Intracellular Lipid Transport: Methods and Protocols [1st ed.] 978-1-4939-9135-8, 978-1-4939-9136-5

Table of contents :
Front Matter ....Pages i-xi
Analysis of Phosphatidylinositol Transfer at ER-PM Junctions in Receptor-Stimulated Live Cells (Chi-Lun Chang, Jen Liou)....Pages 1-11
Monitoring Non-vesicular Transport of Phosphatidylserine and Phosphatidylinositol 4-Phosphate in Intact Cells by BRET Analysis (Mira Sohn, Daniel J. Toth, Tamas Balla)....Pages 13-22
Development of Nonspecific BRET-Based Biosensors to Monitor Plasma Membrane Inositol Lipids in Living Cells (József T. Tóth, Gergő Gulyás, László Hunyady, Péter Várnai)....Pages 23-34
Following Anterograde Transport of Phosphatidylserine in Yeast in Real Time (Juan Martín D’Ambrosio, Véronique Albanèse, Alenka Čopič)....Pages 35-46
Imaging Lipid Metabolism at the Golgi Complex (Serena Capasso, Giovanni D’Angelo)....Pages 47-56
Advanced In Vitro Assay System to Measure Phosphatidylserine and Phosphatidylethanolamine Transport at ER/Mitochondria Interface (Yasushi Tamura, Rieko Kojima, Toshiya Endo)....Pages 57-67
Measurement of Lipid Transport in Mitochondria by the MTL Complex (Juliette Jouhet, Valérie Gros, Morgane Michaud)....Pages 69-93
Bi- and Trifunctional Lipids for Visualization of Sphingolipid Dynamics within the Cell (Doris Höglinger)....Pages 95-103
Indirect Lipid Transfer Protein Activity Measurements Using Quantification of Glycosphingolipid Production (Anders P. E. Backman, Josefin Halin, Matti A. Kjellberg, Peter Mattjus)....Pages 105-114
Measurement of Intracellular Sterol Transport in Yeast (Neha Chauhan, Julian A. Jentsch, Anant K. Menon)....Pages 115-136
Intracellular and Plasma Membrane Cholesterol Labeling and Quantification Using Filipin and GFP-D4 (Léa P. Wilhelm, Laetitia Voilquin, Toshihide Kobayashi, Catherine Tomasetto, Fabien Alpy)....Pages 137-152
Monitoring and Modulating Intracellular Cholesterol Trafficking Using ALOD4, a Cholesterol-Binding Protein (Shreya Endapally, Rodney E. Infante, Arun Radhakrishnan)....Pages 153-163
Measurement of Lysophospholipid Transport Across the Membrane Using Escherichia coli Spheroplasts (Yibin Lin, Lei Zheng, Mikhail Bogdanov)....Pages 165-180
Preparation of Proteoliposomes with Purified TMEM16 Protein for Accurate Measures of Lipid Scramblase Activity (Janine Denise Brunner, Stephan Schenck)....Pages 181-199
In Vitro Assays to Measure the Membrane Tethering and Lipid Transport Activities of the Extended Synaptotagmins (Xin Bian, Pietro De Camilli)....Pages 201-212
Determining the Lipid-Binding Specificity of SMP Domains: An ERMES Subunit as a Case Study (Andrew P. AhYoung, Pascal F. Egea)....Pages 213-235
In Vitro Measurement of Sphingolipid Intermembrane Transport Illustrated by GLTP Superfamily Members (Roopa Kenoth, Rhoderick E. Brown, Ravi Kanth Kamlekar)....Pages 237-256
Purification and Characterization of Human Niemann–Pick C1 Protein (Xin Gong, Hongwu Qian)....Pages 257-267
In Vitro Strategy to Measure Sterol/Phosphatidylinositol-4-Phosphate Exchange Between Membranes (Nicolas-Frédéric Lipp, Guillaume Drin)....Pages 269-292
Determination of Ligand Binding Affinity and Specificity of Purified START Domains by Thermal Shift Assays Using Circular Dichroism (Danny Létourneau, Jean-Guy LeHoux, Pierre Lavigne)....Pages 293-306
Synthesis of Fluorescent Membrane-Spanning Lipids for Studies of Lipid Transfer and Membrane Fusion (Günter Schwarzmann)....Pages 307-324
Setting Up All-Atom Molecular Dynamics Simulations to Study the Interactions of Peripheral Membrane Proteins with Model Lipid Bilayers (Viviana Monje-Galvan, Linnea Warburton, Jeffery B. Klauda)....Pages 325-339
Back Matter ....Pages 341-345

Citation preview

Methods in Molecular Biology 1949

Guillaume Drin Editor

Intracellular Lipid Transport Methods and Protocols

METHODS

IN

MOLECULAR BIOLOGY

Series Editor John M. Walker School of Life and Medical Sciences, University of Hertfordshire, Hatfield, Hertfordshire AL10 9AB, UK

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

Intracellular Lipid Transport Methods and Protocols

Edited by

Guillaume Drin Institut de Pharmacologie Moléculaire et Cellulaire, Université Côte d’Azur and CNRS, Valbonne, France

Editor Guillaume Drin Institut de Pharmacologie Mole´culaire et Cellulaire Universite´ Coˆte d’Azur and CNRS Valbonne, France

ISSN 1064-3745 ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-4939-9135-8 ISBN 978-1-4939-9136-5 (eBook) https://doi.org/10.1007/978-1-4939-9136-5 Library of Congress Control Number: 2019931902 © Springer Science+Business Media, LLC, part of Springer Nature 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express 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 Press imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer Nature. The registered company address is: 233 Spring Street, New York, NY 10013, U.S.A.

Preface The boundaries and complex architecture of cells rely on membranes that result from the assembly of many lipids, along with proteins. Lipid chemistry is highly diverse with thousands of lipid species identified to date. Inside prokaryotic and eukaryotic cells, lipid species are not randomly distributed between membranes, quite the contrary. In actuality, membranes are endowed with precise physicochemical features and selectively host particular lipids with signaling roles. For instance, it is well described that, in eukaryotic cells, each organelle membrane has its own molecular identity, which is essential for many proteins to function properly and support key functions. At any time, underlying mechanisms adjust and maintain the lipid composition of membranes in the face of permanent processes that locally modify the lipid composition, such as the use of certain lipids by signaling pathways, or the mixing of lipids, as in eukaryotic cells, by vesicular trafficking that causes massive membrane exchange. Those mechanisms are the metabolic pathways that ensure lipid synthesis, interconversion, and degradation as well as processes that ensure lipid transport within and between membranes. Yet, while metabolic processes are well defined, relatively little is known about intracellular lipid transport. A likely reason is that intracellular lipid transport routes are by nature quite elusive. Besides, the mechanisms whereby lipid transfer proteins (LTPs) and scramblases/flippases move lipids between and in organelle membranes, respectively, remain still mysterious. Better deciphering of these transport processes is crucial to apprehend how a cell ensures some of its key functions. For instance, recent studies have unveiled that the signaling competence of cells relies on fast lipid exchange between the endoplasmic reticulum and the plasma membrane. Another key motivation to explore how lipid transporters work relates to the fact that they can be at the origin or involved in some cellular dysfunctions, resulting in acute syndromes. An archetypical example is the Niemann-Pick type C disease which arises from a defect of a LTP called NPC1, causing an aberrant accumulation of sterol in lysosomes. We also know, for instance, that the overactivity of the sterol-transporter STARD3 is a marker of poor prognosis in breast cancer or that lipid transport routes are hijacked by human viruses to remodel organelles into replication sites. Along with this, proof-of-concept studies showed that LTPs are potential pharmacological targets. In parallel with a growing interest in membrane contact sites, which correspond to singular regions where two organelles are in close apposition, and the discovery of new classes of lipid transporters and major advances in solving their tridimensional structure, a great ingenuity is currently deployed to understand at different scale how lipids are accurately displaced between organelles, across long-distance or in membrane contact sites, or inside cellular membranes. The goal of this book is to provide technical approaches to tackle various questions related to intracellular lipid transport. It encompasses chapters from leading experts in this realm who outline step-by-step protocols they developed during recent years. Several chapters, which constitute the first part of this volume, expose methodologies to measure the movement of varied lipid species (glycerophospholipid, sphingolipids, sterol, etc.) between or in organelle membranes, inside eukaryotic cells, including plant cells, or in bacteria. Different approaches are described, among which are cell imaging and bioluminescence resonance energy transfer (BRET), relying on the use of fluorescent lipids or

v

vi

Preface

genetically encoded lipid probes, as well as organelle isolation strategies coupled to lipid analysis. Of a great diversity, these methodologies all offer researchers the ability to follow particular lipid type(s) out of many others in the complex environment of a cell, notably in membrane contact sites, with the highest accuracy and the best possible temporal resolution. The second part of the book gathers in vitro or in silico approaches aiming to define, more from biochemical and structural standpoints, how LTPs or flippases/scramblases precisely function. Protocols that are presented allow for answering questions like, “What sort of lipids are recognized by my favorite LTP? At which speed is my protein capable of moving lipids between two distinct organelles or, in the case of a flippase/scramblase, to displace lipids transversally between the two sides of a membrane?” These approaches are useful to define what the affinity and selectivity of a transporter for lipid ligand(s) are. Moreover, based on the use of artificial membranes called liposomes, these protocols offer a unique way to precisely establish the velocity of transporters while indicating how some membrane parameters (surface charge, fluidity, etc.) or cofactors (ions) influence their activity. One chapter describes protocols to measure how a LTP, working at contact sites, tethers two membranes and transfers lipids in between. One chapter describes the making of new lipids that can be useful to improve in vitro transfer assays. Lastly, a chapter presents how to perform reliable molecular dynamic simulations to describe what likely happens at the atomic level when a LTP docks onto a membrane surface, a step that still remains quite enigmatic. Valbonne, France

Guillaume Drin

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

1 Analysis of Phosphatidylinositol Transfer at ER-PM Junctions in Receptor-Stimulated Live Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chi-Lun Chang and Jen Liou 2 Monitoring Non-vesicular Transport of Phosphatidylserine and Phosphatidylinositol 4-Phosphate in Intact Cells by BRET Analysis . . . . . . . Mira Sohn, Daniel J. Toth, and Tamas Balla 3 Development of Nonspecific BRET-Based Biosensors to Monitor Plasma Membrane Inositol Lipids in Living Cells. . . . . . . . . . . . . . . . . J
ozsef T. T
oth, Gergo˝ Gulya´s, La´szl
o Hunyady, and Pe´ter Va´rnai 4 Following Anterograde Transport of Phosphatidylserine in Yeast in Real Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ˇ opicˇ Juan Martı´n D’Ambrosio, Ve´ronique Albane`se, and Alenka C 5 Imaging Lipid Metabolism at the Golgi Complex . . . . . . . . . . . . . . . . . . . . . . . . . . . Serena Capasso and Giovanni D’Angelo 6 Advanced In Vitro Assay System to Measure Phosphatidylserine and Phosphatidylethanolamine Transport at ER/Mitochondria Interface . . . . . . Yasushi Tamura, Rieko Kojima, and Toshiya Endo 7 Measurement of Lipid Transport in Mitochondria by the MTL Complex . . . . . . Juliette Jouhet, Vale´rie Gros, and Morgane Michaud 8 Bi- and Trifunctional Lipids for Visualization of Sphingolipid Dynamics within the Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Doris Ho¨glinger 9 Indirect Lipid Transfer Protein Activity Measurements Using Quantification of Glycosphingolipid Production . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anders P. E. Backman, Josefin Halin, Matti A. Kjellberg, and Peter Mattjus 10 Measurement of Intracellular Sterol Transport in Yeast . . . . . . . . . . . . . . . . . . . . . . Neha Chauhan, Julian A. Jentsch, and Anant K. Menon 11 Intracellular and Plasma Membrane Cholesterol Labeling and Quantification Using Filipin and GFP-D4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Le´a P. Wilhelm, Laetitia Voilquin, Toshihide Kobayashi, Catherine Tomasetto, and Fabien Alpy 12 Monitoring and Modulating Intracellular Cholesterol Trafficking Using ALOD4, a Cholesterol-Binding Protein . . . . . . . . . . . . . . . . . . . Shreya Endapally, Rodney E. Infante, and Arun Radhakrishnan 13 Measurement of Lysophospholipid Transport Across the Membrane Using Escherichia coli Spheroplasts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yibin Lin, Lei Zheng, and Mikhail Bogdanov

vii

v xi

1

13

23

35 47

57 69

95

105

115

137

153

165

viii

14

15

16

17

18 19

20

21

22

Contents

Preparation of Proteoliposomes with Purified TMEM16 Protein for Accurate Measures of Lipid Scramblase Activity . . . . . . . . . . . . . . . . . . Janine Denise Brunner and Stephan Schenck In Vitro Assays to Measure the Membrane Tethering and Lipid Transport Activities of the Extended Synaptotagmins . . . . . . . . . . . . . . . . . . . . . . . Xin Bian and Pietro De Camilli Determining the Lipid-Binding Specificity of SMP Domains: An ERMES Subunit as a Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Andrew P. AhYoung and Pascal F. Egea In Vitro Measurement of Sphingolipid Intermembrane Transport Illustrated by GLTP Superfamily Members. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Roopa Kenoth, Rhoderick E. Brown, and Ravi Kanth Kamlekar Purification and Characterization of Human Niemann–Pick C1 Protein . . . . . . . Xin Gong and Hongwu Qian In Vitro Strategy to Measure Sterol/Phosphatidylinositol-4-Phosphate Exchange Between Membranes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nicolas-Fre´de´ric Lipp and Guillaume Drin Determination of Ligand Binding Affinity and Specificity of Purified START Domains by Thermal Shift Assays Using Circular Dichroism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Danny Le´tourneau, Jean-Guy LeHoux, and Pierre Lavigne Synthesis of Fluorescent Membrane-Spanning Lipids for Studies of Lipid Transfer and Membrane Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ¨ nter Schwarzmann Gu Setting Up All-Atom Molecular Dynamics Simulations to Study the Interactions of Peripheral Membrane Proteins with Model Lipid Bilayers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Viviana Monje-Galvan, Linnea Warburton, and Jeffery B. Klauda

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

181

201

213

237 257

269

293

307

325 341

Contributors ANDREW P. AHYOUNG  Department of Biological Chemistry, David Geffen School of Medicine, University of California, Los Angeles, CA, USA; Department of Early Discovery Biochemistry, Genentech Inc., South San Francisco, CA, USA ´ VERONIQUE ALBANE`SE  Institut Jacques Monod, CNRS, Universite´ Paris Diderot, Sorbonne Paris Cite´, Paris, France FABIEN ALPY  Institut de Ge´ne´tique et de Biologie Mole´culaire et Cellulaire (IGBMC), Institut National de la Sante´ et de la Recherche Me´dicale (INSERM), U1258, Centre National de la Recherche Scientifique (CNRS), UMR7104 and Universite´ de Strasbourg, Illkirch, France ˚ bo Akademi ANDERS P. E. BACKMAN  Faculty of Science and Engineering, Biochemistry, A University, Turku, Finland TAMAS BALLA  Section on Molecular Signal Transduction, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD, USA XIN BIAN  Department of Neuroscience and Cell Biology, Howard Hughes Medical Institute, Program in Cellular Neuroscience, Neurodegeneration, and Repair, and Kavli Institute for Neuroscience, Yale University School of Medicine, New Haven, CT, USA MIKHAIL BOGDANOV  Department of Biochemistry and Molecular Biology, University of Texas Health Science Center at Houston McGovern Medical School, Houston, TX, USA; Department of Biochemistry and Biotechnology, Institute of Fundamental Medicine and Biology, Kazan Federal University, Kazan, Russian Federation RHODERICK E. BROWN  The Hormel Institute, University of Minnesota, Austin, MN, USA JANINE DENISE BRUNNER  Laboratory of Biomolecular Research, Division of Biology and Chemistry, Paul Scherrer Institute, Villigen, Switzerland SERENA CAPASSO  Institute of Protein Biochemistry, National Council Research, Naples, Italy; IRBM, Rome, Italy CHI-LUN CHANG  Janelia Research Campus, Howard Hughes Medical Institute, Ashburn, VA, USA NEHA CHAUHAN  Department of Biochemistry, Weill Cornell Medical College, New York, NY, USA ˇ OPICˇ  Institut Jacques Monod, CNRS, Universite´ Paris Diderot, Sorbonne Paris ALENKA C Cite´, Paris, France JUAN MARTI´N D’AMBROSIO  Institut Jacques Monod, CNRS, Universite´ Paris Diderot, Sorbonne Paris Cite´, Paris, France GIOVANNI D’ANGELO  Institute of Protein Biochemistry, National Council Research, Naples, Italy PIETRO DE CAMILLI  Department of Neuroscience and Cell Biology, Howard Hughes Medical Institute, Program in Cellular Neuroscience, Neurodegeneration, and Repair, and Kavli Institute for Neuroscience, Yale University School of Medicine, New Haven, CT, USA GUILLAUME DRIN  Universite´ Coˆte d’Azur, CNRS, Institut de Pharmacologie Mole´culaire et Cellulaire, Valbonne, France

ix

x

Contributors

PASCAL F. EGEA  Department of Biological Chemistry, David Geffen School of Medicine, University of California , Los Angeles, CA, USA; The Molecular Biology Institute, University of California, Los Angeles, Los Angeles, CA, USA SHREYA ENDAPALLY  Department of Molecular Genetics, University of Texas Southwestern Medical Center, Dallas, TX, USA TOSHIYA ENDO  Faculty of Life Sciences, Kyoto Sangyo University, Kyoto, Japan XIN GONG  Department of Biology, Southern University of Science and Technology, Shenzhen, Guangdong, China; Department of Molecular Biology, Princeton University, Princeton, NJ, USA VALE´RIE GROS  Laboratoire de Physiologie Cellulaire et Ve´ge´tale, UMR 5168 CNRS-CEAINRA-Universite´ Grenoble Alpes, Grenoble, France GERGO˝ GULYA´S  Faculty of Medicine, Department of Physiology, Semmelweis University, Budapest, Hungary ˚ bo Akademi University, JOSEFIN HALIN  Faculty of Science and Engineering, Biochemistry, A Turku, Finland DORIS HO¨GLINGER  Heidelberg University Biochemistry Center, Heidelberg, Germany LA´SZLO´ HUNYADY  Faculty of Medicine, Department of Physiology, Semmelweis University, Budapest, Hungary; MTA-SE Laboratory of Molecular Physiology, Budapest, Hungary RODNEY E. INFANTE  Department of Molecular Genetics and Internal Medicine, and Center for Human Nutrition, University of Texas Southwestern Medical Center, Dallas, TX, USA JULIAN A. JENTSCH  Department of Biochemistry, Weill Cornell Medical College, New York, NY, USA; Institute for Pathophysiology, West German Heart and Vascular Centre, University Hospital Essen, Essen, Germany JULIETTE JOUHET  Laboratoire de Physiologie Cellulaire et Ve´ge´tale, UMR 5168 CNRSCEA-INRA-Universite´ Grenoble Alpes, Grenoble, France RAVI KANTH KAMLEKAR  Department of Chemistry, School of Advanced Sciences, VIT, Vellore, Tamil Nadu, India ROOPA KENOTH  Department of Chemistry, School of Advanced Sciences, VIT, Vellore, Tamil Nadu, India ˚ bo Akademi MATTI A. KJELLBERG  Faculty of Science and Engineering, Biochemistry, A University, Turku, Finland JEFFERY B. KLAUDA  Department of Chemical and Biomolecular Engineering, University of Maryland, College Park, MD, USA; Biophysics Program, University of Maryland, College Park, MD, USA TOSHIHIDE KOBAYASHI  Universite´ de Strasbourg and Laboratory of Bioimaging and Pathologies, Centre National de la Recherche Scientifique (CNRS), UMR7021, Illkirch, France RIEKO KOJIMA  Faculty of Science, Yamagata University, Yamagata, Japan PIERRE LAVIGNE  Faculte´ de Me´decine et des Sciences de la Sante´, De´partement de Biochimie, Universite´ de Sherbrooke, Sherbrooke, QC, Canada JEAN-GUY LEHOUX  Faculte´ de Me´decine et des Sciences de la Sante´, De´partement de Biochimie, Universite´ de Sherbrooke, Sherbrooke, QC, Canada DANNY LE´TOURNEAU  Faculte´ de Me´decine et des Sciences de la Sante´, De´partement de Biochimie, Universite´ de Sherbrooke, Sherbrooke, QC, Canada YIBIN LIN  Division of Infectious Diseases, Department of Pediatrics, Center for Antimicrobial Resistance and Microbial Genomics, University of Texas Health Science Center at Houston McGovern Medical School, Houston, TX, USA JEN LIOU  Department of Physiology, UT Southwestern Medical Center, Dallas, TX, USA

Contributors

xi

NICOLAS-FRE´DE´RIC LIPP  Universite´ Coˆte d’Azur, CNRS, Institut de Pharmacologie Mole´ culaire et Cellulaire, Valbonne, France ˚ bo Akademi PETER MATTJUS  Faculty of Science and Engineering, Biochemistry, A University, Turku, Finland ANANT K. MENON  Department of Biochemistry, Weill Cornell Medical College, New York, NY, USA MORGANE MICHAUD  Laboratoire de Physiologie Cellulaire et Ve´ge´tale, UMR 5168 CNRSCEA-INRA-Universite´ Grenoble Alpes, Grenoble, France VIVIANA MONJE-GALVAN  Department of Chemistry, The University of Chicago, Chicago, IL, USA; Department of Chemical and Biomolecular Engineering, University of Maryland, College Park, MD, USA HONGWU QIAN  Department of Molecular Biology, Princeton University, Princeton, NJ, USA ARUN RADHAKRISHNAN  Department of Molecular Genetics, University of Texas Southwestern Medical Center, Dallas, TX, USA STEPHAN SCHENCK  Laboratory of Biomolecular Research, Division of Biology and Chemistry, Paul Scherrer Institute, Villigen, Switzerland GU¨NTER SCHWARZMANN  LIMES, c/o Kekule´-Institut f. Organische Chemie und Biochemie, Rheinische Friedrich-Wilhelms-Universit€ at Bonn, Bonn, Germany MIRA SOHN  Section on Molecular Signal Transduction, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD, USA YASUSHI TAMURA  Faculty of Science, Yamagata University, Yamagata, Japan CATHERINE TOMASETTO  Institut de Ge´ne´tique et de Biologie Mole´culaire et Cellulaire (IGBMC), Institut National de la Sante´ et de la Recherche Me´dicale (INSERM), U1258, Centre National de la Recherche Scientifique (CNRS), UMR7104 and Universite´ de Strasbourg, Illkirch, France DANIEL J. TOTH  Section on Molecular Signal Transduction, Eunice Kennedy Shriver National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD, USA JO´ZSEF T. TO´TH  Faculty of Medicine, Department of Physiology, Semmelweis University, Budapest, Hungary; Faculty of Medicine, Department of Anesthesiology and Intensive Therapy, Semmelweis University, Budapest, Hungary PE´TER VA´RNAI  Faculty of Medicine, Department of Physiology, Semmelweis University, Budapest, Hungary; MTA-SE Laboratory of Molecular Physiology, Budapest, Hungary LAETITIA VOILQUIN  Institut de Ge´ne´tique et de Biologie Mole´culaire et Cellulaire (IGBMC), Institut National de la Sante´ et de la Recherche Me´dicale (INSERM), U1258, Centre National de la Recherche Scientifique (CNRS), UMR7104 and Universite´ de Strasbourg, Illkirch, France LINNEA WARBURTON  Department of Chemical and Biomolecular Engineering, University of Maryland, College Park, MD, USA; Biophysics Program, University of Maryland, College Park, MD, USA ´ LEA P. WILHELM  Institut de Ge´ne´tique et de Biologie Mole´culaire et Cellulaire (IGBMC), Institut National de la Sante´ et de la Recherche Me´dicale (INSERM), U1258, Centre National de la Recherche Scientifique (CNRS), UMR7104 and Universite´ de Strasbourg, Illkirch, France LEI ZHENG  Department of Biochemistry and Molecular Biology, Center for Membrane Biology, University of Texas Health Science Center at Houston McGovern Medical School, Houston, TX, USA

Chapter 1 Analysis of Phosphatidylinositol Transfer at ER-PM Junctions in Receptor-Stimulated Live Cells Chi-Lun Chang and Jen Liou Abstract Phosphatidylinositol (PI) is an inositol-containing phospholipid synthesized in the endoplasmic reticulum (ER). PI is a precursor lipid for PI 4,5-bisphosphate (PI(4,5)P2) in the plasma membrane (PM) important for Ca2+ signaling in response to extracellular stimuli. Thus, ER-to-PM PI transfer becomes essential for cells to maintain PI(4,5)P2 homeostasis during receptor stimulation. In this chapter, we discuss two live-cell imaging protocols to analyze ER-to-PM PI transfer at ER-PM junctions, where the two membrane compartments make close appositions accommodating PI transfer. First, we describe how to monitor PI (4,5)P2 replenishment following receptor stimulation, as a readout of PI transfer, using a PI(4,5)P2 biosensor and total internal reflection fluorescence microscopy. The second protocol directly visualizes PI transfer proteins that accumulate at ER-PM junctions and mediate PI(4,5)P2 replenishment with PI in the ER in stimulated cells. These methods provide spatial and temporal analysis of ER-to-PM PI transfer during receptor stimulation and can be adapted to other research questions related to this topic. Key words PI, PI(4,5)P2 replenishment, PI(4,5)P2 biosensor, PI transfer protein, ER-PM junctions

1

Introduction Phosphatidylinositol (PI) 4,5-bisphosphate (PI(4,5)P2) is an inositol-containing phospholipid primarily present in the inner leaflet of the plasma membrane (PM) [1, 2]. PI(4,5)P2 regulates a myriad of cellular functions at the PM, including membrane trafficking, ion channel activity, cytoskeleton dynamics, and storeoperated Ca2+ entry [1–3]. In addition, PI(4,5)P2 is essential for mediating Ca2+ signaling in response to extracellular stimuli. Receptor-activated phospholipase C (PLC) hydrolyzes PI(4,5)P2 in the PM to generate inositol 3,4,5-triphosphate (IP3), a cytosolic second messenger leading to Ca2+ release from the endoplasmic reticulum (ER) and activation of Ca2+ signaling events [4]. Animal cells have developed elaborate mechanisms to achieve rapid PI(4,5) P2 replenishment following receptor-induced consumption to maintain PI(4,5)P2-mediated functions and homeostasis. The

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_1, © Springer Science+Business Media, LLC, part of Springer Nature 2019

1

2

Chi-Lun Chang and Jen Liou

ability of cells to replenish PI(4,5)P2 during receptor stimulation is dependent on the level of its precursor lipid, PI in the ER [5–8]. Phosphatic acid in the PM generated following PI(4,5)P2 hydrolysis activates Nir2 [5, 9], a cytosolic PI transfer protein (PITP), to mediate PI(4,5)P2 replenishment dependent on PI in the ER, suggesting ER-to-PM PI transfer [5]. Nir2 and its homolog, Nir3, function at ER-PM junctions, where the ER and PM form close appositions within 20 nm [10]. At ER-PM junctions, Nir2 and Nir3 contact both membrane compartments simultaneously and mediate ER-to-PM PI transfer in exchange of PMto-ER phosphatidic acid transfer [5, 9, 11]. Once delivered to the PM, PI is phosphorylated by PI 4-kinase and PI 4-phosphate (PI4P) 5-kinase to generate PI(4,5)P2 [1]. Disrupting the connection at ER-PM junctions reduces the capability of PI(4,5)P2 replenishment following receptor stimulation [3, 11]. Direct visualization of PI in live cells would be ideal to understand the dynamic regulation of ER-to-PM PI transfer at ER-PM junctions. Nonetheless, this is not feasible because there is no PI biosensor available and PI in the PM is likely to be short-lived before its conversion to PI4P and PI(4,5)P2. Thus, measuring PI(4,5)P2 replenishment using PI (4,5)P2 biosensor and PITP accumulation at ER-PM junctions during receptor stimulation are suitable readouts for ER-to-PM PI transfer. Live-cell fluorescence microscopy is a powerful tool in understanding dynamic information of cellular functions. This chapter first describes a protocol to measure consumption and replenishment of PI(4,5)P2 during receptor stimulation by a green fluorescence protein-tagged PLCδ-pleckstrin homology domain (GFP-PLCδ-PH) as a PI(4,5)P2 biosensor in live cells [12, 13]. In combination with total internal reflection fluorescence (TIRF) microscopy, which selectively illuminates fluorescence within 100 nm of the PM [14], our approach provides a straightforward pipeline of measurement and analysis of PI(4,5)P2 levels in the PM, indicative of ER-to-PM PI transfer, during receptor stimulation in live cells. PI(4,5)P2 replenishment occurs within seconds following receptor-induced hydrolysis. At endogenous receptor level, addition of high concentration (100 μM) of histamine to HeLa cells leads to a shallow consumption phase showing only ~10% reduction of PI(4,5)P2 in the PM before a rapid replenishment phase that restores PI(4,5)P2 back to basal level within 30 s (Fig. 1a). It is difficult to visualize and analyze such a small change and a short-lived process in live cells. To overcome this difficulty, we apply histamine H1 receptor (H1R) overexpression to augment PI (4,5)P2 consumption. HeLa cells with H1R overexpression exhibit a steep consumption phase with up to ~70% reduction of PI(4,5)P2 in the PM immediately after stimulation (Fig. 1b). This enhanced consumption phase is followed by a prolonged replenishment phase manifesting an initial partial recovery and a subsequent steady state.

Analyzing PI Transfer at ER-PM Junctions in Live Cells

3

Fig. 1 Monitoring PI(4,5)P2 levels in the PM during receptor stimulation. Dynamic changes of PI(4,5)P2 levels induced by 100 μM histamine monitored by TIRF microscopy in HeLa cells transfected with (a) GFP-PLCδ-PH or (b) H1R and GFP-PLCδ-PH. Consumption and replenishment phases are labeled with red and green, respectively. Mean  SEM (Standard Error of the Mean) is shown. Figure in (b) was originally published in Cell Reports, 5 (2013) 813–825. © Chang and Cell Reports, 2013

At the steady state, only ~60%, instead of 100% of PI(4,5)P2 in the PM was restored, reflecting the equilibrium between the enhanced consumption and the endogenous replenishment mechanism. Thus, receptor overexpression enables a bigger dynamic range and temporal window for monitoring PI(4,5)P2 levels in the PM in live cells during receptor stimulation. Next, we describe an alternative approach for analyzing ER-toPM PI transfer via visualizing Nir2 translocation to ER-PM junctions during receptor stimulation, a necessary step for Nir2 to mediate PI(4,5)P2 replenishment in stimulated cells [5, 8]. Nir2 tagged with a red fluorescence protein (Nir2-mCherry) localizes in the cytosol in resting cells and displays punctate distribution following receptor stimulation when monitored by confocal microscopy (Fig. 2a). This process, namely Nir2 translocation to ER-PM junctions, is better visualized with TIRF microscopy as Nir2mCherry shows little signal at the TIRF focal plane before stimulation (Fig. 2b). Following histamine treatment, Nir2 translocates to ER-PM junctions as labeled by MAPPER, a synthetic marker for visualization of ER-PM junctions in live cells [11]. This approach provides spatial and temporal analysis of Nir2 activation and function during receptor stimulation. In this chapter, we describe live-cell imaging protocols to visualize and to quantitatively analyze PI(4,5)P2 level in the PM and Nir2 translocation to ER-PM junctions during receptor stimulation, as readouts for ER-to-PM PI transfer. We provide the basic principles and discuss several technical notes as we want to make these protocols straightforward and reproducible. Importantly, these protocols require few fluorescence channels and minimal cell perturbations; therefore, they can be adapted to study ER-toPM PI transfer in a variety of cellular processes.

4

Chi-Lun Chang and Jen Liou

Fig. 2 Nir2 Translocation to ER-PM Junctions. (a) Nir2-mCherry translocation induced by 100 μM histamine monitored by confocal microscopy in HeLa cells transfected with H1R. Scale bar, 10 μm. (b) Nir2-mCherry translocation to ER-PM junctions induced by 100 μM histamine monitored by TIRF microscopy in MAPPERexpressing HeLa cells transfected with H1R. Scale bar: 2 μm. This figure was originally published in Cell Reports, 5 (2013) 813–825. © Chang and Cell Reports, 2013

2

Materials Prepare all solutions with molecular biology and cell biology grade reagents. All materials in contact with cells should be sterile. In this protocol, we use HeLa cells with H1R overexpression as a model system. Other cell types, including HEK 293 cells and mouse embryonic fibroblasts, can also be used.

2.1 Cell Culture and Transfection

1. HeLa cells (CCL-2 line) from American Type Culture Collection. 2. Culturing medium: Eagle’s minimum essential medium (EMEM) supplemented with 10% fetal bovine serum and 1 penicillin and streptomycin solution. 3. Trypsin-EDTA solution (0.25%). 4. Ca2+/Mg2+-free PBS: 138 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, and 1.5 mM KH2PO4, pH 7.4. 5. Hemocytometer. 6. Lab-Tek chambered cover glasses or MatTek dishes. 7. TransIT-LT1 transfection reagent (Mirus). TransIT-LT1 has good transfection efficiency but low cytotoxicity to HeLa cells. 8. Opti-MEM.

Analyzing PI Transfer at ER-PM Junctions in Live Cells

2.2

Plasmids

5

1. H1R. Other plasmids containing PLC-activating receptors, such as M1 muscarinic acetylcholine receptor, are also suitable (cDNA Resource Center and Addgene). 2. GFP-PLCδ-PH as PI(4,5)P2 biosensor (Addgene). 3. MAPPER for labeling ER-PM junctions (please send request to Dr. Jen Liou). 4. Nir2-mCherry (or Nir3-mCherry, please send request to Dr. Jen Liou). 5. Plasmids with a protein of interest.

2.3

TIRF Microscopy

1. Extracellular buffer (ECB): 125 mM NaCl, 5 mM KCl, 1.5 mM MgCl2, 20 mM HEPES-NaOH, pH 7.4, 10 mM glucose, and 1.5 mM CaCl2. 2. Histamine stock solution (1000): Histamine (Sigma Aldrich) is dissolved at 100 mM in sterile water. Store the stock solution at 4  C. 3. A TIRF microscope with live-cell and time-lapse imaging capabilities.

3

Methods

3.1 Measurement of PI(4,5)P2 Replenishment During Receptor Stimulation 3.1.1 Cell Culture and Transfection

Performed all these steps at 37  C or with pre-warmed reagents. 1. Culture HeLa cells in cell culture dishes or flasks in an incubator supplemented with humidified air and 5% CO2. Change media routinely every 2–3 days. When the cells reach ~90% confluency, subculture the cells. Do not overgrow HeLa cells; this may lead to poor performance in subsequent experiments. 2. Rinse HeLa cells with 1 PBS, add trypsin solution (1 mL per T-25 flask), and incubate for 2–3 min until the cells detach. Add 3 volumes of culture medium to neutralize trypsin solution and pipet the cell suspension a few times. 3. Collect the cell suspension in a conical tube and pellet the cells by centrifugation at 200  g for 2–5 min. Remove supernatant and resuspend cells in 1 mL of culture medium. 4. Count the cells using a hemocytometer and seed 10,000 cells in 300 μL of medium into each well of an 8-well Lab-Tek chambered cover glass. This seeding density will result in ~50% confluency. Other types of imaging cover glasses may also be used. Because HeLa cells attach well to imaging cover glasses, no prior coating is required. 5. Perform transfection 24 h after plating the cells. Mix 0.3 μL of TransIT-LT1 with 50 ng of H1R plasmid and 25 ng of GFP-PLCδ-PH plasmid in 30 μL of Opti-MEM and incubate

6

Chi-Lun Chang and Jen Liou

the mixture at room temperature for 15 min. Remove the medium from the chambered cover glass and replace with 270 μL of fresh medium with 30 μL of transfection mixture in each well. Incubate the transfected cells for 16–20 h before imaging. 3.1.2 Live-Cell Imaging

Experiments are performed at room temperature. 1. Remove medium from the transfected chambered cover glass, rinse once with 300 μL of ECB and add 200 μL of ECB. Incubate the chambered cover glass at room temperature for another 10–15 min avoiding light exposure before experiment. This step allows the cells to reach equilibrium with the ECB and room temperature. 2. Prepare 5 histamine solution by adding 1 μL of 1000 histamine stock solution to 200 μL of ECB. Vortex briefly and spin down. 3. Image cells using TIRF microscopy. Find the desired cells and set up time-lapse imaging process. In general, acquiring an image every 10–15 s is fast enough to monitor the dynamic change in PI(4,5)P2 levels in the PM. 4. Start the imaging process. After taking images of resting cells for 30 s–1 min, add 50 μL of 5 histamine solution. A marked reduction in GFP-PLCδ-PH intensity is expected immediately after histamine addition (Fig. 3a, middle panel). Add the treatment in between frames so the imaging process will proceed uninterrupted. To rapidly generate a homogenous histamine solution of 100 μM, carefully pipet up and down a few times without touching any parts of the imaging cover glasses. 5. Continue imaging for another 5–10 min until no obvious change in GFP-PLCδ-PH intensity was observed. A partial recovery in GFP-PLCδ-PH intensity can be observed within 2 min (see Fig. 3a, right panel).

3.1.3 Image Analysis

1. Open acquired images using Fiji (https://fiji.sc/). 2. Open the plugin Time Series Analyzer V3 (the plugin can be downloaded here: https://imagej.nih.gov/ij/plugins/timeseries.html) and two windows (Time Series V3_0 and ROI Manager) will appear. 3. Use the polygon selection tool to draw a region of interest (ROI) covering the PM area stably attached throughout the imaging process. Once done, click on “Add” (hotkey “t”) in the ROI Manager window. Repeat this step to select more ROIs and a region avoiding cells as a background measurement.

Analyzing PI Transfer at ER-PM Junctions in Live Cells

7

Fig. 3 Analysis of PI(4,5)P2 replenishment and Nir2 translocation to ER-PM junctions during receptor stimulation. (a) Changes in intensity of GFP-PLCδ-PH following histamine treatment in HeLa cells cotransfected with H1R monitored by TIRF microscopy. Representative images are shown. Scale bar: 10 μm. (b) Example traces of typical PI(4,5)P2 consumption-replenishment during receptor stimulation as described in (a). Traces from four individual cells are shown. (c) Example traces of Nir2 translocation to ER-PM junctions during receptor stimulation as described in Fig. 2b. Individual traces from 29 ER-PM junctions (gray) and the derived average trace (red) in a cells are shown

4. Go to Analyze > Set Measurements and check the mean gray value. Once done, click “Get Average” in Time Series V3_0 window and time trace data for the individual ROI will be generated. 5. Copy the time trace data into a spreadsheet. Subtract the background data from the ROI data and normalize the background-subtracted data to time zero (the frame right before the addition of 5 histamine solution). 6. Plot and analyze the traces. Please refer to examples in Fig. 3b.

8

Chi-Lun Chang and Jen Liou

3.2 Measurement of PITP Translocation to ER-PM Junctions During Receptor Stimulation

1. Transfect HeLa cells on imaging cover glasses with 50 ng of H1R plasmid, 40 ng of MAPPER plasmid, and 60 ng of Nir2mCherry plasmid (or Nir3-mCherry) following the procedure described in Subheading 3.1.

3.2.1 Transfection 3.2.2 Live-Cell Imaging

1. Prepare cells and set up imaging process as described in Subheading 3.1. Use MAPPER channel to find an optimal TIRF focal plane for imaging. Nir2 is in the cytosol and shows no concentration at ER-PM junctions in resting cells; thus, the Nir2 channel is not suitable to set up the focal plane for the experiment (Fig. 2b, upper panels). 2. Start the imaging process. After taking images of resting cells for 30 s–1 min, add 50 μL of 5 histamine solution. Nir2 accumulation at ER-PM junctions (colocalization with MAPPER) is detectable within 30 s and reaches a plateau within 1–2 min after histamine addition (Fig. 2b, bottom panels).

3.2.3 Image Analysis

1. Open acquired images using Fiji and open the plugin Time Series Analyzer V3 as described in Subheading 3.1. 2. Use the oval selection tool to draw an ROI that covers an ER-PM junction using the MAPPER channel. Once done, click on “Add” (hotkey “t”) in the ROI Manager window. Repeat this step to select 15–30 representative ER-PM junctions in each cell as well as a background region. 3. Right click on the images to duplicate Nir2 channel and check “show all” in the ROI manager window. The same ROIs selected in MAPPER channel will appear in the duplicated Nir2 channel. An alternative approach is to directly draw ROIs covering Nir2 puncta in a poststimulation (~2 min) image of Nir2 channel when MAPPER channel is not available. 4. Set Measurements to mean gray value. Click “Get Average” in Time Series V3_0 window and time trace data for the individual ROI will be generated. 5. Copy the results into a spreadsheet. Subtract the background from the ROI data and normalize the background-subtracted data to time zero. 6. Plot and analyze the traces. Please refer to examples in Fig. 3c.

Analyzing PI Transfer at ER-PM Junctions in Live Cells

4

9

Notes 1. A PM marker without affinity to PI(4,5)P2 would be ideal to include, at least in initial experiments, to make sure that there is no significant photobleaching and the cells do not move out of the TIRF plane during the experiment. 2. Due to the nature of transient transfections, the expression level of H1R varies from cell to cell. Cells occasionally exhibit overconsumption when H1R level is too high (Fig. 4). In these cells, PI(4,5)P2 consumption is too strong and dominates the endogenous replenishing mechanisms leading to no or little PI (4,5)P2 replenishment. In contrast, cells with low H1R overexpression show underconsumption and a very narrow window of PI(4,5)P2 replenishment, similar to HeLa cells without H1R overexpression. Please be aware of these two types of cells as they should be excluded from data analysis to avoid biased interpretation. It would be useful to optimize transfection condition for suitable H1R expression levels or alternatively have cell lines stably expressing H1R. 3. GFP-PLCδ-PH domain is the most commonly used biosensor for PI(4,5)P2 in live cells. Nonetheless, PLCδ-PH domain also binds to IP3 generated via PI(4,5)P2 hydrolysis [15], which may cause inaccurate measurements of PI(4,5)P2 in the PM using fluorescence microscopy. Therefore, validation of the results using other approaches is recommended. For example, Tubby-GFP is another well-documented PI(4,5)P2 biosensor for live cells with no significant affinity for IP3 [16]. PI(4,5)P2 immunostaining is another approach to visualize PI(4,5)P2 in fixed cells [17].

