borders. Channel histograms can be adjusted individually in the Display adjustment module. Using the Animation tool, animations can be produced by adding Key frames at various locations and angles guiding the relative trajectory of the camera and the 3D stack

36

Amir J. Bidhendi et al.

Fig. 3 A single optical section of a z-stack of Arabidopsis MAP4 GFP marker line used for AI-based image enhancement using Topaz Lab’s Gigapixel AI. (a) Confocal micrograph of microtubules marked with GFP-MAP4. (b and d) Different magnification levels of the micrograph in (a) before AI-based enhancement. (c and e) Areas in (b, d) after AI-based enhancement. Apparent uneven abundance of microtubules across the field of view in (a, b, c) result from outward bulging of cells. The single optical slice shown captures only the most elevated areas of the bulges. Scale bars = 20 μm (a), 10 μm (b, c), and 3 μm (d, e)

4. From the viewfinder, select and zoom in a suitable region (Fig. 3a). 5. Choose the scaling factor or set the desired width and height of the image for image enlargement. Select the AI model that produces the best result through trial and error adjusting the method parameters. This can be performed using the side-byside view to compare the original image (Fig. 3b, d) with the updated result (Fig. 3c, e). 6. The determined settings are automatically applied to all selected images. Use “Save all” to batch process and export the image series.

3D Visualization of Microtubules

37

7. To use the processed images for 3D visualization, the image series exported from Gigapixel AI can be imported into Imaris. To this end, in Imaris, add the folder containing the images to the “observed folder” path as described before. 8. Right-click on the thumbnail of the image appearing in the folder in Arena, and choose “Configure file series.” Adjust the series parameters, particularly the z-step, which can be obtained from the metadata of the original stack. 9. Convert the series by right-clicking on the thumbnail and choosing “Convert to Imaris Native File Format.” The enhanced 3D stack can now be visualized in Surpass environment as described previously (see Notes 22–24).

4

Notes 1. To choose the optimal fluorescent protein line for an experiment, one must consider several parameters including but not limited to: • Emission and excitation wavelengths to avoid or at least minimize spectral overlap with those of other fluorescent probes such as dyes used for labeling cell wall polysaccharides. • Brightness and photostability (resistance to photobleaching) of a given fluorescent protein. • Likelihood of the fluorescent protein oligomerization as well as potential changes in the functioning of the attached protein. 2. For live-cell imaging of cortical microtubules, the commonly used lines involve microtubule-binding protein markers such as MAP4 and MAP65 and tubulin markers such as TUA5, TUA6, and TUB6. 3. Seeds and plants of any mutant line must be discarded according to corresponding protocols and regulations such as by autoclaving them before disposal. 4. The presented method for seed sterilization and stratification eliminates the need for additional steps such as seed dehydration after sterilization and works best if the seeds are to be used within a short period (e.g., 1 week). Keeping the seeds in the fridge substantially longer may affect their viability, germination, and growth rate. 5. To avoid condensation and droplets forming on the inside of the lid or on the media, which can promote microbial growth, leave the plate lids partly open.

38

Amir J. Bidhendi et al.

6. Allow sufficient spacing between the seeds (e.g., 5 mm from each side) to prevent the entangling of seedling radicles as they grow and facilitate their harvesting in the next steps. Avoid transfer of excess liquid with the seeds from the Eppendorf tubes. 7. Seedlings typically germinate between 48 and 72 h after sowing. For visualization of microtubules using GFP lines, cotyledons of seedlings between 2 and 5 days after germination were found to be most amenable to study. At later stages, the cotyledons grow too large and form curved surfaces preventing proper mounting (if mounted on microscope slides with coverslips). On the other hand, visualizing microtubules of the epidermal layer of the cotyledon at the early stages (e.g., 1–2 days after germination) will allow approaching the abaxial side of the cotyledon very close to the coverslip because the cotyledons are closed. If the cotyledons are studied at later stages, particularly for time-lapse studies, we recommend the use of double-sided tape as a spacer. Beneath the cotyledon, place either a thin layer of solid MS media or a tape to minimize the distance between the cotyledon and the coverslip. In this case, the abaxial or the adaxial sides of the cotyledon can be visualized while minimizing the compressive forces from the coverslip (see also Note 11). 8. A second probe can be used to illuminate the cell borders. This can be achieved using fluorescent probes with affinity to either the plasma membrane or cell wall polysaccharides. If a dye is chosen, it should be suitable for live-cell imaging: it must be easy to apply to the samples and should not interfere with cell growth over the course of time-lapse experiments. A common plasma membrane stain is the lipophilic dye FM4-64. Some of the common cell wall dyes are PI, Pontamine fast scarlet 4B (S4B), and calcofluor white [9]. Calcofluor white is a fluorescent blue dye, whereas PI and S4B are fluorescent red dyes. Calcofluor white and S4B label cellulose with different affinities, while PI binds the negatively charged pectin in the wall. All these probes have been successfully used for live-cell imaging [9, 11, 16]. PI was particularly useful in the case of pavement cells because its application is easy, it readily penetrates the cuticular layer—at least in cotyledons—and it does not interfere with cell growth at moderate concentrations. 9. Always consult the material safety data sheets (MSDS) for any new material used in experiments. PI is a known mutagen and must be handled with care. The liquid and solid waste including the rinsing solutions as well as tools and containers must be discarded and cleaned according to safety and hazardous waste regulations.

3D Visualization of Microtubules

39

10. The cytoskeleton is highly dynamic with high rates of polymer growth and depolymerization. Therefore, snapshot images of cytoskeletal configuration must be interpreted accordingly, and, in some cases, time-lapse imaging may provide better insight than single snapshots. 11. Microtubules are known to respond to changes in mechanical stress. Indeed, pressing the seedlings between coverslip and glass slide can lead to altered microtubule dynamics and organization [17]. While at the subcellular scales, the mechanical stresses are presumably sufficiently large to overcome contact pressure-induced microtubule reorientation, the tissue-level organization of the cytoskeleton might still be affected to some degree and at least transiently. Therefore, using immersion objectives eliminating the need for coverslips is highly preferred. 12. The procedure listed in this chapter is tailored to one-time observations of the seedlings using a non-immersion microscope objective. If long-term time-lapse experiments are intended, seedlings must be handled extremely gently to avoid mechanical damage, ensuring that radicles are in contact with medium at all times and that sample transfer to and from growth chambers between imaging sessions is swift. For this purpose, microscope mountable growth slides can be invaluable [9, 11]. 13. Laser scanning confocal microscopes produce high-quality low-noise z-stacks. However, the exposure times in this imaging modality can sometimes be too long for time-lapse experiments leading to photobleaching of the fluorescent proteins or phototoxicity. These issues can be reduced by using a spinning disk confocal microscope. 14. The optimal percent of the laser power will depend on various parameters such as the quality of labeling, wavelength, scanning speed, as well as total power output of the laser. 15. Ideally, the acquired images are obtained at maximum resolution (in terms of the number of pixels) with a pinhole set as close as possible to one Airy unit diameter to minimize out of focus light. However, in practice, compromises need to be made to obtain z-stacks at reasonable speed to minimize sample damage. The scanning parameters will largely depend on factors such as the type of fluorophore, the quality of labeling, as well as the depth of the sample to be scanned. 16. Undersampling—using a slice distance larger than optimal— can lead to spatial aliasing reducing the usefulness of micrographs, particularly for publication purposes. Undersampling can be helpful to gain speed when image quality is not a concern, especially during the initial phase of trial and error and parameter optimization. Oversampling, on the other hand,

40

Amir J. Bidhendi et al.

can produce high-quality images but at the expense of the risk of photobleaching. This can lead to non-uniform signal intensity through the depth of the z-stack as the sample is getting increasingly bleached. 17. Sample drift in the 3D and temporal dimensions, e.g., due to the movement of the sample during the acquisition or due to plant growth, can represent an experimental challenge. If the drift occurring during a single z-stack and acquisition cannot be mitigated by modifying the mounting method, we recommend a faster acquisition time. For a time-lapse experiment, sample drift may occur in xy and z-directions. We recommend using a larger scanning area and larger upper and lower limits so that there are at least three z-steps above and below the top and bottom surfaces of the sample or the target structure, respectively. If drift occurs during the study, it may be corrected during post-processing using tools such as the 4D drift correction plugin in ImageJ [18]. 18. Insufficient scanning range can render the data useless or even misleading. Unfortunately, gaps between neighboring cells resulting from an insufficient depth of the z-scan have sometimes been mistakenly referred to as the middle lamella—the adhesive layer between plant cells. Choosing a z-level that captures the inner periclinal walls and adjacent mesophyll, therefore, ensures that such errors in interpretation are avoided and that abundant information for post-acquisition analysis is generated. 19. Confocal micrographs are generally saved automatically with their metadata. Metadata contain information such as the objective used, wavelengths, and pinhole diameter. The file formats vary between microscope brands and models. For Zeiss microscopes, z-stacks are generally stored as .lsm or .czi formats that can be viewed using open-source or proprietary software such as Zeiss Zen, Imaris, or ImageJ. 20. Nonlinear adjustments such as gamma correction must be reported along with the results. Adjustments such as brightness and contrast must be performed identically for the control and other conditions, certainly if fluorescence intensity is to be compared. 21. In some situations, judicious application of post-processing routines can aid in interpreting the data, however. Suboptimal image quality may be inevitable, for example, when higher scanning speeds are required to minimize photobleaching, when samples are poorly mounted, or when signal intensity decreases with increasing scanning depth. The resulting loss of image quality renders the recognition of spatial localization and distribution of microtubules challenging. Recent years have seen a growing number of tools developed to automate feature

3D Visualization of Microtubules

41

detection from microscopy images. FibrilTool [15] and SOAX [19] are examples of toolboxes that enable automated and reproducible detection and quantification of fibrillar structures such as microtubules which can benefit from improved image quality. AI-based packages specially trained for handling microscopical images, such as NIS.ai by Nikon, TruAI by Olympus, and ZEN Intellesis by Zeiss, show a great potential in establishing trainable image segmentation and enhancement pipelines for tasks such as deblurring, denoising, removing out of focus light, and feature recognition. Interestingly, we observed that even some of the general-purpose AI-based image enhancement packages not specifically trained to handle microscopical images can produce stunning results. 22. These image manipulation procedures can easily introduce false information, e.g., non-existing microtubule branches. The enhancement parameters and results must be scrutinized and considered in this context. 23. Studies are warranted to evaluate the efficacy and accuracy of image enhancement using such tools. We performed some tests by adding simple noise and blur effects to original images followed by rescuing the images using Gigapixel AI. While the results were stunning (not shown), an in-depth study of noise and blur relevant to microscopy experiments must be performed. 24. Ultimately, as with any image manipulation, these image modifications should be approached with utmost caution and disclosed when submitted for publication.

Acknowledgments This study was supported by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada (NSERC) to A.G. and by the Canada Research Chairs Program. Image acquisition was performed at the McGill University Multi-Scale Imaging Facility, Sainte-Anne-de-Bellevue, Que´bec, Canada. References 1. Chebli Y, Bidhendi AJ, Kapoor K, Geitmann A (2021) Cytoskeletal regulation of primary plant cell wall assembly. Curr Biol 31:R681– R695 2. Bidhendi AJ, Altartouri B, Gosselin FP, Geitmann A (2019) Mechanical stress initiates and sustains the morphogenesis of wavy leaf epidermal cells. Cell Rep 28:1237–1250.e6

3. Landrein B, Hamant O (2013) How mechanical stress controls microtubule behavior and morphogenesis in plants: history, experiments and revisited theories. Plant J 75:324–338 4. Sampathkumar A, Krupinski P, Wightman R, Milani P, Berquand A, Boudaoud A, Hamant O, Jo¨nsson H, Meyerowitz EM (2014) Subcellular and supracellular mechanical stress prescribes cytoskeleton behavior in

42

Amir J. Bidhendi et al.

Arabidopsis cotyledon pavement cells. Elife 3: e01967 5. Chan J, Coen E (2020) Interaction between autonomous and microtubule guidance systems controls cellulose synthase trajectories. Curr Biol 30:941–947.e2 6. Bidhendi AJ, Geitmann A (2016) Relating the mechanics of the primary plant cell wall to morphogenesis. J Exp Bot 67:449–461 7. Cosgrove, Daniel J (2022) Building an extensible cell wall. Plant Physiology 189:1246– 1277 8. Bidhendi AJ, Geitmann A (2018) Tensile testing of primary plant cells and tissues. In: Geitmann A, Gril J (eds) Plant Biomechanics: From Structure to Function at Multiple Scales. Springer International Publishing, Cham, pp 321–347 9. Bidhendi AJ, Chebli Y, Geitmann A (2020) Fluorescence visualization of cellulose and pectin in the primary plant cell wall. J Microsc 278: 164–181 10. Robinson S, Durand-Smet P (2020) Combining tensile testing and microscopy to address a diverse range of questions. J Microsc 278:145– 153 11. Altartouri B, Bidhendi AJ, Tani T, Suzuki J, Conrad C, Chebli Y, Liu N, Karunakaran C, Scarcelli G, Geitmann A (2019) Pectin chemistry and cellulose crystallinity govern pavement cell morphogenesis in a multi-step mechanism. Plant Physiol 181:127–141

12. Bidhendi AJ, Geitmann A (2019) Geometrical details matter for mechanical modeling of cell morphogenesis. Dev Cell 50:117–125.e2 13. Marc J, Granger C, Brincat J, Fisher D, Kao T, McCubbin A, Cyr R (1998) A GFP-MAP4 reporter gene for visualizing cortical microtubule rearrangements in living epidermal cells. Plant Cell 10:1927–1940 14. Murashige T, Skoog F (1962) A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiol Plant 15:473– 497 15. Boudaoud A, Burian A, Borowska-Wykre˛t D, Uyttewaal M, Wrzalik R, Kwiatkowska D, Hamant O (2014) FibrilTool, an ImageJ plug-in to quantify fibrillar structures in raw microscopy images. Nat Protoc 9:457–463 16. Anderson CT, Carroll A, Akhmetova L, Somerville C (2010) Real-time imaging of cellulose reorientation during cell wall expansion in Arabidopsis roots. Plant Physiology 152:787–796 17. Jacques E, Verbelen J-P, Vissenberg K (2013) Mechanical stress in Arabidopsis leaves orients microtubules in a ‘continuous’ supracellular pattern. BMC Plant Biol 13:163 18. Parslow A, Cardona A, Bryson-Richardson RJ (2014) Sample drift correction following 4D confocal time-lapse imaging. J Vis Exp e51086 19. Xu T, Vavylonis D, Tsai F-C, Koenderink GH, Nie W, Yusuf E, I-Ju Lee, Wu J-Q, Huang X (2015) SOAX: A software for quantification of 3D biopolymer networks. Sci Rep 5:9081

Chapter 4 Quantitative Analysis of Microtubule Organization in Leaf Epidermis Pavement Cells Sandra Klemm, Jonas Buhl, Birgit Mo¨ller, and Katharina Bu¨rstenbinder Abstract Leaf epidermis pavement cells form highly complex shapes with interlocking lobes and necks at their anticlinal walls. The microtubule cytoskeleton plays essential roles in pavement cell morphogenesis, in particular at necks. Vice versa, shape generates stress patterns that regulate microtubule organization. Genetic or pharmacological perturbations that affect pavement cell shape often affect microtubule organization. Pavement cell shape and microtubule organization are therefore closely interconnected. Here, we present commonly used approaches for the quantitative analysis of pavement cell shape characteristics and of microtubule organization. In combination with ablation experiments, these methods can be applied to investigate how different genotypes (or treatments) affect the organization and stress responsiveness of the microtubule cytoskeleton. Key words Leaf epidermis, Pavement cells, Microtubules, PaCeQuant, FibrilTool, OrientationJ, CytoskeletonAnalyzer2D

1

Introduction Pavement cells form the ground cell type in the leaf epidermis and exhibit a wide variety of shapes across different species [1, 2]. In many species, including the model plant Arabidopsis thaliana, pavement cells acquire highly complex, jigsaw puzzle-like shapes with wavy interlocking patterns at their anticlinal walls (Fig. 1a, b). During pavement cell morphogenesis, their shape complexity increases, which requires precise growth coordination within a tissue of mechanically coupled cells [3]. Because of their shape complexity, pavement cells have developed into a popular model system to study morphogenesis at the cell and tissue scale. Pavement cell shape formation is governed by mechanical stresses at the cell and tissue level [4] and involves mechanical wall heterogeneities along their anticlinal walls [5]. Microtubules are enriched in neck regions (Fig. 1d), where they guide plasma membrane-localized

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_4, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

43

44

Sandra Klemm et al.

a

b

Periclinal wall

Anticlinal wall

d

c Neck

Lobe LEC

Neck region 0 Stomata guard cell

255

3-cell-junction

Fig. 1 Characteristics of pavement cell shapes and microtubule organization. (a) Micrograph showing microtubules (black) in the adaxial side of cotyledons in 5-day-old Arabidopsis seedlings expressing the GFP-MAP4 microtubule marker. Scale bar, 50 μm. (b) Pavement cells display large shape complexity at their anticlinal walls. (c) Anticlinal walls form interlocking protrusions and indentations, referred to as lobes and necks, respectively. Cell wall properties differ at three-cell-junctions (violet) and in regions adjacent to stomata guard cells (blue) compared to two-cell-junctions between neighboring pavement cells. The LEC serves as measure of maximal stress (orange circle), and microtubules tend to align in neck regions (orange rectangle). (d) Fluorescence intensities of microtubule signals in a pavement cell pseudo-colored with the FIRE lookup table and magnification of a neck region (box)

cellulose synthase complexes, thereby enforcing cell walls locally [6]. Microtubule ordering correlates with the orientation of subcellular stress fields that originate from the complex geometries of pavement cells [4]. Perturbation of mechanical forces at the tissue scale induces reordering of microtubules along these supracellular stress patterns. Tissue-level stress fields thus can override cell

Cytoskeleton Organization in Pavement Cell Morphogenesis

45

shape-dependent microtubule organization in pavement cells. Pharmacological and genetic studies demonstrated strong defects in pavement cell shape morphogenesis upon disruption of the microtubule cytoskeleton or of components of Rho of Plants GTPase (ROP) signaling modules [7–13]. Similarly, albeit to a weaker extent, disturbances of cell wall composition lead to aberrant pavement cell morphogenesis. Recent data suggest that mechanical signaling is based on the receptor kinase FERONIA, which binds cell wall pectins and activates ROP6 signaling preferentially in neck regions [10, 11]. In addition, phytohormone signaling contributes to spatially confined reorganization of the actin and microtubule cytoskeleton in lobes and necks, respectively [10– 13]. Collectively, upon activation via ROP signaling, the microtubule cytoskeleton thus presumably translates mechanical forces into growth regulation through local cell wall reinforcements via cellulose deposition [14, 15]. In this chapter, we present an overview of the methods that are used to quantify microtubule organization in pavement cells in the context of altered shape complexity or in response to changes in tissue-scale mechanical forces using microtubule marker lines and confocal laser-scanning microscopy. Quantitative analyses of such images are often focused on measuring local microtubule densities and orientations. Target microtubule densities are usually quantified in terms of normalized intensity values, often averaged over local regions of interest (ROIs). To assess microtubule orientation, measures are commonly derived from local structure or nematic tensors resulting in estimations of local structure anisotropy or coherence values and fiber orientations. Common tools for such protocols are, e.g., FibrilTool [16] and OrientationJ [17]. Alternatively, methods that work at a more global and structural characterization of the overall appearance of the microtubule cytoskeleton have been proposed. While anisotropy and orientation measurements are most reliable for clearly visible and highly contrasting fibers, texture measures are always applicable. CytoskeletonAnalyzer2D [27], for example, extracts for each cell, a characteristic fingerprint vector that describes the structural appearance of the cytoskeleton and yields a flexible basis for quantitative comparisons of cytoskeleton appearance among, for example, individual cells or complete populations. Often, microtubule orientation is measured in 3D projections covering the upper half of the cell, from the anticlinal to outer periclinal surface. The ROIs over which measures are averaged are sometimes selected manually, to, for example, consider only lobes or necks (Fig. 5b) or regions connecting necks (Fig. 1c) separately [12]. In more automated approaches, complete cells are analyzed using manually or automatically extracted cell contours as input, or microtubule orientation is measured within the largest empty circle (LEC) [7]. The LEC and its midpoint have become an established

46

Sandra Klemm et al.

measure to characterize locations and directions of maximal mechanical stress within cells. Sometimes sliding window approaches along cell contours are applied. These provide continuous profiles of intensity or coherence values, compared to locally sampled data resulting from analysis of a few selected ROIs only. While the higher degree of automation enables the analysis of larger and more representative data sets of cells, averaging coherence, orientation, or intensity values over larger ROIs such as complete cell areas always bears the risk that average values lose explanatory power, in particular in pavement cells with large local variation in microtubule organization. Microtubule organization and alignment, however, are linked to cell shape complexity at anticlinal walls. Hence, correlating local microtubule measures with necks and lobes or, more generally, local cell wall curvature provides specific information on microtubule organization in pavement cells. Local LECs (LLEC) have proven suitable to measure local curvatures along cell contours [7]. The LLEC at a single-pixel position along a cell contour refers to the largest circle that best fits the local contour. By calculating LLEC radii for all points along a contour, curvature profiles can be extracted. Correlating these profiles with local microtubule orientations or densities as, for example, proposed in protocols like MtCurv by [8] or in the microtubule-contour analysis tool developed by [7] enables a joined-up analysis of mechanical stress, cell shape complexity, and local microtubule appearance.

2

Materials

2.1 Plant and Microscopy Material

1. Arabidopsis thaliana wild-type and mutant seeds expressing a fluorescent protein-fused microtubule marker such as GFP-MAP4, GFP-MBD, GFP-TUB6, or mCherry-TUA5 [18] grown vertically on plates containing ATS media (0.5% (w/v) agar gel, 1% (w/v) sucrose) [19] in a growth chamber under long-day conditions (16 h light, 8 h dark) at 21
C. 2. Membrane or cell wall dyes such as FM4-64, propidium iodide, or calcofluor white with emission profiles distinct from the microtubule marker or dual plasma membrane and microtubule marker lines, such as LTi6b-GFP/mCherry-TUA5 [7]. 3. Low-melting agarose. 4. Glass slides and coverslips. 5. Razor blades, sharp needle, or tweezers.

2.2

Software Tools

1. Fiji, PaCeQuant, CytoskeletonAnalyzer2D, and other MiToBo tools. Fiji [20] can be downloaded from its website https://fiji.sc. Installation instructions are provided on the corresponding download page https://imagej.net/software/fiji/downloads.

Cytoskeleton Organization in Pavement Cell Morphogenesis

47

PaCeQuant, FeatureColorMapper, and some other tools mentioned below are part of MiToBo [21]. MiToBo can be installed in Fiji by activating its update site (see Note 1): (a) Run Fiji by double-clicking its executable (Windows) or running the starter (Linux/macOS). (b) Select from the menu bar “Help” –> “Update. . .” which will execute Fiji’s updater. (c) Click on the button “Manage update sites” at the bottom, select “MiToBo” from the list of sites, and then click “Close.” (d) Click “Apply Changes” in the updater window which will install MiToBo with all its plugins and all other (potentially available) updates into your local Fiji application. (e) Restart “Fiji.” The MiToBo plugins are accessible either directly via entries in MiToBo’s menu entry under “Plugins” or through the “MiToBo Runner.” 2. The FibrilTool macro [16] can be downloaded from https:// www.nature.com/articles/nprot.2014.024, Supplementary Data 1. To permanently install the macro in your local Fiji installation, copy the file to the folder “macros/toolsets” of your Fiji installation. For further details, refer to the installation instructions in ref. [16]. The extension macro to run FibrilTool in batch mode and corresponding installation instructions can be found on GitHub, https://github.com/ marionlouveaux/FibrilTool_Batch. 3. OrientationJ [17] is available in Fiji via the update site of the Biomedical Image Group, “BIG-EPFL.” Activation in Fiji follows the same procedure as described in Subheading 2.2, step 1.

3

Methods

3.1 Imaging of Cell Contours and Microtubules in Pavement Cells

The methods described here can be used to quantify pavement cell shape features and microtubule organization in cotyledons of transgenic Arabidopsis seedlings expressing a fluorescent protein-tagged microtubule marker. To support automatic segmentation, labeling of cell contours by suitable dyes or fluorescent markers is recommended (see Note 2). 1. For comparative analysis of microtubule organization and pavement cell shape, grow transgenic Arabidopsis seedlings expressing a suitable microtubule marker, such as GFP-MAP4 (see Subheading 2.1, step 1) in the respective genetic background, and label cell contours by staining with the lipophilic dye FM4–64 (see Notes 3 and 4).

48

Sandra Klemm et al.

2. Mount the sample on cover slides, and image pavement cells by generating z-stacks in two-channel settings, e.g., excite with a 488 nm and a 555 nm laser, and detect emission between 490 and 535 nm and between 560 and 630 nm for detection of GFP and FM4–64, respectively (see Notes 5–7). Adjust laser intensity relative to the fluorescence intensities of your construct of choice. Include the upper half of the cell in the z-stack, and make sure anticlinal walls are reached, but avoid including the lower half. Save images in meta-format including all metainformation, e.g., lsm or similar formats. 3. To analyze microtubule organization in response to supracellular stress patterns, perform ablation experiments by either manual wounding/ablation or by laser ablation, and track microtubule organization over time (see Notes 8 and 9). 3.2 Cell Segmentation and Quantification of Pavement Cell Shape Features

Mutations or pharmacological treatments that affect microtubule organization often also affect pavement cell shape in various ways, e.g., by reducing lobe number, lobe growth, or lobe shape. Vice versa, altered cell shapes lead to formation of different stress fields, which in turn affect microtubule organization. Microtubule organization and cell shape thus are closely linked and should be analyzed in parallel. Several tools are available to quantitatively assess pavement cell shape characteristics, such as LobeFinder [22], Gravis [23], or PaCeQuant [24]. Here, we briefly introduce how PaCeQuant can be used to quantify and visualize pavement cell shape features in an exemplary input data set from Arabidopsis wildtype seedlings and the iqd5–1 mutant, which is defective in the microtubule-associated protein IQD5 [9]. 1. Cell contours can be segmented automatically with PaCeQuant, which is optimized for 2D confocal input images (Fig. 2a, b) [24–26]. Extract the channel corresponding to the cell outline from the two-channel input images (see Subheading 3.1) in Fiji, and reduce the stack to either a single image covering the anticlinal wall or create a maximum projection of the corresponding three to four z-images (see Note 10). Save images containing cell contours according to the folder organization required for full PaCeQuant functionality (see https://mitobo.informatik.uni-halle.de/index.php/ Applications/PaCeQuant#Sample_data). 2. Open PaCeQuant and select “SEGMENTATION_ONLY” to run the automatic segmentation. As output, label images (with suffix “grayscale-result.tif”), and ZIP archives with all detected ROIs will be saved in a “result” subfolder. Manual correction of segmentation results can be conducted using the LabelImageEditor, which is part of MiToBo or in Fiji (see Notes 11 and 12). ROIs of segmented cells can be used as input for quantification of microtubule organization, e.g., in FibrilTool or

Cytoskeleton Organization in Pavement Cell Morphogenesis

Circularity

1 (round)

j

30 µm2

LEC

6 µm2

LEC

g

k 0.30 0.25 0.20 0.15 0.10 0.05 0.00

0.30 0.25 0.20 0.15 0.10 0.05 0.00

p=0.78

Area [µm2]

f

6000 5000 4000 3000 2000 1000 0

l p=0.008

30

LEC [µm2]

Segmentation

d

iqd5-1

0 (complex)

Anisotropy [a.u.]

c

Cell contour

b

Col-0

i

Input image

iqd5-1

Anisotropy [a.u.]

Col-0

a

49

25 20 15 10

0.30 0.25 0.20 0.15 0.10 0.05 0.00

p=0.07

m Circularity

h

Anisotropy [a.u.]

e

Manual ROI

5

Col-0

0.8 0.7 0.6 0.5 0.4 0.3 0.2

iqd5-1

Fig. 2 Quantification of microtubule organization and pavement cell shape. (a) Input images of cotyledons of 5-day-old Arabidopsis seedlings of wild type (Col-0) and the iqd5–1 mutant expressing the GFP-MAP4 microtubule marker (green). The plasma membrane is stained with FM-4-64 (magenta). (b) Inverted blackand-white images of microtubule signals shown in (a). Cell contours obtained by segmentation with PaCeQuant are shown in orange. (c–e) Magnifications of the framed regions in (b) with ROIs (orange) covering complete cells (c), LECs (d), or manually selected “neck” regions (e). Cell contours are depicted in green in (d) and (e). Scale bars, 100 μm (a, b) and 50 μm (c). (f–h) Quantification of microtubule anisotropy using FibrilTool in complete cells (f), LECs (g), and manual ROIs (h) of Col-0 (n ¼ 121) and iqd5–1 (n ¼ 81). Box plots were generated with BoxPlotR. (i–j) Visualization of pavement cell shape feature parameters in heat maps with FeatureColorMapper. Heat maps are shown for circularity (i) and areas of LECs (j). (k–m) Quantification of cell area (k), LEC area (l), and circularity (m) in segmented pavement cells of Col-0 and iqd5–1. Violin plots were generated with the R package PaCeQuantAna

50

Sandra Klemm et al.

CytoskeletonAnalyzer2D (see Subheading 3.3), and for quantification of pavement cell shape in PaCeQuant. 3. Load segmentation data in PaCeQuant’s batch mode, and run “FEATURES_ONLY” to calculate pavement cell shape features (see Note 13), which will be saved in text files. For visualization in box and/or violin plots and statistical analysis of pavement cell shape feature data, a supplementary R script “PaCeQuantAna” is provided (Fig. 2k–m). To overlay shape characteristics with individual cells, e.g., for displaying circularity values or the area of LECs, a FeatureColorMapper is included in the MiToBo operator runner (Fig. 2i, j) (see Note 14). 3.3 Quantification of Microtubule Organization Within Defined Regions of Interest

To quantitatively assess differences in microtubule organization, for example, between wild type and mutants with altered pavement cell morphogenesis, measures of local structure anisotropy and fiber orientation are commonly applied to user-defined ROIs, such as complete cells, LECs, or manually defined ROIs [7, 12] (Fig. 2c–e). These measurements can be performed, for example, with FibrilTool [16] (Figs. 2f–h, 3f, 4e, g) or OrientationJ [17] (Figs. 3b–e, 4f, h–k). In addition, instead of measuring properties of individual fibers, texture analysis can be applied to quantify the local structural appearance of the cytoskeleton which is particularly beneficial in the case of degraded fibers. CytoskeletonAnalyzer2D [27] (Fig. 3g–i) extracts fingerprint vectors from local binary patterns to characterize microtubule patterns, which has proven helpful in classifying microtubule-associated proteins based on their impact on cytoskeleton structure [28]. Combined with ablation experiments (Fig. 4), these experiments provide information on effects of, e.g., individual mutations on overall microtubule organization and possible differences in their response to supracellular stress patterns. To quantify microtubule organization and/or cytoskeleton patterns, proceed as follows: 1. Select ROIs, e.g., either complete cells, LECs, or manually selected regions, e.g., corresponding to areas connecting necks (Fig. 2c–e) (see Notes 15–17), and generate maximum projections of z-stack images as input data. To quantify microtubule organization using FibrilTool or OrientationJ, refer to Subheadings 3.3, steps 2 and 3, respectively. Usage of CytoskeletonAnalyzer2D is described in Subheading 3.3, step 4. 2. FibrilTool enables analysis of microtubule orientation and anisotropy from single-channel input images of microtubules or two-channel input images of microtubules and, for example, the plasma membrane. Double-click on the FibrilTool icon to select the color channels for analysis. Choose “G” for the green channel in the case of a MAP4-GFP marker being used as the microtubule marker. Load ROIs into Fiji, and select a ROI,

Cytoskeleton Organization in Pavement Cell Morphogenesis

b

iqd5-1

+90° +45°

-45°

-90

-90°

Col-0 iqd5-1 0.15 0.1 0.05 0

e

f

0.5 0.4 0.3 0.2 0.1 0.0

Col-0

iqd5-1

1

Anisotropy [a.u.]

cell 1 cell 2 cell 3 cell 1 cell 2 cell 3 iqd5-1

iqd5-1

Col-0

Col-0

cell 1 cell 2 cell 3 cell 1 cell 2 cell 3

Coherency [a.u.]

1

i

Coherency

d

0.2 0.1 0.0

0.3

1.0 0.9

0.8 0.7 0.6 0.5 0.4

Cluster probability

h

0 +90

0 +90

c

0

Orientation

g

iqd5-1

Col-0 Color-survey

Col-0

-90

a

51

0.30 0.25 0.20 0.15 0.10 0.05 0.00

Col-0 iqd5-1

Fig. 3 Overview of commonly used tools for microtubule pattern analysis. (a) Example images for microtubule analysis in pavement cells of wild type (Col-0) and the iqd5–1 mutant. ROIs (orange) label cell contours. Scale bars, 25 μm. (b–d) Results of microtubule analysis with OrientationJ. Visual representation of microtubule orientation in HSB-colored images (b). Orientation (c) and coherency (d) maps of input images and histograms of microtubule and coherency distributions, respectively, within ROIs. (e, f) Quantitative analysis of coherency by OrientationJ (e) and of anisotropy by FibrilTool (f) in pavement cells of Col-0 (n ¼ 3) and iqd5–1 (n ¼ 3). (g– i) Pattern analysis using CytoskeletonAnalyzer2D. Cluster distribution (g) and probability (h) in example pavement cells of wild type (Col-0) and the iqd5–1 mutant, calculated in tiles of 32  32 pixel. Heat map of pairwise distances between feature vectors in three cells of Col-0 and iqd5–1 each (i)

which is now marked in the image. Click on the FibrilTool icon and afterwards on the marked ROI. A log window will pop up in which information about image name (first column), average fibril angle (sixth column), and anisotropy (seventh column) is provided [16]. Repeat for other ROIs if necessary, or use the batch mode macro (https://github.com/marionlouveaux/ FibrilTool_Batch). Copy and paste measurements directly from the log window, or save data by pressing “Ctrl” + “S.” In batch mode, as output a Log.txt file will be generated containing all FibrilTool measurements (see Notes 18 and

Sandra Klemm et al.

e Cell contour

c

d

6 h after ablation

*

*

LEC

g

0.30 0.25 0.20 0.15 0.10 0.05 0.00

0.30 0.25 0.20 0.15 0.10 0.05 0.00

f

0h

6h

h

0h

6h

Coherency [a.u.]

*

0 h after ablation

0.5 0.4 0.3 0.2 0.1 0.0

Coherency [a.u.]

*

b Input image

Segmentation

0 h after ablation 6 h after ablation

Anisotropy [a.u.]

a

Anisotropy [a.u.]

52

0.5 0.4 0.3 0.2 0.1 0.0

0h

6h

0h

6h

timepoint after ablation timepoint after ablation +90° +45°

k

0

Relative distribution

Color survey

i

-45° -90°

Orientation

j

0h

6h

1 0,8 0,6 0,4 0,2 0 -90

0

+90

Angle [°]

Fig. 4 Quantification of microtubule orientation in response to tissue ablation. (a, b) Inverted black-and-white images of microtubule signals and cell outlines (orange) (a) and original input images of GFP-MAP4 signals (green) and plasma membranes (magenta) (b) from cotyledons of 5-day-old wild-type (Col-0) seedlings 0 h and 6 h after ablation. The ablation site is marked with asterisks. Scale bars, 100 μm. (c, d) Magnifications of the framed regions in (a) with ROIs (orange) of complete cells (c) or LEC (d). (e–h) Quantification of microtubule anisotropy and coherency in complete cells (n ¼ 109 cells) (e, f) or the LECs (n ¼ 105 cells) (g, h) with FibrilTool (e, g) or OrientationJ (f, h), respectively. ** denotes statistically significant differences (p < 0.01). (i, j) Results of the OrientationJ output data for microtubule orientation in HSB-colored images (i) and orientation maps (j). (k) Histogram of relative distributions of microtubule orientation angles within the ROIs shown in (i) at 0 h and 6 h after ablation

Cytoskeleton Organization in Pavement Cell Morphogenesis

53

19). Several web-based tools are available for easy and userfriendly graphical visualization of data in box plots or violin plots, such as PlotsOfData (https://huygens.science.uva.nl/ PlotsOfData/) or BoxPlotR (http://shiny.chemgrid.org/ boxplotr/), and for statistical analysis, such as https://astatsa. com/OneWay_Anova_with_TukeyHSD/. 3. The OrientationJ plugin provides multiple tools for analyzing microtubule orientation and coherency (similar to anisotropy values calculated by FibrilTool) in input images, which can be accessed via a drop-down menu. Selecting the OrientationJ Analysis tool (“Plugins” > “OrientationJ” > “OrientationJ Analysis”) opens a graphical user interface (GUI), in which analysis parameters can be defined. By default, only colorsurvey analysis is activated. To start the microtubule analysis, additionally select “Coherency” and “Orientation,” and hit the “Run” button. The output data include a color-survey image (Figs. 3b, 4i), in which microtubule orientations are depicted in HSB (hue, saturation, brightness) color coding and images showing orientation maps (Figs. 3c, 4j) and coherency maps (Fig. 3d), which are provided for the complete input image. To extract values only for a ROI within the input image, open the ROI in the respective output images (e.g., in the orientation map), and generate a histogram via Fiji’s “Analyze” > “Histogram” operator. Alternatively, the OrientationJ Measure tool can be used for analysis within defined ROIs. Load the input image and the desired ROI in Fiji, and start the OrientationJ Measure tool (“Plugins” - > “OrientationJ” - > “OrientationJ Measure”). Press “t” to add the ROI to the ROI manager, select the ROI, and click on the “Measure” button to start the measurement. The results can be copied by clicking on “Copy results” and used for graphical visualization and statistical analysis (see Note 20). 4. CytoskeletonAnalyzer2D requires a z-projection of microtubule signals and a (set of) ROI(s) of desired cells as input. The ROI file must be named similar to the corresponding z-projection file, followed by the appendix “-mask,” and input data must be provided in a defined folder setup as described in https://mitobo.informatik.uni-halle.de/index. php/Applications/CytoskeletonAnalyzer2D. Start CytoskeletonAnalyzer2D from the MiToBo Runner, choose the folder with input data, and select “IJ_ROIs” as boundary file format. Adjust the size of the tiles as well as the number of feature clusters according to your input data. For the beginning, the standard parameters are recommended (see Note 21). To start the analysis, hit the “Run” button. The output data, containing information on cluster distribution and cluster probability (Fig. 3g, h), is stored in the corresponding input folders. A

54

Sandra Klemm et al.

supplementary R script is provided for visualization of pairwise distances between cell and group distribution vectors in heat maps (Fig. 3i). 3.4 Quantification of Microtubule Enrichment at Necks and Lobes

A specific characteristic of pavement cells is the formation of lobes and necks along their anticlinal walls. Microtubules are enriched in neck regions, and the degree of neck enrichment is often diminished when shape complexity and lobe formation are reduced [6, 7, 12]. To quantify microtubule density along anticlinal walls, fluorescence intensity is measured in lobes and necks within manually defined ROIs [12] or in more automatic approaches relative to contour curvature, as applied in MtCurv [8] (see Note 22) or in Eng et al. [7], where microtubule-contour coefficients are provided (see Note 23). Here, we summarize two common methods to quantify microtubule density in lobes and necks (see Subheading 3.4, step 1) or relative to contour curvature (see Subheading 3.4, step 2) (Fig. 5). 1. Define ROIs in necks and lobes (Fig. 5a–c) by using a suitable selection tool in Fiji (see Notes 15, 16, 24, and 25). Open the channel containing microtubule signals, and select “Mean gray value” via “Analyze” > “Set Measurements. . .” for measurement of fluorescence intensities. Select all ROIs in the ROI manager, and press “M” on the keyboard or “Measure” in the ROI manager. A “Result” window will open containing output data. Load the output in the spreadsheet program of your choice, and use “Mean gray values” of your samples for statistical analysis and graphical visualization in, e.g., box plots (Fig. 5d). 2. To correlate the local curvature of the cell contour with microtubule densities, an estimation of the local curvature is extracted for each pixel position along a contour (see Note 25). In PaCeQuant, curvatures are calculated using an approach based on the algorithm proposed by Freeman and Davis [29]. A curvature value of 0 refers to a straight contour section, while positive and negative values refer to concave or convex sections, respectively. We normalize the curvature values to a range of 1.0 to 1.0, assuming a maximum curvature of +/30 (Fig. 5e, f), which was never exceeded in our data. For each contour pixel, a value for the local microtubule density is estimated, similar to the microtubule-contour coefficient quantified by [7]. To this end, around each contour pixel, a circular area with a radius of eight pixels is considered, and from all pixels lying within this circle and within the cell, the average intensity is calculated as an estimate for the local microtubule density. Intensities are scaled to map the maximal intensity value occurring in the eight-pixel-wide zone of interest along the interior of the cell contour to a value of 1.0. Once

Cytoskeleton Organization in Pavement Cell Morphogenesis

a

iqd5-1 Lobe

b

Manual ROIs Lobes and Necks

Col-0

55

Neck

d Zoom

c

+30 max. curvature

Contour curvature Microtubule signal

e

g Zoom

f

min. curvature -30

Fig. 5 Quantification of microtubule enrichment at necks and lobes. (a, b) Inverted black-and-white images of microtubule signals from cotyledons of 5-day-old wild-type (Col-0) and iqd5–1 mutant seedlings with manually selected ROIs (a), corresponding to lobes (yellow) and necks (blue) (a) as indicated in (b). (c) Magnifications of the framed regions in (a). (d) Normalized gray values of microtubule signals measured in necks and lobes of Col-0 (lobes, n ¼ 22; necks, n ¼ 25) and iqd5–1 (lobes, n ¼ 8, necks; n ¼ 13). * and ** denote statistically significant differences *, p < 0.05; (**, p < 0.01). (e) Heat map of curvatures along the cell contour (blue, 30
to red, +30
) and microtubule signals at the anticlinal to outer periclinal wall within eightpixel distance from the contour. (f) Magnifications of the framed regions in (e). Scale bars, 50 μm (a), 10 μm (c), 25 μm (e), 5 μm (f). (g) Scatterplot showing distributions of fluorescence intensity values relative to local curvature in single cells of Col-0 and iqd5–1 (e)

56

Sandra Klemm et al.

for each contour pixel values for the local curvature and the local microtubule density are given, either for each cell or all cells of a specific genotype, these values can be plotted against each other (Fig. 5g). In addition, linear regression lines can be fit to these data where the slope of the line yields a measure of correlation. Negative slopes correspond to higher-intensity values in neck regions than in lobe regions, while positive slopes indicate the opposite.

4

Notes 1. MiToBo requires Java 8 to run properly. Later Java versions may work as well, but have not been tested extensively so far. 2. Labeling of cell contours enables automatic segmentation of cells in input images and subsequent quantification of shape features, for example, by PaCeQuant [24] or similar tools. Alternatively, cell contours or other ROIs must be selected manually (see Note 15). 3. We recommend that microtubule marker constructs are introduced by crossing instead of transformation to avoid positional effects of the marker construct and to recover wild type and mutants from the F2 generation to avoid differences in microtubule marker expression due to possible transgenerational silencing. 4. Incubate seedlings for 15 min in 25 ng/ml FM4–64 for efficient staining of cell contours in cotyledons of 5-day-old seedlings. Different concentrations and incubation times may be required for seedlings at different developmental stages and for other dyes. Alternatively, lines harboring a microtubule and plasma membrane marker can be imaged (see Subheading 2.1, step 2). 5. Place a drop of ATS media or low-melting agar on the cover slide. Remove seedling from the staining solution, and rinse briefly in water. Cut the cotyledons, and place them on the cover slide. For time series analyses, use low-melting agar, and place the whole seedling on the cover slide to reduce stress. Make sure the same side of the cotyledon is imaged (adaxial or abaxial) and to capture similar regions of the cotyledons/leaves for all genotypes or treatments. 6. Image resolution and magnifications must be adjusted according to the size of the cotyledons and pavement cells. Ideally, images contain several pavement cells to provide sufficient information for meaningful quantification of shape characteristics and at the same time enable imaging of individual microtubules with sufficiently high resolution for subsequent

Cytoskeleton Organization in Pavement Cell Morphogenesis

57

analysis. To capture the entity of microtubules, z-stacks should cover the outer periclinal to anticlinal wall, and images are acquired with optimal z-distances. 7. Subsequent automatic segmentation using PaCeQuant works best with confocal input images in which cell contours are continuously labeled in single optical sections. For automatic extraction of the input plane, we recommend generating all z-stacks in the identical direction (either starting from the midplane (anticlinal walls) or from the outer periclinal surface). 8. Mount the seedlings on a cover slide as described in Note 5. Use sterile sharp tweezers, needle, or razor blade for mechanical ablation of pavement cells. Alternatively, cells can be cut using laser ablation [4]. Mark the ablation side on the cover slide, and acquire images of microtubules directly after preparing the sample as reference for the 0-h time point. Image identical regions at 2 to 6 h after ablation to track microtubule reorganization in response to supracellular stress patterns. We recommend starting with 2-h time intervals initially [7]. 9. Comparative analysis of microtubule orientation or anisotropy in wild-type vs. cell-shaped mutants only provides correlative information as shape and microtubule organization are closely connected. By combining the analysis of pavement cell shape and microtubule organization with measures of microtubule reorganization in response to ablation, additional information regarding the impact of a given mutation on microtubule-stress responsiveness is obtained. 10. We recommend generating and using macros in Fiji for automatic processing of larger sets of input data. 11. LabelImageEditor supports fast and easy corrections of label images, including the removal of wrongly annotated cells, or the correction of local segmentation inaccuracies by local removal of regions, local freehand corrections, or addition of new cell boundaries [26]. 12. Using Fiji’s ROI manager users can manually correct ROIs extracted with PaCeQuant and/or add additional ROIs, for example, if individual cells were not correctly identified in the automatic segmentation. See Note 15 for additional information on how to add ROIs. 13. PaCeQuant calculates values for 28 shape features, including area and circularity, LEC, and lobe number. In addition, PaCeQuant provides optional functionality for quantification of lobe characteristics at two- and three-cell junctions [24]. 14. Information how to use the R script PaCeQuantAna and FeatureColorMapper can be found at https://mitobo.informatik. uni-halle.de/index.php/Applications/PaCeQuant under “Complementary Tools.”

58

Sandra Klemm et al.

15. ROIs can be manually defined and saved in Fiji. Use a selection tool (e.g., oval, polygon, or freehand) to define your ROIs. To smoothen the contour, use the “Edit- > Selection- > Fit Spline” operator. To create the same distance between the selected points, you can perform the Edit- > Selection- > Interpolate operator. Add a ROI in the ROI manager by pressing “t” or save a single ROI by selecting the ROI in the ROI manager, and select “More” and “Save.” To save all ROIs at once, make sure to select all ROIs, and again select “More” and “Save,” which will create a zip file containing all ROIs. It is recommended to save ROIs for each image separately. Load your ROIs by dragging and dropping the single ROI or the ROI containing zip file into Fiji’s status bar. 16. Note that the area of LECs relative to cell area differs depending on shape complexity. In less complex cells, i.e., cells with less pronounced lobes and necks, relative LEC areas will be larger than in complex cells (Fig. 2j, l). Observed statistically significant differences in microtubule anisotropy (Fig. 2g) may thus result from differences in average ROI sizes rather than actual differences in microtubule organization [16]. 17. In addition to ROI selection, signal intensity can affect measurements, such as FibrilTool output. Therefore, try to use input images with similar signal intensities for all sample sets, and avoid regions with saturating pixels. 18. FibrilTool aims to quantify the local anisotropy of filamentous structures and, in the case of significant anisotropy, to extract the orientation of the fibers. It relies on the nematic tensor which exploits correlations in the intensity gradients of image pixels. The tensors are averaged over a region of interest, and anisotropy is calculated from the difference in the eigenvalues of the average nematic tensor. If both eigenvalues are almost equal, the gradients do not show any clear dominating direction, leading to anisotropy values close to zero. Contrarily, a large difference between the eigenvalues refers to more aligned and directed local intensity structures. In this case the eigenvector corresponding to the larger eigenvalue defines the direction of local alignment and yields an estimate of the local structural orientation. 19. Note that FibrilTool always provides an orientation and an anisotropy value for a given ROI. As also stated in the FibrilTool paper [16], these values may or may not be representative. In particular, avoid too high or too low contrasts in your ROIs. The calculated orientation is only meaningful if the anisotropy values are significant with the level of significance depending on your application. According to the FibrilTool authors, this level should be carefully determined by experts manually inspecting ROIs and extracted values for plausibility.

Cytoskeleton Organization in Pavement Cell Morphogenesis

59

20. OrientationJ [17] relies on calculation of the local structure tensor over a given ROI. It may be seen as a weighted autocorrelation matrix of the intensity gradient vectors within the ROI. The relationship between the two eigenvalues of the structure tensor characterizes the local intensity landscape, i.e., its directional coherency. If both eigenvalues are almost equal, the coherency defined by the difference of the eigenvalues divided by their sum is close to zero. If one eigenvalue is significantly greater than the other, coherency approaches 1, indicating significant anisotropy. Again the eigenvector corresponding to the larger eigenvalue refers to the dominant direction within the ROI. Like FibrilTool OrientationJ will always provide an estimate for the local coherency and orientation (see Note 19); however, both values always have to be interpreted jointly since orientations are only reliable if there exists significant anisotropy in the local structures with the ROI. 21. One important parameter of the CytoskeletonAnalyzer2D is the tile size. The size should be adjusted to the resolution of your images. Tiles of size 16  16 have proven suitable in many experiments [27, 28]. Note that each cell should be covered by a sufficiently large number of tiles to provide a robust data basis for clustering and fingerprint vector extraction. Hence, the image resolution should be high enough to account for sufficient tiles per cell. 22. The “MtCurv” protocol proposed by [8] aims to quantify microtubule densities along cell contours. Manually drawn contours of cells are first shrunk by a width of five pixels to ensure that the shrunken contours are lying entirely within the target cells. Subsequently at each contour pixel, a circle is fit passing through the pixel and its precursor and successor pixels. The circle radii yield estimates for the local curvatures of the contour, while from the positions of the circle centers, concave and convex contour sections are derived. Finally, microtubule density is measured by normalized intensities along the contour points. However, it remains unclear if just the intensity at the pixel position itself is considered or some kind of averaging in a local neighborhood takes place. 23. Eng et al. [7] propose the microtubule-contour analysis for joint analysis of local cell shape and microtubule densities. They apply a sliding window approach along the contour of a cell and measure local curvature and microtubule intensities at each window position. Local curvature is derived from the inverse of the radius of a circle fit to the contour within each window, and local microtubule intensity is measured by averaging normalized intensities along a line perpendicular to the contour at each position and directed toward the cell interior.

60

Sandra Klemm et al.

Subsequently, intensities and local curvature values of each cell are correlated by linear regression analysis, yielding the gradient of microtubule intensities over curvatures, i.e., the microtubule-contour coefficient for each specific cell. 24. Try to use the same size of polygons for all ROIs (see Notes 16 and 17). Compared to more automatic approaches (see Notes 22 and 23 and Subheading 3.4, step 2), manually selected ROIs usually exclude the tip of lobes, i.e., the microtubules closest to the anticlinal wall, while those are enriched in ROIs at necks. 25. Only include lobes and necks at two-cell-junctions between neighboring pavement cells. Cell wall properties, stress patterns, and consequently also microtubule organization significantly differ at tri-cellular junctions and at walls adjacent to stomata guard cells [4, 30].

Acknowledgments Katharina Bu¨rstenbinder would like to thank the Deutsche Forschungsgemeinschaft (DFG) (grant numbers BU2955/2-1 and BU2955/1-2), the German-Israeli Foundation for Scientific Research and Development (GIF) (grant number G-1482-423.13/2018), and the Leibniz Association for funding. References 1. Jacques E, Verbelen JP, Vissenberg K (2014) Review on shape formation in epidermal pavement cells of the Arabidopsis leaf. Funct Plant Biol 41:914–921 2. Vofely RV, Gallagher J, Pisano GD, Bartlett M, Braybrook SA (2019) Of puzzles and pavements: a quantitative exploration of leaf epidermal cell shape. New Phytol 221:540–552 3. Zhang CH, Halsey LE, Szymanski DB (2011) The development and geometry of shape change in Arabidopsis thaliana cotyledon pavement cells. BMC Plant Biol 11 4. Sampathkumar A, Krupinski P, Wightman R, Milani P, Berquand A, Boudaoud A, Hamant O, Jonsson H, Meyerowitz EM (2014) Subcellular and supracellular mechanical stress prescribes cytoskeleton behavior in Arabidopsis cotyledon pavement cells. elife 3 5. Majda M, Grones P, Sintorn IM, Vain T, Milani P, Krupinski P, Zagorska-Marek B, Viotti C, Jonsson H, Mellerowicz EJ et al (2017) Mechanochemical polarization of contiguous cell walls shapes plant pavement cells. Dev Cell 43:290–304

6. Belteton SA, Li WL, Yanagisawa M, Hatam FA, Quinn MI, Szymanski MK, Marley MW, Turner JA, Szymanski DB (2021) Real-time conversion of tissue-scale mechanical forces into an interdigitated growth pattern. Nat Plants 7:826–841 7. Eng RC, Schneider R, Matz TW, Carter R, Ehrhardt DW, Jonsson H, Nikoloski Z, Sampathkumar A (2021) KATANIN and CLASP function at different spatial scales to mediate microtubule response to mechanical stress in Arabidopsis cotyledons. Curr Biol 31:3262– 3274 8. Wong JH, Kato T, Belteton SA, Shimizu R, Kinoshita N, Higaki T, Sakumura Y, Szymanski DB, Hashimoto T (2019) Basic Proline-rich protein-mediated microtubules are essential for lobe growth and flattened cell geometry. Plant Physiol 181:1535–1551 9. Mitra D, Kumari P, Quegwer J, Klemm S, Mo¨ller B, Poeschl Y, Pflug P, Stamm G, Abel S, Bu¨rstenbinder K (2019) Microtubuleassociated protein IQ67 DOMAIN5 regulates

Cytoskeleton Organization in Pavement Cell Morphogenesis morphogenesis of leaf pavement cells in Arabidopsis thaliana. J Exp Bot 70:529–543 10. Lin W, Tang W, Pan X, Huang A, Gao X, Anderson CT, Yang Z (2022) Arabidopsis pavement cell morphogenesis requires FERONIA binding to pectin for activation of ROP GTPase signaling. Curr Biol 32(497–507): e494 11. Tang W, Lin W, Zhou X, Guo J, Dang X, Li B, Lin D, Yang Z (2022) Mechano-transduction via the pectin-FERONIA complex activates ROP6 GTPase signaling in Arabidopsis pavement cell morphogenesis. Curr Biol 32(508–517):e503 12. Lauster T, Sto¨ckle D, Gabor K, Haller T, Krieger N, Lotz P, Mayakrishnan R, Spa¨th E, Zimmermann S, Livanos P et al (2022) Arabidopsis pavement cell shape formation involves spatially confined ROPGAP regulators. Curr Biol 32:532–544 13. Zhang C, Lauster T, Tang W, Houbaert A, Zhu SF, Eeckhout D, De Smet I, De Jaeger G, Jacobs TB, Xu T et al (2022) ROPGAPdependent interaction between brassinosteroid and ROP2-GTPase signaling controls pavement cell shape in Arabidopsis. Curr Biol 32: 518–531 14. Landrein B, Hamant O (2013) How mechanical stress controls microtubule behavior and morphogenesis in plants: history, experiments and revisited theories. Plant J 75:324–338 15. Belteton SA, Li WL, Yanagisawa M, Hatam FA, Quinn MI, Szymanski MK, Marley MW, Turner JA, Szymanski DB (2021) Real-time conversion of tissue-scale mechanical forces into an interdigitated growth pattern. Nat Plants 7:989–989 16. Boudaoud A, Burian A, Borowska-Wykret D, Uyttewaal M, Wrzalik R, Kwiatkowska D, Hamant O (2014) FibrilTool, an ImageJ plug-in to quantify fibrillar structures in raw microscopy images. Nat Protoc 9:457–463 17. Pu¨spo¨ki Z, Storath M, Sage D, Unser M (2016) Transforms and operators for directional bioimage analysis: a survey. Adv Anat Embryol Cell Biol 219:69–93 18. Celler K, Fujita M, Kawamura E, Ambrose C, Herburger K, Holzinger A, Wasteneys GO (2016) Microtubules in plant cells: strategies and methods for immunofluorescence, transmission electron microscopy, and live cell imaging. Methods Mol Biol 1365:155–184 19. Lincoln C, Britton JH, Estelle M (1990) Growth and development of the axr1 mutants of Arabidopsis. Plant Cell 2:1071–1080

61

20. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B et al (2012) Fiji: an open-source platform for biological-image analysis. Nat Methods 9: 676–682 21. Mo¨ller B, Glaß M, Misiak D, Posch S (2016) MiToBo - a toolbox for image processing and analysis. J Open Res Software 4:e17 22. Wu TC, Belteton S, Pack J, Szymanski DB, Umulis D (2016) LobeFinder: a convex hullbased method for quantitative boundary analyses of lobed plant cells. Plant Physiol 171: 2331–2342 23. Nowak J, Eng RC, Matz T, Waack M, Persson S, Sampathkumar A, Nikoloski Z (2021) A network-based framework for shape analysis enables accurate characterization of leaf epidermal cells. Nat Commun 12 24. Mo¨ller B, Poeschl Y, Plo¨tner R, Bu¨rstenbinder K (2017) PaCeQuant: a tool for highthroughput quantification of pavement cell shape characteristics. Plant Physiol 175:998– 1017 25. Mo¨ller B, Poeschl Y, Klemm S, Bu¨rstenbinder K (2019) Morphological analysis of leaf epidermis pavement cells with PaCeQuant. Methods Mol Biol 1992:329–349 26. Poeschl Y, Mo¨ller B, Mu¨ller L, Bu¨rstenbinder K (2020) User-friendly assessment of pavement cell shape features with PaCeQuant: novel functions and tools. Methods Cell Biol 160:349–363 27. Mo¨ller B, Zergiebel L, Bu¨rstenbinder K (2019) Quantitative and comparative analysis of global patterns of (microtubule) cytoskeleton organization with CytoskeletonAnalyzer2D. Methods Mol Biol 1992:151–171 28. Bu¨rstenbinder K, Mo¨ller B, Plo¨tner R, Stamm G, Hause G, Mitra D, Abel S (2017) The IQD family of calmodulin-binding proteins links calcium signaling to microtubules, membrane subdomains, and the nucleus. Plant Physiol 173:1692–1708 29. Freeman H, Davis LS (1977) A corner-finding algorithm for chain-coded curves. Ieee T Comput 26:297–303 30. Belteton SA, Sawchuk MG, Donohoe BS, Scarpella E, Szymanski DB (2018) Reassessing the roles of PIN proteins and anticlinal microtubules during pavement cell morphogenesis. Plant Physiol 176:432–449

Chapter 5 Single-Cell Confinement Methods to Study Plant Cytoskeleton Pauline Durand-Smet, Antoine Chevallier, Le´ia Colin, Alice Malivert, Isaty Melogno, and Olivier Hamant Abstract Progress in cytoskeletal research in animal systems has been accompanied by the development of single-cell systems (e.g., fibroblasts in culture). Single-cell systems exist for plant research, but the presence of a cell wall hinders the possibility to relate cytoskeleton dynamics to changes in cell shape or in mechanical stress pattern. Here we present two protocols to confine wall-less plant protoplasts in microwells with defined geometries. Either protocol might be more or less adapted to the question at hand. For instance, when using microwells made of agarose, the composition of the well can be easily modified to analyze the impact of biochemical cues. When using microwells in a stiff polymer (NOA73), protoplasts can be pressurized, and the wall of the well can be coated with cell wall components. Using both protocols, we could analyze microtubule and actin dynamics in vivo while also revealing the relative contribution of geometry and stress in their self-organization. Key words Protoplasts, Single cell, Cytoskeleton, Geometry, Mechanical stress

1

Introduction In contrast to animal cells, where shape directly relates to the actomyosin cortex behavior in most cases, plant cell shape relies on the mechanical properties of cell walls. However, the synthesis and remodeling of the cell wall is highly dependent on the cytoskeleton: actin filaments and microtubules guide vesicle trafficking, and thus the delivery of the elements of the wall matrix (pectin, hemicellulose, structural and regulatory proteins, and peptides). In addition, a population of microtubules, called cortical microtubules (CMTs), are located below the plasma membrane and usually guide the trajectory of the cellulose synthase complex [1]. Because cellulose microfibrils are the load-bearing components of the wall [2], CMT behavior can indirectly modulate the mechanical stiffness and anisotropy of plant cells. Conversely, in the absence of

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_5, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

63

64

Pauline Durand-Smet et al.

microtubules, cells, tissues, and organs switch to isotropic growth [3]. Plants are thus ideal systems to relate cytoskeleton behavior to cellular and multicellular shape changes. Factors controlling the plant cytoskeleton include biochemical regulators (e.g., calcium or hormones), physical factors (e.g., light or mechanical stress), and geometry (e.g., sharp cell edges) [4, 5]. Disentangling the relative contributions of these factors can be a challenge in the living tissue and thus call for single-cell systems, working in parallel, to complement these in vivo studies. Studies at the single-cell scale in the animal field are well developed and already proved useful data for cytoskeletal quantitative analyses [6, 7]. In recent years, single-cell scale approaches have been adapted for plant systems to study a variety of responses such as rheology [8, 9], polarity and cell growth [10–12], or cell wall regeneration [12]. While imaging the cytoskeleton in an isolated plant cell can be done relatively easily, the variability of cell shape calls for an approach with better control of cell geometry for reproducible experiments. Here we present two experimental approaches that allow a systematic control of plant cell shape and can be further used for the study of the effect of mechanical stress on the cytoskeleton at the single-cell level. In addition, cell geometries that are not naturally present in plants can be tested with these methods allowing the quantification of the effect of specific geometrical parameters. In the first method, the cells are confined in agarose wells. The relatively low water potential leads to a deflated state of the protoplasts which allows them to follow the microwells contours very closely. In this configuration, the effect of the geometrical constraint on the cytoskeletal organization can be studied. In the second protocol, the protoplasts are confined inside rigid wells, and the osmotic potential of the surrounding medium can be changed to pressurize the protoplast. This leads to transiently induced mechanical stress in the cell cortex, the effect of which can be quantified on the cytoskeleton.

2 2.1

Materials Plant Material

1. Arabidopsis (Col-0) transformed with p35S::GFP-MBD, a microtubule marker [13]. 2. Arabidopsis (Col-0) transformed with p35S::FABD-GFP, an actin marker [14].

2.2

Media

1. MS medium: Murashige and Skoog medium. 2. Callus induction medium (CIM): 3.8 g/L B5 salt mix, 25 g/L glucose, 0.625 g/L MES, 1.25 mL Gamborg vitamins, pH adjusted to 5.7 with KOH, 62.5 μg/L kinetin, 625 μg/L 2,4D, 10 g/L Phytoagar.

Confined Plant Cell Method

65

3. Protoplast culture medium (PCM): ½MS supplemented with 2.0 mg/L IAA, 0.5 mg/L 2,4D, 0.5 mg/L IPAR, 0.4 M glucose [15]. 4. Solution A: D-Mannitol adjusted to 600 mOsmol/L, 2 mM CaCl, 2 mM MgCl2, 10 mM MES, pH adjusted at pH 5.5 with KOH. 5. Solution B: Identical to solution A except for D-mannitol adjusted to 280 mOsmol/L. 6. Cell wall enzyme solution: Cellulysine 17 mg/mL (Calbiochem), 17 mg/mL Cellulase RS, 0.4 mg/mL Pectolyase Y23, 1 mM L-ascorbic acid, 3.5 mg/mL BSA in solution A. The enzyme solution is sterilized by filtration. 2.3 Callus and Protoplast Generation

1. Petri dishes. 2. Micropore tape. 3. Cell strainer, 75 μm.

2.4 Microwell Fabrication

1. SU8 mold. 2. AutoCAD design software. 3. A film (or mask) with geometrical patterns. 4. Silicon wafer. 5. Polydimethylsiloxane (PDMS). 6. SU8-2025 (MicroChem). 7. Spinner. 8. SU8 developer. 9. Hot plates. 10. UV-KUB (Kloe´ SA).

2.5 Confinement in the Microwell

1. Coverslips. 2. Agarose 1.5% (w/v). 3. Wafer with SU8 patterns (see Fig. 1). 4. Clean Petri dishes. 5. PDMS posts (or several posts with the same height to elevate the wells array).

2.6

Microscopy

1. Upright scanning or spinning confocal microscope with 63× oil or 63× water immersion objectives. 2. Inverted scanning or spinning confocal microscope with 63× to 100× objectives. 3. Scanning electron microscope (FEI Sirion).

66

Pauline Durand-Smet et al.

Fig. 1 Examples of microwells used in this study. Silicon wafer with arrays of SU8 features, PDMS polymer, or agarose is poured directly on the wafer and then peeled off to create the microwell structures. Bottom-left picture: scanning electron microscope picture of the SU8 features with different geometries. Bottom-right picture: microwells ready to be loaded with protoplasts 2.7 For Image Analysis

1. Zeiss, Leica, or MetaMorph software (image acquisition). 2. Fiji with FibrilTool or OrientationJ plugin. 3. Subcellular fibrillar tool (SFT) [16].

3 3.1

Methods Seedling Growth

1. Seeds are first sterilized in 70% ethanol for 5 min, and then rinse with sterilized water three times. 2. Seeds are evenly dispersed on top of Petri dishes filled with 20 ml of ½ MS medium. The plates are then sealed with Micropore tape. 3. Seeds on plates are stratified for 48 h at 4 °C in the dark. 4. Plates are then transferred into a 19 °C continuous light chamber or directly placed in 21 °C long-day chambers (16 h light, 8 h dark). 5. Seedlings are ready to use 7 to 15 days after germination.

3.2 Protoplast Preparation from Seedlings

1. Protoplasts can be generated directly from seedlings [15]. Ten 7- to 15-day-old seedlings are incubated in 1.1 mL cell wall enzyme solution in an Eppendorf tube. Seedlings need to be fully immersed. The incubation lasts for 1 to 2 h under slow agitation on a rotating wheel (12 rpm) at room temperature or lasts overnight under agitation (40 rpm) (see Note 1). 2. After incubation, the seedling suspension is gently mixed up and down ca. five times using a 1 mL pipette to break the tissue and release the protoplasts. Once generated, handle protoplasts carefully with gentle pipetting to avoid shear stress.

Confined Plant Cell Method

67

3. Samples are centrifuged at 1000 rpm for 4 min, the supernatant is discarded, and protoplasts are washed for 1 min with 1 mL of solution A (see Note 2). 4. A second centrifugation (1000 rpm, 4 min) is performed, and most of the supernatant is discarded, keeping 200 μL of a suspension of protoplasts in solution A. 5. (Optional): Protoplasts can then be placed at 4 °C for 30 min to remove microtubule bundles. The complete digestion of the cell wall can be confirmed by the absence of calcofluor (cellulose dye) and propidium iodide (pectin dye) staining (see Note 3). 3.3 Protoplast Preparation from Calli

Because seedling protoplasts can be heterogeneous and are mainly originating from mesophyll cells, protoplasts can also be obtained from callus. Calli are first generated from seedlings, and then callus protoplasts are obtained by a combination of cell wall degradation and hypo-osmotic shock [16]. 1. To generate callus cultures, roots from 2-week-old seedlings are collected, chopped into thin sections with sharp tweezers, and then transferred onto Petri dishes containing solid callus induction medium (CIM). 2. Plates are sealed with Micropore tape and transferred at 25 °C in the dark. 3. Calli are sub-cultured every 2 weeks. This provides tissues with much more homogeneous cells than whole seedlings. 4. Calli are gently mixed in a 15 mL tube with 5.5 mL of cell wall enzyme solution and incubated for 2 h with rotation (60 rpm) at 25 °C (see Note 4). 5. The mixture is centrifuged at 800 rpm for 3 min, and the cells in the pellet are resuspended in 5 mL of washing medium solution A for 5 min. 6. Cells are centrifuged again at 800 rpm for 3 min, the supernatant is removed, and 5 mL of hypo-osmotic medium solution B is added to release protoplasts. 7. After 15 min of gentle shaking (30 rpm), protoplasts are sorted from aggregates by filtration through a 75 μm mesh. 8. Fresh protoplasts are transferred into the protoplast culture medium (PCM) and then directly loaded on the microwell array.

3.4 Microwell Mold Fabrication

To make microwells, a mold needs to be made, i.e., a silicon wafer with pillars that will create the wells in either agarose or polymers (e.g., NOA73). Such a mold is fabricated following standard microfabrication techniques [17]. Shapes of interest are designed with AutoCAD [18]. Protoplasts generated with the protocol described

68

Pauline Durand-Smet et al.

in Subheading 3.3 have diameters of about 25 ± 8 μm. Therefore, for each shape, we designed molds of between 15 and 40 μm in diameter. A silicon master with SU8 features needs to be created in a clean room following standard micro-lithography techniques. 1. The wafer is cleaned with methanol, acetone, and isopropanol and then rinsed with deionized water and dried with clean air. 2. The wafer is then heated on a hot plate at 200 °C for 15 min. 3. SU8 2025 is poured on top of the wafer and then spin-coated following the SU8 instructions and adjusting the speed such that the final height of the feature is 20 ± 0.5 μm (Fig. 1). 4. After a first baking step (2 min at 65 °C and then 5 min at 95 ° C), the wafer is exposed under UV light through the film containing the features designed with AutoCAD. 5. The wafer is then transferred onto the hot plates for another baking step (1 min at 65 °C and 5 min at 95 °C). 6. The wafer is then placed in a bath with SU8 developer for about 3 min (or until the non-exposed SU8 is fully dissolved). The height of the features can then be measured with a scanning electron microscope. 7. The silicon wafers are then rinsed with ethanol and water and air-dried and can be used to prepare the microchambers. Once the silicon wafer is ready, one can make agarose or NOA73 microwells. 3.5 Agarose Microwell Fabrication and Protoplast Confinement

1. PCM medium containing 1.5% (w/v) agarose is poured over the silicon wafer and allowed to cool to form a gel. 2. Once gelled, the agarose microwells can be carefully detached from the wafer (Fig. 1). 3. The microwell array is then deposited on PDMS posts located at the bottom of a plastic Petri dish. Fresh PCM medium is poured around the microwell array such that the air/liquid interface almost reaches the top of the microwell block (Fig. 2a). This step prevents the agarose from drying throughout the experiment. At that stage, the microwells are ready to load protoplasts. 4. Protoplasts are simply plated on top of the microwells. It takes 0.5–1 h for them to stabilize at the bottom of a microwell by sedimentation (Fig. 2a). 5. Protoplasts exhibit shapes dictated by the wells and show polymerized cytoskeletons several hours following plating, indicating that their general physiology is not perturbed. Only protoplasts that are deformed inside the microchambers are considered for analysis. Small protoplasts that fit at the bottom of the chambers without being deformed into a square, triangle, or other shapes remain spherical.

Confined Plant Cell Method

69

Fig. 2 Schematic representation of the devices used to confine protoplasts in agarose or NOA73 microwells. (a) A drop of protoplast suspension is deposited on top of the agarose microwells. After sedimentation of the protoplasts (about 30 min to 1 h), a coverslip is carefully added, and the sample is ready to image under an upright microscope. (b) A drop of a solution containing a suspension of protoplasts is deposited on microwells in a glass-bottom dish (e.g., an Ibidi dish) (1). Close-up of microwells containing protoplasts in a 600 mOSMOL mannitol solution (2). Once in microwells, protoplasts are ready to be imaged (4). In this scheme, protoplasts are pressurized using a hypo-osmotic solution (280 mOsmol mannitol (3), as in [20]). Adapted from [18] and [20] with permission and under a Creative Commons Attribution 4.0 International License)

3.6 NOA73 Microwell Preparation and Protoplast Confinement

1. PDMS wafers are pressed into a drop of ultraviolet (UV) curable monomers of Norland Optical Adhesive 73 (NOA73), with the microwell structure facing down, beforehand deposited in a glass-bottom dish (e.g., Ibidi dish) (35 mm). 2. UV curing is performed either with UV-KUB 9 (200 mW/ cm2) for 30 s or with UV-KUB 2 (40 mW/cm2) for 3 min. 3. Glass-bottom dishes are filled with 3 mL of water, and the PDMS wafers are carefully removed with tweezers to avoid the formation of air bubbles in the microwells. For a more detailed protocol, see [19]. 4. Microwells in glass-bottom dishes are washed with solution A before protoplast confinement. 5. After removing solution A from the dish, 50 μL of protoplast solution is deposited on the microwells and allowed to sediment at room temperature for 5 min. 6. The dish is then centrifuged at 300 × g for 3 min (using a plate holder). This sedimentation step increases the number of confined protoplasts in the microwells. 7. After centrifugation, 3 mL of solution B are slowly added to induce a hypo-osmotic shock (Fig. 2b).

70

3.7

Pauline Durand-Smet et al.

Imaging

Confined protoplasts can be imaged from the top with an upright microscope or from the bottom with an inverted microscope (see Note 5). 1. Once the protoplasts are plated into the microchambers, a coverslip is carefully added on top of the array before imaging under an upright microscope (Fig. 2a). 2. Many different cell compartments, organelles, or structures can be imaged once the protoplasts are lodged in the microchambers. In our previous studies [18, 20], we focused on the cytoskeleton (actin and microtubules). 3. Protoplasts lodged in microchambers are imaged with a Zeiss LSM780 or Zeiss LSM880 Airyscan confocal laser scanning microscope with a 63× oil objective and z-stacks of cells with 0.18 μm intervals are obtained for 3D reconstruction. 4. Protoplasts can also be imaged using an inverted microscope as illustrated in Fig. 2b. Images are acquired on a spinning-disk confocal microscope (Yokogawa W1 on Nikon T1E), an inverted confocal LSM710 microscope (Zeiss) using 100× lens (oil immersion, N.A. = 1.45 [Nikon] and 1.46 [Zeiss]), or an LSM800 inverted confocal microscope (Zeiss) using 63× lens (oil immersion, N.A. = 1.4; Fig. 2b). Examples of protoplasts confined in microwells with both methods are provided in Fig. 3. 5. Protoplasts in agarose wells display low water potential which allows them to follow the microwells contours very closely (Fig. 3a), and the effect of geometry can be investigated. 6. For protoplasts confined inside rigid NOA73 wells, the osmotic potential of the surrounding medium can be changed to pressurize the protoplast against the stiff walls of the well (Fig. 3b). Only the vertical sides of the protoplasts in the well are flattened upon pressurization. Mechanical stress builds up in the free curved edges, and the effect of tensile stress in those areas can be analyzed. 7. The protoplasts are usually observed 30 min to 1 h after being plated; thus the organization we observed forms on a time scale faster than hours (see Note 6). In these experiments, the steady-state organization of the cytoskeletal network is observed (see Note 7).

3.8 Quantification of Cytoskeleton Organization: Average Cytoskeleton Organization

1. Images from the LSM780 confocal microscope were processed with ImageJ [21]. 2. Zen 2.3 software is used to process the images acquired with the LSM880 Airyscan confocal microscope. The software processes all airy channels in order to obtain images with enhanced spatial resolution in 3D [22].

Confined Plant Cell Method

71

Fig. 3 Examples of confined protoplasts using both methods. (a) Protoplast expressing the MBD-GFP reporter in a rectangular agarose microwell. Orthogonal sections show that the top, bottom, and sides of the protoplast are flattened by the confinement (scale bar = 10 μm). (b) Pressurized protoplasts in NOA73 microwells: p35S:: GFP-LTI6b expressing protoplasts display fluorescence in their plasma membrane (mostly), allowing the assessment of the protoplast deformation upon confinement. Orthogonal sections (right panels) show that only the sides of the protoplasts are flattened by the confinement and the top and bottom of the protoplast display a constant curvature. The bottom protoplast is larger than the top one and is thus more deformed upon confinement (scale bar = 10 μm). Adapted from [18] and [20] with permission and under a Creative Commons Attribution 4.0 International License

3. After processing, z-stacks are projected on one plane with maximum intensity projection with ImageJ to create 2D images. 4. The ImageJ plugin FibrilTool [23] is used to quantify the average orientation and the anisotropy of the cytoskeletal networks. The average orientation gives information on the direction of the network, while the anisotropy gives an estimation of how well the filaments are aligned with each other. 5. Regions of interest are drawn manually to define the contours of the cells on the 2D pictures, and nematic tensors of microtubule and actin filament arrays inside the region of interest (ROI) are obtained using FibrilTool (Fig. 4a). 6. FibrilTool directly provides the average orientation of the network inside the ROI and the anisotropy of the network with a score between 0 and 1 for every cell. The full angle distribution can be obtained with the OrientationJ plugin [24]. 3.9 Quantification of Cytoskeleton Organization: Local Cytoskeleton Organization

1. The first five slices (5 μm) of the confocal stack are used to make a maximal projection. This removes most of the cytoplasmic signal in the images, including cytoplasmic microtubules. 2. Then, all projections corresponding to one technical replicate are assembled as a stack, projections are aligned with the ImageJ plugin AlignSlice, and all images are cropped to keep the central part of the protoplast (to avoid artifacts on the sides of the cells due to the projection).

72

Pauline Durand-Smet et al.

Fig. 4 Cytoskeleton organization quantification. (a) Examples of the FibrilTool analysis run on a spherical protoplast expressing the MBD-GFP reporter (top left) and a protoplast confined in a rectangular well (top right). The yellow line represents the ROI in which CMT orientation has been performed (scale bar = 10 μm). Distribution of MT average angle measured with FibrilTool (left, spherical protoplasts; right rectangular protoplasts). (b) Microtubule orientation analysis: example of microtubule signal (p35S::GFP-MBD) in a protoplast confined in 15 × 20 μm microwell (scale bar = 5 μm). The dotted red line represents the ROI in which CMT orientation has been performed (left). The orientation of CMTs in each ROI is color-coded (middle). Polar histograms represent the CMT angle distribution for the protoplast. Each bar corresponds to an angle range of 9°. Schematic representations of CMT orientation are indicated on the plot. Analysis has been performed using SFT (right). Adapted from [20] with permission and under a Creative Commons Attribution 4.0 International License

3. Finally, a Gaussian blur (radius, 2.00 pixels) is applied on the images. Microtubule orientation analysis is performed using subcellular fibrillar tool (SFT [16]). This method is based on nematic tensors and provides orientation and anisotropy of the microtubule arrays. 4. Each protoplast is divided in ROIs, and the orientation of each ROI is then analyzed. In our study, this represented 394 ROIs (i.e., CMT fragment angles) per protoplast. The orientation is measured from an imaginary horizontal line, and the angle between this line and the ROIs is calculated. 5. A polar histogram is constructed, displaying the global probability for each orientation to belong to a class of angles, ranging from 0° to 90° (Fig. 4b). Thus, ten classes of angles of 9° are formed, from 0°–9° to 81°–90°.

Confined Plant Cell Method

73

6. We use the probability density function where (probability) = sum of (probability density function) × (bin width) over bins. When ten classes of angles (0–90°) are defined, bin width is (π/2)/10  0.157. 7. The OrientationJ plugin, distribution in the ImageJ software, can be used for microtubule orientation analysis. As in SFT, protoplasts are divided into several ROIs (one ROI = eight pixels). The orientation of CMTs in each ROI is measured. For each technical replicate, all of the orientations are then distributed into six classes of angles, ranging from 0° to 90°, each class measuring 15°. 8. Data are normalized and histograms are constructed, displaying the proportion of ROIs in each class of angles. Both of these methods (i.e., SFT and OrientationJ plugin) allow a very local analysis of each microtubule orientation.

4

Notes 1. Digestion time highly depends on the local room temperature; thus, the agitation time should be adjusted accordingly. Here, room temperature is 20–25°C, and overnight corresponds to 20°C. 2. When the quality of the protoplasts is poor, it is necessary to add an extra washing step with solution A after the enzyme digestion step. 3. We could not detect cell walls in protoplasts that were cast in the wells for the first few hours. It usually starts forming between 12 and 24 h under standard culture conditions [25]. 4. When calli are transferred to the enzyme solution for cell wall digestion, crushing the calli against the tube side with a grinding pestle improves protoplast production. 5. When using such a minimal growth medium (see the composition of solution A and B), the protoplast survival rate is low (many dead protoplasts are visible 2 days after being confined in the microwells, using both methods). 6. When using an upright microscope with agarose microwells, the top coverslip needs to be deposited really gently to avoid damaging cells. If imaging is done later, close the Petri dish with Parafilm to avoid dehydration. 7. Chlorophyll exhibits very strong autofluorescence in red. Using protoplasts derived from calli can be convenient as they do not contain chloroplasts. Conversely, chloroplast autofluorescence shifts to other wavelengths when mesophyll protoplast is dying providing a natural proxy for the protoplast viability.

74

Pauline Durand-Smet et al.

References 1. Paredez AR, Somerville CR, Ehrhardt DW (2006) Visualization of cellulose synthase demonstrates functional association with microtubules. Science 312:1491–1495. https://doi.org/10.1126/science.1126551 2. Zhang Y, Yu J, Wang X et al (2021) Molecular insights into the complex mechanics of plant epidermal cell walls. Science 372:706–711. https://doi.org/10.1126/science.abf2824 3. Corson F, Hamant O, Bohn S et al (2009) Turning a plant tissue into a living cell froth through isotropic growth. Proc Natl Acad Sci U S A 106:8453–8458. https://doi.org/10. 1073/pnas.0812493106 4. Ambrose C, Allard JF, Cytrynbaum EN, Wasteneys GO (2011) A CLASP-modulated cell edge barrier mechanism drives cell-wide cortical microtubule organization in Arabidopsis. Nat Commun 2:430. https://doi.org/10. 1038/ncomms1444 5. Kirchhelle C, Garcia-Gonzalez D, Irani NG et al (2019) Two mechanisms regulate directional cell growth in Arabidopsis lateral roots. eLife 8. https://doi.org/10.7554/eLife. 47988 6. The´ry M, Pe´pin A, Dressaire E et al (2006) Cell distribution of stress fibres in response to the geometry of the adhesive environment. Cell Motil Cytoskeleton 63:341–355. https://doi. org/10.1002/cm.20126 7. Bao M, Xie J, Katoele N et al (2019) Cellular volume and matrix stiffness direct stem cell behavior in a 3D microniche. ACS Appl Mater Interfaces 11:1754–1759. https://doi. org/10.1021/acsami.8b19396 8. Durand-Smet P, Chastrette N, Guiroy A et al (2014) A comparative mechanical analysis of plant and animal cells reveals convergence across kingdoms. Biophys J 107:2237–2244. https://doi.org/10.1016/j.bpj.2014.10.023 9. Durand-Smet P, Gauquelin E, Chastrette N et al (2017) Estimation of turgor pressure through comparison between single plant cell and pressurized shell mechanics. Phys Biol 14: 055002. https://doi.org/10.1088/14783975/aa7f30 10. Zaban B, Liu W, Jiang X, Nick P (2014) Plant cells use auxin efflux to explore geometry. Sci Rep 4:5852. https://doi.org/10.1038/ srep05852 11. Chan J, Mansfield C, Clouet F et al (2020) Intrinsic cell polarity coupled to growth axis formation in tobacco BY-2 cells. Curr Biol. https://doi.org/10.1016/j.cub.2020.09.036

12. Chen L, Han Z, Fan X et al (2020) An impedance-coupled microfluidic device for single-cell analysis of primary cell wall regeneration. Biosens Bioelectron 165:112374. https://doi.org/10.1016/j.bios.2020. 112374 13. Hamant O, Heisler MG, Jo¨nsson H et al (2008) Developmental patterning by mechanical signals in Arabidopsis. Science 322:1650– 1655. https://doi.org/10.1126/science. 1165594 14. Ketelaar T, Allwood EG, Anthony R et al (2004) The actin-interacting protein AIP1 is essential for actin organization and plant development. Curr Biol 14:145–149. https://doi. org/10.1016/j.cub.2004.01.004 15. Mathur J, Koncz C, Szabados L (1995) A simple method for isolation, liquid culture, transformation and regeneration of Arabidopsis thaliana protoplasts. Plant Cell Rep 14:221– 226. https://doi.org/10.1007/BF00233637 16. Tsugawa S, Hervieux N, Hamant O et al (2016) Extracting subcellular fibrillar alignment with error estimation: application to microtubules. Biophys J 110:1836–1844. https://doi.org/10.1016/j.bpj.2016.03.011 17. Weibel DB, Diluzio WR, Whitesides GM (2007) Microfabrication meets microbiology. Nat Rev Microbiol 5:209–218. https://doi. org/10.1038/nrmicro1616 18. Durand-Smet P, Spelman TA, Meyerowitz EM, Jo¨nsson H (2020) Cytoskeletal organization in isolated plant cells under geometry control. Proc Natl Acad Sci U S A 117:17399–17408. https://doi.org/10.1073/pnas.2003184117 19. Gao X, Stoecklin C, Zhang Y et al (2018) Artificial microniche Array with spatially structured biochemical cues. Methods Mol Biol 1771:55–66. https://doi.org/10.1007/ 978-1-4939-7792-5_5 20. Colin L, Chevallier A, Tsugawa S et al (2020) Cortical tension overrides geometrical cues to orient microtubules in confined protoplasts. Proc Natl Acad Sci U S A 117:32731–32738. https://doi.org/10.1073/pnas.2008895117 21. Schindelin J, Arganda-Carreras I, Frise E et al (2012) Fiji: an open-source platform for biological-image analysis. Nat Methods 9: 676–682. https://doi.org/10.1038/nmeth. 2019 22. Korobchevskaya K, Lagerholm B, ColinYork H, Fritzsche M (2017) Exploring the potential of airyscan microscopy for live cell

Confined Plant Cell Method imaging. Photo-Dermatology 4:41. https:// doi.org/10.3390/photonics4030041 23. Boudaoud A, Burian A, Borowska-Wykre˛t D et al (2014) FibrilTool, an ImageJplug-in to quantify fibrillar structures in raw microscopy images. Nat Protoc 9:457–463. https://doi. org/10.1038/nprot.2014.024 24. Rezakhaniha R, Agianniotis A, Schrauwen JTC et al (2012) Experimental investigation of collagen waviness and orientation in the arterial

75

adventitia using confocal laser scanning microscopy. Biomech Model Mechanobiol 11:461– 473. https://doi.org/10.1007/s10237-0110325-z 25. Kuki H, Higaki T, Yokoyama R et al (2017) Quantitative confocal imaging method for analyzing cellulose dynamics during cell wall regeneration in Arabidopsis mesophyll protoplasts. Plant Direct 1:e00021. https://doi. org/10.1002/pld3.21

Chapter 6 Documentation of Microtubule Collisions with Myosin VIII ATM1 Containing Membrane-Associated Structures Eduard Belausov, Vikas Dwivedi, Sela Yechezkel, Sefi Bar-Sinai, and Einat Sadot Abstract Collisions of microtubules with membrane-associated structures containing myosin VIII were recently described, and these data suggested that such collisions can happen between microtubules and other membrane-associated proteins. Such collisions may contribute to a coordinated organization between microtubules and membrane-associated proteins especially in cases of low lateral diffusion rates of the protein. Coordinated organization of cortical cytoskeleton and membrane structures can have consequences on membrane compartmentalization and downstream signaling. Here we describe a way to analyze collisions of cortical microtubules and membrane-associated proteins by confocal microscopy. In addition, we describe a tool to measure and quantify these collisions. Key words Collisions, Microtubules, Myosin VIII, Membrane-associated proteins, Live cell imaging, Confocal microscopy

1

Introduction Plant cortical MTs form parallel arrays beneath the cell plasma membrane [1–4]. The cytoplasm in plant cells forms a narrow layer between the plasma membrane and the tonoplast, which is the membrane of the large vacuole. The vacuole is responsible for the turgor pressure pushing against the cell wall, leaving a narrow space for the cytoplasm. The plasma membrane is rich in protein clusters of various sizes, densities, and functions, as well as in various nanodomains [5], making the inner surface area of the membrane rough and dynamic. Therefore, cortical microtubules that elongate in the cytoplasm beneath the membrane have to find their way between obstacles of different sizes and densities, making occasional collisions conceivable events. Coordinated localization of membrane proteins and cortical microtubules are often observed, and these events are usually explained to be a

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_6, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

77

78

Eduard Belausov et al.

consequence of the putative fences that microtubules form which then restrict the free lateral diffusion of plasma membrane proteins [6–8]. However, some membrane proteins like myosin VIII, which exhibit a low lateral diffusion, can be found in coordinated localization with microtubules [9]. Therefore, collisions of microtubules with such protein containing clusters might contribute to the observed coordinated localization [9]. In animal cells, it has been shown that microtubules collide with focal adhesions [10–13]. Focal adhesions are built from multiple protein complexes that bind actin stress fibers to the membrane via transmembrane-domain proteins from the integrin family, which in turn bind components of the extracellular matrix. In animal cells, MTs are likely to be targeted to focal adhesions along actin filaments [12]. In plants, it is still to be determined. We describe here, step by step, how to follow and quantify collisions between microtubules labeled by EB1 and membraneassociated clusters containing the myosin VIII ATM1 tail.

2

Materials Prepare all solutions in double-distilled water, and all reagents are of analytical grade. Prepare media and solutions, and store at the indicated temperature.

2.1 Seeds and Plant Growth Materials

1. Seeds of Nicotiana benthamiana. 2. Square plastic pots of 10 × 10 × 9 cm. 3. Germination soil contained three layers: 1/3 of the pot, on the bottom Tuff “Odem 93,” than 2/3 peat “Ram 8” and on top a thin layer of “Zohar 4” which contains crushed Styrofoam with peat. Detailed information on soil composition can be found at https://tuff.co.il/ (see Note 1).

2.2 Plasmids and Agrobacterium Strains

1. Plasmids used for infiltration (Table 1). 2. Agrobacterium tumefaciens strain GV3101 (see Note 2).

Table 1 The plasmids used in this work Constructs

Plasmid

Selection

References

mCherry/ATM1IQ-tail domain (At3g19960)

pART27

Spectinomycin

[9, 15]

EB1b-YFP (At5g62500)

pART27

Spectinomycin

[9]

Measurements of Microtubule and Membrane Protein Collisions

2.3 Media, Buffers, and Solutions

79

1. YEB medium: 1.0 g/L yeast extract, 5.0 g/L beef extract, 5.0 g/L peptone, 5.0 g/L sucrose, 0.5 g/L MgSO4.7H2O, pH 7.0. Add 900 mL of ddH2O in a beaker, then add all the ingredients, and mix until completely dissolved. Adjust pH with 1 M NaOH. Complete to 1 L with ddH2O using a cylinder. Aliquot into 250 mL bottles. Sterilize by autoclave. 2. Antibiotic stock solutions in ddH2O: 25 mg/mL gentamycin; 100 mg/mL spectinomycin. Freeze in aliquots at -20 °C. 3. Stock solution of 1 M acetosyringone: dissolve 196 mg in 1 mL dimethyl sulfoxide (DMSO). Freeze in aliquots at -20 °C. 4. Stock solution of 500 mM MES: dissolve 4.88 g of MES in 50 mL of ddH2O. Store at 4 °C. 5. Stock solution of 20 mM Na3PO4.12H2O (trisodium orthophosphate): dissolve 0.38 g of Na3PO4.12H2O in 50 mL of ddH2O. Store at 4 °C. 6. Infiltration medium (made fresh): In a 50 mL tube, add 250 mg D-glucose, 5 mL MES stock solution, 5 mL Na3PO4.12H2O stock solution, 5 μL 1 M acetosyringone stock solution; make up to 50 mL with ddH2O, and mix well (see Note 3).

2.4 Microscopy and Image Analysis

1. Double-sided adhesive tape. 2. 1 mL needleless syringe. 3. Imaris software package (Bitplane, Oxford Instruments). 4. Python simulation tool. 5. Confocal microscope with a 63× water immersion objective.

3

Methods Carry out all procedures at room temperature with the exception of the growth of Agrobacterium and the storage of stock solutions. Agrobacterium infiltration to Nicotiana benthamiana is done according to Sparkes and colleagues [14].

3.1 Nicotiana benthamiana Plant Growth

1. Germinate 20–30 seeds in 10 cm square plastic pots in a controlled growth room at 23 °C with 16 h light and 8 h dark photoperiod. 2. After 10 days thin the seedlings, and leave four seedlings per pot. 3. Three weeks after thinning, the plants will have reached the optimal developmental stage to be used for Agrobacterium infiltration (i.e., 4–5 weeks old). At this stage, the plants have five to six true leaves and are without flower buds (see Note 4).

80

Eduard Belausov et al.

3.2 Culturing and Preparation of the A. tumefaciens Suspension

1. Use 50 μL of bacterial culture from a glycerol stock (see Note 5) of the A. tumefaciens strain GV3101 containing the desired plasmid (Table 1). Inoculate 2 mL YEB starter cultures containing 25 mg/L gentamycin and 100mg/L spectinomycin in a capped 12 mL plastic culture tube. Incubate the starter cultures at 28 °C under agitation (220 rpm) for 24–48 h to allow growth. 2. Inoculate 50 μL of starter cultures into 10 mL YEB media containing the same antibiotics in a 50 mL polypropylene tube with screw cap, and grow over night at 28 °C with shaking (220 rpm). 3. Measure the OD600 of the overnight cultures with a spectrophotometer. The OD should be in the range of 1.0–1.5. 4. Pellet the 2 mL culture cells by a gentle centrifugation for 10 min at 1000–3200 × g. Wash the cells twice with infiltration media. Finally, suspend the cells in infiltration media to the required final OD600 = 0.6–0.7 for each marker, and mix EB1b-YFP/mCherry-ATM1 at 1:3 for infiltration (see Note 6).

3.3 Leaf Selection, Agroinfiltration, and Plant Incubation

1. Before infiltration (2–3 h), give the plants an extra amount of water. 2. Use the third, fourth, and fifth leaf (from the bottom) for infiltration. 3. Mark the leaves to which infiltration is done by drawing a circle with a marker pen. 4. Make a small hole with a needle at the abaxial side of the leaf in the middle of the circle. Fill a 1 mL needleless syringe with the Agrobacterium culture; tighten the syringe onto the hole on the abaxial side while applying counter pressure to the adaxial side of the leaf with your finger (see Note 7). 5. Push the piston slowly so that the Agrobacterium suspension goes into the intercellular spaces of the leaf, and make sure not to infiltrate too much; an area of 5–8 mm in diameter is sufficient (see Note 8). 6. Keep the plants in the growth room away from direct lamp light. 7. After 24–48 h, the expression of fluorescent protein can be seen using a confocal microscope.

3.4 In Planta Localization of Proteins

1. Harvest the infiltrated leaves 48 h after infiltration. 2. Cut a 5 × 5 mm piece of leaf (as flat as possible and with no vascular bundles), and place with the abaxial side facing the coverslip on a microscope slide, with a double-sided adhesive tape pasted on it (Fig. 1a–c) (see Note 9).

Measurements of Microtubule and Membrane Protein Collisions

81

Fig. 1 Mounting of N. benthamiana leaf pieces on the slide. (a) Cut double-sided tape 22 × 40 mm. (b) Paste it on the glass. (c) Take off upper paper. (d) Put two 5 × 5 mm flat pieces of leaf from a region that was infiltrated. (e) Put a drop of water on each piece. (f) Cover with a coverslip

3. Place one or two such pieces in the center of the slide, gently pressing a piece of leaf to the adhesive surface, and mount them with a small water drop (Fig. 1d–e) (see Note 10). 4. Cover object with a coverslip, and stick the glass along the edges of the slide (Fig. 1f) (see Note 11). 5. Use a 63× water immersion objective. 6. Adjust image acquisition parameters for the best signal-noise ratio, and use line average mode for image quality improving (see Note 12). 7. Focus specimen on the upper part of the epidermal cells for the best imaging of cortical microtubules beneath the plasma membrane (see Note 12). 8. Capture time series movies using 512 × 512 pixels of frame size with line average of 2. 9. Take 50 frames with 1 s/frame rate. 10. EB1-YFP excitation (like GFP in this case) 488 nm, emission 500–550 nm, RFP excitation 552 nm, emission 560–640 nm (see Note 13). 3.5 Image Analysis with Imaris

1. Run Imaris and load the images files created by the microscope (we work with a Leica SP8). 2. Enter a movie from Imaris Arena. 3. Create a Spot object for each channel (see Note 14) by clicking the Spots icon ( ) and selecting the required parameters (see Note 15). For the membrane protein channel, use the Different Spot Diameters option. 4. If you are not going to use the batch option (see Note 16), save all Imaris statistics for both channels. The statistics files are used by the simulation tool later on. In the following instructions, we assume that the Imaris batch option is being used.

82

Eduard Belausov et al.

5. To use the batch option, click the creation icon ( Store Parameters for Batch.

), and select

6. Check carefully the accuracy of spots and tracks captured by Imaris. Make manual corrections if necessary. 7. Click the Arena icon and save the analyzed movie. 8. In the Arena window, right-click the creation icon ( ), and select Run Batch. Imaris creates Spots and Tracks for all movies in the Arena window and creates a batch icon. 9. Right-click the batch icon ( ), and select Export Statistics. Imaris creates an Excel file with statistics data for all the movies in the Arena window. 3.6 The Simulation Tool

The simulation tool (written in Python [9]) provides data and animation that help focus on selected tracks and adds analysis data that is not provided by Imaris. Track and spot statistical data from Imaris were used to draw the microtubule tracks (yellow lines or white lines for tracks that encountered collisions) and membrane proteins spots (red circles or white circles for membrane protein spots which encountered collisions with microtubules) (Fig. 2 and see Note 17). The simulation tool also enables us to focus on one selected microtubule track by using different color and line widths. In addition to statistical data from Imaris, the simulation provides the following information: 1. Collision distance—the distance between the border of the membrane protein spot and the center of the EB1 signal for each pair of microtubule signal and membrane protein spot (colliding pair) in all frames. The border of the membrane

Fig. 2 Demonstration of the process of quantifying collisions of microtubules with spots of ATM1 tail. (a) A snapshot from the time lapse movie. In red are spots of the ATM1 tail and in green, EB1, a marker for plus end elongating microtubules. (b) The spots and tracks captured by Imaris. (c) Frame 50 from the simulation showing in white spots and tracks which have undergone a collision

Measurements of Microtubule and Membrane Protein Collisions

83

protein spots were estimated by using the Different Spot Diameters option in Imaris. Whenever the distance was below the membrane protein spot radius plus a certain value (a configuration parameter, here we used 0–100 nm), we considered it to be a collision, and the colliding pair was highlighted. 2. The number of collisions per microtubule track and per membrane protein spot and the percentage of colliding tracks or spots. 3. The speed of the microtubule tracks near the collision; near collision duration is set by two configuration parameters, and these are the number of frames before and after the collision period. 4. The difference between the speed and straightness of the microtubule track before and after the collision compared to the whole track. 5. The number and percentage of microtubule tracks that went through acceleration or deceleration, or were not affected by the collision, compared to the average speed of the whole track. 6. The percent of acceleration events that were associated with a decrease in straightness. 7. The maximum distance by which membrane proteins were shifted from their initial position. These data were used as a measure for lateral diffusion. 8. The simulation tool also enabled us to export data to csv files for further processing as well as saving animation gif or mpg files. For the full description of the simulation data, please refer to Subheading 3.8. 3.7 Using the Simulation Tool

1. To run the simulation, open a terminal window, go to the folder where ImarisSim.py is located, and type “python ImarisSim.py” (see Note 18); a new window should be displayed (Fig. 3). 2. Control window parameters: • EB track file—The absolute path of the Imaris statistics Excel file containing microtubule track information. • ATM track file—The absolute path of the Imaris statistics Excel file containing membrane protein spot information. • Collision distance—The distance between a microtubule head and a membrane protein under which collision was detected. • Collision range—The number of time frames before and after a collision that define a collision proximity.

84

Eduard Belausov et al.

Fig. 3 The control window of the simulation

• Marker size factor—Used to control spot marker display size in the simulation window. • Animation interval—The number of milliseconds between two displayed frames. • Highlight track—A track number to be highlighted. Used to focus on a specified track behavior. • Highlight marker—A spot marker number to be highlighted. Used to focus on a specified spot behavior. • Fix ATM pos*—If selected, membrane protein spot initial position is displayed throughout the entire simulation. • Fix ATM diam*—If selected, membrane protein spot initial diameter is displayed throughout the entire simulation. • Start sim button—Used to start the simulation. • Pause button—Used to pause the simulation. • Save data button—Used to save simulation data into a csv file. • Create movie button—Used to create a mpg or gif movie file of the simulation. • Exit button—Used to exit the simulation. * Done for purposes of collision presentation only but not for measurements. This is because sometimes at frame 50:50 a white spot no longer touches a white track. We found that freezing the positioning of the spots at time 0 provides the visual justification of the white color given by the simulation (see Note 14).

Measurements of Microtubule and Membrane Protein Collisions

85

3. To start the simulation, click the Start sim button. The simulation tool start playing the movie, drawing spots, tracks, and collisions. 4. To pause the simulation, click the Pause button. Click Play (same button) to continue the simulation. 5. The current frame and the total number of frames are displayed in the upper left side of the window. When the last frame is displayed, the simulation stops. 6. To save the analysis data in a csv file, click the Save data button. The csv file is created in the current folder. 7. To create a movie file in mpeg or gif format, click the Create movie button. 8. A movie file with the entire simulation is created in the current folder. 9. To exit the simulation, click the Exit button. 3.8 The Simulation Measurements File

The csv file created by the simulation contains the data measurements obtained during the simulation. It can be opened with Excel or any other text editor. It contains one Excel tab with four sections: 1. Collisions data • EB track Id—The track number of a microtubule assigned by Imaris. • ATM Id—The spot number of a membrane protein spot assigned by Imaris. • Collision length—The distance of the overall movement of the microtubule tip throughout the entire collision frames (see Note 19). • Collision displacement—The distance between the microtubule tip position at the first collision frame and at the last collision frame. • Collision straightness—Collision displacement divided by collision length. • Track straightness—Track displacement divided by track length (calculated by Imaris). • Straightness diff—Collision straightness.

straightness

minus

track

• Collision speed (μm/s)—Track mean speed during the collision. • Track speed (μm/s)—Track mean speed (calculated by Imaris). • Speed diff—Collision speed minus track speed.

86

Eduard Belausov et al.

2. EB collisions totals • Track Id—The track number of a microtubule assigned by Imaris. • Collisions—Number of collisions of the microtubule tip detected throughout the simulation. • Collision %—Percentage of colliding microtubule tips out of all microtubules tracks. • Straightness—Track straightness (track displacement divided by track length calculated by Imaris). • Speed (μm/s)—Track mean speed (calculated by Imaris). 3. ATM collisions totals • Spot id—The membrane protein spot number assigned by Imaris. • Collisions—Number of collisions of the membrane protein detected throughout the simulation. • Collision %—Percentage of colliding membrane protein spots out of all membrane protein spots. • Area—The area of the membrane protein spot (calculated by Imaris). • Max movement—The maximum distance from the initial position made by the membrane protein spot during the simulation. • Max movement second—The second in which max movement was measured for the membrane protein. 4. Collisions totals • Accelerated—Number of microtubules showing accelerated rate of elongation after a collision. • Accelerated %—Percentage of accelerated microtubules out of all microtubule tracks. • Decelerated—Number of microtubules showing decelerated rate of elongation after a collision. • Decelerated %—Percentage of decelerated microtubules out of all microtubule tracks. • Unaffected—Number of microtubules for which rate of elongation did not change after a collision. • Unaffected %—Percentage of unaffected microtubules out of all microtubule tracks. • Accelerated/decreased straightness—Number of accelerated microtubules for which straightness decreased after collision.

Measurements of Microtubule and Membrane Protein Collisions

87

• Accelerated/decreased straightness %—Percentage of accelerated microtubules for which straightness decreased after collision out of all accelerated microtubule tracks.

4

Notes 1. We optimized the combination of soil for germination and growth of N. benthamiana plants. The combination of this soil worked best. 2. The GV3101 strain is commonly used in our laboratory. Other Agrobacterium strains including EHA105 can also be used. 3. The infiltration media should be freshly prepared. The use of old solution might result in lower expression and sometimes in no expression. 4. Plants should be at the correct stage. Plants at early stages are fragile and may not recover from the infiltration and at later growth stages show less expression. 5. For the preparation of glycerol stocks: from the 2 mL starter culture, add 0.7 mL of culture to a sterile 1.5 mL microcentrifuge tube with 0.3 mL of 50% sterile glycerol solution (with water). Mix the solution by inversion, and quickly place into liquid N2. Store the samples in the -80 °C only. 6. When the cultures have grown until the desired OD600, it is no longer necessary to work under sterile conditions. 7. Chose the leaf area without the main vascular bundle system. 8. Wear glasses and gloves during syringe infiltrations. 9. To avoid specimen drift during image acquisition, leave the slide on the stage of the microscope for a few minutes so that the temperature and pressure under the coverslip stabilize. 10. Double-sided adhesive tape (width 22 mm) pasted on the microscope slide is an additional measure to prevent specimen drift during image acquisition. 11. Be careful not to damage the specimen by applying too strong pressure on the coverslip. 12. During microscopy, the best results are obtained from cells with an intermediate level of fluorescence signal from a more central, undamaged part of the leaf section. 13. When you use a microscope with the option for spectral emission separation, you can change the range of the band for the detectable emission signal using the prism-based tunable multiband spectral detector. In our case this is 500–550 nm for YFP and 560–640 nm for RFP. To reduce the detection of chlorophyll autofluorescence in the RFP channel, the upper maximum of detectable band should be no more than 640 nm.

88

Eduard Belausov et al.

14. This is an optional choice. 15. Make sure to select the right Max Distance, so that Imaris creates the tracks properly (we usually choose 2 μm as Max Distance, but this should be carefully examined). 16. We recommend using the batch option in Imaris if possible. This saves a lot of time and provides more accurate data, as manual editing is minimized. 17. Our movies show two channels, one for the microtubules and the other for myosin VIII ATM1. 18. To run the simulation, you need to have Python installed. 19. We usually chose one frame before and three frames after collision. References 1. Wasteneys GO (2002) Microtubule organization in the green kingdom: chaos or self-order? J Cell Sci 115(Pt 7):1345–1354 2. Elliott A, Shaw SL (2018) Update: plant cortical microtubule arrays. Plant Physiol 176(1): 94–105 3. Hardham AR, Gunning BE (1978) Structure of cortical microtubule arrays in plant cells. J Cell Biol 77(1):14–34 4. Ledbetter MC, Porter KR (1963) A “microtubule” in plant cell fine structure. J Cell Biol 19(1):239–250 5. Jarsch IK, Konrad SS, Stratil TF et al (2014) Plasma membranes are subcompartmentalized into a Plethora of coexisting and diverse microdomains in Arabidopsis and Nicotiana benthamiana. Plant Cell 26(4):1698–1711 6. Danek M, Angelini J, Malinska K et al (2020) Cell wall contributes to the stability of plasma membrane nanodomain organization of Arabidopsis thaliana FLOTILLIN2 and HYPERSENSITIVE INDUCED REACTION1 proteins. Plant J 101(3):619–636 7. Martinie`re A, Lavagi I, Nageswaran G et al (2012) Cell wall constrains lateral diffusion of plant plasma-membrane proteins. Proc Natl Acad Sci 109(31):12805–12810 8. Oda Y, Fukuda H (2012) Initiation of cell wall pattern by a Rho- and microtubule-driven symmetry breaking. Science 337(6100): 1333–1336

9. Bar-Sinai S, Belausov E, Dwivedi V et al (2022) Collisions of cortical microtubules with membrane associated Myosin VIII Tail. Cell 11(1): 145 10. Geiger B, Avnur Z, Rinnerthaler G et al (1984) Microfilament-organizing centers in areas of cell contact: cytoskeletal interactions during cell attachment and locomotion. J Cell Biol 99(1 Pt 2):83s–91s 11. Rinnerthaler G, Geiger B, Small JV (1988) Contact formation during fibroblast locomotion: involvement of membrane ruffles and microtubules. J Cell Biol 106(3):747–760 12. Seetharaman S, Etienne-Manneville S (2019) Microtubules at focal adhesions – a doubleedged sword. J Cell Sci 132, jcs232843 13. Efimov A, Schiefermeier N, Grigoriev I et al (2008) Paxillin-dependent stimulation of microtubule catastrophes at focal adhesion sites. J Cell Sci 121(Pt 2):196–204 14. Sparkes IA, Runions J, Kearns A et al (2006) Rapid, transient expression of fluorescent fusion proteins in tobacco plants and generation of stably transformed plants. Nat Protoc 1(4):2019–2025 15. Golomb L, Abu-Abied M, Belausov E et al (2008) Different subcellular localizations and functions of Arabidopsis myosin VIII. BMC Plant Biol 8:3

Chapter 7 Imaging the Plant Cytoskeleton by High-Pressure Freezing and Electron Tomography Janice Pennington and Marisa S. Otegui Abstract Electron tomography (ET) imaging of high-pressure frozen/freeze-substituted samples provides a unique opportunity to study structural details of organelles and cytoskeletal arrays in plant cells. In this chapter, we discuss approaches for sample preparation by cryofixation at high pressure, freeze substitution, and resin embedding. We also include pipelines for data collection for electron tomography at ambient temperature, tomogram calculation, and segmentation. Key words Microtubules, Actin filaments, High-pressure freezing, Freeze substitution, Electron tomography

1

Introduction The plant cytoskeleton comprises microtubules (MTs) and actin filaments that are part of different dynamic arrays during the cell cycle. MTs are polymers of 13 protofilaments made of α- and β-tubulin heterodimers. The end with exposed β-tubulin is the dynamic plus end, whereas the opposite end is called the minus end. A MT itself is a hollow cylinder of approximately 24 nm in diameter [1]. Plant MTs play important roles in determining cell shape, cell expansion, and mediating intracellular traffic. During their cell cycle, plant cells develop dynamic MT arrays that are critical for cell expansion and division. Thus, the cortical MT arrays control the deposition of cell wall material and the direction of cell expansion during interphase [2]: the preprophase band, which forms during G2, marks the site where the cell plate will fuse with the parental plasma membrane later during cytokinesis [3]; the mitotic spindle mediates chromosome segregation during mitosis and meiosis; and the phragmoplast coordinates the assembly of the cell plate during cytokinesis [4].

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_7, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

89

90

Janice Pennington and Marisa S. Otegui

Actin filaments consist of two-stranded polymers of globular actin, 5–7 nm in diameter [5]. The actin filaments are critical for establishing cell polarity and asymmetry, cellular differentiation and shape, cellular trafficking, and organelle movement, for cell wall assembly and remodeling [6]. The cytoskeleton is regulated by MT- and actin-binding proteins that influence the dynamics of the two polymers [1, 7, 8]. In addition, both types of cytoskeletal element associate with motor proteins that regulate the cytoskeletal arrays as well as intracellular transport [9]. As a result, the cytoskeleton is a highly integrated system of dynamic filaments that regulate many aspects of cellular function and, also, responses to environmental and developmental cues. Whereas live imaging using fluorescent reporters provides a direct window into the dynamics of MTs and actin filaments, the relatively low resolution of light-based microscopy modalities typically fails to resolve individual cytoskeletal filaments. In this case, the use of transmission electron microscopy (TEM) and electron tomography can greatly improve our understanding of this essential component of plant cells [3, 10–13]. The preservation of MT and actin filaments for TEM imaging can be challenging, as they are very sensitive to changes in the cellular physicochemical environment. High-pressure freezing is a physical fixing process that preserves biological samples close to their native state in vitreous ice [14]. The ice in the sample is then substituted at low temperature with a fixative dissolved in an organic solvent, followed by resin embedding and sectioning [15]. Sections can be then imaged by conventional TEM or electron tomography, by which threedimensional (3D) reconstruction of cellular areas can the reconstructed from a series of 2D TEM images collected at different tilt angles [16]. The axial resolution of an electron tomogram is typically one to two times of the lateral (x–y) resolution. For example, the resolution of a well-preserved high-pressure frozen/freezesubstituted and resin-embedded sample can be almost isotropic at 4–5 nm [17, 18]. We discuss in this chapter approaches for sample preparation by high-pressure freezing and freeze substitution for the best preservation and contrast of the plant cytoskeleton and a pipeline for image tomography collection and tomogram calculation using open-source software [16, 18].

2 2.1

Materials Plant Material

1. Seven-day-old Arabidopsis thaliana seedlings grown on agar plates or any other plant material under study.

Electron Tomography of the Plant Cytoskeleton

2.2 High-Pressure Freezing and Freeze Substitution

91

1. High-pressure freezer (e.g., Leica ICE HPF). 2. Specimen carriers (3 or 6 mm in diameter), e.g., type B specimen carriers (Ted Pella, Redding, CA). 3. Cryoprotectant: 0.1 M sucrose. 4. Dissecting tools (needles, razor blades). 5. Cryovials containing 1.5 mL of 2% OsO4 in anhydrous acetone (see Note 1). 6. Aluminum block with 13 mm holes (wide enough to fit cryovials). 7. Container (e.g., Styrofoam box) with dry ice. 8. Eponate 12 resin. 9. Flat embedding molds. 10. Plastic mounting cylinders.

2.3 Preparation of Grids and Sections for Electron Tomography

1. Lead citrate staining solution: Boil the milliQ water to eliminate CO2, and let the water cool down. Place 6 mL of boiled milliQ water into a 15 mL conical tube, and add 330 mg of lead nitrate. Sonicate for 5 min until lead is completely dissolved. Add 440 mg of sodium citrate, and mix gently to get a milky white solution, generating lead citrate. In a second 15 mL conical tube, add 5 mL of boiled milliQ water, and dissolve 200 mg of sodium hydroxide (see Note 2). Take 2 mL of the sodium hydroxide solution, and add to the lead citrate solution, mix until solution clears, and add milliQ water to a final volume of 12 mL. Cover the conical tube with aluminum foil, and store at 4 °C. Do not use the solution if a precipitate can be seen at the bottom of the tube. 2. Notch Beryllium/Copper slot grids (e.g., Synaptec, 2 × 1 mm notch grid). 3. 1.2% Pioloform in chloroform. 4. 70% ethanol. 5. Alconox. 6. Lens tissue paper. 7. Film casting device (Electron Microscopy Sciences). 8. Ultramicrotome. 9. Carbon sputter coater. 10. 2% uranyl acetate in 70% methanol. 11. 0.2 μm filter. 12. 10 nm colloidal gold particles.

92

Janice Pennington and Marisa S. Otegui

2.4 Image Acquisition and Software

1. Intermediate voltage (200 or 300 kV) electron microscope (e.g., Thermo Fisher Talos F200C), equipped with high-tilt rod for tomographic image acquisition. 2. Software: SerialEM [18–20] for image acquisition and IMOD [16] package for tomogram reconstruction (can be downloaded from http://bio3d.colorado.edu/docs/software. html).

3

Methods

3.1 High-Pressure Freezing

High-pressure freezing preserves biological samples in vitreous ice. Vitreous ice is amorphous and does not contain crystals. If the sample is accidentally warmed above -135 °C, ice crystals may grow, leading to cell rupturing and poor ultrastructural preservation. To prevent damage, high-pressure frozen samples must be handled with pre-cooled forceps and remain in liquid nitrogen or nitrogen vapors. Here, we explain how to high-pressure freeze root tips from Arabidopsis seedlings in a Leica ICE high-pressure freezer (Fig. 1a). Samples need to be small enough to fit into specimen carriers that are 0.3 mm deep and either 3 or 6 mm wide. 1. Place a half-cylinder cartridge with a middle plate into the loading station and a half-cylinder cartridge into the cover of the loading station (Fig. 1b).

B

A

half cylinder cartridge

middle plates 3 mm carrier

loading station

6 mm carrier

half cylinder cartridge

Fig. 1 High-pressure freezing using B-type metallic specimen carriers in a Leica ICE high-pressure freezer. (a) Overview of the sample loading station. (b) Mounting of specimen carriers, middle plates, and half-cylinder cartridges

Electron Tomography of the Plant Cytoskeleton

93

2. Quickly excise Set Target. Press Autofocus in the Image Tab. Check the results in the log file. If the target defocus is not reached, press Autofocus again. 5. Take a Record image of the area. Choose Process > Mid Max Mean. Read the counts accumulated in the log file. Increase or decrease intensity to achieve 9000 to 11,000 counts. 6. Choose Tilt Series > Add Extra Output > OK. A small window will appear on the screen which will show the tilt series as it is being acquired. 7. Choose Tilt Series > Setup, and choose tilt series angular range, usually 65° to -65°, number of tilts (base increment) 1 to 1.5°, magnification, pixel size (pixel size should be 1 nm or smaller).

Electron Tomography of the Plant Cytoskeleton

B

A notch

97

notch

90°

slot grid

Fig. 2 Placement of a notch slot grid in a Fischione tomographic holder to collect axis a (a) and axis b (b) with a 90° rotation from each other

Name the tilt series name_a.st (a-axis), and save it. When the sample is rotated, the second tomogram will be named name_b.st (b-axis). 8. Choose Close Column Valves. 9. Choose the information that will be stored with your tomogram (e.g., magnification). 10. Choose End Loop instead of Start. You will be prompted to choose a folder to save your tomogram, and then you will be prompted to name the tomogram. 11. The section will walk up to the highest tilt and stop. End Loop allows you to check the image, and make sure it is centered and in focus. If everything looks well, check Resume in the ddd control, and the tilt series will begin. 12. When the collection of the tilt series has finished, save the log file to your folder, go to Tilt series > Terminate, and reset the tilt to 0. 13. For each tomogram your folder will include the log file (name. log), tilt series (name.st), and mdoc (name.mdoc). Include any images that you want associated with the tomogram. 14. Rotate the sample in the tomography holder 90° (Fig. 2), and collect the b-axis (name1b.st) tomograms. 3.7 Data Collection: Acquiring Data for a Montaged, DoubleAxis Tomogram

1. Go to the Magnification to give a pixel size of 1 nm or less. 2. Choose Calibrate > Image and Stage Shift > Image Shift. 3. Go to File > New Montage. For example, an image of 45,000× magnification, Bin = 2, pixel size = 0.66 nm. Choose number of pieces (e.g., 3 × 2). 4. In the Montage Control Panel, choose Prescan to see if the montage covers the desired area. Set up the tilt series as explained in Subheading 3.6, step 4. Click on Start in the Montage Control Panel to collect the tilt series (a-axis). 5. Rotate sample, and collect a montage (b-axis) reversing the number of pieces from 3 × 2 to 2 × 3.

98

Janice Pennington and Marisa S. Otegui

3.8 Calculation and Segmentation of Tomograms

Once images along the two orthogonal axes are collected (either from single or montaged frames), the two tilt series files can be used to calculate a dual-axis electron tomogram using backprojection algorithms in software packages such as IMOD [16]. IMOD is a collection of different programs. The graphical interphase of IMOD for calculation of tomograms is called Etomo and provides a sequence of processes required for calculating each single-axis tomogram (along axis a and b) and for their subsequent combination into a dual-axis tomogram. 1. Open Etomo in a terminal, and select Build Tomogram in the Front Page menu (Fig. 3a). 2. When asked for Dataset name (Fig. 3b), enter the name of the “a stack” (name_a.st), and click Scan Header to automatically extract pixel size and image rotation from the file. Fiducial diameter (e.g., 10 nm if 10 nm gold particles were applied to the section) needs to be entered manually. Then, press Create Com Script to generate all the required files for the following steps (Fig. 3c). 3. Run Pre-processing to eliminate pixels with extreme intensity values. 4. Run Coarse Alignment to calculate translational alignment transforms using cross-correlation. 5. Run Fiducial Model Generation. This step can be done either manually or automatically to select gold particles as fiducials in the middle image of the tilt series (close to 0°). Between 30 and 50 gold particles distributed in both surfaces are recommended. A model or “seed” is generated that is in turn used to track the position of the initially selected gold fiducials in all views/angles of the tilt series. 6. Run Fine Alignment to generate transforms to align the set of tilt images using the fiducial model generated in step 5. This is an iterative process, and in each round, the positions of each fiducial is fitted to a mathematical model, generating a predicted position for each fiducial in each tilt view. As the predicted and real positions are compared, a residual error for each fiducial is generated. This error can then be used to fix the position of fiducials in the model. The final aligned image tilt series is generated with the final sets of calculated transforms. 7. Run Tomogram Positioning to generate a low-resolution tomogram to define the boundaries of the sample within the volume. 8. Run Final Alignment. 9. Run Tomogram Calculation to generate a tomogram using a backprojection algorithm. We usually use the simultaneous iterative reconstruction technique (SIRT) that combines backprojection and reprojection. If this option is chosen, enter the number of iterations (e.g., 10).

Electron Tomography of the Plant Cytoskeleton

99

Fig. 3 Calculation and segmentation of dual-axis tomograms using IMOD. (a) Front page of Etomo with options for different types of reconstruction and analyses. (b) Etomo setup option for tomography reconstructions. Option for dual- versus single-axis reconstruction and single-frame versus montaged tilt series is available here. (c) Created Com Scripts guide the user through the sequential process to calculate single- and dual-axis tomograms. (d) Tomographic slice and (d) tomographic segmentation of cortical microtubules (MT) and a potential actin filament (arrowheads) in a cortical array of a high-pressure frozen/freeze-substituted Arabidopsis root cell. Note the plus end with curved protofilaments (asterisk) in d. Scale bars = 50 nm

10. Once tomograms are calculated from each of the tilt series, run “Tomogram Combination” step aligns, and combine them in one single tomogram.

100

Janice Pennington and Marisa S. Otegui

3.9 Tomogram Segmentation

1. Once a dual-axis tomogram has been generated, the 3dmod program within IMOD allows for tomogram visualization, quantitative analysis, and segmentation. 2. Through segmentation, either manual or semi-automated, the boundaries of structures within tomograms are defined as contours in individual tomographic slices. Although automated and semi-automated segmentation approaches are preferred to minimize this time-consuming process and reduce user’s error or bias, at least for cellular tomography, segmentation still heavily relies on manual methods. 3. We usually perform manual segmentation in 3dmod from IMOD followed by surface rendering by imodmesh (see Note 7).

4

Notes 1. Prior to high-pressure freezing, prepare 2% OsO4 in acetone in a hood and wearing gloves. Store aliquots of 1.5 mL in cryovials under liquid nitrogen. 2. Adding this amount of sodium hydroxide should result in a pH of 12. If the pH is not 12, the stain may not be as effective as needed. 3. The sample is frozen when the needle valve opens and pressurized liquid nitrogen enters the pressure chamber. The rise of the pressure curve should be shorter than 10 ms to ensure that the sample is pressurized before cooling. At 150 ms after pressure initiation, the pressure curve should be between 2100 and 2350 bar and drop precipitously after 350 to 400 ms, as the pressurized liquid nitrogen is released through the nozzle after freezing. The temperature should drop at a rate of 15,000 to 25,000 K/s. The temperature curve should cross 0 °C between 1700 and 2000 bar. 4. Although an acetonic solution of OsO4 is an excellent freezesubstitution medium for plant tissues, addition of 0.1% uranyl acetate can improve contrast of microtubules and actin filaments. Enhanced contrast of microtubules can also be achieved bringing samples in 2% OsO4 from room temperature to 40 °C for 2 h, followed by 5% uranyl acetate in methanol at 4 °C for 2 h [3]. 5. Performing freeze substitution under agitation improves solution exchange [21]. However, agitation is difficult to achieve in commercial freeze-substitution devices (e.g., Leica AFS), resulting in extended freeze-substitution times of 3–4 days. In recent years, some agitation modules compatible with Leica AFS units have been developed [24].

Electron Tomography of the Plant Cytoskeleton

101

6. There are commercial (e.g., Grid Staining Matrix System from Ted Pella) to stain up to 25 grids simultaneously. 7. Besides IMOD, other software packages for tomogram segmentation include both open-source programs such as Chimera from University of California-San Francisco [25], Fiji/ ImageJ [26], and VolRover from University of Texas-Austin [27] and the commercial software packages Amira (Thermo Fisher Scientific) and MATLAB (MathWorks).