Fig. 4 Example traces of underconsumption (green) and overconsumption (red)

10

Chi-Lun Chang and Jen Liou

4. Original studies used confocal microscopy to characterize GFP-PLCδ-PH as a PI(4,5)P2 biosensor in live cells [12, 13]. Confocal midsection images were taken to detect GFP-PLCδ-PH signal in the PM. In our experience, the midsection of HeLa cells tends to move following treatment during live-cell imaging. This situation makes quantification more difficult and increases the noise of the results. The adherent surface of HeLa cells usually shows little or no movement throughout the imaging process. Thus, using TIRF microscopy not only reduces phototoxicity but also leads to easier quantification and more reliable and consistent results. 5. MAPPER labels ER-PM junctions by targeting the ER and PM simultaneously, meaning that MAPPER provides tethering at ER-PM junctions. Similar to other tethers at ER-PM junctions [18, 19], MAPPER overexpression at higher levels leads to expansion of ER-PM junctions. The expanded ER-PM junctions manifest patch-like structures, instead of puncta, as visualized by fluorescence microscopy. To avoid experimental artifacts, cells with patch-like ER-PM junctions should be excluded. It would be useful to optimize transfection condition for suitable MAPPER expression. Selecting stable MAPPERexpressing cell lines with puncta-like ER-PM junction is also recommended. 6. MAPPER targets to ER-PM junctions via binding to PI(4,5)P2 in the PM. Thus, a reduction of MAPPER signal can sometimes be observed following receptor stimulation due to reduction of PI(4,5)P2, especially in H1R overexpressing cells. The decrease in MAPPER intensity does not necessarily represent reduced ER-PM junctions since there are other endogenous tethers to maintain ER-PM junctions. Similarly, Nir2 accumulation at ER-PM junctions in stimulated cells may not indicate an increase in ER-PM junctions. It should be noted that many proteins come and go at ER-PM junctions without affecting these subcellular structures.

Acknowledgements This work was supported by National Institutes of Health grant GM113079. J. Liou is a Sowell Family Scholar in Medical Research. The authors declare no competing financial interests. References 1. Balla T (2013) Phosphoinositides: tiny lipids with giant impact on cell regulation. Physiol Rev 93:1019–1137

2. Di Paolo G, De Camilli P (2006) Phosphoinositides in cell regulation and membrane dynamics. Nature 443:651–657

Analyzing PI Transfer at ER-PM Junctions in Live Cells 3. Chen YJ, Chang CL, Lee WR et al (2017) RASSF4 controls SOCE and ER-PM junctions through regulation of PI(4,5)P2. J Cell Biol 216:2011–2025 4. Berridge MJ (2009) Inositol trisphosphate and calcium signalling mechanisms. Biochim Biophys Acta 1793:933–940 5. Chang CL, Liou J (2015) Phosphatidylinositol 4,5-bisphosphate homeostasis regulated by Nir2 and Nir3 proteins at endoplasmic reticulum-plasma membrane junctions. J Biol Chem 290:14289–14301 6. Kim YJ, Guzman-Hernandez ML, Balla T (2011) A highly dynamic ER-derived phosphatidylinositol-synthesizing organelle supplies phosphoinositides to cellular membranes. Dev Cell 21:813–824 7. Agranoff BW, Bradley RM, Brady RO (1958) The enzymatic synthesis of inositol phosphatide. J Biol Chem 233:1077–1083 8. Chang CL, Liou J (2016) Homeostatic regulation of the PI(4,5)P2-Ca(2+) signaling system at ER-PM junctions. Biochim Biophys Acta 1861:862–873 9. Kim YJ, Guzman-Hernandez ML, Wisniewski E et al (2015) Phosphatidylinositol-phosphatidic acid exchange by Nir2 at ER-PM contact sites maintains phosphoinositide signaling competence. Dev Cell 33:549–561 10. Chang CL, Chen YJ, Liou J (2017) ER-plasma membrane junctions: Why and how do we study them? Biochim Biophys Acta 1864:1494–1506 11. Chang CL, Hsieh TS, Yang TT et al (2013) Feedback regulation of receptor-induced Ca2+ signaling mediated by E-Syt1 and Nir2 at endoplasmic reticulum-plasma membrane junctions. Cell Rep 5:813–825

11

12. Stauffer TP, Ahn S, Meyer T (1998) Receptorinduced transient reduction in plasma membrane PtdIns(4,5)P2 concentration monitored in living cells. Curr Biol 8:343–346 13. Varnai P, Balla T (1998) Visualization of phosphoinositides that bind pleckstrin homology domains: calcium- and agonist-induced dynamic changes and relationship to myo-[3H]inositol-labeled phosphoinositide pools. J Cell Biol 143:501–510 14. Steyer JA, Almers W (2001) A real-time view of life within 100 nm of the plasma membrane. Nat Rev Mol Cell Biol 2:268–275 15. Ferguson KM, Lemmon MA, Schlessinger J et al (1995) Structure of the high affinity complex of inositol trisphosphate with a phospholipase C pleckstrin homology domain. Cell 83:1037–1046 16. Quinn KV, Behe P, Tinker A (2008) Monitoring changes in membrane phosphatidylinositol 4,5-bisphosphate in living cells using a domain from the transcription factor tubby. J Physiol 586:2855–2871 17. Hammond GR, Schiavo G, Irvine RF (2009) Immunocytochemical techniques reveal multiple, distinct cellular pools of PtdIns4P and PtdIns(4,5)P(2). Biochem J 422:23–35 18. Giordano F, Saheki Y, Idevall-Hagren O et al (2013) PI(4,5)P(2)-dependent and Ca(2+)regulated ER-PM interactions mediated by the extended synaptotagmins. Cell 153:1494–1509 19. Varnai P, Toth B, Toth DJ et al (2007) Visualization and manipulation of plasma membraneendoplasmic reticulum contact sites indicates the presence of additional molecular components within the STIM1-Orai1 Complex. J Biol Chem 282:29678–29690

Chapter 2 Monitoring Non-vesicular Transport of Phosphatidylserine and Phosphatidylinositol 4-Phosphate in Intact Cells by BRET Analysis Mira Sohn, Daniel J. Toth, and Tamas Balla Abstract Non-vesicular lipid transport via lipid transfer proteins (LTPs) at membrane contact sites (MCSs) is critical for the maintenance of the lipid composition of biological membranes. The ability to measure lipid transfer activity of diverse LTPs in live cells without interrupting the fine structural organization is essential to better understand the contribution of non-vesicular lipid transport to membrane organization. Here, we describe a semiquantitative method to analyze phosphatidylinositol 4-phosphate (PI4P) and phosphatidylserine (PS) changes at the plasma membrane (PM) as they relate to LTP functions. This live cell method is based on bioluminescence resonance energy transfer (BRET) measurements between a luciferase-tagged lipidrecognizing module and a PM-targeted fluorescent acceptor. Oxysterol-binding protein-related protein (ORP) 5 is a PI4P/PS lipid transfer protein which is stably tethered to the ER and also dynamically interacts with PM PI4P/PI(4,5)P2 via its pleckstrin homology (PH) domain. We show that this method is able to detect PI4P and PS changes in the PM upon acute recruitment of an ORP5 construct to the PM. This method is convenient and robust, utilizing population of cells in 96-well plates analyzed in a plate reader. We will also highlight potential further applications extending the method for other LTPs and other lipid cargoes. Key words Lipid transfer proteins, Membrane contact sites, Plasma membrane, ORP5, Phosphatidylinositol 4-phosphate, Phosphatidylserine, Bioluminescence resonance energy transfer

1

Introduction Membrane contact sites (MCSs) are specialized regions of biological membranes where different organelles are juxtapositioned with their membranes forming tight contacts separated by only 10–30 nm. MCSs are highly conserved in eukaryotes from yeasts to mammals [1, 2]. MCSs are important as sites where intracellular organelles dynamically exchange various molecules. Lipids, in particular, have a highly unfavorable energetics when crossing from one membrane to another via the aqueous phase due to their hydrophobicity. Lipids can be transported between

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_2, © Springer Science+Business Media, LLC, part of Springer Nature 2019

13

14

Mira Sohn et al.

membranes in a number of ways, but mostly use either vesicular lipid transport or lipid transfer proteins (non-vesicular lipid transport) [3]. Non-vesicular lipid transport utilizes numerous lipid transfer proteins (LTPs) to deliver lipid molecules at MCSs with a wide variety of cargo specificity and intracellular localization. LTPs are classified by their structural features. For instance, Oxysterolbinding protein (OSBP) family or OSBP-related proteins (ORPs), which are found in both mammals (12 homologs) and yeast (7 homologs), possess an OSBP-related domain (ORD) as the site for lipid transfer [4]. Some ORPs also possess FFAT motifs to interact with the ER-protein VAP-A and VAP-B, and N-terminal PH domains to recognize phosphatidylinositol 4-phosphate (PI4P) or other phosphoinositides [4]. The ORD domain of OSBP and the yeast protein Osh4 can transport both PI4P and cholesterol [5, 6] while the ORD domain of ORP5 and its yeast homolog Osh6p are capable of transporting PI4P from the PM to the ER and phosphatidylserine (PS) in the reverse direction [7, 8]. In the following sections we will describe how to examine the lipid transfer function of LTPs using ORP5 as an example. Most commonly, lipid transfer activities of LTPs have been examined using recombinant LTPs and liposomes of various lipid compositions applied to in vitro lipid transfer assays. This method measures the ability of an LTP to exchange lipids between donor and acceptor liposomes of specific lipid compositions. While these assays have been and remain extremely useful to study these processes under well-defined in vitro conditions, they have several limitations. First, they require highly purified recombinant proteins that may work differently in isolation than they do inside the cells where some other proteins or lipids will affect their functions. Second, the lipid composition of the liposomes is only a best estimate of what is found inside the cell in the real membranes. Third, the ratio of aqueous to lipid phase in the cell is vastly different from that used in the in vitro lipid transfer reactions. An additional problem arises if a fluorescent lipid analogue is used, which may behave very differently from its endogenous counterparts. Some of these caveats are overcome when studying LTPs (usually as GFP-tagged constructs) in live cells and study their distribution and movements. This can be combined with fluorescent lipid sensors to monitor changes in lipid distribution using confocal microscopes [9]. However, changes observed in these microscopy studies are not easy to quantitate and the individual cell-to-cell variations make it difficult to make firm conclusions without analyzing a large number of cells with proper statistics and without bias. Here we describe in detail a method that we developed to analyze the lipid transfer function of ORP5 based on bioluminescence resonance energy transfer (BRET) analysis. This method

BRET-Based Quantitation of PI4P/PS Transfer Events

15

combines the power of quantitation of changes in specific lipid species at the PM. We utilized chimeric ORP5 by replacing its pleckstrin homology (PH) domain which binds the PM via PI4P and PI(4,5)P2 with an FK506-binding protein12 (FKBP12) module. Combined with the expression of a PM-targeted FRB module, this system allows an acute establishment of an ER-PM contact upon addition of rapamycin and hence acutely turn on the lipid transfer process. In this present protocol, we combine this recruitable system with BRET analysis to measure PM PI4P or PS changes upon activation of the ORP5-mediated lipid transfer process. The original design of the BRET system was published by the Varnai group [10] and they describe the basic principles in a separate chapter (see Chapter 3) in this volume. Our method monitors PI4P and PS changes in a 96-well format read by a simple fluorescence plate reader. It is a perfect complement of microscopy studies where similar lipid changes can be monitored in single cells, but BRET takes out the bias from the measurement and immediately gives a cell population average. This method should be quite useful in cellular knockdown or knockout studies where population measurements are more informative. Lastly, this method can be extended to various FKBP-fused LTPs and to various membrane contacts in addition to the ER-PM contact sites.

2 2.1

Materials Cell Seeding

1. HEK-AT1 cell line: HEK293 cell stably expressing the rat AT1a angiotensin receptor [11]. The parent HEK293 cells are available from ATCC and are equally useable with this protocol. HEK-AT1 cells are cultivated in Dulbecco’s Modified Eagle Medium with high glucose (4500 mg/L) containing 10% (v/v) Fetal Bovine Serum (FBS) supplemented with 1% Penicillin/Streptomycin and 5 μg/mL of PlasmocinTM Prophylactic (InvivoGen) which is used to ensure mycoplasma-free condition (see Note 1). PlasmocinTM Prophylactic is used at higher concentration (25 μg/mL) only for one week after thawing the cells. Cells were grown in T80 tissue culture flasks with filter caps and incubated in 5% humidified CO2 incubator at 37  C. All experiments are performed in cell passages from 4 to 17 after thawing (see Note 1). 2. Poly-L-lysine (Sigma): 100 diluted in sterilized phosphate buffer saline (PBS), filtered (0.22 μm pore size), and stored in a 4  C refrigerator.

2.2

Transfection

1. DNA constructs: L10-mVenus-T2A-sLuc-P4M(2) to monitor PM PI4P [10]. L10-mVenus-T2A-sLuc-Lact-C2 was generated by utilizing the PS reporter Lact-C2 [12] as described

16

Mira Sohn et al.

previously and used to measure PM PS [13]. As a PM-anchoring recruiter, Lyn-targeted FRB (PM2-FRB), generated in the Meyer lab [14], is used. mCherry-tagged FKBPORP5 was created by replacing the PH domain with an FKBP12 module as described in our recent publication [15] (see Note 2).These plasmids are available upon request from either the Varnai group (Budapest, Hungary) or the Balla group (NIH, USA). 2. Effectene Transfection Reagent kit (Qiagen) containing EC buffer, Effectene solution, and Enhancer solution. 2.3 BRET Measurement

1. Modified Krebs-Ringer buffer: 120 mM NaCl, 4.7 mM KCl, 0.7 mM MgSO4, 10 mM glucose, and 10 mM HEPES-NaOH, pH 7.4 (see Note 3). 2. Ca2+-containing Krebs-Ringer buffer: Modified Krebs-Ringer buffer with 2 mM CaCl2 (final concentration). 3. Coelenterazine h working solution: Coelenterazine h (Regis Technologies Inc) is dissolved in 100% ethanol to make a 500 μM stock concentration stored unfrozen in a 20  C freezer with no light (see Note 4). Dilute stock solution to 1/40 in Ca2+-containing Krebs-Ringer buffer to get a final concentration of 12.5 μM coelenterazine h. Prepare freshly. 4. Rapamycin working solution: Rapamycin (MilliporeSigma) is dissolved in DMSO at 100 μM and stored at 20  C (see Note 5). Dilute rapamycin stock solution to 1/100 in Ca2+-containing Krebs-Ringer buffer (final 1 μM concentration, see Note 5). Prepare freshly. 5. Control solution: Dilute DMSO to 1/100 in Ca2+-containing Krebs-Ringer buffer. 6. Tristar2 LB 942 Multimode Microplate Reader (Berthold Technologies). The plate reader is equipped with 540/40 nm (mVenus fluorescent measurement) and 475/20 nm (luciferase measurement) emission filters.

3 3.1

Methods Cell Seeding

1. Coat nontransparent white 96-well plates with 200 μL/well of diluted poly-L-lysine in tissue culture hood (room temperature) for 30 min–1 h. 2. Remove poly-L-lysine and wash the plate once with 200 μL of PBS 1 per well. 3. Seed 30,000 cells per well in triplicates in 200 μL of medium (see Table 1 and Note 6). 4. Incubate the cells for one day before transfection.

BRET-Based Quantitation of PI4P/PS Transfer Events

17

Table 1 Cell seeding in a 96-well plate ➀ ➀ ➀ ➁ mCherry mCherry mCherry mCherry-FKBPORP5

➁ mCherry-FKBPORP5

➁ mCherry-FKBPORP5

➂ ➂ ➂ ➃ mCherry mCherry mCherry mCherry-FKBPORP5

➃ mCherry-FKBPORP5

➃ mCherry-FKBPORP5

Cells were seeded in triplicates as displayed above. The cells are transfected with the indicated construct plus the noncolored plasma membrane-targeted FRB and the PM-BRET sensor for the particular lipid in question. Group ➀ and ➁ are subjected to DMSO treatment while ➂ and ➃ are to rapamycin treatment. It is highly recommended that this grouping and relative positioning in the 96-well plate is changed from experiment-to-experiment to avoid possible systemic errors biasing the final conclusion

Table 2 Transfection mixture (A total of 1.8 μg per transfection groupsa (300 ng/well  6 wells) of DNA mixture is prepared according to the following table) Control

FKBP-ORP5

EC buffer

180 μL (30 μL/well  6 wells)

180 μL (30 μL/well  6 wells)

PM2-FRB

600 ng (100 ng/well  6 wells)

600 ng (100 ng/well  6 wells)

FKBP-ORPs

mCherry 600 ng (100 ng/well  6 wells)

mCherry-FKBP-ORP5 600 ng (100 ng/well  6 wells)

BRET sensor

L10-mVenus-T2A-sLuc-P4M(2) 600 ng (100 ng/well  6 wells, for PI4P measurement) or L10-mVenus-T2A-sLuc-Lact-C2 600 ng (100 ng/well  6 wells, for PS measurement)

a

Since we plan to use control and treatment (both in triplicates), our transfection groups are calculated for 6 plates

3.2

Transfection

1. Dilute plasmids to 100 ng/μL concentration each (see Note 7). 2. Mix plasmids according to Table 2 (see Note 8). 3. Add 2.4 μL of Enhancer solution per well (a total of 14.4 μL for 6 wells). 4. Vortex for 2–3 s and incubate for 5 min at room temperature. 5. Add 3 μL of Effectene solution per well (a total of 18 μL for 6 wells). 6. Vortex for 2–3 s and incubate for 10 min in room temperature. 7. Add 1/6 of transfection mixture to each well. 8. Incubate at 37  C for 1 day in a tissue culture incubator.

3.3 BRET Measurement

1. Set up the plate reader. With the Berthold Tristar2 LB 942 reader we set up measurements at every 15 s, and in each

18

Mira Sohn et al.

cycle, mVenus and luciferase signals are measured for 0.25 s, each (see Note 10). 2. Remove medium from the 96-well plate and wash the cells once with 200 μL of Krebs-Ringer buffer. 3. Incubate the cells in 50 μL of Krebs-Ringer buffer at room temperature for 30 min (see Note 9). 4. Pipette 40 μL of coelenterazine h working solution (final 5 μM coelenterazine h) to all wells and immediately measure basal level BRET signals with emission wavelengths (λem) for both Renilla luciferase (465–485 nm) and mVenus (520–560 nm) for 4–5 min. 5. Pipette 10 μL of control solution or rapamycin working solution (final 100 nM concentration) and immediately start BRET measurements for 30 min (see Note 10). 3.4

Data Analysis

1. Calculate mVenus/luciferase ratio from each emission intensity at each time point of the measurement (see Note 8). 2. Calculate mean values of triplicates (see Note 10) (Fig. 1).

Fig. 1 Quantitation of PM PI4P (a) and PS (b) changes using BRET analysis under acute recruitment of ORP5 to the PM. Representative raw BRET ratios from a single experiment are shown. One day after transfection, BRET analysis was performed before and after PM recruitment of ORP5 using rapamycin (100 nM) treatment for the indicated number of cycles (15 s each). The values plotted are means of triplicate measurements where the BRET ratio was calculated by dividing the Venus counts by the luciferase counts at each time points. Note the differences in the initial BRET values justifying the needs for appropriate controls. The ratios of rapamycin/ DMSO values in each group are close to one

BRET-Based Quantitation of PI4P/PS Transfer Events

19

Fig. 2 Final results from three independent BRET experiments showing changes in PM PI4P (a) and PS (b) after acute recruitment of ORP5 to the PM. In this plot rapamycin/DMSO ratio values are calculated and normalized to the mean ratio value obtained before rapamycin addition in each of the three experiments. Means  S.E.M. (Standard Error of the Mean) are shown from the three experiments, each performed in triplicates

3. Analyze relative change in lipid levels upon PM recruitment of FKBP-ORP5 by calculating rapamycin/DMSO values at each time point of measurement. 4. For an absolute comparison between control and FKBPORP5, normalize each value (rapamycin/DMSO) to the mean BRET values obtained before rapamycin addition (see Notes 10 and 11) (Fig. 2).

4

Notes 1. Testing mycoplasma contamination regularly and maintaining mycoplasma-free condition is essential for obtaining reliable and reproducible data. In addition, culturing and passaging cells under consistent conditions (e.g., confluency, passages used in experiments) will ensure reproducibility of experiments. 2. When recruiting proteins to the PM in BRET experiments one has to consider that an mRFP or mCherry-tagged protein recruited to the PM can start “stealing” energy from the luciferase-excited Venus, thereby decreasing the BRET signal. This may appear as if the PM lipid measured by BRET is decreasing in response to the recruited protein. To test for this “FRET contamination” of the BRET signal, one has to perform some controls. The best is to use a LTP construct mutated in its lipid cargo binding site to make it functionally dead. We found that in the case of the mCherry-FKBP12tagged ORP5, which occupies only a small fraction of the PM after PM recruitment because of its ER-anchoring, this FRET

20

Mira Sohn et al.

contamination is negligible. However, when a cytosolic mCherry-FKBP12 protein was used as a control, this FRET contamination was significant. Therefore, for most of our BRET applications, we used a “dark” (nonfluorescent) recruiter and an iRFP-tagged recruitable construct to avoid these FRET-based artifacts. 3. Modified Krebs-Ringer buffer was stored at room temperature and freshly prepared with CaCl2 addition (final 2 mM concentration) prior to individual experiment. 4. Coelenterazine h is light sensitive. Keeping coelenterazine h in darkness is important for reproducibility of BRET analysis. In addition, it is also recommended that the stock solution is kept in screw-capped tubes to minimize evaporation of ethanol. 5. By making aliquots (10 μL each), freezing and thawing cycle of rapamycin was minimized. 6. Although it is impossible to check with microscope if cells are seeded evenly in the nontransparent 96-well plate, even cell seeding is critical to ensure good efficiency and consistency of transfection. 7. Using midiprep quality DNA is highly recommended to maintain high transfection efficiency. Due to low DNA amount used in 96-well scale transfections, plasmid DNA was diluted to 100 ng/μL concentration with ultra-pure water. This procedure improves accuracy of pipetting during transfection. Meanwhile, minimizing freeze-thawing cycles of diluted DNA is important to prevent DNA denaturation. 8. The BRET value can be influenced by co-transfection efficiency of the additional individual DNAs. Especially, expression levels of mVenus and luciferase in the BRET sensor can affect the absolute BRET values. Thus, it is recommended that the emission intensity of luciferase (475/20 nm) is checked for each transfection groups to make sure they are comparable within the various groups to be compared. 9. It may be required to repeat the same experiment in 37  C and to check the temperature dependency of the process in question. 10. Although we set the plate reader to measure for 15 s per cycle and 0.25 s per each emission, each cycle will run with a delay while the plate is moved from well to well (less than 15 s) and therefore the time stamp for each measurement could be different from the set cycle time. This is an important consideration when setting up too many wells in one plate. This time delay is reflected in the final figures presented as real running time including delays which resulted in longer measurement time than 30 min.

BRET-Based Quantitation of PI4P/PS Transfer Events

21

11. The raw counts in the Luc channel in our plate reader are between 250,000 and 350,000. These numbers decline with time as the coelenterazine substrate is consumed. The counts in the Venus channel are also between 250,000 and 350,000. The ratio of Venus/Luc is expected to be in the range of 0.8–1.0 in case of a strong BRET signal. This BRET ratio (used as the raw BRET value) also changes in time; therefore, it is always important to have a control group that receives only solvent treatment. This control serves as a reference for any treatment regime. The result then is expressed as a BRET ratio of treated/control. Occasionally these final ratios show a slight difference even before any treatment (even if they should not), in which case the data can be adjusted to correct this difference. However, caution should be used, because the expression level of the BRET constructs has an impact on the absolute BRET ratios. When using expression of other plasmids along with the BRET construct, they can have an effect on the expression level of the latter. It is important to use appropriate control DNAs to ensure equal BRET construct expression. Our practice is to check if the counts in the Luc channel are comparable between different transfections or knockdowns groups. In some cases it may be even better to run the plates through in fluorescent mode determining the Venus expression level to evaluate whether the BRET components are equally expressed among treatment groups.

Acknowledgments We are grateful for DNA constructs provided by the Grinstein, Meyer, Yin, and Varnai laboratories. This work was supported by the intramural research program of the Eunice Kennedy Shriver NICHD, at the National Institutes of Health. References 1. Gatta AT, Levine TP (2017) Piecing together the patchwork of contact sites. Trends Cell Biol 27:214–229 2. Saheki Y, De Camilli P (2017) Endoplasmic reticulum-plasma membrane contact sites. Annu Rev Biochem 86:659–684 3. Lahiri S, Toulmay A, Prinz WA (2015) Membrane contact sites, gateways for lipid homeostasis. Curr Opin Cell Biol 33:82–87 4. Kentala H, Weber-Boyvat M, Olkkonen VM (2016) OSBP-related protein family: mediators of lipid transport and signaling at membrane contact sites. Int Rev Cell Mol Biol 321:299–340

5. de Saint-Jean M, Delfosse V, Douguet D et al (2011) Osh4p exchanges sterols for phosphatidylinositol 4-phosphate between lipid bilayers. J Cell Biol 195:965–978 6. Mesmin B, Bigay J, Moser von Filseck J et al (2013) A four-step cycle driven by PI(4)P hydrolysis directs sterol/PI(4)P exchange by the ER-Golgi tether OSBP. Cell 155:830–843 7. Chung J, Torta F, Masai K et al (2015) Intracellular transport. PI4P/phosphatidylserine countertransport at ORP5- and ORP8mediated ER-plasma membrane contacts. Science 349:428–432

22

Mira Sohn et al.

8. Moser von Filseck J, Copic A, Delfosse V et al (2015) Intracellular transport. Phosphatidylserine transport by ORP/Osh proteins is driven by phosphatidylinositol 4-phosphate. Science 349:432–436 9. Hammond GR, Balla T (2015) Polyphosphoinositide binding domains: key to inositol lipid biology. Biochim Biophys Acta 1851:746–758 10. Toth JT, Gulyas G, Toth DJ et al (2016) BRET-monitoring of the dynamic changes of inositol lipid pools in living cells reveals a PKC-dependent PtdIns4P increase upon EGF and M3 receptor activation. Biochim Biophys Acta 1861:177–187 11. Hunyady L, Baukal AJ, Gaborik Z et al (2002) Differential PI 3-kinase dependence of early and late phases of recycling of the internalized AT1 angiotensin receptor. J Cell Biol 157:1211–1222

12. Yeung T, Gilbert GE, Shi J et al (2008) Membrane phosphatidylserine regulates surface charge and protein localization. Science 319:210–213 13. Sohn M, Ivanova P, Brown HA et al (2016) Lenz-Majewski mutations in PTDSS1 affect phosphatidylinositol 4-phosphate metabolism at ER-PM and ER-Golgi junctions. Proc Natl Acad Sci U S A 113:4314–4319 14. Inoue T, Heo WD, Grimley JS et al (2005) An inducible translocation strategy to rapidly activate and inhibit small GTPase signaling pathways. Nat Methods 2:415–418 15. Sohn M, Korzeniowski M, Zewe JP et al (2018) PI(4,5)P2 controls plasma membrane PI4P and PS levels via ORP5/8 recruitment to ER-PM contact sites. J Cell Biol 217:1797–1813

Chapter 3 Development of Nonspecific BRET-Based Biosensors to Monitor Plasma Membrane Inositol Lipids in Living Cells Jo´zsef T. To´th, Gergo˝ Gulya´s, La´szlo´ Hunyady, and Pe´ter Va´rnai Abstract There are several difficulties to face when investigating the role of phosphoinositides. Although they are present in most organelles, their concentration is very low, sometimes undetectable with the available methods; moreover, their level can quickly change upon several external stimuli. Here we introduce a newly improved lipid sensor tool-set based on the balanced expression of luciferase-fused phosphoinositide recognizing protein domains and a Venus protein targeted to the plasma membrane, allowing us to perform Bioluminescence Resonance Energy Transfer (BRET) measurements that reflect phosphoinositide changes in a population of transiently transfected cells. This method is highly sensitive, specific, and capable of semiquantitative characterization of plasma membrane phosphoinositide changes with high temporal resolution. Key words Phosphoinositide, BRET, Lipid-binding domain, Plasma membrane, Biosensor

1

Introduction Detecting intracellular lipids and among them phosphoinositides (PPIn) in particular is a challenging task as they are present in small amounts and show rapid changes and high metabolic turnover [1]. During the last four decades several techniques have been developed that aimed to detect these small changes with high sensitivity and good temporal and spatial resolution [2]. Isotopic labeling of cells with myo-[3H]-inositol or 32P-phosphate followed by lipid separation with different types of chromatography (e.g., paper chromatography, high performance liquid chromatography) were the first experiments able to detect PPIns [3–6]. The major limitations of these methods were that only with the use of complicated membrane separation procedures was it possible to detect PPIns with subcellular resolution, and these techniques lacked the ability to distinguish between different PPIn isomers. The latter remains a problem even for more novel mass spectrometry detection methods [7–11].

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_3, © Springer Science+Business Media, LLC, part of Springer Nature 2019

23

24

Jo´zsef T. To´th et al.

Using microscopy and fluorescently tagged lipids or labeled PPIn specific antibodies can overcome these problems, but these techniques also have their limitations. When using fluorescently labeled lipids, their modified structure and hydrophobicity can influence their role in the enzymatic routes and their incorporation into membrane bilayers, and makes it very difficult to interpret the results regarding endogenous lipids [12, 13]. When applying antibodies it is always doubtful whether they can distinguish between regio-isomers, and their use is limited to fixed cells [14, 15]. Another method to make lipid changes visible is the application of labeled specific lipid-binding proteins or lipid-binding domains (LBD) [16–19]. Specific LBDs are already available for all seven PPIns [20, 21]. When the level of a certain PPIn in an organelle membrane increases, these LBDs will also be enriched there and this can be followed easily by epifluorescence or confocal microscopy. These experiments enable the detection of massive lipid changes with good temporal and subcellular resolution but the quantification of results is complicated and limited only to enormous, in some cases nonphysiological changes. These LBDs, however, can also be used for more sensitive and semiquantitative resonance energy transfer methods, e.g., Fo¨rster resonance energy transfer (FRET) [22] or BRET [23]. An energy transfer signal occurs when there is molecular proximity (approximately 100 A˚) between two fluorescently labeled proteins (in the case of FRET) or between a fluorescently tagged (usually with yellow fluorescent protein [YFP] or its improved version such as Venus) and a bioluminescent luciferase-bound protein [24]. When creating PPIn recognizing biosensors we preferred BRET over FRET as BRET measurements do not involve donor excitation and therefore the analysis of BRET data does not require corrections for bleedthrough of fluorescent proteins, a major complication for FRET data analysis (Fig. 1a) [16]. The energy transfer signal that we get can be either specific when there is a specific connection between the tagged molecules, or nonspecific when the signal is a result of their colocalization in the same membrane region [23]. Fusing a specific membranetargeting sequence to the energy transfer acceptor Venus and a LBD to the donor luciferase enables the measurement of lipid changes in one particular membrane compartment. When the tagged membrane contains the target PPIn in relatively high amounts, LBD-luciferase constructs will accumulate there which generates a (nonspecific) energy transfer signal between the LBD-luciferase and membrane-Venus. On the contrary, a reduction in PPIn levels causes the LBD-luciferase to dissociate from the membrane and will result in a reduced signal (Fig. 1b) [25, 26]. The main difficulty of the method is that it is necessary that cells express both energy donor and energy acceptor molecules, and to make experiments more comprehensive it is required to express

BRET-Based Inositol Lipid Biosensors

25

Fig. 1 Schematic representation of bioluminescence resonance energy transfer (BRET) and its application in the quantitative measurement of inositol lipid changes. (a) The BRET approach is based on the measurement of emitted light at the wavelength of the donor and acceptor. The ratio of these values correlates with the efficiency of energy transfer and represents the molecular proximity of the luciferase enzyme (Luc), the donor that generates the light by oxidizing its substrate, and the acceptor fluorescent protein (Venus) that is excited by the energy transfer and emits at a longer wavelength. (b) Membrane localization of a lipid-binding domain (LBD) can be measured in cells by calculating the energy transfer between the luciferase fused to the LBD and Venus targeted to the membrane. Membrane recruitment of the probe increases the BRET ratio values, while cytoplasmic translocation of the probe results in a decrease. Since there is no direct interaction between the donor and the acceptor fusion proteins, this is considered a nonspecific energy transfer and depends on the expression level of the proteins

them in a constant ratio. To achieve this we designed these sensor sets in a single plasmid configuration where the two fusion proteins are separated by the T2A viral peptide interrupting the polypeptide chains during translation [27]. Figure 2 shows the general design of the plasmids. This method allows a single mRNA molecule to code for two separate proteins with a theoretical ratio of 1:1, and thus

26

Jo´zsef T. To´th et al.

Fig. 2 Generation of the inositol lipid biosensors. The first step is to select the lipid-binding domain. Since the method is based on the specific interaction between the various inositol lipids and their lipid-binding proteins, first and foremost, well characterized and validated lipid-binding domains need to be selected that are generally accepted and used, e.g., in confocal measurements as fluorescently tagged inositol lipid sensors. A list of these sensors can be found in recent reviews [16]. Our choices were the P4M domain (546–647) of the SidM protein (Q29ST3) (in tandem) for PtdIns(4)P, the PH domain (1–170) of PLCδ1 (P51178) for PtdIns(4,5)P2, the PH domain (266–399) of GRP1 (O08967) with a nuclear export signal (NES, ALQKKLEELELDE) for PtdIns (3,4,5)P3 and the PH domain (109–334) of TAPP1 (Q9HB21) (also in tandem) for PtdIns(3,4)P2. Other domains can be selected for additional sensors, and the method also allows the screening of domains as potential biosensors in the future. We used the GFP-tagged form of each sensor as a starting point. The second step is to replace the fluorescent protein with luciferase. Although many different versions of luciferase enzymes are available, we chose to replace the fluorescent protein with a mutant version (S87R, M185 V, K189 V, V267I) of Renilla luciferase (P27652) [28] which we have been using for many years in our laboratory. Although there are smaller and more stable versions of this enzyme, a great advantage of Renilla luciferase is that the inexpensive and easily available Coelenterazine h can be used as substrate. Most probably, other versions can also be applied with an appropriate substrate and acceptor fluorophore. Keep the original linker sequence of the sensors. For control experiments (to check the localization and the degradation of the fusion proteins) we also generated constructs which contain Cerulean in place of the luciferase. The third step consists of creating a construct that drives the expression of the fluorescent protein Venus on the cytoplasmic surface of the membrane on which the inositol lipid is supposed to be measured. Until now only the plasma membranetargeted Venus has been used, in which the N-terminal 10 amino acid residues of Lck (MGCVCSSNPE) provide the stable recruitment of the protein mostly to the membrane [29].In the last step, the coding sequence of the T2A peptide (EGRGSLLTCGDVEENPGP) is inserted in frame between the two other sequences without a STOP codon. The advantage of this viral T2A peptide is that during translation a molecular cleavage occurs within it, leading to the expression of two separate proteins in equimolar amounts [27]

only transfection of the cells with one plasmid is required for the experiments [16, 25]. Here, we describe a BRET-based methodology, as opposed to directly monitoring translocation of a fluorescently tagged biosensor with microscopic techniques, which has the advantage of simpler quantification, higher sensitivity, and compared to biochemical assays with the application of organelle-selective biosensors, we can achieve PPIn region-isomer detection signal with good temporal and compartment-specific spatial resolution. Nevertheless, BRET is a relatively easy-to-use technique and its popularity will only increase because of its statistical robustness, lower equipment costs, and time commitment.

BRET-Based Inositol Lipid Biosensors

2 2.1

27

Materials Cell Transfection

1. Cultured cells of interest: HEK293T, HEK293, COS-7, Neuro 2A, CHO, HeLa cell lines were tested. For one 96-well plate, about 8–10 million cells grown on a 10 cm Petri dish are required. 2. Opti-MEM medium (Gibco). 3. DMEM (Lonza) supplemented with 10% (v/v) of fetal bovine serum (FBS, Euro Clone). 4. 0.25% trypsin solution. 5. Plasmid DNA (mammalian expression vectors with CMV promoter and kanamycin resistance) coding for the inositol lipid sensors or only the cytoplasmic luciferase enzyme. 6. Transfection reagent: GeneCellin (BioCellChallenge) for HEK293T or Lipofectamine 2000 (Thermo Fisher Scientific) for other cell types. 7. Cell culture microplate, 96-well, flat bottom, white (Greiner Bio-One), coated with poly-lysine or not depending on the cell type that is used.

2.2 BRET Measurement

1. Plate reader with the capacity of measuring luminescent light with high sensitivity (at least 7 amol ATP/well) and with the appropriate filters (480/20 nm filter for Renilla luciferase and 530/20 nm filters for Venus). Fluorescence measurement capability with the appropriate filters or monochromators is highly recommended, because it enables monitoring the level of expression by measuring the protein fluorescence (excitation wavelength at 510 nm and emission wavelength at 535 nm). Temperature control, mixing function, and low dead space injectors (as many as possible) are also recommended (see Note 1). 2. Acquisition software: Time lapse protocol is required that includes the change of fluorescence and luminescence modes, as well as the control of injectors. 3. Stock solution of 0.5 mM coelenterazine h (Regis Technologies) in 100% ethanol. Prepare 1 mL stocks that can be stored at
20  C for months. Dark, screw-cap micro tubes should be used to decrease light-induced degradation and evaporation. It is diluted freshly just before the application in the BRET measurement. 4. Measuring solution (Krebs-Ringer buffer): 10 mM HEPESHCl, pH 7.4, 120 mM NaCl, 4.7 mM KCl, 1.2 mM CaCl2, 0.7 mM MgSO4, 10 mM glucose. 5. Computer with any data analysis software.

28

3

Jo´zsef T. To´th et al.

Methods

3.1 Planning of the BRET Experiment

Considering that measurements are carried out in duplicate or triplicate and those non-stimulated controls must also be included, a high number of the wells is required to allow the simultaneous measurement of several conditions as well as cells transfected with different sensors. Using only the central 6  10 wells of a 96-well plate (we have found that the wells on the edge are not very reliable and therefore are used only for non-transfected cells and for preliminary experiments) and applying only one stimulus (which requires typically two non-stimulated and three stimulated wells distributed randomly), twelve biosensors can be measured (2 in one row, 12 in 6 rows). Obviously, application of various stimuli or pretreatment reduces the number of the biosensors, but still, careful planning is required prior to the experiment to determine both the transfection and the measurement protocol.

3.2

Transfection is carried out in Opti-MEM medium, according to the manufacturer’s protocol,

Cell Transfection

1. For one well of the 96-well plate, mix 25 μL of Opti-MEM containing 1.5 μL of GeneCellin or 0.5 μL of Lipofectamine 2000 with 25 μL of Opti-MEM containing 0.05 μg of plasmid of the biosensor. 2. Incubate at room temperature for 30 min. 3. During that incubation time, cells cultured in a 10 cm tissue culture Petri dish are trypsinized using 5 mL of 0.25% trypsin solution for 5 min at 37  C. Centrifuge the cells at low speed for 5 min. Resuspend the cell in 3 mL of Opti-MEM. A cell density of 750,000 cells/mL is set. 4. For each well, 100 μL of the cell suspension (containing 75,000 cells) is added to 50 μL of the DNA/transfection reagent slurry, then mixed and pipetted into the well. Note that the wells at the edge of the plate must only contain 100 μL of the cells. 5. After 4–6 h, add 100 μL of FBS-supplemented (10%) DMEM to each well. 3.3 BRET Measurement

1. Measurements are performed 24–26 h after transfection and start with the change of medium to 50 μL/well of the KrebsRinger buffer (see Note 2). 2. Preparation of the equipment. Make sure that the plate reader is switched on and reached the set temperature, which is usually 37  C. Fill up the injectors (if used) and open the protocol file (see Note 3). 3. Optional step: measure the fluorescence of Venus prior to the addition of luciferase substrate. Usually the entire plate is measured.

BRET-Based Inositol Lipid Biosensors

29

4. Dilute 0.5 mM stock solution of coelenterazine h with preheated measuring solution (40 dilution). 5. Add 40 μL to each well included in the run (this additional 2.5 dilution achieves the final concentration of 5 μM). The plate has to be briefly shaken. 6. Luminescent measurement. The measurement is started and interrupted several times if drugs are added manually. Either using the injectors or pipettes, drugs are given in a volume of 10 μL, which results in a 10 times dilution for the first addition, 11 times dilution for the second addition, and so on. The plate has to be shaken manually or by the equipment after the administration of drugs. 3.4

Data Analysis

1. Expression level of the constructs. If step 3 in Subheading 3.2 was performed (measuring the Venus fluorescence), the calculation of the average of the wells transfected with the same constructs reflects the expression level of proteins. The error is also useful as it reveals how similar the parallels were. The fluorescence of non-transfected cells can be used for autofluorescence correction. 2. Analysis of the luminescence curves—The kinetics and also the values of both emission curves (emission at 480–530 nm) are very similar, because the level of energy transfer is usually small; therefore, it is enough to analyze only one of them (we usually analyze the 480 nm). Emission curves have a characteristic kinetic (see Fig. 3, upper middle graph). The initial fast increase reflecting the diffusion of the substrate into the cells is always followed by a slower decrease, which is due to the irreversible degradation of the enzymatic activity. The disappearance of the signal is a limitation regarding the length of the measurement. Usually the average of the three biggest values of the corresponding wells are selected (red circle), and used for the calculation of the average, which also reflects the expression level of the sensor proteins. 3. Calculate the BRET ratio values by dividing the emission intensity measured at 530 nm with the one measured at 480 nm. If the applied stimulus resulted in a change compared to the non-stimulated controls, two separate groups of curves should be seen (see Fig. 3, upper right graph). 4. We are always expecting a time-dependent spontaneous change of the BRET ratio values even in the case of the non-stimulated cells. This change is characteristic for the various sensors, and can be greatly reduced by preincubating the cells in the measuring solution at 37  C (see Note 2). To maintain the absolute values of the BRET ratio, this spontaneous change is determined by calculating the difference between each data point

30

Jo´zsef T. To´th et al.