Acknowledgments This work was supported by grant NSF MCB 2114603 to MSO. References 1. Hashimoto T (2015) Microtubules in plants. The Arabidopsis Book 13:e0179–e0179 2. Paradez A, Wright A, Ehrhardt DW (2006) Microtubule cortical array organization and plant cell morphogenesis. Curr Opin Plant Biol 9:571–578 3. Takeuchi M et al (2016) Single microfilaments mediate the early steps of microtubule bundling during preprophase band formation in onion cotyledon epidermal cells. Mol Biol Cell 27:1809–1820 4. Smertenko A (2018) Phragmoplast expansion: the four-stroke engine that powers plant cytokinesis. Curr Opin Plant Biol 46:130–137 5. Li J, Blanchoin L, Staiger CJ (2015) Signaling to actin stochastic dynamics. Ann Rev Plant Biol 66:415–440 6. Li J, Arieti R, Staiger CJ (2014) Actin filament dynamics and their role in plant cell expansion. In: Plant cell wall patterning and cell shape, pp 127–162 7. Duckney P et al (2021) NETWORKED2subfamily proteins regulate the cortical actin cytoskeleton of growing pollen tubes and polarised pollen tube growth. New Phytol 231:152–164 8. Smertenko AP et al (2004) The Arabidopsis microtubule-associated protein AtMAP65-1: molecular analysis of its microtubule bundling activity. Plant Cell 16:2035–2047 9. Wang P, Hawkins TJ, Hussey PJ (2017) Connecting membranes to the actin cytoskeleton. Curr Opin Plant Biol 40:71–76 10. Austin JR II, Segui-Simarro JM, Staehelin LA (2005) Quantitative analysis of changes in spatial distribution and plus-end geometry of microtubules involved in plant-cell cytokinesis. J Cell Sci 118:3895–3903

11. Segui-Simarro JM et al (2004) Electron tomographic analysis of somatic cell plate formation in meristematic cells of Arabidopsis preserved by high-pressure freezing. Plant Cell 16:836– 856 12. Otegui MS, Staehelin LA (2000) Syncytialtype cell plates: a novel kind of cell plate involved in endosperm cellularization of Arabidopsis. Plant Cell 12:933–947 13. Otegui MS, Staehelin LA (2004) Electron tomographic analysis of post-meiotic cytokinesis during pollen development in Arabidopsis thaliana. Planta 218:501–515 14. Gilkey J, Staehelin LA (1986) Advances in ultrarapid freezing for the preservation of cellular structure. J Electron Microsc Tech 3:177– 210 15. McDonald KL (2014) Out with the old and in with the new: rapid specimen preparation procedures for electron microscopy of sectioned biological material. Protoplasma 251:429–448 16. Kremer JR, Mastronarde DN, McIntosh JR (1996) Computer visualization of threedimensional image data using IMOD. J Struct Biol 116:71–76 17. Weiner E et al (2022) Electron microscopy for imaging organelles in plants and algae. Plant Physiol 188:713–725 18. Mastronarde DN (2005) Automated electron microscope tomography using robust prediction of specimen movements. J Struct Biol 152:36–51 19. Mastronarde DN (1997) Dual-axis tomography: an approach with alignment methods that preserve resolution. J Struct Biol 120: 343–352 20. Mastronarde DN (2008) Correction for non-perpendicularity of beam and tilt axis in

102

Janice Pennington and Marisa S. Otegui

tomographic reconstructions with the IMOD package. J Microsc 230:212–217 21. McDonald KL (2014) Rapid embedding methods into epoxy and LR white resins for morphological and immunological analysis of cryofixed biological specimens. Microsc Microanal 20:152–163 22. McDonald KL, Webb RI (2011) Freeze substitution in 3 hours or less. J Microsc 243:227– 233 23. Reynolds ES (1963) The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J Cell Biol 17:208–212

24. Reipert S et al (2018) Agitation modules: flexible means to accelerate automated freeze substitution. J Histochem Cytochem 66:903–921 25. Yang Z et al (2012) UCSF Chimera, MODELLER, and IMP: an integrated modeling system. J Struct Biol 179:269–278 26. Schindelin J et al (2012) Fiji: an open-source platform for biological-image analysis. Nat Methods 9:676–682 27. Zhang Q, Bettadapura R, Bajaj C (2012) Macromolecular structure modeling from 3D EM using VolRover 2.0. Biopolymers 97:709–731

Chapter 8 Confocal Microscopic Assays of Mitotically Active Proteins in an Agrobacterial Infiltration-Based, Cell Division-Enabled Leaf System of Tobacco (Nicotiana benthamiana) Yuh-Ru Julie Lee, Calvin H. Huang, and Bo Liu Abstract The production of tissues and organs in plants is brought about by mitotic cell divisions, starting from the zygote. Successful mitosis and cytokinesis harness the functional input of proteins that are expressed in cell cycle-dependent manners to regulate cytoskeletal reorganization and intracellular motility. Fluorescence microscopic assays of mitotically active proteins have been dependent on time-consuming transformation experiments in a host plant or cultured cells. To facilitate the detection and observation of cell cycledependent localization and dynamics of plant proteins, we demonstrate, in this chapter, a transiently induced cell division system in Nicotiana benthamiana, named the cell division-enabled leaf system (CDELS). Plasmid constructs which express the D-type cyclin along with a fluorescent fusion protein (s) of interest are delivered to the leaves of N. benthamiana by agrobacterial infiltration. Ectopic expression of cyclin D induces leaf epidermal cells to re-enter mitosis and subsequently cytokinesis, allowing the dynamic localization of fluorescent fusion protein(s) to be observed throughout the course of mitotic cell division using live-cell fluorescence microscopy. This effective approach not only allows one to detect mitotic activities of novel proteins but also record their dynamics and relationship with others during mitosis and cytokinesis in a greatly shortened period of time. Key words Cell division, Cyclin D, Cytokinesis, Microtubules, Mitosis, Transient expression, Nicotiana benthamiana

1

Introduction Mitotic cell division results in the faithful segregation of replicated chromosomes during mitosis and the partition of two reformed nuclei, the cytoplasm, and organelles during cytokinesis into two daughter cells. These sophisticated events are brought about by the orchestrated actions of microtubules and microfilaments that go through dramatic reorganization into distinct arrays through the different stages of cell division. Proteins devoted to such cytoskeletal reorganization and associated activities are often expressed in a cell cycle-dependent manner [1, 2]. Rapid advances in live-cell

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_8, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

103

104

Yuh-Ru Julie Lee et al.

imaging technologies have made it possible to investigate the dynamic localization of proteins expressed during cell division. To date, live-cell imaging of proteins in dividing plant cells has been largely dependent on the establishment of stably transformed plants or cultured plant cells that express proteins of interest fused with fluorescent proteins (e.g., GFP). The generation of such stably transformed lines often takes months or longer and can be labor intensive and costly. Therefore, this inspired us to develop an inducible system for the assay of protein localization and function during cell division that is efficient, relatively fast, technically undemanding, and cost-effective. Agrobacterial infiltration in tobacco (Nicotiana benthamiana) is a mature, popular, and straightforward method of transient gene expression and has often been employed to detect the localization of fluorescent protein fusions in differentiated leaf epidermal cells in just a few days [3]. Taking these advantages of agrobacterial infiltration, we have developed a transient cell division-enabled leaf system (CDELS) that causes differentiated cells in the leaves of N. benthamiana to enter mitosis upon induction by the ectopic expression of D-type cyclin [4]. To initiate the G1-to-S phase transition in the cell cycle, the expression of D-type cyclin proteins (CYCDs) activates the cyclin-dependent kinase CDKA to phosphorylate the retinoblastoma (Rb) protein. As a consequence of the phosphorylation, the transcription factor E2F is liberated from the Rb inhibition and promotes cell cycle progression [5]. While mutant plants with reduced CYCD expression show a decrease in cell division in Arabidopsis thaliana, overexpression of CYCD proteins leads to the production of multicellular trichome cells that are otherwise unicellular in the wild-type plant [6]. Additionally, it has been reported that the overexpression of CYCD proteins induces differentiated leaf pavement cells to re-enter the cell division cycle [7]. Inspired by these earlier discoveries, experiments demonstrated that when N. benthamiana leaves were infiltrated with agrobacteria carrying the plasmid which constitutively expresses Arabidopsis CYCD proteins, the epidermal cells re-entered the cell division cycle, and therefore, the entire cell division cycle could be recorded by confocal microscopy [4]. The ectopic expression of the transcription factor E2Fb, which is a downstream effector of the CDKA/CYCD activation, also induced cell division in N. benthamiana [8, 9]. In Fig. 1a, using the CDELS and confocal microscopy, multiple epidermal cells at different stages of mitosis can be visualized. Mitotic microtubule arrays of the preprophase band, spindle, and phragmoplast are shown with their corresponding chromosomal configurations (Fig. 1b–d). Recently, cell cycle-dependent protein activities have been observed for components of the TPLATE complex which are involved in vesicle tethering by employing CDELS [10]. The CDELS technology offers exploratory screens

CDELS, Cell Division-Enabled Leaf System

105

Fig. 1 Epidermal cells undergoing mitosis and cytokinesis after CYCD induction by using the CDELS. (a) Multiple mitotic spindles (arrows) and phragmoplasts (arrowheads) are visible in an image created by Z-projection based on maximum intensity. (b–d) Individual epidermal cells with major mitotic microtubule arrays, including the preprophase band (b) detected at the G2 phase prior to nuclear envelope breakdown, the metaphase spindle with chromosomes aligned at the metaphase plate (c), and the phragmoplast between two reforming daughter nuclei (d). Microtubules are labeled by GFP-tubulin and pseudocolored in green; chromosomes are labeled by histone H1.2-TagRFP and pseudocolored in magenta. Cell outlines become visible due to GFP-tagged microtubules at the cell cortex (a). Scale bars, 20 μm in (a) and 5 μm in (b–d)

of novel proteins that may be involved in cell division. For example, this technology allowed the determination of the mitotic activity of 1 of the 61 microtubule-based motor proteins found in Arabidopsis thaliana [4]. Here, we have summarized the cell division-enabled leaf system (CDELS) which is based on the ectopic expression of CYCD in N. benthamiana leaves via agrobacterial infiltration that can be employed to investigate mitotic cell division in plants.

2 2.1

Materials Plant Materials

1. Nicotiana benthamiana: seedlings were grown in soil in a growth chamber with a regime of 16 h of light and 8 h of dark at 25  C. Plants of approximately 4 weeks old produce leaves ideal for agrobacterial infiltration (see Note 1).

106

Yuh-Ru Julie Lee et al.

2.2 Plasmids and Agrobacterium Strain

1. Plasmids for ectopic CYCD expression [4]: pGWB2-CYCD3;1 (Addgene plasmid #140750) and pGWB660-CYCD3;1 (Addgene plasmid #140751). 2. Plasmids for expressing proteins of interest in fusions with GFP or other fluorescent proteins (see Notes 2 and 3). 3. Plasmids for expressing markers in the cell cycle stages: histone H1.2(p):H1.2-TagRFP, chromosome marker [4] or TUB6(p): mCherry-TUB6, a microtubule marker [11]. Both can be co-infiltrated so that specific mitotic stages can be discerned by the characteristic chromosomal or microtubular configurations (see Note 4). 4. Plasmid for expressing the P19 protein of tomato bushy stunt virus that suppresses gene silencing (see Note 5). 5. Agrobacterium tumefaciens strain: GV3101.

2.3 Agrobacterial Infiltration

1. Infiltration buffer: 10 mM MES pH 5.6, 10 mM MgCl2, and 150 μM acetosyringone. The following stock solutions are used to prepare the BCDAT medium: 500 mM MES pH 5.6 with KOH, 500 mM MgCl2, 100 mM acetosyringone made in DMSO. Stock solutions of MES and MgCl2 are stored at 4  C; aliquots of acetosyringone are stored at 20  C. The infiltration buffer is freshly prepared prior to each use. 2. Syringe needle. 3. 1 mL blunt-tipped syringe.

2.4 Microscopy Materials

3

1. Confocal fluorescence microscope (see Note 6). 2. Glass slides and coverslips.

Methods

3.1 Growth and Preparation of an Agrobacterial Suspension for Infiltration

1. Transform competent cells of the Agrobacterium tumefaciens strain GV3101 separately with the plasmids expressing AtCYCD3;1 driven by the CaMV35S promoter, the GFP fusion of the protein to be tested, the P19 protein, and a mCherry/TagRFP tagged microtubule or chromatin marker. 2. Inoculate transformed GV3101 cells of each line separately in liquid LB medium supplemented with selection antibiotics according to the plasmids used, and culture the cells at 28  C by shaking at 180 rpm. 3. Once the cultured GV3101 cells reach the optical density (OD) of OD600 circa. 0.8–1.0, collect the cells by centrifugation, and then resuspend the cell pellets in the infiltration buffer. Typically, the agrobacterial cells are suspended at OD600 of 1.0. For co-infiltrations, agrobacterial cells

CDELS, Cell Division-Enabled Leaf System

107

containing the different plasmids are mixed in equal volumes. An agrobacterial suspension of 1 mL is used for the infiltration of a single leaf. A suspension containing the mixed agrobacteria is incubated for 3 h at room temperature prior to infiltration (see Note 7). 4. A step-by-step protocol for agrobacterial infiltration has been described clearly in the literature and widely used for transient gene expression [3, also see Chap. 6). 3.2 Live-Cell Imaging with Confocal Fluorescence Microscopy

1. The optimal time for observing cells in mitosis and cytokinesis in the leaf epidermis is at 36–48 h after agrobacterial infiltration. 2. To prepare slides for confocal microscopy, a leaf segment (approximately 1  1 cm2 in dimensions) is dissected from the infiltrated area and mounted in water between a microscope slide and a coverslip. The epidermal cells at the abaxial side of the leaf are chosen for microscopic observations because the adaxial side of the leaf often has bubbles trapped by trichomes that result in optical artifacts (see Note 8). 3. For dual-channel imaging of GFP and the red fluorescent protein of either mCherry or TagRFP under a confocal microscope, samples are excited by 488 and 561 nm lasers. Microscope acquisition parameters are set up according to the manufacturer’s instruction to minimize non-specific leaked fluorescent signals or autofluorescence from leaf cells (see Notes 9 and 10). 4. Still images are commonly acquired in the XY dimensions. Subcellular localization experiments often require the collection of images of different optical sections by applying the Z-stack function in order to reconstruct informative threedimensional data. To determine the dynamics of a protein during cell division, images are acquired with designated time intervals in order to generate time-lapse movies.

4

Notes 1. The healthy condition of the tobacco plants is a critical factor that contributes to the efficiencies of transient expression and mitotic induction in the CDELS experiments. Over-watering can compromise sufficient expansion of the leaves and lead to early development of an inflorescence. 2. To make a plasmid construct that expresses a fluorescent fusion of a targeted protein, the native promoter of the corresponding gene is the preferred choice over the constitutively expressed CaMV35S promoter. CaMV35S promoter-driven overexpression often can lead to mislocalization of proteins. For example,

108

Yuh-Ru Julie Lee et al.

Fig. 2 Different localization patterns of a CENH3 fusion protein when expressed under different promoters. (a) The cell infiltrated with the plasmid of AtCENH3( p):GFP-AtCENH3 shows that the GFP-AtCENH3 fusion protein localizes to the centromeres. (b) The cell infiltrated with the plasmid of CaMV35S:GFP-AtCENH3 shows that the GFP-AtCENH3 fusion decorates the entire chromosomes. The GFP-AtCENH3 signals are pseudocolored in green, and mitotic spindles are labeled by mCherry-TUB6 and pseudocolored in magenta. Scale bar, 5 μm

the Arabidopsis-centromeric histone variant CENH3 is detected at the centromere of the chromatid [12], which is faithfully reproduced in tobacco cells when the AtCENH3(p): GFP-CENH3 construct is used (Fig. 2a). In contrast, when the fusion protein is expressed upon infiltration using the CaMV35S:GFP-CENH3 construct, the GFP signal decorates the entire chromosomes (Fig. 2b). 3. The fluorescent protein tag may be fused to either the amino or the carboxyl terminus of the target protein. When choosing which end, consideration must be toward minimizing possible interference of any functional domains in the protein. This is because GFP or other fluorescent proteins may interfere with polypeptide folding resulting in the masking of any functional domains that are necessary for the correct localization and/or functionality of the protein. 4. Microtubules and chromosomes generate distinct configurations as a cell progresses through the cell division cycle. It is helpful, therefore, to include a plasmid construct which expresses a fluorescent fusion protein of histone or tubulin in the CDELS experiments. This will allow identification of the different phases of the cell division cycle. For example, the localization of the KNOLLE/SYP111 protein, a SNARE protein which is specifically required for vesicle fusion in cell plate formation [13], can be detected with the help of a cell stage marker of histone H1.2-TagRFP or mCherry-TUB6 as shown in Fig. 3.

CDELS, Cell Division-Enabled Leaf System

109

Fig. 3 Localization of Arabidopsis KNOLLE/SYP111 protein in mitotic cells with the cell division stage marker histone H1.2-TagRFP for chromosomes or mCherry-TUB6 for microtubules. The GFP-KNOLLE/SYP111 signal (pseudocolored in green) appears between two sets of segregated chromosomes (pseudocolored in magenta), at anaphase (a) and telophase (b). (c) The GFP signals remain at the developed cell plate when the nuclear envelope is formed around the two daughter nuclei. (d) The GFP-KNOLLE/SYP111 signal (pseudocolored in green) appears at the midzone of the phragmoplast microtubules (pseudocolored in magenta) in an early stage of cytokinesis. (e) At a later stage of cytokinesis, GFP-KNOLLE/SYP111 remains at the cell plate and across the midline following the disassembly of phragmoplast microtubules toward the center of the phragmoplast. Scale bar, 5 μm

5. To maximize ectopic protein expression, having the tomato bushy stunt virus P19 protein co-expressed is strongly recommended in the infiltration experiments. The P19 protein functions as a potent suppressor of gene silencing [14] so that the expression of the proteins of interest is not silenced. 6. A confocal microscope is used for live-cell imaging so that outof-focus signals can be excluded. When a specimen contains multiple layers of cells of heterogeneous tissues, it becomes challenging to acquire satisfactory images using widefield fluorescence microscopes because the out-of-focus fluorescence signal often masks the specific signal at a defined focal plane. 7. Note that optimal density of agrobacterial cells for protein expression in infiltration experiments depends on individual proteins in terms of the peak of expression and protein abundance.

110

Yuh-Ru Julie Lee et al.

8. If an inverted microscope is used for live-cell imaging, a wide coverslip, e.g., 24  50 mm2, may be used to hold the leaf specimen. To do so, a drop of water is applied to the coverslip for placing the leaf segment with its abaxial side facing the coverslip. The leaf segment is overlaid with a piece of a 2% phytagel block containing ½ Murashige and Skoog salts and MES (pH 5.7 with KOH). The moist condition created by the phytagel block will prevent the leaf specimen from drying out, and the weight of the phytagel block will also help flatten the tissue and stop the specimen drifting during prolonged imaging. 9. When two or more fluorescent fusion proteins in a specimen are imaged, a potential problem is the bleed-through fluorescence signals caused by overlapping emission spectra. Although it is less of a problem when the fluorescent fusion proteins render strong signals, such a problem may complicate the interpretation of experimental results when the signals are imbalanced and when one is significantly weaker than the other. The bleedthrough problem can be minimized by choosing fluorescent proteins with well-separated excitation and emission spectra. For example, the combination of GFP and mCherry is wildly used in dual localization experiments in plant cells. Furthermore, the problem may be minimized by applying the protocol of sequential scanning associated with laser scanning confocal microscopy which prevents multiple fluorescent proteins from being excited simultaneously. 10. A major challenge in imaging green plant tissues by fluorescence microscopy is the interference of autofluorescence signals emitted by chloroplasts. A simple solution is to avoid collecting emission signals above 630 nm. This would eliminate the major autofluorescence signals generated by chlorophylls, the primary chloroplast pigments.

Acknowledgments We are grateful to Drs. T. Nakagawa, E. Park, S. Dinesh-Kumar, M. Marimuthu, L. Comai, M. Sato, T. Kato, and T. Hashimoto for sharing their plasmids and seeds described here and Dr. J. Xu for his pioneering effort in developing the CDELS. This work was supported by the NSF of the USA (MCB-1616076 and MCB-1920358) and USDA/NIFA under an AES hatch project (CA-D-PLB-2260-H) to BL.

CDELS, Cell Division-Enabled Leaf System

111

References 1. Lee YRJ, Liu B (2016) Cytokinesis. In: Assmann S, Liu B (eds) Plant cell biology. Springer, New York. https://doi.org/10. 1007/978-1-4614-7881-2_9-1 2. Liu B, Lee YRJ (2022) Spindle assembly and mitosis in plants. Annu Rev Plant Biol 73: 227. https://doi.org/10.1146/annurevarplant-070721-084258 3. Sparkes IA, Runions J, Kearn A, Hawes C (2006) Rapid, transient expression of fluorescent fusion proteins in tobacco plants and generation of stably transformed plants. Nat Protoc 1:2019–2025. https://doi.org/10. 1038/nprot.2006.286 4. Xu J, Lee YRJ, Liu B (2019) Establishment of a mitotic model system by transient expression of the D-type cyclin in differentiated leaf cells of tobacco (Nicotiana benthamiana). New Phytol 226:1213–1220. https://doi.org/10. 1111/nph.16309 5. Polyn S, Willems A, De Veylder L (2015) Cell cycle entry, maintenance, and exit during plant development. Curr Opin Plant Biol 23:1–7. https://doi.org/10.1016/j.pbi.2014.09.012 6. Schnittger A, Scho¨binger U, Bouyer D, Weinl C, Stierhof Y-D, Hu¨lskamp M (2002) Ectopic D-type cyclin expression induces not only DNA replication but also cell division in Arabidopsis trichomes. Proc Natl Acad Sci U S A 99:6410–6415. https://doi.org/10.1073/ pnas.092657299 7. Weimer AK, Matos JL, Sharma N, Patell F, Murray JA, Dewitte W, Bergmann DC (2018) Lineage- and stage- specific expressed CYCD7;1 coordinates the single symmetric division that creates stomatal guard cells. Development 145:dev160671. https://doi. org/10.1242/dev.160671 8. Jimenez-Go´ngora T, Tan H, Lozano-Durán R (2019) Transient overexpression of E2Fb triggers cell divisions in pavement cells of

Nicotiana benthamiana leaves. Plant Cell Rep 38:1465–1471. https://doi.org/10.1007/ s00299-019-02457-3 9. Jimenez-Go´ngora T, Tan H, Lozano-Durán R (2022) Transient overexpression of E2Fb to induce cell divisions in Nicotiana benthamiana pavement cells. Methods Mol Biol 2382:115– 127. https://doi.org/10.1007/978-1-07161744-1_7 10. Yperman K, Papageorgiou AC, Merceron R, De Munck S, Bloch Y, Eeckhout D, Jiang Q, Tack P, Grigoryan R, Evangelidis T, Van Leene J, Vincze L, Vandenabeele P, Vanhaecke F, Potocky´ M, De Jaeger G, Savvides SN, Tripsianes K, Pleskot R, Van Damme D (2021) Distinct EH domains of the endocytic TPLATE complex confer lipid and protein binding. Nat Commun 12:3050. https://doi. org/10.1038/s41467-021-23314-6 11. Nakamura M, Ehrhardt DW, Hashimoto T (2010) Microtubule and katanin-dependent dynamics of microtubule nucleation complexes in the acentrosomal Arabidopsis cortical array. Nat Cell Biol 12:1064–1070. https://doi.org/ 10.1038/ncb2110 12. Ravi M, Chan SW (2010) Haploid plants produced by centromere-mediated genome elimination. Nature 464:615–618. https://doi. org/10.1038/nature08842 13. Enami K, Ichikawa M, Uemura T, Kutsuna N, Hasezawa S, Nakagawa T, Nakano A, Sato MH (2009) Differential expression control and polarized distribution of plasma membraneresident SYP1 SNAREs in Arabidopsis thaliana. Plant Cell Physiol 50:280–289. https:// doi.org/10.1093/pcp/pcn197 14. Scholthof HB (2006) The Tombusvirusencoded P19: from irrelevance to elegance. Nat Rev Microbiol 4:405–411. https://doi. org/10.1038/nrmicro1395

Chapter 9 Assessment of Spindle Shape Control by Spindle Poleward Flux Measurements and FRAP Bulk Analysis Sabine Mu¨ller Abstract In plants, the segregation of genetic material is achieved by an acentrosomal, mitotic spindle. This macromolecular machinery consists of different microtubule subpopulations and interacting proteins. The majority of what we know about the assembly and shape control of the mitotic spindle arose from vertebrate model systems. The dynamic properties of the individual tubulin polymers are crucial for the accurate assembly of the spindle array and are modulated by microtubule-associated motor and non-motor proteins. The mitotic spindle relies on a phenomenon called poleward microtubule flux that is critical to establish spindle shape, chromosome alignment, and segregation. This flux is under control of the non-motor microtubule-associated proteins and force-generating motors. Despite the large number of (plant-specific) kinesin motor proteins expressed during mitosis, their mitotic roles remain largely elusive. Moreover, reports on mitotic spindle formation and shape control in higher plants are scarce. In this chapter, an overview of the basic principles and methods concerning live imaging of prometa- and metaphase spindles and the analysis of spindle microtubule flux using fluorescence recovery after photobleaching is provided. Key words Arabidopsis, CLSM, Spindle, Microtubule dynamics, Cell division, Poleward flux, Kymograph, Fluorescence recovery after photobleaching

1

Introduction The mitotic spindle is a conserved assembly of microtubules formed during the cell division of eukaryotic cells. This array is responsible for the faithful partitioning of genetic material and thus essential for life, while much of our knowledge regarding its dynamics originates from work on vertebrate and in vitro models [1–3]. The mitotic spindle consists of different subpopulations of microtubules extending from two opposing, in plant cells acentrosomal, spindle poles toward the cell equator/spindle midzone (Figs. 1 and 2a). Subpopulations of microtubules attach to sister chromatids at kinetochores, which are multi-protein complexes located at the outer layer of the centromeres. These microtubules

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_9, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

113

114

Sabine Mu¨ller

Non-Kinetochore Microtubules

Kinetochore

Kinetochore Microtubules

Interpolar Microtubules

(+) (+) (+)

(+)

(-) (-)

(+)

(-) (-)

(-) (-) Pole (-) (-)

Flux

(+)

Sister Chromatids

Fig. 1 Schematic representation of the metaphase spindle. Kinetochore microtubules (thick, cyan lines) extend from the poles with their minus ends (
) and attach to kinetochores (grey spheres) of sister chromatids (yellow) with their plus ends (+). Non-kinetochore microtubules (grey lines) extend from the poles. Interpolar microtubules overlap with their plus ends in the spindle midzone/ equator

often form morphologically distinct thick bundles or K fibers and facilitate chromosome alignment and segregation. In addition, non-kinetochore microtubules extend from the poles. Those that overlap at the spindle midzone, termed interpolar microtubules, contribute to pole movement during anaphase [4, 5]. The inherent properties of individual microtubules that stochastically switch between phases of polymerization (rescue) and phases of depolymerization (shrinking/catastrophe) at their ends are crucial features of the mitotic spindle (Fig. 2b). This behavior termed dynamic instability contributes to essential functions such as spindle formation, positioning and shape, kinetochore capture, and chromosome congression and segregation [5, 6]. The less dynamic minus ends face the poles, while the highly dynamic plus ends face the spindle midzone (Fig. 1). A phenomenon characterizing the dynamicity of spindle microtubules is poleward flux: the disassembly of the microtubule minus ends at the poles and polymerization at their plus ends in the spindle midzone, facilitating the minus end directed flow of tubulin subunits through the spindle (Figs. 1 and 2c). The functions of poleward flux include spindle length control

Analysis of Spindle Size Maintenance

a

115

c Tubulin Heterodimer

Pole (-)

Protofilament

(-)

Kinetochore (+)

t0

(+) 25 nm

b

t1

Polymerization

GTP GTP GDP

t2

GDP

Depolymerization

Flux

Fig. 2 Schematic representation of microtubule structure and dynamics. (a) Microtubule lattice, protofilament of alpha- and beta-tubulin heterodimers, is indicated. Polarity of microtubule with exposed beta-tubulin at the plus end (+) and exposed alpha-tubulin at the minus end (
). (b) Dynamic instability is exemplified on microtubule plus end but also occurs on the minus end. Stochastic switching between phases of polymerization by addition of GTP-bound tubulin heterodimers and phases of depolymerization by loss of oligomers. (c) Illustration of a single kinetochore microtubule, exhibiting polymerization at the kinetochore-associated plus end and depolymerization at the minus end facing the pole. The dark tubulin dimers represent the bleached region within the spindle microtubule polymers. The region can be tracked over time to determine the rate of spindle microtubule flux. (Modified from Ref. [4])

of the steady-state metaphase spindle and chromosome segregation in anaphase, and its mis-regulation could increase the probability of erroneous chromosome distribution [5, 7]. These functions require rapid, flexible, and adaptive changes in the steady-state levels of microtubule polymerization/depolymerization during specific time points of the cell cycle, which in turn may translate into the movement of chromosomes or spindle poles. Generally, microtubule-associated proteins and forcegenerating kinesin motor proteins facilitate such changes and orchestrate the assembly and shape/size of the spindle [3, 5, 8]. Although genes encoding spindle assembly factors are well conserved in plants, with the notable lack of centrosomal components and cytoplasmic dynein, few proteins have been characterized for their functions in the mitotic spindle. [9–13]. Likewise, analysis of poleward flux in prometa- and metaphase spindles, when comparing wild-type versus mutants or effects of drug treatment, for example, is not widely exercised in the plant field [14, 15]. Nevertheless, plant genomes host a large number of mitotic and plant-specific kinesins [11, 16], whose mechanistic contributions to spindle formation and size maintenance are unclear. Often, the lack of strong, morphological mutant

116

Sabine Mu¨ller

phenotypes due to functional redundancies hinders attempts of characterization and improvement of our understanding. Nevertheless, it may be rewarding to invest in the analysis of spindle formation and shape control in mutants of spindle-localized proteins [14]. Unravelling the mitotic contribution of the latter may improve our perception of spindle formation and spindle size control in dividing plant cells. Here, we provide a step-by-step protocol for the quantification of prometa- and metaphase spindle flux (adapted from [8, 15]) and bulk fluorescence recovery after photobleaching analysis, using confocal time-lapse imaging and kymograph (space-time-plot) analysis. These methods aim to contribute to the insightful analysis of plant mitosis and facilitate the understanding, analysis, and quantification of cytoskeletal dynamics that drive poleward flux.

2

Materials

2.1

Plant Material

1. Arabidopsis (Col-0) transformed with p35S::GFP-TUA6:green fluorescent protein (GFP) fused to alpha-tubulin 6 (TUA6) from Arabidopsis, expressed under the control of the 35S promoter [14]. Other microtubule marker lines are also available [17–20].

2.2

Plant Growth

1. Eppendorf tubes, 1.5 mL. 2. Square Petri dishes, 125 mm  125 mm. 3. Laboratory glass bottles. 4. 6% sodium hypochlorite in distilled water. 5. Distilled water. 6. Murashige and Skoog (MS, Duchefa) growth medium [21]. 7. Agar. 8. Cold room or refrigerator. 9. Plant growth chamber adjustable to an appropriate temperature, illumination, and photoperiod. 10. Set of pipettes and pipette tips.

2.3 Sample Preparation

1. Imaging chambers (e.g., LAB-TEK II Chambered Coverglass). 2. Small, rectangular (e.g., 18  18 mm) coverslips. 3. Forceps. 4. Pasteur pipettes. 5. Falcon tubes. 6. Aliquots of 0.5% agar in MS medium. 7. Liquid growth medium. 8. Kim wipes.

Analysis of Spindle Size Maintenance

2.4 Microscope and Software

117

1. Leica TCS SP8 CLSM (Leica Microsystems, Manheim, Germany) equipped with resonant and point scanner and photomultiplier tubes (PMT) for detection. 2. Leica Application Suite X software. 3. Image processing software: ImageJ (https://imagej.nih.gov/ ij/) or Fiji (https://fiji.sc/). 4. Excel or comparable data analysis software.

3

Methods

3.1 Plant Material and Growth Conditions

1. Under sterile conditions (clean bench), sterilize Arabidopsis GFP-TUA6 seeds (or your favorite microtubule reporter line) for 5 min using 6% sodium hypochlorite. 2. Subsequently, wash the seeds three times with sterilized distilled water. Then, plate the seeds on solid MS medium supplemented with 1% agar in square Petri dishes. 3. Stratify seeds for 24–48 h at 4  C, and place plates vertically in a plant growth chamber at 22  C and 16 h light/8 h dark photoperiod. Grow seedlings for 3–5 days.

3.2 Sample Preparation

1. An hour before the imaging session starts, prepare a small amount (i.e., 10 mL) of MS medium supplemented with 0.5% agar. 2. Boil the media and wait until cool, i.e., “hand warm” (see Note 1), and then pipet 2 mL into the imaging chamber. The medium should evenly cover the coverslip of the chamber and solidify. 3. For sample preparation, use the 18  18 mm coverslip to vertically cut through the agar in the middle of the imaging chamber. Be careful not to break or etch the glass bottom of the imaging chamber. Then use the 18  18 mm coverslip to slide underneath the medium and lift the latter at an angle. Add a drop of the liquid medium onto the coverslip of the chamber and put three to five seedlings in place (see Note 2). Carefully lower the agar layer, and concomitantly remove the 18  18 mm coverslip. Avoid trapping air bubbles (see Note 3). 4. Drain any access liquid using Kim wipes. 5. Wait a few minutes for the samples to settle and stabilize. The specimen is ready for time-lapse imaging.

3.3

FRAP Imaging

1. Turn on the imaging system, and start the microscope software.

118

Sabine Mu¨ller

2. Turn on the appropriate lasers for imaging. Here we used the 488 nm laser line of an Argon laser for excitation of GFP. Adjust the Argon laser power to 20% or appropriate to your imaging system. Set the laser power of the 488 nm laser line to 5–15% for imaging. These settings should be sufficient to excite the GFP fluorescence of the GFP-TUA6 fusion protein expressed in Arabidopsis. For image acquisition, use a line average of 16–32 frames, when using the resonant scanner. Higher line averaging would improve image quality only marginally, but the image acquisition would take longer, thereby increasing the photobleaching rate due to long exposure of the sample to the laser (see Note 4). 3. Turn on the 405 nm laser, and set to 100% for photobleaching (see Note 5). 4. For fluorescence detection purposes, use standard PMTs. The detection wavelength range for GFP fluorescence may be adjusted between 505 and 550 nm. Adjust gain and offset of the PMT as appropriate. 5. Place the imaging chamber with the specimen on the microscope stage. Select the most suitable objective lens for imaging. Here, we employed a 63 water immersion objective lens with a numerical aperture (NA) ¼ 1.20. 6. Using the bright-field illumination, move the stage to find your specimen, i.e., the root tip, into the center of the field of view. 7. Find a prometa- or metaphase cell, and focus on a median plane in live mode (see Notes 6 and 7). Rotate the scan field to align the spindle axis with the x-axis of the scan field. 8. Zoom in and center the cell of interest. The zoom factor may be set to 4 or 5. Re-adjust laser power and PMT gain, if needed. 9. Switch to FRAP wizard under TCS SP8 (see Note 8). The wizard guides you through the steps to set up the FRAP. 10. Set the scan dimensions to xyt for time-lapse imaging, and set the time interval to 0.5–2 s (see Note 9). 11. In Leica Application Suite X, activate the FRAP booster. 12. Define parameters for photobleaching. Select the laser lines 405 nm at 100% for bleach steps (see Note 10). 13. Select bleaching method “Zoom In.” Set background to zero. Remove bleach steps from the series after the scan. Use the same laser settings and regions of interest (ROI) for all image series. 14. Set the number of pre-bleach images to 3. Set the number of bleaching steps to 10. 15. Set the number of post-bleaching images to 100.

Analysis of Spindle Size Maintenance

a

c

119

b post-bleach

pre-bleach -1.5 s

0s

-1.5 s

e

t0

18.5 s

t 10 t (n-1) tn

d 1s

v=Δd/Δt t d

3 μm

Fig. 3 Analysis of spindle tubulin flux. (a) Images of GFP-TUA6 in metaphase. Frames before FRAP (pre-bleach, –1.5 s). Arrowheads indicate the spindle pole axis. (b) Frames after FRAP at 0 s and 18.5 s (here, equals tau, fluorescence half time). Orange-dashed boxes outline the FRAP region. (c) Example of individual images of a time series (indicated by red arrow) and region of interest (ROI, dashed line). (d) Space-time plot (kymograph) representing the distance, delineated by the ROI (d, x-axis), plotted against time (t, y-axis). The slope of the red line along the contrast edge is proportional to the velocity of tubulin flux. (e) Montage of the region indicated by the blue-dashed box in (a) of FRAP time series. Montage comprises frame 1 to frame 43 with time increment 2 at 0.5 s time interval. The red line indicates fluorescence recovery over time. Arrows in (a), (b), and (c) delineate the metaphase plate. Scale bars, 3 μm

16. Define a rectangular ROI parallel to the chromosomes in the spindle midzone (Fig. 3b), and take a snapshot. Save user settings to perform FRAP using identical conditions. 17. Start FRAP experiments. 3.4 Kymograph Analysis of Spindle Flux

1. Among the simplest ways to analyze the dynamic properties of cytoskeletal elements is to use a kymograph. Use ImageJ [22] or its Fiji distribution [23]. 2. For kymograph analysis, you will need the “Multi Kymograph” plugin, which can be downloaded from http://www.embl.de/ eamnet/html/body_kymograph.html. The website also provides very useful step-by-step instructions to kymograph analysis (see Note 11). 3. Open the FRAP series in ImageJ.

120

Sabine Mu¨ller

4. If using LAS AF from Leica, pre- and post-bleaching images are saved in separate image series (Fig. 3a–c). For kymograph analysis, you may use the post-bleaching series only. Optional: Concatenate (Image > Stacks > Tools > Concatenate) pre- and post-bleach images of the respective FRAP experiment. 5. For minor drift in x, y, use “StackReg plugin” for image registration. This allows adjusting for sample drift within a reasonable range. You can find this plugin under the “Registration” option in “Plugins.” Adjust brightness and contrast. Save the registered and adjusted stack as *.tif file. 6. Now, use the “straight line” tool to select a line (Line width “10”) across the spindle parallel to the spindle axis and perpendicular to the spindle equator with chromosomes in negative stain (Fig. 3b). Inspect every image in each stack to make sure that the line selection covers the entire spindle including the bleached region. Save the selection in the ROI manager. 7. Click “Multi Kymograph,” and set the “Line width” to “1.” The kymograph will appear in a pop-up window (Fig. 3d). This kymograph image represents the trajectory of the signal intensities across the selected line (in step 6) over the recorded period. In the case of the spindle and bleached region, you expect to see a characteristic signal contrast edge (Fig. 3d, e). 8. In the kymograph, the slope of the contrast edge at the border of the fluorescence signal and bleach region is proportional to the speed of spindle poleward flux (Fig. 3d). To determine the rate of spindle flux, use the “Straight line” tool, and select the signal contrast edge in the kymograph image (Fig. 3d). 9. Now click on “Read velocities from tsp” to automatically extract the velocity. In a pop-up window, the read-out specifying dx (distance, d) and dy (time, t) values in pixels and the velocity (dx/dy, proportional to Δd/Δt) are shown. Save the selection in the ROI manager. 10. Export the tabular information from step 8 into an excel spreadsheet. Since the measurements are pixel values, you need to calculate the velocities in nm/sec. Multiply “dy” by the time interval between frames (i.e., 0.5 s) and “dx” by the pixel size. v¼

ðdx  pixel sizeÞ ðdy  time intervalÞ

You will find time interval (i.e., 0.5 s) and pixel size (i.e., 60 nm) in metadata. If you use Leica Application Suite X, you can find this information under “Properties.” “Pixel size” refers to the size of one pixel in the acquired image in physical units (nanometers or micrometers). “Time interval” refers to

Analysis of Spindle Size Maintenance

121

the time between two frames/stacks that are acquired. You may also convert to velocities to μm/min (see Note 12). 11. Perform the speed calculation for various time series, and calculate the mean and standard deviation for spindle poleward flux. 12. For the illustration of poleward flux, arrange a montage of 10-pixel-wide slices along the pole-to-pole axis of every second or third image of the time series, using the montage tool (Fig. 3e). 3.5 Fluorescence Recovery After Photobleaching Analysis

In addition to flux analysis, the FRAP data may provide further insight into spindle microtubule dynamics. For instance, the average fluorescence half time (tau) of bleached regions may be determined, when comparing conditions or genotypes. For rapid analysis of multiple FRAP series, we use a set of plugins devised for bulk analysis (Stowers Plugins collection, https://research. stowers.org/imagejplugins/zipped_plugins/). Therefore, install the Jay_Plugins.jar in ImageJ. 1. Open the concatenated (pre-and post-bleaching), stack registered, and adjusted FRAP series in ImageJ (see Subheading 3.4, step 4). 2. Apply subtract background (rolling ball 100 pixels) to each series. 3. Select the bleached region (ROI), according to the ROI in the FRAP experiment, using the rectangular selection tool. Save the selection in the ROI manager. 4. In the Plugins menu, select Plugins Stowers > Jay Unruh > Image Tools, and apply the “Create spectrum jru v1” plugin to automatically measure and plot the average signal intensity in ROIs for individual time points in each FRAP experiment. Plots may be saved as .pw2 files by clicking the save button at the bottom of the plot windows or going to Stowers > Jay Unruh > Trajectory Tools > export plot jru v1 (see Note 13). 5. Combine plots of individual FRAP experiments (Stowers >>combine all trajectories jru v1, select combine all series), and normalize (normalize trajectories jru v1, option min_max). 6. In the Plot window of the combined plot (Fig. 4a), click on List and copy the combined values to excel (x, frames; y, normalized signal intensity). Reorganize the Excel table into a first column showing x1 values (# of frames is the same for each experiment; therefore you may delete all xn, but x1), followed by the y1–yn columns (normalized signal intensities). 7. Back in ImageJ, compute the microtubule turnover (half time of signal recovery, tau, s) for each experiment using Trajectory Tools > batch FRAP fit jru v1. In the Options window, set the

Sabine Mu¨ller

a

b average curve fitted curve

c

d

0.6 normalized intensity

122

average curve fitted curve

0.4 0.2

Tau 18.7 s ± 5.5 0.0 0

5

11 16 21 27 32 37 42 48 t [sec]

Fig. 4 FRAP analysis. (a) Individual, normalized FRAP curves plotted into one graph. (b) The average (black line) of the individual curves in (a) and fitted (blue) curve are plotted in one graph. (c) Table showing computed fluorescence half time (tau) for individual curves. (d) Average curve (black line) with standard deviation and fitted curve (orange, corresponding to the blue line in (b))

number of pre-bleach images (if applicable) and the number of frames to analyze (i.e., 3 and 103 in the presented experiments). A plot showing the average curve and the fitted curve will open (Fig. 4b).

Analysis of Spindle Size Maintenance

123

8. Use List in the Plot window to open the table with values for the average and fitted curve (Fig. 4c). Again, copy the values for the fitted curve, and paste them into Excel. Y1 are the intensities of the average curve, while Y2 shows the intensities of the fitted curve. 9. Create an Excel graph showing the average intensity values and their standard deviation (calculated from the individual normalized intensities described in step 5) together with the fitted curve (Fig. 4d). Important: convert frames (x values) into respective time intervals in Excel! When comparing conditions, determine significant differences with an appropriate statistic tool (e.g., one-way ANOVA), and generate a box plot.

4

Notes 1. Make sure the medium is still liquid, but cool enough, to prevent the thin coverslip of the chamber from cracking. 2. In the experiments presented in this chapter, we used 3–4-dayold seedlings. Seedlings older than 5 days were considered unsuitable for imaging due to the increased plant size that makes continuous, stable observation difficult and amplifies chances of sample drift. Any microtubule marker line may be used for the analysis, although fusion with a bright and bleaching-resistant fluorophore is most desired [14, 17– 20]. However, different reporters may interfere with normal microtubule function. For example, overexpression of microtubule reporters may produce artifacts, such as helical organ growth. One should be aware of potential artifacts, if and when the reporters start to impact the normal behavior of microtubules in the plant life cycle. Furthermore, variations in growth and environmental conditions differ between laboratories which may explain some of the deviations in imaging and quantitative measurements that may be found in the literature. 3. For a more detailed description of sample preparation for live cell imaging, see the video protocol from Peterson and Torii [24]. 4. Exposure time may be reduced by choosing a lower-resolution format, i.e., 512  256 pixels or less. Make sure pixel size remains below the optical resolution of your imaging system. 5. If the exposure with 405 nm is not sufficient for effective bleaching within ten frames, set additional lasers, i.e., 561 nm and/or laser lines of the Argon laser to 100%. 6. To quickly find suitable prometa- and metaphase cells, use low digital zoom or an objective with lower magnification first.

124

Sabine Mu¨ller

7. Imaging of dividing cells located deep in the root (e.g., vascular tissue) should be avoided unless specifically desired. In general, cells from outer tissue layers are more suitable for live imaging in terms of fluorescence detection. 8. A detailed manual for FRAP using Leica SP8 can be found on the microscope manufacturers’ website. 9. Spindle microtubules are usually highly dynamic. Therefore, a short time interval is most useful. However, avoid photobleaching during imaging. 10. Optional: additional laser lines, i.e., 458 nm, 488 nm, 514 nm, and 561 nm, may be set to 100% and used if the region of interest is not efficiently bleached within ten frames). 11. Particularly important is the tsp.txt file in order to read velocities. After starting ImageJ, click Plugins > Macro > Install, and select tsp050706.txt in the kymograph plugin folder. Now you should find the function “Read velocities from tsp” in the macro drop-down menu. 12. An alternative method to calculate velocities from kymographs is given in Livanos et al. [25]. 13. Reimported via Trajectory Tools > import plot jru v1.

Acknowledgments S.M. is supported by the German Research Foundation (Heisenberg funding MU3133/8-1). References 1. Brugue´s J, Needleman D (2014) Physical basis of spindle self-organization. PNAS 111(52): 18496–18500 2. Prosser SL, Pelletier L (2017) Mitotic spindle assembly in animal cells: a fine balancing act. Nature Rev Mol Cell Bio 18:187–201 3. Pavin N, Tolic´ IM (2016) Self-organization and forces in the mitotic spindle. Annu Rev Biophys 45:279–298 4. Kline-Smith SL, Walczak CE (2004) Mitotic spindle assembly and chromosome segregation: refocusing on microtubule dynamics. Mol Cell 15:317–327 5. Barisic M, Rajendraprasad G, Steblyanko Y (2021) The metaphase spindle at steady state – mechanism and functions of microtubule poleward flux. Semin Cell Dev Biol 117: 99–117

6. Mitchison T, Kirschner M (1984) Dynamic instability of microtubule growth. Nature 312:237–242 7. Dumont S, Mitchison TJ (2009) Force and length in the mitotic spindle. Curr Biol 19: R749-61 8. Buster DW, Zhang D, Sharp DJ (2007) Poleward tubulin flux in spindles: regulation and function in mitotic cells. Mol Biol Cell 18: 3094–3104 9. Zhang H, Dawe RK (2011) Mechanisms of plant spindle formation. Chromosom Res 19: 335–344 10. Nebenfu¨hr A, Dixit R (2018) Kinesins and myosins: molecular motors that coordinate cellular functions in plants. Annu Rev Plant Biol 69:329–361 11. Miki T, Naito H, Nishina M, Goshima G (2014) Endogenous localizome identifies

Analysis of Spindle Size Maintenance 43 mitotic kinesins in a plant cell. PNAS 111: E1053–E1061 12. Yamada M, Goshima G (2017) Mitotic spindle assembly in land plants: molecules and mechanisms. Biology 6:6 13. Lipka E, Herrmann A, Mueller S (2015) Mechanisms of plant cell division. WIREs Dev Biol 4:391–405 14. Herrmann A, Livanos P, Zimmermann S, Berendzen K, Rohr L, Lipka E, Mu¨ller S (2020) KINESIN-12E regulates metaphase spindle flux and helps control spindle size in Arabidopsis. Plant Cell 33:27–43 15. Leong SY, Edzuka T, Goshima G, Yamada M (2020) Kinesin-13 and Kinesin-8 Function During Cell Growth and Division in the Moss Physcomitrella patens. Plant Cell 2020: tpc.00521.2019 16. Vanstraelen M, Inze´ D, Geelen D (2006) Mitosis-specific kinesins in Arabidopsis. Trends Plant Sci 11:167–175 17. Marc J, Granger CL, Brincat J, Fisher DD, Kao T, McCubbin AG, Cyr RJ (1998) A GFP-MAP4 reporter gene for visualizing cortical microtubule rearrangements in living epidermal cells. Plant Cell 10:1927–1940 18. Lipka E, Gadeyne A, Sto¨ckle D, Zimmermann S, De Jaeger G, Ehrhardt DW, Kirik V, Van Damme D, Mu¨ller S (2014) The phragmoplast-orienting Kinesin-12 class proteins translate the positional information of the preprophase band to establish the cortical division zone in Arabidopsis thaliana. Plant Cell 26:2617–2632

125

19. Lindeboom JJ, Nakamura M, Hibbel A, Shundyak K, Gutierrez R, Ketelaar T, Emons AM, Mulder BM, Kirik V, Ehrhardt DW (2013) A mechanism for reorientation of cortical microtubule arrays driven by microtubule severing. Science 342:1245533 20. Yoneda A, Akatsuka M, Hoshino H, Kumagai F, Hasezawa S (2005) Decision of spindle poles and division plane by double preprophase bands in a BY-2 cell line expressing GFP-tubulin. Plant Cell Physiol 46:531–538 21. Murashige T, Skoog F (1962) A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiol Plant 15:473– 497 22. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez J-Y, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A (2012) Fiji: an opensource platform for biological-image analysis. Nat Methods 9:676–682 23. Schneider CA, Rasband WS, Eliceiri KW (2012) NIH image to ImageJ: 25 years of image analysis. Nat Methods 9:671–675 24. Peterson KM, Torii KU (2012) Long-term, high-resolution Confocal Time lapse imaging of Arabidopsis Cotyledon epidermis during germination. JoVE 2012:e4426 25. Livanos P, Chugh M, Mu¨ller S (2017) Analysis of Phragmoplast Kinetics During Plant Cytokinesis. In Plant Protein Secretion: Methods and Protocols, L. Jiang, ed (New York, NY: Springer New York), pp. 137–150

Chapter 10 Expansion Microscopy of Plant Cells (PlantExM) Timothy J. Hawkins, Joanne L. Robson, Bethany Cole, and Simon J. Bush Abstract Expansion microscopy (ExM) achieves super-resolution imaging without the need for sophisticated superresolution microscopy hardware through a combination of physical and optical magnification. Samples are fixed, stained, and embedded in a swellable gel. Following cross-linking of fluorophores to the gel matrix, the components of the sample are digested away and the gel expanded in water. Labeled objects which are too close to be resolved by diffraction-limited microscopy are moved far enough apart that these can now be resolved as individual objects on a standard confocal. Originally developed for animal cells and tissues, ExM for plants requires the additional consideration of cell wall digestion. Super-resolution can be limited in plants due to the size of cells, light scattering of tissues, and variations in refractive index. By removing the components which cause these limitations, ExM opens up the possibility of super-resolution at depth within plant tissues for the first time. Here we describe our method for PlantExM which is optimized for cytoskeleton resolution, which, when also coupled with compatible optical super-resolution technologies, can produce images of the plant cytoskeleton in unprecedented detail. Key words Super-resolution, Expansion microscopy, Fluorescence, Tobacco bright yellow 2 (BY2), Roots, Lightsheet, Cell division, Clearing

1

Introduction Expansion microscopy (ExM), developed by Boyden and colleagues [1], is a microscopy methodology capable of producing superresolution images on conventional microscopy equipment through a combination of both physical and optical magnifications. Furthermore, the technique acts as a clearing method, to allow superresolution microscopy deep within complex tissues by removing the light-absorbing and light-scattering components, leaving only a three-dimensional fingerprint of labeled structures within the optically inert gel (transparent and index matched to water). The initial study utilized a swellable polymer network of sodium acrylate (cross-linked by acrylamide and N-N’-methylenebisacrylamide) generated within and around the specimen. Monomers and crosslinkers are diffused into the sample before initiating free-radical

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_10, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

127

128

Timothy J. Hawkins et al.

polymerization. The label used to highlight proteins or structures of interest is then covalently cross-linked to the polymer. Following this the tissue is destroyed with the addition of protease to remove any components that would limit expansion, homogenizing the mechanical properties of the tissue and allowing the sample to expand isotropically. Upon the addition of water, the polymer network physically expands or swells isotropically, by a factor of approximately 4.5, resulting in physical magnification where labels spaced closer than the diffraction limit are separated allowing them to now be optically resolved. The labeling and cross-linking system developed by Boyden used a trifunctional construct comprising of a methacryloyl group, for direct incorporation into the polymer, a chemical fluorophore, and an oligonucleotide, complementary to a second oligonucleotide bound to the secondary antibody. This construct acts as a cross-linker between the secondary antibody and the gel network, marking those epitope sites with a fluorescent label in a protease compatible manner. Using spinning disk microscopy, significant resolution improvements were observed for HEK cells, labeled for microtubules and clathrin-coated pits, when comparing the pre- and postexpanded cell. The full width at half maximum (FWHM) of these observed microtubules was used to estimate the effective resolution of ExM at approximately 60 nm, a figure comparable with the diffraction-limited confocal lateral resolution of 250 nm divided by the expansion factor of 4.5. However, this initial method relied on the creation of expensive custom-made DNA-labeled antibodies as well as trifunctional constructs. Furthermore, genetically encoded fluorophores could not be imaged without antibody labeling. Work by both Chozinski et al. [2] and Tillerberg et al. [3] closely followed and widened the potential uptake of this method with the introduction of new polymer-linking methods which enabled the use of conventional fluorophore-labeled antibodies and fluorescent proteins. In these methodologies treatment of the sample within the gel with MA-NHS (methacrylic acid N-hydroxysuccinimide ester) or glutaraldehyde following staining provided a sufficient number of direct chemical linkages between the antibody and the hydrogel so that following protease digestion the antibody fragments remain linked to the hydrogel during expansion. Specifically, protein side chains within the antibody are modified with acryloyl moieties. Chozinski and colleagues used the technique to examine early anaphase PtK1 cells producing clear images of the mitotic spindle with distinctly resolved attachments between kinetochore fiber microtubule bundles and chromosomes. Again, considering the FWHM of observed immunostained microtubules gave a similar estimated expansion-corrected lateral spatial resolution of approximately 65 nm. These protein-anchoring approaches allow for the use of commercially available fluorescent secondary antibodies, and

Expansion Microscopy of Plant Cells

129

as such we used this approach as a basis on which to develop our plant methodology. Since the technique’s first publication, numerous groups have further developed the methodology and widened its application with new variants. Modification of polymer and anchoring chemistry to give improved expansion factors including iterative expansion, iExM [4] achieving a 20-fold expansion with a cleavable crosslinker or X10 utilizing a modified gelation mix to provide tenfold expansion [5]. Modification of cross-linking and sample disruption to allow the retention of fluorescent protein markers such as GFP, ProExM (protein retention ExM) [3], MAP (magnified analysis of the proteome) [6], and UltraExM [7] or the addition of collagenase for imaging deep into collagenous tissues [2, 8, 9]. ExM has also been used in combination with other microscopy technologies including lightsheet [10] to increase throughput and image larger volumes and in combination with super-resolution microscopy to further extend the achievable resolution including ExSIM [11, 12], ExSMLM [13], and STED [7, 14]; however these nanoscopy methods can require further improvements in staining densities. Finally, the ExM variant ExFISH allows RNA anchoring and in situ hybridization FISH for imaging of RNAs [15, 16]. Several groups have also extensively characterized the isotropy of expansion and found the mean error introduced by ExM to be approximately 1–5% of the measurement length. Plant cells and tissues pose significant challenges to light microscopy techniques, in particular super-resolution, because of the light scattering and refractive properties of the cell wall, density, and complexity of tissue and the presence of inherent lightabsorbing molecules. Furthermore, imaging at multi-cell depth in plant tissues is difficult, if not impossible for current super-resolution techniques. Considering these challenges expansion microscopy could significantly impact plant super-resolution microscopy by abolishing these limitations; however the techniques developed so far, although suitable for human cells and tissues, are fundamentally unsuitable and do not work with plant material. We reasoned that principally the cell wall would be the main limiting feature to ExM in plant species, constraining the cell from expanding and unaffected by the proteinase treatment. The cell wall can be successfully removed by enzymatic digestion, but the constituents of the wall would need to be taken into account when choosing the required mixture of enzymes. Furthermore, this enzymatic treatment needs to be conducted in a two-stage process, firstly as a short-duration mild treatment immediately following fixation to produce windows through which the primary antibodies can enter and then a second longer treatment following gelation but preceding protease treatment to fully destroy the remaining cell wall and allow unimpeded expansion. The digestion cannot be achieved in a single step as such a dramatic cell wall removal before

130

Timothy J. Hawkins et al.

labeling and cross-linking would disrupt the cell architecture resulting in artifactual cellular organization or total disruption in the final image. Our initial studies focused on suspension culture cells (tobacco bright yellow 2, BY2) as these cells are generally isolated or in short chains with minimal cell-to-cell contacts while also providing an excellent cell type in which to use ExM to fully explore the dense microtubular division arrays which would greatly benefit from this technique. Here fixation is carried out in a microtubule stabilization buffer and using a mixture of both paraformaldehyde and glutaraldehyde to allow rapid and slow fixation with different cross-linker lengths enabling capture of the cortical array as well as the dense division arrays. To aid gel casting, giving gel cubes of consistent dimensions and increasing throughput, we created a 3D-printed casting plate (Fig. 1a).

Fig. 1 Gel casting plate and diagrammatic representation of protocol. (a) 3D-printed ExM gel casting plate aids gel casting, produces gel cubes of consistent dimensions, and increases throughput. The cut corner allows for orientation. (b) Diagrammatic outline of the fixation, cell digestion, cross-linking, and expansion process. Plant-specific steps are highlighted in italics. 3D-printed chamber. (c) Expanded gel sample in iBidi 35 mm glass-bottom dish. Normal lens (Zeiss Plan-Apochromat 63×/1.40 NA Oil DIC M27) vs LD lens (Zeiss LD LCI (Long Distance Live Cell Imaging) Plan-Apochromat 63×/1.20 NA Immersion Correction DIC M27). Diagram represents the difference in working distance and as such axial imaging capabilities following 4–5× expansion

Expansion Microscopy of Plant Cells

131

Our protocol is illustrated in Fig. 1b. This stepwise fixation, mild cell wall digestion, labeling, cross-linking, and gelation followed by a strong cell wall digestion, protease K treatment, and expansion produces well-labeled, intact microtubule arrays with an average expansion factor of 3.9 but ranging up to 5.1. However, a 4× linear expansion corresponds to a 64-fold volumetric expansion, and therefore the ability to image deep into the expanded sample is limited by the working distance of the lens. Although mammalian cell culture cells can be successfully imaged with convectional immersion lenses, this challenge is increased when imaging plant cells or tissues due to their larger starting size. To be able to routinely image the full cortical and cytokinetic array without significant loss of resolution, we needed to employ a long working distance high numerical aperture lens (LD LCI Plan-Apochromat 63× NA 1.2 with Immersion Correction) (Fig. 1c). As discussed above, several groups have extended the achievable resolution of ExM by combining with super-resolution microscopy techniques. These have been successful but face issues with the depth of imaging at high resolution, labeling density, or limited expansion due to the buffers required. Considering these challenges and the expanded plant cell size, we considered Airyscan confocal to be an ideal companion technology to PlantExM. As well as being less susceptible to optical aberrations at depth, the increased sensitivity provided by the Airyscan detector allowed us to effectively image the lower fluorescent intensity following expansion due to probe loss and separation of fluorescence sources, effectively diluting the signal. Figure 2a shows the interphase cortical microtubule array of a PlantExM-processed BY2 cell. The image shows the clarity achievable with PlantExM, revealing the individual microtubules within a cortical bundle. To assess achieved resolution, the FWHM can be measured. This is a Gaussian fit to an image of a sub-resolution point source or fiber and gives a good approximation of achievable resolution. The cross-sectional profile of expanded microtubules yielded an average Gaussian-fitted FWHM of 256 ± 17.18 nm (mean ± s.d.). When this is divided by the estimated expansion factor, this gives an achieved FWHM of 64 ± 4.29 nm (mean ± s. d.). This value is consistent with the microtubule FWHM reported in other animal cell ExM studies and the theoretical size of an indirectly immunolabeled microtubule, where each antibody contributes 8.75 nm to the observed size (25-nm microtubule + 4 × 8.75-nm antibody = 60 nm). To illustrate the technique’s utility for studying plant cell division, Fig. 3 shows PlantExM images of all division arrays demonstrating the detail that can be revealed within these complex arrays. Figure 4 shows an anaphase spindle processed with PlantExM and zoomed in views of three regions. Here we can clearly see the organization of individual microtubules within the array; the complexity of the inter-woven

132

Timothy J. Hawkins et al.

Fig. 2 (a) An interphase BY2 cell expanded showing cortical array. (b) Boxed-out area (red square) shows zoomed details of array. Scale 1 μm. (c) An area of exactly the same size taken from a cortical array imaged with Airyscan but without expansion. Scale 1 μm. Lines across image show examples of those MTs measured to calculate FWHM. Considering expansion factor here of 4, the microtubule FWHM can be calculated as nm. Ten measurements were taken at different locations along ten microtubules (n = 100). Example FWHM profile (d)

Fig. 3 Expanded BY2 cells at different stages of mitosis, interphase cortical array (a), pre-prophase band (PPB) (b), prophase spindle (c), metaphase spindle (d), anaphase spindle (e), cytokinetic phragmoplast (f), late cytokinetic phragmoplast (reformed nuclear envelope) (g). Upper row—tubulin staining in green. Middle row—nuclear stain DAPI. Lower row—Merge. Scale = 5 μm

filaments entering and spanning the equator of the spindle (Fig. 4a), individual microtubules splaying out at the edge of the spindle (Fig. 4b), and the collection of microtubules at the pole of the spindle with singular microtubules extending upward (Fig. 4c). Following the success of the protocol with suspension culture cells, we further developed the approach to work with tissues and seedlings. The main element requiring adjustment was the cell wall enzyme mix. As mentioned above this should be changed to

Expansion Microscopy of Plant Cells

133

Fig. 4 Expanded BY2 cell anaphase spindle. Boxed-out areas show close-up details of the complex organization revealed by the technique at different zones within the spindle. (a) Equator (b). Array edge (c). Spindle pole (a, scale = 5 μm; b and c, scale = 2.5 μm)

include enzymes that will digest all of the distinct components of the cell wall which can differ by cell type and tissue. We tested a selection of mixture combinations and incubation periods. For the root example shown in Fig. 5, we found that a 45 min staining digestion and a 2 h post-gelation digestion with a mixture of driselase and macerozyme worked the best. Other mixtures included cellulase and pectolyase but showed no improvements with root tissue; however these may be necessary for other tissues. Arabidopsis seedling root tips were effectively expanded with excellent preservation of signal and morphology (Fig. 5). To enable us to image all cell layers throughout the root effectively, we needed to use lightsheet microscopy with a long working distance water-dipping lens (W Plan-Apochromat 20× NA 1.0 WD 2.4 mm). This approach also had the benefit of fast volumetric imaging for high-throughput imaging. As continuing to cast in a cube and slice was easier to manipulate than casting in columns, we again designed a 3D-printed manifold to allow us to attach the gel slice to a capillary for affixing into a Zeiss Lightsheet Z.1 for imaging and rotation. Here we could image microtubule arrays in all cell layers from one epidermis to the other (Fig. 6). This is significant as not only does this provide resolution improvement for lightsheet imaging of plants, but it effectively clears the tissue to allow imaging at full depth which can still be limited in conventional lightsheet imaging due to sheet penetration, shadowing, or light scattering. Although there are clearing methods available for plants, in particular ClearSee [17], these do not improve resolution and can take weeks to months to clear the tissue. These clearing

134

Timothy J. Hawkins et al.

Fig. 5 Arabidopsis root. Whole mount microtubule immunolabeling and PlantExM imaged using LSCM. (a) Processed root before expansion (gelled and protease-cleared alone) imaged with Airyscan 880 20× lens. Scale = 10 μm (b). Zoomed area showing spindle. (b) Same spindle imaged post expansion with LWD 63× lens. Scale = 5 μm

Fig. 6 Arabidopsis root. Whole mount microtubule immunolabeling and PlantExM. (a) Full-resolution image acquired with Zeiss Lightsheet Z.1 (20×). Boxed-out shows zoom examples of division arrays within Z series. Scale = 100 μm. Montage showing image acquisition throughout the full volume of the expanded root. Acquired image stack 673 slices, slice every 697 nm, total stack depth 489.26 μm. Every 28th image shown, e.g., every 19.5 μm

Expansion Microscopy of Plant Cells

135

methods also result in samples in solutions that are not suitable for imaging in a water immersion system (e.g., urea, xylitol, and sodium deoxycholate), necessitating special lenses and chambers which can survive such corrosive chemicals. Therefore, PlantExM opens up the opportunity for super-resolution, tissue depth imaging within water-based lightsheet systems. However, when working with highly lignified tissues such as the vasculature, we found that these are still retarded in their expansion due to the inability of our enzyme mix to degrade this robust polymer, particularly further up the root and as such expansion was limited to the root cap, meristem, root hairs, epidermis, cortex, endodermis, and pericycle. To date published use of the expansion microscopy technique in plants has been limited to a single-journal article [18] where it was used with whole mount immunostaining of ovule and seeds to identify active transcription in zygotes before the first cell division. However, in that instance the technique was essentially employed as a clearing solution to image through the seed coat and make interior tissues more accessible to reagents for more consistent immunostaining but not for nanoscopy. Furthermore, due to signal loss, protease treatment was not used resulting in significantly limited anisotropic expansion of only 1.3- to 2-fold. Our methodology discussed in this chapter is designed for nanoscopy and achieves plant super-resolution images consistent with those in animal cells with fourfold expansion and further improved resolution with Airyscan detection and processing. We will continue to develop the technique with advancements in polymer chemistry, cross-linking, and image processing and in particular lignin solubilization to further widen applicable tissues. For example, new image processing methods for Airyscan data such as Joint DeConVolution (jDCV) can further improve the resolution and clarity available in these acquired Airyscan datasets (Fig. 7). Currently this approach will have major benefits for super-resolution imaging of plant division helping to reveal the fine architectural arrangements of cytoskeletal arrays and associated protein distribution.

2

Materials

2.1 For Tobacco Bright Yellow 2 (BY2) Cell Maintenance

1. BY2 media: For a liter of liquid medium, add 30 g sucrose, 4.3 g of Murashige and Skoog (MS) basal salts, 100 mg of inositol, 1 mL of 1 mg/mL thiamine stock, 200 μL of 1 mg/ mL 2,4-D stock, 200 mg of KH2PO4 adjusted to pH 5.8 with 1 M NaOH. 2. 250 mL conical flasks. 3. Rotary shaker. 4. 10 mL serological pipettes.

136

Timothy J. Hawkins et al.

Fig. 7 BY2 PlantExM image (a). Standard Airyscan processing (b). Airyscan Processing jDCV (joint iterative DeConVolution)

5. Electric pipette controller. 6. Sterile flow hood. 7. 70% ethanol. 2.2 For Fixation and Immunofluorescence of BY2 Cells

1. 3-day-old BY2 cells. 2. 40–60 μm mesh strainer. 3. Humidity chamber. 4. Six-well plate. 5. PEM buffer: 50 mM PIPES, 5 mM EGTA, 2 mM MgSO4. 6. Fixative: 3.7% paraformaldehyde, 0.02% glutaraldehyde (GA), 0.5% Triton X-100 in PEM buffer (see Note 1). 7. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl, 10 mM NaHPO4, 2 mM KH2PO4, pH to 7.2. 8. Protease inhibitors mix: 1 mM PMSF, 40 mM leupeptin, and 20 mM pepstatin A. 9. Cell wall digestion buffer: 1% macerozyme R-10, 0.2% pectolyase, and 0.4 M mannitol in PEM buffer supplemented with protease inhibitors mix. 10. Poly-L-lysine. 11. 22 mm × 22 mm #1.5 coverslips coated with poly-L-lysine. 12. Block buffer: 2% (w/v) BSA in PBS buffer. 13. Primary antibody rat anti-tubulin alpha (Bio-Rad MCA78G): dilution 1:100 in block buffer.

Expansion Microscopy of Plant Cells

137

14. Secondary antibody goat anti-rat Alexa Fluor 488 (Thermo Fisher A21208): dilution 1:100 in block buffer. 15. Parafilm. 16. Hoechst 3342. 2.3 For Gelation and Expansion

1. Cross-linking solution: 0.25% GA in PBS. 2. Monomer solution: 8.6% sodium acrylate, 2.5% acrylamide, 0.15% N,N’-methylenebisacrylamide, 11.7% sodium chloride in PBS. 3. Gelation solution: monomer solution plus 0.2% TEMED and 0.2% APS. 4. 3D-printed gelation tray. 5. Cell wall digestion buffer: as above excluding the protease inhibitors. 6. TAE buffer: 40 mM Tris–HCl pH 7.6, 20 mM acetic acid, 1 mM EDTA. 7. Digestion buffer: 1× TAE, 0.5% Triton X-100, 0.8M guanidine HCl, with 8 units/mL proteinase K (New England Biolabs, P8107S). 8. 37 °C incubator. 9. Square petri dish. 10. Distilled water. 11. DNA stain: Hoechst® 33342 (Life Technologies®) at dilution of 1:5000 in MQ water.

2.4

Microscopy

1. iBidi 35 mm glass-bottom dishes. 2. Scalpel. 3. Projector sheet cut to ~2 cm × 5 cm. 4. Zeiss 880 laser scanning confocal microscope with Airyscan. 5. Zeiss long working distance (LWD) lens: (0.49 mm) LD LCI Plan-Apochromat 63×, NA 1.2, Immersion Correction. 6. Zeiss Zen Black software. 7. Zeiss Lightsheet Z.1. 8. Superglue. 9. 3D-printed manifold (see Note 2).

2.5 Fixation of Arabidopsis Seedlings

1. Arabidopsis seedlings. 2. 1.3% sodium hypochlorite. 3. ½ MS agar plates. 4. Sterile flow hood. 5. Growth cabinet.

138

Timothy J. Hawkins et al.

6. Mild cell wall digestion buffer: 0.2% driselase, 0.15% macerozyme in 2 mM MES pH 5.0 in PEM buffer. 7. Membrane permeabilization buffer: 3% IGE-PAL CA-630, 10% DMSO in PEM buffer. 8. Other buffers as for BY2 cells.

3

Methods

3.1 Maintenance of BY2 Tobacco Cell Culture

1. 50 mL of BY2 media dispensed into 250 mL conical flasks, neck covered with triple layer of foil, and sterilize by autoclaving. 2. BY2 cells are sub-cultured every 7 days, and 3–3.5 mL of cells are extracted from flask under sterile conditions and inoculated into two fresh 50 mL of media as sister cultures. 3. Sub-culture from one sister culture, and leave other sister cultures untouched as a back-up culture for use in the case of contamination. 4. Place each flask onto a rotary shaker at 114 rpm, in the dark at 23 °C.

3.2 Fixation of BY2 Cells and Immunofluorescence

1. Aliquot of 1–2 mL of 3-day-old BY2 cells taken from culture. 2. Sieve through 40–60 μm mesh strainer inside a 50 mL beaker to remove culture media. 3. Apply fixative immediately for 35 min at room temperature. 4. Wash cells three times in PBS buffer. 5. Apply cell wall digestion buffer for 10 min. 6. Wash cells three times in PBS buffer. 7. Apply fixed cells to several poly-L-lysine coated #1.5 coverslips. 8. Incubate in a humidity chamber for 20–30 min to allow fixed cells to attach to the coverslips. 9. Transfer coverslips with attached BY2 cells to a six-well plate. 10. Incubate in block buffer for 20 min. 11. Incubation with primary antibody overnight at 4 °C in a humidity chamber. Add 50 μL of antibody on a piece of Parafilm and cells on coverslip laid into the droplet. 12. Three times 3-min PBS washes done in a six-well plate. 13. Incubation with secondary antibody for 1–2 h at room temperature in the humidity chamber as described above. 14. Three PBS washes in a six-well plate. 15. DNA stain for 10 min followed by three more PBS washes. 16. Check for good staining without allowing sample to dry before proceeding to expansion protocol.

Expansion Microscopy of Plant Cells

3.3 Gelation, Digestion, and Expansion of Cultured Cells

139

1. Incubate fixed and stained cells in cross-linking solution for 10 min. 2. Three times 2 min PBS washes. 3. Incubate in monomer solution for 1 min. 4. Apply gelation solution to gelation tray, and cells then laid onto it and allowed to polymerize for 30 min (see Note 3). 5. Incubate the resulting gel in cell wall digestion buffer for 2 h at 30 °C. 6. Incubate in proteinase K digestion buffer for 1 h at 37 °C. 7. Immerse the gel in an excess volume of water in a square petri dish, and change the water three to four times until the gel is fully expanded. The 1.5 cm2 gelation tray should produce an expanded gel of 6–7 cm2, therefore 4–4.5× expansion. Measure the expanded gel and take note. 8. Incubate in DNA stain for 10 min.

3.4 Arabidopsis Seedling Sterilization and Growth

1. Sterilize Arabidopsis seeds in 1.3% sodium hypochlorite for 10 min. 2. Rinse ten times in water. 3. Gently place the seeds on plates, and evenly space out along a line. 4. Wrap the plate in foil, and put into a fridge to vernalize for 2 days. 5. Transfer the plate to a growth cabinet to grow vertically.

3.5 Fixation of Arabidopsis Seedlings and Immunofluorescence

1. Gently transfer 3–4-day-old seedlings directly into fixative for 1 h. 2. Five times 5-min PEM washes. 3. Apply mild cell wall digestion buffer for 45 min. 4. One 5 min PEM wash. 5. Apply membrane permeabilization buffer for 20 min. 6. One 5 min PEM buffer wash. 7. Incubate the seedlings in block buffer for 20 min. 8. Incubate them with primary antibody for 24–48 h at 4 °C. 9. Three PBS washes. 10. Incubate the seedlings with secondary antibody for 24–48 h at 4 °C. 11. Three PBS washes. 12. DNA stain for 10 min followed by three more PBS washes. 13. Check for good staining without allowing sample to dry before proceeding to expansion protocol.

140

Timothy J. Hawkins et al.

3.6 Gelation, Digestion, and Expansion of Seedlings

1. Incubate fixed and stained seedling in cross-linking solution for 1 h. 2. Three times 5 min washes in PBS. 3. Incubate the seedlings in monomer solution for 45 min at 4 °C. 4. Apply gelation solution to the seedlings, and gently lay between two coverslips to form a very thin layer of gel, and allow to polymerize for 2 h. 5. Cut away excessive gel from around the seedling, and measure its size. 6. Incubate the resulting gel in cell wall digestion buffer for 2 h at 30 °C. 7. Quick-wash in TAE buffer. 8. Incubate the gel in digestion buffer overnight at 37 °C. 9. Immerse the gel in an excess volume of water, and change the water three to four times until the gel is fully expanded. Measure the expanded gel, and take note for expected expansion factor. 10. Incubate gel in DNA stain for 10 min.

3.7 Confocal Microscopy and Analysis

1. Apply 10 μL of poly-L-lysine to a glass-bottom dish for approximately 1 min. 2. Use scalpel to cut a 1 cm2 piece of gel, and use projector sheet to transfer it to the glass-bottom dish ensuring cell side is touching the glass. 3. Check the gel piece does not move; the poly-L-lysine should hold it in place. 4. Place the dish onto the confocal microscope, and image for the desired fluorophores using Airyscan mode. 5. Raw Airyscan images processed with automatic-detected Wiener filter. 6. To calculate the resolution, draw a line perpendicular to several microtubules in several images, take an average of ten measurements in each area using Adrian’s FWHM plugin for Fiji (https://imagej.nih.gov/ij/plugins/fwhm/) divided by the expansion factor to give measured resolution.

3.8 Lightsheet Microscopy and Analysis

1. Use scalpel to cut approximately 1 cm2 piece of gel around the seedling (see Subheading 3.6). 2. Superglue the gel piece to a coverslip. 3. Superglue the coverslip with gel to 3D-printed manifold. 4. Manifold attaches to capillary size 4, Zeiss sample holder stem, and disc for insertion into Lightsheet Z.1. 5. Sample rotated appropriately and imaged for the desired fluorophores. 6. Resolution calculations as in Subheading 3.7, step 6.

Expansion Microscopy of Plant Cells

4

141

Notes 1. Add paraformaldehyde powder into PEM buffer, heat the mixture, and stir at 60 °C until a clear solution is formed. Adjust the pH to approximately 7.0 with KOH if required. 2. The manifold allows attachment of gel to capillary (size 4), which in turn attaches in Zeiss capillary stem holder and disc for insertion, rotation, and imaging in the Zeiss Lightsheet Z.1. 3. The use of CAD-designed gelation tray (Fig. 1a) is extremely useful to ensure reproducible sized non-expanded gels which can be measure post-expansion to give an indication of full expansion with water. The grooved pattern of the plastic in the bottom of the tray inherent from the 3D printing machine results in an easily identifiable correct rotation of the gel for imaging; smooth side equates to the side in contact with cells and hence the side which should be imaged. The cut corner of each well also helps for orientation of the gel when imaging pre- and post-expansion; this orientation marker helps home into the correct area of the gel. The six wells of the tray ensure multiple expansions can be done easily, and the uniform size enables to the correct volume of gelation solution to be used each time. It is worth cleaning the gelation before use with 70% ethanol to help the gel come away from the tray post gelation.

References 1. Chen F, Tillberg PW, Boyden ES (2015) Expansion microscopy. Science 347(6221): 543–548 2. Chozinski TJ, Mao C, Halpern AR, Pippin JW, Shankland SJ, Alpers CE, Najafian B, Vaughan JC (2018) Volumetric, nanoscale optical imaging of mouse and human kidney via expansion microscopy. Sci Rep 8(1):10396. https://doi. org/10.1038/s41598-018-28694-2 3. Tillberg PW, Chen F, Piatkevich KD, Zhao Y, Yu CC, English BP, Gao L, Martorell A, Suk HJ, Yoshida F, DeGennaro EM, Roossien DH, Gong G, Seneviratne U, Tannenbaum SR, Desimone R, Cai D, Boyden ES (2016) Protein-retention expansion microscopy of cells and tissues labeled using standard fluorescent proteins and antibodies. Nat Biotechnol 34(9):987–992. https://doi.org/10.1038/ nbt.3625 4. Chang JB, Chen F, Yoon YG, Jung EE, Babcock H, Kang JS, Asano S, Suk HJ, Pak N, Tillberg PW, Wassie AT, Cai D, Boyden ES (2017) Iterative expansion microscopy. Nat Methods 14(6):593–599. https://doi.org/10. 1038/nmeth.4261

5. Truckenbrodt S, Maidorn M, Crzan D, Wildhagen H, Kabatas S, Rizzoli SO (2018) X10 expansion microscopy enables 25-nm resolution on conventional microscopes. EMBO Rep 19(9). https://doi.org/10.15252/embr. 201845836 6. Ku T, Swaney J, Park JY, Albanese A, Murray E, Cho JH, Park YG, Mangena V, Chen J, Chung K (2016) Multiplexed and scalable super-resolution imaging of threedimensional protein localization in sizeadjustable tissues. Nat Biotechnol 34(9): 973–981. https://doi.org/10.1038/nbt. 3641 7. Gambarotto D, Zwettler FU, Le Guennec M, Schmidt-Cernohorska M, Fortun D, Borgers S, Heine J, Schloetel JG, Reuss M, Unser M, Boyden ES, Sauer M, Hamel V, Guichard P (2019) Imaging cellular ultrastructures using expansion microscopy (U-ExM). Nat Methods 16(1):71–74. https://doi.org/10. 1038/s41592-018-0238-1 8. Unnersjo-Jess D, Scott L, Sevilla SZ, Patrakka J, Blom H, Brismar H (2018) Confocal super-resolution imaging of the glomerular

142

Timothy J. Hawkins et al.

filtration barrier enabled by tissue expansion. Kidney Int 93(4):1008–1013. https://doi. org/10.1016/j.kint.2017.09.019 9. Zhao Y, Bucur O, Irshad H, Chen F, Weins A, Stancu AL, Oh EY, DiStasio M, Torous V, Glass B, Stillman IE, Schnitt SJ, Beck AH, Boyden ES (2017) Nanoscale imaging of clinical specimens using pathology-optimized expansion microscopy. Nat Biotechnol 35(8): 757–764. https://doi.org/10.1038/nbt. 3892 10. Horl D, Rojas Rusak F, Preusser F, Tillberg P, Randel N, Chhetri RK, Cardona A, Keller PJ, Harz H, Leonhardt H, Treier M, Preibisch S (2019) BigStitcher: reconstructing highresolution image datasets of cleared and expanded samples. Nat Methods 16(9): 870–874. https://doi.org/10.1038/s41592019-0501-0 11. Cahoon CK, Yu Z, Wang Y, Guo F, Unruh JR, Slaughter BD, Hawley RS (2017) Superresolution expansion microscopy reveals the threedimensional organization of the Drosophila synaptonemal complex. Proc Natl Acad Sci U S A 114(33):E6857–E6866. https://doi.org/ 10.1073/pnas.1705623114 12. Halpern AR, Alas GCM, Chozinski TJ, Paredez AR, Vaughan JC (2017) Hybrid structured illumination expansion microscopy reveals microbial cytoskeleton organization. ACS Nano 11(12):12677–12686. https://doi. org/10.1021/acsnano.7b07200 13. Zwettler FU, Reinhard S, Gambarotto D, Bell TDM, Hamel V, Guichard P, Sauer M (2020)

Molecular resolution imaging by post-labeling expansion single-molecule localization microscopy (Ex-SMLM). Nat Commun 11(1):3388. https://doi.org/10.1038/s41467-02017086-8 14. Pesce L, Cozzolino M, Lanzano L, Diaspro A, Bianchini P (2019) Measuring expansion from macro- to nanoscale using NPC as intrinsic reporter. J Biophotonics 12(8):e201900018. https://doi.org/10.1002/jbio.201900018 15. Chen F, Wassie AT, Cote AJ, Sinha A, Alon S, Asano S, Daugharthy ER, Chang JB, Marblestone A, Church GM, Raj A, Boyden ES (2016) Nanoscale imaging of RNA with expansion microscopy. Nat Methods 13(8): 679–684. https://doi.org/10.1038/nmeth. 3899 16. Wang G, Moffitt JR, Zhuang X (2018) Multiplexed imaging of high-density libraries of RNAs with MERFISH and expansion microscopy. Sci Rep 8(1):4847. https://doi.org/10. 1038/s41598-018-22297-7 17. Kurihara D, Mizuta Y, Sato Y, Higashiyama T (2015) ClearSee: a rapid optical clearing reagent for whole-plant fluorescence imaging. Development 142(23):4168–4179. https:// doi.org/10.1242/dev.127613 18. Kao P, Nodine MD (2019) Transcriptional activation of Arabidopsis Zygotes is required for initial cell divisions. Sci Rep 9(1):17159. https://doi.org/10.1038/s41598-01953704-2

Chapter 11 Microfluidic Device for High-Resolution Cytoskeleton Imaging and Washout Assays in Physcomitrium (Physcomitrella) patens Mari W. Yoshida and Elena Kozgunova Abstract Visualizing cytoskeleton dynamics at high spatiotemporal resolution provides valuable insights into the way the dynamics change as well as its interactions with multiple proteins in order to maintain cellular function. Oblique illumination fluorescent microscopy is a popular technique to image cellular events localized near the plasma membrane. In this chapter, we provide detailed protocols for high-resolution cytoskeleton imaging using protonema and gametophore cells of the moss Physcomitrella (Physcomitrium patens) in the microfluidic device. These include preparation of the polydimethylsiloxane (PDMS) device, culture of moss cells, and both short- and long-term oblique illumination fluorescent microscopy. We also describe how to introduce to, and wash out from, the device chemical compounds, such as microtubule-disrupting drugs, during live-cell imaging. Key words Physcomitrella, Microfluidic, Cytoskeleton imaging, TIRF

1

Introduction Moss Physcomitrium (Physcomitrella) patens became a popular model system in the last few decades, owing to easy culture, fast growth, a predominantly haploid life cycle, and the availability of a wide range of molecular biology methods [1]. Furthermore, Physcomitrella is distinguished by one of the highest rates of homologous recombination among land plants that enable easy genome manipulation, such as target gene knockout and fluorophore tagging to endogenous proteins. It is also amenable to high-resolution live-cell microscopy that is used to study various cytoskeletondependent processes, such as phragmoplast guidance, chromosome separation, and organelle transport [2–4]. Oblique illumination

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_11, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

143

144

Mari W. Yoshida and Elena Kozgunova

fluorescent microscopy has been used to visualize cytoskeleton events that occur close to the cell surface at a near single-molecule level, including microtubule nucleation or movement of cytoskeleton motor proteins. Several techniques have been developed for Physcomitrella livecell imaging. These techniques include moss culture in a thin agar layer on a glass-bottom dish or by placing moss tissue between coverslips for short-term imaging. Both methods have their own limitations. For example, the thin layer of agar is prone to drying, making it challenging to image moss in development or slowgrowing mutants. The first microfluidic device for Physcomitrella was developed to circumvent these problems, enabling long-term culture and imaging at high resolution [5]. On the other hand, covering moss with a coverslip often inhibits cell growth after a few hours and can lead to decreased viability. Therefore, another microfluidic device featuring four independent shallow channels was developed to perform long-term oblique illumination imaging and washout assays [6]. In recent years, microfluidic technology is gaining popularity for plant research [7]. Microfluidic devices can be composed of various materials, including glass, paper, polymers, and other materials. Now, polydimethylsiloxane (PDMS) is widely used to fabricate microdevices for biological applications. PDMS is an inexpensive, gas-permeable, and biocompatible material with excellent optical properties, making it especially suitable for live-cell culture and imaging. Another major advantage of microfluidics is the precise control of the liquid in the channels, which can be useful to study intracellular processes by applying chemical compounds, such as cytoskeleton inhibitors and hormones, during live-cell imaging. Efficient washout of the compound also provides the opportunity to study reversibility of certain processes, for instance, microtubule depolymerization and re-growth. In this chapter, we describe how to fabricate a microfluidic device from PDMS, introduce moss cells into the device, perform short-term and long-term oblique illumination fluorescent imaging, and how to combine high-resolution live-cell imaging with washout assays. We also provide a list of Physcomitrella marker lines that can be used for oblique illumination fluorescent imaging.

2

Materials

2.1 Microdevice Fabrication and Bonding to the GlassBottom Dish

1. Silicon mold (see Note 1). 2. Polydimethylsiloxane (Sylgard 184; Dow Corning). 3. Biopsy punch (1.5 mm diameter). 4. Plasma chamber (YHS-R, SAKIGAKE-Semiconductor).

Microfluidic Chip for Cytoskeleton imaging in Physcomitrella

145

5. Glass-bottom dish (dish diameter 35 mm; glass diameter 27 mm; glass thickness 0.16–0.19 mm) or coverslip (glass thickness 0.16–0.19 mm). 6. Vacuum chamber. 7. Heat chamber (60 °C). 8. Scales. 9. Scalpel, scissors. 2.2 Moss Culture in the Microdevice

General protocols for Physcomitrella culture, maintenance, and genetic transformation are described on the website PHYSCO manual (www.nibb.ac.jp/evodevo/PHYSCOmanual/00Eindex). Here we provide details specific to moss culture in the microdevice: 1. BCDAT growth medium: 5 mM ammonium tartrate, 1 mM MgSO4 7H2O, 1.84 mM KH2PO4, 10 mM KNO3, 1 mM CaCl2 2H2O, 45 μM FeSO4 7H2O, 0.22 μM CuSO4 5H2O, 10 μM H3BO3, 0.23 μM CoCl2 6H2O, 0.1 μM Na2MoO4 2H2O, 0.19 μM ZnSO4 7H2O, 2 μM MnCl2 4H2O, 0.17 μM KI. The following stock solutions are used to prepare the BCDAT medium: 500 mM ammonium tartrate, 100 mM MgSO4 7H2O, 184 mM KH2PO4, pH 6.5 with KOH, 1 M KNO3, 50 mM CaCl2, 4.5 mM FeSO4 7H2O. 2. BCDAT agar plates: BCDAT liquid medium with 0.8% (w/v) agar. 3. Polyethylene tubing (I.D. 0.76 mm; O.D. 1.22 mm, Intramedic). 4. Syringe (1 mL). 5. Needles (0.80 × 38 mm). 6. UVP Crosslinker CL-1000 (Analytik Jena). 7. Growth chamber. 8. 50 mL falcons. 9. Cellophane, pre-cut to fit a standard Petri dish and autoclaved (optional). 10. Tissue homogenizer (OMNI TH with 34750H probe, Omni International), required for washout assays. 11. 50 μm nylon mesh filter fit to the funnel (autoclaved), required for washout assays.

2.3

Microscopy

1. Nikon Ti microscope with a TIRF unit. 2. 100× 1.49 NA lens (Nikon). 3. GEMINI split view (Hamamatsu). 4. EMCCD camera Evolve (Roper).

146

2.4

Mari W. Yoshida and Elena Kozgunova

Washout Assays

1. Syringe pump (YSP-202, YMC). 2. Syringe (1 mL). 3. Needles (0.80 × 38 mm). 4. Polyethylene tubing (I.D. 0.76 mm; O.D. 1.22 mm, Intramedic). 5. BCDAT growth medium.

3

Methods

3.1 Microdevice Fabrication

Our device design is composed of four independent channels (see Fig. 1). Therefore, up to four different marker lines (Table 1) can be cultured in a single device. Both ends of the channel inlet/outlet holes are pierced to introduce or remove liquid. Channel depth is 15μm, which is slightly smaller than an average diameter of ~20μm for Physcomitrella protonema cells. In previous work, several devices of various depths were tested, and 15μm was determined to be the optimal for oblique illumination fluorescent microscopy [6]. 1. For the PDMS pre-polymer mixture, the elastomer base and curing agent are weighed in a ratio of 10:1 (w/w) and vigorously mixed using a 1-ml pipette tip.

Fig. 1 Manufacture of the PDMS device using soft lithography. (a) A step-by-step guide on making a PDMS device for Physcomitrella imaging and attaching it to the coverslip glass. (b) Silicon master mold with six microdevice patterns. (c) A single microdevice featuring four independent channels attached to the coverslip glass. One of the channels is connected to the tubing and filled with red dye for illustration purposes

Miki et al. [15] Nakaoka et al. [16]

Nakaoka et al. [8]

Kosetsu et al. [17]

Microtubules Microtubules/kinesins (72 genes) Augmin2 Augmin4 γ-Tubulin-b XMAP215/microtubules γ-Tubulin-b/microtubules Microtubules/endoplasmic reticulum Microtubules/Golgi Microtubules/mitochondria Microtubules/peroxisome MAP65 (five genes) MAP65/microtubules EB1/microtubules Spiral2a/microtubules Microtubules/CLoG1 Actin Microtubules/actin Actin Actin Formin2A

mCherry-α-tubulin mCherry-α-tubulin/kinesin-Citrine

Aug2-Citrine Aug4-Citrine γ-Tubulin-b-Citrine

PpXMAP215 (PpMOR1)-a-Citrine/mCherry-α-tubulin γ-Tubulin-b-Citrine/mCherry-α-tubulin GFP-α-tubulin/ER-mCherry GFP-α-tubulin/Man1-mRFP GFP-α-tubulin/γ-ATPase-mRFP GFP-α-tubulin/ mCherry-SKL

MAP65-Citrine MAP65-Citrine/mCherry-α-tubulin Citrine-MAP65/ mCherry-α-tubulin EB1-Citrine/mCherry-α-tubulin

SPR2a-Citrine/mCherry-tubulin

mCherry-tubulin/CLoG1-mEGFP

GFP-talin

GFP-α-tubulin/tdTomato-talin

LifeAct-GFP

LifeAct-mCherry

For2A-3XmEGFP

Vidali et al. [24]

Yi et al. [23]

Vidali et al. [22]

(continued)

Yamashita et al. [21]

Finka et al. [20]

Ding et al. [19]

Leong et al. [18]

Hiwatashi et al. [14]

Microtubules KINID1/microtubules

GFP-α-tubulin KINID1-Citrine or GFP/mRFP-α-tubulin

References

Proteins/cellular structures labeled

Marker line

Table 1 P. patens marker lines for oblique illumination fluorescent imaging

Microfluidic Chip for Cytoskeleton imaging in Physcomitrella 147

Proteins/cellular structures labeled MyosinXIa Actin/actin-interacting protein1 arpc4 (Arp2/3 complex) BRICK1 (WAVE/SCAR complex) MyosinVIIIA/microtubules MyosinVIIIA/actin Formin2A/actin Actin/microtubules RAB-E/MyosinXIa Golgi Mitochondrion Peroxisome Mitochondrion Cellulose synthase5 Cellulose synthase8

Marker line

3xmEGFP-myoXIa

LifeAct-GFP/AIP1-mCherry

YFP-ARPC4

BRK1-YFP

MyoVIIIA-GFP/LifeAct-mCherry MyoVIIIA-GFP/mCherry-α-tubulin For2A-GFP/ mCherryα-tubulin LifeAct-GFP/mCherry-α-tubulin

3xmCherry-Rab-E14/MyosinXIa-3xmEGFP

YFP-Man GFP-COX4 CFP-SKL

γ-ATPase-mRFP

mEGFP-PpCESA5 mEGFP-PpCESA8

Table 1 (continued)

Tran et al. [32]

Uchida et al., [31]

Furt et al. [30]

Orr et al. [29]

Wu and Bezanilla [2]

Perroud and Quatrano [28]

Perroud and Quatrano [27]

Augustine et al. [26]

Vidali et al. [25]

References

148 Mari W. Yoshida and Elena Kozgunova

Microfluidic Chip for Cytoskeleton imaging in Physcomitrella

149

2. The solution is degassed (see Note 2) for 10–15 min at room temperature in a vacuum chamber and poured over the silicon mold (see Fig. 1a, b). 3. The silicon mold is placed in the vacuum chamber and again degassed for 20–30 min under vacuum, until no air bubbles are visible (see Note 3). 4. The mold is placed in the pre-heated chamber at 60 °C for 90 min to cure the PDMS (see Note 4). 5. A large square containing all device patterns (see Note 5) is cut using a scalpel and is gently peeled from the mold. 6. Individual device patterns are separated using scissors or a scalpel. The device must be cut to a slightly smaller size than the glass where it will be attached. 7. The inlet/outlet holes (see Note 6) to the microchannel are made using a biopsy punch (see Note 7). 8. Both device and the coverslip glass (or a glass-bottom dish) are prepared for plasma treatment. Surfaces that will be treated with air plasma are cleaned from dust and small particles using scotch tape, which is attached and peeled off multiple times to ensure that all particles are collected (see Note 8). 9. The PDMS device and coverslip glass (alternatively, a glassbottom dish) are both treated with air plasma (see Note 9) for 20 s. The PDMS device must be placed so that the side with the pattern faces upward. Surfaces that were exposed to plasma treatment are immediately pressed together and heated for at least 2 h at 60 °C for irreversible bonding (see Note 10) of the PDMS layer to the glass surface. 3.2 Culture of P. patens Protonema Cells in the Microdevice

Protonema cells are tip-growing cells that appear following spore germination. Protonema grows as a single-cell layer, making it a popular object for microscopy. They have high regeneration ability and grow well in the PDMS device, given regular replacement of the medium [5]. For oblique illumination fluorescent microscopy, we introduce protonema cells to the microfluidic device and culture for 2–3 days to allow regeneration (see Fig. 2a). 1. The device inlet holes are connected to the small pieces of polyethylene tubing, cut to a height that will not prevent dish lid from closing. 2. The device is exposed to a vacuum for a few minutes and immediately filled with BCDAT liquid medium (see Note 11) using a 1 ml syringe with a needle through the tubing. 3. Next, the device is sterilized under UV 2000 × 100μJ/cm2, making sure that the dish cover is open. If the device is attached to the coverslip, a sterile Petri dish is used.

150

Mari W. Yoshida and Elena Kozgunova

Fig. 2 Illustrations of the procedures to introduce moss Physcomitrella to the microdevice channels. (a) General workflow to introduce protonema (tip-growing cells) to the device and a representative image of protonema cells after 3 days of culture in the microdevice. Bar, 100μm. (b) General workflow to introduce gametophore leaf cells to the device and representative image of gametophore cells in the device. Red arrowheads indicate regenerated protonema cells. Bar, 100μm. Adapted from Kozgunova and Goshima, (2019) under a Creative Commons Attribution 4.0 International License

4. (Optional) We usually use protonema cells that are cultured for 4–6 days on the BCDAT agar plate over cellophane. Detailed protocols for routine Physcomitrella culture and maintenance can be found on the PHYSCO manual website (www.nibb.ac. jp/evodevo/PHYSCOmanual/00Eindex). 5. (From here onward, the steps should be carried out in the clean bench/sterile hood) A small piece of protonema tissue (approximately 0.5 cm2) is picked from cellophane or separated from a colony and placed into 15 mL falcon tube with 2~ ml BCDAT liquid medium and briefly disrupted for 5–10 s using a homogenizer. 6. (Required for washout assays, otherwise optional) Solution is filtered through the 50 μm nylon mesh filter into the clean falcon tube. This step separates small clusters of protonema

Microfluidic Chip for Cytoskeleton imaging in Physcomitrella

151

cells (typically one to three cells) and prevents blocking inlets, tubing, and channels with excess cells. 7. Solution from the preceding steps is collected into the 1 mL syringe with a needle and introduced into the device channels through the tubing. 8. (Optional) The microdevice can be viewed using an inverted microscope with a 5× or 10× lens to confirm the presence of protonema cells in the channels. 9. The space around the microdevice is filled with BCDAT medium, and the glass-bottom or Petri dish containing the microdevice is sealed with Parafilm and kept in the growth chamber at 25 °C with continuous light for 2–3 days (or longer; see Note 12) prior to imaging. 3.3 Introducing Gametophore Leaf Cells into the Microdevice

Gametophores are leafy shoots that develop from protonema cells after 2–3 weeks of culture [1]. Unlike tip-growing protonema, gametophore cells expand through diffuse growth, which is typical for most cell types in vascular plants with exception of pollen tubes and root hairs. Consequently, microtubules in gametophore cells are organized in cortical arrays which contrasts with a threedimensional endoplasmic microtubule network in protonema cells [8–10]. Upon damage or excision, some gametophore cells undergo cell reprogramming and turn into tip-growing protonema cells [11]. Using the microfluidic device for long-term imaging, we can observe changes in the microtubule organization inflicted by such reprogramming (see Fig. 2b). 1. The device preparation and sterilization in the same way as for protonema cells. 2. (All steps are carried out in clean bench/sterile hood) Several gametophores (2–3) are picked from the moss colony and placed in the sterile Eppendorf tube containing 1 mL BCDAT medium. 3. Sterile small scissors are used to dissect gametophores into small segments. 4. Alternatively, gametophores can be dissected with a scalpel or razor blade immersed in a drop of BCDAT medium on the glass slide. 5. Gametophores can also be dissected with a homogenizer for 30–60 s to produce smaller pieces that can be introduced into the device. This method requires 10–20 gametophores in 5 ml of BCDAT medium in a 15 mL falcon tube. 6. Medium containing fragments of gametophore is collected into the 1 mL syringe with a needle and introduced into the device channels through the tubing. This typically requires more effort compared to protonema cells as the gametophore fragments are larger.

152

Mari W. Yoshida and Elena Kozgunova

7. (Optional) We check the device under 5× or 10× lens to confirm that gametophore leaf cells are present in the channels (Fig. 2b). 8. Imaging of gametophore leaf cells is performed immediately after the device is prepared. 3.4 Oblique Illumination Fluorescent Imaging of Microtubules

Oblique illumination fluorescent imaging techniques, such as total internal reflection fluorescence (TIRF) microscopy or variableangle epifluorescence microscopy (VAEM), can visualize events that occur close to the cell surface with high resolution. For short-term imaging up to 2–3 h, we used a microdevice attached to the coverslip to reduce the costs. For long-term imaging (48 h), we used the microdevice attached to the glass-bottom dish filled with growth medium to maintain humidity and prevent liquid evaporation from the microdevice channels. The microscope was controlled with NIS elements software version 5.20.02. 1. Glass-bottom dish or the coverslip with the microdevice was placed on the microscope stage. It should be noted that it can be additionally fixed with tape to minimize shifting. We made use of a 100× 1.49 NA lens with immersion oil. 2. After finding cells in the channel, the laser angle was gradually adjusted until individual microtubules were clearly visible (see Fig. 3). 3. Optimal laser power and exposure time must be determined empirically, depending on the expression levels of the

Fig. 3 Short-term and long-term oblique illumination fluorescent imaging of Physcomitrella cells cultured in the microdevice. (a) Short-term imaging of the protonema cells expressing LifeAct-mNeonGreen (actin, green) and mCherry-Tubulin (microtubules, magenta). Bar, 10μm. (b) Long-term imaging of microtubule reorganization in a gametophore cell, expressing GFP-tubulin, from cortical arrays (0 h) to endoplasmic microtubule network (24 h) typical for protonema cells. Images were acquired every 5 min. Bar, 10μm

Microfluidic Chip for Cytoskeleton imaging in Physcomitrella

153

fluorescent markers, experiment duration, photobleaching, etc. For example, while imaging microtubule dynamics in protonema cells expressing mCherry-tubulin, we typically used 561 nm laser with 2 mW power and 800 ms exposure time. 4. In order to simultaneously acquire images from two fluorophores with different emission wavelengths (e.g., mNeonGreen and mCherry; see Fig. 3a), we used the GEMINI split view module with enabled W-VIEW mode. 5. Time-lapse images were usually acquired every 3 s for the short-term imaging or every 5 min for long-term imaging of the reprogramming of gametophore cells (see Fig. 3b). 6. In some cases, enabling the autofocus proved to be very helpful, especially for long-term imaging. 3.5 Performing Washout Assays During Imaging Using the Microdevice

One of the main advantages of a microfluidic device is the precise control over the liquid that fills the channels. It can be used to test the impact of change in the environment, for example, adding various chemical compounds, on various cellular processes. In the previous studies, inhibitors were supplied directly to the liveimaging dish. However, this can often shift the cells, and complete removal of inhibitors from the dish is not feasible. In addition, each experiment requires a separate imaging dish [12, 13]. These issues have been remedied in our microdevice. A single device consists of four independent channels; thus, four washout experiments can be performed using a single device. Furthermore, since channel depth is slightly smaller than average cell size of Physcomitrella, cells are effectively trapped inside the channel and do not shift during liquid perfusion. Here, we describe how to introduce and wash out a microtubule-depolymerizing drug oryzalin to observe microtubule shrinkage and re-growth in protonema cells (see Fig. 4). 1. A piece of tubing long enough to reach from microscope stage to the syringe pump is connected to one of the outlet holes in the microdevice (see Fig. 4a). Next, a coverslip with microdevice is placed on the microscope stage, and its position is fixed with tape. 2. A syringe pump, equipped with a 1 mL syringe and a needle, is connected through the tubing (see Note 13) and is set to the “withdraw” mode and a flow rate of 10μL/min. 3. Suitable cells are selected using a lower magnification lens (e.g., 10 lens). We do not recommend using the cells that block part of the channel as this results in an uneven washout efficiency [6]. In addition, the process of connecting the tubing often damages cells close to the outlet hole, and therefore, we do not recommend imaging cells located near the outlet hole.

154

Mari W. Yoshida and Elena Kozgunova

Fig. 4 Washout assays using the microdevice. (a) Schematic diagram of the washout setup (not to scale). Outlet hole of the microfluidic channel was connected to the syringe pump with tubing. Inlet hole was covered with a drop of liquid to be introduced to the channels (e.g., microtubule-depolymerizing drug oryzalin). (b) Representative images of washing out fluorescent dye Alexa Fluor 488 at a final of concentration 0.5μM. Time stamp min: sec. Bar, 10μm. Graph reflects washout efficiency measured as changes in average fluorescent intensity (AU) in the microfluidic channels before introducing the dye (background), during dye perfusion (0.5-μ M Alexa Fluor 488) and washout. Fluorescent intensity was measured at five randomly selected areas (25 × 25 pixels) close to the protonema cells at each time point; mean intensity is plotted on the graph. Each line represents an individual experiment (n = 8). Fluorescent intensity has reliably returned to background levels after 1 min 30 sec from the start of perfusion. (c) Microtubule nucleation assay in the microdevice. Upper panel shows microtubule depolymerization after 20 μM oryzalin solution was introduced to the channel. Bottom panel shows microtubule re-growth after oryzalin washout. Liquid perfusion using syringe pump was commenced at time 0. Time stamp min: sec. Bar, 10μm. This figure was adapted from Kozgunova and Goshima [6] under a Creative Commons Attribution 4.0 International License

4. The remaining inlet/outlet holes of the microdevice are covered with small pieces of scotch tape to avoid liquid mixing and cross-contamination between the channels. 5. The inlet hole of the corresponding channel is covered with a drop of liquid (100μL) containing 20μM oryzalin in BCDAT medium. All BCDAT medium used in this experiment is passed through a 20μm filter to eliminate inorganic salt precipitates.

Microfluidic Chip for Cytoskeleton imaging in Physcomitrella

155

6. After switching to a 100× 1.49 NA lens with immersion oil, the laser angle is adjusted to image individual microtubules, and the imaging is started. 7. After 1–2 min, the pump is started. After 30–60 s (see Fig. 4b), the medium (see Note 14) inside the channel is replaced with oryzalin-containing medium, and the microtubules begin to depolymerize (see Fig. 4c). 8. After microtubule depolymerization starts, the pump is stopped, and the tubing is disconnected from the needle. 9. Wait until all visible microtubules depolymerize to stop the imaging. This typically takes 5–10 min. 10. If a small amount of oryzalin solution remains on top of the device, gently remove it with a tissue. Cover the inlet hole with a drop of clean BCDAT medium. 11. Repeat steps 6–7. It is necessary to replenish the medium on top of the inlet hole occasionally, while the pump is running (see Note 15). Two to three minutes after oryzalin solution is replaced by the clean medium, microtubules start to re-polymerize (see Fig. 4c). We typically acquired images for 10–15 min.

4

Notes 1. We obtained silicon master mold for the PDMS device from our collaborator and did not prepare it ourselves, as it requires specialized skills and equipment. When properly used, a single silicon mold can be used to fabricate PDMS devices for many years, without any noticeable loss in quality. In brief, a silicon mold used in this study was prepared by spin-coating negative photoresist (SU-83010; MicroChem Corp.) on a silicon wafer. Maskless lithography system (DL-1000; Nano System Solutions, Inc.) was used to create microchannel designs in the photoresist layer. 2. A pre-polymer mixture of PDMS is a high-viscosity solution, in which many air bubbles remain trapped after mixing base with the curing agent. Degassing is an important step as air bubbles can significantly decrease the strength and transparency of the device. 3. If some air bubbles remain, they can be gently pushed toward the edges of the device to the area without microfluidic channels using a pipette tip. 4. We suggest being careful of both over-curing and under-curing PDMS. In the case of over-curing, PDMS tends to become fragile and often cracks when the holes are punched in the

156

Mari W. Yoshida and Elena Kozgunova

device. Under-cured PDMS severely affects the viability of Physcomitrella. 5. We do not recommend cutting the individual devices while PDMS is still attached to the mold. If the scalpel slips, it can easily damage the channels printed on the mold. 6. In our device design, inlet and outlet holes are interchangeable. 7. It is critical to make inlet/outlet holes before the device is attached to the glass with plasma treatment. 8. Small particles and dust significantly reduce efficiency of plasma bonding, and the device might leak if this step is omitted. 9. PDMS oxidation with air plasma changes the surface chemistry and produces silanol terminations (SiOH) on its surface. It is used to covalently bond PDMS to a plasma-oxidized glass surface by the creation of a Si-O-Si bond. 10. The quality of bonding can be tested by trying to peel the PDMS device from the glass surface. 11. This step helps prevent air bubbles in the channels. 12. It is possible to culture P. patens cells for longer periods of time in the PDMS microdevice [5]. However, it is recommended that the BCDAT growth medium be replaced weekly. 13. The length of the tubing affects the rate of pressure built up by the pump and, subsequently, how fast compounds are introduced into the channel. To accurately analyze time-sensitive cellular responses, we recommend using tubing of the same length for all experiments. 14. We suggest testing washout efficiency and how fast the medium in the channels is replaced by using a fluorescent dye, e.g., Alexa488 (see Fig. 4b). 15. If the medium is not added on time, the channel will be filled with air and cannot be used anymore. It is possible to connect the inlet hole to the pipette tip filled with medium or to the Eppendorf tube with another piece of tubing, thus avoiding the need to add the medium manually. However, we did not test these approaches in our setup. References 1. Rensing SA, Goffinet B, Meyberg R et al (2020) The moss Physcomitrium (Physcomitrella) patens: a model organism for non-seed plants. Plant Cell 32:1361–1376. https://doi. org/10.1105/tpc.19.00828 2. Wu S-Z, Bezanilla M (2014) Myosin VIII associates with microtubule ends and together with actin plays a role in guiding plant cell

division. elife 3:1–20. https://doi.org/10. 7554/eLife.03498 3. Kozgunova E, Nishina M, Goshima G (2019) Kinetochore protein depletion underlies cytokinesis failure and somatic polyploidization in the moss Physcomitrella patens. eLife 8:1–16. https://doi.org/10.7554/eLife.43652 4. Yamada M, Goshima G (2018) The KCH kinesin drives nuclear transport and cytoskeletal

Microfluidic Chip for Cytoskeleton imaging in Physcomitrella coalescence to promote tip cell growth in Physcomitrella patens. Plant Cell 30:1496–1510. https://doi.org/10.1105/tpc.18.00038 5. Bascom CS, Wu S-Z, Nelson K et al (2016) Long-term growth of moss in microfluidic devices enables subcellular studies in development. Plant Physiol 172:28–37. https://doi. org/10.1104/pp.16.00879 6. Kozgunova E, Goshima G (2019) A versatile microfluidic device for highly inclined thin illumination microscopy in the moss Physcomitrella patens. Sci Rep 9:1–8. https://doi.org/ 10.1038/s41598-019-51624-9 7. Yanagisawa N, Kozgunova E, Grossmann G et al (2021) Microfluidics-based bioassays and imaging of plant cells. Plant Cell Physiol 62: 1239–1250. https://doi.org/10.1093/pcp/ pcab067 8. Nakaoka Y, Kimura A, Tani T, Goshima G (2015) Cytoplasmic nucleation and atypical branching nucleation generate endoplasmic microtubules in Physcomitrella patens. Plant Cell 27:228–242. https://doi.org/10.1105/ tpc.114.134817 9. Kosetsu K, Murata T, Yamada M et al (2017) Cytoplasmic MTOCs control spindle orientation for asymmetric cell division in plants. Proc Natl Acad Sci 114:E8847–E8854. https://doi. org/10.1073/pnas.1713925114 10. Spinner L, Pastuglia M, Belcram K et al (2010) The function of TONNEAU1 in moss reveals ancient mechanisms of division plane specification and cell elongation in land plants. Development 137:2733–2742. https://doi.org/10. 1242/dev.043810 11. Sato Y, Sugimoto N, Hirai T et al (2017) Cells reprogramming to stem cells inhibit the reprogramming of adjacent cells in the moss Physcomitrella patens. Sci Rep 7. https://doi.org/10. 1038/s41598-017-01786-1 12. Kozgunova E, Higashiyama T, Kurihara D (2016) Cytokinesis defect in BY-2 cells caused by ATP-competitive kinase inhibitors. Plant Signal Behav 11:e1238547. https://doi.org/ 10.1080/15592324.2016.1238547 13. Kozgunova E, Suzuki T, Ito M et al (2016) Haspin has multiple functions in the plant cell division regulatory network. PCP 57:848–861. https://doi.org/10.1093/pcp/pcw030 14. Hiwatashi Y, Obara M, Sato Y et al (2008) Kinesins are indispensable for interdigitation of phragmoplast microtubules in the moss Physcomitrella patens. Plant Cell 20:3094– 3106. https://doi.org/10.1105/tpc.108. 061705 15. Miki T, Naito H, Nishina M, Goshima G (2014) Endogenous localizome identifies

157

43 mitotic kinesins in a plant cell. Proc Natl Acad Sci 111:1053–1061. https://doi.org/10. 1073/pnas.1311243111 16. Nakaoka Y, Miki T, Fujioka R et al (2012) An inducible RNA interference system in Physcomitrella patens reveals a dominant role of Augmin in phragmoplast microtubule generation. Plant Cell 24:1478–1493. https://doi.org/ 10.1105/tpc.112.098509 17. Kosetsu K, De Keijzer J, Janson ME, Goshima G (2013) MICROTUBULE-ASSOCIATED PROTEIN65 is essential for maintenance of phragmoplast bipolarity and formation of the cell plate in Physcomitrella patens. Plant Cell 25: 4479–4492. https://doi.org/10.1105/tpc. 113.117432 18. Leong SY, Yamada M, Yanagisawa N, Goshima G (2018) SPIRAL2 stabilises endoplasmic microtubule minus ends in the moss Physcomitrella patens. Cell Struct Funct 43:53–60. https://doi.org/10.1247/csf.18001 19. Ding X, Pervere LM, Bascom C et al (2018) Conditional genetic screen in Physcomitrella patens reveals a novel microtubule depolymerizing-end-tracking protein. PLoS Genet 14. https://doi.org/10.1371/journal. pgen.1007221 20. Finka A, Schaefer DG, Saidi Y et al (2007) In vivo visualization of F-actin structures during the development of the moss Physcomitrella patens. New Phytol 174:63–76. https://doi. org/10.1111/j.1469-8137.2007.01989.x 21. Yamashita H, Sato Y, Kanegae T et al (2011) Chloroplast actin filaments organize meshwork on the photorelocated chloroplasts in the moss Physcomitrella patens. Planta 233:357–368. https://doi.org/10.1007/s00425-0101299-2 22. Vidali L, Rounds CM, Hepler PK, Bezanilla M (2009) Lifeact-mEGFP reveals a dynamic apical F-actin network in tip growing plant cells. PLoS One 4. https://doi.org/10.1371/jour nal.pone.0005744 23. Yi P, Goshima G (2020) Rho of plants GTPases and cytoskeletal elements control nuclear positioning and asymmetric cell division during Physcomitrella patens branching. Curr Biol 30: 2860–2868.e3. https://doi.org/10.1016/j. cub.2020.05.022 24. Vidali L, Van Gisbergen PAC, Gue´rin C et al (2009) Rapid formin-mediated actin-filament elongation is essential for polarized plant cell growth. Proc Natl Acad Sci U S A 106:13341– 13346. https://doi.org/10.1073/pnas. 0901170106 25. Vidali L, Burkart GM, Augustine RC et al (2010) Myosin XI is essential for tip growth

158

Mari W. Yoshida and Elena Kozgunova

in Physcomitrella patens. Plant Cell 22:1868– 1882. https://doi.org/10.1105/tpc.109. 073288 26. Augustine RC, Pattavina KA, Tu¨zel E et al (2011) Actin interacting protein1 and actin depolymerizing factor drive rapid actin dynamics in Physcomitrella patens. Plant Cell 23: 3696–3710. https://doi.org/10.1105/tpc. 111.090753 27. Perroud PF, Quatrano RS (2006) The role of ARPC4 in tip growth and alignment of the polar axis in filaments of Physcomitrella patens. Cell Motil Cytoskeleton 63:162–171. https:// doi.org/10.1002/cm.20114 28. Perroud PF, Quatrano RS (2008) BRICK1 is required for apical cell growth in filaments of the moss Physcomitrella patens but not for gametophore morphology. Plant Cell 20:411– 422. https://doi.org/10.1105/tpc.107. 053256 29. Orr RG, Furt F, Warner EL et al (2021) Rab-E and its interaction with myosin XI are essential

for polarised cell growth. New Phytol 229: 1924–1936. https://doi.org/10.1111/nph. 17023 30. Furt F, Lemoi K, Tu¨zel E, Vidali L (2012) Quantitative analysis of organelle distribution and dynamics in Physcomitrella patens protonemal cells. BMC Plant Biol 12. https://doi.org/ 10.1186/1471-2229-12-70 31. Uchida M, Ohtani S, Ichinose M et al (2011) The PPR-DYW proteins are required for RNA editing of rps14, cox1 and nad5 transcripts in Physcomitrella patens mitochondria. FEBS Lett 585:2367–2371. https://doi.org/10.1016/j. febslet.2011.06.009 32. Tran ML, McCarthy TW, Sun H et al (2018) Direct observation of the effects of cellulose synthesis inhibitors using live cell imaging of Cellulose Synthase (CESA) in Physcomitrella patens. Sci Rep 8:1–9. https://doi.org/10. 1038/s41598-017-18994-4

Chapter 12 Using Spinning Disk Microscopy to Observe the Mitotic and Cytokinetic Apparatus in Physcomitrium patens Yuji Hiwatashi and Takashi Murata Abstract Protonemata of the moss Physcomitrium patens are ideal structures in which to observe cytoskeletal organization and dynamics. Special care is needed to prepare P. patens cultures for high-resolution microscopy. Here, we describe methods for spinning disk microscopy of dividing P. patens cells expressing sGFP-tubulin and H2B-mCherry, including detailed methods for culturing P. patens. Key words Physcomitrium patens, Protonemal cells, Mitosis, Cytokinesis, Microtubules, Live imaging

1

Introduction Knowledge of the spatiotemporal localization of cytoskeletal components and their regulators is essential for understanding the mechanisms of cytoskeletal organization. Fluorescently tagged proteins are widely used to localize cytoskeletal components and regulators. Fluorescently tagged proteins are usually introduced into plants such as Arabidopsis thaliana by integrating a genomic DNA sequence fused with a fluorescent protein sequence and expressing this sequence under the control of a native promoter. However, complementation of a null mutant is required to confirm that the tagged protein has the same function as the untagged protein. By contrast, inserting a DNA sequence encoding a desired fluorescent protein into a genomic locus is an ideal way to perform fluorescent tagging. The expression level of the tagged protein is expected to be similar to that in untransformed cells, and the function of the fusion protein is easily assessed by examining the phenotype of plants containing the DNA insert. The moss Physcomitrium patens is an ideal material in which to insert fluorescent protein sequences because homologous recombination methods have been established and a full genome sequence is available for this species [1].

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_12, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

159

160

Yuji Hiwatashi and Takashi Murata

However, homologous recombination methods are not available to study cytoskeletal components and regulators in higher plant cells. Localization studies also require the materials to be suitable for high-resolution microscopy without the need for sectioning. Because plant cells are composed of materials with different optical properties [2], it is almost impossible to successfully focus light through multi-layered plant tissues. Therefore, materials for high-resolution microscopy of living cells must have a singlelayered organization. P. patens protonemata fulfill this requirement, as they are composed of linearly attached filamentous cells [3]. Following germination from spores, P. patens forms a filamentous structure called the chloronemata. Chloronemal cells contain many large chloroplasts. An apical cell is located at the tip of a chloronemal filament functions as a stem cell and divides continuously. Chloronemal apical cells gradually differentiate into caulonemal apical cells, which give rise to another filamentous structure. Caulonemal cells contain a few smaller chloroplasts and a more oblique septum than those formed in chloronemata (Fig. 1a). Chloronemata and caulonemata together are referred to as protonemata. Caulonemal apical cells grow more rapidly than chloronemal apical cells; hence, the miotic and cytokinetic apparatus is more easily observed in caulonemal apical cells (see Note 1). Spinning disk microscopy using a multiple pinhole array disk is a good choice for observing the cellular dynamics of protonemata because it is superior to conventional point-scanning confocal microscopy due to its rapid imaging and low phototoxicity [4, 5]. Rapid scanning using a conventional single-point confocal microscope with a rapid-scanning mirror device can cause photobleaching of fluorescent proteins because a stronger laser is focused at the scanned point for a shorter time during image acquisition. In our practice using recent high-performance single-point scanning confocal microscopes with high-sensitivity detectors, we still encounter problems with photobleaching and phototoxicity when observing weakly expressed fluorescently tagged protein under the control of native promoters. Therefore, spinning disk microscopy is the best method for live imaging of cytoskeletal dynamics, and the effects of cytoskeletal regulators on these dynamics, in P. patens protonemata. Indeed, many reports describe the application of spinning disk microscopy in P. patens to study the plant cytoskeleton [6–8]. A disadvantage of spinning disk microscopy is cross-talk among individual pinholes, which reduces the contrast of images via contamination with blurred signals from unfocused planes. However, this cross-talk is practically negligible when using thin tissues. As with other plant materials used for cell biology, the use of P. patens protonema for microscopy requires optimized instrument settings (hardware) and culture conditions. Moss protonemata have well-developed chloroplasts containing chlorophyll.

Spinning Disk Microscopy in Physcomitrium Patens

161

a chl

sep

b

500-510

chl

chl

nuc

510-520

520-530

530-540

phr 540-550

550-560

560-570

570-580

580-590

590-600

600-610

610-620

nuc 620-630

630-640

640-650

650-660

4000

chl

0

Fig. 1 Morphology and fluorescence of Physcomitrium patens line YT195 chloronemata. (a) A caulonema of line YT195 grown in a glass-bottom dish. sep, septum; chl, chloroplast; nuc, position of the nucleus. Bar ¼ 50 μm. (b) Chlorophyll autofluorescence of line YT195 at different wavelengths. A caulonema was analyzed by spectral imaging using an Olympus FV3000 confocal microscope. The caulonema with a phragmoplast was excited by a 488 nm laser, and fluorescence in a 10 nm window was measured from 500 to 660 nm. Pixel intensities are shown in different colors (color bar on the right). The phragmoplast (phr) was brightest at 510–520 nm, and chlorophyll fluorescence (chl) increased at 640 nm. Note that mCherry was excited by 488 nm light, and nuclear fluorescence (nuc) was observed at ~600 nm. Bar ¼ 10 μm

Chlorophyll autofluorescence often masks weak signals from fluorescent proteins if the filter settings are inadequate. Fluorescence microscopy of P. patens protonemata therefore requires careful consideration of the filters used (see Note 2). In addition, careful

162

Yuji Hiwatashi and Takashi Murata

pre-culture of plants for microscopy is required to obtain normal, regular growth and cell division and for the best performance of optical systems. Here, we describe methods for spinning disk microscopy, with a special focus on culturing methods.

2

Materials

2.1

Plants

Transgenic Physcomitrium patens line YT195 expressing both sGFP-α-tubulin and histone H2B-mCherry is used in this study (see Note 3). This line was obtained from Yuji Hiwatashi’s laboratory, Miyagi University, Japan (https://www.myu.ac.jp/teacher/ food/hiwatash/labo/).

2.2

Culture Media

BCDATG and BCDAT solid media are used for the routine subculture of protonemata. BCDAT medium contains ammonium tartrate as a nitrogen source to promote the propagation of protonemata. BCDATG medium is BCDAT medium supplemented with glucose as a carbon source to enable easy detection of bacterial and fungal contamination (see Note 4). BCD medium is usually supplemented with 1 mM CaCl2 medium (referred to as BCD + 1 mM Ca; see Note 5). This medium is considerably better for observing caulonemal growth and bud formation than BCDATG or BCDAT medium because BCD medium promotes caulonemal differentiation.

2.2.1 Stock Solutions for Culture Medium

Ultrapure water should be used to prepare all media. Sterilization by autoclaving should be performed for 20 min at 120  C. Store all stock solutions at 4  C. 1. Stock B (100): Dissolve 25 g (0.101 M) of MgSO47H2O in 1000 mL of water. The stock solution should be autoclaved. 2. Stock C (100): Dissolve 25 g (0.184 M) of KH2PO4 in 900 mL of water, and adjust to pH 6.5 with 4 M KOH. Add water to a final volume of 1000 mL, and autoclave. 3. Stock D (100): Dissolve 1.25 g (4.5 mM) of FeSO47H2O and 101 g (1.0 M) of KNO3 in 1000 mL of water. This solution must not be autoclaved. Yellow crystals may precipitate during storage at low temperatures; this precipitated solution can be used after mixing well. 4. Alternative TES (1000): Dissolve 55 mg (0.22 mM) of CuSO45H2O, 614 mg (10 mM) of H3BO3, 55 mg (0.23 mM) of CoCl26H2O, 25 mg (0.1 mM) of Na2MoO42H2O, 55 mg (0.19 mM) of ZnSO47H2O, 389 mg 2.0 mM of MnCl24H2O, and 28 mg (0.17 mM) of KI in 1000 mL of H2O. This stock solution should be autoclaved.

Spinning Disk Microscopy in Physcomitrium Patens

163

5. 500 mM ammonium tartrate (100): Dissolve 92.05 g (500 mM) of ammonium tartrate ([CH(OH)COONH4]2) in 1000 mL of H2O. This stock solution should be autoclaved. 6. 50 mM CaCl2 (50): Dissolve 7.35 g (50 mM) of CaCl22H2O in 1000 mL of H2O. This stock solution should be autoclaved. 2.2.2 Preparation of Culture Media

The media should be autoclaved at 120  C for 20 min. For routine culture, after autoclaving, pour solid medium into 9 cm plastic Petri dishes (20 mL/dish), and allow it to solidify. Uncover the Petri dishes, and dry them for 30 min in a laminar flow hood. The Petri dishes can be stored at room temperature. 1. BCD + 1 mM Ca medium: BCD medium supplemented with 1 mM CaCl2. Add 10 mL of stock B, 10 mL of stock C, 10 mL of stock D, 1 mL of alternative TES, and 20 mL of 50 mM CaCl2 to 900 mL of water, and fill to 1000 mL with water. For solid medium, add 8 g of agar per 1000 mL of medium. 2. BCDAT medium [9]: Add 10 mL of stock B, 10 mL of stock C, 10 mL of stock D, 10 mL of 500 mM ammonium tartrate, 1 mL of alternative TES, and 20 mL of 50 mM CaCl2 to 900 mL of water, and fill to 1000 mL with water. For solid medium, add 8 g of agar per 1000 mL of medium. 3. BCDATG medium [9]: BCDAT medium supplemented with 5 g/L glucose.

2.2.3 Reagents and Instruments for Routine Subculture

1. Sterile water. 2. Autoclaved glass test tube (30  120 mm). 3. Forceps. 4. Homogenizer (e.g., Polytron PT1200 and PT-DA 07/2SYNE082). 5. Cellophane discs (50). A non-parametric test is required because the data is not normally distributed. We utilize the Wilcoxon test. For a more detailed discussion and a bootstrap dependent statistical test, see [20].

Acknowledgments Z.B. is funded by the UK Biotechnology and Biological Sciences Research Council EASTBIO Doctoral Training Partnership. The Tilsner lab receives funding from the Scottish Government Rural and Environment Science and Analytical Services (RESAS). References 1. Tilsner J, Nicolas W, Rosado A, Bayer EM (2015) Staying tight: plasmodesmal membrane contact sites and the control of cell-to-cell connectivity in plants. Annu Rev Plant Biol 67: 337–364. https://doi.org/10.1146/annurevarplant-043015-111840 2. Brault ML, Petit JD, Immel F et al (2019) Multiple C2 domains and transmembrane region proteins ( MCTP s) tether membranes at plasmodesmata. EMBO Rep 20:e47182. https://doi.org/10.15252/embr.201847182 3. Roberts AG, Oparka KJ (2003) Plasmodesmata and the control of symplastic transport. Plant Cell Environ 26:103–124. https://doi. org/10.1046/j.1365-3040.2003.00950.x 4. Tilsner J, Amari K, Torrance L (2011) Plasmodesmata viewed as specialised membrane

adhesion sites. Protoplasma 248:39–60. https://doi.org/10.1007/s00709-0100217-6 5. White RG, Barton DA (2011) The cytoskeleton in plasmodesmata: A role in intercellular transport? J Exp Bot 62:5249–5266. https:// doi.org/10.1093/jxb/err227 6. Diao M, Huang S (2021) An update on the role of the actin cytoskeleton in plasmodesmata: a focus on formins. Front Plant Sci 12: 191. https://doi.org/10.3389/FPLS.2021. 647123/BIBTEX 7. Radford JE, White RG (2011) Inhibitors of myosin, but not actin, alter transport through Tradescantia plasmodesmata. Protoplasma 248:205–216. https://doi.org/10.1007/ s00709-010-0244-3

202

Zoe Barr and Jens Tilsner

8. Diao M, Ren S, Wang Q et al (2018) Arabidopsis formin 2 regulates cell-to-cell trafficking by capping and stabilizing actin filaments at plasmodesmata. elife 7:e36316. https://doi. org/10.7554/eLife.36316 9. White RG, Badelt K, Overall RL, Vesk M (1994) Actin associated with plasmodesmata. Protoplasma 180:169–184. https://doi.org/ 10.1007/BF01507853 10. Radford JE, White RG (1998) Localization of a myosin-like protein to plasmodesmata 11. Overall RL, Blackman LM (1996) A model of the macromolecular structure of plasmodesmata. Trends Plant Sci 1:307–311. https:// doi.org/10.1016/1360-1385(96)88177-5 12. Nicolas WJ, Grison MS, Tre´pout S et al (2017) Architecture and permeability of postcytokinesis plasmodesmata lacking cytoplasmic sleeves. Nat Plants 3:17082. https://doi.org/ 10.1038/nplants.2017.82 13. Cui W, Wang X, Lee JY (2015) Drop-ANd-see: a simple, real-time, and noninvasive technique for assaying plasmodesmal permeability. Methods Mol Biol 1217:149–156. https://doi.org/ 10.1007/978-1-4939-1523-1_10 14. Liesche J, Schulz A (2015) Quantification of Plant Cell Coupling with Live-Cell Microscopy. Methods Mol Biol 1217:137–148. https://doi.org/10.1007/978-1-49391523-1_9 15. Ohtsu M, Jennings J, Johnston M et al (2021) Colletotrichum higginsianum effectors exhibit

cell to cell hypermobility in plant tissues and modulate intercellular connectivity amongst a variety of cellular processes 2 3. bioRxiv 426415. https://doi.org/10.1101/2021.01. 13.426415 16. Gerlitz N, Gerum R, Sauer N, Stadler R (2018) Photoinducible DRONPA-s: a new tool for investigating cell-cell connectivity. Plant J 94: 751–766. https://doi.org/10.1111/tpj. 13918 17. Oparka KJ, Roberts AG, Boevink P et al (1999) Simple, but not branched, plasmodesmata allow the nonspecific trafficking of proteins in developing tobacco leaves. Cell 97:743–754. https://doi.org/10.1016/S0092-8674(00) 80786-2 18. Gal-On A, Meiri E, Elman C et al (1996) Simple hand-held devices for the efficient infection of plants with viral-encoding contructs by particle bombardment. J Virol Methods 64:103– 110. https://doi.org/10.1016/S0166-0934 (96)02146-5 19. Deinum EE, Mulder BM, Benitez-Alfonso Y (2019) From plasmodesma geometry to effective symplasmic permeability through biophysical modelling. elife 8:e49000. https://doi. org/10.7554/ELIFE.49000 20. Johnston MG, Faulkner C (2021) A bootstrap approach is a superior statistical method for the comparison of non-normal data with differing variances. New Phytol 230:23–26. https://doi. org/10.1111/nph.17159

Chapter 15 Studying Nuclear Dynamics in Response to Actin Disruption in Planta Joseph F. McKenna and Katja Graumann Abstract The plant nucleus and the actin cytoskeleton are intimately connected. The actin cytoskeleton is pivotal for nuclear positioning, shape, and dynamics. These properties of the nucleus are important for its functions during normal development and in response to external cues such as biotic and abiotic stresses. Moreover, we know that there is a direct physical connection between the actin cytoskeleton and the nucleus which spans the double-membraned nuclear envelope into the nuclear lamina, and this connection is called the linker of nucleoskeleton and cytoskeleton (LINC) complex. Recently a role for actin in regulating internuclear organization via the control of nuclear invaginations has emerged. Therefore, a detailed understanding of nuclear shape, organization, and dynamics and the techniques used to measure and quantify these metrics will allow us to determine and further understand the contribution made by actin to these parameters. The protocols described here will allow researchers to determine the circularity index of a nucleus, quantify nuclear deformations, and determine dynamics of nuclei within plant cells. Key words Actin, Nucleus, Klarsicht ANC-1, Syne homology (KASH), Linker of nucleoskeleton and cytoskeleton (LINC), Confocal microscope, Circularity index, Kymographs

1

Introduction The nucleus is a highly complex compartment whose functions depend on its ordered and dynamic structure. It is highly malleable, being able to change shape and volume, and is highly mobile. These key characteristics of the nucleus are essential to its functions. In plant cells, nuclear dynamics govern a myriad of processes linked to development, growth, fertility, stress responses, and immunity. For example, mechanical stimuli trigger nuclei to move to the stimulation site [1] and reorganization of the actin cytoskeleton [2]. Mechanical pressure can be exerted by neighboring cells with the resulting change in nuclear position determining the cell division plane [1]. This in turn affects tissue organization and function [3]. Mechanical stimuli are also involved in pathogen interactions, where the nuclei move to the wound or infection site [4]. This is

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_15, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

203

204

Joseph F. McKenna and Katja Graumann

thought to elicit a more efficient response with the nucleus as the organizing center for pathogen interactions [1, 4]. Long-range nuclear migrations are associated with tip growth in both pollen tubes and root hairs [4]. Nuclear movement is often contemporaneous with changes in nuclear shape https://www.zotero.org/ google-docs/?iixX9T [4, 5], which in turn, has been linked to changes in chromatin organization and regulation of gene expression [6]. The nucleus is surrounded by a dual-membrane system, the nuclear envelope (NE), which encases the genetic material and the regulatory and structural protein components associated with it. Composed of an outer and inner nuclear membrane (ONM and INM), the NE is emerging as the key regulator of nuclear shape, positioning, and movement [7–11]. Specifically, an evolutionarily conserved bridging complex called linker of nucleoskeleton and cytoskeleton (LINC) mediates these processes. In the LINC complex, cytoskeletal elements including actin associate with ONM intrinsic Klarsicht/Anc-1/Syne-1 homology (KASH) proteins, which in turn interact with INM-localized Sad1/UNc84 (SUN) domain proteins. These associate with nucleoskeletal proteins and chromatin in the nuclear interior completing the bridging [7–9, 12]. Mutations in any LINC component alter nuclear shape and movement highlighting their importance [7–9, 12]. LINC complex components are evolutionary conserved in plants and have been studied in various models including Arabidopsis, maize, rice and Medicago. At the ONM, plant-specific KASH proteins called SINEs, WIPs, and WITs anchor the actin cytoskeleton [7–9]. SINE1 directly interacts with actin, while WIP and WIT heteromers anchor myosin XI-i to the NE [7–9, 13]. The plant SUN proteins resemble other eukaryotic SUN proteins and contain a conserved SUN domain, which mediates the binding to KASH proteins in the periplasmic space. Two of the SUN proteins, AtSUN1 and AtSUN2, have been shown to interact with plantspecific nucleoskeletal components—CRoWded Nuclei 1 (CRWN1, directly) and KAKU4 (indirectly), thereby completing the linkage across the NE to the actin cytoskeleton [12, 14– 16]. Proteins associated with the microtubule cytoskeleton are also present at the ONM and are involved in altering nuclear shape [17]. Evidence is building that the plant NE, and plant LINC, complexes are associated with perinuclear actin governing nuclear dynamics at the molecular level. In plants, actin is positioned in nuclear grooves [18], and perinuclear actin surrounds the nucleus [19]. Pollen tube and root hair growth are dependent on actin. Nuclear shape and movement are affected when actin is perturbed resulting in cessation of tip growth. In addition to actin, WIP and WIT proteins as well as myosin XI-i are required for correct nuclear movement and shape [4]. In leaf epidermal cells, this WIP-WIT-

Studying Nuclear Dynamics in Response to Actin Disruption in Planta

205

myosin-actin linkage to the NE mediates blue light-induced nuclear movement [4, 13]. In guard cells the nucleus is anchored directly to the actin cytoskeleton via SINE1, which contributes to the correct positioning of the nucleus and plays a role in stomatal opening [20] and in plant immunity against oomycete infection [8]. The maize SINE1 homologue MLKS2 also interacts with actin and anchors nuclei in position when transiently expressed in N. benthamiana [11]. Maize knockout lines of MLKS2 have defects in root hair morphology, stomatal complex development, meiosis, and pollen viability and show a reduction in perinuclear actin [11]. The fact that NE-associated actin and LINC components affect nuclear shape, which is linked to changes in chromatin organization and gene expression, is exemplified in mutants of plant LINC components including CRWN1 which derepresses transcriptionally silenced heterochromatin [6]. Recently, it was shown that overexpression of the maize MKAKU41 nuclear lamina protein induces nuclear invaginations. Actin depolymerization increased MKAKU41-induced nuclear changes, therefore demonstrating that the actin cytoskeleton has a role in regulating nuclear deformations [16]. In conclusion, we have an understanding of how links with the actin cytoskeleton regulate nuclear membranes. In this chapter, we detail some common nuclear imaging analysis techniques which allow researchers to determine how nuclear shape, invaginations, and dynamics are altered under different conditions.