Fig. 3 Workflow of the data analysis. The analysis of the measurement of five wells is shown, in which cells were transfected with the same biosensor, and only one stimulus was applied after the tenth data point in three wells, while two wells got the vehicle only (non-stimulated). BRET ratio values that correlate with the lipid levels are calculated by dividing the emission intensity values measured at 530 nm with the ones measured at 480 nm (upper graphs). Spontaneous change calculated from the average of the non-stimulated curves (lower left graph) is used for the correction of the stimulated curves (lower middle graph). Finally, data are expressed as a percentage value, where 100% is the basal BRET ratio and 0% is the value that can be measured using the same equipment and protocol in cells that express luciferase only (in this experiment 0.468). Red circles show the data that reveals the expression level of the biosensors (upper middle graph) and the basal BRET ratio (lower middle graph). In this experiment stimulus was applied manually, which takes time and leads to a huge time gap on the graph. The exact addition of the stimulus cannot be determined as indicated with question mark (lower right graph). This gap can be greatly reduced by using injectors for stimulations

and the average of the three points prior to stimulation (see Fig. 3, lower left graph). 5. Correction of stimulated curves for the spontaneous change of the control. By subtracting the spontaneous change from each of the stimulated curves we get the absolute ratio values that already reflect the membrane localization of the sensor proteins (see Fig. 3, lower middle graph). The average of the three data points prior to stimulation is calculated and considered as basal BRET ratio value (red circle). Although this value depends on many factors (filters, sensor expression level), it can be used to evaluate the resting membrane binding level of the lipidbinding domains. 6. Finally, the normalized BRET ratios, which reflect the change of the inositol lipids, are expressed as a percentage value where 100% is the basal BRET ratio value and 0% is set to the absolute

BRET-Based Inositol Lipid Biosensors

31

minimum of the BRET ratio, determined by wells where cells express the luciferase protein only (Fig. 3, lower middle and right graphs) (see Note 4).

4

Notes 1. There are many plate readers on the market capable of BRET measurements. In our laboratory, the following readers have been used: Mithras LB940 (Berthold Technologies), Varioskan Flash (Thermo Fisher Scientific), CLARIOstar (BMG Labtech). When purchasing a new plate reader, the following issues should be considered: (1) Both fluorescence and luminescence detection. Fluorescence measurement is needed to follow the expression of the fluorescence protein. (2) Wave selection by monochromators or filters. While monochromators are excellent for the measurement of the Venus fluorescence, for luminescence (BRET), the application of filters may result in greater signals. (3) Top reading is required when white plates, which give better signals, are used. (4) The number of injectors determines how versatile a measurement can be. (5) The acquisition software and support availability are also important issues. 2. We noticed that after changing the media to the Krebs-Ringer solution, incubation of the cells at 37  C for 30–60 min greatly reduces the spontaneous change of the BRET ratio values, most probably by stabilizing the fluorescence of the Venus protein. Pretreatment of the cells can be performed during this time. 3. The following points should be considered for the creation of the right measurement protocol. First, BRET measurements are typically carried out in several runs. The number of wells included in one run depends on many different factors, such as time resolution, signal-to-noise ratio, and exposure time. Measurement of the two emission intensities (at emission wavelength of 480–530 nm) cannot happen exactly at the same time (which would be ideal), but at least has to occur sequentially for each well (in the so-called “by well” configuration, as opposed to “by plate”). The usual exposure time is a few hundred milliseconds (250 ms in our experiments) for each wavelength, which means that together with the mechanical movement of the plate, the measurement of one well requires 0.75–1.5 s. For example, when two rows (20 wells) are measured in a run, roughly 3–4 measurement points can be achieved in every minute. This number can be increased by decreasing the exposure time, but that in turn will increase the signal-to-noise ratio. Better time resolution can be achieved by

32

Jo´zsef T. To´th et al.

including fewer wells in each run, which in turn increases the number of runs and the length of the entire experiment. Based on our experience, the total length of the experiment (with preincubation) should be limited to 120–150 min. 4. To get the ratio value corresponding to 0% energy transfer, additional experiments are required in which cells express the cytoplasmic luciferase enzyme, but no forms of the Venus protein. Depending on the optical parameters of the equipment (e.g., filters) this number can be quite different. The advantage of this normalization is that experiments carried out using different equipment will be comparable, plus the effect of different expression levels of the biosensors achieved in the various experiments is also reduced. Figure 4 reveals an

Fig. 4 Effects of the activation of a Gq-coupled receptor on plasma membrane inositol lipids. HEK293 cells expressing rat AT1a receptor were stimulated with submaximal concentration of angiotensin II (AngII). Cells were transiently transfected with the biosensors indicated on the figure. For these experiments, only two rows of a 96-well plate were used without the first and last wells of each row. Means  S.D., n ¼ 4

BRET-Based Inositol Lipid Biosensors

33

example of an experiment in which four different sensors were used.

Acknowledgments This work was supported by the Hungarian National Research, Development and Innovation Fund Grants NKFIH K105006 (to PV), NVKP 16-1-2016-0039 (to LH). The technical assistance of Kata Szabolcsi and Da´niel Nagy is highly appreciated. References 1. Balla T (2013) Phosphoinositides: tiny lipids with giant impact on cell regulation. Physiol Rev 93:1019–1137 2. Idevall-Hagren O, De Camilli P (2015) Detection and manipulation of phosphoinositides. Biochim Biophys Acta 1851:736–745 3. Hokin LE, Hokin MR (1958) Phosphoinositides and protein secretion in pancreas slices. J Biol Chem 233:805–810 4. Hokin MR, Hokin LE (1953) Enzyme secretion and the incorporation of P32 into phospholipides of pancreas slices. J Biol Chem 203:967–977 5. Nakanishi S, Catt KJ, Balla T (1995) A wortmannin-sensitive phosphatidylinositol 4-kinase that regulates hormone-sensitive pools of inositolphospholipids. Proc Natl Acad Sci U S A 92:5317–5321 6. Traynor-Kaplan AE, Harris AL, Thompson BL et al (1988) An inositol tetrakisphosphatecontaining phospholipid in activated neutrophils. Nature 334:353–356 7. Hsu FF, Turk J (2000) Characterization of phosphatidylinositol, phosphatidylinositol-4phosphate, and phosphatidylinositol-4,5bisphosphate by electrospray ionization tandem mass spectrometry: a mechanistic study. J Am Soc Mass Spectrom 11:986–999 8. Kielkowska A, Niewczas I, Anderson KE et al (2014) A new approach to measuring phosphoinositides in cells by mass spectrometry. Adv Biol Regul 54:131–141 9. Wenk MR, Lucast L, Di Paolo G et al (2003) Phosphoinositide profiling in complex lipid mixtures using electrospray ionization mass spectrometry. Nat Biotechnol 21:813–817 10. Kraft ML, Klitzing HA (2014) Imaging lipids with secondary ion mass spectrometry. Biochim Biophys Acta 1841:1108–1119 11. Wakelam MJ (2014) The uses and limitations of the analysis of cellular phosphoinositides by

lipidomic and imaging methodologies. Biochim Biophys Acta 1841:1102–1107 12. Golebiewska U, Kay JG, Masters T et al (2011) Evidence for a fence that impedes the diffusion of phosphatidylinositol 4,5-bisphosphate out of the forming phagosomes of macrophages. Mol Biol Cell 22:3498–3507 13. Lipsky NG, Pagano RE (1983) Sphingolipid metabolism in cultured fibroblasts: microscopic and biochemical studies employing a fluorescent ceramide analogue. Proc Natl Acad Sci U S A 80:2608–2612 14. Hammond GR, Dove SK, Nicol A et al (2006) Elimination of plasma membrane phosphatidylinositol (4,5)-bisphosphate is required for exocytosis from mast cells. J Cell Sci 119:2084–2094 15. Hammond GR, Schiavo G, Irvine RF (2009) Immunocytochemical techniques reveal multiple, distinct cellular pools of PtdIns4P and PtdIns(4,5)P(2). Biochem J 422:23–35 16. Varnai P, Gulyas G, Toth DJ et al (2017) Quantifying lipid changes in various membrane compartments using lipid binding protein domains. Cell Calcium 64:72–82 17. Varnai P, Balla T (2008) Live cell imaging of phosphoinositides with expressed inositide binding protein domains. Methods 46:167–176 18. Balla T, Varnai P (2009) Visualization of cellular phosphoinositide pools with GFP-fused protein-domains. Curr Protoc Cell Biol 24 (24):24 19. Stauffer TP, Ahn S, Meyer T (1998) Receptorinduced transient reduction in plasma membrane PtdIns(4,5)P2 concentration monitored in living cells. Curr Biol 8:343–346 20. Hammond GR, Balla T (2015) Polyphosphoinositide binding domains: key to inositol lipid biology. Biochim Biophys Acta 1851:746–758

34

Jo´zsef T. To´th et al.

21. Stahelin RV, Scott JL, Frick CT (2014) Cellular and molecular interactions of phosphoinositides and peripheral proteins. Chem Phys Lipids 182:3–18 22. van der Wal J, Habets R, Varnai P et al (2001) Monitoring agonist-induced phospholipase C activation in live cells by fluorescence resonance energy transfer. J Biol Chem 276:15337–15344 23. Toth DJ, Toth JT, Gulyas G et al (2012) Acute depletion of plasma membrane phosphatidylinositol 4,5-bisphosphate impairs specific steps in endocytosis of the G-protein-coupled receptor. J Cell Sci 125:2185–2197 24. Hamdan FF, Percherancier Y, Breton B et al (2006) Monitoring protein-protein interactions in living cells by bioluminescence resonance energy transfer (BRET). Curr Protoc Neurosci 5(5):23 25. Toth JT, Gulyas G, Toth DJ et al (2016) BRET-monitoring of the dynamic changes of inositol lipid pools in living cells reveals a PKC-dependent PtdIns4P increase upon EGF

and M3 receptor activation. Biochim Biophys Acta 1861:177–187 26. Kuo MS, Auriau J, Pierre-Eugene C et al (2014) Development of a human breast-cancer derived cell line stably expressing a bioluminescence resonance energy transfer (BRET)-based phosphatidyl inositol-3 phosphate (PIP3) biosensor. PLoS One 9:e92737 27. Szymczak AL, Workman CJ, Wang Y et al (2004) Correction of multi-gene deficiency in vivo using a single ‘self-cleaving’ 2A peptide-based retroviral vector. Nat Biotechnol 22:589–594 28. Woo J, von Arnim AG (2008) Mutational optimization of the coelenterazine-dependent luciferase from Renilla. Plant Methods 4:23 29. Gulyas G, Radvanszki G, Matuska R et al (2017) Plasma membrane phosphatidylinositol 4-phosphate and 4,5-bisphosphate determine the distribution and function of K-Ras4B but not H-Ras proteins. J Biol Chem 292:18862–18877

Chapter 4 Following Anterograde Transport of Phosphatidylserine in Yeast in Real Time Juan Martı´n D’Ambrosio, Ve´ronique Albane`se, and Alenka Cˇopicˇ Abstract In order to understand how lipids are sorted between cellular compartments, kinetic assays are required to selectively follow the transport of lipid species in cells. We present here a microfluidics-based protocol to follow the transport of phosphatidylserine (PS) in yeast cells from the site of its synthesis, the endoplasmic reticulum (ER), to downstream compartments, primarily the plasma membrane under our conditions. This assay takes advantage of yeast cells lacking Cho1, the enzyme responsible for PS synthesis. Lyso-PS can be added exogenously and is taken up by the cells and converted to PS. Because acylation of lyso-PS to PS appears to occur at the ER, anterograde transport of PS from the ER can then be followed by fluorescent microscopy using the specific PS reporter C2Lact-GFP. We describe the construction of the required cho1Δ yeast strain and the preparation of lyso-PS. We present an example of the use of this assay to follow the activity of the yeast PS transport proteins Osh6 and Osh7. Key words Phosphatidylserine, Lipid transport, Fluorescence microscopy, Fluorescent probe, Microfluidics, Saccharomyces cerevisiae, Real-time assay

1

Introduction Recent years have brought a huge increase in our understanding of the function of the so-called lipid transport proteins, in no small part thanks to development of sophisticated in vitro assays that monitor the activity of purified lipid transporters on synthetic bilayers [1, 2]. However, assays in cells are required to understand how these proteins operate in the complex cellular environment. This is not an easy task, as lipids are very difficult to follow inside the cells. Monitoring their transport requires kinetic assays and specific labeling of individual lipid species. This is especially difficult for monitoring the anterograde transport of newly synthesized cellular lipids. Radioactive labeling of lipids through precursors can be very specific but usually lacks the temporal resolution required to measure the kinetics of the transport process, which generally happens

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_4, © Springer Science+Business Media, LLC, part of Springer Nature 2019

35

36

Juan Martı´n D’Ambrosio et al.

on the order of minutes [3, 4]. An alternative is to use fluorescent labeling coupled with fluorescent microscopy, either by making the lipid itself fluorescent or by using a fluorescent probe that specifically recognizes the lipid in question. Phosphatidylserine (PS) distribution can be followed thanks to a genetically encoded lactadherin C2 domain (C2Lact) fluorescent probe [5, 6]. C2Lact binds to membrane-embedded PS species and does not appear to be significantly affected by the surrounding membrane environment in an in vitro analysis [7]. When expressed in various wild-type cells, it accumulates at the plasma membrane (PM) and in endocytic compartments, because the cytoplasmic leaflets of these membranes are enriched in PS [5, 8, 9]. In yeast, C2Lact-GFP signal is largely confined to the PM, but can also be observed accumulating at the ER, on vesicles or on the vacuolar surface under mutant conditions [9–11]. In cells depleted of PS, however, C2Lact is largely cytosolic. In budding yeast and in plants, PS synthesis is controlled by a single gene, CHO1/PSS1, whose product resides in the endoplasmic reticulum (ER) [12, 13]. Null mutants in this gene are devoid of PS, but they can still grow, and their cellular processes are not grossly perturbed [8, 11, 14]. However, PS can be added exogenously to cho1Δ yeast or plants in the form of lyso-PS, which can cross the cell membrane and, once inside, gets acylated into PS by a poorly understood mechanism [11, 15]. Exogenous addition of lyso-PS to cho1Δ cells was used to analyze the role of two yeast lipid exchange proteins, Osh6 and Osh7, in the transport of PS from the ER to the PM [9, 10]. This strategy has allowed us to demonstrate that counter transport of phosphoinositide-4-phosphate enables PS transport by Osh6/7 [10]. Here, we describe a method to follow anterograde transport of PS from the ER in real time in a quantitative manner by fluorescence microscopy, using the PS probe C2Lact-GFP (Fig. 1a, b). Our protocol requires the use of a microfluidics device, but can be adapted to imaging on a coated glass slide [9, 10]. We also present a simpler batch protocol that can be used for a qualitative assessment of cell’s ability to incorporate PS into the PM membrane (Fig. 1c). Because a cho1Δ yeast strain is required for this assay, we first present a protocol to generate this deletion mutant by transformation and homologous recombination. We then explain the preparation of lyso-PS emulsions, followed by the lyso-PS transport assay. Finally, we describe our method of quantifying PS distribution between the yeast cytosol, the ER, and the PM using the C2Lact-GFP reporter (Fig. 2).

Fig. 1 Changes in PS subcellular levels in cho1Δ yeast strains after lyso-PS addition. (a) Mutant cho1Δ cells were incubated with Lyso-PS in a microfluidics chamber and C2Lact-GFP localization was followed for 20 min; select time-points are shown. Initially, C2Lact-GFP shows a cytosolic localization. A transient PM signal is usually observed after 3 min, corresponding to the time required for lyso-PS to reach the cells. Internalized lyso-PS is acylated into PS and after a lag period, C2Lact-GFP signal can be transiently observed at the ER (in this case at 11 min, see Note 12). PS is then quickly transported to the PM, reflected in large increase in C2Lact-GFP at the cell surface. (b) Cells lacking PS transport proteins Osh6/7 (cho1Δ osh6/7Δ strain) display the same timing of C2Lact-GFP appearance at the ER, but unlike in cho1Δ cells, C2Lact-GFP remains at the ER during the time of live imaging. (c) PS transport assay performed in batch in cho1Δ cells. After 30 min of lysoPS treatment, C2Lact-GFP signal is mostly at the PM, whereas longer incubations lead to a loss of PM signal and increase in cytosolic C2Lact-GFP. Filled and dashed arrows indicate transient PM and ER C2Lact-GFP signal, respectively. Scale bar: 4 μm

38

Juan Martı´n D’Ambrosio et al.

Fig. 2 Quantification of PS distribution in cells using C2Lact-GFP reporter. Using Image J software, an intensity profile is plotted from a line crossing a cell (lower panel). Plateau profile is given for the cytosolic C2Lact-GFP in cho1Δ cells lacking PS (a, left panel). Upon relocalization of C2Lact-GFP to the PM after lyso-PS addition, a large increase in peripheral cellular fluorescence peaks can be observed (a, middle panel). Peaks corresponding to cortical and perinuclear ER are shown in the cho1Δ osh6/7Δ mutant (a, right panel), in accordance with the localization of the ER marker Sec63-GFP (b). Scale bar: 4 μm

2

Materials Use sterile technique when working with yeast: sterilize solutions and plastic consumables, clean bench with 70% ethanol, and use a Bunsen burner to reduce contaminations. Regularly change solutions and monitor for contaminations. Lipid work requires a hood due to toxic solvents (e.g., chloroform) and the use of protective gloves and clothing. Make sure to use solvent-resistant materials (glass vials, Hamilton syringe). Dispose of solvents according to regulations.

2.1 Yeast Growth and Transformation

1. Rich medium (YPD): 1% (w/v) yeast extract, 2% (w/v) peptone, 2% (w/v) dextrose (liquid or with 2% (w/v) agar). Sterilize by autoclaving. For plates, cool to ~50
C and pour about 25 mL per plate. If using antibiotic selection, add desired antibiotic to cooled media. 2. Synthetic drop-out medium (SD): 0.67% (w/v) yeast nitrogen base without amino acids, 2% (w/v) dextrose, amino acid dropout supplement (e.g., without histidine, without histidine and uracil; the exact amount depends on the supplement). Liquid or with 2% (w/v) agar, sterilize by autoclaving. For growing cho1Δ yeast, supplement with 1 mM ethanolamine (from 500 mM stock solution).

Phosphatidylserine Transport Assay in Yeast

39

3. Plasmids: pRS416-LactC2-GFP (Haematologic technologies), pFA6a-His3MX6 or plasmids bearing other deletion markers for generating gene disruption cassettes [16, 17]. 4. PCR cassette for generating cho1Δ::HIS3 genomic replacement: generated by high-fidelity PCR using pFA6a-His3MX6 and the following forward and reverse primers [16]: TTTTCATTTTGTGGGTGATTGTCATTTTTAGTTGTCT ATTTGATTCAATCcgtacgctgcaggtcgac and AAAGTT ATATGTACAAATTTTTTTTGACGCCAGGCATGAA CAAAAACTACatcgatgaattcgagctcg. PCR reaction (50 μL total volume) contains the following: 30 ng of plasmid DNA, 0.25 μM each primer, 250 μM each dNTPs, 1 Herculase reaction buffer, Herculase II fusion DNA polymerase (Avanti), and is performed using the following program: 2 min at 95
C; 30 cycles of 20 s at 95
C, 30 s at 55
C, 1 min at 72
C; a final extension for 7 min at 72
C. Verify by agarose gel electrophoresis the presence of a single band (see Note 1). 5. Taq’ozyme purple mix 2 (Ozyme). 6. Transformation solution I: 100 mM LiOAc, 10 mM Tris–HCl, pH 7.5, 1 mM EDTA. Prepare from stock solutions: 1 M LiOAc, 1 M Tris–HCl, pH 7.5, 500 mM EDTA. Sterilize by filtering through a 0.22 μm filter or by autoclaving. 7. 40% (w/v) PEG-4000: Weigh 40 g of PEG-4000, slowly add to a beaker containing 70 mL of transformation solution I, stir until dissolved, and then adjust volume to 100 mL. Sterilize by filtering through a 0.22 μm filter. 8. 10 mg/mL single-stranded DNA from salmon testes (Sigma): boil 5 min at 95
C before use and cool on ice. 2.2 Preparation of Lyso-PS

1. 18:1 Lyso-PS (1-oleoyl-2-hydroxy-sn-glycero-3-phospho-Lserine) at 5 mg/mL in chloroform (Avanti Polar Lipids). Store at 20
C under argon in small aliquots to minimize freeze-thawing. Do not use after expiration date (1 year). 2. Glass vials (1 mL) with teflon caps. 3. Hamilton syringe (100 μL). 4. Argon gas. 5. Bovine serum albumin (BSA), fatty-acid free (Sigma): Prepare 6.1 mg/mL solution in liquid SD medium, dissolve by gentle agitation and sterilize by filtering through a 0.22 μm filter. This solution can be stored at 4
C.

40

Juan Martı´n D’Ambrosio et al.

2.3 Fluorescent Imaging and Image Analysis

1. Microscope DMI6000 (Leica), equipped with an oil immersion plan apochromat 100 objective NA 1.4, a QuantEM cooled EMCCD camera (Photometrics), and a spinning-disk confocal system CSU22 (Yokogawa). 2. MetaMorph 7 software (Molecular Devices). 3. CellASIC ONIX Microfluidic Perfusion Platform, driven by CellASIC ONIX software (Millipore). 4. YO4C microfluidics chamber for haploid yeast (Millipore). 5. Glass slides and coverslips. 6. Image analysis: Fiji (Image J, version: 2.0.0-rc-19/1.49m), Excel (Microsoft) or similar spreadsheet software.

3

Methods

3.1 cho1Δ Strain Preparation

Yeast homologous recombination can be used for specifically directing the insertion of a disruption DNA cassette into the yeast genome and thus generating a deletion mutation of the desired gene. This transformation method is based on the procedure of Gietz et al. [18]. 1. Day 1: Inoculate 10 mL of YPD medium with a single yeast colony of the starting haploid yeast strain. Incubate at 30
C in a shaker overnight (190 rpm). 2. Day 2: Measure the optical density at 600 nm (OD600) of the overnight culture. Dilute the culture into 50 mL of medium in a sterile 250 mL flask so that the final OD600 is 0.2. 3. Incubate at 30
C, shaking at 190 rpm, until OD600 ¼ 0.8–1.0 (duplication time for wild-type yeast ~90 min). 4. Spin cells at 1000  g for 2 min at room temperature (the samples should be kept at room temperature from here on). Remove supernatant and resuspend cell pellet in 1 mL of sterile water. 5. Spin cells at 1000  g for 2 min and remove liquid. Add 500 μL of transformation solution I. Resuspend cells by swirling or gently vortexing. 6. For each transformation aliquot 50 μL of cells to a sterile 1.5 mL tube. We strongly recommend to set up a control tube without DNA. 7. Add 5 μL (~0.5–1 μg) of the cho1Δ PCR cassette and 5 μL of carrier single-stranded salmon sperm DNA. Mix gently. 8. Add 300 μL of 40% PEG solution and mix by inverting 2–3 times (do not vortex, as yeast cells are more fragile in the presence of PEG).

Phosphatidylserine Transport Assay in Yeast

41

9. Incubate the tubes at 30
C for 30 min. 10. Transfer to 42
C for 15 min. 11. Allow to cool and spin the cells at 300  g for 2 min. Resuspend cell pellet in 50 μL of sterile water. For a selection by auxotrophy (e.g., His), spread on SD-His plates containing 1 mM ethanolamine. For a selection by antibiotic resistance, resuspend gently in 500 μL of YPD medium and incubate at 30
C for 1 h before spreading on plates containing the corresponding antibiotic. 12. Incubate at 30
C for ~4 days (see Note 2). 13. Day 6: Streak out single colonies on a new selection plate (see Note 3). Check gene disruption by performing PCR on genomic DNA [19] using a primer in the promoter of the deleted gene and one primer in the auxotrophy or antibiotic resistance gene. Alternatively, genomic DNA can be prepared from a 5 mL overnight culture by disrupting cells using bead lysis (10 min vortex at high speed at 4
C with 0.5 mm glass beads) and isolating DNA from the cell lysate using a plasmid mini-prep kit. In the final step, DNA is eluted in 25 μL of elution buffer. PCR reaction (20 μL total volume) contains 3 μL of genomic DNA (~300 ng), 0.25 μM each primer and 1 Taq’ozyme purple mix 2 (Ozyme, France), and is performed using the following program: 5 min at 95
C; 30 cycles of 30 s at 95
C, 30 s Tm-5
C, 1 min/Kb 72
C, 1 min 72
C; final extension 7 min at 72
C. Verify by agarose gel electrophoresis the presence of a single band. 14. Once verified, the cho1Δ strain should be transformed with the plasmid carrying the PS-sensor C2Lact-GFP following the same protocol as described above (see Note 4). 3.2 PS Transport Assay

1. Day 1: Start 5 mL of cho1Δ culture at 30
C in appropriate SD medium supplemented with 1 mM ethanolamine with shaking at 190 rpm overnight (see Note 5). 2. Day 2: Prepare lyso-PS emulsion. Using a Hamilton syringe, transfer 24 μL of stock lyso-PS chloroform solution (5 mg/ mL) into a glass vial (see Note 6). 3. Dry off chloroform under argon to make a film on the vial (see Note 7). 4. Dried lyso-PS can be solubilized directly in water solutions. Add 1 mL of SD medium, vortex, and incubate for 30 min at 37
C with occasional vortexing. This yields 216 μM lyso-PS emulsion (see Note 8). 5. Alternatively, lyso-PS can be dissolved in a BSA solution, which improves lyso-PS solubility. Dissolve dried lyso-PS in 0.5 mL of BSA solution, then add 0.5 mL of SD medium. This yields

42

Juan Martı´n D’Ambrosio et al.

lyso-PS:BSA complexes at 4:1 stoichiometric ratio. However, we do not recommend using this preparation for kinetic assays (see Note 9). 6. Dilute lyso-PS to a working concentration of 108 μM, vortex again, keep at room temperature (see Note 10). 7. Check that yeast cultures are in mid-logarithmic phase (OD600 ~ 0.6), note their OD600 and harvest 3 mL by centrifugation (1000  g). Discard supernatant, wash cells once, and resuspend them in SD medium. In our hands, the optimal final concentration of cells is 1 OD600 in 160 μL for microfluidicsbased assay, or 1 OD600 in 50 μL for batch lyso-PS assay, but this should be adjusted depending on strain and absorbance reader. 8. Performing lyso-PS time-course experiment: Pipette SD medium, lyso-PS emulsion, and yeast suspension into appropriate wells of the Y04C microfluidics plate. Flow SD medium into observation chamber, then inject cells and maintain them in a constant flow of fresh medium at 3 psi, corresponding to a flow rate of ~20 μL/h (see Note 11). At time 0, switch the flow to lyso-PS emulsion. Take cell images with a spinning-disk microscope every 2 min for 20–30 min in several different positions (see Note 12). We usually take 5 z-sections separated by 0.5 μm, imaging cells in four different positions (see Note 13) (Fig. 1a, b). 9. Alternatively, for a batch lyso-PS incubation: Place cells in 1.5 mL tube and resuspend in SD medium or in lyso-PS emulsion. Image cells on glass slides at 0 min (SD medium) for 20–30 min after lyso-PS addition on a spinning-disk microscope in 5 z-sections separated by 0.5 μm (see Note 14) (Fig. 1c). 3.3 Quantitative Image Analysis

To monitor the distribution of PS in individual cells during the lyso-PS time-course, the distribution of C2Lact-GFP is quantified in one dimension in selected cells at selected time-points of the timecourse. The pixel-intensity profile from a line crossing a single cell is plotted and peaks are analyzed and identified using Image J (Fiji) following the procedure described below (Fig. 2). 1. Select the best z section of each time-point and make a copy (image > duplicate). 2. Concatenate your time-point (Image > Stacks > Tools > Concatenate).

images

3. Correct cell translation using the plugin StackReg (Plugins > StackReg > Transformation > Translation) (see Note 15).

Phosphatidylserine Transport Assay in Yeast

43

4. Select the time-points to be analyzed, depending on the timing of signal transition, following an evaluation by eye. Delete the other time-points from the stack. 5. Open ROI manager (Analyze > Tools > ROI manager) and check “show all.” 6. Draw a line through a single cell and press “add [t]” in the “ROI manager.” Repeat the same procedure with as many cells as possible. To avoid bias, select the cells to be analyzed at t ¼ 0 (before lyso-PS addition), and analyze C2Lact-GFP signal in all selected cells at all selected time-points. 7. Select the first line in the left panel of ROI manager and plot pixels profile (Analyze > Plot profile). 8. Find peaks (BAR>Data analysis > find peaks). Check: min peak amplitude 10–100, exclude peaks on edges of plot, list values. Copy all data into a spreadsheet. 9. Repeat this analysis for all selected cells in the next selected time-point, each time generating a new spreadsheet. 10. Use the average of the two peripheral maximum peak values (corresponding to PM or cortical ER) and the two highest internal peaks for analysis. Subtract from these peaks the average value of all minima within the cell to get the relative heights of peripheral and internal maxima. Normalize these values with average fluorescence intensity of each cell, i.e., average intensity of all pixels within the two peripheral peaks (see Note 16).

4

Notes 1. In general, a 50 nucleotide homology region in the forward and reverse primer (depicted in capital letters) is sufficient for homologous recombination into the target locus. We order desalted primers (basic level of purification). 2. Cells deleted for CHO1 grow considerably more slowly compared to wild-type cells; therefore, a longer period of incubation is needed to obtain colonies after transformation. 3. Make sure to take a single colony for checking your gene disruption by PCR and for further work, any single wild-type cells can interfere with the result. 4. Transformation with a plasmid is generally much more efficient; therefore, a lower quantity of cells can be used. However, cho1Δ strains have lower transformation efficiency than wildtype cells, and we use 3–5 mL of cells per transformation reaction. 1 mM ethanolamine should be added whenever cho1Δ cells are grown in minimal media.

44

Juan Martı´n D’Ambrosio et al.

5. Since cho1Δ are slow growing, we recommend starting overnight cultures 15–20 h before performing the experiment. Fresh transformants should be used (plates not older than 2 weeks), because cho1Δ strains do not keep well on SD plates. 6. Great care should be taken in storing and handling lyso-PS. Always store it in its original solvent (usually chloroform) at 20
C under argon gas and limit the number of freeze-thaw cycles. Lyso-PS cannot be stored longer than 1 year. For a shorter time (1–4 weeks), lyso-PS can also be stored at 20
C in dried aliquots that are ready for solubilization. Quality of the reagent can be checked by dynamic light scattering after preparing the lyso-PS emulsion. An emulsion prepared from fresh lyso-PS contains primarily particles with sizes in the 10 nm range, which would be the expected size of lyso-PS micelles. In emulsions of old lyso PS, particles are more heterogenous, with peak values around 1 nm. 7. This is very fast, but make sure that all chloroform has evaporated. Five minutes with continuous gas flow is largely sufficient. Whereas chloroform itself is toxic for cells, we find that the presence of any organic solvent (DMSO, ethanol) in the lyso-PS emulsion will interfere with the PS transport assay. 8. Even with greatest care, we find that there is some variability in the quality of lyso-PS emulsions, which is reflected in variability in the length of the lag phase in PS transport assay (time between the first contact of lyso-PS with cells and the appearance of C2Lact-GFP signal at the ER; see Fig. 1a and Suppl Fig. 8 in ref. [10]). This is likely due to variations in the active lyso-PS concentrations, as well as the nonlinear response of cells to lyso-PS. We also note that sonication of lyso-PS emulsions or adjusting the pH in the range from 5.5 to 7.4 did not help with decreasing this variability. 9. We find that the addition of BSA to lyso-PS changes the cellular response by eliminating the lag phase in the assay. This is likely due to increased active concentration of lyso-PS in the medium. 10. Lyso-PS water emulsion should be prepared fresh, kept at room temperature in a low-adsorption tube (glass vial or similar), and used within a few hours (4 h maximum, with some variability between different preparations). Do not put on ice or freeze. 11. We wash the observation chamber with growth medium for 5–10 min before injecting the cells. Under our experimental settings, it takes 3 min to exchange buffer in the microfluidics chamber. 12. There is some variability in the timing of the cellular response to lyso-PS, which is likely due to variability in lyso-PS active

Phosphatidylserine Transport Assay in Yeast

45

concentration in the emulsion. Three minutes after injection of lyso-PS (time 0), we observe a small and transient increase in C2Lact-GFP fluorescence at the PM, likely reflecting fast transport of lyso-PS across the cell membrane (note that the 3 min time-point corresponds to the void volume of the microfluidics plate at our selected flow rate). C2Lact-GFP signal then returns to the initial state, suggesting that lyso-PS did not accumulate at the PM but started being transported to the ER. We note that in vitro, C2Lact binds to both PS and lyso-PS in liposomes (G. Drin, personal communication). Eight to fifteen minutes after start, an increase in C2Lact-GFP signal at the ER can be observed, followed by a fast transition to C2Lact-GFP at the cell periphery in wild-type cells (see Fig. 1a and Suppl Fig. 8 in ref. [10]). This variable lag phase (time between the first increase in C2Lact-GFP fluorescence at the cell periphery and its appearance at the ER) reflects nonlinearity of the cellular response to lyso-PS and suggests that a threshold level of PS at the ER has to be reached before export of PS from the ER is activated. 13. To facilitate quantification, select imaging fields with intermediate cell concentration, containing 10–25 cells, but not large clusters. Whereas cells are generally well immobilized in the microfluidics chamber, some movement may occur during the time-course. For this reason, we image several z-sections at each time-point. 14. We find that 20–30 min is the optimal time to detect an increase of C2Lact-GFP signal at the PM, and possibly at the ER, although the signal increase at the ER is more transient, even under mutant conditions. We do not recommend performing batch incubation in a quantitative manner. 15. ImageJ Plugin websites: StagReg: An ImageJ plugin for the recursive alignment of a stack of images. Philippe The´venaz, Biomedical Imaging Group, Swiss Federal Institute of Technology Lausanne. http://bigwww.epfl.ch/thevenaz/stackreg/. BAR version 1.5.2-SNAPSHOT 2017-04-08 A curated collection of Broadly Applicable Routines for ImageJ. Ferreira et al. (2017). DOI https://doi.org/10.5281/zenodo.495245 16. We use an Excel spreadsheet for sorting and manipulating raw data imported from Image J. This excel sheet identifies two maximal peripheral and internal peaks and normalizes their values to average cellular fluorescence.

Acknowledgments We thank Jackson-Verbavatz and Leon labs for support, A-C Gavin for sharing yeast strains, and the ImagoSeine facility at the Institut Jacques Monod, member of the France-Bio-Imaging national

46

Juan Martı´n D’Ambrosio et al.

research structure (ANR-10-INBS-04). This work is funded by the French National Research Agency grant ExCHANGE (ANR-16CE13-0006) and by the CNRS. References 1. Wong LH, Copic A, Levine TP (2017) Advances on the transfer of lipids by lipid transfer proteins. Trends Biochem Sci 42:516–530 2. Moser von Filseck J, Vanni S, Mesmin B et al (2015) A phosphatidylinositol-4-phosphate powered exchange mechanism to create a lipid gradient between membranes. Nat Commun 6:6671 3. Kannan M, Lahiri S, Liu LK et al (2017) Phosphatidylserine synthesis at membrane contact sites promotes its transport out of the ER. J Lipid Res 58:553–562 4. Georgiev AG, Johansen J, Ramanathan VD et al (2013) Arv1 regulates PM and ER membrane structure and homeostasis but is dispensable for intracellular sterol transport. Traffic 14:912–921 5. Yeung T, Gilbert GE, Shi J et al (2008) Membrane phosphatidylserine regulates surface charge and protein localization. Science 319:210–213 6. Andersen MH, Graversen H, Fedosov SN et al (2000) Functional analyses of two cellular binding domains of bovine lactadherin. Biochemistry 39:6200–6206 7. Del Vecchio K, Stahelin RV (2018) Investigation of the phosphatidylserine binding properties of the lipid biosensor, Lactadherin C2 (LactC2), in different membrane environments. J Bioenerg Biomembr 50:1–10 8. Platre MP, Noack LC, Doumane M et al (2018) A combinatorial lipid code shapes the electrostatic landscape of plant endomembranes. Dev Cell 45:465–480 e411 9. Maeda K, Anand K, Chiapparino A et al (2013) Interactome map uncovers phosphatidylserine transport by oxysterol-binding proteins. Nature 501:257–261 10. Moser von Filseck J, Copic A, Delfosse V et al (2015) INTRACELLULAR TRANSPORT. Phosphatidylserine transport by ORP/Osh proteins is driven by phosphatidylinositol 4-phosphate. Science 349:432–436

11. Fairn GD, Hermansson M, Somerharju P et al (2011) Phosphatidylserine is polarized and required for proper Cdc42 localization and for development of cell polarity. Nat Cell Biol 13:1424–1430 12. Atkinson K, Fogel S, Henry SA (1980) Yeast mutant defective in phosphatidylserine synthesis. J Biol Chem 255:6653–6661 13. Yamaoka Y, Yu Y, Mizoi J et al (2011) PHOSPHATIDYLSERINE SYNTHASE1 is required for microspore development in Arabidopsis thaliana. Plant J 67:648–661 14. Natarajan P, Wang J, Hua Z et al (2004) Drs2p-coupled aminophospholipid translocase activity in yeast Golgi membranes and relationship to in vivo function. Proc Natl Acad Sci U S A 101:10614–10619 15. Riekhof WR, Wu J, Gijon MA et al (2007) Lysophosphatidylcholine metabolism in Saccharomyces cerevisiae: the role of P-type ATPases in transport and a broad specificity acyltransferase in acylation. J Biol Chem 282:36853–36861 16. Longtine MS, McKenzie A 3rd, Demarini DJ et al (1998) Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 14:953–961 17. Janke C, Magiera MM, Rathfelder N et al (2004) A versatile toolbox for PCR-based tagging of yeast genes: new fluorescent proteins, more markers and promoter substitution cassettes. Yeast 21:947–962 18. Gietz D, St Jean A, Woods RA et al (1992) Improved method for high efficiency transformation of intact yeast cells. Nucleic Acids Res 20:1425 19. Looke M, Kristjuhan K, Kristjuhan A (2011) Extraction of genomic DNA from yeasts for PCR-based applications. BioTechniques 50:325–328

Chapter 5 Imaging Lipid Metabolism at the Golgi Complex Serena Capasso and Giovanni D’Angelo Abstract The development of fluorescence-based molecular imaging has revolutionized cell biology allowing the visualization of specific biomolecules at the microscopic and, more recently, at the nanoscopic scale while in their relevant biological contexts. Nonetheless, despite the imaging toolkit for biologists interested in exploring the subcellular localization and dynamics of proteins and nucleic acids has expanded exponentially in the last decades, the means to visualize and track lipids in cells did not develop to the same extent until recently. Here we described some basic fluorescence-based techniques that can be used in standard cell biology laboratories to visualize subcellular pools of specific lipids and to evaluate their regional metabolism. Specifically, here we focus on the imaging-based analysis of phosphoinositide and sphingolipid metabolism at the Golgi complex. Key words Immunofluorescence, Sphingolipids, Phosphatidylinositol-4-Phosphate, Diacylglycerol, Golgi complex

1

Introduction Lipids are the building blocks of biological membranes and in eukaryotic organisms they are differently distributed among endomembranes of the pre- and post-Golgi compartments. Indeed, extensive lipid remodeling occurs at the Golgi membranes where sphingolipids and sterols are delivered and metabolized by molecular machines that rely on the local enrichment of Phosphatidylinositol-4-Phosphate (PtdIns(4)P) for their functioning. Recently was reported that PtdIns(4)P turnover at the Golgi complex is regulated by the sphingolipid metabolic flux, thus revealing a two-way relationship between sphingolipid and phosphoinositide metabolism at the Golgi complex. In the light of this and of similar evidence on the crosstalk among co-compartmentalized lipid metabolisms, a need is emerging for a dynamic and multiplexed lipid assessment strategy at specific subcellular locations. Nevertheless, while achieving a space-resolved quantitation of lipid metabolisms is the far-reaching aim of laboratories specialized in this specific task, the toolkit of probes and methods available to non-specialists

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_5, © Springer Science+Business Media, LLC, part of Springer Nature 2019

47

48

Serena Capasso and Giovanni D’Angelo

for the indirect imaging of selected lipids at specific cell compartments is constantly expanding. Here, taking advantage on our experience in visualizing sphingolipids and phosphoinositide metabolism at the Golgi complex, we introduce the readers to these basic techniques and reagents. 1.1 Sphingolipid Metabolism at the Golgi Complex

The biosynthesis of sphingolipids starts with the production of sphinganine from serine and palmitoyl-CoA operated by SerinePalmitoyl Transferase (SPTLC) [1], followed by the conversion of sphinganine to dihydroceramide operated by ceramide synthases 1–6 (CERS1–6) [2] and finally to ceramide (Cer) thanks to the action of dihydroceramide desaturase (DES) [3]. SPTLC, CERSs, and DES are all Endoplasmic Reticulum (ER)-localized enzymes and, as a consequence, de novo Cer production is completed within ER membranes [4]. Once produced in the ER, Cer is transported to the Golgi for its conversion into complex sphingolipids [5]. Specifically, Cer can be transported to the cis Golgi and be converted to Glucosylceramide (GlcCer) by the enzyme GlcCer synthase (GCS) or transferred non-vesicularly to the Trans-Golgi network (TGN) by a specific Cer-Transfer protein (CERT) and converted to Sphingomyelin (SM) [1] in a reaction operated by the enzymes SM synthases 1–2 (SMS1–2). Similarly, GlcCer produced at the cis-Golgi can be transported progressively through the Golgi cisternae and converted in ganglio-series glycosphingolipids or shunted to the TGN by the GlcCer transfer protein FAPP2 for the production of globo series GSLs [6, 7]. Importantly both CERT and FAPP2 rely on their binding to the Golgi pool of the phosphoinositide PtdIns(4)P for their dynamic association to the Golgi complex [5]. As a consequence, PtdIns(4)P is required for Cer and GlcCer non-vesicular transport to the TGN and for the efficient production of SM and globo series GSLs [5].