2 2.1

Materials Equipment

1. A fluorescent widefield microscope can be used for determining circularity index with appropriate filters (see Note 1). 2. A confocal laser scanning microscope is best used for quantifying nuclear deformations or generating kymographs.

2.2 Microscopy Consumables

1. Gilsen pipette and appropriate tip. 2. Surgical tape. 3. Microscope glass slides. 4. No. 1.5 coverslips.

2.3 Image Analysis Tools

1. FIJI-equipped image J (v1.53n).

2.4

1. 1-cm2 leaf sections of 6-week old N. benthamiana plants grown in a 16:8 light dark cycle and transiently expressing fluorescent markers as described in [21].

Plant Samples

2. Microsoft Excel.

206

Joseph F. McKenna and Katja Graumann

2. Whole mount 5-day-old Arabidopsis thaliana seedlings germinated and grown on ½ strength MS plates with MES in 16:8long-day conditions. 2.5 Chemicals and Constructs

1. Latrunculin-B (Lat-B): working concentration at 25 μM. 2. DAPI staining buffer: for 1 mL, add 3 μL of 20% Triton X-100 and 1 μL of 50 mg/mL DAPI stain. 3. 35S:LBR-GFP: a nuclear envelope marker [14]. 4. MKAKU41: a nuclear lamin protein [16].

3

Methods

3.1 Mounting of Samples for Imaging

1. For DAPI staining, incubate the plant sample (e.g., seedling or tissue section) for 15 min in DAPI staining buffer (see Note 2). 2. Mount leaf or seedling sample, used for all imaging, by placing an 80 μl drop of water on a microscope slide using a Gilsen pipette. 3. Place the plant sample on top using forceps or a razor blade. 4. Using the Gilsen, place a drop of water in the middle of a no. 1.5 coverslip, and lower this onto the sample. 5. Use 3 M micropore surgical tape to wrap around each side of the coverslip to ensure the coverslip stays in place and that movement is restricted.

3.2 Analysis of Circularity Index (CI) to Determine Nuclear Shape

1. Export z-stack images as tiff files creating one folder for each z-stack (see Note 3). 2. Open FIJI. 3. Under “File,” select “Import,” and select “Image Sequence.” 4. Open the first image in the series, and the program will automatically open the entire stack if the images have been numbered in sequence. 5. Select “Image,” select “Stacks,” and select “3D project” to generate a flattened 3D projection of the z-stack. Save this as a tiff file. 6. Using the line drawing tool, draw a straight line (press shift key) the length of the size bar. 7. Select “Analyze,” select “Set Scale,” and enter the know distance of the scale bar. Enter “um” as unit of length to represent micrometer. Tick the global box if this scale is applicable to following projections. 8. Using the line drawing tool, draw a straight line (press shift key) along the widest section of the nucleus (Fig. 1a).

Studying Nuclear Dynamics in Response to Actin Disruption in Planta

207

Fig. 1 Using ImageJ to quantify nuclear shape. (a) DAPI-stained Arabidopsis root hair nuclei with length (yellow line) and width (red line) indicated. Size bar, 10 μm. (b) Example FIJI output of width and length measurements of nuclei in (a)

9. Select “Analyze,” select “Measure” and the value of the width will be displayed in an output window (Fig. 1b). 10. Using the line drawing tool, draw a straight line (press shift key) along the longest section of the nucleus (Fig. 1a). 11. Select “Analyze,” select “Measure,” and the value of the length will be displayed in an output window (Fig. 1b). 12. Save the results (see Note 4). 13. Calculate CI with the following equation: CI=width/length. 14. A CI value close to 1 indicates a round nucleus with 1 being a perfect sphere. The lower the CI value, the more elongate the nucleus is in shape. 3.3 Quantifying Nuclear Envelope Deformations

Perturbing the actin cytoskeleton can cause deformations where the nuclear morphology has been changed and invaginations have occurred [16]. These invaginations contain nuclear envelope markers which are normally localized to the nuclear periphery. A useful analysis is to determine the rate of this nuclear invagination. 1. For actin perturbation, actin can be depolymerized with 25 μM Lat-B incubation for 1 h [22]. Alternatively, the overexpression of proteins which are known or thought to localize to the nuclear envelope and also interact with the actin cytoskeleton, such as MKAKU41 [8, 11], can be used. 2. To determine the amount of nuclear deformation, visualize a nuclear envelope marker on a confocal microscope (OD600 of 0.1 at 3 days post-infiltration for LBR-GFP). Use appropriate excitation wavelength, and collect appropriate emission wavelength. If imaging two constructs, use sequential line scanning to ensure fluorescence blead through is avoided.

208

Joseph F. McKenna and Katja Graumann

3. Digitally zoom into the nucleus, and focus at its widest point so that the nuclear membrane is in clear focus. Collect a single image of one nucleus, and move onto the next. In order to ensure statistical robustness, collect ten samples from each biological repeat three times, and perform three experimental repeats for a total of 90 nuclei. 4. In FIJI, draw a region of interest around the whole nuclear exterior, next to the membrane marker using the polygon tool (red box, Fig. 2a; see Note 5). This is the “Whole nucleus ROI.” 5. Save this ROI using ROI manager by selecting Analyze > Tools > ROI manager in FIJI (Fig. 2a, red boxes). 6. Rename with the filename, and save ROI so it can be identified later if required (Fig. 2b, see Note 6). 7. Now draw an ROI internal to the nuclear periphery, and save in the ROI manager. This is the sub-periphery ROI. 8. In ROI manager, with both ROIs selected, click show all (Fig. 2b, red box), both ROIs will now appear on the image file. 9. Select the image file, and measure the fluorescence in both ROIs by clicking Ctrl + M (windows) or command + M (Mac), and alternatively click measure in ROI manager. 10. Copy the results window into Excel (see Note 7). Divide the sub-periphery intensity value (RawIntDen) by the whole nucleus (see Note 8). This will result in a ratio of the sub-periphery over the whole nuclear fluorescence. 11. Repeat on all control datasets and experimental ones in which invaginations are induced and the actin cytoskeleton is perturbed. 12. Figure 2c gives examples of control and nuclear invaginationinduced nuclei with the corresponding sub-periphery/wholenuclear fluorescence ratio value. 13. Data can then be collated and statistical analysis performed to determine how perturbing the actin cytoskeleton alters nuclear invaginations. 3.4 Tracking Nuclear Movement Using Temporal Color-Coded Projections

In plants, nuclei are extremely dynamic and move in response to abiotic and biotic stresses including pathogen perception and during light stress. This movement is driven by the actin cytoskeleton as actin is known to interact with the nuclear envelope via LINC complexes. Therefore, to understand links between the actin cytoskeleton and nuclear membranes, we need to be able to determine changes in nuclear movement. Here we describe three simple ways to determine nuclear movement: temporal color-coded projections in time, kymographs, and manual particle tracking.

Studying Nuclear Dynamics in Response to Actin Disruption in Planta

209

Fig. 2 Determining sub-periphery/whole nuclear fluorescence ratio. (a) Screenshot of FIJI showing the polygon image tool used to create ROI and selection sequence for running ROI manager (red boxes). (b) Image showing example ROIs for sub-periphery and whole nucleus from a control LBR-GFP dataset and ROI manager window showing controls required to rename ROIs so they can be later identified, and both ROIs are displayed in the image at once. (c) Example data from the LBR-GFP only control and coexpression with maize MKAKU41 which is known to induce nuclear invaginations. Sub-periphery/whole nucleus fluorescence ratios for these examples are shown. Scale bar denotes 5 μm

1. Image nuclei using a standard confocal system with appropriate laser settings for excitation and emission wavelength. 2. Focus the image on the cross section of the nucleus; do not zoom in as a larger field of view to track nuclear movement from its original position is required.

210

Joseph F. McKenna and Katja Graumann

Fig. 3 Tracking nuclear movement. (a) Workflow in FIJI for generating a temporal color-coded projection of nuclear movement. (b) Line selection tool and location of line length readouts and FIJI workflow for generating a kymograph of nuclear movement. (c) FIJI workflow for manual tracking of nuclei in FIJI. (d) Example data for each of the analysis methods described above including associated kymographs and total distance moved. Scale bar denoted 5 μm. Red boxes denote important selection steps

3. Collect time course data at appropriate intervals (e.g., 2 s over 2 min; 120 s total; see Note 9). 4. In FIJI, open the data file to be analyzed, and select Analyze > Tools > Scale bar (see Note 10). 5. Select an appropriate-sized scale bar for the image, select overlay, and tick hide text. The scale bar is added at this point so on the final image it appears white. 6. In FIJI select Image > Hyperstacks and temporal color-coded projection (Fig. 3a). 7. In the popup window, the LUT can be changed if required. The start and end frame can also be altered, but ensure, between the different conditions, that the time course length (start and end frame) is always the same (Fig. 3a). 8. If this is the first sample being analyzed, tick create time color scale bar, and a color-coded scale bar will be generated along with the data image. 9. Click OK. 10. The image generated will show nuclear positions at all of the different time points in different colors (see Note 11). 3.5 Tracking Nuclear Movement Using Kymographs

1. Open the data file to be analyzed, and use the scroll bar at the bottom (Fig. 3b, red box) to scroll through the image and determine where the nucleus moves from and to. This is important as the line drawn to generate the kymograph should encompass both the start and end of movement. 2. In the FIJI control tab, click on the line selection (Fig. 3b, red box), and draw a line through the nucleus, through the path it moves and intersecting through both sides of the nuclear envelope.

Studying Nuclear Dynamics in Response to Actin Disruption in Planta

211

3. If you need to adjust the line, it can be lengthened from the end by clicking the small box at the end (Fig. 3b, red box). If you need to move the line, select the small box in the middle; click and drag (Fig. 3b, red box). 4. Make a note of the line length, and ensure this length is kept the same for all kymographs generated for datasets that are to be compared (see Note 12). 5. Generate a kymograph by selecting Analyze > Multi Kymographs > Multi kymographs. 6. Select linewidth one on popup window and click OK. 7. With the generated kymograph, select Image > Overlay > Flatten, and save the generated image. 3.6 Manual Particle Tracking of Nuclear Movement

1. Open the data file to be analyzed in FIJI (see Note 13). 2. Select Plugins > Tracking > Manual tracking (Fig. 3c, red box). 3. In the popup window, ensure the parameters are correct (time interval, x/y calibration, etc.). Data from modern confocal systems will include all of this in the file metadata, but processed or older data might need these parameters to be input. 4. Select Add track (Fig. 3c, red box). 5. In the image file, click on either the leading edge of the nucleus or the central point of the nucleus. Example data here show the leading edge. 6. The file will automatically move to the next time point. Select the same point on the nucleus. Continue until the end of the dataset. 7. When completed click End track on the manual tracking window (Fig. 3c, red box). 8. If you want to display the movement on your data file, select Overlay Dots and Lines (Fig. 3c, red box). Movie files of your data can now be generated if required showing tracked nuclear movement. 9. To track another data file, open and click Add track and continue. The results window will classify each tracked object separately so when they become output they can be easily identified as track no. 1, 2, etc. 10. When tracking has finished, copy the results window into Microsoft Excel, and determine the sum of the distance moved across time points for each tracked object (see Note 14). Remember not to include the first point as this always shows up as -1.

212

Joseph F. McKenna and Katja Graumann

11. The collected data can now be used for statistical analysis and graph generation if required. 12. Examples of all three of these nuclear movement analyses are shown in Fig. 3d.

4

Notes 1. Nuclei can be observed without the optical sectioning ability of a confocal microscope, and some measurements such as the circularity index or particle tracking can be performed in this way. However, for other methods the improved signal-to-noise ratio of a confocal system is required. 2. DNA stains such as DAPI or SYBR green can be used. Alternatively, fluorescent protein markers that are expressed and that localize to the nucleus, for example, fluorescent protein fusions with a nuclear localization signal (NLS), can also be used. 3. Ensure that scale/size bars are added to the images before export. 4. Many measurements as required can be added to the output file, including from different images. The user must keep track of which measurements represent which nuclei, width and length. Saved result files can be opened in Excel. 5. Free-hand or other methods of ROI generation can be used; we just found the polygon tool the easiest. 6. We save the ROI files in the same folder as the raw image data in a sub-folder. Files are named with the identical name to the image file with the exception of the ROI label. 7. Instead of copying the results window, it can also be saved as a . csv file and opened in Excel. 8. If RawIntDen column does not show, in FIJI click Analyze > Set measurements, and ensure integrated density is ticked. 9. Always check the minimum time period required for imaging your sample, and adjust your imaging conditions accordingly. It is best to keep image acquisition frames down to a minimum, and set a delay in imaging to ensure your sample is not damaged. 10. If you do not desire a scale bar on your temporal color-coded projections, then steps 1 and 2 can be omitted. 11. The image generated will not have the file metadata, and hence it is much easier to add a scale bar pre-temporal color-coded projection file creation. 12. Lines used for kymograph generation can be saved using ROI manager as described above.

Studying Nuclear Dynamics in Response to Actin Disruption in Planta

213

13. Manual tracking was the easiest available to quantify the data generated. However, if a nucleoplasm marker such as Hoechst or SV40-GFP is used, trackmate and automatic tracking could be attempted. 14. The results window can also be saved as a .csv file and opened in Excel.

Acknowledgments We would like to acknowledge the bioimaging unit at Oxford Brookes University for access to the confocal microscope systems used for generation of the example images presented. We would also like to express our sincere thanks to Professor David Evans for his unwavering support to both authors over the years and his passionate and infectious dedication to understanding the plant nucleus. References 1. Qu L-H, Sun M-X (2007) The plant cell nucleus is constantly alert and highly sensitive to repetitive local mechanical stimulations. Plant Cell Rep 26:1187–1193. https://doi. org/10.1007/s00299-007-0343-6 2. Hardham AR, Takemoto D, White RG (2008) Rapid and dynamic subcellular reorganization following mechanical stimulation of Arabidopsis epidermal cells mimics responses to fungal and oomycete attack. BMC Plant Biol 8:63. https://doi.org/10.1186/1471-2229-8-63 3. Smith LG (2001) Plant cell division: building walls in the right places. Nat Rev Mol Cell Biol 2 : 3 3 – 3 9 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 8 / 35048050 4. Griffis AHN, Groves NR, Zhou X, Meier I (2014) Nuclei in motion: movement and positioning of plant nuclei in development, signaling, symbiosis, and disease. Front Plant Sci 5. https://doi.org/10.3389/fpls.2014.00129 5. Chytilova E, Macas J, Sliwinska E et al (2000) Nuclear dynamics in Arabidopsis thaliana□V. Mol Biol Cell 11:9 6. Poulet A, Duc C, Voisin M et al (2017) The LINC complex contributes to heterochromatin organisation and transcriptional gene

silencing in plants. J Cell Sci 130:590–601. https://doi.org/10.1242/jcs.194712 7. Zhou X, Graumann K, Evans DE, Meier I (2012) Novel plant SUN–KASH bridges are involved in RanGAP anchoring and nuclear shape determination. J Cell Biol 196:203– 2 1 1 . h t t p s : // d o i . o r g / 1 0 . 1 0 8 3 / j c b . 201108098 8. Zhou X, Graumann K, Wirthmueller L et al (2014) Identification of unique SUN-interacting nuclear envelope proteins with diverse functions in plants. J Cell Biol 205:677–692. https://doi.org/10.1083/jcb. 201401138 9. Graumann K, Vanrobays E, Tutois S et al (2014) Characterization of two distinct subfamilies of SUN-domain proteins in Arabidopsis and their interactions with the novel KASHdomain protein AtTIK. J Exp Bot 65:6499– 6512. https://doi.org/10.1093/jxb/eru368 10. Gumber HK, McKenna JF, Estrada AL et al (2019) Identification and characterization of genes encoding the nuclear envelope LINC complex in the monocot species Zea mays. J Cell Sci 132:jcs221390. https://doi.org/10. 1242/jcs.221390

214

Joseph F. McKenna and Katja Graumann

11. Gumber HK, McKenna JF, Tolmie AF et al (2019) MLKS2 is an ARM domain and Factin-associated KASH protein that functions in stomatal complex development and meiotic chromosome segregation. Nucleus 10:144– 166. https://doi.org/10.1080/19491034. 2019.1629795 12. Graumann K (2014) Evidence for LINC1SUN associations at the plant nuclear periphery. PLoS One 9:e93406. https://doi.org/10. 1371/journal.pone.0093406 13. Tamura K, Iwabuchi K, Fukao Y et al (2013) Myosin XI-i links the nuclear membrane to the cytoskeleton to control nuclear movement and shape in Arabidopsis. Curr Biol 23:1776– 1781. https://doi.org/10.1016/j.cub.2013. 07.035 14. Graumann K, Runions J, Evans DE (2010) Characterization of SUN-domain proteins at the higher plant nuclear envelope. Plant J 61: 134–144. https://doi.org/10.1111/j. 1365-313X.2009.04038.x 15. Goto C, Tamura K, Fukao Y et al (2014) The novel nuclear envelope protein KAKU4 modulates nuclear morphology in Arabidopsis. Plant Cell 26:2143–2155. https://doi.org/ 10.1105/tpc.113.122168 16. McKenna JF, Gumber HK, Turpin ZM et al (2021) Maize (Zea mays L.) Nucleoskeletal proteins regulate nuclear envelope remodeling and function in stomatal complex development and pollen viability. Front. Plant Sci 12:

645218. https://doi.org/10.3389/fpls.2021. 645218 17. Goswami R, Asnacios A, Milani P et al (2020) Mechanical shielding in plant nuclei. Curr Biol S0960982220304292. https://doi.org/10. 1016/j.cub.2020.03.059 18. Collings DA, Carter CN, Rink JC et al (2000) Plant nuclei can contain extensive grooves and invaginations, Plant Cell:16. https://doi.org/ 10.1105/tpc.12.12.2425 19. Traas JA, Doonan JH, Rawlins DJ et al (1987) An actin network is present in the cytoplasm throughout the cell cycle of carrot cells and associates with the dividing nucleus. J Cell Biol 105:387–395. https://doi.org/10. 1083/jcb.105.1.387 20. Biel A, Moser M, Meier I (2020) A role for plant KASH proteins in regulating stomatal dynamics. Plant Physiol 182:1100–1113. https://doi.org/10.1104/pp.19.01010 21. Sparkes IA, Runions J, Kearns A, Hawes C (2006) Rapid, transient expression of fluorescent fusion proteins in tobacco plants and generation of stably transformed plants. Nat Protoc 1:2019–2025. https://doi.org/10. 1038/nprot.2006.286 22. McKenna JF, Rolfe DJ, Webb SED et al (2019) The cell wall regulates dynamics and size of plasma-membrane nanodomains in Arabidopsis. Proc Natl Acad Sci 116:12857–12862. https://doi.org/10.1073/pnas.1819077116

Chapter 16 Cytoskeleton Remodeling in Arabidopsis Stigmatic Cells Following Pollination Lucie Riglet and Isabelle Fobis-Loisy Abstract In plants, the first interaction that occurs between the male gametophytes (pollen grains) and the stigmatic epidermis of the female organ is crucial for successful reproduction. The stigma consists of a dome of flaskshaped cells specialized in pollen capture. In these stigmatic cells, the cytoskeleton network (cortical microtubules and actin microfilaments) actively responds to pollen contact and undergoes dynamic remodeling required for successful pollen acceptance to occur. Here, we have designed several microscopy mountings to monitor stigmatic cytoskeleton dynamics. These designs are based on the constraints linked to the tightly regulated pollen–stigma interaction and depend upon the experimental goal, either a static view or live-cell imaging. Key words Stigma cytoskeleton, Pollen, Confocal microscopy, Live imaging

1

Introduction The plant cytoskeleton is a filamentous network consisting of cortical microtubules (CMTs) and actin microfilaments (AMFs). This network undergoes dynamic changes in response to environmental stimuli and plays key roles in both biotic and abiotic responses [1]. During reproduction, when the pollen grains land on the stigma surface, the cytoskeleton of the stigmatic cells actively responds to pollen arrival. CMT breakdown and polymerization of AMFs occur in these cells at the pollen attachment sites, and such cytoskeleton reorganization is required for successful pollen acceptance [2–4]. Capturing these dynamic events by imaging the stigmatic cell is not trivial. In Arabidopsis, the whole stigma measures around 3 mm and consists of a three-dimensional dome composed of hundreds of flask-shaped cells (papillae, Fig. 1). Each papilla cell is covered by a thick cuticular layer that generates background autofluorescence and prevents absorption of conventional plant cell dyes and drugs [4, 5]. In addition, monitoring cytoskeleton

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_16, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

215

216

Lucie Riglet and Isabelle Fobis-Loisy

Fig. 1 Pollination in Arabidopsis. (a) An Arabidopsis flower bud at the end of stage 12, observed by scanning electron microscopy (SEM). The stigma is highlighted by a white square. (b) Enlarged view of the stigma, composed of elongated papilla cells. (c) During pollination, pollen grains released from the anthers land on the stigmatic cells. Pollen grains have been highlighted in red and papilla cells in green for a better visualization. Scale bars, 50 μm

dynamics is challenging, since CMTs are, for instance, highly sensitive to pressure and can be destabilized in contact with the microscope coverslip [5]. Finally, the pollen–stigma interaction is tightly regulated. Very soon after contact with the papilla surface, the incoming pollen grain triggers water fluxes from the stigmatic cell to hydrate and reach a critical water content required for germination of a pollen tube [4, 6, 7]. It has been shown that a high humidity environment can stimulate pollen hydration bypassing the stigma control [4, 8]. Then, controlling the humidity at the vicinity of the stigma is necessary to properly image the pollenstigma interaction. Considering these physiological constraints, we have designed different microscopy mountings to monitor the stigmatic cytoskeleton, by either static viewing or live-cell imaging. We also describe a procedure to quantify fibril orientation in papillae using FibrilTool [9] as well as a method to destabilize the papilla cytoskeleton using drugs.

Cytoskeleton Dynamics in the Plant Stigma

2 2.1

217

Materials Plant Material

1. Arabidopsis thaliana (Col-0). 2. pSLR1::MAP65.1-citrine transgenic expressed marker for CMTs [5].

Arabidopsis:

stigma-

3. pSLR1::LifeActin-Venus transgenic expressed marker for AMFs [4].

Arabidopsis:

stigma-

4. pACT11::RFP transgenic Arabidopsis: pollen grain/tubeexpressed construction [10]. 2.2

Media

1. Medium A: ½ Murashige and Skoog (MS) medium. Weight 0.22 g MS basal salt mixture and 1 g of sucrose. Add water to a volume of 90 ml and mix. Adjust pH to 5.7 +/- 0.1 with KOH and make to 100 ml with water. Sterilize by autoclaving. Keep at room temperature. 2. Medium B: MS solid medium. Weight 0.44 g MS basal salt mixture and 1 g of sucrose. Add water to a volume of 90 mL and mix. Adjust pH to 5.7 +/- 0.1 with KOH and make up to 100 mL with water. Transfer to an autoclavable bottle and add 0.7 g of plant agar. Sterilize by autoclaving and let dry at room temperature. Keep at room temperature. When needed, warm for a few minutes in a microwave oven until the medium has melted.

2.3

Microscopes

Different microscopes were used depending on the experiment, and these are listed below: 1. Upright Leica SP8 microscope using a 25× objective (numeric aperture 0.95, water immersion). 2. Inverted Zeiss LSM800 microscope (AxioObserver Z1) using a 40× Plan-Apochromat objective (numerical aperture 1.3, oil immersion). 3. Inverted Zeiss microscope (AxioObserver Z1) equipped with a spinning disk module (CSU-W1-T3, Yokogawa) using a 40× Plan-Apochromat objective (numerical aperture 1.1, water immersion).

2.4

Other Materials

1. Forceps (tip thickness 0.01 mm). 2. Cellulose filter paper. 3. Slide, small coverslip (22 × 22 mm), large coverslip (22 × 60 mm). 4. 0.2 mL PCR tube. 5. High-vacuum grease. 6. Lanolin.

218

Lucie Riglet and Isabelle Fobis-Loisy

Fig. 2 Static view of the stigmatic cytoskeleton. (a) The pistil is cut beneath the style transverse to its length (black line) before mounting. (b) To get the most suitable papillae to image, deposit grease plugs around the stigma, place a microscope coverslip on these plugs, and gently press downward (black arrows) to lay down the top periphery papillae. Some papillae, directly in contact with the coverslip will be damaged (highlighted in purple). The non-damaged cells just beneath the coverslip (highlighted in dark green) and almost perpendicular to the confocal laser path (blue arrow) are imaged. (c) Confocal mounting in medium A. (d, e) Confocal images of papilla cells expressing MAP65.1-citrine (d) or LifeActin-Venus (e). Scale bars, 5 μm 2.5 Computer Programs Required

1. Image J software (https://imagej.nih.gov/ij/) to process images and movies. 2. Image J FibrilTool plugin [9].

3

Methods Plants were grown in a growth chamber under a long day cycle of 16 h light/8 h dark at 21 °C/19 °C with a relative humidity ca. 60%. All procedures should be carried out at room temperature unless otherwise stated.

3.1 Static View of the Stigmatic Cytoskeleton: Analysis of Cytoskeleton Fiber Orientation

The mounting shown in Fig. 2 can be used to observe stigmas at all developmental stages from stage 12 to 15 as described in [5]. We are equipped with an inverted microscope, but this mounting can also be used with an upright microscope. 1. Emasculate floral buds from the MAP65.1-citrine line (to visualize the CMTs) or LifeActin-Venus line (to visualize the AMFs).

Cytoskeleton Dynamics in the Plant Stigma

219

2. In the middle of the microscope slide, deposit four small plugs of grease. 3. With a scalpel, cut the pistil just beneath the style transverse to its length (Fig. 2a). 4. Gently deposit the cut pistil in between the four grease plugs. 5. Rapidly place a small microscope coverslip on the top of the grease plugs and gently press downward (Fig. 2b). The coverslip needs to touch the papillae without damaging the cells (see Note 1). 6. Add a droplet of medium A to fill the space between the slide and the coverslip (see Note 2). 7. Observe under an inverted Zeiss LSM800 microscope using a 40× Plan-Apochromat objective (oil immersion). 8. From here (steps 8–14), we describe the procedure for quantifying the CMTs anisotropy as an example, but the same procedure can be applied to the AMFs. Quantitative analyses of the average orientation and anisotropy of fibers can be assessed from images obtained with the methodology described in steps 1–7 (Fig. 2d, e). 9. Install the FibrilTool plugin on ImageJ (instructions can be retrieved from [9]). 10. Open your confocal image. 11. Create a maximum projection of half the volume of the stigmatic cell, and only measure the fiber organization of the upward face of the papilla (close to the objective lens). 12. Double-click on the Tool symbol and choose the channel corresponding to the fibril signal. 13. Select your region of interest (ROI) using the Polygon tool. 14. Click on FibrilTool and then on your ROI (Fig. 3a). A segment with a specific orientation appears. This corresponds to the average orientation of your fibril network within your ROI. A window also opens (Fig. 3c), with a series of numbers in order of appearance (Fig. 3b): file name, ROI number, x-coordinate of ROI center, y-coordinate of ROI center, ROI surface area, average fibril orientation, anisotropy, and coordinate of ROI vertices. Anisotropy values range from zero to one; zero indicates pure isotropy and one complete anisotropy. 3.2 Static View of the Stigmatic Cytoskeleton: Cytoskeleton Destabilization

To perform drug delivery to papilla cells without applying liquid directly on the stigma, we applied lanolin paste around the style as described in [5]: 1. Prepare the lanolin paste containing the oryzalin as follows: add 1 volume of the oryzalin stock solution (888 μg/mL) to

220

Lucie Riglet and Isabelle Fobis-Loisy

Fig. 3 Quantification of the cytoskeleton anisotropy using FibrilTool. (a) A region of interest (green polygon) is defined to estimate the CMT anisotropy in papillae expressing MAP65-citrine. The blue line corresponds to the average orientation of the fibril network estimated by FibrilTool. The number corresponds to the region of interest (number one). Scale bars, 5 μm. (b) Result window obtained from ImageJ using FibrilTool plugin. (c) To ease the identification of the results, data obtained are analyzed in an Excel file sheet, with their corresponding column names. Anisotropy value is framed in black

2 volumes of melted (55 °C) lanolin, and mix thoroughly until the formation of a homogeneous emulsion. 2. Under a stereomicroscope, apply the oryzalin containing lanolin just beneath the stigmatic cells, around the style (Fig. 4a, b). Be careful not to directly touch the stigmatic cells with lanolin. 3. Wait for 4 h at 21 °C. 4. After 4 h, observe the stigma using the mounting described in 3.1. We observed that lower oryzalin concentrations produce incomplete CMT destabilization (Fig. 4c). 5. Here, we have described the procedure to deliver oryzalin, but the same method could be applied to other drugs. 3.3 Dynamic View of the Stigmatic Cytoskeleton Remodeling upon Pollination, Using an Inverted Microscope

To visualize stigmatic CMT and AMF remodeling in response to pollen arrival, it is necessary to avoid liquid excess since it can perturb the control of pollen hydration and germination at the stigma surface. In addition, to visualize the cytoskeleton in papilla cells, the stigma should be maintained alive throughout the experiment. 1. Emasculate floral buds from the MAP65.1-citrine line at the end of stage 12 [11]. 2. Detach a mature anther from a stage 14 flower of pACT11:: RFP Arabidopsis. 3. Under a stereomicroscope, pollinate the stigma surface by gently brushing the stigma with the dehiscent anther (Fig. 5a). Ideally, cover the stigma with a single layer of pollen grains (see Note 3). 4. Immediately after pollination, cut the pistil in the middle of the ovary, transverse to its length. 5. Deposit the pollinated pistil upside down in the center of a large coverslip.

Cytoskeleton Dynamics in the Plant Stigma

221

Fig. 4 Application of cytoskeleton destabilization drugs. (a) Schematic view of a treated pistil showing the position of the lanolin paste (orange ring). (b) Top view of the treated stigma, observed by SEM. The lanolin paste is highlighted in orange, the basal part of the pistil in green for a better visualization. (c) Papillae expressing MAP65-citrine after 4 or 6 h of oryzalin treatment. CMT depolymerization is complete and homogeneous after 4 h of treatment at 888 μg/mL. At 404 μg/mL and 666 μg/mL, CMTs are not completely destabilized. Scale bars, 50 μm. Reproduced from Riglet et al., 2020 with permission from eLife, under the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)

6. Place four grease plugs around the cut pistil. 7. Cover the cut part of the ovary with a strip of cellulose paper (0.5 cm x 3 cm) humidified with medium A. Be careful to avoid medium excess on the cellulose strip (see Note 2). 8. At the opposite side of the strip, place a block of medium B (square of 0.5 cm2), to maintain a constant humidity (Fig. 5b). 9. Gently place a small coverslip on the top of the grease plugs, and gently press downward; not too much so as to avoid papilla damage and water release (see Note 1). Do not add any medium between the coverslips (see Note 2). 10. Observe under a Zeiss microscope equipped with a spinning disk module using a 40× Plan-Apochromat objective (numerical aperture 1.1, water immersion).

222

Lucie Riglet and Isabelle Fobis-Loisy

Fig. 5 Live imaging of cytoskeleton dynamics upon pollination (inverted mounting). (a) pACT11-RFP pollen grains are deposited on a MAP65-citrine stigma (1). Pollinated pistil is cut in the middle of the ovary transverse to its length (black line) (2). (b) The cut pistil is mounted without any medium and observed under confocal microscopy. (c) Time-lapse imaging of stigmatic CMT dynamics upon pollination. t0 corresponds to the pollen germination time point. Pollen outline and pollen tube path are highlighted in red for a better visualization. No major alteration of stigmatic CMT arrays is visible. Scale bars, 10 μm. Reproduced from Riglet et al., 2020 with permission from eLife, under the Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)

11. Acquire serial-confocal images of the entire volume of the stigma every μm and every minute. This mounting coupled with a spinning confocal microscope is particularly adapted for high-speed optical sectioning over short timescales to image the CMT remodeling in stigmatic cells along the pollen tube path [5] (Fig. 5c). 12. The protocol described above can also be used to study AMFs. 3.4 Dynamic View of the Stigmatic Cytoskeleton Remodeling upon Pollination, Using an Upright Microscope

1. A small humidity chamber is designed to control the humidity in the vicinity of the stigma. In the center of a microscope slide, deposit a droplet of medium B, and let this dry for at least 30 minutes. 2. A PCR tube is cut in half, and its conical extremity is perforated with a needle. The pierced and cut PCR tube is embedded into the solid droplet of medium B (Fig. 6a, b). 3. Place four grease plugs around the humidity chamber. 4. Emasculate floral buds from the LifeActin-Venus line at the end of stage 12 [11]. 5. Cut the pistil in the middle of the ovary, transversally to its length, and introduce vertically into the pierced PCR tube of the humidity chamber. 6. Detach a mature anther from a stage 14 flower of pACT11RFP Arabidopsis.

Cytoskeleton Dynamics in the Plant Stigma

223

Fig. 6 Live imaging of cytoskeleton dynamics upon pollination (upright mounting). (a) To prepare the humidity chamber, pierce the bottom of the PCR tube with a needle, and cut the tube in half (black line). (b) A LifeActinVenus pistil is cut in the middle of the ovary and introduced in the pierced and cut PCR tube embedded into a droplet of solid medium B. Then, pACT11-RFP pollen grains are deposited on the stigma. (c) Confocal mounting. (d) Time-lapse imaging of stigmatic AMF dynamics upon pollination. The first image was recorded 10 min after pollen deposition. Actin filaments focus toward the pollen grain contact site. 22 min after pollen deposition, a ring of AMFs around the emerging pollen tube is visible (white arrow). Pg, pollen grain; pt.: pollen tube Scale bars, 10 μm

7. Under a stereomicroscope, pollinate the stigma surface by carefully brushing the stigma with the dehiscent anther. Ideally, cover the stigma with a single layer of pollen grains (see Note 3). 8. Immediately after pollination, gently place a small coverslip on top of the grease plugs and gently press downward (Fig. 6c). The coverslip needs to touch the papillae without damaging the cells and avoid any water release (see Note 1). Do not add any medium between slide and the coverslip (see Note 2). 9. Observe under an inverted Zeiss LSM800 microscope using a 40× Plan-Apochromat objective (oil immersion). Acquire serial-confocal images of the entire volume of the stigma every μm and every 2 to 5 min (Fig. 6d). As pollination is performed just before observation and directly in the humidity chamber, this mounting is particularly adapted to catch the very early cellular changes occurring in the stigmatic cells after pollen attachment. Moreover, contrary to the previous mounting, the humidity level is better controlled which is suitable for assessing pollen hydration and germination [4]. 10. The protocol described above can also be used to study CMTs.

224

4

Lucie Riglet and Isabelle Fobis-Loisy

Notes 1. A better visualization of the fluorescent signal is obtained from papillae just beneath the coverslip and almost perpendicular to the confocal laser path. To obtain accurate images of papillae, we use grease plugs around the stigma to cushion the coverslip and reduce cellular damage. We gently press on it to lay down papillae at the top periphery of the stigma dome so that they are as perpendicular as possible to the laser path (Fig. 2b). 2. When non-pollinated papillae are imaged, mounting in a liquid medium is suitable. However, we observed that pollen germination and pollen tube growth are inhibited in liquid; thus avoid using wet mounting for time-lapse imaging. To capture dynamic events occurring in stigmatic cells in response to pollen, a dry mounting is more adapted. Be careful not to press too much on the coverslip on top of the stigma in order to avoid cell damage and water release. 3. To avoid too much pollen grain deposition, under a stereomicroscope, start by brushing the dehiscent anther on a tissue to release excess pollen and then gently brush the stigma to deposit a single layer of pollen grains. Do not use a paintbrush or another instrument (forceps, needle, etc.) to deposit pollen. We observed that touching pollen grains with a paintbrush or another instrument to transport them drastically reduces their hydration rate on the stigma surface.

References 1. Wang X, Mao T (2019) Understanding the functions and mechanisms of plant cytoskeleton in response to environmental signals. Curr Opin Plant Biol 52:86–96. https://doi.org/ 10.1016/j.pbi.2019.08.002 2. Samuel MA, Tang W, Jamshed M, Northey J, Patel D, Smith D, Siu KWM, Muench DG, Wang Z-Y, Goring DR (2011) Proteomic analysis of brassica stigmatic proteins following the self-incompatibility reaction reveals a role for microtubule dynamics during pollen responses. Mol Cell Proteomics 10(M111):011338. h t t p s : // d o i . o r g/ 1 0 . 1 0 7 4 / m c p . M 1 1 1 . 011338 3. Iwano M, Shiba H, Matoba K, Miwa T, Funato M, Entani T, Nakayama P, Shimosato H, Takaoka A, Isogai A, Takayama S (2007) Actin dynamics in papilla cells of Brassica rapa during self- and cross-pollination. Plant Physiol 144:72–81. https://doi.org/10. 1104/pp.106.095273

4. Rozier F, Riglet L, Kodera C, Bayle V, Durand E, Schnabel J, Gaude T, Fobis-Loisy I (2020) Live-cell imaging of early events following pollen perception in self-incompatible Arabidopsis thaliana. J Exp Bot 71:2513–2526. https://doi.org/10.1093/jxb/eraa008 5. Riglet L, Rozier F, Kodera C, Bovio S, Sechet J, Fobis-Loisy I, Gaude T (2020) KATANIN-dependent mechanical properties of the stigmatic cell wall mediate the pollen tube path in Arabidopsis. eLife 9:e57282. https://doi.org/10.7554/eLife.57282 6. Hiroi K, Sone M, Sakazono S, Osaka M, Masuko-Suzuki H, Matsuda T, Suzuki G, Suwabe K, Watanabe M (2013) Time-lapse imaging of self- and cross-pollinations in Brassica rapa. Ann Bot 112:115–122. https://doi. org/10.1093/aob/mct102 7. Wang L, Clarke LA, Eason RJ, Parker CC, Qi B, Scott RJ, Doughty J (2017) PCP-B class pollen coat proteins are key regulators of the hydration checkpoint in Arabidopsis

Cytoskeleton Dynamics in the Plant Stigma thaliana pollen–stigma interactions. New Phytol 213:764–777. https://doi.org/10.1111/ nph.14162 8. Safavian D, Jamshed M, Sankaranarayanan S, Indriolo E, Samuel MA, Goring DR (2014) High humidity partially rescues the Arabidopsis thaliana exo70A1 stigmatic defect for accepting compatible pollen. Plant Reprod 27:121–127. https://doi.org/10.1007/ s00497-014-0245-z 9. Boudaoud A, Burian A, Borowska-Wykre˛t D, Uyttewaal M, Wrzalik R, Kwiatkowska D, Hamant O (2014) FibrilTool, an ImageJ

225

plug-in to quantify fibrillar structures in raw microscopy images. Nat Protoc 9:457–463. https://doi.org/10.1038/nprot.2014.024 10. Rotman N, Durbarry A, Wardle A, Yang WC, Chaboud A, Faure J-E, Berger F, Twell D (2005) A novel class of MYB factors controls sperm-cell formation in plants. Curr Biol 15: 244–248 11. 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

Chapter 17 Investigation of ROP GTPase Activity and Cytoskeleton Dynamics During Tip Growth in Root Hairs and Pollen Tubes Lei Zhu and Ying Fu Abstract Pollen tubes and root hairs are typical tip-growing cells and are employed as model systems to study plant cell polarity. Previous studies have shown that the Rho family ROP GTPase plays a critical role in the regulation of pollen tube and root hair growth. Periodically, activated ROP GTPase coordinates with the tip-focused calcium gradient, to regulate actin dynamics and vesicle trafficking. Moreover, microtubules are also involved in organelle movement and growth directionality. Here, we describe methods for analyzing the spatiotemporal localization and activity of ROP, cortical microtubule organization, and F-actin dynamics in pollen tubes and/or root hairs. Key words Active ROPs, Cytoskeleton, Pollen tubes, Root hairs, Spinning disc confocal microscopy, Arabidopsis thaliana

1

Introduction The tip growth of pollen tubes and root hairs is a rapid and highly polarized process that occurs exclusively at the apical region. Therefore, in addition to their important biological function during plant growth and development, both the pollen tube and the root hair also serve as model systems for studying signal transduction and regulatory mechanisms controlling plant cell polarity and polarized growth [1]. ROP belongs to a Rho-like plant-specific small GTPase subfamily that plays crucial roles in signal transduction in many physiological processes in plants [2]. As a binary molecular switch that transmits extracellular signals to intracellular biological events, the rapid conversion of ROPs between the GTP-bound (active) and the GDP-bound (inactive) forms allows a fast response to cellular processes [3, 4]. The role of ROPs in the regulation of tip growth has been described. ROP2 is a positive regulator of root hair initiation and tip growth in Arabidopsis, and the active ROP2 is located

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_17, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

227

228

Lei Zhu and Ying Fu

in the apical plasma membrane (PM) of the root hair tip [5, 6]. ROP4, ROP6, and ROP11 also accumulate at the root hair initiation site to regulate root hair growth [5, 7, 8]. In addition, ROP1 is highly expressed in pollen, and its GTP-bound activated form is localized on the apical PM of growing pollen tubes, thereby supporting the normal polar growth of pollen tubes [9]. Activated ROP GTPase interconnects with tip-focused calcium gradient, actin dynamics, and vesicle trafficking. The regulation of the organization and the dynamics of the cytoskeleton are important for normal tip growth [1]. Microtubules are not thought to directly regulate tip growth, but to support the slow and shortrange transport of vesicles and organelles in pollen tubes. Depolymerization of microtubules can result in straighter pollen tubes [10], suggesting that they play an important role in maintaining the directionality of polar growth [11–13]. Therefore, the use of live cell imaging to investigate the growth process of the root hair and the pollen tube tip is very useful in understanding the signal transduction and regulatory mechanisms of apical growth. In this chapter, we present several methods that are used to monitor the growth process of pollen tubes and root hairs, to detect the activity and location of the ROP protein, and to assess the organization and dynamics of the cytoskeleton in Arabidopsis pollen tubes and root hairs [14–17]. We describe the protocols for in vitro germination of pollen tubes, preparation of root hair growth medium, immunofluorescence techniques, and confocal observation approaches, which can be applied to study the regulatory network involving the ROP signaling pathway and the cytoskeleton involved in maintaining growth polarity and maintaining tip cell elongation.

2

Materials

2.1 Plant Materials and Transgenic Reporters

1. Arabidopsis thaliana ecotype Columbia-0 as the wild type. 2. Lat52:lifeact-mEGFP transgenic lines, an F-actin reporter, expressed in pollen tubes [18, 19]. 3. 35S:GFP-fABD2-GFP transgenic lines, an F-actin reporter, expressed in root hairs [20]. 4. 35S:GFP-ROP2 [16] or Super::RIC4ΔC-GFP [21] transgenic lines, used as an active ROP reporter.

2.2 Solutions and Antibodies

1. Seedling growth medium: one-half-strength Murashige and Skoog salts (1/2 MS) with 0.9% (w/v) Phytagel plant tissue culture agar.

Investigation of ROP GTPase Activity and Cytoskeleton Dynamics During Tip. . .

229

2. Pollen grain germination medium: 5 mM CaCl2, 5 mM KCl, 1 mM MgSO4, 0.01% H3BO3, 10% (w/v) sucrose, and 1.5% (w/v) low-melting agarose (Invitrogen), pH 7.5 (adjusted with 0.1 M NaOH). 3. Fixative formulation: 2.5% (v/v) paraformaldehyde, 2 mM MgCl2, 2 mM EGTA, 5% (w/v) sucrose, in 50 mM PIPES buffer, pH 6.9. 4. PEM buffer: 50 mM PIPES buffer, 2 mM MgCl2, 2 mM EGTA, pH 6.9. 5. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4, pH 7.4. 6. Enzymatic solution: 1% (w/v) pectinase, 1% (w/v) cellulose in PEM buffer. 7. Anti-fluorescence quenching solution: 0.1% paraphenylene diamine, 50% glycerol, dissolved in PBS buffer, pH 9.0. 8. Mouse anti-α-tubulin monoclonal antibody (#T5168, SigmaAldrich). 9. Alexa 488 (or fluorescein isothiocyanate, FITC)-conjugated goat anti-mouse IgG (#ab150113, Abcam). 2.3 Microscopy and Image Analysis Software

1. Humidity chamber. 2. Glass slides. 3. Forceps. 4. ImageJ software.

3

Methods

3.1 Pollen Tube Germination In Vitro

1. Prepare fresh pollen germination medium. 2. Add 2 mL of the above medium to a 35 mm Petri dish. After the medium has solidified, cut off the four sides, retain the central part of the medium, and flip it over on the dish. This side is flatter, and pollen will spread along it (see Note 1). 3. Collect newly blooming flowers, and spread pollen evenly on the surface of the medium. Use approximately four to six flowers per medium-sized dish (see Note 2). 4. Seal the Petri dishes with Parafilm to maintain humidity; then wrap the Petri dishes with tinfoil, and allow the pollen to germinate in the dark at 25 °C for 3–5 h. 5. Approximately 4 h after germination, capture images of pollen tubes using a microscope equipped with a CoolSNAP HQ CCD camera. At this time, the wild-type pollen germination rate can usually reach about 70–80% (see Note 3).

230

Lei Zhu and Ying Fu

3.2 Immunofluorescent Staining of Microtubules in Arabidopsis Pollen Tubes

1. Allow the wild-type and mutant pollen to germinate for 4 h on a 10 × 10 mm section of solid medium in vitro. 2. Add fixative to the solid medium to fix the germinated pollen tubes at room temperature for 90 min. 3. Remove the fixative solution and rinse the solid medium three times in PEM buffer for 10 min each at room temperature. 4. Carry out cell wall digestion with enzymatic solution for 1 h at room temperature; rinse in PEM buffer three times for 10 min each. 5. The pollen tubes are incubated with primary antibody (mouse anti-α-tubulin monoclonal antibody [1:600 dilution in PBS buffer with 5% glycine]) at 4 °C overnight; then rinse in PBS buffer three times for 10 min. 6. Apply an Alexa Fluor® 488 or fluorescein-isothiocyanate (FITC)-conjugated goat anti-mouse IgG (1:400 dilution in PBS buffer) as the secondary antibody for 1 h at 37 °C. 7. After rinsing in PBS three times, mount the tubes in an antifluorescence quenching solution and then observe, and acquire the images by confocal microscopy (Fig. 1).

3.3 Observation of FActin Dynamics in Tip Growth Cells

1. To observe F-actin organization, obtain single (for the apical region) or z-stack-projections (for the shank region) of optical sections (0.5 μm) of root hairs expressing 35S:GFP-fABD2GFP or pollen tubes expressing Lat52:lifeact-mEGFP. 2. Time-lapse images are taken at 30 s intervals over a 10 min period using 150 ms exposure times for growing pollen tubes (Fig. 2) or at 5 s intervals over a 10 min period using 200 ms exposure times for growing root hairs. 3. All fluorescence images of pollen tubes or root hairs are recorded by spinning disc confocal microscopy equipped with a CCD camera and a × 100 1.45–numerical aperture objective (or a × 60 1.42–numerical aperture objective). GFP is excited using 488-nm argon lasers, with emission captured using 525 ± 5.5 nm filters.

3.4 Preparation of Root Hair Growth Medium

Root hairs are very sensitive plant material. They can be damaged during the preparation of the sample, which prevents them from growing. To monitor the dynamics of root hair growth, a special medium device has been developed for the microscopic observation of the growing root hairs. 1. Prepare 1/2 MS solid and liquid media. 2. Position two 1 mL tip sets together as a stand in the center of a 90 × 90 mm Petri dish, prepare 24 × 50 mm coverslips, and autoclave together with the 1/2 MS medium.

Investigation of ROP GTPase Activity and Cytoskeleton Dynamics During Tip. . .

231

Fig. 1 Immunofluorescent labeling of microtubules in pollen tubes. Microtubules are organized as bundles in the shank region, while few microtubules are detected in the apical region of the pollen tube. Bar = 10 μm

Fig. 2 Time-lapse images showing the fine actin filaments visible at the apex and actin cables arranged along the shanks of a wild-type pollen tube expressing Lat52:lifeact-eGFP. Bar = 5 μm

3. On an ultra-clean bench, place 3 to 4 mL of sterilized 1/2 MS medium on the coverslip. After this solidifies, cut three of the edges (the top edge, the left, and the right edge) to form a horizontal section along the long axis (Fig. 3a).

232

Lei Zhu and Ying Fu

Fig. 3 Schematic diagram showing the preparation of root hair growth medium and chamber. A. Top view of Arabidopsis seeds seeded on the upper edge of the root hair growth medium. B. Side view of the moisturizing chamber. The oblique coverslip allows the seeds to grow against the coverslip after germination, facilitating subsequent observation

4. Add 1/2 MS medium to the Petri dish and rely on its solidification to fix the holder in place. 5. Transfer the coverslip containing the 1/2 MS solid medium to the Petri dish, and place it on the holder at an angle, with the horizontal cut side facing upward. Then add 4 to 5 mL of 1/2 MS medium to the bottom side of the coverslip to hold it in place after solidification (Fig. 3b). In addition, 1/2 MS liquid medium is then added to the Petri dish to maintain a moist environment for the growth of the seedlings. 6. The seeds are sterilized for 15 min in 0.5% sodium hypochlorite and treated at 4 °C in the dark for 2 days. Place the sterilized and cold imbibed seeds on the horizontal section prepared in step (3) (see Note 4). The Petri dishes are then sealed with Parafilm and transferred to an illuminated incubator. 3.5 Measurement of Root Hair Growth Rate

1. The seeds of wild-type Arabidopsis or mutants are germinated on a coverslip layered with root hair medium, as indicated above. 2. After 5 days, the coverslips containing the seedlings are transferred directly to the stage of an inverted microscope and visualized at ×40 magnification using a spinning disc confocal microscope.

Investigation of ROP GTPase Activity and Cytoskeleton Dynamics During Tip. . .

233

3. The region 2 to 4 mm from each primary root tip is chosen to examine the growth rate. Time-lapse images of the best growing root hairs are photographed 60 times at 600 s intervals. 4. The distance between the apex of the root hair in the first picture and in the last is recorded using ImageJ software. The increase in length of each root hair is determined, and the growth rate is calculated. 3.6 Quantitative Analysis of the Fluorescence Intensity of Active ROPs (PM/ Cytosol Ratio) in Root Hair Growth

Observing the localization of the fluorescent protein-fused ROPs (e.g., GFP-ROP2) will not only reflect the dynamic changes of its intracellular localization, but these observations can also be used as an indicator of ROP activity status particularly the PM-associated response to the active ROPs. A GFP-tagged deletion mutant of the ROP effector RIC4 (RIC4ΔC-GFP) has been used to monitor ROP activity in pollen tubes or root hairs [16, 21]. The RIC4ΔC-GFP is a marker used to avoid depolarized growth caused by ROP overexpression. Below, we describe tracking the dynamics of PM-localized GFP-ROP2 in the growth of root hairs as a representative example of the protocol steps. 1. To investigate the sub-localization of fluorescence-fused proteins (e.g., GFP-ROP2 or Super::RIC4ΔC-GFP), acquire timelapse images using 30% laser intensity and operating in the 512 × 512 mode with an exposure time of 200 ms. Capture the images for 10 min at 5 s intervals (see Note 5). 2. To quantify the localization of GFP-ROP2, measure the mean fluorescence intensity of the GFP signal at the apical plasma membrane (along the PM, length 30 μm) and in the cytosol (area, 40 μm2) using ImageJ software. A scheme illustrating how the fluorescence intensity of the PM/Cytosol ratio is measured is shown in Fig. 4. 3. In detail, use a “segmented line” or a “freehand line” drawn along the PM, and then measure the fluorescence intensity (Analyze > Plot Profile > list). Use the “oval selections” to draw a circle around an area of interest, and measure the fluorescence intensity (Analyze > Measure > Mean).

4

Notes 1. If pharmacological treatment is to be performed, wait until the temperature of the melted medium drops to about 60 °C before adding the agent (such as Latrunculin B, an actin polymerization inhibitor). 2. A low pollen grain density is not conducive to in vitro germination, and high density is not conducive to subsequent sample visualization.

234

Lei Zhu and Ying Fu

Fig. 4 Quantitative analysis of the fluorescence intensity (PM/Cytosol ratio) during root hair growth. (a) The schematic diagram shows the measurement of the fluorescence intensity used to calculate the PM/cytosol ratio. The red line indicates the apical PM of the growing root hair (1), and the green circle indicates the cytosol of the root hair apex (2). (b) Active ROP2 is localized to the apical plasma membrane of growing root hairs that express GFP-ROP2. Bar = 5 μm

3. Below this rate of germination indicates poor pollen activity or inappropriate in vitro germination conditions. 4. Note that the distance between seeds should not be too far apart when seeding, as it is not conducive to root hair growth. Seeds can be seeded quite densely, as long as the root hairs of two adjacent seedlings do not overlap to interfere with the subsequent observation. 5. Ensure that pollen tubes and root hair cells remain in the growing state throughout the observation. As with ROP2, the active ROP2 signal will accumulate at the apical PM but will disappear from the PM once the root hairs stop growing.

Acknowledgements This work was supported by the Chinese Ministry of Science and Technology (grant No. SQ2019YFE011441 to Y.F. and L. Z) and the National Natural Science Foundation of China (grant no. 32070311 and 31571384 to L.Z.; 32061143018, 91735305 and 91854119 to Y.F.).

Investigation of ROP GTPase Activity and Cytoskeleton Dynamics During Tip. . .

235

References 1. Lee YJ, Yang Z (2008) Tip growth: signaling in the apical dome. Curr Opin Plant Biol 11:662– 671 2. Yang Z (2002) Small GTPases: versatile signaling switches in plants. Plant Cell (Suppl) 14: S375–S388 3. Nibau C, Wu HM, Cheung AY (2006) RAC/ROP GTPases: ‘hubs’ for signal integration and diversification in plants. Trends Plant Sci 11:309–315 4. Yang Z, Fu Y (2007) ROP/RAC GTPase signaling. Curr Opin Plant Biol 10:490–494 5. Jones MA, Shen JJ, Fu Y, Li H, Yang Z, Grierson CS (2002) The Arabidopsis Rop2 GTPase is a positive regulator of both root hair initiation and tip growth. Plant Cell 14:763–776 6. Jones MA, Raymond MJ, Yang Z, Smirnoff N (2007) NADPH oxidase dependent reactive oxygen species formation required for root hair growth depends on ROP GTPase. J Exp Bot 58:1261–1270 7. Molendijk AJ, Bischoff F, Rajendrakumar CS, Friml J, Braun M, Gilroy S, Palme K (2001) Arabidopsis thaliana Rop GTPases are localized to tips of root hairs and control polar growth. EMBO J 20:2779–2788 8. Bloch D, Lavy M, Efrat Y, Efroni I, BrachaDrori K, Abu-Abied M, Sadot E, Yalovsky S (2005) Ectopic expression of an activated RAC in Arabidopsis disrupts membrane cycling. Mol Biol Cell 16:1913–1927 9. Gu Y, Vernoud V, Fu Y, Yang Z (2003) ROP GTPase regulation of pollen tube growth through the dynamics of tip-localized F-actin. J Exp Bot 54:93–101 10. Cheung AY, Wu HM (2008) Structural and signaling networks for the polar cell growth machinery in pollen tubes. Annu Rev Plant Biol 59:547–572 11. Ketelaar T, Faivre-Moskalenko C, Esseling JJ, de Ruijter NC, Grierson CS, Dogterom M, Emons AM (2002) Positioning of nuclei in Arabidopsis root hairs: an actin-regulated process of tip growth. Plant Cell 14:2941–2955

12. Van Bruaene N, Joss G, Van Oostveldt P (2004) Reorganization and in vivo dynamics of microtubules during Arabidopsis root hair development. Plant Physiol 136:3905–3919 13. Mathur J (2004) Cell shape development in plants. Trends Plant Sci 9:583–590 14. Zhu L, Zhang Y, Kang E, Xu Q, Wang M, Rui Y, Liu B, Yuan M, Fu Y (2013) MAP18 regulates the direction of pollen tube growth in Arabidopsis by modulating F-actin organization. Plant Cell 25:851–867 15. Zhang Y, Kang E, Yuan M, Fu Y, Zhu L (2015) PCaP2 regulates nuclear positioning in growing Arabidopsis thaliana root hairs by modulating filamentous actin organization. Plant Cell Rep 34:1317–1330 16. Kang E, Zheng M, Zhang Y, Yuan M, Yalovsky S, Zhu L, Fu Y (2017) The microtubule-associated protein MAP18 affects ROP2 GTPase activity during root hair growth. Plant Physiol 174:202–222 17. Boavida LC, McCormick S (2007) Temperature as a determinant factor for increased and reproducible in vitro pollen germination in Arabidopsis thaliana. Plant J 52(3):570–582 18. Riedl J, Crevenna AH, Kessenbrock K, Yu JH, Neukirchen D, Bista M, Bradke F, Jenne D, Holak TA, Werb Z, Sixt M, Wedlich-Soldner R (2008) Lifeact: a versatile marker to visualize F-actin. Nat Methods 5:605–607 19. Vidali L, Rounds CM, Hepler PK, Bezanilla M (2009) Lifeact-mEGFP reveals a dynamic apical F-actin network in tip growing plant cells. PLoS One 4:e5744 20. Wang YS, Yoo CM, Blancaflor EB (2008) Improved imaging of actin filaments in transgenic Arabidopsis plants expressing a green fluorescent protein fusion to the C- and N-termini of the fimbrin actin-binding domain 2. New Phytol 177:525–536 21. Hwang JU, Gu Y, Lee YJ, Yang Z (2005) Oscillatory ROP GTPase activation leads the oscillatory polarized growth of pollen tubes. Mol Biol Cell 16:5385–5399

Chapter 18 Functional Analysis of Phospholipid Signaling and Actin Dynamics: The Use of Apical Growing Tobacco Pollen Tubes in a Case Study Teresa Braga, Fernando Vaz Dias, Marta Fratini, Susana Serrazina, Ingo Heilmann, and Rui Malho´ Abstract Signaling molecules are crucial to perceive and translate intra- and extracellular cues. Phosphoinositides and the proteins responsible for their biosynthesis (e.g., lipid kinases) are known to influence the (re)organization of cytoskeletal elements, namely, through interaction with actin and actin-binding proteins. Here we describe methods to functionally characterize lipid kinases and their phosphoinositide metabolites in relation to actin dynamics. These methods include GFP-tagged protein expression followed by timeresolved live imaging and quantitative image analysis. When combined with biochemical and interaction studies, these methods can be used to correlate signaling with actin dynamics, microfilament assembly, and intracellular trafficking, linking structure and function. Key words Actin imaging, Diacylglycerol kinases, Fab1 kinases, PIPkinases, Phosphoinositide biosensors, Nicotiana tabacum, Tip growth

1

Introduction Apical tip growth is a form of cell extension common to all eukaryotes. This growth form serves as a paradigm for cell polarity since cell extension is restricted to a narrow zone at the apex [1]. Thus, apical tip growth is particularly suitable for investigations on signal transduction, cytoskeleton and wall structure, membrane dynamics, and even cell–cell communication [2]. The development of molecular biosensors in parallel with functional analysis of fusion proteins highlighted the importance of phosphoinositides (PIs)—and the proteins responsible for their metabolism—in all aspects of cellular and organismic development. In plant cells, the events modulated by PIs include cytoskeletal dynamics, ion fluxes, regulation of reactive oxygen species, membrane trafficking, and cell wall secretion [3]. Specifically, the

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_18, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

237

238

Teresa Braga et al.

interaction between PIs and the actin cytoskeleton seems to involve Rac/Rho GTPase-mediated recruitment of lipid kinases [4]. In this chapter, we describe imaging methodologies for the observation of individual or co-expressed fluorescent fusion proteins in pollen tubes (biolistic). We focus on methods to map lipid kinases and PI biosensors, individually or simultaneously with imaging of actin dynamics. These methods can however be extended to any protein of interest putatively involved in cytoskeletal regulation (e.g., GTPases). In addition to the characterization of subcellular localization and dynamics, the results obtained with these methodologies may complement further interaction studies resorting to FRET/FLIM or BiFC techniques (see Chap. 29). PIs and the proteins responsible for their metabolism can be found in all cellular compartments and frequently exhibit distinct distributions [5]. Our examples thus cover imaging of lipid kinases located in cytosol or membrane compartments and of PI biosensors based on specific lipid-binding domains [6–8]. Although Arabidopsis thaliana remains as the best model for genetic and molecular analysis, the use of heterologous systems for protein localization studies is quite disseminated in the community with Nicotiana (tabacum and benthamiana) being extensively used in protein trafficking studies. Arabidopsis pollen is prone to several artifacts when performing dynamics of protein localization; the tubes are very sensitive to laser irradiation, frequently change focal plane, and experience significant alterations of growth rate that result in altered tip diameter and morphological shape. By contrast, in tobacco pollen, these problems are significantly mitigated, and tobacco pollen tubes have been successfully used by different groups [6–10], and, to the best of our knowledge, no published data has been shown to be artifactual.