1.2 Phosphoinositide Metabolism at the Golgi-ER Interface

Phosphoinositides are derivatives of phosphatidyl-inositol (PtdIns) where hydroxyl groups in positions 3, 4, and 5 of the inositol ring are phosphorylated [8]. All the possible phospho-combinations in these three locations are found in mammals resulting in three different PtdIns mono phosphates (i.e., PtdIns(3)P; PtdIns(4)P, and PtdIns(5)P), three PtdIns bis phosphates (i.e., PtdIns(3,4)P2; PtdIns(3,5)P2, and PtdIns(4,5)P2), and one PtdIns tris phosphate (i.e., PtdIns(3,4,5)P3). The combined action of specific PtdIns kinases and phosphatases and their subcellular distribution dictate the local concentration of the different phosphoinositides [9] that, as a consequence, are unevenly distributed among cell membranes. The predominant phosphoinositide found at the Golgi complex is the PtdIns(4)P that is locally produced by two PtdIns-4-kinases (i.e., PI4KIIIβ and PI4KIIα) [9]. The Golgi pool of PtdIns(4)P is turned over mainly by dephosphorylation operated by the PtdIns (4)P-phosphatase Sac1 [9]. Interestingly, while PtdIns(4)P is

Lipid Imaging at the Golgi Complex

49

produced at the Golgi complex, Sac1 is an integral ER membranelocated enzyme, thus implying that for Sac1-mediated dephosphorylation PtdIns(4)P should be made available to Sac1 active site [5]. Two models have been proposed, where ER-localized Sac1 acts on Golgi PtdIns(4)P in “trans” at sites of close apposition of the two organelles (i.e., ER-Golgi membrane contact sites), or in “cis” following either Sac1 relocation to the Golgi or PtdIns(4)P transport to the ER. This last model was proven to be valid in a series of reports culminating with the study by Antonny and coworkers where the lipid transfer protein Oxysterol-Binding-Protein1 (OSBP1) was shown to counter-transfer cholesterol and PtdIns(4) P between the TGN to ER [10] and to present PtdIns(4)P to Sac1 for its dephosphorylation. Along similar lines, recent reports have shown that PtdIns produced in the ER is transported to Golgi by lipid transfer proteins of the PITP family [11] for its conversion into PtdIns(4)P, thus closing a metabolic cycle at the ER-Golgi interface [10]. 1.3 The Sphingolipid/ Phosphoinositide Metabolic Crosstalk at the Golgi Complex

Sphingolipid and phosphoinositide metabolisms at the Golgi complex are intimately interconnected. On one hand this is due to the fact that CERT and FAPP2 are recruited to the TGN by the interaction of their Pleckstrin Homology (PH) domains with PtdIns(4)P [12], which implies that efficient delivery of GlcCer and Cer to the TGN requires adequate steady-state PtdIns(4)P levels to be present at the Golgi [13]. According to this notion, both interruption of PtdIns(4)P production and increased PtdIns (4)P disposal impact SM and GSL synthesis [5]. On the other hand, it has been recently reported that sustained SM production at the TGN triggers a signaling cascade leading to OSBP1 phosphorylation and increased PtdIns(4)P turnover, thus establishing a negative feedback loop involved in the homeostatic control of sphingolipid production at the Golgi complex [5].

1.4 The Basic Toolkit for Lipid Imaging at the Golgi Complex

Different imaging strategies can be used to study the regulation of SLs and phosphoinositide metabolism/trafficking at the Golgi.

1.4.1 Imaging PtdIns(4)P at the Golgi

Mammalian cells have multiple pools of PtdIns(4)P with enrichments being reported at the TGN, plasma membrane (PM), and endo/lysosomal compartments [14]. A specific antibody for the immune-detection of PtdIns(4)P was characterized by Hammond and coworkers [15]. This antibody is able to recognize either the PM or the TGN PtdIns(4)P pool, depending on conditions used for the staining. A specific Golgi staining protocol (Fig. 1) was used to analyze PtdIns(4)P levels at the Golgi complex. It is described in detail below along with some caveats about the reproducibility linked to the procedure.

50

Serena Capasso and Giovanni D’Angelo

Fig. 1 Immunofluorescence images of HeLa cells fixed and stained for nuclei (DAPI), cis-Golgi marker (G95) and with a specific anti-PtdIns(4)P antibody as detailed in the Subheading 3.1. Cells are fixed and examined under confocal laser microscope Zeiss LSM 700. Optical confocal sections were taken at 1 Airy unit. Scale bar: 10 μm

An alternative method to reveal specific PtdIns(4)P pools at different cellular compartments (including the one at the Golgi) is to use fluorescently tagged PtdIns(4)P biosensors [16]. Several protein domains are indeed recruited at specific subcellular membranes by their virtue to bind PtdIns(4)P [17]. These include the PH domains of FAPP1, FAPP2, CERT, OSBP1 for the TGN PtdIns(4)P pool [16] and those of ORP5 and ORP8 for the PM PtdIns(4)P pool (Fig. 2). Along similar lines a biosensor for multiple PtdIns(4)P pools has been obtained by the use of the PtdIns(4) P-binding domain (P4M) of the protein SidM from the bacterial pathogen Legionella pneumophila [14]. In particular, when transfected in HeLa cells fluorescently tagged P4M-SidM, it reveals PtdIns(4)P at the Golgi, PM, and endo/lysosomal compartments [14] . Here is provided basic protocols for the transfection and detection of these different biosensors in living and fixed cells by fluorescence microscopy. 1.4.2 Revealing the Sphingolipid Metabolic Flux at the Golgi Complex

Cer synthesized in the ER is converted to SM at the TGN in a reaction where a phosphocholine moiety is transferred from a phosphatidylcholine molecule to Cer to yield SM and diacylglycerol (DAG) [18]. The synthesis of DAG can serve as a proxy for the evaluation of SM synthetic rate at the TGN and can be visualized by specific fluorescence- or FRET-based biosensors. Here we describe the use of a biosensor based on the C1α domain of PKD (PKD-C1α), which is recruited to TGN membranes due to its binding to DAG [18], to evaluate DAG at the Golgi. Importantly, the very nature of the biosensors described in this chapter allows their combined use for the simultaneous detection of different metabolic reactions or subcellular lipid pools. Here specifically we show the simultaneous detection of DAG and PtdIns(4)P at the Golgi by the combined use of PKD-C1α and of an anti- PtdIns(4)P antibody (Fig. 3).

Lipid Imaging at the Golgi Complex

51

Fig. 2 HeLa cells transfected with GFP-tagged ORP5-PH, ORP8L-PH, FYVE, or PLC-delta-PH. Fixed and examined under confocal laser microscope Zeiss LSM 700. Optical confocal sections were taken at 1 Airy unit. Scale bar: 10 μm

2

Materials

2.1 PtdIns(4)P Staining

The staining of the Golgi pool of PtdIns(4)P was performed as described in reference [15]. 1. Primary purified Anti-PtdIns(4)P IgM (Echelon Bioscience). 2. Secondary anti-mouse donkey IgM conjugated with Alexa Fluor 488, Alexa Fluor 568, or Alexa Fluor-633 (Thermo Fisher Scientific). 3. Phosphate Buffered Saline (PBS, 1): 1.5 mM KH2PO4, 8 mM Na2HPO4, 2.7 mM KCl, 137 mM NaCl, pH 7.4. 4. PBS/NH4Cl solution: PBS 1 containing 50 mM NH4Cl, pH 7.4.

52

Serena Capasso and Giovanni D’Angelo

Fig. 3 Cells expressing the GST-tagged DAG sensor PKD-C1α domain (C1-alpha) were treated for the indicated times with sphingolipid precursor D-C6-Ceramide (10 μM). Cells are fixed, permeabilized, and stained with DAPI (blue), anti-GST (red), and anti-PtdIns(4)P (green) antibodies. Cells are examined under confocal laser microscope Zeiss LSM 700. Optical confocal sections were taken at 1 Airy unit. The Golgi-associated C1-alpha fraction at the different time points is analyzed using ImageJ software. Scale bar: 10 μm

5. Buffer A: 20 mM PIPES-NaOH, pH 6.68, 137 mM NaCl, 2.7 mM KCl (see Note 1). 6. Cell fixation solution: 2% (v/v) paraformaldehyde in PBS 1, stored at 4  C. 7. Membrane permeabilization solution: Digitonin (Calbiochem) is dissolved in H2O to prepare a concentrated stock solution

Lipid Imaging at the Golgi Complex

53

(20 mM). The stock solution is further diluted in buffer A to obtain a working solution of 20 μM. 8. Blocking buffer 1: 5% (v/v) FBS (Fetal Bovine Serum) in buffer A. 9. Confocal laser microscope: Zeiss LSM 700 (Carl Zeiss) or similar. 2.2 PtdIns(4)PProbes and DAG Staining

1. HeLa cells (ATCC). 2. Plasmids: ORP5-PH- and ORP8L-PH-GFP constructs from Pietro De Camilli (Yale School of Medicine), FYVE-domain from Antonella De Matteis (Telethon Institute of Genetics and Medicine) and PH-PLCδ GFP construct from R. Polishchuk (Telethon Institute of Genetics and Medicine). 3. Primary anti-GST antibody (Sigma-Aldrich). 4. Secondary antibodies IgG donkey anti-mouse or anti-rabbit Alexa-Fluor-568, 488, and 633 conjugated (Thermo Fisher Scientific). 5. TransIT-LT1 Transfection Reagent. 6. Phosphate Buffered Saline (PBS, 1): see Subheading 2.1. 7. Cell fixation solution: Dilute 4% paraformaldehyde (v/v) (final concentration) in PBS 1 and store at 4  C. 8. Blocking buffer 2: 0.05% saponin, 0.5% BSA, 50 mM NH4Cl in PBS 1. 9. Mowiol solution: Dissolve 20 mg of mowiol in 80 mL of PBS 1. The solution is stirred overnight and centrifuged for 30 min at 12,000  g. 10. N-C6:0-D-erythro-sphingosine (D-C6-ceramide) (Matreya) is dissolved in ethanol to prepare a concentrated stock solution (10 mM). The stock solution is further diluted in medium to obtain a working solution of 10 μM. 11. Confocal laser microscope: Zeiss LSM 700 (Carl Zeiss) or similar.

3

Methods

3.1 PtdIns(4)P Staining

1. HeLa cells are grown on 24-mm coverslips under controlled atmosphere in the presence of 5% CO2 at 37  C. At 80% confluence, cells are washed three times with 500 μL of PBS/NH4Cl solution for each well at room temperature (see Note 2). 2. Cells are fixed with 500 μL of 2% paraformaldehyde solution for 15 min at room temperature.

54

Serena Capasso and Giovanni D’Angelo

3. Cells are permeabilized for 7 min with 500 μL of 20 μM digitonin in buffer A at room temperature (see Note 3). 4. HeLa cells are blocked with 500 μL of blocking buffer 1 for 45 min at room temperature. 5. Anti-PtdIns(4)P monoclonal mouse antibody is diluted (1/400) in 500 μL of blocking solution 1 for 1 h at room temperature. 6. Cells are washed three times with 500 μL of buffer A at room temperature. 7. Anti-mouse IgM secondary antibody is diluted (1/1000) in 500 μL of blocking buffer 1 for 45 min at room temperature. 8. Cells are postfixed for 5 min with 500 μL of 2% paraformaldehyde solution and washed with PBS/NH4Cl solution at room temperature. 9. Coverslip is washed once with bi-distilled water and mounted on glass microscope slides with Mowiol solution at room temperature. 10. Immunofluorescence samples were examined under confocal laser microscope. Optical confocal sections were taken at 1 Airy unit. Images were analyzed using ImageJ software. 3.2 PtdIns(4)PProbes Staining

1. HeLa cells are grown on 24-mm coverslips under controlled atmosphere in the presence of 5% CO2 at 37  C. The day after, cells are transfected with PtdIns(4)P-fluorescent probes for 24 h to recognize PtdIns(4)P in the cells. The transfection reaction is performed using TransIT-LT1 Transfection Reagent according to the manufacturer protocol. 2. After 24 h, cells are fixed with 500 μL of 2% paraformaldehyde solution for 15 min and washed three times with 500 μL of PBS 1 at room temperature. 3. Coverslip is washed once with bi-distilled water and mounted on glass microscope slides with Mowiol solution. 4. Immunofluorescence samples were examined under confocal laser microscope. Optical confocal sections were taken at 1 Airy unit. Images were analyzed using ImageJ software.

3.3

DAG Staining

1. HeLa cells are grown on 24-mm coverslips under controlled atmosphere in the presence of 5% CO2 at 37  C. The day after, cells are transfected with the GST-tagged PKD-C1α domain to recognize DAG synthesis at the Golgi complex. The transfection reaction is performed using TransIT-LT1 Transfection Reagent. 2. After 24 h of transfection, cells are treated with 500 μL of cell medium supplemented with 0.5 μL of D-C6-ceramide (stock solution 10 mM) at final concentration of 10 μM and

Lipid Imaging at the Golgi Complex

55

incubated under controlled atmosphere in the presence of 5% CO2 at 37  C for indicated time (Fig. 3). 3. Cells are fixed with 500 μL of 4% paraformaldehyde for 15 min and washed three times with 500 μL of PBS 1 at room temperature. 4. Cells are incubated with 500 μL of blocking buffer 2 for 20 min at room temperature. 5. Coverslip with cells is incubated with the primary anti-GST antibody (1/500) for 1 h. 6. Cells are washed three times with 500 μL of PBS and incubated with a secondary antibody conjugated with Alexa-Fluor-568, 488, or 633 diluted at 1/1000 in 500 μL of blocking buffer 2 for 45 min. 7. Coverslip is washed once with bi-distilled water and mounted on glass microscope slides with Mowiol. 8. Immunofluorescence samples were examined under confocal laser microscope. Optical confocal sections were taken at 1 Airy unit. Images were analyzed using ImageJ software.

4

Notes 1. The pH of buffer A used for PtdIns(4)P staining is 6.68 and it is characterized by a white color. Despite that, during the pH adjustment procedure with 5 M NaOH, when the final pH 6.68 is reached, the solution becomes clear. 2. Cells confluence is an important and critical step in PtdIns(4)P staining. As a consequence, it is recommended to plate a number of cells at 80–85% of confluence for PtdIns(4)P staining. 3. Based on experimental evidences, it is suggested to use digitonin action for 7 min when cells are at 80–85% confluent. However, the incubation time may vary depending on cells confluency.

References 1. Hanada K, Kumagai K, Yasuda S et al (2003) Molecular machinery for non-vesicular trafficking of ceramide. Nature 426:803–809 2. Hannun YA, Obeid LM (2008) Principles of bioactive lipid signalling: lessons from sphingolipids. Nat Rev Mol Cell Biol 9:139–150 3. Hu W, Ross J, Geng T et al (2011) Differential regulation of dihydroceramide desaturase by palmitate versus monounsaturated fatty acids: implications for insulin resistance. J Biol Chem 286:16596–16605

4. Breslow DK, Weissman JS (2010) Membranes in balance: mechanisms of sphingolipid homeostasis. Mol Cell 40:267–279 5. Capasso S, Sticco L, Rizzo R et al (2017) Sphingolipid metabolic flow controls phosphoinositide turnover at the trans-Golgi network. EMBO J 36:1736–1754 6. D’Angelo G, Polishchuk E, Di Tullio G et al (2007) Glycosphingolipid synthesis requires FAPP2 transfer of glucosylceramide. Nature 449:62–67

56

Serena Capasso and Giovanni D’Angelo

7. D’Angelo G, Uemura T, Chuang CC et al (2013) Vesicular and non-vesicular transport feed distinct glycosylation pathways in the Golgi. Nature 501:116–120 8. Tolias KF, Cantley LC (1999) Pathways for phosphoinositide synthesis. Chem Phys Lipids 98:69–77 9. Ebrahimzadeh Z, Mukherjee A, Richard D (2018) A map of the subcellular distribution of phosphoinositides in the erythrocytic cycle of the malaria parasite Plasmodium falciparum. Int J Parasitol 48:13–25 10. Mesmin B, Bigay J, Moser von Filseck J et al (2013) A four-step cycle driven by PI(4)P hydrolysis directs sterol/PI(4)P exchange by the ER-Golgi tether OSBP. Cell 155:830–843 11. Hsuan J, Cockcroft S (2001) The PITP family of phosphatidylinositol transfer proteins. Genome Biol 2: Reviews3011.1–Reviews3011.8 12. Lemmon MA (2010) Chapter 136—Pleckstrin homology (PH) domains. In: Bradshaw RA, Dennis EA (eds) Handbook of cell signaling, 2nd edn. Academic Press, San Diego, pp 1093–1101

13. Cheong FY, Sharma V, Blagoveshchenskaya A et al (2010) Spatial regulation of Golgi phosphatidylinositol-4-phosphate is required for enzyme localization and glycosylation fidelity. Traffic 11:1180–1190 14. Hammond GR, Machner MP, Balla T (2014) A novel probe for phosphatidylinositol 4-phosphate reveals multiple pools beyond the Golgi. J Cell Biol 205:113–126 15. Hammond GR, Schiavo G, Irvine RF (2009) Immunocytochemical techniques reveal multiple, distinct cellular pools of PtdIns4P and PtdIns(4,5)P(2). Biochem J 422:23–35 16. Wuttke A, Idevall-Hagren O, Tengholm A (2010) Imaging phosphoinositide dynamics in living cells. Methods Mol Biol 645:219–235 17. Levine TP, Munro S (1998) The pleckstrin homology domain of oxysterol-binding protein recognises a determinant specific to Golgi membranes. Curr Biol 8:729–739 18. Baron CL, Malhotra V (2002) Role of diacylglycerol in PKD recruitment to the TGN and protein transport to the plasma membrane. Science 295:325–328

Chapter 6 Advanced In Vitro Assay System to Measure Phosphatidylserine and Phosphatidylethanolamine Transport at ER/Mitochondria Interface Yasushi Tamura, Rieko Kojima, and Toshiya Endo Abstract A number of previous studies have shown that phospholipid molecules come and go between the endoplasmic reticulum (ER) and mitochondrial membranes while the molecular basis of non-vesicular phospholipid transport is still not understood well. In this chapter, we describe an optimized method that uses membrane fractions isolated from yeast cells to directly analyze phospholipid transport between the ER and mitochondria. With this assay, we are able to assess not only the ER-to-mitochondria but also mitochondria-to-ER transports at the same time. We believe that this assay system can accelerate the research on inter-organelle phospholipid trafficking. Key words Mitochondria, Endoplasmic reticulum, Phospholipid transport, In vitro reconstitution, Yeast

1

Introduction Phospholipids are synthesized from the simple phospholipid, phosphatidic acid (PA) through several modification reactions. To generate a complete set of phospholipids, the precursor phospholipids have to shuttle between the ER and mitochondrial inner membranes, where most phospholipids are synthesized [1–3]. In yeast, CDP-diacylglycerol (DAG) synthase Cds1 localized in the ER first couples PA with CTP to generate CDP-DAG, which is a highenergy intermediate phospholipid [4]. Then, Cho1 and Pis1, which are phosphatidylserine (PS) and phosphatidylinositol (PI) synthases localized in the ER, substitute the CDP moiety of CDP-DAG with serine and inositol to produce PS and PI, respectively [5, 6]. Importantly, PS is transported from the ER to mitochondria and converted to phosphatidylethanolamine (PE) by PS decarboxylase Psd1, which is localized in the mitochondrial inner membrane. PE, synthesized in mitochondria, is then transported

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_6, © Springer Science+Business Media, LLC, part of Springer Nature 2019

57

58

Yasushi Tamura et al.

Fig. 1 Phospholipid biosynthetic pathways in the ER and mitochondria. Broken lines indicate phospholipid movements. OM Mitochondrial outer membrane, IM mitochondrial inner membrane, IMS Mitochondrial intermembrane space, PA Phosphatidic acid, CDP-DAG CDP-diacylglycerol, PS phosphatidylserine, PE phosphatidylethanolamine, PC phosphatidylcholine, PI phosphatidylinositol, PGP phosphatidylglycerolphosphate, PG phosphatidylglycerol, CL cardiolipin

back to the ER and methylated by two ER-resident methyltransferases Cho2 and Opi3, generating phosphatidylcholine (PC) [7]. In addition, to synthesize the mitochondrial signature phospholipid, cardiolipin (CL), PA has to move to the inner leaflet of the mitochondrial inner membrane where mitochondrial CDP-DAG synthase Tam41 resides [8]. The phospholipid synthetic pathways in yeast are summarized in Fig. 1. Although it is clear that phospholipid transport reactions between the ER and mitochondria are keys to the phospholipid biosynthesis, the question of how phospholipids travel between these organelles has not been studied extensively, probably due to the lack of a practically useful in vitro assay system to analyze these processes. To overcome this situation, we recently developed a useful in vitro assay system that can assess PS transport from the ER to mitochondria as well as PE transport from mitochondria to the ER [9]. With this system, we were able to show that the ERMES (ER-Mitochondria Encounter Structure) complex, which directly tethers the ER and mitochondrial outer membranes [10], facilitates phospholipid transfer from the ER to mitochondria [9]. This in vitro assay complements the standard in vitro lipid transfer assays between the membranes, which utilizes purified lipid transfer proteins and liposomes [11]. We thus believe that the in vitro assay systems are powerful tools to study how phospholipids travel between these organelles.

In Vitro Phospholipid Transfer Assay Between the ER and Mitochondria

2

59

Materials

2.1 Yeast Growth Media

1. YPD medium: 1% (w/v) yeast extract, 2% (w/v) polypeptone, and 2% (w/v) glucose (see Note 1). 2. SCD medium: 0.67% (w/v) yeast nitrogen base without amino acids, 0.5% (w/v) casamino acid, 2% (w/v) glucose, 20 μg/mL each of adenine, uracil, L-tryptophan and L-histidine and 30 μg/mL each of L-leucine and L-lysine (see Note 2).

2.2 Buffers and Reagents for Membrane Preparation

1. Alkaline buffer: 0.1 M Tris–HCl, pH 9.5, 10 mM dithiothreitol (DTT). Combine 10 mL of 1 M Tris–HCl, pH 9.5, stock buffer, 154 mg of DTT, and dH2O to bring the total volume up to 100 mL. Need to be freshly prepared. 2. Spheroplast buffer: 20 mM Tris–HCl, pH 7.5, 1.2 M sorbitol. Mix 500 mL of 2.4 M sorbitol (stock solution, stored at 4
C), 20 mL of 1 M Tris–HCl, pH 7.5 (stock buffer, stored at room temperature), and dH2O to bring the total volume up to 1 L. Store at 4
C. 3. Breaking buffer: 20 mM Tris–HCl, pH 7.5, 0.6 M mannitol, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride (PMSF). Mix 500 mL of 1.2 M mannitol (stock solution, stored at 4
C), 20 mL of 1 M Tris–HCl, pH 7.5 (stock buffer), 2 mL of 0.5 M EDTA (stock solution, stored at room temperature), and dH2O to bring the total volume up to 1 L. Store at 4
C. Dilute 1 M PMSF stock 10 times in isopropanol, and freshly add to ice-cold breaking buffer with stirring before use. 4. SEM buffer: 250 mM sucrose, 10 mM MOPS-KOH, pH 7.2, 1 mM EDTA. Mix 250 mL of 2 M sucrose (stock solution, stored at 4
C), 10 mL of 1 M MOPS-KOH, pH 7.2 (stock buffer), 2 mL of 0.5 M EDTA (stock solution), and dH2O to bring the total volume up to 1 L. Store at 4
C. 5. SDS 0.6% (w/v) in dH2O. 6. Zymolyase 20 T.

2.3 Stock Solutions for In Vitro Phospholipid Transport Assay

1. 1 M Tris–HCl, pH 7.5 buffer. 2. 2 M sucrose stock solution. 3. 1 M KCl stock solution. 4. 100 mM CTP (cytidine triphosphate) solution. 5. 1 M MgCl2 stock solution. 6. 100 mM MnCl2 stock solution. 7. 32 mM S-adenosylmethionine (AdoMet). 8. 100 μCi/mL [14C(U)]-L-serine or 50 μCi/mL serine.

L-[3-

14

C]-

60

Yasushi Tamura et al.

9. Chloroform/methanol (2:1, v/v) mix. 10. 0.1 N HCl, 0.1 M KCl solution. 2.4 Thin-Layer Chromatography (TLC) and Radioimaging

1. 1.8% (w/v) boric acid in ethanol. 2. Sample concentrator with heat blocks and multiple gas delivery needles. 3. TLC developing tank. 4. Pre-coated TLC-plates SILGUR-25-C/UV25 (20  20 cm). 5. Developing solvent: Chloroform/ethanol/H2O/triethylamine (30:30:7:35, v/v/v/v). 6. Incubator oven. 7. Phosphorimager screen and cassettes. 8. Phosphor Screen Scanner.

3

Methods

3.1 Isolation of Membrane Fractions Containing Both ER and Mitochondrial Membranes

1. Cultivate yeast cells in YPD medium overnight to stationary phase (see Note 3). 2. Inoculate 500 μL of the overnight preculture to 1 L of SCD medium and cultivate in an incubator at 30
C for 15 h with vigorous shaking at ~180 rpm (see Note 4). The typical optical density at wavelength of 600 nm (OD600) after the cultivation is ~1.5 (see Note 5). 3. Transfer the culture to a centrifuge bottle and centrifuge it at 5000  g for 10 min to collect the cells. 4. Measure the wet weight of the cells. 5. Resuspend the cells in 40 mL of spheroplast buffer, then transfer them to a 50 mL tube, and centrifuge at 2000  g for 5 min to pellet the cells. Discard the supernatant. 6. Resuspend and incubate the cells in 40 mL of alkaline buffer in a 30
C water bath for 15 min under mild agitation. 7. Centrifuge the suspension, discard the supernatant, and wash the cells with 40 mL of spheroplast buffer as in step 5. 8. Resuspend the cells in 40 mL of spheroplast buffer with Zymolyase 20 T (5 mg for 1 g yeast wet weight) to digest cell wall. Incubate for 30 min in a 30
C water bath under mild agitation (see Note 6). 9. Centrifuge the suspension, discard the supernatant, and wash the cells with ice-cold breaking buffer and centrifuge at 2000  g for 5 min. From this step, keep the sample below 4
C.

In Vitro Phospholipid Transfer Assay Between the ER and Mitochondria

61

10. Resuspend the spheroplast cells in 40 mL of ice-cold breaking buffer and homogenize 20 times in a dounce tissue grinder with the tight pestle on ice. 11. Centrifuge the suspension at 2000  g for 5 min to precipitate unbroken cells and cell debris including nuclear membranes. 12. Transfer the supernatant to a new 50 mL tube carefully not to transfer the pellet together. 13. Centrifuge the supernatant at 12,000  g for 10 min to sediment heavy membrane fractions (HMFs) containing the ER and mitochondria. 14. Resuspend the pellet in 40 mL of SEM buffer and centrifuge as in step 13. 15. Resuspend the pellet in ~1 mL of SEM buffer. 16. Add 10 μL of the membrane suspension to 990 μL of 0.6% SDS solution and boil for 5 min. 17. Measure Abs280 of the sample. OD280 ¼ 0.21 is assumed to correspond to a concentration of protein equal to 10 mg/mL. Other methods for quantification of protein concentration can be also used. 18. Make aliquots of the membrane suspension and store them at 80
C after the snap-freezing in liquid nitrogen. We routinely set the protein concentration of the suspension to 10 mg/mL and make 200 μL of aliquots (2 mg of protein). 3.2 Phospholipid Transport Assay

1. Mix the stock solutions as described in Table 1. 2. Thaw 2 mg of frozen membrane suspension quickly in hand or water bath at 30
C and then transfer onto ice. 3. Centrifuge the membrane suspension at 12,000  g for 5 min at 4
C. 4. Remove the supernatant and resuspend well the pellet with 980 μL of assay buffer by pipetting with a yellow 200 μL tip. 5. Make 98 μL of aliquots in a 2 mL safe-lock tube (see Note 7). 6. Preincubate the aliquot in a 30
C water bath for 2 min. 7. Add 2 μL of synthesis.

14

C-serine solution to the aliquot to start PS

8. Incubate for a certain period of time (for example, 10, 20, 30, and 40 min as shown in Fig. 2) in a 30
C water bath (see Note 8). 9. Stop each reaction by adding 900 μL of chloroform/methanol (2:1, v/v) to the sample, mix it with short vortex, and place it on ice. 10. After stopping all the reactions, vortex all the samples again thoroughly for 15 min at room temperature.

62

Yasushi Tamura et al.

Table 1 Preparation of assay buffer (for 10 reactions) 11 Reaction volume (μL) 1 M Tris–HCl, pH 7.5 buffer (final concentration 20 mM) 2 M sucrose solution (final concentration 300 mM)

22 165

1 M KCl solution (final concentration 40 mM)

44

100 mM CTP solution (final concentration 2 mM)

22

32 mM AdoMet solution (final concentration 1 mM)

34.4

5 mM MnCl2 solution (final concentration 0.1 mM)

22

100 mM MgCl2 solution (final concentration 2 mM)

22

dH2O

746.6

Total

1078

11. Spin down. 12. Add 200 μL of 0.1 N HCl, 0.1 M KCl solution. 13. Vortex for 5 min thoroughly at room temperature. 14. Centrifuge at 1000  g for 5 min at room temperature. 15. Remove aqueous (upper) phase. 16. Dry the organic solvent to concentrate lipids under a stream of nitrogen gas (see Note 9). 17. Dissolve in 20 μL of chloroform. The sample can be stored at 30
C for a week. 3.3

TLC Analysis

3.3.1 Preparation of TLC Plates

1. Soak a TLC plate in 1.8% boric acid solution until the plate is completely wetted. 2. Dry the TLC plate in a fume hood for 5 min. 3. Bake the TLC plate in a 100
C oven for 15 min.

3.3.2 Preparation of TLC Tank

1. Stand two pieces of filter paper, which is cut to fit to the size of TLC tank. 2. Pour the developing solvent to a TLC tank and leave to stand for over 2 h until the chamber is equilibrated with the solvent.

3.3.3 TLC Separation and Autoradiography

1. Apply a whole amount (20 μL) of the sample prepared in step 17 of section Subheading 3.2 to a TLC plate (see Note 10). 2. Place the TLC plate in the TLC tank after drying the spotted samples. 3. Develop the plate for approximately 2 h.

In Vitro Phospholipid Transfer Assay Between the ER and Mitochondria

63

Fig. 2 (a) A representative TLC result of phospholipid transfer assay in vitro. Yeast membrane fractions containing the ER and mitochondria were incubated with 14C-serine for the indicated time periods. Phospholipids were extracted and analyzed by TLC and radioimaging. (b) Diagram showing the principle of the in vitro phospholipid transfer assay. 14C-labeled PS synthesized in the ER is transported to mitochondria and is converted to PE. PE is then transported back to the ER and methylated to be PC. Broken lines indicate phospholipid movements. (c) Representative quantifications of total and relative amounts of PS, PE, and PDME +PC to total phospholipids are shown

4. Take the TLC plate out from the TLC tank and dry it completely in a fume hood. To quickly dry the TLC plate, a hair drier may be used. 5. Cover the TLC plate with plastic wrap and expose it to a phosphorimager screen in a cassette overnight. 6. Scan the phosphorimager screen using a Phosphor Screen Scanner. 3.3.4 Quantification

1. As shown in Fig. 2, in vitro synthesized PS, PE, PC, and phosphatidyldimethylethanolamine (PDME) should be observed. 2. Quantify intensities of these phospholipid signals using an image processing software (for example, ImageJ, see Note 11). 3. Calculate the relative amounts of PS, PE, and PDME+PC to total phospholipids or PE (see Fig. 2 and Note 12).

64

4

Yasushi Tamura et al.

Notes 1. Dissolve 10 g of yeast extract and 20 g of peptone in 900 mL of dH2O (distilled water), and 20 g of glucose in 100 mL of dH2O, separately. Combine them after autoclave. 2. Dissolve 6.7 g of yeast nitrogen base without amino acids and 5 g of casamino acid in 900 mL of dH2O, and 20 g of glucose in 100 mL of dH2O, separately. Combine them after autoclave. Then, add 10 mL of 100  complete supplement (2 mg/mL each of adenine, uracil, L-tryptophan and L-histidine and 3 mg/mL each of L-leucine and L-lysine). 3. Although we routinely use YPD for preculture media, this is not mandatory. Other media such as SCD can be used and the results should not be affected. The typical OD600 of the saturated preculture is over 8. 4. Synthetic and fermentable media such as SCD have to be used for this assay. We confirmed that membrane fractions prepared from yeast cells cultivated in rich media such as YPD have less activities to produce phospholipids in vitro due to decreased expression levels of phospholipid synthetic enzymes. Moreover, we noticed that the ER and mitochondrial membranes are not co-fractionated in the same 12,000  g pellet fractions when yeast cells are grown in non-fermentable media containing glycerol or lactate as a carbon source (Fig. 3). Recently, Psd1 was reported to show dual localization in mitochondria and the ER [12]. However, our 12,000  g pellet fraction for the in vitro experiments (prepared from SCD-grown cells) did not show a clear ER-localized (N-glycosylated) form of Psd1 (Fig. 3). 5. The final optical density is important to obtain reproducible results. In other words, the growth phase or cell density may be important factor to determine the efficiency of the phospholipid transport between the ER and mitochondria. Although we have not carefully tested the effects of the final optical density on the phospholipid transport in vitro, the phospholipid transport tends to be more efficient when we use yeast culture with lower optical density for membrane preparations (unpublished results). Therefore, we recommend using membrane fractions prepared from cell cultures with the same optical densities when one wants to compare the phospholipid transfer rates quantitatively. 6. Dissolve Zymolyase 20T in spheroplast buffer freshly before use. Zymolyase is an enzyme that digests yeast cell wall. This step is essential for efficient disruption of yeast cells by homogenization.

In Vitro Phospholipid Transfer Assay Between the ER and Mitochondria

65

Fig. 3 (a) Yeast cells cultivated in YPLac, YPD, or SCD were subjected to subcellular fractionation. 12 k, 25 k, and 40 k represent 12,000, 25,000, and 40,000  g pellet fractions containing different yeast membranes. These membrane fractions were analyzed by SDS-PAGE, followed by immunoblotting using the indicated antibodies against phospholipid synthetic enzymes, ER (ER), and mitochondrial (Mito.) marker proteins. (b) The 12,000  g pellet fractions isolated from yeast cells cultivated in YPLac, YPD, or SCD were incubated with 14 C-serine for the indicated periods of time. Total phospholipids were extracted and analyzed by TLC and radioimaging

7. Since the samples that contain radioisotope must be vortexed thoroughly in organic solvent, it is important to keep the lid tightly closed in order for the contaminated solvent not to be leaked out. 8. To perform time-course experiments, we routinely put a sample into a 30
C water bath every 20 s intervals (total 6 samples in 2 min-preincubation), then add 2 μL of 14C-serine to the sample which has been incubated for 2 min at 30
C. 9. We use a sample concentrator with heat blocks and multiple gas delivery needles to efficiently evaporate organic solvent of the multiple samples at the same time. 10. We use a TLC plate that contains separate lanes with sample concentration zones so that we can use a regular pipet with tip for loading samples. 11. When you use [14C(U)]-L-serine as a substrate, you have to multiply signal intensities of phospholipids except PS by 1.5 because one of three radioactive carbons is removed upon decarboxylation of PS. When you use L-[3-14C]-serine, the signal intensities can be directly subjected to quantification.

66

Yasushi Tamura et al.

12. PS/total shows the PS transport rate from the ER to mitochondria while PE transport rate from mitochondria to the ER can be assessed on the basis of PE/total, (PDME+ PC)/total or (PDME+ PC)/PE. We confirmed that PS to PE and PE to PC conversions depend on phospholipid transports but not their syntheses because overexpression of phospholipid synthetic enzymes such as Psd1, Cho2, or Opi3 did not affect the conversion rates. In addition, we found that the decreased amount of PS synthesis does not interfere with the phospholipid conversion rates [9].These findings indicate that phospholipid transports between the ER and mitochondria can be assessed by the conversions of PS and PE to PE and PC, respectively. On the other hand, it was recently reported that over 20% of Psd1 is localized in the ER in addition to mitochondria when cells are cultivated in a synthetic fermentable medium [12]. That study questioned the validity of our present in vitro assay using yeast membrane fractions prepared from synthetic fermentable medium (SCD)-grown cells. However, our Western blotting did not show clear Psd1 signals that migrate slower due to its N-glycosylation (Fig. 3). We are therefore skeptical about the substantial dual localization of Psd1 and believe that the effects of ER-localized Psd1 on the present in vitro assay is only marginal, if any.

Acknowledgments This work was supported by JSPS KAKENHI (Grant Numbers 15H05595 and 17H06414 to Y.T. and 15H05705 and 22227003 to T.E.). References 1. Flis VV, Daum G (2013) Lipid transport between the endoplasmic reticulum and mitochondria. Cold Spring Harb Perspect Biol 5: a013235 2. Tamura Y, Sesaki H, Endo T (2014) Phospholipid transport via mitochondria. Traffic 15:933–945 3. Tatsuta T, Scharwey M, Langer T (2014) Mitochondrial lipid trafficking. Trends Cell Biol 24:44–52 4. Shen H, Heacock PN, Clancey CJ et al (1996) The CDS1 gene encoding CDP-diacylglycerol synthase in Saccharomyces cerevisiae is essential for cell growth. J Biol Chem 271:789–795 5. Clancey CJ, Chang SC, Dowhan W (1993) Cloning of a gene (PSD1) encoding phosphatidylserine decarboxylase from Saccharomyces

cerevisiae by complementation of an Escherichia coli mutant. J Biol Chem 268:24580–24590 6. Nikawa J, Yamashita S (1984) Molecular cloning of the gene encoding CDPdiacylglycerolinositol 3-phosphatidyl transferase in Saccharomyces cerevisiae. Eur J Biochem 143:251–256 7. Kodaki T, Yamashita S (1987) Yeast phosphatidylethanolamine methylation pathway. Cloning and characterization of two distinct methyltransferase genes. J Biol Chem 262:15428–15435 8. Tamura Y, Harada Y, Nishikawa S et al (2013) Tam41 is a CDP-diacylglycerol synthase required for cardiolipin biosynthesis in mitochondria. Cell Metab 17:709–718

In Vitro Phospholipid Transfer Assay Between the ER and Mitochondria 9. Kojima R, Endo T, Tamura Y (2016) A phospholipid transfer function of ER-mitochondria encounter structure revealed in vitro. Sci Rep 6:30777 10. Kornmann B, Currie E, Collins SR et al (2009) An ER-mitochondria tethering complex revealed by a synthetic biology screen. Science 325:477–481

67

11. Kawano S, Tamura Y, Kojima R et al (2018) Structure-function insights into direct lipid transfer between membranes by Mmm1Mdm12 of ERMES. J Cell Biol 217:959–974 12. Friedman JR, Kannan M, Toulmay A et al (2018) Lipid homeostasis is maintained by dual targeting of the mitochondrial PE biosynthesis enzyme to the ER. Dev Cell 44:261–270

Chapter 7 Measurement of Lipid Transport in Mitochondria by the MTL Complex Juliette Jouhet, Vale´rie Gros, and Morgane Michaud Abstract Membrane biogenesis requires an extensive traffic of lipids between different cell compartments. Two main pathways, the vesicular and non-vesicular pathways, are involved in such a process. Whereas the mechanisms involved in vesicular trafficking are well understood, fewer is known about non-vesicular lipid trafficking, particularly in plants. This pathway involves the direct exchange of lipids at membrane contact sites (MCSs) between organelles. In plants, an extensive traffic of the chloroplast-synthesized digalactosyldiacylglycerol (DGDG) to mitochondria occurs during phosphate starvation. This lipid exchange occurs by non-vesicular trafficking pathways at MCSs between mitochondria and plastids. By a biochemical approach, a mitochondrial lipoprotein super-complex called MTL (Mitochondrial Transmembrane Lipoprotein complex) involved in mitochondria lipid trafficking has been identified in Arabidopsis thaliana. This protocol describes the method to isolate the MTL complex and to study the implication of a component of this complex (AtMic60) in mitochondria lipid trafficking. Key words Mitochondria, Lipid transfer, MTL complex, CN-PAGE, Mass spectrometry

1

Introduction Eukaryotic cells are composed of different organelles that allow the compartmentalization of cellular functions. Each organelle is delineated by at least one membrane composed of a specific assembly of lipids and proteins. This specific composition defines the identity of each organelle and is required to ensure their proper functions. Lipids play multiple roles in the regulation of the architecture and the function of membranes. They are synthesized in different cell compartments and have to be properly distributed to other organelles. Two main lipid trafficking pathways have been described: the vesicular and the non-vesicular pathways. The vesicular pathway is involved in lipid and protein trafficking to the endomembrane system and has been well studied in different organisms [1, 2]. The non-vesicular pathway involves the exchange of lipids directly between two membranes, usually at contact sites (MCSs),

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_7, © Springer Science+Business Media, LLC, part of Springer Nature 2019

69

70

Juliette Jouhet et al.

when membranes are at a distance
30 nm [3]. This pathway requires the establishment of MCSs, by the so-called tethering proteins or complexes, followed by the transfer of the lipid from one membrane to another. Identification of proteins involved in non-vesicular lipid trafficking in cells is limited mainly by technical issues. In vivo strategies to investigate the role of a protein on lipid trafficking are mostly based on (1) the analysis of the steady state level of lipids in cells or organelles in a mutant or (2) experiments based on lipid labeling. However, the steady state of a lipid relies on several metabolic processes besides trafficking, such as synthesis, degradation, or modification. Lipid labeling experiments can be limited by the availability of labeled (fluorescent, radioactive, etc.) substrates that can be easily uptake by cells and behave like endogenous lipids. In addition, because of the presence of several redundant pathways, the effect of a protein on lipid trafficking can be masked in a mutant. Several in vitro assays using liposomes and purified proteins have also been developed to study the ability of a protein to transfer lipids [4, 5]. However, these assays are adapted for soluble proteins and some specific factors, such as lipid composition of the donor and acceptor liposomes or the presence of partners, might be required to detect and measure the transfer activity of a protein by such assays. Recently, we took advantage of the massive lipid remodeling triggered by phosphate starvation to identify proteins involved in lipid trafficking to mitochondria in Arabidopsis thaliana [6–8]. During phosphate starvation, the phospholipids present in mitochondria membranes are partially degraded to release phosphate [6]. To maintain the integrity of mitochondrial membranes, a massive transfer of the non-phosphorous galactoglycerolipid digalactosyldiacylglycerol (DGDG) occurs from chloroplast to mitochondria [6]. DGDG is synthesized in plastids and is mainly retained in this compartment under normal growth condition [9]. During Pi starvation the level of DGDG in mitochondria drastically increases from 2% to 20% [6, 8]. This transfer occurs at MCSs between chloroplasts and mitochondria, which number increases during Pi starvation [6]. Thus, this particular situation constitutes a powerful tool to study lipid trafficking to mitochondria as we can (1) turn on/off the traffic of DGDG to mitochondria by simply modifying the Pi concentration in the medium and (2) easily follow the content of this natural non-labeled lipid inside mitochondria by classical methods of lipid analysis such as mass spectrometry or thin layer chromatography. Thus, by looking for mitochondrial complexes enriched in DGDG during Pi starvation, we identified on Clear Native Polyacrylamide Gel (CN-PAGE) a super-complex called Mitochondrial Transmembrane Lipoprotein complex (MTL) [8, 10]. The MTL complex is present in cells grown in the presence or in the absence of Pi. Proteomic analyses have

Lipid Trafficking in Plant Mitochondria

71

revealed the presence in this complex of proteins located mostly in the mitochondrial membranes [8]. However, some components are also located in other cell compartments such as plastids, suggesting the presence of the MTL complex at membrane contact sites. Furthermore, this complex is dynamic and both its lipid and protein composition vary during Pi starvation. Among this complex, a protein localized in the inner membrane of mitochondria, AtMic60, has been further investigated for a putative role in lipid trafficking [8, 10]. This protocol describes the strategy used to reveal the key role of AtMic60 in mitochondrila lipid remodeling during Pi starvation [8]. We started from callus cell cultures that allow to easily obtain liquid cultured cell lines from mutant plants. After purification of mitochondria from atmic60 mutant callus, the involvement of AtMic60 on lipid trafficking is investigated by analyzing (1) the in vitro incorporation of radiolabeled DGDG inside the MTL complex and (2) the lipidome of the MTL complex compared to the lipidome of mitochondrial membranes (Fig. 1).

2

Materials Prepare all reagents and media with ultrapure water. For cell cultures, always work in sterile condition under a laminar flow hood. For lipid extraction and analysis, always use glass vessels and never plastic with organic solvents. Wash all vessels without detergent but with distilled water, then ethanol. All solvents must be for Analysis grade except hexane that must be for GC grade. Chloroform should be ethanol stabilized. Diligently follow all waste disposal regulations when disposing waste materials.

2.1 Plant Material and Culture

1. Arabidopsis thaliana Col0 plants are used as wild type (WT). A T-DNA insertion line (SALK_087650c) [11] in AtMic60 gene (AT4G39690) in Col0 background is used as an example to study a defect in lipid trafficking [8]. 2. MS +Pi medium: 4.41 g/L Murashige and Skoog (MS) basal salts, 1 MS Vitamin Solution, 1.5% (w/v) sucrose, 1.2 mg/L 2,4-dichlorophenoxyacetic acid. The pH is adjusted to 5.7 with NaOH and the medium is autoclaved. 3. MS –Pi medium: 4.41 g/L MS basal salts without Pi, 1 MS Vitamin Solution, 1.5% (w/v) sucrose, 1.2 mg/L 2,4-dichlorophenoxyacetic acid. The pH is adjusted to 5.7 with NaOH and the medium is autoclaved. 4. MS seedlings: 4.41 g/L MS with basal salts and vitamins, 0.5% (w/v) sucrose, and 0.5 g/L MES. The pH is adjusted to 5.7 with KOH. Agar is added at 0.8% (w/v) and the medium is autoclaved.

72

Juliette Jouhet et al.

atmic60 callus+Pi atmic60 callus -Pi

Col0 callus+Pi Col0 callus -Pi

1- Galactoglycerolipids labelling and transfer (UDP-[14C]-galactose) OE

Mitochondria purification (Discontinous Percoll gradient)

MGDG

DGD

UDP-Gal*

Membrane purification Mitochondria

Complexes solubilization

Isolation of MTL complex by CN-PAGE 1- Analysis of UDP-[14C]galactose incorporation into MTL complex

Fixation / Staining

Lipid extraction

2- Analysis of MTL complex lipids by mass spectrometry

2- Analysis of mitochondrial membranes lipids by mass spectrometry

Fig. 1 Overview of the methods used to analyze defects in lipid trafficking in the MTL complex of atmic60 mutant callus. The dashed boxes indicate the two strategies used: (1) study of the in vitro synthesis and transfer of radiolabeled galactoglycerolipids in the MTL complex; (2) study of the mitochondrial membranes and MTL complex lipidome. DGDG: digalactosyldiacylglycerol, DGD: DGDG synthase, MGDG: monogalactosyldiacylglycerol, OE: outer envelope of plastids, Pi, phosphate, UDP-Gal*: UDP-[14C]-galactose

5. MS callus: 4.41 g/L MS, 3% (w/v) sucrose, and 1.2 mg/L of 2,4-dichlorophenoxyacetic acid. The pH is adjusted to 5.7 with KOH. Agar is added at 0.8% (w/v) and the medium is autoclaved. 6. Ethanol 70% (v/v) in water. 7. Sterilization solution: bleach 0.4% active chlorine in water. 8. Agarose 0.2% (w/v) in water. The solution is autoclaved. 9. Sterile ultrapure water (prepared by purifying deionized water, to attain a sensitivity of 18 MΩ.cm at 25  C).

Lipid Trafficking in Plant Mitochondria

2.2 Mitochondria Purification from Callus Liquid Cultures

73

1. Grinding buffer: 0.3 M mannitol, 15 mM 3-(N-morpholino) propanesulfonic acid (MOPS), pH 8, 2 mM ethylene glycol-bis (2-aminoethylether)-N,N,N0 ,N0 -tetraacetic acid (EGTA), pH 8, 0.6% (w/v) polyvinylpyrrolidone K25 (PVP25). Just before use, 0.5% (w/v) bovine serum albumin (BSA), 1 mM dithiothreitol (DTT), 1 mM phenylmethylsulfonyl fluoride (PMSF), 5 mM aminocaproic acid, and 1 mM benzamidine are added. 2. Washing buffer: 0.3 M mannitol, 10 mM MOPS-NaOH, pH 7.4. Just before use, add 1 mM PMSF, 5 mM aminocaproic acid, and 1 mM benzamidine. 3. CX2: 0.6 M mannitol, 20 mM MOPS-NaOH, pH 7.2, 2 mM EGTA. Just before use, 0.2% (w/v) BSA is added. 4. Percoll 40: 40% (v/v) Percoll, 1 of CX2. 5. Percoll 28: 28% (v/v) Percoll, 1 of CX2. 6. Percoll 23: 23% (v/v) Percoll, 1 of CX2. 7. Mortar and pestle. 8. Sand. 9. 50 mL polycarbonate tubes.

2.3 Galactoglycerolipid Labeling and Transfer with UDP-[14C]Galactose

1. UDP-[14C]-galactose: 167 bq/nmoL.

2.4 Mitochondria Membranes Purification and Solubilization

1. Swelling buffer: 10 mM MOPS pH 7.4. Just before use, 1 mM PMSF, 5 mM aminocaproic acid, and 1 mM benzamidine are added.

10

mM

UDP-[14C]-galactose

2. Washing buffer: 0.3 M mannitol, 10 mM MOPS, pH 7.4.

2. Membrane buffer: 50 mM imidazole-HCl, pH 7, 0.5 M aminocaproic acid, 1 mM EDTA. EDTA solution is adjusted to pH 8 using NaOH. 3. DDM 1%: A stock solution 10% (w/v) of n-dodecyl β-D-maltoside (DDM) is prepared in water. Then, this solution is diluted ten times in water to obtain a DDM 1% (w/v) solution stored at 20  C. 4. Loading buffer 5: 300 mM Tris–HCl, pH 6.8, 50% (v/v) glycerol, 0.5% (w/v) bromophenol blue in water. Store at 20  C.

2.5

CN-PAGE

Prepare the gel mix just before use. 1. Concentration gel: 3% acrylamide/bis-acrylamide 37.5/1 (stock solution at 30%, Bio-Rad), 0.125 M Tris–HCl, pH 6.8, 0.01% (w/v) ammonium persulfate (APS), 0.04%

74

Juliette Jouhet et al.

(v/v) N,N,N0 ,N0 -tetramethylethane-1,2-diamine (TEMED). Add the TEMED just before pouring the gel. 2. Separation gel 3.5%: 3.5% acrylamide/bis-acrylamide 37.5/1 (stock solution at 30%, Bio-Rad), 0.125 M Tris–HCl, pH 8.8, 0.01% (w/v) APS, 0.04% (v/v) TEMED. Add the TEMED just before pouring the gel. 3. Separation gel 12%: 12% acrylamide/bis-acrylamide 37.5/1 (stock solution at 30%, Bio-Rad), 0.125 M Tris–HCl, pH 8.8, 0.01% (w/v) APS, 0.04% (v/v) TEMED. Add the TEMED just before pouring the gel. 4. Migration buffer (5): 0.25 M Tris, 1.92 M glycine. Adjust the pH to 8.3 with HCl. Store the solution at 4  C. Dilute five times this solution in MilliQ water just before use to obtain the migration buffer. 5. Gradient mixer, 15 mL volume. 6. Peristaltic pump. 2.6 Fixation and Staining of CN-PAGE

1. Fixation solution: 40% (v/v) ethanol, 10% (v/v) acetic acid. 2. Staining solution: 34% (v/v) methanol, 17% (w/v) ammonium sulfate, 0.5% (v/v) acetic acid, 0.1% (w/v) Coomassie blue G250. 3. Ultrapure water (prepared by purifying deionized water, to attain a sensitivity of 18 MΩ.cm at 25  C).

2.7

Lipid Extractions

1. Corex tubes 15 mL high strength. 2. Boiling ethanol. 3. Methanol/chloroform (1:2, v/v). 4. Quartz wool (see Note 1). 5. Potters pestles (see Note 2). 6. Evaporator under argon (see Note 3). 7. Ultrapure water (prepared by purifying deionized water, to attain a sensitivity of 18 MΩ.cm at 25  C). 8. NaCl 1% (w/v) in ultrapure water. 9. Hemolysis glass tube for single use.

2.8

Methanolysis

1. PYREX® tubes of 20 mL (160 mm  16 mm), tubes in borosilicate glass (high thermal resistance) with screw top. Screw caps in phenolic resin with rubber seal lined with inert PTFE. 2. Standard C15 (fatty acid with 15 carbons) solution: Dissolve pentadecanoic acid (in powder, purity 99%) in chloroform/ methanol (1:2, v/v) to prepare a solution at 0.5 mg/mL (see Note 4).

Lipid Trafficking in Plant Mitochondria

75

3. Dry hot bath. 4. Methanolysis buffer: 2.5% (v/v) sulfuric acid (H2SO4 24N) in methanol (see Note 5). 5. Ultrapure water (prepared by purifying deionized water, to attain a sensitivity of 18 MΩ.cm at 25  C). 6. Hexane for GC analysis. 7. Hemolysis glass tube for single use. 2.9 Total Fatty Acid Quantification by Gas Chromatography– Flame Ionization Detector (GC-FID)

1. Vials for automatic sampler with an insert of 250 μL and 9 mm screw caps with PTFE seal. 2. Hexane for GC analysis. 3. Column BPX70 (70% Cyanopropyl Polysilphenylene-siloxane) for GC with a length of 30 m, an internal diameter of 0.22 mm and a film thickness of 0.25 μm. 4. Gas Chromatography–Flame Ionization Detector. 5. Commercial FAME standard solution to calibrate the GC-FID retention time.

2.10 Lipid Class Quantification by Liquid Chromatography– Mass Spectrometry (LC-MS)

1. Internal standard solution: prepare 1 mL of stock solution for each internal standard at 1.25 mM in chloroform/methanol (2:1,v/v) (see Note 6). Internal standards for PE 18:0–18:0 and PC 18:0–18:0 (Avanti Polar Lipids). DGDG 16:0–16:0 was synthesized chemically [12]. As an alternative, DGDG 18:0–18:0 can be obtained from purchased natural extract (Avanti Polar Lipids) and hydrogenated (see Note 7). To prepare the internal standard solution, in a 50 mL volumetric flask, add 50 μL of each individual standard and complete to 50 mL with chloroform/methanol (2:1, v/v). Prepare 1 mL aliquots of the solution in hermetically sealed vials and store at 20  C. 2. Hemolysis glass tube for single use. 3. Evaporator under argon (see Note 3). 4. High-Pressure Liquid Chromatography (HPLC) coupled to a triple quadripole mass spectrometer (see Note 8). 5. Mobile phase A: hexane/isopropanol/water/ammonium acetate 1 M, pH 5.3 (625:350:24:1, v/v/v/v). 6. Mobile phase B: isopropanol/water/ammonium acetate 1 M, pH 5.3 (850:149:1, v/v/v).

3

Methods

3.1 Preparation of Callus Liquid Cultures

1. Put approximately 20 seeds of WT or mutant plants in a 2 mL Eppendorf tube for sterilization. 2. Add 1 mL of ethanol 70% and incubate the seeds during 2 min at room temperature under agitation.

76

Juliette Jouhet et al.

3. Discard ethanol and add 1 mL of sterilization solution (see Note 9). 4. Incubate for 12 min at room temperature under agitation. 5. Discard the sterilization solution and wash the seeds four times with 1 mL of sterile water. 6. Resuspend the seeds in 100 μL of 0.2% agarose. 7. Spread the seeds one by one on MS seedling plates with a pipette, tape the plates with a porous adhesive, and allow the stratification of the seeds in the dark at 4  C for 48 h. 8. Incubate the plates under long-day condition (16 h in the light/8 h in the dark) at 22  C for 2 weeks. 9. Carefully cut with a sterile razor leaves from seedlings and place them on MS callus plates with the epidermal layer contacting the medium. 10. Incubate the plates under continuous light. Callus will be formed after 4–5 weeks. 11. To maintain callus, every 4 weeks, cut small pieces of callus with a sterile razor blade and transfer them on a new MS callus plate. 12. To obtain liquid cultures, add around ten 4-week-old callus to 200 mL of MS +Pi medium and incubate under continuous light at 22  C on a rotary shaker at 125 rpm. Callus are then subcultured every 7 days in 200 mL of MS +Pi medium (see Note 10). 13. For one experiment, 5 mL of sedimented callus (see Note 10) of 7-day-old cultures are washed three times with MS +Pi or MS –Pi medium in a 50 mL sterile tube and used to inoculate 200 mL of MS +Pi or MS –Pi medium in a 1 L flask. For a typical experiment, three flasks of MS +Pi medium and four flasks of MS –Pi medium are inoculated. To compare the results, experiments from Col0 and atmic60 grown in the presence or the absence of phosphate are performed the same day, meaning that four mitochondria purifications are performed at the same time. 14. Callus are grown in a continuous light under agitation at 125 rpm for 4 days to be in exponential phase of lipid remodeling for mitochondria purification. 3.2 Mitochondria Purification from Callus Liquid Cultures

To preserve mitochondria integrity and functions, each step of the purification has to be performed at 4  C with all materials (mortar, tubes, rotors, etc.) and buffers prechilled at 4  C. Mitochondria have to be manipulated carefully. Particularly, during resuspension steps, cut tips have to be used (see Note 11). 1. Filter each cell flask on round Whatman filter paper using a filtration unit. Cells are weighed and placed on a mortar (see

Lipid Trafficking in Plant Mitochondria

77

Note 12). Typically, we obtain about 10 and 8 g of callus grown for 4 days in MS +Pi or MS –Pi medium, respectively. 2. In a cold room, add 30 mL of grinding buffer and 15 mL of sand for 64 g of callus to the mortar. Grind the cells with a pestle until obtaining a homogenous suspension. This takes generally 6–8 min to obtain a complete grinding. 3. Add grinding buffer (20 mL) to the lysate in order to facilitate the transfer into 500 mL centrifuge bottles. Further wash the mortar with 20 mL of grinding buffer. 4. Centrifuge cell lysates at 700  g for 5 min at 4  C to pellet unbroken cells, heavy material, and sand. 5. Centrifuge the supernatant two times at 3000  g for 5 min at 4  C in 50 mL polycarbonate centrifuge tubes. 6. Pellet crude mitochondria by a centrifugation of the supernatant at 20,000  g for 15 min at 4  C. 7. Carefully resuspend mitochondria in 2 mL of washing buffer using a paintbrush and completely dissociate the mitochondria in a 5 mL Potter homogenizer by three gentle strokes. Use cut tips to manipulate crude mitochondria. 8. Prepare four discontinuous percoll gradients (one for each condition): pour 5 mL of Percoll 40 at the bottom of 32 mL ultracentrifuge tubes for swinging rotor. Then, gently add 20 mL of Percoll 23 and 10 mL of Percoll 18 (see Note 13). The three layers should be clearly visible. 9. Carefully add 2 mL of crude mitochondria on the top of each gradient with a pipette. 10. Centrifuge the tubes at 70,000  g for 45 min at 4  C in a swinging rotor, with standard acceleration at the beginning of the run but a slow deceleration at the end to preserve the gradient layers. 11. After centrifugation, mitochondria are located at the 23/40% interface (see Note 14). 12. Carefully aspirate and discard the upper part of the gradient. 13. Pipette the mitochondria fraction and place them in a 50 mL polycarbonate centrifuge tube (one per gradient). Fill the tube with washing buffer. 14. Centrifuge at 20,000  g for 15 min at 4  C. 15. Carefully discard the supernatant (see Note 15). Fill the tubes with washing buffer and centrifuge again at 20,000  g for 15 min at 4  C. 16. Discard the supernatant and resuspend carefully the pellet in 2 mL of washing buffer.

78

Juliette Jouhet et al.

17. Mitochondria are then transferred in 2 mL microfuge tubes and centrifuged at 12,000  g for 10 min at 4  C. 18. Resuspend mitochondria in 100–200 μL of washing buffer and determine the concentration of mitochondria in terms of protein quantity by a Bradford assay [13]. 3.3 Galactoglycerolipid Labeling and Transfer with UDP-[14C]Galactose

1. The incorporation of UDP-[14C]-galactose in the MTL complex is performed in a final volume of 250 μL of washing buffer containing 750 μg of mitochondria proteins, 1 mM DTT and 1 mM MgCl2. 2. Start the reaction by adding UDP-[14C]-galactose to a final concentration of 1 mM (see Note 16). 3. Incubate the samples for 30 min at 22  C. 4. To eliminate unincorporated UDP-[14C]-galactose, pellet the mitochondria by a 10-min centrifugation at 12,000  g at 4  C. 5. Discard the supernatant and resuspend the mitochondria in 500 μL of washing buffer.

3.4 Membrane Purification and Complex Solubilization

1. Mitochondria membranes are broken by osmotic shock. After the last wash, transfer the mitochondria (radiolabeled or not) in a 1.5 mL ultracentrifuge tube and pellet mitochondria for 10 min at 12,000  g at 4  C. 2. Resuspend the mitochondria in 1 mL of swelling buffer and incubate the suspension for 5 min on ice. 3. Mitochondria breakage is achieved by mixing the mitochondria for 10 s at 500 rpm with a shaker. 4. Pellet the membranes by a centrifugation at 100,000  g for 20 min at 4  C. 5. After centrifugation, discard the supernatant and resuspend the membranes in 30–50 μL of membrane buffer for labeling experiment (Fig. 1, step 1) or in 100–300 μL for lipidomic analyses (Fig. 1, step 2). 6. Estimate the concentration of mitochondrial membranes proteins using a Bradford assay [13]. We usually obtain about 50–100 μg of mitochondrial membranes proteins per g of callus. 7. To isolate and analyze the MTL complex, mitochondrial complexes are solubilized with DDM. Membranes containing 20 μg of proteins are solubilized and loaded into one lane of the CN-PAGE gel (see Note 17). For lipidomic analysis of the MTL complex, complexes from four lanes are pooled, meaning that 80 μg of membrane proteins have to be solubilized per experiment. For membranes containing 20 μg of proteins, the solubilization is performed in a total volume of 20 μL of

Lipid Trafficking in Plant Mitochondria

79

membrane buffer by the addition of DDM at a final concentration of 1.5 μg per μg of membrane proteins to a 1.5 mL ultracentrifuge tube (see Note 18). 8. Solubilization is achieved by pipetting slowly the mix 30 times followed by 5-min incubation on ice. 9. After incubation, pellet the insoluble materials by a 20-min centrifugation at 100,000  g at 4  C. 10. After centrifugation, carefully pipette the supernatant without touching the pellet and mix it with loading buffer 5 (5 μL for 20 μL of supernatant). 3.5 Isolation of the MTL Complex by CN-PAGE 3.5.1 Preparation of the CN-PAGE

1. The native gels are prepared in a Bio-Rad Mini-Protean® gel system with 1-mm glass plates (see Note 19). 2. A 15 mL gradient mixer and a peristaltic pump are used to cast the gel. Put the chamber on a magnetic stirrer. Link the first chamber of the gradient mixer to the peristaltic pump. Before use, wash the gradient mixer and the pump with 30 mL of ultrapure water. Then, close the connections between the chamber 1 and the pump and between the two chambers and fix the end of the pump tube on the top of the glass plates. 3. Add a small magnetic stirring bar in the first chamber. 4. For one gel, prepare 3 mL of separation gel 12% and 3.5 mL of separation gel 3.5%. 5. Right away after the addition of TEMED, load 2 mL of separation gel 12% and 2.5 mL of separation gel 3.5% in the first and second chamber, respectively, and turn on the stirrer in the first chamber (see Note 20). 6. Turn on the pump (flux of 3 mL/min), open the connection between the chamber 1 and the pump tube, and immediately open the connection between the chambers to allow the formation of the gradient. Wait until all the separation gel fill the glass plates. Usually, with these volumes of separation gels, about 1 cm remains from the top of the gel to the top of the plate to allow the addition of the concentration gel. 7. Carefully add water to the top of the gel. This allows making a straight border between the concentration and the separation gel. Wait about 1 h for a complete polymerization of the separation gel. 8. Remove residual water from the chamber by putting the chamber upside-down on an adsorbing paper. 9. Prepare (about 2 mL) and pour immediately the concentration gel on the top of the separation gel and add a 10-well comb,

80

Juliette Jouhet et al.

while avoiding air bubble formation. Wait about 1 h for a complete polymerization. 3.5.2 Loading of Samples and Separation of the Complexes

1. Place the gel in the migration tank and fill the tank with the migration buffer. 2. Gently clean the wells and load the samples in the wells (25 μL/ well) (see Note 21). 3. Operate the migration of the gel at 80 V until the bromophenol blue reaches the bottom of the gel. This takes around 2 h–2 h 30 min (see Note 22).

3.5.3 Fixation and Staining of CN-PAGE

1. After the migration of the native gel, remove the gel from the glass plates and cut the concentration gel. Fix the gel in 25 mL of fixation solution for at least 30 min at room temperature (see Note 23). 2. After the fixation, replace the fixation solution with 25 mL of staining solution and incubate overnight at room temperature (see Note 24). 3. Discard the staining solution and wash the gel with ultrapure water and several incubations at room temperature. 4. The MTL complex appears as a wavy band around 800 kDa (Fig. 2a). In Col0, a decrease of the MTL complex molecular weight is observed in –Pi compared to +Pi. However, a smaller shift of the MTL complex is observed in atmic60 callus suggestive of an impairment in MTL complex remodeling during Pi starvation.

3.6 Analysis of [14C]DGDG Incorporation into the MTL Complex

1. If an UDP-[14C]-galactose labeling is performed (Fig. 1, step 1), dry the gel on a Whatman paper for 2 h using a gel dryer. 2. Expose a phosphorimager screen to the dried gel and reveal the radioactive signal with a laser scanner (see Note 25). Usually, an exposure of 3 days is enough to visualize the radiolabeled galactoglycerolipids present in the MTL complex (Fig. 2b). In a typical experiment for Col0, an increase by a factor 2 of the radioactive signal is observed in the MTL complex in the –Pi compared to the +Pi condition (Fig. 2b) [8]. In atmic60 callus, a decrease in the incorporation of [14C]-galactoglycerolipids is observed in the MTL complex in –Pi compared to Col0 callus (Fig. 2b). These results indicate that the rate of incorporation of DGDG into the MTL complex is altered in the absence of AtMic60, suggesting an involvement of AtMic60 in lipid trafficking during Pi starvation.

Lipid Trafficking in Plant Mitochondria

a

81

b

Callus Pi MW

720

*

C

C

m

m

C

C

m

m

+

-

+

-

+

-

+

-

*

*

*

*

*

*

*

480 242 146 66

Coomassie

[14C] labelling

Fig. 2 In vitro synthesis and transfer of radiolabeled galactoglycerolipids in the MTL complex of Col0 and atmic60 callus grown in the presence (+) or absence () of phosphate (Pi). After labeling, the MTL complex has been isolated by CN-PAGE. A Coomassie staining of the CN-PAGE (a) and its exposure to a phosphorimager plate (b) are shown. The MTL complex is labeled by an asterisk. C: Col0, m: atmic60, MW: molecular weight (in kDa)

1. Start the lipid extraction from a pellet of mitochondrial membranes containing at least 200 μg of proteins prepared according to Subheading 3.4, step 5 (see Note 26).

3.7 Analysis of the Lipid Composition of Mitochondrial Membranes and of the MTL Complex

2. Warm up a water bath with a hot block inside. Experiment can start when the water temperature is above 80  C. At the same time, warm up 25 mL of absolute ethanol until it boils, in an Erlenmeyer flask closed with double aluminum foil.

3.7.1 Lipids Extraction from Mitochondrial Membranes

3. Lipid extraction: resuspend the mitochondrial membranes in 1 mL of boiling ethanol, mix by pipetting, and transfer the mitochondrial membranes in a hemolysis tube. 4. Close the tube with double aluminum foil and dispose the tube in the hot block in the water bath for 5 min with some shaking using a pestle (see Note 27). This step inhibits phospholipase D activity. 5. Remove the tube from the hot block, check the ethanol volume left in the tube, and if necessary complete up to 1 mL with boiling ethanol. Add 0.5 mL of methanol to directly rinse the pestle in the tube. Remove the pestle and add 2 mL of chloroform. Blow argon in the solvent mixture during 1 min to remove oxygen (see Note 28). Close the tube with aluminium foil and leave it for 1 h at room temperature. 6. Filter the liquid in a 15 mL Corex tube with a funnel plugged with ethanol washed glass wool to remove cell debris. Add 0.75 mL of chloroform/methanol (2:1, v/v) to the Corex tube containing the cell debris to rinse it and pour the liquid in the funnel.

82

Juliette Jouhet et al.

7. Remove the funnel and add 1.25 mL of NaCl 1% to the filtrate mixture. Blow argon inside the tube to mix up solvents during 1 min. Centrifuge for 10 min at 1000  g to separate the organic and aqueous phase. 8. Collect the lower phase (organic phase) with a Pasteur pipette (see Note 29) and transfer it in a clean hemolysis glass tube. Dry all the solvents by blowing argon on top of the liquid. 9. Rinse the tube with 200 μL of chloroform and dry lipids by blowing argon on top of the liquid. Close the tube and store the lipid extract at 20  C. Lipid extracts are stable for several months. 3.7.2 Lipids Extraction from CN-PAGE

1. To perform a lipidomic analysis of the MTL complex (Fig. 1, step 2), cut four bands of MTL complex per sample (the equivalent of 80 μg of proteins of solubilized membranes) and pool them in a 5 mL hemolysis glass tube (see Note 30). As a control, two other bands located in different parts of the gel and a band corresponding to a lane without any protein are also cut to be analyzed and treated like the MTL complex sample (Fig. 3a). 2. The lipids are then extracted with a protocol adapted from Bligh and Dyer [14]. Put 500 μL of water and 1875 μL of chloroform/methanol (1:2, v/v) onto the acrylamide bands. 3. Vortex and incubate the samples for 1 h at room temperature under agitation to extract the lipids from the gel. 4. Add 625 μL of water and 625 μL of chloroform to promote the formation of two phases. 5. Vortex and centrifuge the tubes during 5 min at 1000  g. 6. Transfer the lower organic phase in a new hemolysis tube and dry it under argon. 7. Perform a second extraction from the aqueous phase with 625 μL of chloroform. 8. After a 5-min centrifugation at 1000  g, pool the second organic phase with the first one and dry the lipids under argon. Lipids can be stored in a dried state at 20  C before analysis by mass spectrometry.

3.7.3 Production of Fatty Acid Methyl Esters (FAMEs) from Glycerolipids and Free Fatty Acids by Transesterification with Methanol (Methanolysis)

This step is performed only for lipids extracted from mitochondrial membranes to estimate the quantity of lipids that will be used for mass spectrometry analysis. The quantity of lipids extracted from CN-PAGE is too low and all the extracted lipids will be used for mass spectrometry analysis. 1. Turn on the hot block at 100  C and pre-warm an aliquot of standard C15 solution (see Note 31).

Lipid Trafficking in Plant Mitochondria

a

Callus

C +

Pi

C -

m +

83

m No - protein MW

MTL 720 480

Crt1 242 66 20

Crt2

b

Mitochondria membranes

MTL complex *

60

*

60

*

*

*

*

Col0_+Pi C o l0 _ + P i

*

*

Col0_-Pi C o l0 _ - P i

*

C P

D D

G

P

P

D G

atmic60_-Pi m ic 6 0 _ - P i

E

0

G

0 C

20

E

20

D

atmic60_+Pi m ic 6 0 _ + P i

**

**

P

40

G

Mol %

* 40

Fig. 3 Lipidomic analysis by mass spectrometry of the mitochondrial membranes and MTL complex of Col0 and atmic60 callus grown in the presence (+) or absence () of phosphate (Pi). (a) CN-PAGE bands analyzed by mass spectrometry. The MTL complex and two others complexes located at different parts of the gel were analyzed (crt1 and crt2). A control corresponding to a band cut in a lane without proteins was also included. (b) Lipid content of mitochondrial membranes and MTL complex of callus grown in + or –Pi. No lipids were detected in the control bands. Only DGDG, PC, and PE were detected in the MTL complex. Consequently, only these three lipids have been analyzed in mitochondrial membranes. The results are expressed in mol%. DGDG: digalactosyldiacylglycerol; PE: phosphatidylethanolamine; PC: phosphatidylcholine, C: Col0, m: atmic60, MW: molecular weight (in KDa). Statistical significance of the data was evaluated with two-tailed unpaired t-test using GraphPad software. *p < 0.05, **p < 0.01, n ¼ 3

2. Resuspend the lipid extract in 500 μL of chloroform. Collect 50 μL of the lipid extract and transfer it into the methanolysis tube. 3. With a Hamilton syringe, add 10 μL of C15 solution (5 μg/ tube) to the methanolysis tube. 4. Add 3 mL of methanolysis buffer in all the methanolysis tubes. Each tube must contain 10 μL of C15 solution, 50 μL of lipid extract, and the methanolysis buffer. Close tightly the glass

84

Juliette Jouhet et al.

tubes, vortex briefly, and incubate the tubes for 1 h at 100  C for the methanolysis reaction (esterification reaction) to occur. 5. Take tubes out of the hot block and let them cool down for 5 min at room temperature. Stop the reaction by adding 3 mL of water. 6. Add 3 mL of hexane to extract the FAMEs and vortex vigorously. Wait at least 20 min at room temperature to allow the proper formation of two phases. At this step, the sample can be stored a few days at 4  C if necessary. Take the upper phase (hexane phase) containing the FAMEs, transfer it in a hemolysis glass tube and dry it under argon. 7. Repeat step 6 to re-extract FAMEs from the methanol-water phase by adding again 3 mL of hexane in the methanolysis tube. Pour the upper phase in the hemolysis glass tube previously used in step 6 and dry it under argon. 8. To concentrate the FAMEs at the bottom of the glass tube, rinse the tube wall with 200 μL of hexane. Allow the liquid to rest at the bottom of the tube and dry it gently under argon. 9. Store the FAMEs at 20  C or proceed with GC analysis. 3.7.4 Quantification of FAMEs by Gas Chromatography–Flame Ionization Detector (GC-FID)

1. Resuspend the FAMEs in 50 μL of hexane and transfer the FAMEs in an insert vial. Tightly seal the vial to avoid any evaporation. 2. Inject 2 μL of the sample into the GC-FID on a BPX70 column. Nitrogen is used as carrier gas with 3.5 mL/min constant flow compensation, a split ratio of 13.3:1, an injection temperature of 200  C, a detector temperature of 280  C. The oven temperature range starts at 130  C, holds for 7.5 min at 130  C, ramp up to 180  C at 3  C/min, and holds for 10 min at 180  C. This allows FAMEs separation from 12 C up to 24 C in function of the chain length and the number of desaturation. 3. Each FAME is identified by comparison of its retention times with those of standards. FID response is dependent on the mass of the FAME; therefore, each FAME will be quantified by the surface peak method using C15 surface peak for calibration with the following equation:

Quantity in μg o f FAME ¼

Area o f FAME peak  Quantity in μg o f C15 ð5μgÞ Area o f C15 peak

4. By adding each FAME quantity and taking into account the volume used for the methanolysis, GC-FID analysis gives the total fatty acid content of the lipid extract in μg as well as its fatty acid composition. Fatty acid content in nmol can be established taking into account the molecular weight of each fatty acid.

Lipid Trafficking in Plant Mitochondria 3.7.5 Quantification of Lipid Molecules by LC/MS/MS

85

1. All the lipids extracted from CN-PAGE or 25 nmoL of fatty acids from mitochondrial membranes lipid extract are dissolved in 100 μL of internal standard solution. 2. Lipids are then separated by HPLC and quantified by ESI-MS/ MS. Lipid classes are separated using an HPLC system on a 150 mm  3 mm (length  internal diameter) 5 μm diol column at 40  C and mobile phases A and B. The injection volume is 20 μL, corresponding to 5 nmoL of total fatty acid (see Note 32), and each sample is injected three times as technical replicates. After 5 min, the percentage of B is increased linearly from 0% to 100% in 30 min and stays at 100% for 15 min. This elution sequence is followed by a return to 100% A in 5 min and an equilibration for 20 min with 100% A before the next injection, leading to a total runtime of 70 min. The flow rate of the mobile phase is 200 μL/min (see Note 33). 3. Mass spectrometric analysis is done on a triple quadripole mass spectrometer (see Note 34). The quadrupoles Q1–Q3 are operated at widest and unit resolution, respectively. Specific MRM (Multiple Reaction Monitoring) scans used to define molecular species need to be known (see Note 35). As MTL complex contains a small amount of lipids, only the main lipids (PC, PE, and DGDG) are detected and quantified (see Note 36). Negligible amounts of lipids are detected in control bands (Fig. 3a, crt1 and crt2) showing that the MTL complex is truly enriched in lipids [8]. Table 1 describes the acquisition settings for the analyzed glycerolipids. 4. To quantify each molecule of lipid, each chromatographic peak is integrated by the mass spectrometer software. The area of the peak is proportional to the quantity of the molecule. Because mass spectrometry detection efficiency is always molecule dependent, the internal standard is used to correct molecule bias quantification. Each molecular species (ms_LIP) of a lipid class (lc_LIP) is corrected by its corresponding internal standard (lc_is) to neglect the matrix effect by applying the following formula with Q corresponding to the quantity in pmol: Qðms LIPÞ ¼

Area ðms LIPÞ ∗Qðlc isÞ Area ðlc isÞ

Then all molecules of the same lipid class are summed to obtain the quantity of one lipid class: X Q ðms LIPÞ Q ðlc LIPÞ ¼

DGDG-36-2

DGDG-36-3

DGDG-36-4

DGDG-36-5

DGDG-36-6

DGDG-34-0

DGDG-34-1

DGDG-34-2

DGDG-34-3

DGDG-34-4

DGDG-34-5

DGDG-34-6

DGDG-32-1

DGDG-32-2

DGDG-32-3

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG-36-0

DGDG-36-1

DGDG-36-2

DGDG

DGDG

DGDG

Segment 2: from 7 to 16 min

DGDG-36-1

DGDG

Segment 1: from 0 to 7 min

962

964

966

904

906

908

926

928

930

932

934

936

938

954

956

958

960

962

964

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

621

623

625

563

565

567

585

587

589

591

593

595

597

613

615

617

619

621

623

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

8

8

8

8

8

8

8

8

8

8

8

8

8

8

8

8

8

8

8

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Compound group Compound name Prec Ion MS1 Res Prod Ion MS2 Res Dwell Fragmentor Collision energy Cell accelerator voltage Polarity

Table 1 Transition and mass spectrometer parameters used to detect and quantify Arabidopsis lipids

86 Juliette Jouhet et al.

DGDG-36-3

DGDG-36-4

DGDG-36-5

DGDG-36-6

DGDG-34-0

DGDG-34-1

DGDG-34-2

DGDG-34-3

DGDG-34-4

DGDG-34-5

DGDG-34-6

DGDG-32-1

DGDG-32-2

DGDG-32-3

DGDG-32-0

PE-36-1

PE-36-2

PE-36-3

PE-36-4

PE-36-5

PE-36-6

PE-34-1

PE-34-2

PE-34-3

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

DGDG

Std DGDG

PE

PE

PE

PE

PE

PE

PE

PE

PE

714

716

718

736

738

740

742

744

746

910

904

906

908

926

928

930

932

934

936

938

954

956

958

960

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

573

575

577

595

597

599

601

603

605

569

563

565

567

585

587

589

591

593

595

597

613

615

617

619

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

20

20

20

20

20

20

20

20

20

8

8

8

8

8

8

8

8

8

8

8

8

8

8

8

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

(continued)

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Lipid Trafficking in Plant Mitochondria 87

PE-32-1

PE-32-2

PE-36-0

PE

PE

std PE

PC-36-2

PC-36-3

PC-36-4

PC-36-5

PC-36-6

PC-34-0

PC-34-1

PC-34-2

PC-34-3

PC-34-4

PC-32-0

PC-32-1

PC-32-2

PC-36-0

PC

PC

PC

PC

PC

PC

PC

PC

PC

PC

PC

PC

PC

Std PC

790

730

732

734

754

756

758

760

762

778

780

782

784

786

788

748

688

690

712

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

Widest

184

184

184

184

184

184

184

184

184

184

184

184

184

184

184

607

547

549

571

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

Unit

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

30

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

135

34

34

34

34

34

34

34

34

34

34

34

34

34

34

34

20

20

20

20

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

7

Prec Ion precursor ion, Res resolution, Prod Ion product ion, DGDG digalactosyldiacylglycerol, PE phosphatidylethanolamine, PC phosphatidylcholine

PC-36-1

PC

Segment 3: from 16 to 40 min

PE-34-4

PE

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Positive

Compound group Compound name Prec Ion MS1 Res Prod Ion MS2 Res Dwell Fragmentor Collision energy Cell accelerator voltage Polarity

Table 1 (continued)

88 Juliette Jouhet et al.

Lipid Trafficking in Plant Mitochondria

89

Data are then expressed in mol% of PC, PE, and DGDG in the MTL complex or in mitochondrial membranes (Fig. 3b). Typically, in Col0 callus, we observe an increase in the DGDG content and a decrease in PE and PC in both mitochondrial membranes and MTL complex in -Pi. However, in the absence of AtMic60, the level of DGDG barely increases during Pi starvation. Interestingly, whereas the level of PE slightly decreases in mitochondrial membranes, it increases in the atmic60 MTL complex during Pi starvation, suggesting that PE accumulates inside the MTL complex. These results show that the traffic of both DGDG and PE is altered in atmic60 mitochondria during Pi starvation.

4

Notes 1. Quartz wool corresponds to pure SiO2 wool with a fiber thickness from 4 to 12 μm. It does not react with solvent and does not contain lipids. 2. Rinse potter pestles with distilled water, then ethanol. 3. If the evaporator bath could warm, set the temperature at 40  C maximum. It will decrease the evaporation time. 4. To prepare the standard solution at 0.5 mg/mL: Put 25 mg of C15 in a 1.5 mL tube. Add 1 mL of chloroform/methanol (1:2, v/v). Agitate the mixture manually. Transfer this solution in a 50 mL graduated flask. Add chloroform/methanol (1:2, v/v) to a volume of 50 mL. Put a glass cap and parafilm around. Invert the flask several times. Put some milliliters of this C15 solution in a little beaker (to avoid evaporation) and prepare 500 μL aliquots. Store them at 20  C. 5. Prepare the solution at 4  C to avoid overheating of the solution. Add 400 mL of methanol to a glass bottle, under agitation. Then, slowly add 10 mL of sulfuric acid. Put a glass cap and store at room temperature. 6. All the stock standard solutions need to be quantified by GC-FID to check that concentrations are accurate. If necessary, the volume can be adjusted to be sure that the standard solution is at 1.25 μM for each standard. 7. Solubilize 5 mg of DGDG in 1 mL of chloroform in a glass vial (HPLC type). Add a small spatula of PtO2 powder (catalyzer). Add a small magnetic stirring bar (4 mm). Label on the surface of the vial where the level of liquid is. Close the vial. Insert two thin needles in the center of the lid to allow hydrogen to get in and out. Needles need to be above the liquid. Prepare an Erlen with a lid resistant to chloroform with two holes, one for the entrance of hydrogen and the other one, serving as a “safety

90

Juliette Jouhet et al.

valve,” constituted of a Pasteur pipette with a finger of cut latex glove. Put the HPLC vial into the Erlen with the gas entrance of the Erlen connected to the first needle of the vial (use silicon tube and junctions). Put the Erlen above a magnetic stirrer under the fume hood. Open the hydrogen bottle or generator to deliver a maximum of 3.5 bars to fill the Erlen with hydrogen and then reduce the pressure to 1 bar. Leave overnight under small stirring. The following day recover the lipid by adding chloroform and collecting the liquid. Quantify and check the DGDG composition by methanolysis and GC-FID (see Subheadings 3.7.3 and 3.7.4). 8. Here, settings are described for an HPLC Agilent 1200 and a triple quadripole Agilent 6460. Parameters might vary for other instruments. 9. To avoid collecting seeds, centrifuge the tubes a few seconds at low speed and pipette carefully the solution without touching the seeds. 10. To obtain stable culture, transfer around 50 mL of the 7-day-old culture into a sterile 50 mL tube. Wait until the callus sediment (the volume of callus should not exceed 5 mL) and throw away the excess of medium. Recover the callus by resuspending them with fresh medium and transfer them into the culture flask. Complete the medium to 200 mL. 11. Materials for mitochondria preparations (tubes, glassware, etc.) have to be detergent free. It is thus recommended to dedicate a batch of material to mitochondria preparation and to wash the material with distillated water only. 12. After filtration, cells have to be maintained at 4  C before grinding. Thus, a prechilled mortar is stored on ice close to the balance. Cells are immediately deposited inside the mortar after weighting. 13. To deposit the 23% and 18% Percoll layers on top of the Percoll 40% layer, we use a peristaltic pump. The tubes containing the 5 mL of Percoll 40% are slightly inclined on ice and the exit tube of the pump is fixed at the top of the 32 mL tube. The entry tube of the pump is dived on a falcon containing 20 mL of Percoll 23% and the pump is turned on. Usually, we use a flow rate of 1.5 mL/min to layer the Percoll 23% and 18% solutions. During the distribution with the pump, the solution has to slide on the surface of the tube. If a drop falls directly on the gradient, it can create perturbation of the gradient. 14. Mitochondria purified from MS –Pi condition are more contaminated by the plastid fraction at this step.