2

Materials

2.1 Plant Material and Microscopy

1. Nicotiana tabacum: grown (for ~5–7 weeks) in plant cabinets with a 16-h-light/8-h-dark photoperiod (and at 25/21 °C) and universal substrate soil with vermiculite (6:1). 2. Fresh or -80 °C stored N. tabacum pollen: for storage, collect pollen from freshly opened flowers into microcentrifuge tubes and freeze directly. 3. Pollen tube growth media: 5% sucrose, 15 mM MES-KOH pH 5.9, 0.03% casein acid–hydrolysate, 1 mM CaCl2, 1 mM KCl, 0.8 mM MgSO4, 1.6 mM H3BO3, 30 μM CuSO4, 12.5 % PEG-6000, 10 μg/mL rifampicin, filter sterilized and stored at -20 °C [11]. 4. Phytagel.

Phospholipid Signaling and Actin Dynamics in Apical Growth

239

5. Glass coverslips (24 × 60 and 24 × 24 mm), square petri dishes. 6. Confocal microscope (laser scanning or spinning disk) with water-immersion objectives (e.g., x63, 1.15NA). 2.2

Plasmids

1. pGreen pLat52 mGFP4 (original plasmid pGreen 0000, [12]) for N. tabacum pollen transformation (see Note 1). 2. In vivo biosensors used for purposes described in this chapter include RedStar-PLCPH [9] and YFP-PH [10] for PtdIns(4,5) P2, GFP-FYVE [13] for PtdIns(3)P, and mTalin [10] or LifeAct for actin [14].

2.3 Biolistic Transient Transformation

1. Helium-driven PDS-1000/He particle delivery system. 2. Plasmid-coated gold microcarriers (1 μm diameter): in 50% glycerol prepare 50 μL of gold microcarriers at 5 μg/μL. 3. Spermidine. 4. CaCl2 solution: 2.5 M stock in water. 5. 1100 psi (or 1350 psi) rupture disks, microcarriers, and stopping screens (Bio-Rad). 6. Cellulose acetate filter: 47 mm diameter, 0.45 μm pore or 50 mm diameter, 0.2 μm pore. 7. Buckner flask and funnel, vacuum pump. 8. Bench centrifuge and vortex.

2.4 Image Analysis Software Required

3

1. ImageJ / FIJI and/or ILASTIK. 2. Adobe Photoshop /PowerPoint.

Methods

3.1 Transient Expression of Constructs in Nicotiana Tabacum Pollen Tubes

The method described here can be used for the efficient transient transformation of Nicotiana tabacum pollen tubes with fluorescent protein fusion constructs. Prior to transformation and for the preparation of two shots per sample, 1 μm of gold particles must be coated with 2–6 μg of the desired plasmid of interest (steps 1– 4). 1. In an 1.5 ml reaction tube, wash the gold particles with ethanol 100%, and let it rest; remove ethanol and repeat twice of the procedure. 2. To 50 μg of particles, add 2–6 μL of plasmid DNA (1 μg/μL], vortex, and precipitate the DNA with 50 μL of CaCl2 (2.5 M) and 20 μL of spermidine (0.1 M) (see Note 2). Vortex for 10 min. 3. Spin, remove the supernatant, and wash with 200 μL of 100% ethanol; vortex, spin, and remove the supernatant again.

240

Teresa Braga et al.

4. Resuspend in 30 μL of 100% ethanol, equally distribute between two microcarriers, and let dry for 10 min. Use immediately after pollen plating. 5. Collect mature pollen grains from two to four recently opened flowers (or stored pollen at -80 °C). 6. Resuspend pollen in 4 mL of the growth media followed by filtering the pollen grains onto a cellulose acetate filter pre-wetted with 1 mL of the growth medium. Transfer the pollen to semi-solid growth medium (3 mL with 0.5% phytagel in a 5.5 cm plate) by placing the filter upside down on the medium and remove the filter. 7. In the helium-driven particle accelerator (PDS-1000/He), place the filtered pollen grains 10 cm away from the rupture disc, and transform with the plasmid-coated particles using 1100 (or 1350) psi rupture disks and a vacuum of 28 inches of mercury. 8. After bombardment, equally divide the medium onto three microscope glass coverslips. Maintain the coverslips inside a humidified petri dish (e.g., with soaked filter paper) at 20–25 °C, preferentially in the dark (see Note 3). 3.2 Imaging Lipid Kinases Fusion Constructs and Phosphoinositide Biosensors

1. Perform biolistic transformation as indicated above, and 4 to 6 h after the bombardment, view samples on an inverted confocal microscope (see Note 4). Maintain the temperature and humidity during imaging as best as possible (see Note 5). 2. For imaging settings, prioritize low expression levels and low laser intensity (e.g., 8bits). 3. Correlate pixel length to microns using the following path: Image > Properties > Pixel width/height. 4. Use the bandpass filter to reduce the background noise of the image using the Fiji software default setting (Process > FFT > Bandpass Filter) (see Note 4). 5. Adjust the brightness and contrast of the image so that the microtubules can be visualized (Image > Adjust > Brightness/ Contrast).

260

Shuwei Wang et al.

Fig. 2 Image processing of microtubules based on the Fiji software. (a) Original image. (b) Image processed by a bandpass filter and adjusted for brightness/contrast. (c) Image processed by the unsharp mask filter. (d) Threshold processed image. (e) Skeletonized image. Bar = 10 μm

6. Use the unsharp mask filters to increase the contrast of the image (Radius = 1 pixel, Mask Weight = 0.6) (Process > Filters > Unsharp Mask). 7. Use the Despeckle plugin to reduce the background noise (Process > Noise > Despeckle). 8. Distinguish the microtubules from the background in the image using the Threshold plugin to finally produce a binary image (Image > Adjust > Threshold) (see Note 5). 9. Use the Despeckle plugin to reduce the background noise (Process > Noise > Despeckle). 10. Activate the Skeletonize plugin, and the microtubules are converted into one-pixel-thick segments (Process > Binary > Skeletonize) (see Note 6). 11. Use the Region of Interest (ROI) Manager to select multiple ROIs and the skeletonized degree of these ROI regions in the original image (Analyze > Tools > ROI manger). Then rightclick Duplicate on the image. These ROI regions will be displayed as an independent cropped image. 12. Use the AnalyzeSkeleton (2D/3D) plugin to analyze the skeletonized MTs and obtain the branch information data (Plugins > Skeleton > AnalyzeSkeleton (2D/3D)). 13. Copy the branch information data to the new EXCEL table. Use the branch distance (i.e., the length of the skeletonized MT), and calculate the ROI area to determine the MT density. The latter is defined as LMT/AROI; LMT and AROI represent the length of all skeletonized microtubules and the area of the ROI, respectively.

Microtubule Rearrangement in Response to ABA

4

261

Notes 1. Microtubules are highly dynamic and sensitive, so the process of transferring seedlings to NaCl medium and later sample preparation should be gentle. 2. It is necessary to use multiple layers of the double-sided tape to make a certain thickness between the slide and the cover glass so that the seedling will not be crushed during observation. At the same time, ensure that the cover glass and the slide are secured to prevent any loss of focus during image acquisition. 3. The image with a high signal-to-noise ratio is better for the subsequent image processing. 4. We use the whole image for the bandpass filter using the Fiji software default setting. If the ROI area of the image is cropped in advance, the large/small structure setting in the bandpass filter will be reset. 5. Threshold segmentation is very important for later image skeletonization. The Fiji software provides an Auto-Threshold function, which can preliminarily screen the images and then choose the appropriate threshold algorithm. 6. In order to determine the accuracy of the skeletonized MTs, the skeletonized images can be merged with the original image.

References 1. Munns R, Tester M (2008) Mechanisms of salinity tolerance. Annu Rev Plant Biol 59: 651–681 2. Mulet JM, Campos F, Yenush L (2020) Editorial: ion homeostasis in plant stress and development. Front Plant Sci 11:618273 3. Yang YQ, Guo Y (2018) Elucidating the molecular mechanisms mediating plant saltstress responses. New Phytol 217:523–539 4. Gong ZZ, Xiong LM, Shi HZ et al (2020) Plant abiotic stress response and nutrient use efficiency. Sci China Life Sci 63:635–674 5. Dou LR, He KK, Higaki T et al (2018) Ethylene signaling modulates cortical microtubule reassembly in response to salt stress. Plant Physiol 176:2071–2081 6. Li CJ, Lu HM, Li W et al (2017) A ROP2RIC1 pathway fine-tunes microtubule reorganization for salt tolerance in Arabidopsis. Plant Cell Environ 40:1127–1142 7. Yang ZJ, Wang CW, Xue Y et al (2019) Calcium-activated 14-3-3 proteins as a molecular switch in salt stress tolerance. Nat Commun 10:1199 8. Wang C, Li JJ, Yuan M (2007) Salt tolerance requires cortical microtubule reorganization in Arabidopsis. Plant Cell Physiol 48:1534–1547

9. Wang SH, Kurepa J, Hashimoto T et al (2011) Salt stress-induced disassembly of Arabidopsis cortical microtubule arrays involves 26S proteasome-dependent degradation of SPIRAL1. Plant Cell 23:3412–3427 10. Marc J, Granger CL, Brincat J et al (1998) A GFP–MAP4 reporter gene for visualizing cortical microtubule rearrangements in living epidermal cells. Plant Cell 10:1927–1939 11. Schindelin J, Arganda-Carreras I, Frise E et al (2012) Fiji: an open-source platform for biological-image analysis. Nat Methods 9: 676–682 12. Arganda-Carreras I, Ferna´ndez-Gonza´lez R, ˜ oz-Barrutia A et al (2010) 3D reconstrucMun tion of histological sections: application to mammary gland tissue. Microsc Res Tech 73: 1019–1029 13. Young K, Morrison H (2018) Quantifying microglia morphology from photomicrographs of immunohistochemistry prepared tissue using ImageJ. J Vis Exp 136:e57648 14. Fujita S, Pytela J, Hotta T et al (2013) An atypical tubulin kinase mediates stress-induced microtubule depolymerization in Arabidopsis. Curr Biol 23:1969–1978

Chapter 21 Analysis of Actin Array Rearrangement During the Plant Response to Bacterial Stimuli Bingxiao Wang, Minxia Zou, Qing Pan, and Jiejie Li Abstract Plants are constantly exposed to various environmental stresses, among which, microbial pathogens are one of the major threats. Studies have shown that the host actin cytoskeleton undergoes active rearrangement during the plant–microbe interaction. This actin remodeling is required for plant resistance to bacterial infection. In this chapter, we introduce a protocol routinely used in our laboratory to investigate actin dynamics in response to bacterial cues. We describe the bacterial inoculation methods, plant sample preparation, and imaging techniques used to monitor actin responses in different Arabidopsis cell types including epidermal cells from light-grown leaves and dark-grown hypocotyls, as well as guard cells. We further introduce a high-throughput image analysis method for quantifying cytoskeletal changes. This protocol has allowed us to dissect the host cell contribution to actin remodeling and identify actin-binding proteins as stimulus-response regulators of the cytoskeleton. Key words VAEM, Confocal microscopy, Live-cell imaging, Actin rearrangement, Bacterial pathogen

1

Introduction Plants are constantly exposed to a variety of microbial pathogens. These parasitic microbes invade the host tissue and exploit the resources of the plant to derive nutrition. However, plants deploy a robust multilayered immune system to defend against the invading pathogens [1]. Perception of microbial cues or microbe-associated molecular patterns (MAMPs) by cell surface receptors activate pattern-triggered immunity (PTI) responses such as mitogen-associated protein kinase (MAPK) signaling, Ca2+ spiking, oxidative burst, and the production of defense hormones [2]. Stomatal closure, known as stomatal immunity, is also activated as a part of PTI responses to restrict pathogen entry [3]. These robust defense reactions can often successfully abrogate the pathogen infection.

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_21, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

263

264

Bingxiao Wang et al.

The plant actin cytoskeleton provides a dynamic structural framework and controls diverse cellular processes such as vesicular trafficking, cargo delivery, and organelle distribution [4]. Emerging evidence demonstrated the importance of the host actin cytoskeleton during the plant–microbe interaction. It has been shown that a rapid increase in the abundance of actin filaments takes place in Arabidopsis epidermal cells upon treatments with the bacterial pathogen Pseudomonas syringae pv. tomato (Pst) DC3000 as well as diverse MAMPs [5–7]. This actin response is considered as a novel hallmark of PTI [8]. Disrupting host actin leads to impaired plant resistance to bacterial infection [5–7]. In addition, actin remodeling is also required for stomatal immunity [9, 10]. In guard cells, actin filaments are radially oriented in open stomata and reorganized into a randomly organized actin array after treatment with bacteria or MAMP. In closed stomata, the actin filaments are rearranged and bundled preferentially as long cables in the longitudinal direction. Mutants that fail to reorganize the actin array have significantly compromised stomatal immunity [10]. In this chapter, we describe the sample preparation and imaging techniques used to monitor actin dynamics during Arabidopsis–bacterial interaction using different host cell types: epidermal cells from light-grown leaves and dark-grown hypocotyls, as well as guard cells. We further introduce the high-throughput image analyses routinely used in our laboratory to quantify the cytoskeletal changes. This protocol has allowed comparative studies between treatments and mutant plant lines to be performed. Therefore, this protocol provides an exceptional opportunity to dissect the host cell contribution to actin remodeling and identify actin-binding proteins as stimulus–response regulators of the cytoskeleton [8].

2 2.1

Materials Materials

1. Arabidopsis stable lines expressing the actin filament marker GFP-fABD2 [11] or GFP-lifeAct [10]. 2. Pseudomonas syringae pv. tomato (Pst) DC3000: stored at 80 °C in KB medium with 20% glycerol. 3. Glass slides and coverslips. 4. One-ml needleless syringe and spray bottle. 5. Eye scissors and dissecting forceps.

2.2

Reagents

1. ½ Murashige and Skoog (MS) medium: 2.2 g/l MS basal medium, 10 g/l sucrose, and 8 g/l plant agar, pH 5.8 with NaOH. Sterilize by autoclaving. 2. MAMP stock solutions: flg22 (amino acid sequence: QRLST GSRINSAKDDAAGLQIA ), elf18 (amino acid sequence:

Actin Remodeling in Arabidopsis under Biotic Stresses

265

SKEKFERTKPHVNVGTIG), or chitin (C9752, Sigma) is dissolved in PBS buffer to 1 mM. Divide the stock solutions into 10 μl aliquots, stored at -80 °C. 3. PBS buffer: 8 g/L NaCl2, 0.2 g/L KCl, 1.44 g/L Na2HPO4, 0.24 g/L KH2PO4, pH 7.4 with HCl. 4. KB medium: 2% (w/v) tryptone, 1% (v/v) glycerol, 0.15% (w/v) K2HPO4, 0.15% (w/v) MgSO4. 5. Bacterial suspension buffer: 10 mM MgCl2 with 1/5000 (v/v) of silwet-77. 6. Stomata-opening buffer: 50 mM KCl, 10 mM CaCl2, 10 mM MES, pH 6.15.

3

Methods

3.1 Preparation of MAMP and Bacterial Working Solutions

1. Thaw the bacterial glycerol stocks in a labtop cooler; transfer 20 μl of bacterial stock to 5 ml of fresh KB medium supplemented with appropriate antibiotics. Grow the bacteria in the shaker at 28 °C for 2 days. 2. Centrifuge the bacteria at 4000 rpm for 3 min in a benchtop centrifuge, pull off the supernatants, and add bacterial suspension buffer, and adjust OD600 to 0.2. For DC3000, an OD600 = 0.2 is approximately 1 × 108 colony-forming units (cfu)/ml. 3. Make final spray solution: dilute the above suspension (OD600 = 0.2) to 5 × 105 cfu/ml in suspension buffer (1: 200 dilution). 4. For making the MAMP working solutions, thaw aliquots of the MAMP stock solutions in a labtop cooler; dilute the stock with distilled water to 1 μM (1:1000 dilution) unless indicated otherwise (see Note 1).

3.2 Bacteria or MAMP Treatment of Epidermal Pavement Cells

1. For bacterial treatment, plants are grown under long-day lighting conditions (16 h light, 8 h dark) at 22 °C for 10 d. 2. Spray plants with spray bottle containing bacterial solution, and keep plants in the growth chamber for 6 h. Cotyledons detached with eye scissors are placed on the slide, mounted in water and covered with a coverslip. 3. For MAMP treatment, plants are grown under long-day lighting conditions at 22 °C for 3 weeks. 4. The abaxial (under) sides of rosette leaves are infiltrated with the desired MAMPs. Use a 1 ml needleless syringe to pressure– infiltrate the intercellular space of the leaf (see Note 2). 5. Cover plants with a clear plastic tray cover for 30 min. The infiltrated area is cut out and placed on the slide, mounted in

266

Bingxiao Wang et al.

MAMP working solutions (see Subheading 3.1 step 4) and covered with a coverslip before observation (see Note 3). 3.3 MAMP Treatment of Dark-Grown Hypocotyls

1. Surface-sterilized seeds are plated onto 1/2 MS plates. After exposure to white light for 4 h, plates are wrapped with three layers of aluminum foil and placed vertically in the growth chamber at 22 °C for 5 days. 2. Whole seedlings are soaked in the desired MAMPs for 5 min. Use dissecting forceps, transfer seedlings carefully to the slide, add a small amount of MAMP working solutions, and gently cover it with a coverslip (see Notes 4–6).

3.4 MAMP Treatment of Guard Cells

1. Cotyledons from 10-day-old seedling are detached with eye scissors and incubated in stomata-opening buffer at 22 °C under constant illumination for 2 h so that the stomata are fully open. 2. Incubate cotyledons in stomata-opening buffer containing 10 μM flg22. After 1 h treatment, cotyledons are placed on the slide, mounted in the same solution and covered with a coverslip.

3.5 Live-Cell Imaging

1. We image fields of guard cells or epidermal pavement cells with a PerkinElmer UltraView Vox spinning disk microscope equipped with a 60 × 1.42 numerical aperture objective. Cells at the abaxial side of leaves are imaged by collecting 24 steps of 0.5 μm each starting at the plasma membrane (see Notes 7–8). 2. We image epidermal cells of dark-grown hypocotyls with VAEM (variable-angle epifluorescence microscope) equipped with a 100× objective by collecting single optical sections from the cortical cytoplasm. Cells from the basal third of hypocotyls are documented with a series of overlapping VAEM images (see Note 8).

3.6 Image Process and Analysis of Epidermal Pavement Cells

The architecture of cortical actin arrays is measured with two metrics: skewness and density [12]. Skewness analysis is used to monitor the extent of actin filament bundling, and the filament density is calculated as the percentage of occupancy of GFP signal in each image. Images are analyzed with Fiji using Higaki’s macro available at http://hasezawa.ib.k.i-tokyo.ac.jp/zp/ Kbi/HigStomata. 1. For the skewness analysis: open collected stacks with Fiji; these will be grayscale 16 bit image stacks. Make a maximum intensity projection (Image > Stack > Z-project > Maximum Intensity > OK). Save maximum intensity images to a new folder.

Actin Remodeling in Arabidopsis under Biotic Stresses

267

Fig. 1 Density analysis of images from epidermal pavement cells after mock and DC3000 treatment. (a, b) Original confocal images taken from 10-day-old cotyledons after treatment with mock (a) or DC3000 (b) for 6 h. (c, d) The respective binary image of (a) and (b) after threshold setting. The density values are indicated in the upright corners. Cells treated with bacteria have a more abundant actin array compared with cells treated with mock. Scale bar = 10 μm

2. Import maximum intensity images as an image sequence (File > Import > Image Sequence >8 bit (check) sort numerically (check) > OK). Import skewness macro. Run analysis. 3. For the density analysis: open collected stacks with Fiji. Apply a 1.0 radius Gaussian blur to an image stack (Process > Filters > Gaussian blur). Save the Gaussian blur stacks in a new folder. 4. Make a maximum intensity projection from the Gaussian blur stack. Apply high-bandpass filter to maximum intensity images (Process > FFT > Bandpass Filter). Set filter large structures down to 40 pixels. Save processed images in a new folder (see Note 9). Import processed maximum intensity images as an image sequence. Set threshold (Image > Adjust > threshold; see Note 10), import density macro, and run analysis (Fig. 1; see Note 11).

268

Bingxiao Wang et al.

Fig. 2 Density analysis of images from epidermal cells of dark-grown hypocotyls. (a) Original VAEM images. (b) Fields of epidermal cells are cropped into micrographs. (c) The binary black and white image after threshold setting used for density analysis. Scale bar = 100 μm 3.7 Analyzing Images from Epidermal Cells of Dark-Grown Hypocotyls

1. VAEM images are cropped (Fig. 2; see Notes 12 and 13) and analyzed in Fiji.

3.8 Analyzing Images of Guard Cells

1. Open collected stacks with Fiji, and make a maximum intensity projection. The areas containing stomata are cropped (Fig. 3a; see Note 13). Save the cropped images to a new folder.

2. Skewness and density analysis is performed as described in Subheading 3.6 with a few modifications. As there are no z-series projections and VAEM generates high-contrast images, we do not apply Gaussian blur, high-bandpass filter, or skeletonization processing steps.

2. Skewness and density analysis is performed as described in Subheading 3.5 step 1. 3. To measure filament angle in guard cells, each actin filament is marked with segmented lines. The average angle of these segmented lines relative to the width of the stomatal pore is quantified (Fig. 3b).

4 Notes 1. Divide MAMP stock solution into small aliquots. Avoid repeated freezing and thawing so that MAMP peptides (i.e., flg22, elf18) stay active and immune responses can be successfully elicited. 2. A small amount of MAMP solution, approximately 100 μl, is enough to infiltrate one leaf. The infiltrated leaf area should appear water-soaked.

Actin Remodeling in Arabidopsis under Biotic Stresses

269

Fig. 3 Filament angle measurement in guard cells. (a) The cropped image of stomata for analysis. (b) The zoomed image from the selected area in (a) (white box). Filament a is segmented to line 2 and 3; θ1 and θ2 indicate the angle of line 2 and 3 relative to the width of the stomatal pore (line 1), respectively. The angle of the filament a is calculated as an average of θ1 and θ2

3. It is easier to obtain better images with small and flat leaf samples. Image chambers are recommended. We usually place two parallel strips of double-sided tape on both ends of the coverslip to create an image chamber. 4. The dark-grown hypocotyls are very fragile. When transferring seedlings to the glass slide, avoid damaging the hypocotyls by holding the cotyledon or root gently with dissecting forceps. 5. 10 μl of MAMP solution is usually added between the glass slide and coverslip. Remove excess solution with a tissue. Once mounted, the seedlings should not move freely under the coverslip. 6. Elf18 and chitin can elicit actin responses in epidermal cells of dark-grown hypocotyls, but flg22 cannot. 7. Images are randomly taken from different fields. Five images are usually collected from one plant. Silencing of GFP-fABD2 expression is common in pavement cells, especially in 3-weekold rosette leaves. In order to obtain enough samples to analyze, preparing extra plants is always recommended. 8. A fixed exposure time and gain setting are selected such that single actin filaments can be seen but the pixel intensities of higher-order actin structures are not saturated. 9. The Gaussian blur processing step needs to be performed before making the image projection. The high-bandpass filter processing step occurs after making the image projection. Skeletonizing images reduces the contribution of large structures such as actin bundles; we do not apply skeletonization steps for density analysis.

270

Bingxiao Wang et al.

10. Threshold settings are set to include all actin filaments, and then images are converted to binary black and white images. The same threshold settings are applied to all images collected from one experiment. 11. At least 50 images are required for skewness and density analysis. 12. When calculating the density value of actin in epidermal cells from dark-grown hypocotyls, fields of cells are cropped into small micrographs because the original VAEM images often contain empty spaces that will affect the result (Fig. 2a). The same size crop box is used for all images collected from one experiment. The box size used in our lab is 300 pixels × 130 pixels (length × width). 13. The same size images are required for skewness and density macros. References 1. Ngou BPM, Ding P, Jones JDG (2022) Thirty years of resistance: Zig-zag through the plant immune system. Plant Cell 34(5):1447–1478 2. Macho AP, Zipfel C (2014) Plant PRRs and the activation of innate immune signaling. Mol Cell 54(2):263–272 3. Melotto M, Zhang L, Oblessuc PR, He SY (2017) Stomatal defense a decade later. Plant Physiol 174(2):561 4. Li J, Blanchoin L, Staiger CJ (2015) Signaling to actin stochastic dynamics. Annu Rev Plant Biol 66:415–440 5. Henty-Ridilla JL, Shimono M, Li J, Chang JH, Day B, Staiger CJ (2013) The plant actin cytoskeleton responds to signals from microbeassociated molecular patterns. PLoS Pathog 9(4):e1003290 6. Henty-Ridilla JL, Li J, Day B, Staiger CJ (2014) Actin Depolymerizing Factor4 regulates actin dynamics during innate immune signaling in Arabidopsis. Plant Cell 26:340–352 7. Li J, Henty-Ridilla JL, Staiger BH, Day B, Staiger CJ (2015) Capping protein integrates multiple MAMP signalling pathways to modulate actin dynamics during plant innate immunity. Nat Commun 6:7206

8. Li J, Staiger CJ (2018) Understanding cytoskeletal dynamics during the plant immune response. Annu Rev Phytopathol 56:513–533 9. Shimono M, Higaki T, Kaku H, Shibuya N, Hasezawa S, Day B (2016) Quantitative evaluation of stomatal cytoskeletal patterns during the activation of immune signaling in Arabidopsis thaliana. PLoS One 11(7):e0159291 10. Zou M, Guo M, Zhou Z, Wang B, Pan Q, Li J, Zhou J-M, Li J (2021) MPK3- and MPK6mediated VLN3 phosphorylation regulates actin dynamics during stomatal immunity in Arabidopsis. Nat Commun 12(1):6474 11. Sheahan MB, Staiger CJ, Rose RJ, McCurdy DW (2004) A green fluorescent protein fusion to actin-binding domain 2 of Arabidopsis fimbrin highlights new features of a dynamic actin cytoskeleton in live plant cells. Plant Physiol 136:3968–3978 12. Higaki T, Kutsuna N, Sano T, Kondo N, Hasezawa S (2010) Quantification and cluster analysis of actin cytoskeletal structures in plant cells: role of actin bundling in stomatal movement during diurnal cycles in Arabidopsis guard cells. Plant J 61:156–165

Chapter 22 Live-Cell Imaging of Cytoskeletal Responses and Trafficking During Fungal Elicitation Amber J. Connerton, Stefan Sassmann, and Michael J. Deeks Abstract Understanding the mechanisms driving plant defense responses holds the promise to provide new means to reinforce plant defense both through agrochemicals and targeted genetic improvement. The capability to quantify impacts of phytopathogens on subcellular dynamics is particularly important when elucidating the role of specific virulence mechanisms that make contributions toward infection success but do not individually alter disease outcome. Acquiring these data requires an investigator to achieve the successful handling of both plant and microbe prior to observation and an appreciation of the challenges in acquiring images under these conditions. In this chapter we describe a protocol to support the observation of cytoskeletal dynamics surrounding sites of fungal interaction, specifically the powdery mildew Blumeria graminis f.sp. hordei on the surface of Arabidopsis thaliana. Furthermore, we also describe a procedure to expose etiolated (dark-grown) hypocotyls to a molecular pattern to activate defense responses in the absence of a phytopathogen with the aim of observing localized actin-dependent trafficking. Key words Actin, Arabidopsis, Powdery mildew, Blumeria, Bgh, Etiolated hypocotyls, Formin, Confocal microscopy

1

Introduction One of the earliest observations made in the field of plant cell biology was the corruption of plant tissues by pathogenic fungi. Robert Hooke’s Micrographia is considered the foundation of cell theory for multicellular organisms, but it also contains the first use of a microscope to illustrate a fungal disease of plants, specifically the eruption of spores from rose leaves infected with a phytopathogenic rust. The resistance of plants to diseases caused by rusts, mildews, and blasts has been linked to the functions of the plant actin cytoskeleton [1–3]. The partnership between actin and immunity stretches beyond the laboratory to agriculture, where at least one disease susceptibility allele, mlo, has been associated with the modulation of actin dynamics. Wild-type Mlo of barley appears to constrain actin dynamics during defense responses, and its

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_22, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

271

272

Amber J. Connerton et al.

function is required for powdery mildew invasion of host tissues [1]. Naturally occurring mlo loss-of-function alleles have become common additions to the genotypes of spring barley cultivars and provide an exemplar for engineering basal resistance through cell biology. Gene editing technologies, and their growing acceptance in law and public opinion, mean that the opportunities to engineer improvements to defense are now limited by our understanding of the signaling and molecular dynamics underpinning the plant immune system. If we were to understand how virulent diseases suppressed the defense activities of the actin cytoskeleton at the molecular genetic level, we would have a list of target genes suitable for allele replacement strategies to improve crop resilience. Hypothesis testing requires reliable quantitative data to compare actin dynamics between genotypes and stimuli. This chapter describes the procedures for growing, infecting, and imaging fluorescently labeled live Arabidopsis leaf epidermal cells with powdery mildew to provoke a non-host immune response. It also describes a second procedure to elicit and image pathogen-responsive trafficking. Both methods have been used extensively to understand FORMIN-driven actin dynamics at fungal interaction sites [4]. The methods deliberately include instructions for high-quality plant care as this is the most important factor in producing reliable and reproducible results at the microscope.

2

Materials

2.1 Bulk Growth of Bgh Powdery Mildew

1. Fresh Blumeria graminis f. sp. hordei conidiospores. 2. Golden Promise Hordeum vulgare (barley) seed. 3. A dedicated infection growth cabinet with 16 h photoperiod and approximately 120 mol m-2 s-1 photosynthetically active radiation (PAR). 4. A “clean” cabinet for healthy barley growth to provide material for infection. 5. Vacuum formed 24 × 5 cm × 5 cm growth cells. 6. Two propagator growth trays (without basal holes). 7. Two propagator lids. 8. Levington M2 compost. 9. Vermiculite.

2.2 Bgh Inoculation of Arabidopsis Leaves

1. 9 cm Petri dish. 2. Whatman filter paper. 3. 3 M micropore tape. 4. Dissection scissors.

Cytoskeletal Responses During Fungal Elicitation

273

5. Forceps/tweezers. 6. Artist’s paintbrush. 7. Black craft card. 8. Parafilm. 9. Aluminum foil. 2.3 Mounting BghInfected Arabidopsis Leaves

1. Standard glass microscopy slides. 2. Vaseline. 3. 3 M micropore tape. 4. Cotton buds. 5. Dissection scissors. 6. 22 × 50 mm coverslips. 7. Forceps/tweezers. 8. Flutec PP11 perfluorocarbon (F2 Chemicals, UK).

2.4 Confocal Observation of Appressorium Interaction Sites

1. Laser-scanning microscope.

confocal

microscope

or

spinning

2.5 Growing Elongated Hypocotyls for Elicitation with PAMPs

1. Sodium hypochlorite (bleach), approximately 5–12% w/v.

disk

2. 10 M hydrochloric acid. 3. Sterile (autoclaved) water. 4. 96 square well boxes. 5. 250 μl PCR tubes. 6. Laminar flowhood or biological safety cabinet. 7. Half-concentration Murashige and Skoog (MS) growth medium: 0.8% w/v agar, 2.2 g of MS salts, adjusted to pH 5.7 using KOH. 8. Aluminium foil.

2.6 Exposing DarkGrown Hypocotyls to PAMP Elicitors

1. Crab/shrimp chitin granules (Merck/Sigma Aldrich). 2. Driselase (Merck/Sigma Aldrich). 3. Trichoderma viride chitinase (Merck/Sigma Aldrich). 4. 100 mM sodium phosphate pH 6.1. 5. Glycerol.

2.7 Mounting Hypocotyls for LiveCell Imaging

1. MP1 molecular pattern cocktail (see Subheadings 2.6 and 3.6 to produce MP1) or an alternative elicitor. 2. 100 mM pH 7.5.

2-(N-morpholino)ethanesulfonic

acid

(MES)

274

Amber J. Connerton et al.

3. Standard glass microscopy slides. 4. Vaseline. 5. 22 × 50 mm coverslips. 6. Forceps/tweezers. 7. 3 M micropore tape.

3

Methods

3.1 Bulk Growth of Bgh Powdery Mildew

1. For the majority of our experiments, we use a species of powdery mildew, Blumeria graminis f. sp. hordei (Bgh), UK isolate, CC/133 from NIAB Cambridge, as an antagonist to elicit plant immune responses. Isolate CC/133 can be substituted for any field isolate, but some aspects of host interaction may vary according to genotype. To keep a fresh supply of 4-weekold plants for propagating the mildew, new barley plants are sown weekly in the “clean” cabinet with 16 h photoperiod, 200 mol mol m-2 s-1 PAR, 20 °C, 60% RH (see Note 1). 2. Bgh is cultivated on Golden Promise Hordeum vulgare (barley) in growth cabinets at 17 °C, 60% relative humidity (RH). 3. To achieve small-scale weekly culturing of Bgh, separate 4 modules from a 24-module seed tray, and fill with Levington M2 compost mixed with vermiculite at a 3:1 ratio (three scoops of compost to one scoop of vermiculite). Shake the modules to ensure there are no air pockets and top up with compost mixture as needed. 4. Soak the compost in water and place four barley seeds in a square on the surface of the compost to ensure even spacing. Push each seed into the growing media at a depth of twice the diameter of the seed. 5. Place the modules on a tray and move to a plant growth cabinet dedicated to infection-free growth. 6. Check water daily (pots should be heavy to lift but not saturated with water). Consistent watering and mild conditions promote healthy growth to support the subsequent biotrophic interaction. 7. The 4-week-old barley plants are added weekly to the existing collection of infected plants used to cultivate the Bgh. To infect the new plants, shake the previous week’s infected plants over the new ones to distribute the conidiospores (spores generated for asexual reproduction). 8. Place a propagator lid on top of the freshly infected barley to retain humidity, while the spore attachment and host penetration process begins.

Cytoskeletal Responses During Fungal Elicitation

275

9. At this point the need to preserve high humidity outweighs the damage caused by overwatering. We suggest maintaining 2.5 cm (1 inch) depth of water in the tray containing the plants. 10. Remove the propagator lid 72 h after the inoculation to prevent premature senescence and decomposition of older leaves. 11. Older plants dominated by dead leaves should be removed weekly (approximately after 2 weeks residency) as dead material does not support Bgh and encourages saprophytic fungal growth. As with all aspects of phytopathogen handling, transport, and experimentation, please consult local guidance on compliance with biosafety legislation to dispose of infected plant material (see Note 2). 12. The plant waste must then be safely disposed of through autoclaving without risking the contamination of healthy plants. Prior to autoclaving, the plants should be contained in bags to minimize spore release. 3.2 Bgh Inoculation of Arabidopsis Leaves

Typically, we infect detached Arabidopsis thaliana leaves from 2- to 4-week-old plants, carefully standardizing plant age and developmental identity within each experiment. We use typical long-day growth conditions (16-h light, 21 °C day, 19 °C night, 55/60% RH) to produce healthy plants for infection. Once Arabidopsis plants are at the target age, leaves can be cut and prepped for infection (see Note 3). 1. First, cut the filter paper to the size of the Petri dish, and layer two pieces at the bottom of the dish. 2. Soak with water to keep humidity levels high for overnight incubation (pour off any excess). 3. Carefully cut the leaf petioles using dissection scissors making sure to leave as much stem as possible. This will enable easy manipulation of leaves with tweezers. When choosing a leaf developmental stage, avoid taking the oldest or youngest leaves. The youngest leaves have the smallest cells relative to Bgh, and this can obfuscate localized subcellular responses, while the oldest leaves can be undergoing senescence and are more likely to have lower transgene expression. 4. Carefully place with the adaxial surface facing upward, and secure in place with micropore tape on the stems; two rows of four leaves will comfortably fit in one Petri dish (Fig. 1a). Labels can be written (with care) on the micropore tape. 5. We use spores from barley plants 7 days after their original inoculation (Fig. 1b). The freshest conidiophores (colonies of conidiospores) on infected barley leaves are snow-white in color. Older, brown conidiophores should be avoided to ensure high germination rates.

276

Amber J. Connerton et al.

Fig. 1 Blumeria graminis leaf inoculation. (a) Petri dish 9 cm with two layers of wet filter paper, with Arabidopsis thaliana leaves secured in place with micropore tape. To fit eight leaves, two rows can be arranged. (b) Barley leaves 7 days post infection ready for spore collection with paint brush as shown. This image was taken before knocking the leaves. The spores will fall off with the slightest movement as is evident by the small cluster seen on the card. With this in mind, be very careful when moving the tray to access the leaves with the most spores. (c) Close up of snow-white Bgh spores after gentle tapping with the paintbrush. (d) Final stage of inoculating Arabidopsis leaves, using the paint brush to gently brush spores onto the prepared leaves.

6. Gently knock the barley leaves with a healthy spore population with a paintbrush via a tapping motion while holding the black craft card underneath to catch the falling spores. The spores should look white against the black card (Fig. 1c). A fully infected barley leaf provides enough spores to infect approximately four Arabidopsis leaves. 7. Dust the collected spores over the Arabidopsis leaves in the Petri dish (Fig. 1d). It is good practice to aim for the middle of leaves, so you always know which section of leaf to take later for imaging. 8. Once all the leaves have been dusted, wrap the sides of the plates in Parafilm to maintain humidity, and cover the stack of plates in aluminum foil. 9. Place in the mildew infection cabinet to allow the spores to germinate. Reducing exposure to light contributes to preserving fluorescent transgenic markers, as growth lights could lead to bleaching and oxidative stress in the detached leaves. 3.3 Mounting BghInfected Arabidopsis Leaves

At approximately 24 h after initial contact with Bgh, the infected leaves can be taken out of the growth cabinet and mounted for observation under the microscope. Developed appressoria and host defense sites should be prevalent by 16 h post infection. Forty-eight hours post infection will yield a greater proportion of mature host defense sites, but younger (developing) sites will be rare. Ensure that the work area is dust- and grease-free as coverslips will contact the worksurface during this procedure. Pre-clean work areas with detergent, water, and 70% ethanol. Alternatively, fresh foil can be used as a clean surface for assembling slides. Handle slides and coverslips from the edges.

Cytoskeletal Responses During Fungal Elicitation

277

Fig. 2 Mounting inoculated leaf tissue. (a) Micropore tape was secured to the desk and a coverslip placed on top, making sure to leave enough of a window for the sample (b) Petroleum jelly box drawn on slide with a cotton swab to create a seal when all the slide components of slide are sandwiched together. (c, d) Dissection scissors were used to cut a small square of tissues from the inoculated leaves. (e) All components required to assemble the slide. First the cover slip with the micropore tape was placed on the desk with the sticky side of the tape facing up. Mounting media was added, and the leaf tissue square was placed on top of the coverslip with the waxy cuticle (and germinated spores) facing down. (f) The slide was slowly lowered onto the coverslip making sure the leaf tissue was within the petroleum jelly box. (g) Assembled slide after the micropore tape was wrapped around to hold the assembly together

1. To mount the leaf tissue, first take a glass slide and draw a box with a cotton bud sparingly covered with petroleum jelly (Vaseline) to create the seal (Fig. 2b). 2. Next cut a small square (approximately 5 × 5 mm) of the leaf with dissection scissors from the middle where the mildew was dusted, avoiding the central leaf vein (Fig. 2c, d), as the vein impedes mounting and sample focusing due to the resulting long and varied working distance from the objective lens. 3. Add around 70 μL of the mounting medium of your choice to the center of the coverslip. Using tweezers place the leaf square on a 22 × 50 mm coverslip with the adaxial surface facing downwards (Fig. 2e). 4. Align the slide, and carefully lower it down onto the coverslip using the weight of the slide to assist the process (Fig. 2f). Mounting the specimen in this way reduces the number of bubbles that interfere with imaging. 5. Finally, secure the coverslip in place with micropore tape (Fig. 2g). Optionally you can pre-prepare coverslips with micropore tape so that the ends of the tape strips are left sticky side up at the edges of the coverslip (Fig. 2a). This minimizes crushing of the sample when securing the coverslip to the slide (Fig. 2f).

278

Amber J. Connerton et al.

6. Optimization of mounting media used in live-cell imaging can be a simple remedial approach that can significantly improve the optical quality of images. Using perfluoroperhydrophenanthrene (PP11) as a mounting medium has been shown to reduce light scattering and improve fluorescence signal compared to mounting the samples in water [5] (see Note 4). 3.4 Confocal Observation of Appressorium Interaction Sites

For the most part, imaging constraints, challenges, and optimization strategies applied to Bgh interaction sites are equivalent to other plant live-cell imaging scenarios. Exact settings will vary considerably according to microscope, constructs, and processes under observation. Host response sites surrounding fungal appressoria are rarely within a single focal plane or in close contact with the coverslip (see Note 5). Autofluorescence from multiple tissue layers usually limits the use of epifluorescence microscopy to pre-screening samples before the use of higher specification modalities for quantitative imaging. The majority of our data for hypothesis testing are generated using laser scanning confocal microscopy and spinning disk confocal microscopy. Figure 3 shows an example sample of captured data. Bghchallenged cells generate a cell wall apposition (CWA) that is approximately 5 μm in both depth and diameter (Fig. 3b). This consists of autofluorescent material deposited between the plasma membrane and cell wall and can under some circumstances present a significant barrier to imaging labeled features at the response site.

Fig. 3 Live-cell imaging of plant cells 48 h post Bgh infection. (a) Transmission image of germinated Bgh conidiospore. The spore has produced a short primary germ tube (PGT) and large appressorial (or “secondary”) germ tube (AGT). The tip of the “hockey stick” AGT houses the appressorium (labeled as App.) that drives a penetration peg into the host cell. This image is atypical as all labeled features are present within the same focal plane. (b) Laser scanning confocal imaging of actin labeled with GFP-Lifeact (green). This image is a maximum projection of 20 slices with a total Z-depth of 14 μm and represents a typical depth of imaging required to capture host cell cytoplasmic features surrounding the site of fungal penetration. GFP emission is captured between 505 and 530 nm with excitation at 488 nm. The magenta channel uses an excitation wavelength of 543 nm and an emission window of 580–650 nm. White arrow indicates the autofluorescent cell wall apposition (CWA), which is also referred to as a papilla. Yellow lines show cell boundaries. (c) Merged image of a (processed to isolate object edges) and b to show alignment of fungal and plant features. Scale bar is 20 μm

Cytoskeletal Responses During Fungal Elicitation

279

It is essential to compare unfamiliar or previously uncharacterized CWA-localized labels with control Bgh infections (for instance, using wild-type host material infected with Bgh to compare with infected plants expressing fusion proteins) to prevent misidentification of autofluorescent structures as labeled features. CWAs are produced in response to cell penetration attempts by the appressorium, a specialized fungal structure associated with the appressorial germ tube (AGT; Fig. 3a). Primary germ tubes (PGTs) can also stimulate lesser CWA-like responses and cause increased autofluorescence. The thinner PGTs are produced rapidly and in higher numbers than the later developing AGTs, and suboptimal infection conditions often result in PGT production without subsequent AGT formation. 3.5 Growing Elongated Hypocotyls for Elicitation with PAMPs

Contacts between hyphae and plant cells are asynchronous, and appressoria can obstruct imaging of the plant cytoplasm directly beneath. Figure 3b demonstrates that light scattering increases, while signal strength decreases in locations where the plant cell cortex is imaged beneath fungal structures. TIRF/VAEM of the plant cell surface is rarely possible in areas where fungal structures prevent coverslip contact. Some insight into pathogen-induced actin dynamics and actin-dependent trafficking can instead be achieved by stimulating host responses using molecular methods [4, 6]. Various molecules produced and released by microbial phytopathogens, such as bacterial flagellin and fungal chitin, have been termed pathogen-associated molecular patterns (PAMPs) due to their ability to activate host receptors and prime defense responses. PAMPs alone will stimulate specific “global” actin network changes [6]. Moreover, localized actin-dependent trafficking can be activated by PAMP stimulation [7]. We have developed a simple workflow to elicit defense responses including actin-dependent trafficking from Arabidopsis dark-grown hypocotyls. Etiolated hypocotyls are a convenient tissue for TIRFM/VAEM due to their long, elongated epidermal cells that make good contact with the coverslip. The walls and cuticles of these cells are relatively thin, which also aids TIRFM. 1. To sterilize Arabidopsis seed for this procedure, expose the seeds to chlorine gas for 5 h in a desiccator by mixing 100 mL bleach with 3 mL of HCl (see Note 6). 2. After sterilization, suspend seeds in sterile Milli-Q® water and leave to rest for 7 days in the dark at 4 °C. 3. We use sterile 96 square well boxes with lids to grow microbefree Arabidopsis. Each cell is approximately 2.5 mL in volume (Fig. 4a; see Note 7). Under a laminar flow hood, open an autoclaved 96 square well box and place 2–4 sterilized PCR tubes per line to be imaged into the wells at least 2 rows in from the outer edge (Fig. 4b).

280

Amber J. Connerton et al.

Fig. 4 Growth and mounting of etiolated hypocotyls. (a) Pre-sterilized 96 cell plastic boxes are used to support 250 μl PCR tubes containing 100 μL of solidified plant growth medium. (b) Sterilized seed are suspended in water, imbibed at 4 °C for 5 days, and are then pipetted onto the surface of the agar. (c) Following 5 days of dark growth, the seedlings have developed elongated hypocotyls and are beginning to emerge from the top of the tubes. Elicitor solution can be added to the volume of the tube above the agar. (d) Hypocotyls are mounted by using fine forceps to lift the hypocotyls from the tubes by their base. When submerged in elicitor solution, the seedlings can also be carefully lifted by the cotyledons; however the cotyledons can detach if this is attempted with dry seedlings. Black arrow indicates mounted seedling

4. To each PCR tube, add 100 μL of molten 1/2MS growth medium. 5. Allow the growth medium to cool and solidly in the PCR tubes to avoid exposing the seeds to heat shock. 6. While waiting for the 1/2MS to cool, add in sterile Milli-Q® water to the central wells to maintain the humidity within the box. 7. Carefully pipette two to five seeds onto the agar surface in each tube (Fig. 4b). A small quantity of water on the surface will cause no issues at this stage. 8. Tap the box firmly on a flat surface to make sure all seeds are in contact with the growth medium. 9. Finally, close the lid and wrap in a double layer of aluminum foil to exclude all light. Place the box into a growth cabinet for 5 days. 3.6 Exposing DarkGrown Hypocotyls to PAMP Elicitors

We use a cocktail of fungal material and enzymes named MP1 to approximate pre-exposure to cell wall irritating fungal activity. In addition to the PAMP chitin, MP1 also contains a highly dilute mix of fungal-derived cell wall digesting enzymes (driselase) to promote the release of damage-associated molecular patterns (DAMPs). The MP1 method was developed to image in TIRFM the actindependent trafficking of Arabidopsis FORMIN4-GFP which is only expressed in aboveground tissue following exposure to Bgh [4]. Other specific elicitors can be easily substituted for MP1: 1. To produce MP1, suspend chitin granules to a concentration of 100 μg/mL in a 0.004% w/v driselase solution. 2. Resuspend 40 mg of lyophilized chitinase stock in 400 μL of 100 mM sodium phosphate (pH 6.1) and 400 μL glycerol.

Cytoskeletal Responses During Fungal Elicitation

281

3. Activate the suspension by adding the Trichoderma chitinase to a final concentration of 0.05 mg/L. Leave stirring at room temperature for 1 h to allow the endochitinase to generate chitin oligomers. 4. Freeze 1 mL aliquots at -20 °C. Bring individual aliquots to room temperature before use. 5. Fully submerge Arabidopsis hypocotyls in MP1 by using a pipette to fill the PCR tubes. Incubate in the dark for 4 h prior to imaging to elicit a response (see Note 8). 3.7 Mounting Hypocotyls for LiveCell Imaging

The elicitor-exposed hypocotyls are mounted using a similar procedure to that described in Subheading 3.3. The protocol below highlights key differences. 1. We use mounting medium designed to maintain continual elicitation during imaging. Produce a fresh working stock on the day of imaging containing 10% v/v MP1 and 10 mM MES (see Note 9). 2. Pipette 70 μL of mounting medium on to a coverslip with pre-attached micropore tape as described in Subheading 3.3. 3. Use fine forceps to gently remove hypocotyls by their base from the PCR tubes. Transfer two to the coverslip with mounting medium. 4. Prepare a Vaseline seal around the slide area for the sample (see Subheading 3.3, step 1). 5. Attach the slide to the coverslip, taking care not to apply excessive pressure to the slides to avoid compressing the epidermal cells (see Note 10). 6. We only quantify samples under the microscope (Fig. 5) within a 30 min window following mounting to minimize the potential impact of confounding stresses from the mounting process (see Note 11).

4

Notes 1. This interaction is biotrophic and requires healthy host tissue to support fungal development. The powdery mildew strain is kept in a separate, isolated growth cabinet, and when experiments or maintenance of the isolate is conducted, the researcher will avoid (for that working day) all subsequent contact with healthy non-infected plants. The researcher will also wear a dedicated laboratory coat and nitrile gloves to be disposed of immediately after finishing any procedure at the mildew growth cabinet. A spray bottle of 70% v/v ethanol can be stationed at the incubator to be used to decontaminate the

282

Amber J. Connerton et al.

Fig. 5 Example of imaged hypocotyls following exposure to elicitor. (a) Spinning disk microscopy image of hypocotyls co-expressing GFP-Lifeact (green) and FORMIN4-tdTomato (magenta) following 4 h of exposure to MP1. The positions of cell borders are shown in yellow. (b) Enlarged view of area in a marked by a dashed box. FORMIN4-tdTomato has accumulated in a deposit and shows a punctate distribution. (c) An enlarged view of GFP-Lifeact within the same area shows a coincident fine F-actin structure (indicated by white arrow). Scale bar is 10 μm

area after access to the cabinet. These precautions ensure that plants not intended for infection are kept uncontaminated and the strain’s isolation is maintained. Field strains of Bgh and other powdery mildews can invade from the outside environment, and this requires vigilant protection, particularly for barley plants before their entry into the designated mildew cabinet. 2. In the UK, an endemic field strain of powdery mildew is unlikely to require licensing, but there may be some exceptions, especially if the strain is highly virulent and invasive or experiments are at risk of cross-contaminating work with other phytopathogens. Our local procedure for disposal of Bgh-infected barley is treatment in an autoclave at 121 °C for 45 min. 3. It is important to check plants daily to keep the growth media damp to the touch but not drenched as over- (or under-) watering can cause stress responses that can affect the host– pathogen interaction. 4. PP11 is a perfluorocarbon with high saturation of atmospheric gases and a low surface tension, low enough to infiltrate the stomatal aperture. This results in the spongy mesophyll airspaces becoming occupied with PP11, without having any toxic or hypoxic effects. The refractive index of PP11 more

Cytoskeletal Responses During Fungal Elicitation

283

closely matches the indices of the vacuole and cytoplasm than air, reducing optical aberrations and improving image quality for both transmission and fluorescence microscopy [5]. Although Bgh does not enter the deep leaf space, its presence on the surface can trap air bubbles when using water-based mounting media, and we have found the low surface tension properties of PP11 to be advantageous when used with Bgh-infected leaf surfaces. 5. This does not preclude total internal reflection microscopy (TIRFM) or variable angle epifluorescence microscopy (VAEM), but the random orientations of pathogen contact sites and appressoria mean that large proportions of instrument time can be spent locating suitable sites when using these modalities, and typically only small sectors of cell cortex are correctly aligned for imaging. 6. Limiting chlorine exposure to 5 h prevents reductions in germination efficiency. Other sterilization strategies are equally valid; however, we find the chlorine method convenient when sterilizing many lines in parallel. It should be noted this method can reduce germination frequency if used to sterilize a small number of seeds. When using this method, be cautious of exposure to fumes, and perform all preparation and exposure stages in a fume hood with appropriate PPE. 7. If these boxes are unavailable, it is possible to use old tip boxes which are still sealable, but these can be less stable in supporting PCR tubes. Most importantly, all the materials need to be autoclavable for repeated use as we pre-sterilize this equipment before each experiment. 8. The incubation time of 4 h may be varied according to the nature of the experiment. It is best practice to stagger the time intervals when MP1 is added to each tube to ensure imaging occurs at an exact standard time point (e.g., 4 h) after elicitor contact. 9. For negative control samples, we replace MP1 with a mock solution that contains the appropriate concentration of sodium phosphate and glycerol but lacks driselase, chitinase, and chitin. 10. Dark-grown hypocotyl epidermal cells undergoing rapid cell expansion are more delicate than leaf epidermal cells as they have minimal cell wall thickness and are under high turgor. 11. In addition to standardizing the time window for observation, it is also necessary to select hypocotyls of equivalent developmental age and to image within a defined zone along the hypocotyl length. Cells at the base of the hypocotyl expand earlier than cells close to the cotyledons. We therefore image epidermal cells within the central third of the hypocotyl length.

284

Amber J. Connerton et al.

Acknowledgments The authors would like to thank James Sheffield for his support in producing Figs. 1, 2, and 4. This work was supported by the BBSRC (BB/T008741/1, BB/M024172/1). References 1. Miklis M et al (2007) Barley MLO modulates actin-dependent and actin-independent antifungal defense pathways at the cell periphery. Plant Physiol 144(2):1132–1143 2. Zhang B et al (2017) TaADF4, an actindepolymerizing factor from wheat, is required for resistance to the stripe rust pathogen Puccinia striiformis f. sp. tritici. Plant J 89(6): 1210–1224 3. Jarosch B et al (2005) RAR1, ROR1, and the actin cytoskeleton contribute to basal resistance to magnaporthe grisea in barley. MPMI 18(5): 397–404 4. Sassmann S et al (2018) An Immune-Responsive Cytoskeletal-Plasma Membrane Feedback Loop in Plants. Current Biology 28:2136–2144

5. Littlejohn GR et al (2014) An update: improvements in imaging perfluorocarbon-mounted plant leaves with implications for studies of plant pathology, physiology, development and cell biology. Front Plant Sci 5:140 6. Henty-Ridilla JL et al (2013) The plant actin cytoskeleton responds to signals from microbeassociated molecular patterns. PLoS Pathog 9(4):e1003290 7. Underwood W, Somerville SC (2013) Perception of conserved pathogen elicitors at the plasma membrane leads to relocalization of the Arabidopsis PEN3 transporter. Proc Natl Acad Sci U S A 110(30):12492–12497

Chapter 23 Visualization and Quantification of the Dynamics of Actin Filaments in Arabidopsis Pollen Tubes Qiaonan Lu, Xiaonan Liu, Xiaolu Qu, and Shanjin Huang Abstract The actin cytoskeleton plays an essential role in the regulation of polarized pollen tube growth, and its functions are dictated by its spatial organization and dynamics. Here we describe an assay to monitor the dynamics of actin filaments decorated with Lifeact-mEGFP in Arabidopsis pollen tubes using spinning disk confocal microscopy and measuring the parameters associated with their dynamics. The method allows us to assess the dynamics of actin filaments in growing Arabidopsis pollen tubes. Key words Actin, Actin dynamics, Lifeact-mEGFP, Arabidopsis thaliana, Pollen tube

1

Introduction Pollen tube formation is an essential step during sexual reproduction in flowering plants, whose growth is subject to tight regulation and has been subject to intensive scrutiny for decades. Pollen tube growth is astonishingly rapid, with a growth rate that can reach up to 1 cm/h in the style [1]. To support such rapid tip growth, the pollen tube requires the presence of an active intracellular transport system enabling the delivery of materials necessary for cell wall synthesis and membrane expansion to the tip. Within this framework, a dynamic actin cytoskeletal system is absolutely required for rapid polarized pollen tube growth by choreographing exo- and endocytotic vesicle traffic [2–4]. Decades of studies have shown that the actin cytoskeleton assumes a unique spatial distribution in pollen tubes [5–12]. Specifically, actin filaments are arrayed into longitudinally oriented actin cables in the shank, which act as the railway tracks for driving vesicle traffic and organelle movement to generate a reverse fountain pattern of cytoplasmic streaming

Supplementary Information The online version contains supplementary material available at https://doi.org/ 10.1007/978-1-0716-2867-6_23. Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_23, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

285

286

Qiaonan Lu et al.

[13, 14]. Whether there are actin filaments, and also how they are organized at the tip, has been subject to much debate over the years. This is likely because the actin filaments are so dynamic that is hard to monitor in real time and analyze quantitatively. With the introduction of appropriate actin markers in combination with various actin defective mutants, and the employment of high spatiotemporal resolution imaging technology, we are able to reveal the organization and dynamics of actin filaments at the pollen tube tip in unprecedented detail. Here we present assays to directly visualize the dynamics of actin filaments decorated with LifeactmEGFP [15] and quantify their dynamic parameters in Arabidopsis pollen tubes growing in vitro. This method is adapted from our previously published methods [16–25].

2 2.1

Materials Plant Materials

2.2 Materials and Chemical Reagents for Pollen Germination

1. Arabidopsis (Col-0) expressing Lat52::Lifeact-mEGFP, [22]. 1. 9 cm
9 cm glass Petri dish (NORMAX, catalog number: 5058546). 2. 12 cm
12 cm square Petri dish. 3. Microscope slide. 4. Parafilm. 5. Stock solution of pollen germination medium: 1% (w/v) H3BO3, 100 mM CaCl2, 100 mM Ca(NO3)2, 100 mM MgSO4. Make all stocks with Milli-Q water; keep them at 4  C and used within 6 months. 6. Pollen germination medium (PGM): 18% (w/v) sucrose, 0.01% (w/v) H3BO3, 1 mM CaCl2, 1 mM Ca(NO3)2, 1 mM MgSO4, pH 6.9–7.0, adjusted with 1 M KOH (see Note 1). To make solid PGM, agarose at 0.8% (w/v) was added and melted in a microwave oven (see Note 2).

2.3

Microscopy

1. Coverslips. 2. Glass slides. 3. Glass bottom dish (In Vitro Scientific, catalog number: D35-20-1-N). 4. An inverted microscope (Olympus IX83) equipped with spinning disk head (Yokogawa CSU-X1) and electron multiplying CCD camera (Andor iXON). 5. 100
Universal Plan Super Apochromat (UPLSAPO 100
, numerical aperture of 1.4).

Objective

Actin Dynamics in Arabidopsis Pollen Tubes

2.4 Data Analysis Software

287

1. ImageJ (https://imagej.nih.gov/ij). 2. Excel (Microsoft). 3. IBM SPSS Statistics (https://www.ibm.com/cn-zh/ products/spss-statistics) or other software for statistical data analysis.

3

Methods

3.1 Preparation of Arabidopsis Pollen Germination Medium (PGM)

1. Prepare individual stocks with Milli-Q water (see Subheading 2.2, item 5). 2. Prepare 100 mL liquid pollen germination medium (see Subheading 2.2, item 6); transfer 30 mL liquid PGM to a 50 mL clean conical flask; add 0.8% agarose to make solid PGM. Melting agarose with microwave oven, spread the medium on two 9 cm
9 cm glass Petri dishes. For one round 9 cm glass culture dish, preparing 15 mL PGM is enough (Fig. 1a). 3. Cut solid PGM into 1 cm
1 cm pieces. Place the medium pieces onto microscope slides (Fig. 1b). 4. Put the microscope slide into humid 12 cm
12 cm square Petri dish. Place two microscope slides underneath to avoid water contamination (Fig. 1c).

3.2 Pollen Germination and Pollen Tube Growth

1. Arabidopsis thaliana plants expressing Lat52::Lifeact-mEGFP were cultured in soil in a growth chamber under 16 h light/8 h dark photoperiod at 22  C. Add an appropriate amount of fertilizer to facilitate the growth of Arabidopsis plants, and water them regularly and properly to obtain healthy plants having well-developed flowers. 2. Harvest fresh opened Arabidopsis flowers from Lat52::LifeactmEGFP transgenic plants during the period from 9:30 am to 12:00 am (see Note 3). For the observation of actin in pollen tubes, collect pollen grains from ten freshly opened Arabidopsis flowers. 3. Immediately dip the fresh flowers onto the PGM block gently to spread pollen grains (Fig. 1b). 4. Culture pollen grains in an incubator at 28  C for 2–3 h under high humidity. 5. Move the slide with the solid PGM slice to the microscope stage to check pollen germination and pollen tube growth under a microscope. After culture for 2–3 h under 28  C, the average length of pollen tubes would reach about 200 μm (Fig. 1d), which would now be ready for microscopic visualization and image acquisition.

288

Qiaonan Lu et al.

Fig. 1 Preparation of pollen germination medium and Arabidopsis pollen germination. (a) Preparation of pollen tube germination medium. Cut the medium into a 1 cm
1 cm small piece with a blade. (b) Gently dip the flowers on the medium to spread the Arabidopsis pollen grains. (c) Preparation of a humid dish to culture the pollen at 28  C. (d) Image of pollen tubes after 2 h of culture at 28  C. Scale bar ¼ 100 μm 3.3 Image Collection Using Spinning Disk Confocal Microscopy

1. To visualize the dynamics of actin filaments, the solid PGM slice is moved into a glass bottom dish with the side holding the germinating pollen facing downward. Add some PGM to keep the level of humidity up (see the whole procedure in Electronic Supplementary Video 1). 2. Move the glass bottom dish onto the microscope stage, and initially observe using a fluorescence microscope at low magnification to find the microscopic field with the pollen tubes of interest (see Note 4). 3. Add a drop of immersion oil onto the top of the 100
UPLSAPO objective, and switch to high-magnification observation.

Actin Dynamics in Arabidopsis Pollen Tubes

289

4. Use a mercury lamp to excite pollen tubes of interest and find the focal plane with the sharpest image of Lifeact-mEGFPdecorated actin filaments by using the fine adjustment control. 5. Acquire high-quality time-lapse Z-stack images with an inverted microscope equipped with a spinning disk head. Excite Lifeact-mEGFP-decorated actin filaments with a 488 nm argon laser and emission filter (Bandpass 525/45). The time-lapse Z-series images are collected with a 512
512 electron multiplying CCD camera using MetaMorph software (version 7.8, https://www.moleculardevices. com/). The Z-step size is set at 0.5 μm, and time intervals between the Z-series images are set at 2 s. The time-lapse Z-stack images are saved as .tiff files (see Note 5). 6. The data file is opened in ImageJ software, and the Z projection image is generated with the “Z project” tool in the “ImageStacks” menu (see the representative z-project time-lapse images in Electronic Supplementary Video 2). 3.4 Analysis of the Overall Polymerization of Apical Actin Filaments from Plasma Membrane at the Pollen Tube Tip Using Kymographs

1. Most quantitative image processing is completed in ImageJ. Firstly, use the Hyperstacks function to separate the t-axis and z-axis to obtain the x-y-z-t image (Fig. 2a). Conduct singlelayer observation for analyzing the dynamics of individual actin filaments (Fig. 2a). 2. Use maximum intensity projected time-lapse images for the analysis of real-time intensity or growth rate of the whole pollen tube (Fig. 2b). In the growing wild-type (WT) pollen tube, a bright actin fringe structure can be visualized within the apical region (Fig. 2b). In pollen tubes with impaired actin polymerization, such as in the prf4 prf5 mutant, the actin fringe is short and dim (Fig. 2c). 3. Next, we generated a kymograph diagram to observe the state of actin filaments at the tip of the pollen tubes [17, 18, 23] (see Notes 6 and 7). 4. A segmented line along the top of the pollen tube was drawn overtime. Adjust the line width to 3 μm by double-clicking the line button. In ImageJ, line width is determined by pixel number. Input the pixel number represented by 3 μm according to the scale (Fig. 3a, 1). 5. Run the Stack ProfileData (http://rsb.info.nih.gov/ij/ macros/StackProfileData.txt) plug-in in ImageJ to obtain the intensity profile along the line in the whole stack (Fig. 3a, 2). 6. Delete non-intensity value, such as axis information, and then save the file as a text file. Import the text file into ImageJ to obtain the kymograph diagram (Fig. 3a, 3, b, a, b). A bright actin fringe structure is observed and is highlighted by the red

290

Qiaonan Lu et al.