Lipid Trafficking in Plant Mitochondria

91

15. At this step, the mitochondria pellet is not stable. Discard the supernatant by carefully pipetting the upper part of the tube. Leave around 5 mL of buffer in the tube. 16. If labeling is performed, from this step, all samples will be radioactive and all materials and wastes in the subsequent steps have to be treated in consequence in appropriate waste disposals. 17. We usually load 20 μg of membranes per lane. It is possible to load a higher quantity if necessary but we recommend loading no more than 30 μg per lane such as to not disturb the migration of the MTL complex. 18. For 20 μg of membranes, 30 μg of DDM is added. Thus, we add 3 μL of DDM 1% (w/v) to 20 μg of membrane fraction and adjust the volume to 20 μL with membrane buffer. Higher concentration of DDM can also be used but 1.5% is the reference concentration that we used to analyze the MTL complex. 19. The native gel can be casted the day before the experiment and stored at 4  C. However, do not use a gel that has been stored more than 48 h at 4  C. 20. At this step, it is important to proceed as fast as possible to avoid the polymerization of the gel in the gradient chamber or in the pump system. If more than one gel is required, it is better to cast one gel and then prepare a new mix of 12% and 3.5% separation gel to eventually cast a second gel. 21. Do not vortex or heat the samples before loading in the gel. 22. Do not perform the migration at a voltage higher than 80 V to avoid the formation of a smear by the MTL complex. 23. The fixation can also be performed overnight. 24. This process of fixation and staining is compatible with a subsequent analysis by mass spectrometry. 25. Add a thin plastic layer that does not absorb the radiation between the dried gel and the screen. 26. Mitochondrial membranes can be stored at 80  C if a long storage is required. 27. Be careful: samples should not dry. If there is almost no liquid left, take out the tube from the bath before the end of the 5-min incubation and continue the protocol. If several extractions are done at the same time, it is recommended to treat each sample one by one at this step. 28. Open the argon very slowly after diving the Pasteur pipette into the liquid to avoid any spilling of the solvent. 29. To avoid any contamination with the upper phase, use a propipette at the top end of the Pasteur pipette. Suck a little bit of air into the pipette, dive it at the bottom of the tube, eject one

92

Juliette Jouhet et al.

or two bubbles that will reject any upper phase that might have penetrated into the pipette, and suck up the lower phase. 30. The bands can be stored at 80  C if a long storage is required. 31. Let the aliquot of standard C15 at room temperature at least for 20 min before use. No overthrow but gentle agitation. 32. We have verified that the method is within a linear range up to 15 nmoL of total fatty acids. 33. The distinct glycerophospholipid classes were eluted successively as a function of the polar head group. Under these conditions, they were eluted in the following order: DGDG, PE, and PC. 34. For a 6460 triple quadrupole mass spectrometer (Agilent) equipped with a Jet stream electrospray ion source the source parameters are the following settings: Drying gas heater: 260  C, Drying gas flow 13 L/min, Sheath gas heater: 300  C, Sheath gas flow: 11 L/min, Nebulizer pressure: 25 psi, Capillary voltage:  5000 V, Nozzle voltage  1000 V. Nitrogen is used as collision gas. 35. To use this method, it is mandatory to know the glycerolipidome of the organism that is studied. Only lipid transitions that are entered in the method will be measured. It is a targeted method. 36. After an initial analysis, only three lipids (PC, DGDG, and PE) were detected in the MTL complex. If all the lipids need to be analyzed, you can find all the transitions and standards we usually use in our lab in [15]. References 1. Dacks JB, Field MC (2007) Evolution of the eukaryotic membrane-trafficking system: origin, tempo and mode. J Cell Sci 120:2977–2985 2. Nebenfu¨hr A (2002) Vesicle traffic in the endomembrane system: a tale of COPs, Rabs and SNAREs. Curr Opin Plant Biol 5:507–512 3. Prinz WA (2010) Lipid trafficking sans vesicles: where, why, how? Cell 143:870–874 4. Raychaudhuri S, Prinz WA (2008) Nonvesicular phospholipid transfer between peroxisomes and the endoplasmic reticulum. Proc Natl Acad Sci U S A 105:15785–15790 5. Schulz TA, Choi MG, Raychaudhuri S et al (2009) Lipid-regulated sterol transfer between closely apposed membranes by oxysterolbinding protein homologues. J Cell Biol 187:889–903 6. Jouhet J, Marechal E, Baldan B et al (2004) Phosphate deprivation induces transfer of

DGDG galactolipid from chloroplast to mitochondria. J Cell Biol 167:863–874 7. Jouhet J, Marechal E, Bligny R et al (2003) Transient increase of phosphatidylcholine in plant cells in response to phosphate deprivation. FEBS Lett 544:63–68 8. Michaud M, Gros V, Tardif M et al (2016) AtMic60 is involved in plant mitochondria lipid trafficking and is part of a large complex. Curr Biol 26:627–639 9. Boudie`re L, Michaud M, Petroutsos D et al (2014) Glycerolipids in photosynthesis: composition, synthesis and trafficking. Biochim Biophys Acta 1837:470–480 10. Michaud M, Prinz WA, Jouhet J (2017) Glycerolipid synthesis and lipid trafficking in plant mitochondria. FEBS J 284:376–390 11. Alonso JM, Stepanova AN, Leisse TJ et al (2003) Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 301:653–657

Lipid Trafficking in Plant Mitochondria 12. Amara S, Barouh N, Lecomte J et al (2010) Lipolysis of natural long chain and synthetic medium chain galactolipids by pancreatic lipase-related protein 2. Biochim Biophys Acta 1801:508–516 13. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254

93

14. Bligh EG, Dyer WJ (1959) A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37:911–917 15. Jouhet J, Lupette J, Clerc O et al (2017) LC-MS/MS versus TLC plus GC methods: consistency of glycerolipid and fatty acid profiles in microalgae and higher plant cells and effect of a nitrogen starvation. PLoS One 12: e0182423

Chapter 8 Bi- and Trifunctional Lipids for Visualization of Sphingolipid Dynamics within the Cell Doris Ho¨glinger Abstract Investigations into lipid localization and transport are often hampered by a lack of methods and tools to faithfully visualize lipids in the context of living cells, since fluorescent modifications drastically change lipid properties. Here, we describe the use of bifunctional as well as trifunctional sphingosine to reveal its subcellular localization via crosslinking, fixation, and specific staining by click reaction with a fluorophore. Additionally, these probes allow investigations into lipid metabolism as revealed by thin-layer chromatography. Key words Bifunctional lipids, Trifunctional lipids, Sphingosine, Lipid metabolism, Lipid localization, Thin-layer chromatography, Crosslinking

1

Introduction Sphingolipids represent one of the major classes of lipids in eukaryotes. They form a complex network which is tightly interconnected across different organelles [1]. Many sphingolipids exhibit bioactive and signaling functionalities, the mechanisms of which have been steadily elucidated in the past years. Lipid identification by mass spectrometry (“lipidomics”) played a big role in furthering our understanding of sphingolipid diversity [2] and is crucial to much of the knowledge we currently have about sphingolipid metabolism. Unfortunately, information about subcellular localization is lost in the lipidomics workflow. This may explain why many questions still remain open with regard to lipid compartmentalization and subcellular transport. In attempting to tackle these questions, fluorescently tagged lipid analogues have been widely applied [3, 4]. However, their use faces drawbacks since fluorescent groups are bulky and severely change the physical and chemical properties of the target lipid. A step toward more “natural” lipid analogues came with the development of polyene lipids, analogues that fluoresce due to conjugated double bonds in the acyl chain [5].

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_8, © Springer Science+Business Media, LLC, part of Springer Nature 2019

95

Doris Ho¨glinger

96

Here, the drawbacks mainly lie in the dimness of the fluorophore as well as the need for two-photon excitation to overcome photobleaching. A new tool to faithfully visualize lipid localization are photoactivatable and clickable (pac) lipids. Initially designed to capture lipid-interacting proteins, as reported using pac-cholesterol [6], the usefulness of this concept to visualize lipid localization was shown using pac-fatty acid [7] as well as pac-sphingosine [8]. Since the photoactivatable diazirine moiety as well as the alkyne click handle are very small modifications, these lipid probes exhibit close to natural properties. They are readily taken up by the endogenous lipid metabolizing machinery and their metabolic conversion rates are comparable to their endogenous lipid counterparts [8]. The ability to crosslink to proteins allows a faithful fixation of the lipid probe followed by covalent attachment of a fluorophore to the alkyne click handle of the lipid, thereby revealing the original localization of the lipid probe at the time of crosslinking. This probe design is also useful for following lipid metabolism since the modified lipid (clicked to a fluorophore) can be visualized by thin-layer chromatography [9]. While this methodology lacks the species-precision of lipidomics, it is a quick, cheap, and easy way to investigate lipid metabolism and to correlate the findings with localization information obtained in microscopy experiments. However, due to the close mimicry of endogenous lipids, pac(or bifunctional) lipids cannot be easily used for investigations of signaling lipids such as sphingosine. These lipids have low abundance and are turned over very quickly. Here, the addition of a “caging” group (see Fig. 1), a photocleavable protection group which prevents metabolism upon addition, is advantageous as the lipid can be released inside cells with a flash of light. A quick succession of uncaging and crosslinking of such “trifunctional”

trifunctional sphingosine

bifunctional sphingosine

OH

OH HO

HO NH2

N N

crosslink

HN

click

N N

O

crosslink

O O

O

click

N

cage

Fig. 1 Chemical structures of bifunctional sphingosine (pacSph) and trifunctional sphingosine. The coumarincage group can be cleaved using UV light with wavelengths >400 nm, while the diazirine crosslinking group requires shorter wavelengths around 355 nm. This enables sequential photoreactions (uncaging and crosslinking) to be performed on the same molecule. The alkyne click moiety is used for post-crosslinking functionalization with fluorophores

Visualization of Functionalized Lipids

97

lipid probes enables visualizing single lipid species, while also giving an excellent starting point for precise pulse-chase experiments [10]. Using bi- and trifunctional lipids, it is now possible to investigate lipid metabolism as well as lipid localization in intact cells using the same probe. Here, we provide streamlined protocols for both assays, as they are often needed to optimize and validate each other.

2

Materials

2.1 Thin-Layer Chromatography

1. Bifunctional sphingosine (pacSph, Avanti Polar Lipids) or trifunctional sphingosine at 10 mM in DMSO (see Note 1). 2. Labeling buffer: 20 mM HEPES-NaOH, pH 7.4, 115 mM NaCl, 1.2 mM MgCl2, 1.2 mM K2HPO4, 11 mM glucose, 1.8 mM CaCl2. To prepare 1 L, weigh 4.77 g of HEPES, 6.73 g of NaCl, 244 mg of MgCl2, 273.9 mg of K2HPO4, 2.18 g of glucose, and 264.6 mg of CaCl2. Add to a graduated 1 L cylinder. Add 900 mL of water and dissolve the salts while stirring. Adjust the pH with NaOH to 7.4 and make up to 1 L with water. Aliquot, filter through a 0.22 μm filter, and store at 4  C. 3. Coumarin-click mix: 43 μM 3-azido-7-hydroxycoumarin and 2 mM [MeCN]4CuBF4 in ethanol. Prepare a stock solution of 3-azido-7-hydroxycoumarin at 44.5 mM in ethanol and a 10 mM stock solution of [MeCN]4CuBF4 in acetonitrile. The latter has to be prepared freshly for every experiment. Then mix 1.2 μL of 3-azido-7-hydroxycoumarin solution with 250 μL of [MeCN]4CuBF4 and 1 mL of ethanol, vortex, and use immediately. 4. TLC solvent 1: Chloroform/methanol/water/acetic acid (65:25:4:1, v/v/v/v). To prepare 190 mL, pipet 2 mL of acetic acid (see Note 2) into 8 mL of water, add 50 mL of methanol, and finally add 130 mL of chloroform (see Note 3). 5. TLC solvent 2: Ethylacetate/cyclohexane (1:1, v/v). 6. HPTLC Silica Gel 60 plate (Merck). 7. Chloroform (reagent grade, >99.8%). 8. Acetic acid: 1% (v/v) in water. 9. TLC developing tank (e.g., Camag twin trough chambers). 10. Gel imager.

2.2

Microscopy

1. Bifunctional sphingosine (pacSph, Avanti Polar Lipids) or trifunctional sphingosine at 10 mM in DMSO (see Note 1). 2. Labeling buffer: as described in Subheading 2.1. 3. Fixation solution: Store methanol at 20  C to ensure sufficiently cold temperature for the fixation reaction.

98

Doris Ho¨glinger

4. Wash solution: Methanol/chloroform/acetic acid (55:10:0.75, v/v/v). To prepare 500 mL, add 5.75 mL of acetic acid (see Note 2) to 385 mL of methanol and mix. Then add 70 mL of chloroform (see Note 3). 5. Phosphate buffered saline (PBS): 137 mM NaCl, 18 mM Na2HPO4, 2.7 mM KCl, 1.5 mM KH2PO4. To prepare a 10 stock, weigh 80 g of NaCl, 14.42 g of Na2HPO4, 2 g of KCl, and 2 g of KH2PO4. Dissolve in 1 L of water, followed by autoclaving. 6. Fluorophore-click mix: 3 μM fluorophore-azide and 2 mM [MeCN]4CuBF4 in PBS. Prepare a stock solution of 2 mM fluorophore-azide (we use Alexa488-azide or Alexa594-azide) in DMSO and keep at 20  C. For each experiment, freshly make a 10 mM stock solution of [MeCN]4CuBF4 in acetonitrile. Prepare the click mix by adding 250 μL of [MeCN]4CuBF4 and 2 μL of fluorophore-azide to 1 mL of PBS and use immediately. 7. UV lamp capable of crosslinking at 365 nm (see Note 4).

3

Methods

3.1 Following Sphingolipid Dynamics by Thin-layer Chromatography

1. Plate cells (see Note 5) in 12-well plates and grow in appropriate medium so they reach a confluency of 90% on the day of the experiment. 2. Prepare bifunctional or trifunctional sphingosine working solutions from DMSO stock via direct dilution in labeling buffer. We recommend using a final concentration of 0.5–5 μM. 3. Remove growth medium from cells and label with working solutions for varying times, either continually or in a pulsechase fashion (see Note 6). 4. Wash with labeling buffer and, if trifunctional sphingosine is used, irradiate with >400 nm light for uncaging (see Note 7). 5. Scrape the cells or add trypsin and quench with growth medium. Spin the cells down (1500  g, 4  C, 5 min) and resuspend in 300 μL of PBS. 6. Extract lipids by addition of 600 μL of methanol and 150 μL of chloroform. Vortex. 7. Precipitate the protein pellet by centrifugation (12,000  g, 4  C, 5 min) and transfer the supernatant in a new 2 mL vial. 8. Add 300 μL of chloroform and 600 μL of acetic acid (1%, v/v, in water). Vortex. 9. Centrifuge at 12,000  g for 5 min at 4  C. This should result in a clear phase-separation.

Visualization of Functionalized Lipids HeLa S1PL-/-

99

HeLa WT

front

Cer

? Sph PC SM

Sph SM origin Std

-

5

10

20

30

45

60 min

Std 1

5 10 20 30 45 min

Fig. 2 Thin-layer chromatography analysis of pacSph metabolism in sphingosine-1-phosphate lyase deficient (S1PL / ) and WT HeLa cells. Lipid extracts of cells, pulsed with 1 μM pacSph for the indicated times, were reacted with 3-azido-7-hydroxycoumarin for visualization. In WT cells phosphatidylcholine (PC) is strongly labeled, as the bifunctional sphingosine probe can be cleaved by S1PL, yielding bifunctional hexadecanal, which can subsequently enter the glycerophospholipid metabolism. In S1PL / cells, Sph is metabolized to ceramide (Cer) and sphingomyelin (SM) as expected

10. Discard the upper (aqueous) phase and transfer the organic phase into a new vial. 11. Remove solvents either in a speed vac or by a stream of nitrogen. 12. Redissolve lipids in 7 μL of chloroform (see Note 3). 13. Add 30 μL of coumarin-click mix (see Note 8). Vortex briefly and spin down. 14. The click reaction can be performed either in a heating block at 37  C without the lid for approximately 3 h (see Note 9) or in a speed vac at 45  C for 20 min. 15. Redissolve samples (see Note 10) and apply to a HPTLC plate (see Note 11). 16. Develop with TLC solvent 1 for 5 cm from origin. 17. Dry the HPTLC plate completely in a fume hood. 18. Develop with TLC solvent 2 for 8 cm from origin. 19. Dry the HPTLC plate completely in a fume hood. 20. Visualize the HPTLC plate (see Fig. 2) using a system capable of excitation at around 400 nm and detection at 475 nm to see the coumarin fluorescence. We use a gel imager equipped with a 400 nm longpass filter. 3.2 Following Sphingolipid Dynamics by Microscopy

1. Plate cells in desired container (see Note 12) and grow in appropriate medium so they reach a confluency of 60–80% on the day of the experiment. 2. Prepare bifunctional or trifunctional sphingosine working solutions from DMSO stock solutions via direct dilution in labeling

100

Doris Ho¨glinger

Fig. 3 Visualization of pacSph localization in pulse-chase experiments. HeLa cells were pulsed with 1 μM pacSph for 5 min, followed by the indicated chase times. Cells were UV-irradiated and fixed, non-crosslinked lipids were washed away and the remaining crosslinked lipids were visualized by click reaction with Alexa488-azide. Scale bars: 10 μm

buffer. We recommend using a final concentration of 0.5–5 μM. 3. Remove growth medium from cells and label with working solutions for varying times (see Note 13). 4. Wash twice with labeling buffer and overlay with 0.5 mL of cold labeling buffer. 5. Place the culture plate on cold metal blocks under the light cone of a UV lamp for the predetermined duration (see Note 14). This photoreaction activates the diazirine crosslinking group and leads to the formation of covalent crosslinks between the lipid probe and proteins in the immediate vicinity. 6. Remove the labeling buffer and fix cells by addition of 1 mL of fixation solution (cold methanol). Keep at 20  C for 20 min. 7. Wash 3 times with 1 mL of wash solution to remove non-crosslinked lipids. 8. Wash twice with 1 mL of PBS and aspirate all liquid. Take care to place the coverslip in the middle of the well. 9. Add 50 μL of click mix (see Note 15) and incubate for 1 h at room temperature in the dark. 10. Wash twice with 1 mL of PBS. 11. Optional: Perform immunofluorescence experiments using markers of subcellular organelles. 12. Mount coverslip on glass slides and image using appropriate microscope settings. For a typical experiment, see Fig. 3.

4

Notes 1. Trifunctional sphingosine is currently only accessible by chemical synthesis as described in [10].

Visualization of Functionalized Lipids

101

2. Glacial acetic acid is highly corrosive to skin and eyes. Wear gloves and goggles, and, if possible, handle acetic acid in a wellventilated fume hood. Ensure there are eyewash stations close by. Always add acetic acid to water or methanol, never the other way around. 3. Chloroform is a suspected carcinogen. Avoid contact with eyes and skin. Always work in a well-ventilated fume hood and use personal protective equipment such as safety goggles, gloves, and a laboratory coat. 4. UV lamps for lipid crosslinking should have high power output in the 350–565 nm wavelength. We have so far tested two suitable models: a Newport mercury-xenon lamp (669241000HXF-R1) as well as a UVP Blak-Ray UV lamp (B-100AP). 5. In wild-type cells, the bifunctional sphingosine backbone can be cleaved by sphingosine-1-phosphate lyase (S1PL), producing (bifunctional) hexadecanal which can subsequently enter the glycerophospholipid pathway. In this case, the diazirine and alkyne modifications will be mainly found in phosphatidylcholine (PC). Therefore, S1PL-deficient cell lines need to be used for the study of sphingolipid metabolism. 6. We suggest to first perform continuous pulse experiments with varying times to investigate uptake of the probe. For studies of metabolism, short pulse times followed by varying chase times are better suitable. A good starting point would be to pulse the cells with probe for 5 min at 37  C, followed by two washes with labeling buffer and incubation for 5 min–2 h in labeling buffer or medium. 7. Uncaging of trifunctional sphingosine is achieved by putting a 400-nm longpass filter in the light path of the UV lamp (see Note 4) to ensure that the diazirine group is not prematurely crosslinked. 8. Remember to include a positive control (pacSph) as well as a negative control (no lipid) for the click reaction. 9. During incubation of the lipids with the click mixture at 37  C, make sure that the heating block is not agitated or covered with a lid. Evaporating solvents will condense on the lid of the vial. This evaporation is crucial to drive the click reaction to completion. Check that no more liquid is observed on the bottom and everything has condensed on the lid of the vial. 10. If the click reaction was performed by incubation at 37  C, redissolve lipids by vortexing the vial after completion of the click reaction. If the click reaction was done in a speed vac, redissolve the lipids by addition of 20 μL of ethanol/acetonitrile (4:1, v/v).

102

Doris Ho¨glinger

Fig. 4 Schematic layout of a thin-layer chromatography plate for analysis of lipid metabolism. Samples are spotted at the baseline 1 cm from the bottom of the plate. It is good practice to keep a distance of at least 1 cm from the right and left side of the plate. The TLC plate is then developed in TLC solvent 1 up to 5 cm from the origin. After drying in a well-ventilated fume hood, the plate is developed in TLC solvent 2 for 8 cm

11. Apply the samples either by pipetting manually or by employing an automatic sample applicator (such as Camaq ATS4) following the pattern in Fig. 4. 12. For imaging experiments, we usually plate cells on 11 mm diameter coverslips placed in the wells of a 24-well plate. The number of wells for use in each experiment depends on the diameter of the UV-light cone. A 10 cm diameter allows for simultaneous illumination of 3  3 wells. In addition, we recommend plating 2–3 coverslips outside the illuminated zone for use as negative controls. Of course, other cell culture vessels (such as chambered cover glasses) can be used as long as they are resistant to UV irradiation during the crosslinking step. 13. Best practice is to optimize labeling times beforehand via TLC experiments as described in Subheading 3.1. As a guide, we recommend incubation times of 5–10 min when using trifunctional sphingosine. Bifunctional sphingosine can be used in continuous labeling or pulse-chase experiments. For pulsechase experiments, we again counsel pulse times of 5–10 min, followed by individual chase times. 14. The duration of crosslinking is dependent on the distance of the sample and the power output of the lamp. These two parameters should be optimized in preceding microscopy experiments for maximal fluorescence intensity. 15. Take care that the click mix forms a stable drop on the coverslip without spreading to the rest of the well. If this is not possible, addition of 200 μL of click mix to the well of the 24-well plate usually results in sufficient coverage to ensure labeling of the whole coverslip.

Visualization of Functionalized Lipids

103

Acknowledgments D.H. is supported by the Deutsche Forschungsgemeinschaft through SFB/TRR 83 and SFB/TRR 186. References 1. Hannun YA, Obeid LM (2018) Sphingolipids and their metabolism in physiology and disease. Nat Rev Mol Cell Biol 19:175–191 2. Quehenberger O, Armando AM, Brown AH et al (2010) Lipidomics reveals a remarkable diversity of lipids in human plasma. J Lipid Res 51:3299–3305 3. Wang T, Silvius JR (2000) Different sphingolipids show differential partitioning into sphingolipid/cholesterol-rich domains in lipid bilayers. Biophys J 79:1478–1489 4. Pagano RE, Sleight RG (1985) Defining lipid transport pathways in animal cells. Science 229:1051–1057 5. Kuerschner L, Ejsing CS, Ekroos K et al (2005) Polyene-lipids: a new tool to image lipids. Nat Methods 2:39–45 6. Hulce JJ, Cognetta AB, Niphakis MJ et al (2013) Proteome-wide mapping of

cholesterol-interacting proteins in mammalian cells. Nat Methods 10:259–264 7. Haberkant P, Raijmakers R, Wildwater M et al (2013) In vivo profiling and visualization of cellular protein-lipid interactions using bifunctional fatty acids. Angew Chemie—Int Ed 52:4033–4038 8. Haberkant P, Stein F, Ho¨glinger D et al (2016) Bifunctional sphingosine for cell-based analysis of protein-sphingolipid interactions. ACS Chem Biol 11:222–230 9. Thiele C, Papan C, Hoelper D et al (2012) Tracing fatty acid metabolism by click chemistry. ACS Chem Biol 7:2004–2011 10. Ho¨glinger D, Nadler A, Haberkant P et al (2017) Trifunctional lipid probes for comprehensive studies of single lipid species in living cells. Proc Natl Acad Sci U S A 114:1566–1571

Chapter 9 Indirect Lipid Transfer Protein Activity Measurements Using Quantification of Glycosphingolipid Production Anders P. E. Backman, Josefin Halin, Matti A. Kjellberg, and Peter Mattjus Abstract Here we summarize how glycosphingolipid production can be followed using metabolic labeling with radiolabeled lipid precursors. No assays are available yet that directly would address the lipid transfer protein activity in vivo. Therefore, these approaches can serve as tools to indirectly study the lipid transfer protein activity in cells, by monitoring their impact on the glycosphingolipid homeostasis. Key words Lipid analysis, Radiolabel, Thin-layer chromatography, Glycolipids

1

Introduction Lipid transfer proteins can be easily analyzed in vitro with different approaches [1–3]. However, one of the challenges regarding their study in vivo is the lack of direct lipid transfer activity assays. Here we present methods to metabolically label cells in order to follow the production of glycosphingolipids. Different treatments of the cells, such as up- and downregulation of the lipid transfer protein expression, are necessary to gain insight into their activity and function [3–8]. By manipulating the synthesis rate of different sphingolipids by using inhibitors or gene silencing it is also possible to follow the expression levels of lipid transfer proteins and other proteins of interest in the sphingolipid pathway. For these types of experiments metabolic labeling of the sphingolipid pool is equally important, because the efficiency of the lipid manipulations need to be followed and analyzed. The incorporation of radioactive isotopes in the sphingolipid pool can be achieved with different precursors. It is up to the user to optimize the time that the radiolabeled precursors are allowed to be metabolized into the lipid pool of interest. In Fig. 1 the pathways of glycosphingolipid synthesis are presented. The radiolabeled products from the cellular lipid metabolism can be analyzed with

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_9, © Springer Science+Business Media, LLC, part of Springer Nature 2019

105

106

Anders P. E. Backman et al.

Fig. 1 The sphingolipid pathway. The radiolabeled compounds used in our metabolic labeling studies are highlighted in the red boxes

standard methods, such as high-performance thin-layer chromatography (HPTLC). The fast and inexpensive TLC of lipids, in particular glycolipids, has still not had its day [9]. TLC is a very powerful method, in particular in combination with mass spectrometric (MS) detection that can resolve the identities of closely related molecular species within the same lipid class. Here we present approaches used in our studies of the glycolipid transfer protein (GLTP) and its effects on the cellular glycolipids. However, these approaches can be adapted to many other types of lipid transfer proteins.

Glycosphingolipid Biosynthesis and Transfer Protein Activity

2

107

Materials Throughout these procedures it is paramount to keep all samples containing organic solvents in laboratory glassware. Glassware to be used for lipid analysis should be cleaned thoroughly and minimal amount of detergents should be used. Cleaning the glassware with organic solvents prior to use is also recommended. TLC silica plates can be primed prior to sample application. A common preconditioning method is to first run the “empty” TLC plate in the development chamber containing the elution solvent to be used, dry the plate in low humidity, and then apply the samples. The TLC plate is then developed in a fresh elution solvent. This minimizes the impact of impurities that will interfere with the samples and the detection methods.

2.1 Synthesis of Radiolabeled Ceramides and Sphinganine

All chemical reagents should be of analytical grade or higher. 1. Hexanoic acid, decanoic acid and hexadecanoic acid. 2. D-erythro-[3-3H]-sphingosine, 50 μCi (PerkinElmer). 3. D-erythro-sphingosine. 4. Sphinganine (Avanti Polar Lipids or equivalent). 5. Tritiated sodium borohydride (NaB3H4, PerkinElmer). 6. N,N-dicyclohexylcarbodiimide (DCC). 7. Dry dichloromethane. 8. PdCl2. 9. Tetrahydrofuran. 10. Diethyleneglycol dimethyl ether. 11. Diethyl ether. 12. TLC plates. 13. TLC solvent system: Chloroform/methanol/water (65:25:4, v/v/v). 14. Brown screw-capped vial (Sigma-Aldrich). 15. Supelco Discovery C18 column (25 cm  21.2 mm, 5 μm particle size).

2.2 Lipid Labeling in Cell

1. [9,10-3H]-palmitic acid, 25 mCi (PerkinElmer). 2. Cholesteryl phosphocholine (CholPC, Avanti Polar Lipids). 3. Phosphate buffered saline (PBS 1): Dissolve in one liter of distilled or deionized water, 8 g of NaCl, 0.2 g of KCl, 1.42 g of Na2HPO4, and 0.24 g of KH2PO4. Adjust the pH to 7.4. Final concentrations are 137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4 and 1.8 mM KH2PO4. 4. Water bath (FinnSonic M3 bath sonicator or similar).

108

Anders P. E. Backman et al.

5. Probe sonifier (Branson 250 probe Emerson Industrial Automation or similar). 2.3 Lipid Analysis After Cell Treatment

1. 1 PBS. 2. 0. 1 M NaOH. 3. HPTLC plates, Silica gel 60 F254, 10  10 cm (Merck). 4. Total lipid extraction solvent system: Hexane/isopropanol (3:2, v/v). Add 30 mL of hexane to 20 mL of isopropanol. 5. TLC solvent system for glycosphingolipids: Chloroform/ methanol/acetone/acetic acid/water (10:2:4:2:1, v/v/v/v). 6. TLC solvent system for globosides: Chloroform/methanol/ 0.2% CaCl2 (w/v) in water (45:55:10, v/v/v). 7. TLC solvent system for phospholipids: Chloroform/methanol/acetic acid/water (50:30:8:3, v/v/v). 8. Chloroform/methanol (2:1, v/v). 9. High quality chain pure and natural phospholipids and glycosphingolipid standards (Avanti Polar Lipids): 1,2-dimyristoylsn-glycero-3-phosphatidylethanolamine (DMPE), bovine brain sphingomyelin (SM), Gaucher liver glucosylceramide (GlcCer), bovine butter lactosylceramide (LacCer), bovine brain total galactosylcerebrosides (containing galactosylceramide (GalCer) both in a hydroxylated (phrenosin) and non-hydroxylated form (kerasin)) and globotriaosylceramide (Gb3). 10. Iodine. 11. Orcinol spray: 0.2% (w/v) orcinol in a 50% H2SO4 solution. 12. Scintillation fluid (Optiphase “Hi phase” scintillation fluid, PerkinElmer-Wallac). 13. Liquid scintillation counter (1216 Rackbeta or similar). 14. Nitrocellulose membrane (Whatman). 15. Anti-glucosylceramide, anti-galactosylceramide and other antiglycosphingolipid antibodies (Glycobiotech GmbH and Matreya). 16. HRP-conjugated Scientific).

mouse

secondary

antibody

(Thermo

17. Enhanced chemiluminescence detection kit (SuperSignal West Femto Maximum Sensitivity Substrate, Thermo Scientific).

3

Methods All methods are done in room temperature unless otherwise instructed.

Glycosphingolipid Biosynthesis and Transfer Protein Activity

3.1 Synthesis of Radiolabeled Ceramides and Sphinganine 3.1.1 Synthesis of [3H]Ceramides

109

Different lengths of ceramides, 3H-C6-ceramide, 3H-C10-ceramide and 3H-C16-ceramide can be prepared from [3-3H]D-erythro-sphingosine and hexanoic, decanoic and hexadecanoic acids using DCC and triethylamine as catalyst. [3H]-C16-ceramide with the radioisotope in the fatty acid can also be prepared from sphingosine and [9,10-3H]hexadecanoic acid. For comprehensive synthesis details see Byun and Bittman 1993 [10]. Ceramides with the 18:1 sphingoid base and different acyl chains were synthesized by coupling the fatty anhydride to the long-chain base in the presence of triethylamine. The fatty anhydride was prepared with DCC. The amount of sphingosine, fatty acid and DCC should be 1:5:5.1. 1. Dry the fatty acid and D-erythro-[3-3H]-sphingosine in separate brown glass vials overnight under vacuum at 40  C. 2. Dissolve the fatty acid in 1 mL of dry dichloromethane and add 100 μL of DCC to the vial. 3. Incubate for 1 h at 45  C with stirring. 4. Dissolve the D-erythro-[3-3H]-sphingosine in 2 mL of dichloromethane and combine with the fatty acid/dichloromethane/DCC mixture. 5. Add 10 μL of trimethylamine and incubate for 1 more hour at 45  C. 6. Dry the mixture under nitrogen. 7. Store at

20  C until HPLC and MS analysis.

8. The product is purified by preparative HPLC on a C18 column using 100% methanol as solvent (flow 1 mL/min at room temperature) and with UV detection at 203 nm [11]. Purity is assessed by molecular identity by ESI-MS. 3.1.2 Synthesis of D-erythro-[4,5-3H]Sphinganine from D-erythro-Sphingosine

Reduction of D-erythro-sphingosine to D-erythro-sphinganine is done with NaB3H4. This method will hydrogenate the sphingosine 4,5-trans double bond with tritium gas generated in situ using palladium as a catalyst [12]. 1. Dissolve 5 mg of D-erythro-sphingosine and 2 mg of PdCl2 in 2 mL of tetrahydrofuran in a brown screw-capped glass vial. 2. NaB3H4 (100 mCi) is dissolved in 50 μL of diethyleneglycol dimethyl ether and added immediately to a brown screwcapped vial. 3. The solution is flushed with nitrogen and the reaction is allowed to proceed for at least 3 h at room temperature under continuous stirring. 4. The reaction is stopped by addition of 2–3 mL of diethyl ether. 5. The products are extracted and dried with nitrogen.

110

Anders P. E. Backman et al.

6. The reaction product is purified on a TLC plate using the solvent system, chloroform/methanol/water (65:25:4, v/v/ v). 7. The radioactive [4,5-3H]-sphinganine is identified on the TLC plate by a sphinganine standard. 8. The [4,5-3H]-sphinganine is extracted from the silica by chloroform/methanol (2:1, v/v). 3.2 Lipid Labeling in Cell 3.2.1 Sphingolipid Labeling in Cell with 3H-Sphinganine and 3H-Sphingosine

The single chain sphinganine and sphingosine is easily taken up from the growth media by cells in culture when added in the form of an ethanol or DMSO solution. The sphingosine and sphinganine solubility in ethanol or DMSO is up to 25 mg/mL. 1. Cultured cells with 80% confluency are labeled with 0.33 mCi/ mL of D-erythro-[4,5-3H]-sphinganine or D-erythro-[3-3H]sphingosine dissolved in ethanol. This leads to a specific label in the sphingolipid pool. 2. The solution is added to the culture medium and allowed to incubate for at least 15 min at 37  C, after which the cells can be treated and subsequently analyzed. Gene-silenced or geneupregulated cells are labeled after the manipulation in the same fashion.

3.2.2 3H-palmitic Acid Labeling of Cells

The palmitic acid will be incorporated into many different classes of lipids, as well as degraded to acetate that is further metabolized. Therefore, this gives a much broader labeling than the 3H-sphingosine approach. 1. Cultured cells with 80% confluency are labeled with 1 mCi/mL [9,10-3H]-palmitic acid dissolved in ethanol and added to the culture medium. The solubility of palmitic acid in ethanol and DMSO is approximately 20 mg/mL. 2. After a 15-min incubation at 37  C, the cells can be treated or harvested for analysis.

3.2.3 Loading of Ceramide Using Cholesterol Phosphocholine (CholPC)

Cells in culture easily take up sphinganine, sphingosine, and free fatty acids; however, ceramide is very hydrophobic and has extremely poor water solubility. Liposomal ceramide complexes can be used to avoid solvents, for instance ceramide/CholPC complexes [13, 14]. We have successfully used this approach to study the effect of different lengths of ceramides on the GLTP expression levels in HeLa cells and human skin fibroblasts [4]. 1. Complexes of CholPC and radiolabeled ceramide (equimolar ratio) are prepared from the organic stock solutions, dried in a glass tube, redissolved in chloroform to ensure proper lipid mixing and dried again.

Glycosphingolipid Biosynthesis and Transfer Protein Activity

111

2. The lipid film is hydrated with PBS 1, pH 7.4, at 55  C for 20 min and sonicated for 10 min using a water bath sonifier followed by sonication with a probe sonifier at room temperature to achieve an opalescent clear solution. 3. The solution is centrifuged at 10,000  g for 10 min to remove any undispersed lipids. 4. Prior to use it is important to recalculate the ceramide concentration based on the specific activity of the radiolabeled ceramide to ensure that the right amount of ceramide is added to the tissue culture. 3.3 Lipid Analysis After Cell Treatment

1. After treatment the cells are washed with PBS and the tissue culture dishes dried in a fume hood.

3.3.1 Lipid and Protein Extraction

2. The total lipids are extracted directly from the cell dishes with total lipid extraction solvent system. 3. The extracted lipids are then transferred to glass tubes and dried under a stream of nitrogen. The dried lipids can be stored at 20  C until analysis is performed. 4. After lipid extraction, the cellular proteins are extracted from the tissue culture dishes with 0.1 M NaOH, and the protein content analyzed using the standard Lowry method [15]. The amount of lipids spotted on the HPTLC plates are usually normalized to the total cellular protein content.

3.3.2 Identification and Quantification of Lipid Species with HPTLC

1. The dried lipid samples are redissolved in total lipid extraction solvent system and analyzed on high-performance thin-layer chromatography (HPTLC) silica plates using different solvent systems. 2. Appropriate volumes of each sample are preferably applied using automated TLC samplers on TLC plates. This ensures high reproducibility and allows for sharp band separations. 3. Use the adequate TLC solvent system listed in Subheading 2.3 for phospholipids and Gb3. For simple glycosphingolipids such as GlcCer, GalCer, and LacCer, use the TLC solvent system for glycosphingolipids. 4. Lipid standards are run in parallel with the samples (see Notes 1 and 2). See Fig. 2 for suitable standards for different solvent systems. 5. Glycosphingolipid migration is visualized using orcinol spray (see Note 3) with subsequent heating for 5 min at 120  C. 6. Other lipids, such as phospholipids, are detected with iodine or copper acetate staining (see Notes 4 and 5).

112

Anders P. E. Backman et al.

Fig. 2 Lipid separation using HPTLC and different solvent systems. (a) Typical HPTLC plate for separation of glucosylceramide (GlcCer), galactosylceramide (GalCer), and lactosylceramide (LacCer) using the solvent system chloroform/methanol/acetone/acetic acid/water (10:2:4:2:1, v/v/v/v). The plates were stained both with the carbohydrate orcinol spray and with copper acetate, with subsequent heating to visualize all lipids (see Note section). (b) The separation of Gb3 from other glycosphingolipids, using the solvent system chloroform/methanol/0.2% (w/v) CaCl2 in water, also stained with both orcinol and copper acetate. Galactosylceramide separates into two bands where the upper is GalCer (kerasin) and the lower hydroxylated Galcer (phrenosine)

7. Silica spots are marked with a pencil and scraped into a scintillation fluid and the radioactivity is measured using a liquid scintillation counter. 8. Counts per minute (cpm) obtained are either normalized to the total protein content for each respective sample or presented as a percentage of the total signal per sample. Varying the incubation time before the cells are harvested will result in different lipid pools being radioactively labeled. This needs to be controlled and optimized for different cell lines. 3.3.3 Dot Blot Analysis of Glucosyland Galactosylceramide

Should there be a need to verify the identity of glycosphingolipid spots on the TLC plate, a dot blot analysis can be performed. 1. The TLC plate is stained with iodine vapor (see Note 3). 2. The spots corresponding to the lipid to be analyzed are scraped into a glass tube. 3. The lipids are extracted from the scraped silica using chloroform/methanol (2:1, v/v) and dried with nitrogen gas, and redissolved in hexane/isopropanol (3:2, v/v). 4. Appropriate amounts are dotted onto a nitrocellulose membrane alongside GlcCer and GalCer standards (2 nmoL/dot). 5. The lipids are then visualized by immunoblotting using primary mouse anti-GlcCer and anti-GalCer antibodies (dilution

Glycosphingolipid Biosynthesis and Transfer Protein Activity

113

1:100–1:500), incubated for 1 h at 37  C [4], and HRP-conjugated mouse secondary antibody (dilution 1:5000). Enhanced chemiluminescence detection kit was used together with standard X-ray film.

4

Notes 1. Since TLC plates vary in quality and the solvent polarity can differ from time to time, standards must be included in each run. The standards are run in parallel with the samples, preferably on both sides of the samples. 2. Heterogeneity in the composition of the fatty acid moieties and the Gb3 head group structure (its isomer isoglobo-Gb3) result in double bands in the HPTLC separation. Both bands should be scraped and analyzed and termed as Gb3. 3. Orcinol spray can be used for carbohydrate detection [16]. The silica plates are sprayed with the orcinol solution in the fume hood and heated at 120  C for 15 min. Carbohydratecontaining lipids appear violet against a white silica background. 4. Visualization of general lipid migration on silica plates can be done using an iodine chamber or with primuline spray. For a primuline spray stock, dissolve 100 mg of primuline in 100 mL of water. Before use, dilute 1 mL of the stock solution with 100 mL of acetone/water (4:1, v/v). Lipids appear as yellow spots under UV340nm. 5. Copper-acetate spray solution containing 3% (w/v) copper acetate in 8% phosphoric acid can also be used to visualize total lipids by heating the plate until the spots become charred. Staining with 10% (w/v) cupric (II) sulfate in 8% phosphoric acid is more independent of the fatty acid unsaturation compared to the 3% copper acetate stain [17].

Acknowledgments This work was supported by Svenska Kulturfonden, Sigrid Juse´lius Foundation, Magnus Ehrnrooth Foundation, Medicinska Under¨ sterbottniska Samfundet. sto¨dsfo¨reningen Liv och H€alsa, Svensk O References 1. Kjellberg MA, Backman APE, Mo¨uts A et al (2017) Purification and validation of lipid transfer proteins. Methods Mol Biol 1609:231–239

2. Mattjus P (2016) Specificity of the mammalian glycolipid transfer proteins. Chem Phys Lipids 194:72–78

114

Anders P. E. Backman et al.

3. Tuuf J, Mattjus P (2014) Membranes and mammalian glycolipid transferring proteins. Chem Phys Lipids 178:27–37 4. Kjellberg MA, Lo¨nnfors M, Slotte JP et al (2015) Metabolic conversion of ceramides in HeLa cells—a cholesteryl phosphocholine delivery approach. PLoS One 10:e0143385 5. Kjellberg MA, Backman AP, Ohvo-Rekil€a H et al (2014) Alternation in the glycolipid transfer protein expression causes changes in the cellular lipidome. PLoS One 9:e97263 6. Kjellberg MA, Mattjus P (2013) Glycolipid transfer protein expression is affected by glycosphingolipid synthesis. PLoS One 8:e70283 7. Tuuf J, Wistbacka L, Mattjus P (2009) The glycolipid transfer protein interacts with the vesicle-associated membrane proteinassociated protein VAP-A. Biochem Biophys Res Commun 388:395–399 8. Tuuf J, Mattjus P (2007) Human glycolipid transfer protein-intracellular localization and effects on the sphingolipid synthesis. Biochim Biophys Acta 1771:1353–1363 9. Fuchs B, Suss R, Teuber K et al (2011) Lipid analysis by thin-layer chromatography-a review of the current state. J Chromatogr A 1218:2754–2774 10. Byun HS, Bittman R (2018) Chemical preparation of sphingosine and sphingolipids: a review of enantioselective synthesis. In: Cevs

G (ed) Phospholipids handbook. CRC Press, Boca Raton, FL, pp 97–141 11. Lutzke BS, Braughler JM (1990) An improved method for the identification and quantitation of biological lipids by HPLC using laser lightscattering detection. J Lipid Res 31:2127–2130 12. Schwarzmann G (1978) A simple and novel method for tritium labeling of gangliosides and other sphingolipids. Biochim Biophys Acta 529:106–114 13. Lo¨nnfors M, La˚ngvik O, Bjo¨rkbom A et al (2013) Cholesteryl phosphocholine—a study on its interactions with ceramides and other membrane lipids. Langmuir 29:2319–2329 14. Sukumaran P, Lo¨nnfors M, La˚ngvik O et al (2013) Complexation of c6-ceramide with cholesteryl phosphocholine—a potent solvent-free ceramide delivery formulation for cells in culture. PLoS One 8:e61290 15. Lowry OH, Rosebrough NJ, Farr AL et al (1951) Protein measurement with the Folin phenol reagent. J Biol Chem 193:265–275 16. Irwin M, Leaver AG (1956) Use of the orcinolsulphuric acid reaction in the positive identification of certain monosaccharides from a salivary mucoid. Nature 177:1126 17. Baron CB, Coburn RF (1984) Comparison of two copper reagents for detection of saturated and unsaturated neutral lipids by charring densitometry. J Liq Chromatogr 7:2793–2801

Chapter 10 Measurement of Intracellular Sterol Transport in Yeast Neha Chauhan, Julian A. Jentsch, and Anant K. Menon Abstract Intracellular sterol transport occurs largely by non-vesicular mechanisms in which sterol transport proteins extract sterol from one membrane and transfer it to another across the cytoplasm. Here we describe a suite of complementary assays to measure intracellular sterol transport in the model eukaryote Saccharomyces cerevisiae, as well as to quantify protein-mediated sterol transport between populations of vesicles in vitro. The in vivo assays can be adapted to study sterol transport in other cell types. Key words Acyl-CoA:sterol acyltransferase (ACAT), Cholesterol, Dehydroergosterol, Ergosterol, Fluorescence microscopy, Fo¨rster resonance energy transfer (FRET), Methyl-β-cyclodextrin, Liposome, Lipid droplets, Plasma membrane, Endoplasmic reticulum, Subcellular fractionation, Detergent-resistant membranes

1

Introduction Sterols, e.g., cholesterol in mammals and ergosterol in yeast (Fig. 1), play a critical role in regulating the fluidity and barrier function of the plasma membrane (PM). The intracellular distribution of sterols is tightly controlled, such that the sterol content of the PM is high (one out of every 2–3 lipids in the PM is a sterol), whereas that of the endoplasmic reticulum (ER) membrane is 7–8fold lower [1]. Sterols are synthesized in the ER and/or imported to meet cellular needs, and transported rapidly within the cell to achieve homeostatic control of sterol content and to maintain the correct sterol distribution [2–4]. Intracellular sterol transport occurs mainly by non-vesicular mechanisms, requiring sterol transport proteins (STPs) [5] (Fig. 2). STPs are cytoplasmic proteins that pick up sterol at one membrane and off-load it at another; they may also be membrane-bound proteins that function at sites of membrane contact where, for example, the ER and PM [6] or the ER and endosomes [7] are closely apposed. Assays to measure intracellular sterol transport have become increasingly important as a variety of STPs have been identified and topographical

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_10, © Springer Science+Business Media, LLC, part of Springer Nature 2019

115

116

Neha Chauhan et al.

Fig. 1 Sterol structures. Mammalian cholesterol, yeast ergosterol, and fluorescent DHE are shown. DHE differs from ergosterol by one double bond (indicated in red). Both DHE and ergosterol can be detected by their UV absorbance making it convenient to monitor them, and their esters, by HPLC using an in-line UV monitor

Fig. 2 Intracellular sterol transport in yeast facilitated by sterol transport proteins (STPs). (Left panel) Exogenous sterols ([3H]cholesterol or DHE, depicted as a yellow oval) can be imported by yeast cells that express the plasma membrane (PM)-localized ABC transporters Aus1 and Pdr11. These transporters are expressed in the upc2-1 strain of yeast cells and also in cells that are incubated under hypoxic conditions. Transport of the imported sterols to the endoplasmic reticulum (ER) is monitored by their conversion to steryl esters, catalyzed by the ER-localized enzymes Are1 and Are2. The steryl esters are stored in lipid droplets (LD) that can be visualized in experiments with DHE. (Right panel) Newly synthesized ergosterol (generated as [3H]ergosterol (depicted as a green oval) by pulse labeling cells with [3H]methyl-methionine) in the ER is transported bidirectionally to the PM; arrival at the PM is detected by measuring the specific radioactivity of [3H]ergosterol in isolated PM (obtained via subcellular fractionation), methyl-β-cyclodextrin extracts of cells, or detergentresistant membranes. Both panels depict a sterol-loaded STP engaged in moving sterol between the ER and PM

connections between intracellular compartments, e.g., contact sites between the ER and PM, have been implicated in lipid exchange [6, 8–13]. Here we describe a suite of complementary assays to visualize and measure the transport of sterols—cholesterol, ergosterol, fluorescent dehydroergosterol (DHE) (Fig. 1)—between the PM and

Intracellular Sterol Transport

117

the ER in the budding yeast Saccharomyces cerevisiae, and to measure STP-facilitated transport of sterols between synthetic lipid vesicles in vitro. Although the in vivo assays are optimized for yeast cells, they can be readily adapted for studying intracellular transport of sterols in mammalian cells. The in vitro assay has broad applicability, enabling mechanistic characterization of STPs. The assays enable a quantitative analysis of intracellular sterol trafficking and provide a means to evaluate candidate elements of the currently poorly defined molecular machinery of sterol transport. In the following sections we present methods to assay the retrograde transport of exogenously supplied sterol (DHE or [3H]cholesterol) from the PM to the ER (including visualization of DHE transport by fluorescence microscopy), the exchange of newly synthesized ergosterol between the ER and PM (detected by three different methods), and the STP-catalyzed exchange of DHE between synthetic liposomes in vitro.

2

Materials Prepare all solutions using double distilled water and analytical grade reagents. Prepare and store all reagents at room temperature (unless indicated otherwise). Follow all waste disposal regulations when disposing waste materials.

2.1 Retrograde Transport of Sterols: Transport-Coupled Esterification of DHE 2.1.1 Loading and Chase of DHE from PM to ER in Yeast Cells

1. Complete synthetic medium (CSM): To prepare 1 L of medium, weigh 1.7 g of yeast nitrogen base (without ammonium sulfate and without amino acids, Sigma Aldrich), 0.79 g of CSM powder (Sunrise Science), 5 g of ammonium sulfate, and 20 g of glucose. Dissolve using a magnetic stirrer and autoclave at 121
C for 15 min [14]. 2. Saccharomyces cerevisiae wild-type and mutant strains to be analyzed. 3. DHE stock solution: Dissolve 5 mg of DHE (Ergosta-5,7,9 (11),22-tetraen-3β-ol; Sigma-Aldrich) in 1 mL of ethanol and store at 20
C (see Note 1). 4. Tween-80/ethanol (1:1, v/v) solution: Add 5 mL of Tween80 to 5 mL of ethanol in a 15 mL plastic screw cap tube. Dissolve by rotating the tube slowly on an end-over-end shaker until the solution is visibly homogenous. 5. Energy poison stock solution: Weigh 1.625 g of NaN3 and 1.05 g of NaF. Add 50 mL of water to prepare a 500 mM NaN3/NaF solution, vortex to dissolve, and store at 4
C (see Note 2).

118

Neha Chauhan et al.

6. Energy poison solution 1 (EP1): 10 mM NaN3/NaF. Add 2 mL of 500 mM NaN3/NaF (energy poison stock solution) to 98 mL of water. 7. Energy poison solution 2 (EP2): 10 mM NaN3/NaF with 0.5% (v/v) Tween-80. Add 2 mL of 500 mM NaN3/NaF (energy poison stock solution) to 97.5 mL of water. Add 500 μL of Tween-80 and dissolve with a magnetic stirrer. Store at 4
C (see Note 3). 8. GasPak™ EZ large incubation container (BD Diagnostics). 9. GasPak™ EZ anaerobe carbon sachets. 2.1.2 Live Cell Fluorescence Microscopy of DHE Redistribution

1. Concanavalin A (Con A)-coated dish: Prepare a solution of Con A in water (0.2% (w/v)) and load 100 μL onto the coverslip insert of a glass bottom dish (MatTek Corporation: 35 mm Dish/No. 1.0 Coverslip/10 mm Glass Diameter/ Uncoated or poly-D-lysine coated). Dry the liquid by incubating the dish at 60
C (this can also be done by incubating the dish at 37
C overnight). Rinse the coverslip a few times with water to remove excess Con A (use 500 μL of water each time). Dry the dish again at 60
C for 15 min. The Con A-coated dish can be made a day in advance and used to adhere yeast cells onto a coverslip/slide for any live microscopy experiment. 2. Leica 1 fluorescence microscope driven by the MetaMorph imaging software and equipped with a filter cube from Chroma Technology with a 355 nm (20 nm bandpass), 365 nm longpass dichromatic filter and 405 nm (40 nm bandpass) emission filter.

2.1.3 Extraction of Lipids from Yeast Cells to Quantify Sterols and Steryl Esters

1. Hexane. 2. Hexane/isopropanol (3:2, v/v). Make fresh. 3. Methanol/isopropanol (1:1, v/v). Make fresh. 4. Glass beads: Disruptor beads, 0.5 mm (Electron Microscopy Sciences) or similar. 5. BeadBug™ microfuge homogenizer (Sigma-Aldrich) or similar. 6. Glass Pasteur pipette. 7. 10 mL glass tubes with a Teflon-lined screw cap. 8. Reversed phase C18 column: Spherisorb ODS2 Column, 80 A˚, 5 μm, 4.6 mm  250 mm (Waters) or similar.

2.2 Retrograde Transport of Sterols: Transport-Coupled Esterification of [3H] cholesterol

1. upc2-1 yeast cells [10]. 2. Tween-80/ethanol (1:1, v/v) solution, energy poison stock solution, energy poison solutions (EP1 & EP2), solvents for lipid extraction: see Subheadings 2.1.1 and 2.1.3.

Intracellular Sterol Transport

119

3. Liquid scintillation cocktail (Research Products International (RPI) or similar). 4. Glass scintillation vials (7 mL). 5. Cholesterol master mix 1 (the recipe is for an assay with 8 time points). To prepare a 10 mM cholesterol stock solution, weigh 19.33 mg of cholesterol in a screw cap glass tube and add 5 mL of chloroform (the solution can be stored at 20
C for extended times). Combine 10 μL of 10 mM cholesterol solution and 70 μL of [3H]cholesterol (American Radiolabeled Chemicals; typically 60 Ci/mmoL; 1 mCi/mL in ethanol, stored at 20
C) in a screw cap glass tube and dry under a stream of nitrogen. Dissolve the dried cholesterol in 450 μL of Tween-80/ethanol (1:1, v/v) solution. Vortex the suspension and allow it to stand at room temperature for 15 min to ensure that the cholesterol is completely dissolved. Take an aliquot (10 μL ~ 2.25 nmoL cholesterol) for liquid scintillation counting. Use the counts obtained (A cpm) to calculate the specific radioactivity (SRMM) of the master mix as SRMM ¼ 0.445·A cpm/nmoL. 6. Pre-coated silica thin layer chromatography (TLC) plate, 20  20 cm (TLC silica gel 60, Merck or similar). 7. Glass tank for developing thin layer chromatogram. 8. Solvent system for developing TLC: Hexane/diethylether/ acetic acid (80:20:2, v/v/v). A volume of 150–200 mL is sufficient for a standard TLC tank. Prepare the solvent fresh and transfer it into the glass tank at least 30 min prior to resolving the TLC plate. 9. Radioactivity TLC analyzer (Raytest or similar). 2.3 Measurement of Acyl-CoA:sterol Acyltransferase (ACAT) Activity 2.3.1 Microsome Preparation

1. Reaction buffer: Dissolve 17.4 g of dipotassium hydrogen phosphate (K2HPO4) in 100 mL of water to obtain a 1 M solution. Dissolve 13.6 g of potassium dihydrogen phosphate (KH2PO4) in 100 mL of water to make a 1 M solution. Mix 80.2 mL of 1 M K2HPO4 and 19.8 mL of 1 M KH2PO4 solution to prepare 100 mL of 100 mM potassium phosphate buffer, pH 7.4. Add 1 mM reduced glutathione (3 mg in 10 mL) to 10 mL of this buffer (see Note 4). 2. Glass dounce homogenizer for 2 mL volumes (Sigma-Aldrich or similar). 3. Micro BCA protein estimation kit (Thermofisher Scientific or similar).

2.3.2 ACAT Assay

1. 10 mM Methyl-beta-cyclodextrin (MβCD) solution: Dissolve 13.3 mg of MβCD (MW ¼ 1330 g/moL) in 1 mL of reaction buffer (see Subheading 2.3.1).

120

Neha Chauhan et al.

2. Cholesterol master mix 2: Mix 10 μL of 10 mM cholesterol solution (see Subheading 2.2) and 12 μL of [3H]cholesterol (see Subheading 2.2) in a glass screw cap tube. Dry the lipids under a stream of nitrogen and resuspend in 100 μL of MβCD solution and vortex well. Heat the suspension at 60
C for 5 min to solubilize cholesterol. Cool to room temperature before use (see Note 5). 3. Fatty acid free bovine serum albumin (BSA) solution: Dissolve 20 mg of fatty acid-free BSA in 1 mL of water. 4. Oleoyl-CoA stock solution: Dissolve 5 mg of oleoyl-CoA (obtained as a lithium salt, Sigma-Aldrich) in 1 mL of water to obtain a 4.8 mM stock. Store at 20
C. 5. Solvents for lipid extraction: see Subheading 2.1.3. 2.4 Transport of Newly Synthesized Ergosterol Between the ER and PM 2.4.1 Pulse Labeling with [3H]methyl Methionine

1. Synthetic dropout medium minus methionine (SD-Met): Prepare the medium as described in Subheading 2.1.1 but instead of CSM powder, add 0.75 g of CSM powder without methionine (Sunrise Science or similar) [14]. 2. [3H]methyl methionine: Aliquot 1 mL of [3H]methyl methionine (American Radiolabeled Chemicals; 1 mCi/mL in ethanol) into a 50 mL plastic tube, dry under a stream of nitrogen, and dissolve in 10 mL of pre-warmed SD-Met medium. 3. Methionine 50 mg/mL: Dissolve 50 mg of methionine in autoclaved distilled water. Filter the solution using a 0.2 μm syringe filter. Store at 4
C.

2.4.2 Subcellular Fractionation

1. Ultra-clear, single-use tubes (13  51 mm, Beckman Coulter). 2. 500 mM NaN3/NaF solution: see Subheading 2.1.1. 3. Gradient buffer: 10 mM HEPES-KOH, pH 7.2, 1 mM EDTA, 0.8 M Sorbitol. To prepare 100 mL buffer, add 10 mL of 100 mM HEPES-KOH buffer (pH 7.2), 200 μL of 500 mM EDTA (see Note 15), and 14.57 g of sorbitol in 60 mL of distilled water. Dissolve the sorbitol and make up the solution to 100 mL with distilled water. 4. 60% (w/w) sucrose solution: Dissolve 120 g of sucrose in 80 mL of gradient buffer. Autoclave the suspension at 121
C for 15 min to dissolve the sucrose. 5. Breaking buffer: 10% sucrose solution (diluted v/v from 60% sucrose solution above), EP1, 1 mM EDTA, 10 mM HEPESKOH buffer, pH 7.2 (see above), 1 protease inhibitor cocktail (Calbiochem or similar). 6. Milk powder 2.5% (w/v): Weigh 0.5 g of fat-free milk powder and dissolve in 20 mL of Tris buffer saline (or phosphate buffer saline) plus 0.1% (v/v) Tween 20.

Intracellular Sterol Transport

121

7. Anti-Pma1 antibody: PM marker (Abcam) 1:10,000 dilution in 2.5% milk powder . 8. Anti-Dpm1 antibody: ER marker (Abcam) 1:500 dilution in 2.5% milk powder. 9. Mouse monoclonal HRP-conjugated secondary antibody (Abcam), 1:10,000 dilution in 2.5% milk powder. 10. Solvents for lipid extraction: see Subheading 2.1.3. 11. 95% (v/v) methanol in distilled water. 12. 10 mM ergosterol standard: Dissolve 19.8 mg of ergosterol in 5 mL of 95% (v/v) methanol in distilled water. Dilute to 100 μM and use it to prepare ergosterol standards in the range of 1–100 μM. 2.4.3 Methyl-β-Cyclodextrin (MβCD) Extraction of Ergosterol

1. MβCD extraction buffer: Prepare a 40 mM solution of MβCD by dissolving 266 mg of MβCD in 5 mL of EP1 (see Subheading 2.1.1). Prepare fresh. 2. 0.2 μm syringe filters. 3. Solvents for lipid extraction: see Subheading 2.1.3. 4. 95% methanol: see Subheading 2.4.2.

2.4.4 Ergosterol in Detergent-Resistant Membranes (DRMs)

1. Resuspension buffer: 100 mM Tris–HCl, pH 7.5, 10 mM EDTA supplemented with protease inhibitor cocktail (Calbiochem or similar). 2. 2% (w/v) Triton X-100 in water. Prepare the solution with distilled water in a 15 mL tube and mix by using an end-overend shaker. Store at 4
C. 3. Solvents for lipid extraction: see Subheading 2.1.3. 4. 95% methanol: see Subheading 2.4.2.

2.5 Analysis of Sterol Transport In Vitro 2.5.1 Liposome Preparation

1. Liposome buffer: 20 mM PIPES-NaOH, pH 6.8, 3 mM KCl, 10 mM NaCl. Store at 4
C (see Note 6). 2. Lipid stock solutions (Avanti Polar Lipids): 1,2-dioleoyl-snglycero-3-phosphocholine (DOPC, 25 mg/mL in chloroform, 31.8 mM), Dansyl-PE (1 mg/mL in chloroform, 1 mM), and DHE (5 mg/mL in chloroform, 12.67 mM). Other lipids obtained as chloroform stocks: 1,2-dioleoyl-sn-glycero-3phosphoethanolamine (DOPE, 25 mg/mL, 33.6 mM), 1,2-dioleoyl-sn-glycero-3-phospho-L-serine (DOPS, 10 mg/ mL, 12.3 mM), Liver L-α-lysophosphatidylinositol (10 mg/ mL, 10.1 mM). 3. 50–100 μL gas-tight glass syringes (e.g., Hamilton) to measure and aliquot the chloroform stocks of lipids mentioned above. 4. Mini Hand Extruder (Avanti Polar Lipids) with two 1 mL gas-tight glass syringes.

122

Neha Chauhan et al.

5. Polycarbonate membranes with 100–200 nm pore size. 6. Membrane supports. 2.5.2 Quantification of Phospholipids

1. Phosphate standard 1: Weigh 0.54 g of Na2HPO4∙7H2O and dissolve in 50 mL of distilled water to prepare a 40 mM solution. Store at 4
C or in aliquots at 20
C. 2. Phosphate standard 2: 4 mM Na2HPO4∙7H2O. Dilute 5 mL of Phosphate standard 1 in 45 mL of distilled water. 3. Phosphate standard 3: 0.4 mM Na2HPO4∙7H2O. Dilute 5 mL of Phosphate standard 2 in 45 mL of distilled water. 4. Perchloric acid (70–72%). 5. 12 mg/mL ammonium molybdate tetrahydrate in distilled water (prepare fresh). 6. 50 mg/mL ascorbic acid in distilled water (prepare fresh).

2.5.3 FRET-Based In Vitro Sterol Transport Assay

1. Quartz cuvette (2 mL) for fluorescence measurements. 2. Magnetic stir bar (flea) that fits into the cuvette. 3. Fluorescence spectrometer equipped with a magnetic stirrer.

3

Methods

3.1 Retrograde Transport of Sterols: Transport-Coupled Esterification of DHE

Saccharomyces cerevisiae does not take up exogenous sterol under aerobic growth conditions; this phenomenon, termed “aerobic sterol exclusion” [15], can be bypassed by growing cells under hypoxic conditions. To measure retrograde transport, DHE is loaded into the PM of yeast cells under hypoxia, conditions under which the sterol import machinery in the PM is expressed and synthesis of endogenous ergosterol is blocked. The cells are chased under aerobic conditions, which allow de novo ergosterol synthesis. During the chase, newly synthesized ergosterol moves to the PM, thereby displacing DHE which is transported to the ER where it is converted to DHE esters by ER-localized sterol acyltransferases (Fig. 2). The movement of DHE from the PM to the ER can be visualized by fluorescence microscopy as a change in the pattern of cell-associated fluorescence from a peripheral ring (PM) to several dots (lipid droplets (LD)) and quantified by organic solvent extraction followed by HPLC analysis to determine the percentage conversion of DHE to DHE esters. Note that visualization of DHE distribution in cells by fluorescence microscopy requires specialized equipment and cannot be done with routine microscopes. Quantification of DHE esterification by HPLC is more readily achieved. Data obtained with this assay are reported in, e.g., references [6, 12, 16, 17] .

Intracellular Sterol Transport 3.1.1 Loading and Chase of DHE from the PM to the ER in Yeast Cells

123

1. Inoculate CSM (see Subheading 2.1.1) [14] (5 mL in 50 mL plastic tubes) at an OD600 ¼ 0.005 from a saturated (OD600 ~ 10) overnight liquid culture of your desired Saccharomyces cerevisiae strain (see Note 7). Add 20 μL of DHE stock solution (final concentration 20 μg/mL DHE) and 100 μL of Tween-80/ethanol solution as a source of fatty acid. Always use an isogenic yeast wild-type strain as a control. It is important to note that the experiment involves taking different time points to monitor the rate of conversion of DHE to DHE ester during aerobic chase. Each time point requires 5 mL of yeast culture, corresponding to a separate inoculation. 2. Culture the yeast cells under hypoxic conditions to ensure DHE loading into the PM (see Note 8). For this, an anaerobic incubation chamber is used (GasPak™ EZ large incubation container). Place the tubes in tube racks (the lid of the tube should be loose to allow exchange of air) and place the rack in the anaerobic incubation chamber. Place three GasPak™ EZ anaerobe carbon sachets in the chamber, seal the chamber, and transfer to a 30
C incubator for 36 h. 3. At the end of the hypoxic incubation, open the chamber and immediately pellet the cells by centrifugation at 3000  g for 5 min at 4
C. Resuspend the cells in 5 mL of CSM supplemented with 100 μL of Tween-80/ethanol, and incubate at 30
C with constant shaking at 180 rpm. Harvest the cells after 0, 60, 90, 120, 180, and 240 min of aerobic growth as in the next step (see Note 9). 4. At the end of each chase point, harvest the cells by adding 5 mL of ice-cold EP2 (energy poisons to stop transport). Pellet the cells by centrifugation at 4000  g for 2 min at 4
C. Repeat this step three times. Perform three more washes with 5 mL of ice-cold EP1. For the 0 min time point, aerobic chase is not required; hence, remove the medium immediately after the 36 h hypoxic incubation and process the cells with energy poisons (see Note 10). 5. After washing, resuspend the cell pellet in 1 mL of ice-cold EP1 and transfer to a 1.5 mL screw cap tube. Take a 50 μL aliquot of the cell suspension for fluorescence microscopy. Pellet the remaining cells by centrifugation at 4000  g for 2 min and store in a 1.5 mL screw cap plastic tube at 80
C until lipid extraction.

3.1.2 Live Cell Fluorescence Microscopy of DHE Redistribution

1. Load 50 μL of cell suspension from each time point onto the Con A-coated coverslip in a culture dish and incubate at 4
C for 10 min to allow the cells to adhere. 2. Aspirate the liquid using a pipette and add 200 μL of ice-cold EP1 over the attached cells. 3. Visualize the cells under the microscope to image DHE with a 63 objective.

124

Neha Chauhan et al.

3.1.3 Extraction of Lipids from Yeast Cells for DHE and DHE Ester Determination

1. Resuspend the aliquot of frozen cells (step 5 in Subheading 3.1.1) in 1 mL of hexane/isopropanol (3:2, v/v). Add 500 μL of glass beads to the suspension and lyse the cells using a BeadBug microfuge homogenizer (4  30 s cycles) (see Note 11). 2. Using a glass Pasteur pipette transfer the sample into a 10 mL glass tube with a teflon-lined screw cap. Wash the glass beads with 2 mL of hexane/isopropanol (3:2, v/v) and pool with the original sample in the glass tube. Incubate the tubes at room temperature for 1 h in an end-over-end shaker. 3. At the end of the incubation, add 500 μL of distilled water to the samples, briefly vortex, and centrifuge the sample at 1500  g for 5 min. 4. Transfer the upper (organic) phase into a fresh tube using a glass Pasteur pipette. Add 2 mL of hexane to the lower phase for re-extraction. Briefly vortex, centrifuge the tubes at 1500  g for 5 min, and pool the organic phases together. Dry the lipid extract under a stream of nitrogen (see Note 12). 5. Resuspend the dried lipids in 400 μL of methanol/isopropanol (1:1, v/v). For HPLC analysis, inject 100 μL of the lipid sample onto a reversed phase C18 column with methanol/isopropanol (1:1, v/v) as the solvent system. Set the flow rate to 1 mL/min. DHE (detected with an in-line UV monitor set at a wavelength of 325 nm) elutes at ~4.5 min and DHE esters elute later as two predominant peaks at ~10–12 min. These retention times are sensitive to the solvent composition and temperature. Represent the data as the percentage of DHE esters (integrated intensity of ester peaks  sum of integrated intensities of the DHE peak plus DHE ester peaks).

3.2 Retrograde Transport of Sterols: Transport-Coupled Esterification of [3H] cholesterol

Transport of exogenously supplied [3H]cholesterol from the PM to the ER in yeast is quantified by the synthesis of [3H]cholesteryl esters, as assessed by thin layer chromatography of lipids extracted from cells. For this assay, the yeast cells must express the Upc2-1 protein, a constitutively active variant of the transcription factor Upc2 that enables the cells to bypass “aerobic sterol exclusion” by constitutively expressing the Aus1 and Pdr11 transporters [18] at the PM (Fig. 2) (upc2-1 yeast cells are described in reference [10]). Although this assay is easier to execute than the DHE uptake assay, here the cells must express Upc2-1 whereas the DHE assay can be performed with any yeast strain. Also, this assay requires working with radioactivity whereas the DHE uptake assay does not. Data obtained with this assay are reported in, e.g., references [6, 10, 12, 16].

Intracellular Sterol Transport 3.2.1 Transport of [3H] cholesterol from the PM to the ER in upc2-1 Yeast Cells

125

1. Grow upc2-1 yeast cells at 30
C overnight in appropriate medium to obtain a saturated culture. Dilute the pre-culture (OD600 of 0.05–0.1 for YPD or OD600 of 0.25 for complete synthetic media) and incubate for 3–4 h till the OD600 ¼ 0.6–0.8. 2. The experiment involves four time points, each requiring 20 OD600 units of cells. Pellet cells in a 50 mL plastic culture tube by centrifugation at 3000  g for 5 min. Resuspend 20 OD600 units of cells in 5 mL of medium, add 50 μL of cholesterol master mix 1, and grow the cells at 30
C. 3. At the end of each time point (10, 20, 40 and 60 min) process the cells with energy poison as described in step 4 in Subheading 3.1.1. 4. Resuspend the cells in 1 mL of ice-cold EP1 and determine the OD600 (to be used for normalizing the data). Re-pellet the cells and store at 80
C until lipid extraction.

3.2.2 Extraction of Lipids from Yeast Cells and Quantification of the Conversion of Cholesterol to Cholesteryl Ester

1. The lipid extraction protocol is the same as described for DHE in Subheading 3.1.3. Once extraction is complete, dissolve the dried lipids in 300 μL of 1:1 (v/v) methanol/isopropanol and take an aliquot (20 μL) of the lipid extract for liquid scintillation counting. 2. Dry the remaining 280 μL of lipid extract under a stream of nitrogen and dissolve in 50 μL of methanol. Carefully spot the entire 50 μL on a pre-coated silica TLC plate using glass capillary tubes. Place the spots 2 cm apart on a 20  20 cm plate. 3. Pour hexane/diethylether/acetic acid (80:20:2, v/v/v) solvent system into a rectangular, glass TLC tank containing a large square of filter paper propped against the back wall to enable faster saturation of the tank with solvent. Cover the tank with a lid and equilibrate for (at least) 30 min before starting chromatography. Resolve the TLC plate. 4. Air-dry the TLC plate and visualize the chromatogram using a radioactivity TLC analyzer (if a detector is not available, the silica may be scraped at 1 cm intervals and taken for liquid scintillation counting to obtain the radio-chromatogram). 5. From the chromatogram determine the relative amounts of cholesterol and cholesteryl ester (Rf values: cholesterol ~0.10, cholesteryl ester ~0.75) as the area under each peak. Using the [3H] content of the lipid extract (step 1 above), the specific radioactivity (SRMM) of the cholesterol master mix 1, and the OD600 of the cells in each harvested aliquot, compute the amount of [3H]cholesteryl esters as pmoL/OD600.

126

Neha Chauhan et al.

3.3 Measurement of Acyl-CoA: Sterol Acyltransferase Activity

3.3.1 Microsome Preparation

Saccharomyces cerevisiae has two acyl-CoA-dependent sterol acyltransferases encoded by ARE1 and ARE2 [19]. Both assays of retrograde sterol transport described above rely on sterol esterification. Hence, as a control it is important to assay ACAT activity and ensure that the retrograde sterol transport readout between different strains/conditions is not due to alterations in acyltransferase activity. 1. Reinoculate cells in CSM (see Subheading 2.1.1) at OD600 ¼ 0.2 from a saturated overnight culture. Grow the cells at 30
C till the OD600 reaches 1. Harvest 100 OD600 units of cells by centrifugation at 3000  g for 5 min. Wash the cells twice with water and re-pellet. 2. Resuspend the cell pellet in 1 mL of reaction buffer, add 500 μL of glass beads and lyse the cells using a BeadBug homogenizer (4  30 s with 30 s on ice, see Note 11). 3. Centrifuge the cell lysate at 3000  g for 5 min at 4
C to pellet the glass beads and cell debris. Transfer the supernatant into fresh tubes and re-pellet at 10,000  g for 5 min. 4. Pellet the microsomes (ER) by ultra-centrifugation of the supernatant at 100,000  g for 60 min. 5. Remove the supernatant and homogenize the membrane pellet in 500 μL of reaction buffer using a small dounce homogenizer. Determine the protein concentration of the homogenate by the Micro BCA method as per the manufacturers’ instructions. Freeze the sample in liquid nitrogen and store at 80
C.

3.3.2 ACAT Assay

1. Mix 100 μL of 1 mM cholesterol master mix 2 (100 nmol) with 40 μL of fatty acid-free BSA (1 mg) and 200 μg of microsomal protein. Make up to 800 μL with reaction buffer. 2. Incubate the reaction mix at 30
C for 15 min before adding 20 nmol of oleoyl-CoA (4.2 μL of the stock solution) to initiate esterification. 3. At the end of each time point (0, 15 and 30 min) remove a 250 μL aliquot (see Note 13) and extract lipids as described in Subheading 3.1.3. 4. Resolve the extracted lipids by TLC and determine the relative amount of [3H]cholesterol and [3H]cholesteryl ester using the TLC analyzer, or by scraping the relevant areas of silica (see details in Subheading 3.2.2). 5. The fraction ( f ) of total radioactivity corresponding to [3H] cholesteryl ester (i.e., f ¼ ([3H]cholesteryl ester)/(([3H]cholesteryl ester) + ([3H]cholesterol)) increases linearly over time during the assay. ACAT activity (nmoles of sterol esterified per mg protein per min) is calculated as 100 nmol/0.2 mg multiplied by the rate of increase of f as a function of time (see Note 14).

Intracellular Sterol Transport

3.4 Transport of Newly Synthesized Ergosterol Between the ER and PM

3.4.1 Pulse Labeling with [3H]methyl Methionine

127

Transport of de novo synthesized ergosterol from the ER to the PM is measured by briefly labeling yeast cells with [3H]methylmethionine to generate a pulse of radiolabeled ergosterol in the ER, chasing with unlabeled methionine for 0–60 min, and detecting arrival of [3H]ergosterol in the PM by one of three methods: (1) subcellular fractionation to isolate a PM fraction, (2) MβCD extraction to sample ergosterol in the outer leaflet of the PM, and (3) isolation of detergent-resistant membranes (DRMs) which provide an approximation of the sterol content of the PM. In each case, [3H]ergosterol in the sample is extracted in organic solvents and the specific radioactivity (SR; ratio of [3H]ergosterol to unlabeled ergosterol) is determined by HPLC analysis and liquid scintillation counting. This value is divided by the specific radioactivity of ergosterol in total cells/unfractionated extract to obtain a relative specific radioactivity or RSR. Data from such assays are reported in several publications, e.g., references [9, 10, 13, 16]. 1. Inoculate 0.25 OD600 units of cells from a saturated, overnight liquid culture of yeast cells (OD600 ~ 10) into 250 mL of CSM and grow at 30
C until OD600 ¼ 1. 2. Harvest the cells by centrifugation at 3000  g for 5 min and wash with 40 mL of pre-warmed synthetic dropout medium minus methionine (SD-Met). 3. Resuspend 200 OD600 unit cells in 10 mL of pre-warmed SD-Met medium containing 1 mCi of [3H]methyl methionine and pulse label for 4 min at 30
C. 4. After the pulse, harvest 100 OD600 units for the 0 min time point by transferring half of the culture to a fresh 50 mL tube. 5. Add 5 mL of EP2 solution to stop the transport and process the sample as in step 4 in Subheading 3.1.1. 6. Transfer the remaining 100 OD600 unit cells to 40 mL of pre-warmed SD medium containing 20 mg/mL of unlabeled methionine (SD + Met). 7. Pellet the cells by centrifugation at 3000  g for 5 min. 8. Resuspend the cells in 5 mL of SD + Met medium and chase for 60 min at 30
C (more time points may be included as required). It is important to include the processing time to the total chase time. 9. After the chase is complete, process the sample with energy poisons as described in step 4 in Subheading 3.1.1.

3.4.2 Transport of Ergosterol from the ER to the PM Assayed by Subcellular Fractionation

1. Resuspend the cells from each chase point (100 OD600 units) in 1 mL of breaking buffer, add 500 μL of glass beads to the suspension and disrupt the cells using a BeadBug homogenizer with 4  30 s cycles (see Note 11).

128

Neha Chauhan et al.

Table 1 Preparation of sucrose solution Sucrose solution

60%

40%

37%

34%

32%

29%

27%

22%

60% (v/v) sucrose solution (mL)

24

16

14.8

13.6

12.8

11.6

10.8

8.8

Gradient buffer (mL)

0

8

9.2

10.4

11.2

12.4

13.2

15.2

Amount added to gradient (mL)

0.2

0.4

0.4

0.6

1

1

0.6

0.6

2. Centrifuge the lysate at 3000  g for 10 min at 4
C and carefully remove the supernatant into a fresh tube. Re-centrifuge at 3000  g for 5 min and collect the supernatant (see Note 16). 3. Prepare sucrose solutions for the discontinuous gradient used for subcellular fractionation (see Table 1). 4. Start preparing the gradient by slowly adding 60% sucrose solution to the bottom of an ultra-clear, single-use tube. Volumes to add are indicated in Table 1. Next, add the 40% sucrose layer by placing the pipette tip as close to the 60% sucrose layer as possible (without touching it) and slowly layering on the 40% sucrose solution. Do not add the sucrose solution against the walls of the tube. Continue adding layers as indicated in Table 1. You can prepare the gradients a few hours in advance and store at 4
C. 5. Carefully load 500 μL of the supernatant on top of the gradient and centrifuge at 160,000  g for 16 h at 4
C in a Beckman Coulter MLS50 rotor. After centrifugation, carefully collect seven fractions of 750 μL each from the top without disturbing the gradient (see Note 17). 6. For analysis by immunoblot, dilute 20 μL of each fraction ten-fold in 10 mM HEPES-KOH buffer (see Subheading 2.4.2) and SDS-PAGE loading dye. Resolve 10 μL of the diluted sample on a 10% SDS-PAGE gel, transfer the proteins to a nitrocellulose membrane, and immunoblot for PM and ER markers. A typical analysis is performed with antibodies against Pma1 (PM marker) and Dpm1 (ER marker) (see Subheading 2.4.2 for dilution and secondary antibody details). Typically, fraction 2 contains ER and fraction 7 contains PM. 7. Use the remainder of each fraction for lipid extraction as described in Subheading 3.1.3. Dissolve the extracted lipids in 100 μL of 95% (v/v) methanol in distilled water and resolve by HPLC (step 6 in Subheading 3.1.3) in the same solvent. With a flow rate of 1 mL/min, ergosterol (detected with an in-line UV monitor set at 280 nm) elutes at 23 min and ergosteryl esters elute at 28 min. As the retention times are

Intracellular Sterol Transport

129

sensitive to temperature and solvent it is advisable to run an ergosterol standard prior to analysis. Calibrate the detector response to ergosterol by measuring a series of standards over a linear response range (analyze 100 μL each of 1 μM to 100 μM ergosterol solutions). 8. Collect the ergosterol and ergosteryl ester fractions directly into 7-mL glass vials that can be used for liquid scintillation counting, dry under a stream of nitrogen, add liquid scintillation cocktail, and determine [3H] counts per min (cpm). 9. Determine the specific radioactivity (SR ¼ cpm  absorbance units (AU)) of ergosterol for ER and PM fractions as well as for whole cells; calculate the relative specific radioactivity (RSRfrac ¼ SRfrac  SRcell) of the sucrose gradient fractions. 3.4.3 Transport of Ergosterol from the ER to the PM Assayed by MβCD Extraction

1. Perform the pulse-chase labeling with [3H]methyl methionine as in Subheading 3.4.1. 2. Harvest 200 OD600 units of cells after 60 min chase and resuspend into 2 mL of ice-cold MβCD extraction buffer (see Note 18). Incubate the cells on ice for 30 min and every 5 min mix the suspension by inversion. 3. Centrifuge the cells at 10,000  g for 10 min at 4
C and carefully collect 1.8 mL of the supernatant with the MβCDergosterol complexes into a fresh tube. 4. Wash the pellet with 500 μL of EP1 solution, re-pellet the cells, suspend the pellet in 1 mL of water, and note the OD600. Store the cell pellet at 80
C till lipid extraction. 5. Pass the 1.8 mL of supernatant with the MβCD-ergosterol complexes through a 0.2 μm syringe filter to ensure removal of all intact yeast cells (see Note 19). Store the supernatant at 80
C until lipid extraction. 6. Extract lipids from the cell pellet as described in Subheading 3.1.3. Resuspend the dried lipids in 1 mL of 95% (v/v) methanol in distilled water and use 100 μL for HPLC analysis. 7. To extract lipids from 1.8 mL of supernatant containing MßCD-ergosterol complexes, add 9 mL of hexane/isopropanol (3:2, v/v) and follow the usual lipid extraction procedure as in Subheading 3.1.3. Resuspend the dried lipid extracts in 120 μL of 95% methanol and use 100 μL for analysis by reversed phase HPLC. 8. Resolve the lipids by HPLC as described in step 7 in Subheading 3.4.2. 9. Collect the ergosterol and ergosteryl ester fractions, dry under a stream of nitrogen, and resuspend in liquid scintillation cocktail to determine the [3H] cpm.

130

Neha Chauhan et al.

10. Express the data obtained from HPLC and scintillation counting as SR (cpm/AU). Thus, calculate the RSR for MβCD extract as RSRMβCD ¼ SRMβCD  SRcell. RSRMβCD directly compares the specific radioactivity of PM-localized ergosterol extracted by MβCD with that of total cellular ergosterol. The RSR calculation also provides a simple way of consolidating data from different experiments where the extent of metabolic labeling can vary. 3.4.4 Transport of Ergosterol from the ER to the PM Assayed by Isolating DetergentResistant Membranes (DRMs)

1. Use the protocol for pulse-chase labeling with [3H]methyl methionine (Subheading 3.4.1) and harvest the 60 min time point (or other time point of interest) by centrifugation at 3000  g for 5 min at 4
C. Resuspend the cells in 1 mL of ice-cold resuspension buffer. 2. Split the cells into two 1.5 mL screw cap plastic tubes each containing 100 OD600 units. Pellet the cells by centrifugation at 3000  g for 5 min at 4
C. Resuspend the cell pellet in 500 μL of ice-cold resuspension buffer with 500 μL of glass beads. Lyse the cells in the BeadBug homogenizer using a 4  30 s cycle with intermittent incubation on ice for 30 s (see Note 11). 3. Remove the unbroken cells and debris by centrifugation at 3000  g for 5 min at 4
C. Carefully transfer the supernatant to prechilled 1.5 mL plastic tubes and centrifuge again at 3000  g for 5 min at 4
C. Repeat the centrifugation step to ensure removal of cell debris. It is important to maintain the cells at 4
C for the extraction of DRMs. 4. Treat 500 μL of one cleared supernatant with detergent. To do this, add an equal volume of ice-cold resuspension buffer plus 2% (w/v) Triton X-100 and incubate on ice for 30 min. Store the supernatant of the second aliquot at 80
C until lipid extraction. 5. Collect DRMs by centrifugation for 30 min at 4
C in a precooled TLA 100.3 rotor at 200,000  g. Discard the supernatant, resuspend the pellet in 500 μL of ice-cold resuspension buffer, and store the DRMs at 80
C until lipid extraction. 6. Subject the cleared supernatant and the resuspended DRMs to lipid extraction (Subheading 3.1.3) and analysis by HPLC as described in steps 7 and 8 in Subheading 3.4.2. Calculate the RSR of the DRM fraction as RSRDRM ¼ SRDRM  SRsup, where SRsup is the specific radioactivity of the cleared supernatant (without Triton X-100).