Fig. 2 Image processing of actin filaments in pollen tubes. (a) Hyperstack image in the left panel displays every single slice of the pollen tube at each time point. Z-slice images at a single time point were shown in the right panel. Single filaments can be visualized, and the dynamic parameters can be determined via hyperstacks. (b) Maximum intensity projected image displays the fluorescence distribution in the whole pollen tube. Timeseries projected images reveal the actin structure dynamics in the right panel. By tracking the fluorescence along the line drawn in the growth direction of the pollen tube, as shown in the left panel, kymograph images can be made. (c) Maximum intensity projected image of a prf4 prf5 pollen tube. Compared to WT pollen tubes, apical actin structure is severely disrupted in the prf4 prf5 pollen tube

arrows. The real-time growth state of the pollen tube can be determined as well. 7. To quantify the fluorescent intensity or growth rate of the pollen tube, the following steps are conducted to initially identify the growing tip of the pollen tube and arrange the tip in a line. First, set a threshold to cover the pollen region in the kymograph diagram (Fig. 3a, 4). Convert to binary image via “Binary-Convert to Mask” function. 8. Multiply the original kymograph image with mask image to subtract the background intensity using the “Image Calculator” function. To regain the original intensity value in the pollen tube region, divide the resulting image by 255 using the “Math-Divide” function (Fig. 3a, 5). Save the final resulting image as “Text Image.” 9. The real-time growth rate of the pollen tube is measured by counting the movement of the position of the first non-zero gray value pixel (Fig. 3a, 6). 10. To analyze the fluorescent intensity in a specific region, arrange the tip in a line first (Fig. 3b, b, d). Open the text image in

Actin Dynamics in Arabidopsis Pollen Tubes

291

Fig. 3 Analysis of apical actin polymerization using kymograph analysis of pollen tubes. (a). Procedure of doing kymograph analysis. (b). Resulting kymograph images processed by (a). a, kymograph of WT pollen tube in Fig. 2b after processing via (a)3. b, kymograph of WT pollen tube after arranging the tip of pollen tube in a row. c, kymograph of prf4 prf5 pollen tube in Fig. 2c. d, kymograph of prf4 prf5 pollen tube after arranging the tip of the pollen tube in a row. Red arrows indicate the actin fringe structure. e, Fluorescent profiles along the yellow 3-μm-width segmented line drawn in Fig. 2b, c at represented time point. The actin fringe in WT is significantly brighter and wider than that in prf4 prf5 pollen tube. Scale bars represent 5 μm in all images

Excel. One cell represents intensity in one pixel. Clear all cells with 0 value. Delete blank cells, and move the rest of the cells below up to arrange all of the tip in a row. Make the rest of the cells into a matrix so that the file can be imported into ImageJ as a text image. 11. Average the fluorescent value in the region of interest to obtain the real-time fluorescent intensity. For instance, the average apical actin intensity can be obtained by measuring the average intensity within the region that is 5 μm distal from the tip. A detailed intensity profile along the segmented line is illustrated in Fig. 3b, e.

292

Qiaonan Lu et al.

3.5 Analysis of Parameters Associated with the Dynamics of Individual Actin Filaments

X-y-z-t images without z-axis integration are used to analyze the dynamics of an individual actin filament (Fig. 2a). Several parameters associated with the dynamics of individual actin filaments, including maximal filament length, maximal filament life time, elongation rate, depolymerization rate, and severing frequency as well as bundling and de-bundling frequencies, are determined. 1. To determine the maximal filament length and maximal filament life time, a sketch map is drawn to show how the dynamic parameters of individual actin filaments are determined. As shown in Fig. 4a, an actin filament began to polymerize at 0 s and reached at its maximal length of 5 μm at 20 s. Subsequently, the filament began to depolymerize and disappeared at 40 s. Therefore, the maximal filament length is 5 μm, and the maximal filament lifetime, from its birth to disappearance, is 40 s. 2. The elongation rate of this actin filament can be obtained by dividing the elongated length by the duration of actin elongation. In this case, the actin elongation rate is 5 μm/ 20 s ¼ 0.25 μm/s. 3. The depolymerization rate can be obtained by dividing the maximal filament length by the duration of actin depolymerization, which is equal to 5 μm/(40 s–20 s) ¼ 0.25 μm/s. 4. Determination of actin filament severing frequency is performed as shown in Fig. 4b. After the maximal length of actin filament reaches 5 μm at 20 s, it began to be severed at 24 s and 32 s and thus generated two breaks. The severed actin filament disappeared completely at 44 s. Based on the criterion described previously [26], average severing frequency is defined as the number of breaks per length of original mother filament per unit time (breaks/μm/s). Therefore, average actin severing frequency can be determined by dividing the number of severing events (2 times) by the maximal length of the filament (5 μm) and the elapsed time from its birth to disappearance (44 s), which equals to two breaks/ (5 μm
44 s) ¼ 0.009 breaks/μm/s. 5. To determine actin filament bundling frequencies, the region of interest is in the sketch map labeled by a yellow frame (Fig. 4c) 5 μm
5 μm ¼ 25 μm2. Actin bundling events are marked with red arrows. There are two bundling events in the 25 μm2 region within 20 s; thus bundling frequency can be determined as two events/(20 s
25 μm2) ¼ 0.004 events/μ m2/s. 6. Actin de-bundling frequency is determined in the same way by simply replacing the number of bundling events with that of de-bundling events [16, 23].

Actin Dynamics in Arabidopsis Pollen Tubes

293

Fig. 4 Schematic diagram showing the measurement of parameters associated with actin filament dynamics in pollen tubes. (a) Time-lapse images showing the growth and shortening of one individual actin filament. This is used as an example showing how we perform quantification of the max. length, max. lifetime, elongation rate, and depolymerization rate of actin filaments. (b) Time-lapse images showing the severing of an actin filament. Red arrows indicate actin filament fragmentation events. (c) Time-lapse images showing the bundling of actin filaments. Red arrows indicate actin bundling events

294

4

Qiaonan Lu et al.

Notes 1. The medium with a pH of 6.9–7.0 works well. Do not add excessive KOH, which can dramatically decrease pollen germination rate. 2. Heating in the microwave oven will cause weight loss of PGM due to evaporation, so do make up the water loss by adding back distilled water. Weigh the conical flask with PGM before and after microwaving. The difference in the weight is evaporated water, and use Milli-Q water to make up the difference. Make sure the glass Petri dishes are horizontal and stationary while cooling the medium. The liquid or solid PGM can be stored at 4  C for no more than 3 days. 3. Choose plants with over five siliques on the main inflorescence stem, and do not choose older plants with few flower buds. 4. Select pollen tubes with normal morphology and a normal cytoplasmic streaming pattern for image collection under the microscope. 5. Keep acquisition parameters the same when quantifying the fluorescent intensity of the actin filaments. Prevent overexposure and photobleaching, especially when acquiring the timelapse images. 6. Set “Brightness and Contrast” to the same range first, when displaying and comparing the fluorescent intensity between different pollen tubes, and then change the image type from 12- or 16-bit image to 8-bit image. 7. Do all the fluorescent intensity quantification work on raw data with the same acquisition parameters.

Acknowledgments This work was supported by grants from National Natural Science Foundation of China (32270338 and 31970180) and the startup fund from Huazhong Agricultural University. We thank Dr. Wanying Zhao for the help in drawing the schematic diagram shown in Fig. 4. References 1. Bedinger PA, Hardeman KJ, Loukides CA (1994) Travelling in style: the cell biology of pollen. Trends Cell Biol 4:132–138 2. Jiang Y, Zhang M, Huang S (2017b) Analysis of actin-based intracellular trafficking in pollen tubes. Methods Mol Biol 1662:125–136

3. Kroeger JH, Daher FB, Grant M, Geitmann A (2009) Microfilament orientation constrains vesicle flow and spatial distribution in growing pollen tubes. Biophys J 97:1822–1831 4. Zhang Y, He J, Lee D, McCormick S (2010) Interdependence of endomembrane trafficking

Actin Dynamics in Arabidopsis Pollen Tubes and actin dynamics during polarized growth of Arabidopsis pollen tubes. Plant Physiol 152: 2200–2210 5. Chen N, Qu X, Wu Y, Huang S (2009) Regulation of actin dynamics in pollen tubes: control of actin polymer level. J Integr Plant Biol 51: 740–750 6. Cheung AY, Wu HM (2008) Structural and signaling networks for the polar cell growth machinery in pollen tubes. Annu Rev Plant Biol 59:547–572 7. Fu Y (2015) The cytoskeleton in the pollen tube. Curr Opin Plant Biol 28:111–119 8. Qu X, Jiang Y, Chang M et al (2015) Organization and regulation of the actin cytoskeleton in the pollen tube. Front Plant Sci 5:786 9. Ren H, Xiang Y (2007) The function of actinbinding proteins in pollen tube growth. Protoplasma 230:171–182 10. Staiger CJ, Poulter NS, Henty JL et al (2010) Regulation of actin dynamics by actin-binding proteins in pollen. J Exp Bot 61:1969–1986 11. Stephan OOH (2017) Actin fringes of polar cell growth. J Exp Bot 68:3303–3320 12. Xu Y, Huang S (2020) Control of the actin cytoskeleton within apical and subapical regions of pollen tubes. Front Cell Dev Biol 8:614821 13. Cai G, Parrotta L, Cresti M (2015) Organelle trafficking, the cytoskeleton, and pollen tube growth. J Integr Plant Biol 57:63–78 14. Hepler PK, Winship LJ (2015) The pollen tube clear zone: clues to the mechanism of polarized growth. J Integr Plant Biol 57:79–92 15. Vidali L, Rounds CM, Hepler PK, Bezanilla M (2009) Lifeact-mEGFP reveals a dynamic apical F-actin network in tip growing plant cells. PLoS One 4:e5744 16. Chang M, Huang S (2015) Arabidopsis ACT11 modifies actin turnover to promote pollen germination and maintain the normal rate of tube growth. Plant J 83:515–527

295

17. Diao M, Li X, Huang S (2020) Arabidopsis AIP1-1 regulates the organization of apical actin filaments by promoting their turnover in pollen tubes. Sci China Life Sci 63:239–250 18. Jiang Y, Chang M, Lan Y, Huang S (2019) Mechanism of CAP1-mediated apical actin polymerization in pollen tubes. Proc Natl Acad Sci USA 116:12084–12093 19. Jiang Y, Wang J, Xie Y, Chen N, Huang S (2017a) ADF10 shapes the overall organization of apical actin filaments by promoting their turnover and ordering in pollen tubes. J Cell Sci 130:3988–4001 20. Lan Y, Liu X, Fu Y, Huang S (2018) Arabidopsis class I formins control membraneoriginated actin polymerization at pollen tube tips. PLoS Genet 14:e1007789 21. Liu X, Qu X, Jiang Y et al (2015) Profilin regulates apical actin polymerization to control polarized pollen tube growth. Mol Plant 8: 1694–1709 22. Qu X, Zhang H, Xie Y et al (2013) Arabidopsis villins promote actin turnover at pollen tube tips and facilitate the construction of actin collars. Plant Cell 25:1803–1817 23. Qu X, Zhang R, Zhang M et al (2017) Organizational innovation of apical actin filaments drives rapid pollen tube growth and turning. Mol Plant 10:930–947 24. Zhang M, Zhang R, Qu X, Huang S (2016) Arabidopsis FIM5 decorates apical actin filaments and regulates their organization in the pollen tube. J Exp Bot 67:3407–3417 25. Zheng Y, Xie Y, Jiang Y, Qu X, Huang S (2013) Arabidopsis actin-depolymerizing factor7 severs actin filaments and regulates actin cable turnover to promote normal pollen tube growth. Plant Cell 25:3405–3423 26. Staiger CJ, Sheahan MB, Khurana P et al (2009) Actin filament dynamics are dominated by rapid growth and severing activity in the Arabidopsis cortical array. J Cell Biol 184: 269–280

Chapter 24 Noninvasive Long-Term Imaging of the Cytoskeleton in Arabidopsis Seedlings Felix Ruhnow, Staffan Persson, and Rene´ Schneider Abstract The preparation of biological samples, especially for live-cell microscopy, remains a major experimental challenge in the lab despite technological advances. In addition, high-resolution microscopy techniques require higher sample quality and uniformity, which is difficult to ensure during manual preparation while maintaining “ideal” growth conditions. In this protocol, we provide a way out by growing Arabidopsis thaliana seedlings directly in an imaging chamber, which eliminates invasive sample preparation directly before imaging. This method hinges on the precise placement of seeds into imaging chambers, which can be grown in conventional climate chambers. We detail three methods to grow hypocotyls, cotyledons, leaves, and roots for high-resolution and long-term imaging of the plant cytoskeleton. Furthermore, we show that the growth and development of seedlings inside the chambers can be externally manipulated by the addition of chemicals. Key words Arabidopsis, Microtubules, Salt stress, Secondary cell walls, Custom-built imaging chambers, Vertical stage

1

Introduction The diversity of fluorescence microscopy methods has exploded in the last two decades. Plant biology benefits greatly from this general trend, as many fluorescence methods and reporters are developed in mammalian systems [1] and later find their application in plants [2–4]. Despite these favorable circumstances and a wide range of fluorescence techniques available [5], daily lab experience shows that it is not the lack of method that slows scientific progress, but often sample preparation. This is particularly true for experiments that aim at capturing cellular processes over a longer time

Supplementary Information The online version contains supplementary material available at https://doi.org/ 10.1007/978-1-0716-2867-6_24. Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_24, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

297

298

Felix Ruhnow et al.

frame such as organ morphogenesis [6, 7], cytoskeletal organization [8–12], and cell wall synthetic processes [13–16]. The major challenge is to transfer the seedling into an imaging chamber without significantly disrupting the developmental process by, for example, changing the environmental conditions. One common strategy is to first grow seedlings on medium plates and then transfer them to an imaging chamber at an appropriate time. The technical requirements for these imaging chambers are variable and range from relatively simple, handmade agarose pads placed onto seedlings to maintain a wet environment to precisionengineered chambers that use microfluidics to enable environmental control over a longer time period [17–19]. An alternative approach we follow here is to grow seedlings directly in the imaging chamber to minimize or prevent changes of environmental conditions during imaging. We present a strategy that uses commercial and custom-made multi-well chambers that can be adhered to glass coverslips, filled with growth medium and handled similarly to commonly used plant culture dishes. This strategy is easy to implement into a plant lab. We additionally describe how these imaging chambers can be used for various established growth assays such as dark-grown hypocotyl assays, root growth assays, and cotyledon and leaf assays. Although seedling growth and development is slower under a layer of medium than on the surface of it [10], the mechanical interference caused by growth through an obstructing layer is actually more in line with physiological conditions, as supported by recent data [20, 21]. Although the seedlings are covered by a layer of agarose in our assays, their growth and development can still be influenced by external manipulations (such as chemical treatments). We demonstrate this using two examples: 1, the reorientation of cortical microtubules after induction of the secondary wall master transcription factor VASCULAR NAC DOMAIN 7 (VND7) using dexamethasone and 2, the depolymerization and recovery of cortical microtubules upon induction of salt stress using NaCl [11, 12, 22].

2

Materials 1. Imaging chambers: either commercially available such as multiwell μ-Slides from Ibidi (two-well or four-well glass bottom; sterile) or custom-built (one-well glass bottom; sterilized; Fig. 1). 2. Sterile plant growth medium: 1/2 Murashige and Skoog (MS)MES medium with 0.8 % plant agar and 1% sucrose (adjusted to pH 5.7 using KOH).

Cytoskeletal Imaging in Plant Seedlings

299

Fig. 1 Strategies for assembling custom-built one-well imaging chambers. Aluminum chambers (a) can be milled using a CNC machine, and standard coverslips (60 mm
26 mm, #1.5) can be attached with vacuum grease. Plastic chambers (b) can be assembled using empty body μ-Slide (two-well), sticky tape for μ-Slide eight-well and standard coverslips (76 mm
26 mm, #1.5). Assembled chambers can be autoclaved. Standard μ-Slide lids fit on both chambers and can be sterilized with EtOH (do not autoclave the lids)

3. Sterilized Arabidopsis seeds stably transformed with fluorescent markers: we used lines expressing mCHERRY-labeled ALPHA-TUBULIN 5 (TUA5) under control of the 35S promoter from cauliflower mosaic virus [12, 23]. For secondary wall studies, this reporter was used in the inducible VND7 background [9, 22], whereas for salt studies, the wild-type Columbia-0 background was used. 4. 10 μM dexamethasone (DEX, Sigma-Merck) or 200 mM NaCl diluted in water or in liquid 1/2 MS medium. 5. Sterile 1 mL pipette tips and toothpicks. 6. Micropore tape (3 M, 12.5 mm wide) or Parafilm (Bemis, cut to 15 mm wide stripes). 7. Square Petri dishes, 120
120
17 mm. 8. Microscope coverslips (60
26 mm or 76
26 mm, grade #1.5). 9. Microscope cover slides (76
26 mm, 1–1.5 mm thickness). 10. Vacuum grease.

300

3

Felix Ruhnow et al.

Methods We germinate and grow seedlings directly inside imaging chambers until they have reached the desired developmental stage to be investigated under the microscope. We tested three different common growth assays: dark-grown hypocotyl assays, root growth assays, and cotyledon and leaf development assays (Fig. 2). Each assay requires slight modifications to the overall protocol of sample preparation that we outline in the following.

3.1 Setting Up Arabidopsis Seedlings for Growth Inside Imaging Chambers

Steps 1–6 should be carried out in a sterile laminar flow cabinet. 1. Place sterile imaging chambers on a sterile and clean surface. For sterilizing custom-built imaging chambers, see Note 1. 2. Heat plant growth medium to 95  C to melt the agarose (see Note 2). 3. Fill the imaging chambers with hot, liquid medium using a pipette. Use 0.5–1 mL of medium for each well of the fourwell chamber slides and 1.5–2 mL for each well of the two-well chamber slides. The custom-built chambers (one well) are filled with 3.5–4 mL of medium. 4. Wait 5 min for the medium to cool and solidify. 5. To transfer sterile seeds to the chambers, use sterile toothpicks to push individual seeds through the solid medium until they touch the glass coverslip (Fig. 3a–b, see Note 4, and Electronic Supplementary Movie 1). For good imaging results, it is important that the seedlings grow as close as possible to the coverslip.

ers

Cotyledon and leaf assay

ers lip

cov

Root assay

cov

lip

Hypocotyl assay

coverslip

Fig. 2 Growth variants of Arabidopsis seedlings in long-term imaging chambers. Seeds germinate inside the medium-filled imaging chambers. Orienting these chambers in specific ways ensures that the seedlings grow and elongate close to the coverslip. A short distance between the epidermis and coverslip is crucial for longterm live-cell imaging

Cytoskeletal Imaging in Plant Seedlings

301

Fig. 3 Sample preparation for imaging elongated hypocotyls. (a) Place the seeds in the bottom third of the chamber. (b) Use a toothpick to carefully push the seeds through the solid medium until they touch the coverslip. (c) Close the chambers and seal them with tape. (d) Place and fix the chambers in a square Petri dish. (e) Orient the dish in a growth incubator, slightly tilted at an angle between 60 and 80 from the horizontal. (f) Examples of seedlings grown in the dark for 4 days. Four-well Ibidi μ-Slide (top) and one-well custom-made chambers (bottom)

6. For imaging elongated hypocotyls, place the seeds in the bottom third of the chamber (Fig. 3a). For imaging roots, cotyledons, or leaves, see Note 3 and Figs. 4 and 5. For the 4-well, 2-well, and custom-built chambers (Fig. 6), use 2–3, 4–6, or 8–12 seeds, respectively. 7. Close the chambers with the provided lids and additionally seal them with micropore tape or Parafilm (Fig. 3c). The custommade chambers can be closed with Ibidi lids or, alternatively, with standard microscope slides attached to the aluminum frame with a thin layer of vacuum grease. 8. Carefully place up to four chambers in a square Petri dish and fix them with tape (Fig. 3d). Additionally sealing the Petri dish with tape, micropore tape, or Parafilm (see Note 5) helps in maintaining constant humidity. 9. If seeds have not been exposed to the cold for a minimum of 48 h, cold-incubate the chambers at 4  C. Subsequently, expose the chambers to a light intensity of at least 100 μE for up to 4 h at 22  C to trigger germination (see Note 6). 10. For imaging elongated hypocotyls, wrap the Petri dish after light exposure with aluminum foil and leave them in a plant incubator at 22  C (see Note 7). 11. Tilt the Petri dish at an angle between 60 and 80 from the horizontal (Fig. 3e). For good imaging results, it is paramount that the Petri dish is tilted properly, i.e., that the hypocotyls grow against the coverslip as they elongate (Fig. 2). If the Petri

302

Felix Ruhnow et al.

Fig. 4 Sample preparation for imaging roots. (a) Place the seeds in the top third of the chamber, and push seeds through the medium until they touch the coverslip. (b) For optimal seedling growth, use a sterile scalpel to cut out the upper third of the agarose. (c) Instead of pushing seeds through the medium, place them at the edge between the remaining agarose and the coverslip. (d) Place and fix chambers in a square Petri dish (cover the bottom of the Petri dish with black paper; do not stack chambers on top of each other), and place the Petri dish in a growth incubator, slightly tilted at an angle between 60 and 80 from the horizontal. (e) Examples of 5-day-old light-grown seedling. (f) When imaging roots of seedlings that are older than 5 days, turn the custom-built chambers by 90 to allow for longer root growth. (g) Time-lapse images of the cortical microtubule array in the same epidermal root cell for 14 h (3i Marianas plant microscope with spinning disk; Nikon CFI Apo TIRF 60
/1.49NA objective; 10min time resolution; scale bar 40 μm)

dish is tilted incorrectly, the hypocotyls will grow away from the coverslip and can no longer be observed due to the relatively short working distance of high-magnification objectives. For imaging roots, cotyledons, or leaves, see Notes 7–9. 12. After 4 days of growing in the dark, unwrap the Petri dish and immediately place the imaging chamber onto the microscope stage (Fig. 3f; see Note 10). For long-term imaging of elongated hypocotyls, strategies for imaging seedlings in their natural, upright position are highly recommended [24]. The remaining imaging chambers can be kept in the dark for up to 24 h (see Note 11). Maintain the tilted configuration. See Notes 12–16 for more technical details on long-term or largescale imaging. 3.2 Long-Term Imaging of Secondary Wall Formation

Induction of secondary wall formation takes approximately 6–12 h before changes on the cytoskeletal level become observable [9]. The induction step can be performed directly inside the imaging chambers, ideally (but not necessarily) in a sterile laminar flow cabinet. 1. Remove the lid of the imaging chamber (see Note 12), and add a solution containing 10 μM DEX; incubate for 10–30 min with the lid closed. For the four-well, two-well, and custom-

Cytoskeletal Imaging in Plant Seedlings

303

Fig. 5 Sample preparation for imaging cotyledons and leaves. (a) Place the seeds in a 3-by-3 grid, and push seeds through the medium until they touch the coverslip. (b) For optimal seedling growth, use autoclaved, 3D-printed square dishes (using heat-resistant material, Fig. 6), place them on a lid, fill them with growth media, and place seeds on top of the media in a 3
3 grid. (c) These dishes fit in the 2-well imaging chambers, and a thin layer of air should remain between growth media and coverslip (to ensure proper opening of the cotyledons and better growth of young seedlings). (d) Place chambers flat in the growth incubator with the coverslip facing upward (avoid light exposure from the side). (e) Large-scale 3D montage of the microtubule network in the same seedling over several days (Electronic Supplementary Movie 2). Acquisition of each montage required 3 h of stable imaging each day, and the seedlings were placed back in the growth incubator after each imaging session (Leica Stellaris 8; HC PL APO 20
/0.75NA objective; 21 tiles, 125 slices; total imaging volume approximately 2 mm
1 mm
0.5 mm with 360 nm resolution in xy and 4 μm in z; scale bar ¼ 1 mm)

B

A

0

9,50

C

22,30

7,00

18,30

R2 ,0

16,30 20,30

Fig. 6 Custom-printed chamber for cotyledon and leaf imaging. (a) Top view, (b) side view, and (c) 3D view of the printed chamber which fits in μ-Slide (two-well). The chamber design STL file for standard 3D printers is supported with the Electronic Supplementary Material 1. (Courtesy of Lasse Gaardsted Flyvholm)

304

Felix Ruhnow et al.

Fig. 7 Time-lapse images of the cortical microtubule array. (a) Re-arrangements of cortical microtubules during secondary wall formation in 4-day-old, VND7-induced seedlings. Scale bar ¼ 20 μm (100
/1.49NA). (b) Transient depolymerization and recovery of cortical microtubules during salt stress (200 mM NaCl). Scale bar ¼ 40 μm (60
/1.49NA). Both examples were recorded with a 3i Marianas plant microscope with spinning disk; Nikon CFI Apo TIRF objectives; 20 min time resolution

built chambers, use 50–200 μL, 200–500 μL, and 1 mL of solution, respectively. Make sure the solution is wetting the entire medium surface. Remove the remaining solution after incubation with a pipette and properly close the lid. 2. Approximately 6 h after induction, cells in the upper hypocotyl and the apical hook will undergo microtubule rearrangements. Focus on these regions to identify induced seedlings. Select multiple regions of interest (cells) along the hypocotyl for automated imaging (2–3 regions per seedling is recommended). 3. Set up automated, three-dimensional time-lapse imaging of multiple cells in different seedlings using N-dimensional acquisition. Since epidermal cells can be found close to the glass coverslip, the cytoskeletal re-organization during secondary wall formation can be imaged at high resolution (Fig. 7a). Imaging seedlings of different genotypes at similar time points and under the same conditions is also possible and depends only on the placement of seeds in the imaging chamber (see Electronic Supplementary Movie 1).

Cytoskeletal Imaging in Plant Seedlings

3.3 Long-Term Imaging of Seedlings During Salt Stress

305

Seedlings grown in the chambers can be exposed to salt stress during long-term imaging. 1. Carefully remove the tape or Parafilm from the chambers (without opening the chamber). 2. Place the chamber on the microscope stage, and screen for seedlings with good microtubule signal. Select multiple regions of interest along the hypocotyl of selected seedlings for imaging (2–3 regions per seedling is recommended). 3. Set up automated, three-dimensional time-lapse imaging of multiple cells in different seedlings using N-dimensional acquisition. 4. Carefully remove the lid of the imaging chamber, and add a solution containing 200 mM NaCl (see Subheading 3.2 for recommendations on volume). Carefully close the lid without moving the imaging chambers. 5. Start N-dimensional acquisition immediately after adding the solution. After approximately 20–30 min, first alterations of cortical microtubule arrays become observable (Fig. 7b).

3.4 Post-processing of Long-Term Recordings (Drift Correction and Maximum Projections)

The following workflow can be used to correct for stage and focus drift, as well as account for plant growth. The following steps can be performed in FIJI [25]: 1. Load image data into FIJI. The Fiji plugin “bioformats” helps to load various raw data formats. 2. Select the cell of interest by drawing a rectangular region around it. 3. Correct for drift using Plugins > Registration > Correct 3D drift. (Use standard settings, but increase “Max shift x” and “Max shift y” if the log says “too large drift along dimension.”) 4. Create maximum projection if a 3D stack was acquired: Image > Stacks > Z Project with “Projection type ¼ Max Intensity.” 5. Optional: Rotate image to align main cell axis along the y-axis by using Image > Transform > Rotate. 6. Create montage of the time-lapse recording: Image > Stacks > Make Montage.

4

Notes 1. The assembled custom-built chambers can be autoclaved (at 121  C) and stored inside 50 mL Falcon tubes or glass staining troughs. 2. Sterile growth medium can be prepared and stored in 2 mL aliquots. Aliquots can then be heated to 95  C using a heat block.

306

Felix Ruhnow et al.

3. For imaging roots, place seeds in the top third of the chamber, and use 2, 4–5, or 8–10 seeds per imaging chamber (Fig. 4a–c). For imaging cotyledons or leaves, use two-well chambers and place nine seeds in a 3-by-3 grid (Fig. 5a, b). 4. It is best to tilt the imaging chamber by 45 while pushing the seeds through the medium. Carefully watch the tip of the toothpick approach the glass coverslip by observing through the transparent chamber wall (Fig. 2, Electronic Supplementary Movie 1). 5. Using a square Petri dish will help growing the plants in standard growth chambers and allows for correct tilting. Adding a wet paper tissue inside the Petri dish helps to maintain humidity. 6. If seedlings are poorly germinating, try exposing the chambers to 100-μE light intensity for up to 24 h before wrapping in aluminum foil. 7. For imaging roots, cotyledons, or leaves, do not wrap the Petri dish in aluminum foil after inducing germination. 8. For imaging roots, tilt the dish slightly (60–80 from the horizontal) so that the roots will grow along the glass coverslip as they elongate downward (Fig. 2). In cases of light-grown seedlings, avoid stacking several chambers above each other in the same Petri dish because light might be blocked from above. Adding a black sheet of paper in the Petri dish (between dish and glass coverslip) helps to prevent the roots from growing away from the coverslip due to light avoidance (Fig. 4d–g). 9. For imaging cotyledons or leaves, place the Petri dish flat in the growth incubator with the glass coverslip of the imaging chambers facing upward (Fig. 5c, d). Avoid light from the side (or bottom) of the incubator so that the cotyledons or leaves grow upward (i.e., toward the light source and thus press against the coverslip; see Fig. 2). 10. The seedling-age can be varied to image specific developmental processes. In our hands, imaging roots in 4- to 5-day-old seedlings was optimal. Imaging cotyledons was successful for 3- to 7-day-old seedling. First true leaves were imaged in 14-day-old seedlings. 11. Hypocotyls will reorient if placed in light or will grow away from the glass coverslip if the imaging chamber is not tilted correctly. 12. The room and microscope temperature are essential for stable imaging, since even minor temperature fluctuations can lead to significant focus drift. To reduce large temperature changes, (a) start the microscope at least 1 h before imaging,

Cytoskeletal Imaging in Plant Seedlings

307

(b) optimize air-conditioning in the microscopy room, (c) use a temperature-controlled objective, and/or (d) activate focus stabilization system (if available). 13. Focus stabilization systems (e.g., Nikon Perfect Focus System, Zeiss Definite Focus, etc.) might have difficulties with the refractive index changes within plant cells and the growth medium. Movement of the imaged cells (either due to plant growth or stage drift) can lead to the disruption of the focus stabilization system. 14. Instead of long-term imaging of the same cells (tracking experiment), the imaging chambers can also be used for large-scale high-resolution imaging. This imaging mode typically requires the acquisition and subsequent 3D stitching of several tiles (Fig. 5e and Electronic Supplementary Movie 2). The total acquisition time can exceed an hour or more, and therefore stable imaging conditions (i.e., absence of stage- or focus-drift) are essential for optimal stitching. 15. Imaging seedlings in chambers can be done in upright and inverted microscope configurations. However, to minimize reorientation of the seedlings during imaging, a vertical stage is recommended when imaging hypocotyls or roots, and an upright microscope configuration is most practical for imaging cotyledons or leaves. 16. The continuous growth of the seedling causes significant sample drift and is therefore the biggest challenge during longterm imaging. Elongating cells will typically move out of the field of view within 1 h and require the stage position to be adjusted accordingly. This can either be done manually by moving the stage between the acquisition time points or by automated scripts that correct for 3D plant growth (e.g., MATLAB tracking in Slidebook 6, 3i Intelligent Imaging Innovations). Such scripts are available upon request from the authors. References 1. Chalfie M, Tu Y, Euskirchen G, Ward WW, Prasher DC (1994) Green fluorescent protein as a marker for gene expression. Science 263: 802–805. https://doi.org/10.1126/science. 8303295 2. Mathur J (2007) The illuminated plant cell. Trends Plant Sci 12:506–513. https://doi. org/10.1016/j.tplants.2007.08.017 3. Haseloff J, Siemering KR, Prasher DC, Hodge S (1997) Removal of a cryptic intron and subcellular localization of green fluorescent protein are required to mark transgenic

Arabidopsis plants brightly. Proc Natl Acad Sci USA 94:2122–2127. https://doi.org/10. 1073/pnas.94.6.2122 4. Marc J, Granger CL, Brincat J, Fisher DD, Kao T, McCubbin AG, Cyr RJ (1998) A GFP-MAP4 reporter gene for visualizing cortical microtubule rearrangements in living epidermal cells. Plant Cell 10:1927–1940. https://doi.org/10.1105/tpc.10.11.1927 5. Colin L, Martin-Arevalillo R, Bovio S, Bauer A, Vernoux T, Caillaud M-C, Landrein B, Jaillais Y (2021) Imaging the living plant cell: from

308

Felix Ruhnow et al.

probes to quantification. Plant Cell. https:// doi.org/10.1093/plcell/koab237 6. Hamant O, Heisler MG, Jo¨nsson H, Krupinski P, Uyttewaal M, Bokov P, Corson F, Sahlin P, Boudaoud A, Meyerowitz EM, Couder Y, Traas J (2008) Developmental patterning by mechanical signals in Arabidopsis. Science 322:1650–1655. https://doi.org/ 10.1126/science.1165594 7. Zhao F, Du F, Oliveri H, Zhou L, Ali O, Chen W, Feng S, Wang Q, Lu¨ S, Long M, Schneider R, Sampathkumar A, Godin C, Traas J, Jiao Y (2020) Microtubule-mediated wall anisotropy contributes to leaf blade flattening. Curr Biol 30:3972–3985.e6. https:// doi.org/10.1016/j.cub.2020.07.076 8. Schneider R, Tang L, Lampugnani ER, Barkwill S, Lathe R, Zhang Y, McFarlane HE, Pesquet E, Niittyla T, Mansfield SD, Zhou Y, Persson S (2017) Two complementary mechanisms underpin cell wall patterning during xylem vessel development. Plant Cell 29: 2433–2449. https://doi.org/10.1105/tpc. 17.00309 9. Schneider R, Klooster KV, Picard KL, van der Gucht J, Demura T, Janson M, Sampathkumar A, Deinum EE, Ketelaar T, Persson S (2021) Long-term single-cell imaging and simulations of microtubules reveal principles behind wall patterning during proto-xylem development. Nat Commun 12: 669. https://doi.org/10.1038/s41467-02120894-1 10. Eng RC, Schneider R, Matz TW, Carter R, Ehrhardt DW, Jo¨nsson H, Nikoloski Z, Sampathkumar A (2021) KATANIN and CLASP function at different spatial scales to mediate microtubule response to mechanical stress in Arabidopsis cotyledons. Curr Biol. https:// doi.org/10.1016/j.cub.2021.05.019 11. Endler A, Kesten C, Schneider R, Zhang Y, Ivakov A, Froehlich A, Funke N, Persson S (2015) A mechanism for sustained cellulose synthesis during salt stress. Cell 162:1353– 1364. https://doi.org/10.1016/j.cell.2015. 08.028 12. Kesten C, Wallmann A, Schneider R, McFarlane HE, Diehl A, Khan GA, van Rossum B-J, Lampugnani ER, Szymanski WG, Cremer N, Schmieder P, Ford KL, Seiter F, Heazlewood JL, Sanchez-Rodriguez C, Oschkinat H, Persson S (2019) The companion of cellulose synthase 1 confers salt tolerance through a Tau-like mechanism in plants. Nat Commun 10:857. https://doi.org/10.1038/s41467019-08780-3 13. Paredez AR, Somerville CR, Ehrhardt DW (2006) Visualization of cellulose synthase

demonstrates functional association with microtubules. Science 312:1491–1495. https://doi.org/10.1126/science.1126551 14. Watanabe Y, Meents MJ, McDonnell LM, Barkwill S, Sampathkumar A, Cartwright HN, Demura T, Ehrhardt DW, Samuels AL, Mansfield SD (2015) Visualization of cellulose synthases in Arabidopsis secondary cell walls. Science 350:198–203. https://doi.org/10. 1126/science.aac7446 15. Watanabe Y, Schneider R, Barkwill S, Gonzales-Vigil E, Hill JL, Samuels AL, Persson S, Mansfield SD (2018) Cellulose synthase complexes display distinct dynamic behaviors during xylem transdifferentiation. Proc Natl Acad Sci USA 115:E6366–E6374. https://doi.org/10.1073/pnas.1802113115 16. Schneider R, Ehrhardt DW, Meyerowitz EM, Sampathkumar A (2022) Tethering of cellulose synthase to microtubules dampens mechanoinduced cytoskeletal organization in Arabidopsis pavement cells. Nat Plants 8:1064–1073. https://www.nature.com/articles/s41477022-01218-7 17. Grossmann G, Guo W-J, Ehrhardt DW, Frommer WB, Sit RV, Quake SR, Meier M (2011) The RootChip: an integrated microfluidic chip for plant science. Plant Cell 23:4234–4240. https://doi.org/10.1105/tpc.111.092577 18. Seerangan K, van Spoordonk R, Sampathkumar A, Eng RC (2020) Long-term live-cell imaging techniques for visualizing pavement cell morphogenesis. Methods Cell Biol 160:365–380. https://doi.org/10. 1016/bs.mcb.2020.04.007 19. Calder G, Hindle C, Chan J, Shaw P (2015) An optical imaging chamber for viewing living plant cells and tissues at high resolution for extended periods. Plant Methods 11:22. https://doi.org/10.1186/s13007-0150065-7 20. Baral A, Aryal B, Jonsson K, Morris E, Demes E, Takatani S, Verger S, Xu T, Bennett M, Hamant O, Bhalerao RP (2021) External mechanical cues reveal a kataninindependent mechanism behind auxinmediated tissue bending in plants. Dev Cell 56:67–80 21. Jonsson K, Lathe RS, Kierzkowski D, RoutierKierzkowska A-L, Hamant O, Bhalerao RP (2021) Mechanochemical feedback mediates tissue bending required for seedling emergence. Curr Biol. https://doi.org/10.1016/j. cub.2020.12.016 22. Yamaguchi M, Mitsuda N, Ohtani M, OhmeTakagi M, Kato K, Demura T (2011) VASCULAR-RELATED NAC-DOMAIN7 directly regulates the expression of a broad

Cytoskeletal Imaging in Plant Seedlings range of genes for xylem vessel formation. Plant J 66:579–590. https://doi.org/10. 1111/j.1365-313X.2011.04514.x 23. Gutierrez R, Lindeboom JJ, Paredez AR, Emons AMC, Ehrhardt DW (2009) Arabidopsis cortical microtubules position cellulose synthase delivery to the plasma membrane and interact with cellulose synthase trafficking compartments. Nat Cell Biol 11:797–806. https:// doi.org/10.1038/ncb1886 24. DeVree BT, Steiner LM, Głazowska S, Ruhnow F, Herburger K, Persson S, Mravec J (2021) Current and future advances in

309

fluorescence-based visualization of plant cell wall components and cell wall biosynthetic machineries. Biotechnol Biofuels 14:78. https://doi.org/10.1186/s13068-02101922-0 25. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez J-Y, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A (2012) Fiji: an opensource platform for biological-image analysis. Nat Methods 9:676–682. https://doi.org/10. 1038/nmeth.2019

Chapter 25 Visualization of Cytoskeleton Organization and Dynamics in Elongating Cotton Fibers by Live-Cell Imaging Guangda Wang, Yanjun Yu, and Zhaosheng Kong Abstract Cotton fibers are extremely elongated single cells and have long been regarded as an ideal model to investigate polarized plant cell elongation. Actin filaments (F-actin), as well as the cortical microtubules (CMTs), play vital roles in polarized cell growth and morphogenesis. We have generated stable transgenic cotton plants expressing fluorescent markers for the actin and microtubule cytoskeletons. Further live-cell imaging identified dynamic features of the F-actin and cortical microtubule (CMT) architectures and discovered that cotton fibers elongate in a unique tip-biased diffuse growth mode. Here, we describe methods for preparing growing cotton fiber samples, as well as the visualization of cytoskeletal organization and dynamics by live-cell imaging. Combined with comprehensive image analyses, these methods can be used to identify how cytoskeleton organization and dynamics determine cell morphogenesis in highly polarized cotton fibers. Key words Cotton fiber, Live-cell imaging, Actin filaments (F-actin), Cortical microtubules (CMTs), Spinning-disc confocal microscopy

1

Introduction Cotton fibers are unbranched seed trichomes, which show a highly polarized elongation pattern after differentiating from the ovule epidermis. As single cells, cotton fibers can elongate up to 2 mm per day during their rapid growth stage [1]. The length of mature cotton fibers can reach up to 5 cm, which is 3000 times greater than their width [2]. Due to their unicellular morphology and the rapid elongation characteristics, cotton fibers have long been considered an ideal model to address the mechanism for plant cell morphogenesis and polarized cell elongation [3]. It is well accepted that actin filaments and cortical microtubules (CMT) play central roles in plant cell elongation and polarity establishment. Accumulating comparative genomic analyses, including GWAS, eQTL, and other omics datasets, has offered numerous candidate genes that may be associated with cotton

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_25, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

311

312

Guangda Wang et al.

fiber elongation, and these include a number of genes encoding cytoskeletal proteins, such as GhACTIN1, GhADF1, GhFIM2, and GhTUA9 [4]. In the past, immunostaining was widely utilized to investigate cytoskeletal structures in cultured cotton fibers, especially during the elongation stage [5]. However, it should be noted that the growth conditions of the cultured cotton fibers were totally different from that of naturally grown fibers and that the fixation process may also have affected the arrangement of F-actin and CMTs. Thus, the results that were obtained from cultured fibers in this way is not sufficient to reveal the cellular mechanisms of cytoskeletal dynamics and cell polarity establishment during cotton fiber elongation in vivo. However, the difficulties encountered in preparing samples for microscopy have impaired live-cell imaging of growing cotton fibers. In addition, cotton transgenic lines expressing appropriate fluorescent cytoskeletal markers are needed. Here, we describe a method for the visualization of F-actin and CMT organization in living cotton fiber cells using a spinning-disc confocal laser scanning microscopy. Stable transgenic cotton lines expressing the fluorescent markers, ABD2-GFP (a reporter for F-actin) or mCherry-EB1b (a marker for tracking microtubule plus-ends), were used to observe F-actin or CMT organization, respectively. Combined with image processing of time series or z-stack confocal images, we have recorded the F-actin and CMT architectures and dynamics in growing cotton fiber cells. Based on these live-cell observations, we discovered that cotton fibers elongate via a unique, tip-biased diffuse growth mode [6]. In summary, the stable transgenic cotton lines expressing ABD2–GFP and mCherry–EB1b markers, as well as the method for preparing samples for live-cell imaging in cotton fibers, offer opportunities for further exploring cytoskeletal dynamics, providing a unique livecell perspective for understanding the polarity and growth of these specialized elongated cells.

2 2.1

Materials Plant Materials

2.2 Solutions and Microscopy Materials

1. Stable transgenic cotton lines carrying the F-actin marker ABD2-GFP or the CMT marker mCherry-EB1b [6]. 1. Liquid 1/2 Murashige and Skoog (MS) medium: Dissolve 0.22 g of MS basal salt mixture in 95 mL ddH2O. Adjust pH to 5.8 ± 0.1 with 1M KOH or HCl. Fill up to 100 mL with ddH2O. After autoclaving at 113 °C for 30 min, store at room temperature (25 °C). 2. Solid 1/2 Murashige and Skoog (MS) medium: Add 0.5 g of phytagel to 100 mL of liquid MS medium, and autoclave at 113 °C for 30 min. Dispense 20 mL sterilized medium in petri dishes (10 cm × 10 cm) for use and storage at room temperature (25 °C).

Observation of Cytoskeletal Dynamics in Elongating Cotton Fibers

313

3. Razor blades and tweezers. 4. Glass slides and cover glasses. 5. Chambered cover glasses. 2.3 Microscopy and Computer Program

1. Spinning-disk confocal microscope (Ultraview VoX, PerkinElmer) equipped with a 100× Plan Apo oil immersion TIRF objective (NA = 1.49). 2. ImageJ (http://rsbweb.nih.gov/ij).

3

Methods All procedures are carried out at room temperature.

3.1 Preparation of Cotton Fibers for LiveCell Imaging

1. Cotton marker lines expressing fluorescent protein fusions were grown in the field under normal farming practices or in a greenhouse with a 16 h day length (28 °C) and 8 h in the dark (25 °C). 2. Flowers were labeled on the day of anthesis (0 DPA). Harvest the fresh whole flowers from the cotton plants in the field or greenhouse (see Note 1). 3. Using clean razor blades and tweezers, remove the petals and sepals. Incise the ovary carefully with a razor blade (Fig. 1). Gently separate the ovules from the ovary with tweezers (see Notes 2 and 3). 4. Using a razor blade, dissect a slice of the surface of an ovule (approx. 4 mm2). Place the slice on a glass slide immediately using tweezers (see Note 4). Use a drop of liquid 1/2 MS medium (20–50 μL) for mounting the coverslip (see Note 5).

Fig. 1 Preparation of cotton fibers. Fresh flowers at 3 DPA were harvested from greenhouse grown plants (left panel, petals are removed). A small piece of ovary wall is removed (middle panel) to expose the ovules (right panel)

314

Guangda Wang et al.

5. Alternatively, the ovule can also be placed on the chambered cover glass without dissection. A piece of trimmed solid 1/2 MS medium can be used here instead of a coverslip (see Note 6). 6. Use small pieces of Whatman filter paper to absorb excess liquid. Stick the cover glass to the slides with tape. 3.2 Image Acquisition

1. Clean the objective with 70% ethanol using lens paper before starting the observation. 2. Live-cell imaging is performed using a confocal laser scanning microscope under 100× oil objective. Use transmission light to find the cotton fiber cells (see Note 7). 3. To detect the F-actin or CMT markers in cotton fibers, use laser lines at 488 nm for GFP excitation and at 561 nm for mCherry excitation. 4. The frame size can be set as 512 × 512. Acquire time-lapsed images with a 5 s interval for 40–60 frames. A z step at 0.5 μm can be used to obtain z-stacks with the maximum speed (see Note 8).

3.3 Image Processing

1. Convert the acquired images into 8-bit TIFF file stacks using ImageJ. 2. Use the rolling ball method with a mean filter to subtract the background, and enhance the signal-noise ratio for confocal images. The “walking average” plugin can be used to reduce the noise in the time-lapsed image series. 3. To detect the CMT organizations in cotton fibers, the frame projection can be used for overlaying mCherry-EB1b signals (Fig. 2). Using the 3-D viewer can generate 3-D optical view of F-actin in acquired z-stack images (Fig. 3). 4. We use a previously reported framework to automatically process and extract networks from the actin cytoskeleton images. The detailed steps for the quantification can be obtained in [7].

4

Notes 1. Keep the flowers as fresh as possible and avoid low temperature storage. It is recommended that flowers are picked with branches that have leaves. 2. Work as gently as possible to avoid additional harm to the ovules. 3. For sequential observation of an individual ovule in one ovary, avoid leaving the ovules exposed in air for a long time, before and after dissecting the ovary.

Observation of Cytoskeletal Dynamics in Elongating Cotton Fibers

315

Fig. 2 Microtubule organization during cotton fiber development. By using confocal microscopy, mCherryEB1b fluorescent signals were observed in living cotton fibers at 0 DPA (left panels) and 3 DPA (right panels). Projection of time-lapsed series of the EB1b fluorescent comets (top panels), showing the microtubule organization (bottom panels). Scale bars, 10 μm

Fig. 3 F-actin structures in living cotton fibers. (a) Extension focus of z-stack confocal image showing the actin filaments in fibers at 3 DPA. (b) Multi-view of 3-D optical images showing the F-actin spatial structures in (a). Scale bar, 5 μm

4. It is better to make only one cut for each ovule. Make the ovule slice as thin as it can be for mounting on the coverslips. 5. ddH2O can also be used immediately for mounting coverslips. But the liquid 1/2 MS medium is recommended for long-time live-cell observation. 6. Using chambered cover glasses may reduce additional harm to the ovules or fiber cells for long-time live-cell observation. But the quality of obtained images can also be affected since the fiber cells are less attached to the cover surface as tightly as using coverslips. 7. Perform the microscopy immediately after mounting the coverslips. The CMTs may depolymerize gradually. 8. Avoid over-scanning at one position to minimize photobleaching, during image acquisition

316

Guangda Wang et al.

References 1. Pfluger J, Zambryski PC (2001) Cell growth: the power of symplastic isolation. Curr Biol 11: R436–R439 2. Kim HJ, Triplett BA (2001) Cotton fiber growth in planta and in vitro. Models for plant cell elongation and cell wall biogenesis. Plant Physiol 127:1361–1366 3. Haigler CH, Betancur L, Stiff MR, Tuttle JR (2012) Cotton fiber: a powerful single-cell model for cell wall and cellulose research. Front Plant Sci 3:104 4. Huang G, Huang JQ, Chen XY, Zhu YX (2021) Recent advances and future perspectives in cotton research. Annu Rev Plant Biol 72:437

5. Wang JA, Wang HY, Zhao PM, Han LB, Jiao GL, Zheng YY, Huang SJ, Xia GX (2010) Overexpression of a Profilin (GhPFN2) promotes the progression of developmental phases in cotton fibers. Plant Cell Physiol 51:1276–1290 6. Yu Y, Wu S, Nowak J, Wang G, Han L, Feng Z, Mendrinna A, Ma Y, Wang H, Zhang X et al (2019) Live-cell imaging of the cytoskeleton in elongating cotton fibres. Nat Plants 5:498–504 7. Breuer D, Nowak J, Ivakov A, Somssich M, Persson S, Nikoloski Z (2017) System-wide organization of actin cytoskeleton determines organelle transport in hypocotyl plant cells (vol 114, pg E5741, 2017). P Natl Acad Sci U S A 114:E6732–E6732

Chapter 26 Methods to Visualize and Quantify Cortical Microtubule Arrays in Arabidopsis Conical Cells Xie Dang, Lilan Zhu, Huibo Ren, and Deshu Lin Abstract Many studies from different model organisms have demonstrated that microtubules are essential for various cellular processes, including cell division, cell morphogenesis, and intracellular trafficking. In interphase plant cells, oriented cortical microtubule arrays are highly characteristic in cells that display various morphologies, such as elongated hypocotyl cells and root cells, jigsaw-puzzled leaf pavement cells, and petal epidermal conical cells. Conical cells represent a specialized epidermal cell type found in the petal epidermis of many flowering plants. It has been suggested that in the model plant Arabidopsis thaliana, the petal adaxial epidermal cells develop from a roughly hemispherical morphology to a conical shape, correlating with the reorientation of cortical microtubules from random to well-ordered circumferential arrays. This chapter presents an overview of the methods available to visualize the microtubule cytoskeleton in living conical cells via confocal microscopy. Key words Plant, Microtubule, Cell biology, Fluorescent-tagged reporters, Confocal microscope, Live-cell imaging, Conical cells, Petal

1

Introduction In plants, the cytoskeleton system is composed of two core components: microtubules and actin filaments. Specific cytoskeleton configurations are essential for diverse cellular processes, such as chromosome segregation, intracellular trafficking, cell motility, cell division, and morphogenesis [1–3]. In interphase plant cells, cortical microtubules that are attached to the plasma membrane (PM) guide the deposition of cellulose microfibrils in the cell wall while serving as tracks for PM-localized cellulose synthase complexes (CSCs) [4–6]. Cell expansion is in part determined by cell wall anisotropy [7–9]. Thus, the organization of cortical microtubule arrays is fundamental to directional cell expansion. Cortical

Authors Xie Dang and Lilan Zhu have equally contributed to this chapter. Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_26, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

317

318

Xie Dang et al.

microtubules are organized into diverse dynamic arrays in the absence of microtubule organizing centers in the various cellular types. During plant growth and development, cortical microtubules are usually organized into aligned arrays that are oriented perpendicularly to the direction of the axis of cell elongation [10]. The plant epidermis has various essential functions, such as protection, interactions with the environment, providing mechanical strength, and controlling tissue and organ growth. To achieve these physiological functions, epidermal cells from various organs have evolved specialized shapes. These include but are not limited to the typical jigsaw-puzzle shape of cotyledon and leaf pavement cells, branched leaf trichomes, elongated root hairs, rectangularshaped root epidermal cells, and specialized petal conical cells [11– 13]. Live-cell imaging studies have demonstrated that the organization of cortical microtubules into well-ordered arrays and the re-organization of these arrays in specialized cells have fascinated plant biologists for a long time. Understanding how cortical microtubules are organized into specific arrays and the underlying regulatory mechanisms are very important but remain a challenging task [14, 15]. It should be noted that angiosperm species usually have specialized epidermal conical cells that are used as a characteristic to identify petals [16–19]. These cells vary greatly in size and angle among different species, playing roles in pollinator attraction, light capture and reflectance, and temperature and wettability maintenance [16]. Thus, understanding how petal cells achieve their characteristic shape is very important. A previous study showed that the MIXTA gene encodes a MYB transcription factor in Antirrhinum majus, with loss-of-function mutants exhibiting flat hexagonal-based cells instead of wild type-like conical cells [17]. Like most angiosperm species, Arabidopsis has conical cells that are decorated with cuticular nanoridges in the petal adaxial epidermis [20]. We use the conical cell of Arabidopsis petals as a model cellular system for the study of the mechanism that regulates cortical microtubule organization [12]. Scanning electron microscopy is often used to visualize the 3D geometric shape of conical cells with high resolution, but this approach is not good for high-throughput experimental analysis. Live-confocal microscope imaging methods together with powerful software [21–23] enable researchers to describe cell morphology and protein localization patterns over the course of their development and have been extensively used in many cellular types of Arabidopsis, such as leaf pavement cells, trichomes, shoot apical meristem cells, and root cells. .By using confocal microscopebased imaging approaches, the Arabidopsis conical cell provides a useful system for investigating microtubule organization and cell morphogenesis [24–27]. Using Arabidopsis plants stably expressing α-Tubulin 6 fused with green fluorescent protein (GFP-TUA6) as a

Cortical Microtubule Arrays in Arabidopsis Conical Cells

319

microtubule reporter [28], we tracked the behaviors of the cortical microtubule cytoskeleton and suggested that the final conical shape is correlated with cortical microtubule reorientation from a disordered into a well-ordered circumferential array over the course of cell development [25]. The organization of cortical microtubules into well-ordered circumferential arrays requires the microtubulesevering protein KATANIN (KTN1), with loss of KTN1 function causing disordered microtubule arrays and wide-angled conical cells [25]. We recently reported that protein phosphatase 2A (PP2A) interacts with KTN1 and they cooperatively regulate the formation of circumferential cortical microtubule arrays during the morphogenesis of conical cells [27]. Here we describe in detail how to visualize conical cells and to quantitatively investigate cortical microtubule orientation in Arabidopsis conical cells using transgenic Arabidopsis plants expressing the GFP-TUA6.

2

Materials

2.1 Plant Materials and Growth Condition

1. Arabidopsis ecotype Columbia (Col-0). 2. ktn1–4, seeds of T-DNA insertion line (SAIL_343_D12) from the ABRC Stock center. 3. GFP-TUA6, a transgenic line used for the microtubule marker [28]. 4. ktn1–4 GFP-TUA6 [25].

2.2 Special Chemicals

1. Propidium iodide: 10 μg propidium iodide (PI) in 1 mL distilled water and stored in 50 μL aliquots at 4 °C.

2.3 Materials for Imaging Conical Cells and Microtubules

1. Microscope slides.

2.4 Special Equipment

1. Hitachi scanning electron microscope (SEM) TM3030Plus with 5 kv voltage mod.

2. 22 × 22 mm, Coverslips. 3. Double-sided tape.

2. Carl Zeiss inverted microscope (Axio Observer A1) equipped with 20× objective lens (0.8 NA) and 10× ocular lens. 3. Carl Zeiss confocal laser scanning microscope (CLSM, LSM 880) equipped with 63× immersion oil objective lens (1.4 NA), 20× objective lens (0.8 NA) and 10 × ocular lens, Airyscan mode, GaAsP detector, 488 nm and 561 nm laser. 2.5

Software

1. Microsoft Office Excel. 2. ImageJ/Fiji (https://imagej.net/software/fiji/).

320

Xie Dang et al.

3. Macro: OrientationJ (http://bigwww.epfl.ch/demo/orienta tion/index.html?applet=measure#soft). 4. GraphPad Prism (https://www.graphpad.com/scientific-soft ware/prism/).

3

Methods

3.1 Imaging Conical Cells Using a TableTop Scanning Electron Microscope

1. Grow plants in walk-in growth chambers under long-day (16 h light/8 h dark, 100 μmol m-2 s-1 light intensity) conditions at 22 °C. 2. Stick a 0.5 × 0.5 cm2 conductive adhesive on the TM3030Plus stage. 3. Collect stage 14 petals, keep adaxial side upward and gently stick them on the conductive adhesive (see Note 1). 4. Select 5 kv voltage to perform a quick preview scan of the sample in standard mode. 5. Adjust the magnification and brightness as needed. 6. After finding a region of interest, scan and save the target region with a slower scanning speed (see Note 2). 7. Stage 14 petals of Col-0 showed a typical conical shape on the adaxial surface and relatively flat cells on the abaxial surface (Fig. 1).

3.2 Imaging Conical Cells Using a Confocal Microscope

1. Prepare a piece of double-sided tape adhered to a microscope slide. 2. Dissect petals from flowers at developmental stage 14. 3. Put petal sample onto the double-sided tape. 4. Fold petal blades in half along the transverse axis, enabling the side view of the conical cells under a microscope (Fig. 2a). 5. Add 10 μg/mL of propidium iodide solution on the petal sample and then put a coverslip on the microscope slide. 6. Incubate the petal sample at room temperature for more than 10 min for staining. 7. Observe conical cells under a confocal microscope (Zeiss LSM 880) at the folded interface. For generating the 3D shape of conical cells, Z-stack confocal images are taken from the top view along the Z axis at steps of 0.8 μm, and these are reconstructed into 3D images using Zeiss LSM 880 software (Fig. 2b).

Cortical Microtubule Arrays in Arabidopsis Conical Cells

321

Fig. 1 Observation of petal epidermal cells using a scanning electron microscope. Left panel shows a wildtype mature flower (stage 14) for observation of epidermal cell shape in Arabidopsis. Right panel shows images from a TM-3000 table-top scanning electron microscope (SEM) view of adaxial and abaxial epidermis. Arrow indicates a region used for observation by SEM. The adaxial epidermis has conical-shaped cells, and the abaxial side has flat-shaped cells. Scale bars = 15 μm 3.3 Live Confocal Imaging of Cortical Microtubules in Conical Cells

1. Dissect mature petals from GFP-TUA6 and ktn1–4 GFP-TUA plants. 2. Fold petal in half along the transverse axis onto the doublesided tape (see Note 3). Add water into the petal samples and put onto a coverslip. 3. Observe the petal samples under a Zeiss LSM 880 confocal microscope with Airyscan, excitation at 488, emission 500–570 nm. 4. For observation of microtubules from the top view of conical cells, we use non-folded petal samples. Take serial optical sections at 0.5 μm increments with a 63× oil lens, and images were projected on a plane (i.e., maximum intensity) using Zeiss LSM 880 software (Fig. 3a). 5. For generating 3D images of microtubules in conical cells, Z-stack confocal images were taken from the top view along the Z axis at steps of 0.5 μm and reconstructed into 3D images using Zeiss LSM 880 software (Fig. 3b). 6. For visualization of microtubules from the side view of the conical cells, we focus on the interface of folded petals and take images for projection (Fig. 3c).

322

Xie Dang et al.

Fig. 2 Observation of conical cells using confocal microscopy. (a) Left panel, fold back a petal to expose the interface, allowing the side view of WT and ktn1–4 conical cells at the folding position under a microscope (right panel). Stage 14 petals from WT and ktn1–4 were used. Samples were viewed by a Zeiss microscope (Observer. A1) with BF (bright field) mode. Samples were stained with PI (propidium iodide) and were then observed using a Zeiss confocal (LSM 880) with 561 nm excitation. Scale bars = 10 μm). (b) 3D reconstruction of conical cells from wild-type and ktn1–4 non-folded petals. Z-stacks of images of PI-stained non-folded petals were taken from the top view along their Z axis at steps of 0.8 μm, using Zeiss LSM 880 software to reconstruct the 3D images

3.4 Quantification of Cortical Microtubules in Conical Cells

1. For quantification of cortical microtubules (Fig. 3c, d), download and install Fiji/ImageJ software (https://imagej.net/ software/fiji/) and OrientationJ Macro (http://bigwww.epfl. ch/demo/orientation/index.html?applet=measure#soft). 2. Set images as 8-bit format and input them into ImageJ. 3. Operate OrientationJ Macro in Fiji/ImageJ; choose a conical cell to measure relevant parameters of microtubules (Fig. 4). 4. Choose the “Freehand tool” to profile the conical cells, and then click the “measure” button in the analytical window; the “Coherency” data will be given in the same window. 5. Save the raw data as an Excel file for quantification. 6. Perform statistical analysis of “Orientation” and “Coherency” of microtubules using Excel and GraphPad Prism applications. A coherency coefficient value close to “1” indicates a strongly coherent orientation of the microtubules.

Cortical Microtubule Arrays in Arabidopsis Conical Cells

323

Fig. 3 Analysis of microtubules orientation from both WT and ktn1–4. (a) Visualization of cortical microtubules in conical cells of both WT and the ktn1–4 mutant stably expressing GFP-TUA6. Confocal images of surface projections from the top view of petal adaxial epidermis. Scale bar = 5 μm. (b) 3D reconstruction. Z-stacks of images were taken from the top view along their Z axis at steps of 0.6 μm, using Zeiss LSM 880 software to reconstruct 3D images. (c) Representative confocal images showing microtubule arrangement from the side view of conical cells from folded petals of WT and the ktn1–4 mutant stably expressing GFP-TUA6. Z-stacks of images were taken along their Z axis at steps of 0.6 μm and were processed with maximum projection. Scale bar = 5 μm. (d) Quantification of the orientation degree of microtubules. The microtubule orientation of WT (n = 49) ranges from -50 to 40 degree, SD (standard deviation) = 26.360. By contrast, the microtubule orientation of ktn1–4 (n = 50) was widely distributed from -90 to 90 degree, SD = 53.659. (e) Quantification of microtubule alignment. The coherency value was measured with “OrientationJ,” an ImageJ plug-in, used for calculating the direction of the fibers. A coherency coefficient value close to “1” represents a strongly coherent orientation of the microtubules. The microtubule alignment of WT is well-ordered compared to that of ktn1–4. Values are means ± SD. WT, n = 49, ktn1–4, n = 50. Significant difference was performed with the Student t-test, p < 0.0001

4

Notes 1. Place the sample carefully to avoid damaging of the petal epidermis. 2. Scan as soon as possible to avoid the collapse of the conical cells. 3. Try to avoid squashing the coverslip that covers the petal surface.

324

Xie Dang et al.

Fig. 4 Analysis of microtubule orientation via OrientationJ, Fiji/ImageJ plug-in. (a) Primary menu of Fiji ImageJ. (b) A confocal image used for microtubule orientation measurement. The yellow box was drawn using the freehand tool indicated in the Fiji/ImageJ menu to define the ROI. The red ellipse indicates microtubule orientation. (c) Work interface of OrientationJ operated in Fiji/ImageJ. Raw data measured by OrientationJ are shown in the table

Acknowledgments D.L. is supported by grants from the National Natural Science Foundation of China (Grants 31822003 and 31771344). References 1. Hashimoto T (2015) Microtubules in plants. Arabidopsis Book 13:e0179 2. Muller S, Wright AJ, Smith LG (2009) Division plane control in plants: new players in the band. Trends Cell Biol 19(4):180–188 3. Goodson HV, Jonasson EM (2018) Microtubules and microtubule-associated proteins. Cold Spring Harb Perspect Biol 10:a022608 4. Baskin TI, Meekes HT, Liang BM, Sharp RE (1999) Regulation of growth anisotropy in well-watered and water-stressed maize roots. II. Role of cortical microtubules and cellulose microfibrils. Plant Physiol 119:681–692 5. Dixit R, Cyr R (2004) The cortical microtubule array: from dynamics to organization. Plant Cell 16:2546–2552 6. Paredez AR, Somerville CR, Ehrhardt DW (2006) Visualization of cellulose synthase

demonstrates functional association with microtubules. Science 312:1491–1495 7. McFarlane HE, Do¨ring A, Persson S (2014) The cell biology of cellulose synthesis. Annu Rev Plant Biol 65:69–94 8. Baskin TI (2005) Anisotropic expansion of the plant cell wall. Annu Rev Cell Dev Biol 21: 203–222 9. Smith LG, Oppenheimer DG (2005) Spatial control of cell expansion by the plant cytoskeleton. Annu Rev Cell Dev Biol 21:271–295 10. Elliott A, Shaw SL (2018) Update: plant cortical microtubule arrays. Plant Physiol 176:94– 105 11. Glover BJ (2000) Differentiation in plant epidermal cells. J Exp Bot 51:497–505

Cortical Microtubule Arrays in Arabidopsis Conical Cells 12. Yang Y, Huang W, Wu E, Lin C, Chen B, Lin D (2019) Cortical microtubule organization during petal morphogenesis in Arabidopsis. Int J Mol Sci 20:4913 13. Liu S, Jobert F, Rahneshan Z, Doyle S, Robert S (2021) Solving the puzzle of shape regulation in plant epidermal pavement cells. Annu Rev Plant Biol 72:525–550 14. Lucas J, Shaw SL (2008) Cortical microtubule arrays in the Arabidopsis seedling. Curr Opin Plant Biol 11:94–98 15. Hamada T (2014) Microtubule organization and microtubule-associated proteins in plant cells. Int Rev Cell Mol Biol 312:1–52 16. Whitney H, Bennett K, Dorling M, Sandbach L, Prince D, Chittka L, Glover BJ (2011) Why do so many petals have conical epidermal cells? Ann Bot 108:609–616 17. Noda K, Glover BJ, Linstead P, Martin C (1994) Flower colour intensity depends on specialized cell shape controlled by a Myb-related transcription factor. Nature 369: 661–664 18. Glover BJ, Martin C (1998) The role of petal cell shape and pigmentation in pollination success in Antirrhinum majus. Heredity 80:778– 784 19. Baumann K, Perez-Rodriguez M, Bradley D, Venail J, Bailey P, Jin H, Koes R, Roberts K, Martin C (2007) Control of cell and petal morphogenesis by R2R3 MYB transcription factors. Development 134:1691–1701 20. Irish VF (2008) The Arabidopsis petal: a model for plant organogenesis. Trends Plant Sci 13: 430–436

325

21. Bassel GW, Smith RS (2016) Quantifying morphogenesis in plants in 4D. Curr Opin Plant Biol 29:87–94 22. Fernandez R, Das P, Mirabet V, Moscardi E, Traas J, Verdeil J, Malandain G, Godin C (2010) Imaging plant growth in 4D: robust tissue reconstruction and lineaging at cell resolution. Nat Methods 7:547–553 23. Ovecˇka M, von Wangenheim D, Tomancˇa´k P, Sˇamajova´ O, Komis G, Sˇamaj J (2018) Multiscale imaging of plant development by lightsheet fluorescence microscopy. Nat Plants 4: 639–650 24. Saffer AM, Irish VF (2017) Isolation of mutants with abnormal petal epidermal cell morphology. Plant Signal Behav 12:e138279 25. Ren H, Dang X, Cai X, Yu P, Li Y, Zhang S, Liu M, Chen B, Lin D (2017) Spatio-temporal orientation of microtubules controls conical cell shape in Arabidopsis thaliana petals. PLoS Genet 13:e1006851 26. Dang X, Yu P, Li Y, Yang Y, Zhang Y, Ren H, Chen B, Lin D (2018) Reactive oxygen species mediate conical cell shaping in Arabidopsis thaliana petals. PLoS Genet 14:e1007705 27. Ren H, Rao J, Tang M, Li Y, Dang X, Lin D (2022) PP2A interacts with KATANIN to promote microtubule organization and conical cell morphogenesis. J Integr Plant Biol. https:// doi.org/10.1111/jipb.13281 28. Ueda K, Matsuyama T, Hashimoto T (1999) Visualization of microtubules in living cells of transgenic Arabidopsis thaliana. Protoplasma 206:201–206

Chapter 27 Studying the Organization of the Actin Cytoskeleton in the Multicellular Trichomes of Tomato Zhijing Xu, Xiaolu Qu, Shuang Wu, and Pengwei Wang Abstract Trichomes are unique polarized cells of the plant epidermis that provide an excellent model system for studying the plant cytoskeleton. Unlike Arabidopsis trichomes that are unicellular with a typical threebranch shape, the trichomes in tomato (Solanum lycopersicum) are multicellular with additional morphology and function diversity. Technically, it is hard to image tomato trichomes at the subcellular level because of their size and because they can be easily damaged. Here, we describe the methods we have used for the visualization and quantification of cytoskeletal arrangements in tomato trichomes which are at different developmental stages, using both live-cell imaging and phalloidin staining after fixation. Key words Cytoskeleton, Trichomes, Tomato, Live-cell imaging, Phalloidin staining

1

Introduction Trichomes are specialized epidermal cells that play important roles in plant resistance to biotic and abiotic stresses [1–3]. Early genetic studies in Arabidopsis identified and characterized a number of mutants that had deformed trichomes, and these deformations were later related to cytoskeletal dysfunction. The best examples are the mutants in components of the SCAR/WAVE and ARP2/3 complexes which are multi-subunit complexes that regulate actin nucleation and branching [4–7]. Loss-of-function mutants in most of the subunits of these complexes show trichome expansion and morphological defects (Fig. 1a–d) caused by cytoskeletal disorganization. Similar phenotypes have been described in many plant species including cotton and tomato [8–10]. Compared to the unicellular trichomes of Arabidopsis, the epidermal trichomes of tomato are multicellular with much more functional and morphological diversity [11]. The structure of tomato trichomes is also

Authors Zhijing Xu and Xiaolu Qu have equally contributed to this chapter. Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_27, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

327

328

Zhijing Xu et al.

Fig. 1 Examples of trichomes in Arabidopsis and tomato. (a, b) SEM images of Arabidopsis trichomes from either wild type (Col-0) or a mutant with a dysfunctional SCAR/WAVE complex (nap1). Scale bar = 100 μm. (c) Arabidopsis trichome (Col-0, nap1) imaged using light microscopy. Scale bar = 50 μm. (e, f) SEM images of tomato trichomes from either wild type or a mutant with a dysfunctional SCAR/WAVE complex (Slscar2). Scale bar = 100 μm. (g) The morphology of different types of tomato trichomes. Scale bar = 30 μm

controlled by the dynamics and organization of the cytoskeleton, and its disruption (either genetically or chemically) also produces distorted and deformed cells [10] (Fig. 1e, f). In most tomato cultivars (e.g., Micro-Tom), there are different types of epidermal trichomes (Fig. 1g), and their distribution shows strong tissue and developmental specificity [10, 11]. For example, type I and II trichomes mainly appear on young leaves and tender stems near the growing point, type IV trichomes mainly appear on the hypocotyls, and type V, type VI, and type VII trichomes are mainly distributed on mature stems [10]. The temporal and spatial distribution of tomato epidermal trichomes is therefore a good system to study cell differentiation, cytoskeletal dynamics, and polarity in a more complex multicellular system. Here, we introduce methods of how to study the cytoskeletal arrangement of tomato epidermal trichomes using both live-cell imaging and phalloidin staining and the procedures that we have used for sample preparation, image collection, and analysis.

Cytoskeleton Organization in Tomato Trichomes

2

329

Materials

2.1 Plant Materials and Sample Preparation

1. Wild-type or transgenic tomato (either Micro-Tom or Ailsa Craig, “AC”) expressing actin cytoskeleton markers, 35S: Lifeact-GFP [10] (see Note 1). 2. 1/2 Murashige and Skoog (MS) medium: 5 g of MS basal salt mixture and 10 g of sucrose in 0.8 L of distilled water. Adjust pH to 5.8 using 1 M KOH, and add distilled water to a final volume of 1 L. Add 8 g plant agar. Autoclave for 20 min at 121 °C, and cool to 45–50 °C before pouring into petri dishes under sterile conditions. 3. Glass slides and coverslips. 4. Razor blade and fine forceps. 5. Nail polish. 6. 75% ethanol. 7. 3% sodium hypochlorite. 8. Sterile petri dishes. 9. Latrunculin B.

2.2 Fixation and Actin Staining

1. PEM buffer: 100 mM PIPES, 5 mM EGTA, 4 mM MgCl2, adjust pH to 6.9 with KOH. 2. Fixation buffer: 2% (w/v) formaldehyde in PEM buffer. 3. Staining buffer: 1% (v/v) glycerol and 0.198 μM Alexa Fluor 488 phalloidin (Invitrogen, A12379) in PEM buffer. 4. Six-well culture plate. 5. Parafilm. 6. Hotplate magnetic stirrer and a magnetic bead. 7. Vacuum pump with a desiccator.

2.3 Microscopy and Computer Programs

1. Confocal laser scanning microscopy (Leica SP8) equipped with 63× oil immersion objectives (NA = 1.4). 2. TM3030 Plus scanning electron microscope (Hitachi). 3. ImageJ/Fiji.

3

Methods

3.1 Treatment of Cytoskeleton Depolymerizing Drugs

In our study, all plants were grown in growth rooms with a 16 h day length (26 °C) and 8 h dark (22 °C) regime (see Note 2). 1. For surface sterilization, wash the seeds with 75% ethanol solution, and soak in 3% sodium hypochlorite for 15 min. Then wash the seeds 3 times in sterile water.

330

Zhijing Xu et al.

Fig. 2 Latrunculin B and oryzalin treatment causes massive disruption to trichome morphology. (a) SEM images of WT trichomes. (b) Treatment with oryzalin resulted in isotopically expanded cells. (c) Treatment with latrunculin B reduced cell elongation. Scale bar = 500 μm

2. Plate the sterilized seeds on solid 1/2 Murashige and Skoog (MS) medium for 7–10 days. 3. Place sprouted tomato seeds on 1/2 MS medium supplemented with 20 mM latrunculin B (an actin depolymerization drug) or 20 mM oryzalin (a microtubule destabilizing drug). The hypocotyl or cotyledon trichomes are imaged 4 days after transfer. 4. Image the samples with normal light microscopy or scanning electron microscopy for the morphological analysis (Fig. 2a–c). 3.2 Preparation of Tomato Trichomes for Live-Cell Imaging

1. Paint a rectangle using nail polish on a microscopy slide, and let this dry at room temperature (Fig. 3a). This generates a small chamber in which the samples are placed. 2. Select the tomato tissues that need to be observed (see Note 3), and carefully dissect them; try not to touch the surface during this procedure. 3. To image the surface of the stem, cut it into thin vertical slices. 4. To image trichomes from leaf tissue, remove the main leaf veins, and cut the flat leaf tissue into thin slices with a sharp razor blade (Fig. 3b). 5. Carefully place the cut sample in the chamber made by the nail polish, mount with distilled water, and place a coverslip on top; seal the edge of the coverslips with nail polish to prevent dehydration (see Note 4). 6. Image the sample using a confocal microscope with the appropriate settings for the desired fluorophores (Fig. 3c, d).

3.3 Tomato Trichome Fixation and Imaging

1. 10-day-old tomato seedlings are used for tissue fixation. 2. Prepare fresh fixative solution. Add paraformaldehyde powder into PEM buffer, heat the mixture, and stir at 60 °C until a clear solution is formed. Adjust the pH to approximately 7.0 with KOH if required.

Cytoskeleton Organization in Tomato Trichomes

331

Fig. 3 Sample preparation and imaging of tomato trichomes. (a) Draw a rectangle on the slide using nail polish. Add approximately 200 μL of water before loading the sample. (b) Thin sections dissected from tomato stem or leaf tissue are used for imaging. (c) Light microscopy of tomato trichome. Scale bar = 100 μm. (d) Confocal microscopy of tomato trichomes labeled using GFP-Lifeact. Scale bar = 50 μm

Fig. 4 Phalloidin staining of tomato trichomes. (a) Actin filaments in the leaf primordia of tomato. The inset shows the corresponding bright field image. (b) Actin filaments in cotyledon pavement cells. (c) Actin filaments in hypocotyl epidermal cells. (d) Actin filaments in different types of tomato trichomes. Scale bar = 40 μm in all images

332

Zhijing Xu et al.

3. We use this method to label actin filaments in hypocotyl, cotyledon, and leaf primordia. To do this, tear off the epidermis of the hypocotyl, cut the cotyledons into 0.5 cm × 0.5 cm pieces, or cut off the leaf primordia. 4. Put these samples into the fixative for 30 min. For better infiltration of the fixative, place the samples in a vacuum for 10 min, and then slowly release the vacuum. Shake the tube gently to remove air bubbles that attach to the samples. 5. Wash samples with PEM buffer for 5 min twice. 6. Place the samples onto coverslips and put them on a Parafilm (see Note 5). 7. Add 70–100 μL of the staining buffer to the samples, and incubate them overnight at 4 °C in a humid chamber. 8. On the same coverslips, wash samples with PEM buffer once before observing using a confocal laser scanning microscopy equipped with a 63× oil objective lens (Fig. 4). 9. Images were stitched in Fiji with plug-in “Stitching” [12] if necessary. 3.4 Anisotropy Analysis Using FibrilTool

We use FibrilTool to quantify the average orientation and anisotropy of the cytoskeleton in a region of interest (ROI) [13]. 1. Download the FibrilTool plug-in (https://www.nature.com/ articles/nprot.2014.024), and install it to ImageJ. To do this, place the downloaded “supplementary data 1.txt” file to the Fiji macro toolset folder, and restart the software. 2. Open Fiji and open an image; click the double arrow in the toolset (Fig. 5a). 3. Select the supplementary data 1 that links to the plug-in (see Note 6). 4. The FibrilTool plug-in appears as a striped square in the toolset, and double click it (Fig. 5b). 5. Select the parameter, such as channel of image, multiplier length (which demonstrates the average direction and anisotropy of the filaments), and the display number of ROIs. 6. Define the ROI using the shape tool, and the data will appear automatically. The angle of the line segment represents the average direction of the array, and the length of the line segment is proportional to the anisotropy of the array (Fig. 5c). 7. Paste output data including the average orientation and anisotropy into the excel sheet named “Supplementary Table 1” for calculation.

Cytoskeleton Organization in Tomato Trichomes

333

Fig. 5 FibrilTool analysis of the cytoskeleton in tomato epidermal trichomes. (a) Open Fiji and open an image; click the double arrow in the toolset. (b) Use the FibrilTool plug-in. (c) Select the parameter and define the ROI; analyze the output data 3.5 Cytoskeleton Dynamics Analysis Using Kymograph

In living cells, the cytoskeleton rearranges dynamically, and this process can be captured and quantified using mathematical models. This is crucial to understanding the underlying mechanism of the process [14]. To do this: 1. Open the software Fiji and the images (Fig. 6a). 2. Select the slice count, and decide whether to display flip vertically, or convert to grayscale (Fig. 6b). 3. Draw a line through where the fiber is to be measured (Fig. 6c). 4. Click “Analyze > Multi kymograph > Multi kymograph.” 5. Select the width of the generated kymograph lines; the recommended width is 5 pixels (Fig. 6d). 6. Kymograph can display fiber movement or static status: the Y-axis represents the time that the image series was taken, and the X-axis represents the displacement of the cytoskeletal movement (calculated by scale).

334

Zhijing Xu et al.

Fig. 6 Kymograph analysis of cytoskeleton dynamics in tomato epidermal trichomes. (a) Open Fiji and open a time series. (b) Select parameters. (c) Draw a line through, and choose Analyze-Multi kymograph and then Multi kymograph. (d, e) Set the linewidth and analyze the output data

4

Notes 1. Micro-Tom and Ailsa Craig (AC) are the most commonly used tomato cultivars, both of which are robust for Agrobacteriamediated transformation. Please refer to [15] for the protocol for tomato transformation. 2. Keep distance between tomato plants to avoid damaging the surface of trichomes during cultivation. 3. Select the tomato tissues that need to be observed. To observe type I, II, and IV epidermal trichomes, we recommend that you select young leaves near the growth tips and hypocotyls; to observe type V, VI, and VII epidermal trichomes, we recommend that you select young stems. 4. For sections from a leaf tissue, put them on microscope slide vertically to avoid physical damage by the coverslips. 5. The staining solution forms a strong surface tension on the Parafilm, and this is helpful in saving the dye.

Cytoskeleton Organization in Tomato Trichomes

335

6. The name of FibrilTool plug-in is shown as “supplementary data 1.txt” in the literature [13]; it can be renamed, but we have kept the original name here for demonstration.

Acknowledgements The project was supported by NSFC grants (no. 32261160371, 92254307, 91854102), the Foundation of Hubei Hongshan Laboratory (2021hszd016) to P.W. References 1. Karabourniotis G, Papadopoulos K, Papamarkou M, Manetas Y (1992) Ultraviolet-B radiation absorbing capacity of leaf hairs. Physiol Plant 86:414–418 2. Werker E (2000) Trichome diversity and development. Advances in botanical research incorporating advances. Plant Pathol 31:1–35 3. Kennedy GG (2003) Tomato, pests, parasitoids, and predators: tritrophic interactions involving the genus Lycopersicon. Annu Rev Entomol 48:51–72 4. Steffen A, Faix J, Resch GP, Linkner J, Wehland J, Small JV, Rottner K, Stradal TE (2006) Filopodia formation in the absence of functional WAVE- and Arp2/3-complexes. Mol Biol Cell 17:2581–2591 5. Nicholson-Dykstra SM, Higgs HN (2008) Arp2 depletion inhibits sheet-like protrusions but not linear protrusions of fibroblasts and lymphocytes. Cell Motil Cytoskeleton 65: 904–922 6. Suraneni P, Rubinstein B, Unruh JR, Durnin M, Hanein D, Li R (2012) The Arp2/3 complex is required for lamellipodia extension and directional fibroblast cell migration. J Cell Biol 197:239–251 7. Wu C, Asokan SB, Berginski ME, Haynes EM, Sharpless NE, Griffith JD, Gomez SM, Bear JE (2012) Arp2/3 is critical for lamellipodia and response to extracellular matrix cues but is dispensable for chemotaxis. Cell 148:973–987 8. Basu D, Le J, El-Essal Sel D, Huang S, Zhang C, Mallery EL, Koliantz G, Staiger CJ, Szymanski DB (2005) DISTORTED3/ SCAR2 is a putative arabidopsis WAVE complex subunit that activates the Arp2/3 complex

and is required for epidermal morphogenesis. Plant Cell 17(2):502–524 9. Yu Y, Wu S, Nowak J, Wang G, Han L, Feng Z, Mendrinna A, Ma Y, Wang H, Zhang X, Tian J, Dong L, Nikoloski Z, Persson S, Kong Z (2019) Live-cell imaging of the cytoskeleton in elongating cotton fibres. Nat Plants 5(5): 498–504 10. Chang J, Xu Z, Li M, Yang M, Qin H, Yang J, Wu S (2019) Spatiotemporal cytoskeleton organizations determine morphogenesis of multicellular trichomes in tomato. PLoS Genet 15(10):e1008438 11. Luckwill LC (1943) The genus Lycopersicon: A historical, biological, and taxonomic survey of the wild and cultivated tomato. Aberd Univ Stud 120:1–44 12. Preibisch S, Saalfeld S, Tomancak P (2009) Globally optimal stitching of tiled 3D microscopic image acquisitions. Bioinformatics 25(11):1463–1465 13. Boudaoud A, Burian A, Borowska-Wykre˛t D, Uyttewaal M, Wrzalik R, Kwiatkowska D, Hamant O (2014) FibrilTool, an ImageJ plug-in to quantify fibrillar structures in raw microscopy images. Nat Protoc 9(2):457–463 14. Mangeol P, Prevo B, Peterman EJ (2016) KymographClear and KymographDirect: two tools for the automated quantitative analysis of molecular and cellular dynamics using kymographs. Mol Biol Cell 27(12):1948–1957 15. Sun HJ, Uchii S, Watanabe S, Ezura H (2006) A highly efficient transformation protocol for Micro-Tom, a model cultivar for tomato functional genomics. Plant Cell Physiol 47(3): 426–431

Chapter 28 Light Microscopy Technologies and the Plant Cytoskeleton Timothy J. Hawkins Abstract The cytoskeleton is a dynamic and diverse subcellular filament network, and as such microscopy is an essential technology to enable researchers to study and characterize these systems. Microscopy has a long history of observing the plant world not least as the subject where Robert Hooke coined the term “cell” in his publication Micrographia. From early observations of plant morphology to today’s advanced superresolution technologies, light microscopy is the indispensable tool for the plant cell biologist. In this mini review, we will discuss some of the major modalities used to examine the plant cytoskeleton and the theory behind them. Key words Light microscopy, Laser scanning confocal microscopy, Super-resolution, Airyscan, Structured illumination, Lightsheet microscopy, Total internal reflection microscopy

1

Introduction Light microscopy is a critical technology for the study of the cytoskeleton in plants, allowing the researcher to visualize and quantify the changes of the networks in development and in response to external cues or stress. It is often microscopic interrogation of the cytoskeletal network of a mutant that can reveal the underlying irregularities which are responsible for the phenotype. The importance of light microscopy to plant cell cytoskeleton research is reflected in the diverse array of modalities which are routinely deployed by plant cytoskeletal researchers, each with their own technical benefits to reveal distinct aspects of the cytoskeleton, be it rapid imaging speed for dynamics, super-resolution for accessory protein distribution and subtle network features, or the long-term imaging for developmental questions. This review seeks to give an overview of some imaging modalities widely used to study the cytoskeleton in plants and the theory behind each technology and its application. While not a comprehensive analysis, this chapter can provide the reader with the background knowledge

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_28, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

337

338

Timothy J. Hawkins

of these imaging technologies to aid in their experimental design and choice of modality for their biological question.

2

Imaging Modalities

2.1 Laser Scanning Confocal Microscopy (LSCM)

Confocal is a mainstay of plant microscopy, particularly for its ability to image the larger cells and complex tissues of plants where widefield microscopy can be hampered by the extensive light scattering. The confocal microscope produces optical sections within a sample through the exclusion of out of focus light from areas outside of the focal plane. This results in an image with increased clarity, signal to noise, and resolution. The confocal microscope has two pinhole apertures, one in front of the laser source (excitation pinhole) and a second before the detector (emission pinhole) (Fig. 1). These pinholes are at conjugate planes leading to the term confocal. The thinness of the optical section is often in the order of a few hundred nanometers but can be varied by the adjusting the aperture of the detector pinhole. For imaging of the cytoskeleton, where subcellular resolution is required with the highest signal to noise to discern filaments and bundles, an aperture of 1 airy unit is typically chosen. This produces a thin optical section, where we have the optimal balance between resolution and sensitivity. For weak samples, it is possible to increase the pinhole aperture to gain more signal, but this is at the expense of confocality and section thinness, and with today’s highly sensitive detector arrays, this is now rarely done. As only a point is illuminated to form an image, the point source needs to be scanned across the sample in a raster pattern. Optical sectioning allows the confocal to acquire a Z-stack, a series of optical sections in the Z dimension achieved by moving the sample relative to the focal plane by small increments, the size of which are calculated using the Nyquist theorem. Once these overlapping sections are generated, these can then be assembled to form a 3D object; a 3D image of the sample which can then be rotated and observed from new angles (Fig. 1b).

Confocal Imaging Considerations Once the lens has been chosen and the channels selected, there are three principle considerations when acquiring images in confocal microscopy, and these are (1) image size, (2) scan speed or pixel dwell time, and (3) the balance between detector settings, e.g., gain and laser intensity. Image size ideally should be calculated according to the Nyquist theorem as discussed above to maximize the resolution. Scanning speed is typically high when selecting an area to image for ease of navigation but is reduced to achieve final images. As the laser scans across the sample, passing over each area of the sample which

Light Microscopy and the Cytoskeleton

339

Fig. 1 (a) Schematic diagram of the light path in a confocal microscope. Laser light is directed into the objective by a beam splitter and focused at the focal plane within the sample. Light emitted from molecules at the focal plane enters the objective lens and passes through the pinhole to reach the detector. Light from out of focus planes in the specimen both above and below are obstructed by the pinhole diaphragm and do not reach the detector. (b) Example confocal image of actin filaments in Nicotiana benthamiana leaf epidermal cells

equates to a pixel, the detector measures photons emitted. If the time the laser spends at each point, the dwell time, is too little, then less photons or signal are collected resulting in an image with low signal to noise. However, if the laser is scanning slowly, it dwells at each pixel for longer, acquiring more signal and producing an image with significantly higher signal to noise (SNR) and clarity. The signal can be improved with the use of either higher laser intensity which may result in phototoxicity and photobleaching or increasing the detector gain. Increasing gain can improve signal without the need for increased radiation. However, increasing detector gain can also introduce noise into the image necessitating a balance between these two. At this stage the pixel values in the image should be evaluated for saturation and range. For example, in 8-bit image, each pixel can have a value between 0 and 255. If a significant number of pixels are saturated at 255, then fine intensity differences in the image are lost which may represent important biological differences in protein distribution. To acquire an optimal image where this full range of the detector is used, the detector gain and laser intensity should be adjusted while using a range indicator palette where pixels at 0 or 255 are highlighted with different colors. These same considerations apply when using camera-based

340

Timothy J. Hawkins

imaging systems but here balancing laser or bulb illumination with camera exposure. Importantly when imaging live samples, often a compromise between all these considerations needs to be selected to scan fast enough to capture dynamic behavior but while still capturing images of high enough SNR without the use of laser powers that would be damaging to the viability of the cell (phytotoxicity) or the emission of the fluorophore (photobleaching). When imaging dynamics in living cells, often spinning disk confocal (SDC) has been chosen over standard point scanners as they can typically be faster and more sensitive and cause less photobleaching which is particularly useful when working with low intensity or sensitive samples. The SDC achieve this though the combination of rapid disc scanning of multiple point sources coupled to sensitive acquisition with EMCCD cameras. A limitation of SDC is that the pinholes in the disc are fixed, and so this is not optimal for all lenses; plus there can be a degree of pinhole cross talk which may need to be resolved with the application of deconvolution. However, while still a useful tool, this gap between SDC and LSCM has reduced dramatically or even been abolished with the development of more modern rapid LSCM scanning and detection methods or even non-confocal implementations such as Lattice SIM. When capturing a series of optical sections to generate a Z-stack, the spacing between each image in the axial or z direction needs to be calculated with the Nyquist theorem. This gives the best volumetric presentation of the cell. However, the large size of a plant cell and the number of sections required can prove challenging for a confocal to capture rapidly enough to view cytoskeletal dynamics across the whole cell. Therefore, rapid dynamic cytoskeletal imaging over short time periods is often confined to a single plane or a small group of planes. As much as 90% of the plant cell volume can be occupied by the vacuole, and much of the cytoskeleton is within a limited strip of cytoplasm at the cortex in interphase; imaging is thus still often limited to a small window in XY. Conversely, if longer-term imaging is required of a larger field of view, or division arrays, then Z-stacks are acquired at wider intervals. To address these limitations, there have been ingenious approaches to imaging the cytoskeleton by capturing short imaging bursts as part of longer-term observations. This combination has been successful in allowing the simultaneous acquisition of longterm changes and short-term rapid dynamics while reducing photobleaching [1]. Going forward, new technologies such as lattice lightsheet will enable the observation of full plant cell volumes at resolutions and speeds rapid enough to capture cytoskeletal dynamics yet over time periods long enough for cell growth and development. As discussed later, to date LLS use with plant samples has been very limited, but with the advent of easy to use inverted

Light Microscopy and the Cytoskeleton

341

systems, this modality’s application to the plant cytoskeleton will certainly grow. 2.2 Super-Resolution Microscopy (SRM)

The use of super resolution in plants slightly lagged behind the uptake seen for, in particular, animal cell culture, due to plantspecific challenges and limitations such as spherical aberrations and light scattering caused by the variety of refractive indices of components of the plant cell; birefringence of the cell wall; and autofluorescence. There are now numerous publications using SRM in plants, but compared to other modalities, it is still a somewhat nascent field for plants. With the advent of a wider selection of commercially available technologies which are easier to use and have better success with plant tissue, this is now a growth area where plant SRM will become routine. Resolution achievable with light microscopy is restricted by the diffraction limit first described by Ernst Abbe. The ability to distinguish two points with a microscope depends on the diffraction of the light from each point source. As this point source of light is diffracted, this pattern produces a series of concentric rings of minima and maxima referred to as Airy disks. The extent to which these disks overlap for adjacent objects determines if these two points can be distinguished as separate objects. The Rayleigh criterion defines two objects as resolved if the distance is sufficiently large enough to separate the Airy disks by at least one minimum (dark fringe between disc and the first ring). The size of these Airy disks is dependent on the wavelength of light and the NA of the lens. Using shorter wavelengths produces smaller Airy disks and a higher resolution, e.g., greater resolution is achievable with blue light compared to red. Therefore, using visible wavelengths of light for microscopy results in a limit for resolution, the diffraction limit, or barrier. For example, the maximum theoretical resolution achievable by conventional microscopy is approximately 200 nm using a 1.4NA lens and 405 nm excitation light. Super-resolution microscopy or sub-diffraction microscopy techniques break this diffraction barrier through a series of different methods. This has been and continues to be a rapidly growing area with different approaches being published on a regular basis. The majority of commercial offerings currently fall into three classical approaches of (1) structural illumination[2], (2) STED [3], or (3) localization microscopy [4] with additional modalities now gaining popularity including systems with novel confocal detectors such as Zeiss Airyscan, SoRa Spinning Disk [5], or the in-camera processing of differences in the symmetry of point emitters, super-resolution radial fluctuations (SRRF) [6] Super-resolution microscopy has been successfully used to study a variety of aspects of cell biology within plant cells including 3D/SR-SIM, organization of the plasmodesmata [7, 8], viral replication/movement complexes [9, 10], cortical microtubules[11–

342

Timothy J. Hawkins

13], plant centromere architecture [14], and dynamics of cellulose synthase [15]; localization microscopy, perinuclear actin [16], cellulose microfibrils [17], cortical microtubules [18], and RNAPII localization within the nucleus [19]; and STED, PIN2 membrane distribution [20] and chromosomal imaging [21]. This only represents a fraction of the field, and it’s worthy to note that the majority of cases have used SIM, while STED is still rarely seen. In this chapter, we will discuss in more detail two of the most wildly used examples of SRM modalities in plants, 3D-SIM and the confocal-based Airyscan. 2.3 ThreeDimensional Structured Illumination Microscopy (3D-SIM)

Structured illumination microscopy can double the spatial resolution of fluorescence microscopy in three dimensions and is based optically on a precise widefield microscopy implementation. However, the illuminating light is spatially structured or rather has a three-dimensional pattern (Fig. 2). This is often created with the use of a mechanically manipulated diffraction grating, spatial light modulator, or reflection beam splitting. SIM utilizes the concept of moire patterns. When two such patterns are superimposed, a beat pattern called moire fringes is produced in the resulting image (Fig. 2c). In 3D-SIM one pattern is the precise special modulation created in the illuminated light (Fig. 2b), and the second is the spatial distribution of the fluorescently labeled details within the sample (Fig. 2a). As these two are superimposed, moire fringes are created in the resulting microscopy image (Fig. 2c). Although sub-diffraction details cannot be observed directly, we can observe the effects they cause through the creation of these moire patterns. These details beyond the diffraction limit are encoded in the image and can be decoded and revealed by computation. In reciprocal or Fourier space, spatial information in the sample occupies different locations with low-resolution information close to the origin whereas high-resolution is further away (Fig. 2d). The lateral resolution limit of a microscope is represented by a circle centered at the origin, whose radius is proportional to the numerical aperture of the lens and inversely proportional to the excitation wavelength. The high-resolution information within the sample which lies beyond this circle cannot be observed and is lost. In frequency space, a sinusoidal illumination pattern corresponds to three frequency points (Fig. 2e). Effectively the product of this pattern and labeling pattern within the sample shifts the information outside the circle into the circle where it is now observable as moire fringes (Fig. 2f) [22]. The high-resolution information at this stage is mixed with low-resolution information, and so to separate these mathematically, a series of images are required where the pattern is moved through a series of phases. Furthermore, this resolution enhancement only occurs perpendicular to the stripes, and so to give isotropic resolution enhancement, the pattern needs to be rotated by 60 degrees (Fig. 2g). Following the same principles, in

Light Microscopy and the Cytoskeleton

343

Fig. 2 3D-SIM microscopy. (a) Pattern representing the labeled details within the specimen below the diffraction limit. This is unknown. (b) The known pattern created in the illuminating light. (c) Demonstration of the appearance of moire fringes as two fine patterns are superimposed or multiplied. (d) Image information represented in Fourier space or the frequency domain. The central circle represents the observable region accessible by a light microscope, whose radius is proportional to the numerical aperture of the lens and inversely proportional to the excitation wavelength. The high-resolution information within the sample which lies beyond this circle cannot be observed and is lost. (e) In frequency space, a sinusoidal illumination pattern corresponds to three frequency points. (f) The frequencies in the light gray circles are translated into the normal observable region and become observable. (g) The structured illumination pattern is applied at three different orientations, so lateral resolution is enhanced isotopically. (h) Representative image acquired using 3D-SIM. Tobacco BY2 suspension culture cell. Immunofluorescence antibody labeling of tubulin in an anaphase spindle

3D-SIM, axial resolution enhancement is achieved through the creation of a distinct illumination pattern in this dimension also. In practice to acquire a 3D-SIM image, the spatially modulated light is moved through five phases, and this movement is then repeated again at two additional angles spaced by 60 degrees at each Z slice within the sample volume. Therefore, the dataset required for a single reconstructed super-resolved section consists of 15 images. As resolution is double in all three dimensions, this results in an eightfold smaller observable volume. 2.4 Airyscan Laser Scanning Confocal Microscopy

Airyscan confocal is now being routinely used for plant samples. Airyscan provides a 1.7× resolution improvement in all dimensions and a 4–8×-fold improvement in SNR. Airyscanning makes use of light that is normally rejected by a standard confocal. In a standard confocal closing, the pinhole will narrow the detection PSF and raise the resolution. However, upon reducing the pinhole diameter, we also reduce detection efficiency giving poor signal to noise (SNR) images. For high-resolution improvement, the pinhole would need to be closed to zero, so a value of 1 AU is usually used, a balance between resolution and SNR. In Airyscanning, the pinhole is removed, and all emitted light is collected by being presented across an array detector, a compound eye, or hexagonal

344

Timothy J. Hawkins

Fig. 3 (a) Schematic representation of the Airyscan detector array. Each detector unit acts as an effective 0.2 AU pinhole. (b) Example Airyscan confocal image of actin in Arabidopsis guard cells labeled with the mNeonGreen-Lifeact actin marker

arrangement of 32 GaAsP (gallium arsenide phosphide) detector elements which function as individual pinholes. Each of these detectors within the array is displaced off the optical axis and is small enough to be considered as an individual pinhole with an aperture of 0.2 AU (Fig. 3a). As the displacement of each element is known, then these measured intensities can be shifted back to the correct position and summed. However, as out of focus light remains at this stage, deconvolution is applied to the image acquired by each individual detector element. Each resulting image is weighted to acknowledge the detector positional information and unique optical transfer function (OTF) to form a reconstructed sum image with increased resolution and SNR. Examples of Airyscan LSCM microscopy of plants include both live and fixed cortical microtubules [13, 23], mitosis/cell division [24], plasma membrane and internal membrane nanodomain imaging [25, 26], subcellular localization of ANNEXIN1 [23], and visualization of fossilized pollen [27]. Although Airyscan does not achieve the same resolution improvements as other SRM technologies, it can image deeper into large plant cells and tissues and is simple to use. Very recently however, new deconvolution approaches such as joint iterative deconvolution (Zeiss Airyscan jDCV) have been designed for this type of data, further increasing the achievable resolution to 90nm. An important recent addition to super-resolution microscopy techniques is that of physical magnification through expansion microscopy allowing resolution improvements using diffraction limited imaging modalities such as spinning disk or confocal [28]. These techniques, importantly, both democratize the access to super-resolution imaging as many SRM technologies can be too costly and permit super-resolution analysis of dense and complex samples such as the brain where super-resolution microscopy would be impossible. We have successfully adapted and further developed

Light Microscopy and the Cytoskeleton

345

Fig. 4 (a) Basic schematic representation of the light path in a conventional Gaussian sheet, lightsheet microscope. Laser light is shaped to produce a laser sheet by a cylindrical lens. The thin midsection or waist of this sheet defines the thickness of the imaged optical section. This sheet is positioned at the focal plane of the imaging objective lens. Typically, this objective is a water dipping lens submerged into the water-filled imaging chamber. The sample is suspended into the chamber in the path of the excitation sheet. Image series are generated by transition of the specimen through the sheet. Unlike other imaging modalities, the sample maybe rotated and a new section series collected. (b) A lightsheet microscopy image of FABD2-GFP labeled actin filaments in a growing Arabidopsis root

the technique for use with plant cells and tissues, and this will be discussed in Chap. 10. 2.5 Lightsheet Fluorescence Microscopy (LSFM)

This modality, often also referred to as selective plane illumination microscopy (SPIM), generates a sheet of laser light which selectively only illuminates a single slice within the specimen. The emitted light is captured by an objective typically at 90 degrees to the excitation source (Fig. 4a). In a conventional lightsheet, the sample is suspended in a water-filled chamber and observed with a water dipping lens. The sample is moved while the sheet remains static, and in doing so, a series of images is produced of different optical sections creating a dataset similar to a z series on the confocal. Here however, the optical section thickness is determined by the thickness of the excitation sheet. Lightsheet microscopy has three significant advantages compared to confocal microscopy which make it a suitable modality for developmental experiments. Firstly, only the section of the sample being imaged within the laser sheet is exposed to the excitation light. Areas of either side of this sheet are not exposed, and so there is a reduction in out of focus light and an increase in the viability of those sample areas with no phototoxicity or photobleaching. This is distinct from confocal microscopy, as although the pinhole enables a singular optical section to be captured, areas of the sample outside of the focal plane still exposed the excitation laser light. As such, as a confocal Z series stack is

346

Timothy J. Hawkins

acquired, out of focus areas of the sample continue to be exposed during each slice acquisition leading to high laser dosage and reduced sample viability. Secondly, lightsheet image acquisition is very fast compared to confocal. With confocal microscopy, image acquisition is determined by the scanning speed of the laser excitation point. However, in lightsheet the whole field of view of the optical section is captured with a EMCCD or SCMOS camera with exposure times typically much shorter than a full confocal laser raster scan over a comparable sample area. Thirdly, because the sample is suspended in the imaging chamber, it can be freely rotated with respect to the illumination sheet. As a result, z series can be acquired of the sample from multiple angles which can then be incorporated together to produce a three-dimensional dataset with optimal clarity throughout. This can be necessary as the integrity of the sheet can be attenuated as it travels deep within the sample or can be obstructed by components of the specimen. One way to improve this is the switching of the sheet projection from right to left or the rotation of the sample creating individual datasets in which these areas are visualized optimally with less image disruption. Combining these series gives a dataset with the best aspects of all image series. This vertical orientation of the specimen suspended in the imaging chamber is ideal for the imaging of root development where the sample is now kept in its natural orientation relative to the direction of gravity during imaging. This allows for correct gravitropic sensing and guidance, while aerial tissues remain aerated. Furthermore, lightsheet microscopy can image all developing root hairs as they grow radially out from the root surface into the 3D space of the imaging chamber and so is an excellent modality to study root hair formation and tip growth [29]. Long-term developmental imaging of plants can last from hours to days and is achieved through the control of environmental conditions with a continuous supply of nutrients within the imaging chamber and the diurnal control of growth lighting in the microscope [30]. The study of Arabidopsis root growth and development has been one of the most popular subjects of lightsheet microscopy because of its small size and thin roots which are almost transparent. However, other plant species have been imaged including Nicotiana benthamiana, Nicociana tobbacum, Lactuca sativa, Medicago sativa, Lycopersicon esculentum, Gerbera hybrida, Hordeum vulgare, and Petunia multiflora [29, 31–34]. In the study of the plant cytoskeleton, lightsheet microscopy has currently been used to examine both actin and microtubule networks and their dynamic accessory proteins. Because of the tissues suitability, this work has generally been focused on the root system. One such example is the imaging of growing roots of GFP-TUA6 expressing katanin1-2 (ktn1-2) mutants which showed ectopic longitudinal cell divisions in the calyptrogen,

Light Microscopy and the Cytoskeleton

347

procambium, and the mature parts of the roots during the formation of lateral root primordia [34]. In addition, lightsheet has also been used successfully to image mitotic microtubule transitions and microtubule orientation during cell division using fluorescently tagged tubulin, microtubule-associated proteins, and microtubule end-binding proteins [12, 24]. Lightsheet microscopy has imaged the effects of the Arabidopsis root hair mutant der-1-3 (deformed root hairs 1), a single-point ACT2 gene mutation, which showed changes in overall actin organization, root growth, and plant development [35]. Furthermore, application of this modality has been extending to other plant species such as Medicago roots where GFP-MBD (Microtubule Binding Domain) allowed quantitative determination of cell divisions in the epidermis and cortex of the root meristem[36] and the actin cytoskeleton using the FADB2 marker in barley. Lightsheet imaging of roots can also enable the live-cell imaging of the root hair interactions with bacteria and the formation of root nodules. FABD2 expressing Medicago plants have been co-cultured with RFP-labeled Sinorhizobium meliloti in the microscope where roots hairs associated with the bacteria show accelerated tip growth and compromised polarity which is correlated with large rearrangements and accumulation of abundant actin filaments [29]. Lightsheet microscopy has one of the most diverse ranges of implementation ranging from commercial systems, homemade and optical engineering to open source projects. A related modality Lattice lightsheet (LLS) uses a lattice illumination sheet which provides improved resolution enabling subcellular volumetric imaging. So far there are few examples of LLS use in plant species. One such example examined the subcellular localization pattern of ANNEXIN1-GFP in root trichoblast cells within the meristematic and elongation root zones where it accumulates around the nucleus, close to the nuclear envelope but later on is enriched in the cortical cytoplasm of the developing bulges of root hairs [23]. Very recently, LLS has also been used to document fine actin filaments labeled with (FABD2-GFP) in young Arabidopsis root hairs at the bulge-to-tip growth transition stage [29]. 2.6 Total Internal Reflection Fluorescence Microscopy (TIRFM)

The medium through which light rays travel affects the speed of those rays/photons. The measure of these properties is the refractive index of the medium. When rays of light travel from one medium to another of a different refractive index, those rays are bent or refracted (Fig. 5a). Indeed, it is this very fundamental property that enables lenses to provide magnification. TIRF microscopy exploits the phenomena of total internal reflection. The angle at which light approaches that interface will affect the angle of refraction, and if this angle exceeds the critical angle, the light rays will be totally internally reflected back into the current medium without entering the second (Fig. 5b, c). In practice, for

348

Timothy J. Hawkins

Fig. 5 Total internal reflection. (a) Incident ray approaches the glass water interface at an angle below the critical angle, and the ray is refracted. (b) When the incident ray approaches the interface at the critical angle, the refracted ray is parallel with the interface. (c) When the incident ray approaches the interface at angles greater than the critical angle, the ray is totally internally reflected and returns into the initial medium. This generates the thin evanescent wave at the interface which selectively excites fluorescent molecules in the cell within 200 nm of the glass providing rapid imaging of the plasma membrane and cellular components in close proximity with improved clarity and SNR. (d) TIRF microscopy of growing Alexa488 conjugated actin filaments on a coverglass

TIRF, this interface is the interface between the glass coverslip and the aqueous medium surrounding the cell or tissue. Importantly TIRF microscopy is made possible by the generation of an electromagnetic wavefront at the interface upon reflection called the evanescent wave (Fig. 5c). The evanescent wave front has limited propagation away from the glass and into the sample generating a highly restricted excitation field adjacent to the interface, in the lower-index medium (Fig. 5c). The field is identical in frequency to the incident light and decays exponentially in intensity with distance from the interface, meaning the field extends maximally 200 nanometers into the specimen. As such TIRF is ideally suited to imaging cellular dynamics at the cell membrane or just below and has been exploited to study focal adhesions, endocytosis, and exocytosis. It is also an ideal technology for imaging cell-free systems. For plant microscopy and the cytoskeleton in particular, the proximity of the interphase cortical array to the membrane and the thin cytosolic compartment under the membrane in high vacuolated cells has provided an excellent subject for the deployment of this technology. Early deployment of TIRF for plant cell biology was used in the visualization of in vivo dynamics of secretory vesicles in pollen tubes [37], the study of the dynamics of DYNAMIN RELATED PROTEINS (DRPs) required for cytokinesis and cell expansion, and plasma membrane markers for the study of plant endocytosis [38, 39]. TIRF microscopy has proven to be a useful tool in the investigation of the plant cytoskeleton both in vitro and in vivo, in particular actin dynamics, bundling and severing, its membrane association and response to pathogens where the TIRF excitation field has provided improved resolution of filaments through greater

Light Microscopy and the Cytoskeleton

349

signal to noise images, and the ability to dissect the network proximal to the membrane. Specific studies have used time-lapse TIRF microscopy to observe real-time “catch and zipper” actin bundle formation by VLN1 and 3 in vitro and quantify the severing of individual actin filaments by VLN3 and its dependence on calcium concentration [40]. Moreover, TIRF microscopy allowed the detailed strategies of actin reorganization in plants to be visualized revealing that F-actin is more dynamic in isotropically expanding cells, regularly changing its density with new filaments able to be assembled from shorter free cytoplasmic fragments, the polymerization from ends of extant filaments, and the severing and end joining of existing polymers [41]. Initially, there was a perception that TIRF was not feasible in plant cells because the cell wall would restrict the penetration of the evanescent field and lead to scattering of illumination, and as such several of these early studies considered the phenomena as variable-angle epifluorescence microscopy (VAEM). Here, instead of an evanescent field, this uses a narrow band of illumination that passes through the sample almost parallel to the coverslip, still yielding a high SNR image of events at or near the plasma membrane of intact cells. Although some studies continue to choose to use VAEM in intact tissues to access slightly deeper events, Vizcay-Barrena et al. were able to demonstrate true TIRF can occur in plant cells and distinguished this from VAEM by observing a second reflected light path in the reverse of the lens [42]. More recently TIRF microscopy has successfully shown that the actin-dependent trafficking of host proteins to discrete sites of fungal pathogen contact includes a specific membrane-integrated formin (FORMIN4) that reinforces local cytoskeletal dynamics during an immune response. TIRF also observed a 2D pattern of formin at the plasma membrane with distinct meso-domains of the key defense protein PEN3, revealing a fine spatial segregation of destinations for actin-dependent immunity cargo at cell wall appositions [43].The improved SNR and thin axial imaging window provided by TIRF microscopy can be combined with structured illumination (TIRF-SIM) to achieve fast super-resolution acquisition rates of microtubule dynamics at the cell cortex in intact plants [13]. Light microscopy remains the indispensable research tool for plant cytoskeleton research. The plant community has utilized the full breadth of advanced light imaging modalities available and continues to embrace the newest optical developments as well and developing new modifications and methodologies to address the unique challenges of plant samples.

350

Timothy J. Hawkins

References 1. Chan J, Calder G, Fox S, Lloyd C (2007) Cortical microtubule arrays undergo rotary movements in Arabidopsis hypocotyl epidermal cells. Nat Cell Biol 9(2):171–175. https://doi.org/10.1038/ncb1533 2. Schermelleh L, Carlton PM, Haase S, Shao L, Winoto L, Kner P, Burke B, Cardoso MC, Agard DA, Gustafsson MG, Leonhardt H, Sedat JW (2008) Subdiffraction multicolor imaging of the nuclear periphery with 3D structured illumination microscopy. Science 320(5881):1332–1336. https://doi.org/10. 1126/science.1156947 3. Hein B, Willig KI, Hell SW (2008) Stimulated emission depletion (STED) nanoscopy of a fluorescent protein-labeled organelle inside a living cell. Proc Natl Acad Sci U S A 105(38): 14271–14276. https://doi.org/10.1073/ pnas.0807705105 4. Rust MJ, Bates M, Zhuang X (2006) Subdiffraction-limit imaging by stochastic optical reconstruction microscopy (STORM). Nat Methods 3(10):793–795. https://doi.org/ 10.1038/nmeth929 5. Azuma T, Kei T (2015) Super-resolution spinning-disk confocal microscopy using optical photon reassignment. Opt Express 23(11): 15003–15011. https://doi.org/10.1364/ OE.23.015003 6. Gustafsson N, Culley S, Ashdown G, Owen DM, Pereira PM, Henriques R (2016) Fast live-cell conventional fluorophore nanoscopy with ImageJ through super-resolution radial fluctuations. Nat Commun 7:12471. https:// doi.org/10.1038/ncomms12471 7. Fitzgibbon J, Bell K, King E, Oparka K (2010) Super-resolution imaging of plasmodesmata using three-dimensional structured illumination microscopy. Plant Physiol 153(4): 1453–1463. https://doi.org/10.1104/pp. 110.157941 8. Knox K, Wang P, Kriechbaumer V, Tilsner J, Frigerio L, Sparkes I, Hawes C, Oparka K (2015) Putting the squeeze on plasmodesmata: a role for reticulons in primary plasmodesmata formation. Plant Physiol 168(4):1563–1572. https://doi.org/10.1104/pp.15.00668 9. Tilsner J, Linnik O, Louveaux M, Roberts IM, Chapman SN, Oparka KJ (2013) Replication and trafficking of a plant virus are coupled at the entrances of plasmodesmata. J Cell Biol 201(7):981–995. https://doi.org/10.1083/ jcb.201304003 10. Linnik O, Liesche J, Tilsner J, Oparka KJ (2013) Unraveling the structure of viral

replication complexes at super-resolution. Front Plant Sci 4:6. https://doi.org/10. 3389/fpls.2013.00006 11. Komis G, Samajova O, Ovecka M, Samaj J (2015) Super-resolution microscopy in plant cell imaging. Trends Plant Sci 20(12): 834–843. https://doi.org/10.1016/j.tplants. 2015.08.013 12. Komis G, Novak D, Ovecka M, Samajova O, Samaj J (2018) Advances in imaging plant cell dynamics. Plant Physiol 176(1):80–93. https://doi.org/10.1104/pp.17.00962 13. Vavrdova T, Krenek P, Ovecka M, Samajova O, Flokova P, Illesova P, Snaurova R, Samaj J, Komis G (2020) Complementary superresolution visualization of composite plant microtubule organization and dynamics. Front Plant Sci 11:693. https://doi.org/10.3389/fpls. 2020.00693 14. Schubert V, Neumann P, Marques A, Heckmann S, Macas J, Pedrosa-Harand A, Schubert I, Jang TS, Houben A (2020) Super-resolution microscopy reveals diversity of plant centromere architecture. Int J Mol Sci 21(10). https://doi.org/10.3390/ ijms21103488 15. Duncombe SG, Chethan SG, Anderson CT (2022) Super-resolution imaging illuminates new dynamic behaviors of cellulose synthase. Plant Cell 34(1):273–286. https://doi.org/ 10.1093/plcell/koab227 16. Durst S, Hedde PN, Brochhausen L, Nick P, Nienhaus GU, Maisch J (2014) Organization of perinuclear actin in live tobacco cells observed by PALM with optical sectioning. J Plant Physiol 171(2):97–108. https://doi. org/10.1016/j.jplph.2013.10.007 17. Liesche JZ, Schulz, A. Super-resolution imaging with Pontamine Fast Scarlet 4BS enables direct visualization of cellulose orientation and cell connection architecture in onion epidermis cells. BMC Plant Biol 13:226 18. Dong B, Yang X, Zhu S, Bassham DC, Fang N (2015) Stochastic optical reconstruction microscopy imaging of microtubule arrays in intact arabidopsis thaliana seedling roots. Sci Rep 5:15694. https://doi.org/10.1038/ srep15694 19. Schubert V, Weisshart K (2015) Abundance and distribution of RNA polymerase II in Arabidopsis interphase nuclei. J Exp Bot 66(6): 1687–1698. https://doi.org/10.1093/jxb/ erv091 20. Kleine-Vehn J, Wabnik K, Martiniere A, Langowski L, Willig K, Naramoto S,

Light Microscopy and the Cytoskeleton Leitner J, Tanaka H, Jakobs S, Robert S, Luschnig C, Govaerts W, Hell SW, Runions J, Friml J (2011) Recycling, clustering, and endocytosis jointly maintain PIN auxin carrier polarity at the plasma membrane. Mol Syst Biol 7: 540. https://doi.org/10.1038/msb.2011.72 21. Kurzbauer MT, Janisiw MP, Paulin LF, Prusen Mota I, Tomanov K, Krsicka O, Haeseler AV, Schubert V, Schlogelhofer P (2021) ATM controls meiotic DNA double-strand break formaand affects tion and recombination synaptonemal complex organization in plants. Plant Cell 33(5):1633–1656. https://doi.org/ 10.1093/plcell/koab045 22. Rego EH, Shao L (2015) Practical structured illumination microscopy. Methods Mol Biol 1251:175–192. https://doi.org/10.1007/ 978-1-4939-2080-8_10 23. Ticha M, Richter H, Ovecka M, Maghelli N, Hrbackova M, Dvorak P, Samaj J, Samajova O (2020) Advanced microscopy reveals complex developmental and subcellular localization patterns of ANNEXIN 1 in arabidopsis. Front Plant Sci 11:1153. https://doi.org/10.3389/ fpls.2020.01153 24. Vyplelova P, Ovecka M, Komis G, Samaj J (2018) Advanced microscopy methods for bioimaging of mitotic microtubules in plants. Methods Cell Biol 145:129–158. https://doi. org/10.1016/bs.mcb.2018.03.019 25. McKenna JF, Rolfe DJ, Webb SED, Tolmie AF, Botchway SW, Martin-Fernandez ML, Hawes C, Runions J (2019) The cell wall regulates dynamics and size of plasma-membrane nanodomains in Arabidopsis. Proc Natl Acad Sci U S A 116(26):12857–12862. https://doi. org/10.1073/pnas.1819077116 26. Liang P, Stratil TF, Popp C, Marin M, Folgmann J, Mysore KS, Wen J, Ott T (2018) Symbiotic root infections in Medicago truncatula require remorin-mediated receptor stabilization in membrane nanodomains. Proc Natl Acad Sci U S A 115(20):5289–5294. https:// doi.org/10.1073/pnas.1721868115 27. Romero IC, Kong S, Fowlkes CC, Jaramillo C, Urban MA, Oboh-Ikuenobe F, D’Apolito C, Punyasena SW (2020) Improving the taxonomy of fossil pollen using convolutional neural networks and superresolution microscopy. Proc Natl Acad Sci U S A 117(45): 28496–28505. https://doi.org/10.1073/ pnas.2007324117 28. Chen FT, Tillberg PW, Boyden ES (2015) Expansion microscopy. Science 347(6221): 543–548 29. Ovecka M, Sojka J, Ticha M, Komis G, Basheer J, Marchetti C, Samajova O, Kubenova L, Samaj J (2022) Imaging plant

351

cells and organs with light-sheet and superresolution microscopy. Plant Physiol 188(2): 683–702. https://doi.org/10.1093/plphys/ kiab349 30. Ovecka M, von Wangenheim D, Tomancak P, Samajova O, Komis G, Samaj J (2018) Multiscale imaging of plant development by lightsheet fluorescence microscopy. Nat Plants 4(9): 639–650. https://doi.org/10.1038/s41477018-0238-2 31. O’Callaghan FE, Braga RA, Neilson R, MacFarlane SA, Dupuy LX (2018) New live screening of plant-nematode interactions in the rhizosphere. Sci Rep 8(1):1440. https://doi. org/10.1038/s41598-017-18797-7 32. Zhang T, Cieslak M, Owens A, Wang F, Broholm SK, Teeri TH, Elomaa P, Prusinkiewicz P (2021) Phyllotactic patterning of gerbera flower heads. Proc Natl Acad Sci U S A:118 (13). h ttps://doi.o rg/10 .1 073/pnas. 2016304118 33. Capua Y, Eshed Y (2017) Coordination of auxin-triggered leaf initiation by tomato LEAFLESS. Proc Natl Acad Sci U S A 114(12):3246–3251. https://doi.org/10. 1073/pnas.1617146114 34. Ovecka M, Luptovciak I, Komis G, Samajova O, Samakovli D, Samaj J (2020) Spatiotemporal pattern of ectopic cell divisions contribute to mis-shaped phenotype of primary and lateral roots of katanin1 mutant. Front Plant Sci 11:734. https://doi.org/10.3389/ fpls.2020.00734 35. Vaskebova L, Samaj J, Ovecka M (2018) Single-point ACT2 gene mutation in the Arabidopsis root hair mutant der1-3 affects overall actin organization, root growth and plant development. Ann Bot 122(5):889–901. https://doi.org/10.1093/aob/mcx180 36. Vyplelova P, Ovecka M, Samaj J (2017) Alfalfa root growth rate correlates with progression of microtubules during mitosis and cytokinesis as revealed by environmental light-sheet microscopy. Front Plant Sci 8:1870. https://doi.org/ 10.3389/fpls.2017.01870 37. Wang X., et al (2006) Imaging of dynamic secretory vesicles in living pollen tubes of Picea meyeri using evanescent wave microscopy. Plant Physiol 141(4): 1591–603 38. Konopka CA, Bednarek SY (2008) Variableangle epifluorescence microscopy: a new way to look at protein dynamics in the plant cell cortex. Plant J 53(1):186–196. https://doi. org/10.1111/j.1365-313X.2007.03306.x 39. Konopka CA, Backues SK, Bednarek SY (2008) Dynamics of Arabidopsis dynamin-related protein 1C and a clathrin light chain at the plasma

352

Timothy J. Hawkins

membrane. Plant Cell 20(5):1363–1380. https://doi.org/10.1105/tpc.108.059428 40. Khurana P, Henty JL, Huang S, Staiger AM, Blanchoin L, Staiger CJ (2010) Arabidopsis VILLIN1 and VILLIN3 have overlapping and distinct activities in actin bundle formation and turnover. Plant Cell 22(8):2727–2748. https://doi.org/10.1105/tpc.110.076240 41. Smertenko AP, Deeks MJ, Hussey PJ (2010) Strategies of actin reorganisation in plant cells. J Cell Sci 123(Pt 17):3019–3028. https://doi. org/10.1242/jcs.071126 42. Vizcay-Barrena G, Webb SE, MartinFernandez ML, Wilson ZA (2011) Subcellular

and single-molecule imaging of plant fluorescent proteins using total internal reflection fluorescence microscopy (TIRFM). J Exp Bot 62(15):5419–5428. https://doi.org/10. 1093/jxb/err212 43. Sassmann S, Rodrigues C, Milne SW, Nenninger A, Allwood E, Littlejohn GR, Talbot NJ, Soeller C, Davies B, Hussey PJ, Deeks MJ (2018) An immune-responsive cytoskeletal-plasma membrane feedback loop in plants. Curr Biol 28(13):2136–2144 e2137. https:// doi.org/10.1016/j.cub.2018.05.014

Chapter 29 Investigating Plant Protein–Protein Interactions Using FRET-FLIM with a Focus on the Actin Cytoskeleton Patrick Duckney and Patrick J. Hussey Abstract The study of protein–protein interactions is fundamental to understanding how actin-dependent processes are controlled through the regulation of actin-binding proteins by their interactors. FRET-FLIM (Fo¨rster resonance energy transfer-fluorescence lifetime imaging microscopy) is a sensitive bioimaging method to detect protein–protein interactions in living cells through measurement of FRET, facilitated by the interactions of fluorophore-tagged fusion protein. As a sensitive and noninvasive method for the spatiotemporal visualization of dynamic protein–protein interactions, FRET-FLIM holds several advantages over other methods of protein interaction assays. FRET-FLIM has been widely employed to characterize many plant protein interactions, including interactions between actin-regulatory proteins and their binding partners. As we increasingly understand the plant actin cytoskeleton to coordinate a diverse number of complex functions, the study of actin-regulatory proteins and their interactors becomes increasingly technically challenging. Sophisticated and sensitive in vivo methods such as FRET-FLIM are likely to be crucial to the study of protein–protein interactions as more complex and challenging hypotheses are addressed. Key words FRET-FLIM, Protein–protein interaction, Confocal microscopy, Imaging, In vivo, Protein dynamics, Noninvasive

1

Introduction The actin cytoskeleton is highly dynamic and undergoes continuous structural reorganization to drive a wide variety of actindependent processes. The dynamic reorganization of F-actin is regulated by a large number of actin-binding proteins (ABPs) which regulate F-actin polymerization, depolymerization, stabilization, and cross-linking [1, 2], and spatiotemporal regulation of ABP activity by their upstream regulators is essential for the coordination of actin dynamics. Therefore, characterization of interactions between ABPs and their interacting proteins is essential to understand how the actin cytoskeleton is regulated to perform specific functions, and investigation of protein–protein interactions

Patrick J. Hussey and Pengwei Wang (eds.), The Plant Cytoskeleton: Methods and Protocols, Methods in Molecular Biology, vol. 2604, https://doi.org/10.1007/978-1-0716-2867-6_29, © The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2023

353

354

Patrick Duckney and Patrick J. Hussey

between ABPs and their regulators is fundamental to plant cytoskeletal research. There exist numerous techniques to assay plant protein–protein interactions. For example, protein affinity purification techniques, such as coimmunoprecipitations and pulldowns, have for a long time been used effectively to isolate and identify proteins in complex with a specific bait protein from tissue lysate. Given the harsh experimental conditions required to purify protein complexes from a lysate, it may not be possible to isolate unstable, promiscuous, or weak interactors, and false-positive results from artificial association of noninteracting proteins may also occur [3]. Difficult-to-purify membrane proteins may also present a challenge in investigating protein interactions [4]. For identification of direct interactions, the yeast-2-hybrid (Y2H) system has also long served as a staple experimental system, relying on the direct interaction between bait and prey constructs. While the traditional Y2H system is limited to investigating cytosolic interactors, the split-ubiquitin Y2H system is able to identify interactors of membrane proteins [5]. However, the Y2H system may not be suitable to study the interactions of certain plant proteins, which may require specific tertiary folding or posttranslational modification conditions which are absent in yeast, or the presence of certain plant tertiary intermediate proteins for interactions to occur [3]. The investigation of protein–protein interactions with live-cell imaging techniques is becoming increasingly popular and holds several advantages over traditional Y2H and affinity purification methods. Imaging-based interaction studies in planta do not require the harsh treatments necessary for affinity purification and study the interactors in their native environments in which they are functional. For example, bimolecular fluorescence complementation (BiFC) is a frequently used live-cell imaging system to detect protein–protein interaction, utilizing the interaction of proteins fused to fluorophore fragments (usually split YFP or GFP) to reconstitute a fully functional fluorophore [6, 7]. A particular benefit of BiFC is to indicate the exact spatial localization of the interaction between proteins, as BiFC fluorescence is only generated at the site of protein interaction [7]. However, BiFC is prone to generating false-positive results due to the inherent self-affinity of split fluorophore fragments and their spontaneous reassembly independent of fusion protein interaction [8]. Split luciferase assays are another live-cell imaging technique for studying protein interaction, based on the reconstitution of functional luciferase (LUC) from split LUC fragments by interaction between split-LUC-fusion protein constructs, to generate bioluminescent light emission. Unlike BiFC, split LUC is not prone to false positives as LUC fragments do not undergo spontaneous reassembly; however it cannot be used to identify the subcellular localization of protein– protein interactions [9, 10]. Recently, rapamycin-dependent

FRET-LIM in Plants

355

delocalization assays for detection of protein–protein interactions have been adapted for use in plants (named “knocksideways in plants”; KSP). KSP utilizes the rapid rapamycin-induced heteromerization of FRB and FKBP tags, so that following rapamycin treatment, FKBP-tagged bait fusion proteins are delocalized to an FRB-tagged anchor along with prey protein constructs, dependent on bait–prey interaction [3]. KSP is a simple and effective experimental system that allows quantification of interaction strength and is not prone to false positives. However, KSP may not be able to detect weak or transient interactions [3] and is dependent on prey protein delocalization, making it potentially unsuitable for membrane-anchored proteins. Here we describe FRET-FLIM (Fo¨rster resonance energy transfer-fluorescence lifetime imaging microscopy), a powerful tool for studying protein–protein interactions in vivo through quantitative measurement of FRET between fluorophore-fusion proteins, dependent on their direct interaction. FRET-FLIM investigation of protein–protein interactions in vivo can be a convenient, noninvasive, and sensitive imaging method to generate quantitative, spatial, and dynamic information about protein–protein interactions. We will discuss the principles of FRET-FLIM, its application for studying protein–protein interactions in plants, and promising technical developments that will advance the investigation of protein interactions in vivo.

2

The Principles of FRET-FLIM FRET-FLIM is the quantitative measurement of the fluorescence lifetime of a fluorescent molecule to detect FRET between two fluorescent molecules in direct proximity, which in the context of cell biology can be indicative of direct physical interaction between fluorescent protein-tagged fusion constructs [11]. Fluorescent proteins have long been used to study protein subcellular localization and dynamics, and colocalization between two fusion proteins has been an instrumental part in demonstrating protein–protein interaction in vivo. One inherent property of fluorescent proteins, as for all fluorescent molecules, is a specific fluorescence lifetime (defined as the time spent in an excited state before emitting a photon and returning to ground state following excitation) [12, 13]. This property, which in the case of fluorescent proteins lies within a range of nanoseconds (