3.5 Analysis of Sterol Transport In Vitro

The assay illustrated in Fig. 3 monitors the transfer of DHE between vesicles by Fo¨rster resonance energy transfer (FRET) using dansyl-phosphatidylethanolamine (dansyl-PE) as the FRET

Intracellular Sterol Transport

131

Fig. 3 In vitro assay for sterol transport. Donor vesicles containing DHE are mixed with acceptor vesicles containing dansyl-PE. Excitation of DHE at 310 nm results in FRET-mediated fluorescence emission from dansyl-PE at 525 nm when the two molecules are in the same vesicle. This occurs when DHE moves from donor to acceptor vesicles. The time-dependent increase in sensitized emission from dansyl-PE is used to measure the rate at which DHE moves from donor to acceptor vesicles, either spontaneously or catalyzed by a STP

acceptor. Donor vesicles containing DHE are mixed with acceptor vesicles containing dansyl-PE in a stirred quartz cuvette in a fluorescence spectrometer. Addition of STP accelerates transfer of DHE between vesicles, detected as sensitized emission from dansyl-PE. Examples of data generated with this assay are reported in references, e.g., [6, 17]. 3.5.1 Liposome Preparation

The protocol describes experiments with vesicles that have a simple composition: donor vesicles contain 80 mol% DOPC and 20 mol% DHE; acceptor vesicles contain 97 mol% DOPC and 3 mol% dansyl-PE. The composition of the vesicles can be varied as needed. Handle lipids with airtight glass syringes. The protocol listed is for a 2 mM lipid solution (see Note 20). Work at room temperature. Protect lipids from light by covering glass tubes with aluminum foil throughout the experiment unless otherwise specified. 1. Combine 50 μL of DOPC solution and 32 μL of DHE solution (see Subheading 2.5.1) in a 9 mL glass tube to make donor vesicles. 2. Combine 61 μL of DOPC solution and 60 μL of dansyl-PE solution (see Subheading 2.5.1) in a 9 mL glass tube to make acceptor vesicles. 3. Dry lipid mixtures under a nitrogen stream. Add 1 mL of liposome buffer to the lipid film and rehydrate lipids by constant orbital shaking at 2000 rpm in a Vibrax for 30 min (see Note 21). 4. Freeze the lipid suspension in liquid nitrogen and thaw by immersing the tube into room temperature water. Repeat the freeze-thaw step five times for each liposome suspension, making sure that the vesicles remain in liquid nitrogen for several minutes each time to ensure complete freezing. The lipid suspension should become milky-clear with a bluish appearance. 5. Assemble the mini-extruder according to the manufacturer’s instructions (see Note 22).

132

Neha Chauhan et al.

6. First extrusion step: Assemble the hand extruder with a 200 nm polycarbonate membrane. Pre-wet the extruder with 1 mL of liposome buffer to reduce the dead volume. Load one of the two syringes with the lipid suspension. Avoid air bubbles in the syringe. Place both syringes into the extruder. Pass 13 times through the 200 nm polycarbonate membrane (it is important to carry out an odd number of passages so that the final vesicle suspension is collected on the “clean” side of the apparatus). Take out the syringes and store the 200 nm liposome solution in a 9 mL glass tube on ice. Disassemble and wash the extruder and syringes with distilled water. 7. Second extrusion step: Assemble the hand extruder with a 100 nm polycarbonate membrane. Extrude the 200 nm liposome solution as described above. Store the final 100 nm liposome solution on ice or at 4
C. Use liposomes within a few days. 3.5.2 Quantification of Phospholipids

1. Prepare inorganic phosphate (Pi) standards: add Na2HPO4.7H2O to distilled water in glass tubes according to Table 2. 2. Prepare samples: Add 10 μL of liposomes (acceptor and donor individually) to 40 μL of distilled water in a glass tube. 3. Add 300 μL of perchloric acid to each glass tube and heat for 1 h at 145
C in a heating block (see Note 23). 4. Remove tubes from heating block. Add 1 mL of water and vortex. Let the tubes cool to room temperature. 5. Add 400 μL of ammonium molybdate tetrahydrate and 400 μL of ascorbic acid. Vortex well. 6. Heat the tubes for 10 min at 100
C in a heating block (see Note 23).

Table 2 Preparation of phosphate standards Sample

Pi (nmoL)

Phosphate standard 2 (μL)

Phosphate standard 1 (μL)

H2O (μL)

a

0

0

0

50

b

2

5

0

45

c

5

12.5

0

37.5

d

10

25

0

25

e

20

0

5

45

f

50

0

12.5

37.5

g

80

0

20

30

Intracellular Sterol Transport

133

7. Remove the tubes from the heating block and cool to room temperature. 8. Measure absorbance at 797 nm. Plot standard curve and interpolate unknowns (see Note 24). 3.5.3 FRET-Based In Vitro Sterol Transport Assay

1. Fluorimeter settings: excitation and emission wavelengths— 310–525 nm, respectively; slit widths, 0.5 mm (excitation) and 1.5 mm (emission), or as needed to obtain a reasonable signal; duration of the measurement—900 s or greater. 2. Calculate the volumes of each acceptor and donor liposome stock solution needed for a 100 μM final concentration in 2 mL of reaction volume. 3. Pipette the donor and acceptor liposomes into the stirred quartz cuvette and bring to a final volume of 1.8 mL with liposome buffer. 4. Prepare a 200 μL sample of 1 μM (or more, as needed) protein of interest or blank (only buffer) in a separate microfuge tube. The concentration of protein can be adjusted but the volume of buffer used for the “blank” sample should be the same as the volume of protein added to have the same dilution effect. 5. Place the cuvette with the 1.8 mL mixture of liposomes into the fluorimeter chamber. Set the stirrer to an appropriate speed. 6. Start the measurement. After 60 s, add the protein or buffer (200 μL) into the cuvette without opening the chamber lid or stopping the measurement (see Note 25). Monitor fluorescence for 900 s. It is important to determine experimentally the maximal FRET value (FRETmax) for a set of measurements by acquiring data for a sufficiently long period for at least some samples. 7. Analyze FRET data; the raw FRET data can be used to determine the rate of sterol transport between donor and acceptor vesicles. Correct the trace by subtracting the lowest value (value obtained right after addition of buffer or protein, corresponding to sample dilution) from each trace point. Fit the trace to a mono-exponential function: constrain the initial fluorescence value to zero and the plateau value to FRETmax. Obtain the transport rate constant k (s1) from the equation FRET signal (t) ¼ FRETmax (1ekt).

4

Notes 1. DHE is supplied as white powder in a brown bottle with a rubber seal. After adding ethanol, vortex well, then place the bottle on a heating block at 95
C for a few minutes to dissolve DHE.

134

Neha Chauhan et al.

2. Be very careful when weighing NaN3 and NaF; wear gloves and a facemask. Check with your Environment Health and Safety department to determine the correct disposal protocol. 3. Tween-80 is highly viscous, so cut off the tip of the pipette with a scissor for easy pipetting. To speed up the process, stir the 10 mM NaN3/NaF solution with a magnetic bead prior to adding Tween-80. 4. The 1 M stock solutions can be made in advance, but the reduced glutathione-containing potassium phosphate buffer should be prepared fresh. 5. After heating, allow the suspended lipids to stand at room temperature for 15 min. You will see some white precipitate on the walls of the glass tube. This is excess MβCD and is normal for this experiment. 6. PIPES does not dissolve at pH 50%) levels of DHE ester formation. 10. For washing the cells, keep the tubes on ice, add the appropriate ice-cold energy-poison solution, vortex briefly, and centrifuge immediately. Do not discard the waste down the sink. 11. BeadBug homogenizer is compatible only with v-bottom, 1.5 mL screw cap tubes. Cells can also be broken with glass beads by vortexing at maximum speed. Do at least 10 cycles of 30 s with 30 s intervals on ice. 12. Re-extraction with hexane does not require additional incubation. Do not pool the entire 3 mL organic phase for drying; instead, dry 1 mL at a time (adding a second quantity of 1 mL after the first amount has dried) to avoid drying the lipids along the walls of the glass tube. You can also use heat along with the nitrogen stream for faster drying.

Intracellular Sterol Transport

135

13. A single 30 min time point can also be performed to compare differences between different yeast deletion strains. 14. Use an acyltransferase deletion mutant (e.g., are1Δare2Δ in yeast) as a negative control [19]. 15. EDTA will not dissolve till the pH of the solution reaches 8. Set the pH to 8 with sodium hydroxide; make the solution to 100 mL with distilled water. Store at room temperature. 16. From 1 mL of the supernatant, collect 700 μL for lipid extraction and sucrose gradient fractionation to avoid any contamination from unbroken cells/debris. 17. Place the tubes on a stable surface and carefully collect fractions starting at the top. We use a 1 mL pipette for collecting the fractions: since the 1 mL pipette tip is broad, modify it by appending a small (10 μL) pipette tip to it. Take fresh tips for each fraction. 18. Do not resuspend by vortexing. Use a pipette to resuspend the cells gently in MβCD extraction buffer. 19. Pre-wet the syringe filter with water prior to filtering the MβCD extracted supernatant to avoid loss of sample. 20. The phospholipid composition of liposomes can be modified by using lipids listed in Subheading 2.5.1. Modifying lipid composition (as described in previous studies) might affect the sterol transport rate. Final lipid concentration may be adjusted according to the experimental set up. To avoid clogging of the polycarbonate membrane, keep the final lipid concentration lower than 4 mM and DHE concentration lower than 30 mol%. 21. For the rehydration step it is important that the lipids form a homogeneous suspension in solution. As an alternative to using a Vibrax, lipids can also be vortexed extensively. 22. https://avantilipids.com/divisions/equipment/mini-extruderassembly-instructions/. 23. Place marbles on tubes to prevent evaporation. All glass tubes have to be placed in the same heating block as the temperature may vary between two aluminum blocks in the same ensemble. 24. The standard liposome preparation should yield 2 μmol lipids in 1 mL, i.e., 2 mM. However, the lipid recovery is typically 60–80% (12–16 nmol per 10 μL). Note also that the phosphate assay readout for the donor liposomes must be corrected to account for their sterol content as sterols are not detected by this assay. 25. A drop in fluorescence will be observed due to dilution of the sample from 1.8 mL to 2 mL.

136

Neha Chauhan et al.

Acknowledgments This work was supported by grant NPRP 7-082-1-014 from the Qatar National Research Fund (A.K.M.) and Graduiertenkolleg 2098, Project 11, from the Deutsche Forschungsgemeinschaft (J-A.J., A.K.M.). References 1. Menon AK (2018) Sterol gradients in cells. Curr Opin Cell Biol 53:37–43 2. Holthuis JC, Menon AK (2014) Lipid landscapes and pipelines in membrane homeostasis. Nature 510:48–57 3. Maxfield FR, Menon AK (2006) Intracellular sterol transport and distribution. Curr Opin Cell Biol 18:379–385 4. Maxfield FR, van Meer G (2010) Cholesterol, the central lipid of mammalian cells. Curr Opin Cell Biol 22:422–429 5. Dittman JS, Menon AK (2017) Speed limits for nonvesicular intracellular sterol transport. Trends Biochem Sci 42:90–97 6. Quon E, Sere YY, Chauhan N et al (2018) Endoplasmic reticulum-plasma membrane contact sites integrate sterol and phospholipid regulation. PLoS Biol 16:e2003864 7. Wilhelm LP, Wendling C, Vedie B et al (2017) STARD3 mediates endoplasmic reticulum-toendosome cholesterol transport at membrane contact sites. EMBO J 36:1412–1433 8. Gatta AT, Wong LH, Sere YY et al (2015) A new family of StART domain proteins at membrane contact sites has a role in ER-PM sterol transport. elife 4:e07253 9. Mesmin B, Pipalia NH, Lund FW et al (2011) STARD4 abundance regulates sterol transport and sensing. Mol Biol Cell 22:4004–4015 10. Raychaudhuri S, Im YJ, Hurley JH et al (2006) Nonvesicular sterol movement from plasma membrane to ER requires oxysterol-binding protein-related proteins and phosphoinositides. J Cell Biol 173:107–119 11. Sullivan DP, Georgiev A, Menon AK (2009) Tritium suicide selection identifies proteins involved in the uptake and intracellular

transport of sterols in Saccharomyces cerevisiae. Eukaryot Cell 8:161–169 12. Georgiev AG, Sullivan DP, Kersting MC et al (2011) Osh proteins regulate membrane sterol organization but are not required for sterol movement between the ER and PM. Traffic 12:1341–1355 13. Gatta AT, Levine TP (2017) Piecing together the patchwork of contact sites. Trends Cell Biol 27:214–229 14. Hanscho M, Ruckerbauer DE, Chauhan N et al (2012) Nutritional requirements of the BY series of Saccharomyces cerevisiae strains for optimum growth. FEMS Yeast Res 12:796–808 15. Lorenz RT, Parks LW (1991) Involvement of heme components in sterol metabolism of Saccharomyces cerevisiae. Lipids 26:598–603 16. Georgiev AG, Johansen J, Ramanathan VD et al (2013) Arv1 regulates PM and ER membrane structure and homeostasis but is dispensable for intracellular sterol transport. Traffic 14:912–921 17. Roelants FM, Chauhan N, Muir A et al (2018) TOR Complex 2-regulated protein kinase Ypk1 controls sterol distribution by inhibiting StARkin domain-containing proteins located at plasma membrane-endoplasmic reticulum contact sites. Mol Biol Cell 29:2128–2136 18. Wilcox LJ, Balderes DA, Wharton B et al (2002) Transcriptional profiling identifies two members of the ATP-binding cassette transporter superfamily required for sterol uptake in yeast. J Biol Chem 277:32466–32472 19. Yang H, Bard M, Bruner DA et al (1996) Sterol esterification in yeast: a two-gene process. Science 272:1353–1356

Chapter 11 Intracellular and Plasma Membrane Cholesterol Labeling and Quantification Using Filipin and GFP-D4 Le´a P. Wilhelm, Laetitia Voilquin, Toshihide Kobayashi, Catherine Tomasetto, and Fabien Alpy Abstract Cholesterol, a major component of biological membranes, is rapidly trafficked and unevenly distributed between organelles. Anomalies of intracellular cholesterol distribution are the hallmark of a number of lysosomal lipid storage disorders. A major methodological obstacle for studying cholesterol trafficking is tracing this molecule in situ. The use of fluorescent probes that specifically bind cholesterol allows the visualization and imaging of cellular cholesterol. Here, we describe a series of assays optimized for quantifying free cholesterol in cell populations and at the single cell level, both at the plasma membrane and inside cells. These methods use two fluorescent probes: the D4 fragment of perfringolysin O fused to GFP (GFP-D4) and the polyene macrolide filipin. First, we report a robust method for quantifying plasma membrane cholesterol by flow cytometry using the GFP-D4 probe. Second, to optically distinguish and quantify intracellular cholesterol accumulation, we have adapted the classical filipin cholesterol staining protocol. Indeed, we observed that treatment of living cells with methyl-β-cyclodextrin, a chemical known to extract cholesterol from the plasma membrane, improves the visualization of the intracellular cholesterol pool with filipin. To complement these staining procedures, we developed an image analysis protocol based on image segmentation to quantify, in a robust manner, intracellular cholesterol stained with filipin. Thus, this chapter is a guideline for cellular cholesterol staining and signal quantification. Key words Cholesterol, Perfringolysin O, Filipin, Plasma membrane, Endosome

1

Introduction Cholesterol is a key molecule in biology, as a major structural component of cell membranes, and as a precursor of other vital compounds, such as hormones and bile [1, 2]. Interestingly, sterol levels are regulated at the subcellular scale. Indeed high, intermediate, and low levels of sterol are found in the plasma membrane, endosome/Golgi apparatus, and mitochondria/endoplasmic reticulum, respectively [3, 4]. This uneven sterol distribution between

Le´a P. Wilhelm and Laetitia Voilquin contributed equally to this work. Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_11, © Springer Science+Business Media, LLC, part of Springer Nature 2019

137

138

Le´a P. Wilhelm et al.

organelles results both from vesicular and non-vesicular transport mechanisms [5]. In particular, a class of proteins named Lipid Transfer Proteins (LTPs) plays a crucial role in the latter process. LTPs are cytosolic proteins which have a pocket that accommodates a lipid molecule; LTPs can capture, move, and deliver this lipid molecule within the cell [6–8]. The study of sterol transport inside cells faces a major obstacle, which is the ability to trace this type of molecule in situ. However, the use of natural fluorescent sterols that closely mimic cholesterol, fluorophore-labeled sterols, or fluorescent probes that can trace cholesterol distribution in cells, have advanced the understanding of sterol transport [9]. In this chapter, we describe the use of two fluorescent probes, GFP-D4 and filipin, to label and quantify plasma membrane and intracellular cholesterol levels. The GFP-D4 probe is a fusion between the fluorescent protein GFP and the D4 fragment of perfringolysin O (referred to as θ toxin), a pore-forming toxin from the bacterium Clostridium perfringens. Perfringolysin O is a cholesterol-dependent cytolysin secreted as a water-soluble monomer that recognizes and binds cholesterol-rich membranes where it oligomerizes and creates pores. The D4 fragment of perfringolysin O is able to bind cholesterol-rich membranes (>30 mol% cholesterol) but is devoid of pore-forming activity [10–13]. Interestingly, mutants of GFP-D4, able to bind membranes with less cholesterol, have been generated [14, 15]. In this chapter, we describe the use of recombinant GFP-D4 to stain intracellular cholesterol, and to label plasma membrane cholesterol followed by its quantification by flow cytometry. Filipin is a polyene macrolide extracted from the bacterium Streptomyces filipinensis [9, 16]. Filipin is naturally fluorescent and specifically binds to sterols but not to esterified sterols. Owing to its toxicity, filipin is used on fixed cells. The use of filipin to label and quantify intracellular cholesterol is technically challenging: first, the probe is very sensitive to photobleaching which impedes its use for quantification; and second, it labels total cellular cholesterol, and especially the plasma membrane, making intracellular cholesterol difficult to image. Thus, to bypass these technical limits, we set up a new protocol, also described in this chapter, which allows the imaging and quantification of intracellular cholesterol using filipin [17]. To avoid filipin photobleaching, we used a new generation confocal microscope equipped with a 355 nm UV-laser and ultrasensitive detectors. In addition, to visualize intracellular cholesterol pools, we treated live cells with methyl-β-cyclodextrin (MβCD), a molecule able to deplete cholesterol from the plasma membrane [18].

Intracellular and Plasma Membrane Cholesterol Labeling and Quantification

2

139

Materials

2.1 Chemicals and Buffers

1. U18666A [(3β)-3-[2-(Diethylamino)ethoxy]androst-5-en17-one hydrochloride]: 1 mg/mL in water (stock solution). Store at
20  C. 2. Methyl-β-cyclodextrin (MβCD) solution: Dissolve 660 mg of MβCD powder in 1 mL of water. Heat at 65  C for 15 min to dissolve, and vortex several times. Store this stock solution (500 mM) at
20  C. Heat at 65  C for 15 min before use. 3. MβCD-cholesterol complex (MβCD/cholesterol molar ratio: 6.5/1) solution: Dissolve 2.65 g of MβCD in 25 mL of buffer (150 mM NaCl, 20 mM HEPES-NaOH, pH 7.5, 1 mM CaCl2, 5 mM KCl, 1 mM MgCl2). Add 0.119 g of cholesterol powder (Sigma), mix, sonicate three times (10 min each) and heat (15 min at 65  C) to solubilize cholesterol. Store this stock solution (80 mM) at
20  C. Heat at 65  C for 15 min before use. 4. Phosphate-buffered saline (PBS) 1: 134 mM NaCl, 2.7 mM KCl, 8.1 mM Na2HPO4, 1.47 mM KH2PO4, pH 7.4. 5. Paraformaldehyde (PFA) stock solution: 16% (w/v) in water (EMS). Store at room temperature (see Note 1). 6. PFA solution (4% (w/v) in PBS 1). Mix 10 mL of PFA stock solution (16%) with 4 mL of PBS 10 and 26 mL of water. Store at 4  C (see Note 1). 7. Filipin solution. To prepare 50 mL of filipin solution (0.1 mg/ mL), first dissolve 5 mg of filipin powder in 200 μL of dimethyl sulfoxide (DMSO). Add this solution dropwise to 50 mL of PBS 1 under constant vortexing. Keep filipin protected from light. Aliquot this solution and store at
20  C. 8. 10% (w/v) bovine serum albumin (BSA) solution: Dissolve 1 g of BSA (Fraction V) in 10 mL of PBS 1. Store the stock solution at
20  C. 9. Blocking solution: 1% (w/v) BSA in PBS 1. 10. Cell culture medium with and without serum. Trypsin solution (~30 μg/mL) and Trypsin-EDTA (~30 μg/mL, 0.53 mM).

2.2 GFP-D4 Production

1. GFP-D4 expression plasmid (pET28/His6-EGFP-D4; RIKEN BioResource Research Center; RDB13961). This plasmid can be amplified in E. coli DH5α bacteria. 2. E. coli BL21 (DE3) competent bacteria. 3. LB broth and LB agar plates containing 40 μg/mL kanamycin. 4. IPTG (isopropyl β-D-1-thiogalactopyranoside): A 1 M stock solution is prepared by dissolving IPTG in water, and stored at
20  C.

140

Le´a P. Wilhelm et al.

5. Conical flasks. 6. Lysis buffer: 50 mM NaH2PO4/Na2HPO4, 300 mM NaCl, 10 mM imidazole, pH 8. Add 58.25 mL of Na2HPO4 (0.2 M), 4.25 mL of NaH2PO4 (0.2 M), 15 mL of NaCl (5 M), and 2.5 mL of imidazole (1 M) to a graduated cylinder. Add water to a volume of 250 mL, mix, and store at 4  C. Before use, add 1 tablet of EDTA-free complete protease inhibitor cocktail (Roche) in 50 mL of lysis Buffer. 7. His-Select Nickel Affinity Gel (Sigma). 8. Elution buffer: 20 mM NaH2PO4/Na2HPO4, 250 mM imidazole, pH 7.4. Add 1.935 mL of Na2HPO4 (0.2 M), 0.565 mL of NaH2PO4 (0.2 M), and 6.25 mL of imidazole (1 M) to a graduated cylinder. Add water to a volume of 25 mL, mix, and store at 4  C. 9. Phosphate buffer (PB): 20 mM NaH2PO4/Na2HPO4,pH 7.4. 10. Ultrafiltration units (Ultra-15 Amicon, MWCO 10 kDa, Millipore). 11. Glycerol (molecular biology grade). 12. Sonicator (Sonics Vibra-Cell 750 W equipped with a 13 mm probe). 13. Chromatography column (Bio-Rad Poly-Prep chromatography column). 14. Peristaltic pump (Gilson). 15. Spectrophotometer (Nanodrop, Thermo Scientific). 2.3 Flow Cytometry and Imaging

1. Round cover glasses (diameter 12 mm, N . 1.5H). 2. Superfrost microscope slides. 3. 24-well plates. 4. P150 plates. 5. Propidium iodide (PI) (50 μg/mL stock solution, stored at 4  C) is diluted in PBS to a final concentration of 0.35 μg/mL. 6. TO-PRO-3 iodide solution: Dilute a 1 mM stock solution in DMSO (ThermoFisher scientific, stored at
20  C) in PBS to a final concentration of 0.35 μg/mL. 7. Hoechst-33258 solution: Prepare a stock solution of the dye at 10 mg/mL in water. Dilute the dye in PBS to a final concentration of 10 μg/mL. 8. Anti-CD63 mouse monoclonal antibody (Developmental Studies Hybridoma Bank (DSHB) H5C6; deposited to the DSHB by August, J.T./Hildreth, J.E.K.). The hybridoma supernatant is diluted 100 times in blocking solution. 9. Mounting medium (ProLong Gold, Invitrogen).

Intracellular and Plasma Membrane Cholesterol Labeling and Quantification

141

10. Confocal laser scanning microscope (Leica SP8) equipped with a 355 nm optically pumped semiconductor laser (Coherent), super-sensitive detectors (HyD, Leica), and a 63 objective (oil, NA 1.4). 11. Image analysis software: Fiji software (http://fiji.sc) or CellProfiler (http://cellprofiler.org/). 12. Flow cytometer (BD FACS Celesta) or similar. 13. Flowjo software (FlowJo LLC) or similar.

3

Methods

3.1 GFP-D4 Production and Purification

1. Transform competent E. coli BL21 (DE3) cells with the pET28/His6-EGFP-D4 vector according to the manufacturer’s instructions. 2. Spread bacteria on LB agar plates with kanamycin and incubate overnight at 37  C. 3. Transfer one colony to 10 mL of LB broth with kanamycin. Incubate for 16 h at 37  C under agitation. 4. Transfer this pre-culture to a 5 L conical flask containing 1 L of LB broth with kanamycin. 5. Incubate at 37  C under agitation. Measure the optical density (OD600nm) at a wavelength of 600 nm with a spectrophotometer. When the OD600nm has reached 0.4–0.6, cool down the culture to 18  C. 6. Add IPTG to a final concentration of 0.4 mM. 7. Incubate under agitation for 16 h at 18  C. 8. Transfer the culture medium to centrifugation tubes. Pellet the cells by centrifugation at 3500  g for 15 min. 9. Discard the supernatant. Resuspend the cell pellet in 100 mL of lysis buffer. 10. Sonicate the suspension for 10 min (40% amplitude, on/off cycles 10s/20s) on ice under constant stirring. 11. Centrifuge twice at 3500  g for 15 min at 4  C. Keep the supernatant. 12. Centrifuge twice at 11,000  g for 30 min at 4  C. Keep the supernatant. 13. Add 1.5 mL of nickel-charged resin into a chromatography column. Connect the column to a peristaltic pump (see Note 2). 14. Wash the resin with 16 mL of lysis buffer (flow rate ~1 mL/ min). 15. Load the supernatant on the resin (see Note 3).

142

Le´a P. Wilhelm et al.

16. Wash the column with 30 mL of lysis buffer. 17. Elute the GFP-D4 probe with 6 mL of elution buffer. 18. The buffer is then exchanged and the probe concentrated with an ultrafiltration unit. First, equilibrate the membrane of a centrifugal filter by adding 5 mL of phosphate buffer. Centrifuge for 10 min at 3500  g. Discard the liquid. 19. Add the GFP-D4 probe to the ultrafiltration unit. Spin at 4500  g for 30 min in a fixed-angle rotor (see Note 4). Discard the flow-through. 20. Add 7 mL of PB buffer and spin at 4500  g for 30 min (see Note 4). Discard the flow-through. 21. Repeat two times the step 20. 22. Final spin at 4500  g and stop when the probe is concentrated in ~1 mL. Recover the probe and measure its concentration with a spectrometer (molar extinction coefficient ε280nm ¼ 65,320 M
1·cm
1; molecular weight: 42.1 kDa). 23. Add 1 mL of glycerol and store the probe at
20  C. 3.2 Plasma Membrane Cholesterol Staining with the GFPD4 Probe, Followed by Confocal Imaging

1. Day 0: Plate cells on a glass coverslip (see Notes 5 and 6). 2. Day 1: Incubate with the GFP-D4 solution (1: 200 in PBS containing 1% BSA) for 30 min at 37  C (see Notes 7 and 8). 3. Quickly wash with 1 mL of PBS at room temperature (see Note 9). 4. Fix the cells with 0.5 mL of PFA solution for 10 min at room temperature. 5. Wash with 1 mL of PBS for 5 min at room temperature. 6. Stain nuclei with the DNA marker Hoechst-33258 for 5 min. 7. Wash with 1 mL of PBS for 5 min at room temperature. 8. Mount the sample (see Note 10). 9. Acquire images with a confocal laser scanning microscope equipped with an Argon Laser (488 nm line for GFP excitation) and a 405 nm diode laser (Hoechst-33258 excitation) (Fig. 1a).

3.3 Plasma Membrane Cholesterol Staining with the GFPD4 Probe, Followed by Flow Cytometry Analysis

1. Day 0: Plate 600,000 cells in a P150 plate. 2. Day 1: Rinse the cells with PBS. Trypsinize the cells and resuspend them in 5 mL of PBS (see Notes 6 and 11). 3. Centrifuge for 5 min at 230  g. 4. Resuspend the cells in 800 μL of PBS. Aliquot in three tubes (250 μL in each tube) to have a technical triplicate. 5. Add 250 μL of GFP-D4 (10 μg/mL) diluted in PBS containing 2% (w/v) BSA in each tube (see Notes 7 and 12).

Intracellular and Plasma Membrane Cholesterol Labeling and Quantification

a

b

MβCD

U18666A

No treatment

GFP-D4 Intracellular staining

No treatment

GFP-D4 Plasma membrane staining

c

143

GFP-D4 Plasma membrane staining FACS analysis

d

GFP-D4 Plasma membrane staining FACS analysis 100

Normalized To Mode

Normalized To Mode

100 80 60 40 20 0

80 60 40 20 0

101

102

103

104

105

Fluorescence intensity (Log)

HeLa (no probe) HeLa HeLa/MβCD treatment HeLa/MβCD+Cholesterol treatment

101

102

103

104

105

Fluorescence intensity (Log)

HeLa (no probe) HeLa HeLa/U18666A treatment

Fig. 1 Plasma membrane and intracellular cholesterol staining with the GFP-D4 probe (a) Plasma membrane cholesterol staining of HeLa cells with the GFP-D4 probe. Live cells were left untreated (top image) or treated with MβCD (10 mM in medium without serum for 30 min at 37  C; bottom image), then stained with the GFP-D4 probe and fixed. The nucleus staining is shown in blue. Scale bar: 10 μm. (b) Intracellular cholesterol staining of HeLa cells with the GFP-D4 probe. Live cells were left untreated (top image) or treated with U18666A (2.5 μg/mL in normal medium for 16 h at 37  C; bottom image), fixed, permeabilized, and then stained with the GFP-D4 probe. Nuclei staining is shown in blue. Scale bar: 10 μm. In panel (a) and (b), a confocal section image and a higher magnification (2.5) image of the area outlined in white are displayed. (c) Flow cytometry analysis of plasma membrane cholesterol staining with the GFP-D4 probe. Live HeLa cells were left untreated (green curve), or were treated with MβCD (10 mM for 30 min at 37  C; purple curve) or with MβCD complexed to cholesterol (500 μM for 2 h at 37  C; blue curve). (d) Flow cytometry analysis of live HeLa

144

Le´a P. Wilhelm et al.

6. Incubate for 30 min at 37  C under gentle agitation. 7. Fix the cells by adding 500 μL of PFA working solution. Mix gently. Incubate for 10 min at room temperature (see Note 1). 8. Pellet the cells by centrifugation for 5 min at 230  g. 9. Resuspend the cells in 1 mL of PBS. 10. Analyze GFP-D4 fluorescence on individual cells by flow cytometry. 11. Single cells are discriminated from doublets using forward scatter height (FSC-H) and forward scatter area (FSC-A) measures. FCS-A and FCS-H are plotted and cells along the diagonal are gated. GFP fluorescence is measured on individual cells using blue laser (488 nm) excitation and a FITC filter (band pass filter 530/30 nm). A minimum of 10,000 events must be counted. 12. Analyze the results with Flowjo software: Viable cells are manually gated using SSC (Side-scattered) and FSC (Forwardscattered) parameters. Data are represented as single-parameter histograms showing GFP fluorescence intensity (x-axis) and the number of events (y-axis). The mean of GFP fluorescence is also measured (Fig. 1c, d). 3.4 Intracellular Cholesterol Labeling with the GFP-D4 Probe

1. Day 0: Plate ~10,000–15,000 cells on a glass coverslip (see Note 5). 2. Day 1: Wash with 1 mL of PBS at room temperature for 2 s (see Note 13). 3. Fix the cells with 500 μL of PFA solution at room temperature for 10 min (see Note 1). 4. Wash twice with 1 mL of PBS at room temperature (5 min). Keep the cells in PBS. 5. Permeabilize the cells using a liquid nitrogen bath: Take the coverslip out of the 24-well plate, remove the excess of PBS, and put the coverslip in liquid nitrogen. Wait for 5 s. Place the coverslip back into the 24-well plate in PBS. 6. Incubate with 500 μL of blocking solution at room temperature for 30 min, under gentle agitation.

ä Fig. 1 (continued) cells which are untreated (green curve), or treated with U18666A (2.5 μg/mL for 16 h at 37  C; yellow curve). After the different treatments, the cells were stained with the GFP-D4 probe, fixed, and GFP-D4 staining intensity was analyzed by flow cytometry. These representative histograms display the number of cells analyzed (normalized to mode) as a function of GFP-D4 fluorescence (log intensity). Unstained cells (gray curve) are shown. Note that HeLa cells are strongly labeled with the GFP-D4 probe; MβCD and MβCD/cholesterol treatments prior to labeling strongly decrease and increase GFP-D4 signal intensity, respectively; U18666A treatment decreases GFP-D4 signal intensity. Indeed, U18666A was shown to induce cholesterol accumulation in the endocytic pathway at the expense of the plasma membrane

Intracellular and Plasma Membrane Cholesterol Labeling and Quantification

145

7. Incubate with the GFP-D4 solution (1:200 in PBS containing 1% BSA) for 1 hour at room temperature (see Notes 7 and 12). 8. Quickly wash with 1 mL of PBS at room temperature (see Note 9). 9. Fix the cells with 500 μL of PFA solution for 10 min at room temperature (see Note 1). 10. Wash with 1 mL of PBS at room temperature for 5 min (see Note 14). 11. Stain nuclei with 1 mL of Hoechst-33258 solution. 12. Wash with 1 mL of PBS at room temperature for 5 min. 13. Mount the sample (see Note 10). 14. Acquire images with a confocal laser scanning microscope equipped with an Argon Laser (488 nm line for GFP excitation) and a 405 nm diode laser (Hoechst-33258 excitation) (Fig. 1b). 3.5 Cholesterol Labeling with Filipin, Followed by Confocal Imaging and Image Analysis 3.5.1 Labeling and Image Acquisition

1. Day 0: Plate ~10,000–15,000 cells on a glass coverslip (see Note 5). 2. Day 1: Wash twice with 1 mL of PBS at room temperature (2 s each). 3. In order to facilitate intracellular cholesterol visualization, remove plasma membrane cholesterol with MβCD: Dilute the MβCD stock 50 times with preheated culture medium without serum, to obtain a 10 mM MβCD concentration. Add 500 μL of MβCD solution onto the cells, and incubate at 37  C for 30 min. 4. Wash twice with 1 mL of PBS at room temperature (2 min each). 5. Fix the cells with 500 μL of PFA solution at room temperature for 10 min (see Note 1). 6. Wash with 1 mL of PBS at room temperature (5 min). 7. Incubate with 500 μL of filipin solution at room temperature for 30 min, under gentle agitation (see Note 7). 8. Remove the filipin solution and wash the cells with 1 mL of PBS at room temperature for 2 min (see Notes 15 and 16). 9. Cells are labeled with propidium iodide (PI) (0.35 μg/mL in PBS) for 5 min, which labels nucleic acids (both DNA and RNA), and thus delineates the cell morphology, an important step for image analysis. Alternatively, cell nuclei are labeled with the DNA marker TO-PRO-3 iodide (0.35 μg/mL in PBS) for 5 min. 10. Wash twice with 1 mL of PBS at room temperature (2 min each time).

146

Le´a P. Wilhelm et al.

Fig. 2 Whole-cell and intracellular cholesterol staining with filipin. (a) Live HeLa cells were fixed and stained with filipin. (b) Cells were treated with MβCD (10 mM in serum-free medium; 30 min at 37  C) before fixation and filipin staining. (c) Live HeLa cells were treated with U18666A to induce cholesterol accumulation in endosomes/lysosomes before fixation and filipin staining. (d) Cells were treated with U18666A and then with MβCD, before fixation and filipin staining. Each panel display a confocal section image and a higher magnification (2.5) image of the area outlined in white. Scale bars: 10 μm. Note that MβCD treatment, by removing cholesterol from the plasma membrane, improves the visualization of intracellular cholesterol pools

11. Mount the sample (see Note 10). 12. Acquire images with a confocal laser scanning microscope equipped with a 355 nm optically pumped semiconductor laser to excite filipin. Acquisitions are made using the photon counting mode with a super-sensitive detector to decrease light dosage and thus filipin photobleaching (Fig. 2). Typically, a single optical section comprising multiple randomly chosen cells is acquired (Fig. 3a). 3.5.2 Quantification of Cholesterol in Discrete Subcellular Structures

We use the Fiji software or CellProfiler to automate the analysis. The quantification of cholesterol accumulation in discrete subcellular structures is based on signal intensity thresholding 1. Individual cells are segmented on images using the PI staining (Fig. 3b). First an intensity threshold is applied to determine the cell contours. Nuclei positions, manually selected on the PI staining image, are used to divide the image into discrete areas with a watershed algorithm (Find maxima). Then, the

Intracellular and Plasma Membrane Cholesterol Labeling and Quantification

147

a Confocal images

Filipin

b Image segmentation

CD63

PI staining

PI

Thresholded image

Segmentation mask

Manual selection of nuclei positions

Nuclei positions mask

Segmented particles

c Intensity thresholding and filipin quantification

Filipin staining

Thresholded image

Segmented image

Signal intensity measurement in each cell

d CD63 mask and filipin quantification

CD63 staining

Thresholded image

Filipin staining in CD63 mask

Segmented image

Fig. 3 Filipin staining quantification by image analysis. (a) HeLa cells were stained with filipin (blue), anti-CD63 antibodies (green), and propidium iodide (PI, red). Images were acquired with a confocal microscope and analyzed with Fiji. Scale bar: 30 μm. (b) Image segmentation: First an intensity threshold is applied to determine the cell contours (Thresholded image). Nuclei positions (Nuclei positions mask), manually selected on the PI staining image (Manual selection of nuclei positions), are used to divide the image into discrete areas with a watershed algorithm (Segmented particles). Then, the thresholded and the segmented images are

148

Le´a P. Wilhelm et al.

thresholded and the segmented images are combined to build a segmentation mask showing the contour of individual cells. 2. Apply a threshold value to the filipin image (see Note 17). 3. Then, apply the segmentation mask and measure filipin staining intensity in individual cells (Fig. 3c). 3.5.3 Quantification of Cholesterol in a Specific Subcellular Compartment

The second method is based on the use of a mask to restrict the analysis to a specific cellular compartment (Fig. 3d; illustrated with a labeling of late endosomes/lysosomes with anti-CD63 antibodies). 1. Individual cells are segmented as described in step 1 of Subheading 3.5.2. 2. Apply an intensity threshold to the image of the subcellular compartment marker staining. 3. Generate a mask from this threshold. 4. Combine the threshold and the segmentation masks, and measure filipin staining intensity in individual cells.

4

Notes 1. Manipulate PFA under a fume hood. Wear protective gloves, protective clothing, and eye protection. 2. Alternatively, use an automated purification system with a pre-packed immobilized metal ion affinity chromatography (IMAC) column. 3. The green color of GFP facilitates the purification since the protein can easily be followed when bound to the resin and during elution. 4. Adjust the centrifugation time to the volume remaining on the filter which should not be lower than ~1 mL. 5. Under sterile condition, put a glass coverslip in each well of a 24-well plate. Add 500 μL of preheated medium and the cell suspension (~10,000–15,000 cells). The confluency should be low (around 20–30%) to have isolated cells.

ä Fig. 3 (continued) combined to generate a mask showing the contour of individual cells (Segmentation mask). (c) A threshold is applied on the filipin staining image to decrease the background signal and to focus the analysis on filipin bright spots corresponding to cholesterol accumulations. The segmentation mask is applied on this thresholded image, and filipin signal intensity is measured in each cell. (d) A threshold is applied on CD63 staining image to delineate late endosomes and lysosomes positive for this marker. The segmentation mask is applied on this thresholded image, and filipin signal intensity is measured in each cell

Intracellular and Plasma Membrane Cholesterol Labeling and Quantification

149

6. For cells which have a tendency to form clusters, trypsin EDTA can be used to detach and dissociate cells before the plating. 7. Protect the samples from light during all steps, once filipin or GFP-D4 has been added. 8. Optional: The cells can be pretreated with MβCD to remove cholesterol from the plasma membrane as negative control. Dilute the MβCD stock 50 times with preheated medium without serum, to obtain a 10 mM MβCD solution in culture medium. Wash the cells twice with PBS. Add 0.5 mL of MβCD solution onto the cells, and incubate at 37  C for 30 min. Remove the MβCD solution and wash the cells once with PBS. 9. This step allows the removal of the excess probe. Remove the probe solution, add 1 mL of PBS at room temperature, and immediately remove it. A longer incubation in PBS would enable the GFP-D4 probe to detach from the membrane before the fixation step. 10. Apply a drop of mounting medium on a glass slide. Take out the cover glass carefully from the well, remove the excess of PBS, and carefully place over the mounting medium with the cells face down. Incubate at least for 2 h at room temperature before use. Slides can be stored at 4  C protected from light. 11. On Day 1, positive and negative control samples can be obtained by adding or removing cholesterol from the plasma membrane before trypsinization. Plasma membrane cholesterol levels can be reduced by MβCD or U18666A treatment, or increased by MβCD-cholesterol treatment. MβCD treatment: Dilute the MβCD stock 50 times with preheated medium without serum, to obtain a 10 mM MβCD solution in culture medium; wash the cells twice with PBS; add 5 mL of MβCD solution onto the cells and incubate at 37  C for 30 min. U18666A treatment: Wash the cells twice with PBS; treat the cells with U18666A (2.5 μg/mL) in normal culture medium for 16 h at 37  C. MβCD-cholesterol treatment: Dilute the MβCD-cholesterol complex stock 160 times with preheated medium without serum, to obtain a 500 μM MβCDcholesterol complex solution in culture medium; wash the cells twice with PBS; add 5 mL of MβCD-cholesterol complex solution onto the cells, and incubate at 37  C for 2 h. 12. The concentration of GFP-D4 has to be adjusted to the samples analyzed. Increasing the concentration of GFP-D4 will increase the fluorescent signal proportionally. 13. On Day 1, positive control samples can be obtained by treating cells with U18666A to induce cholesterol accumulation in endosomes. Treat the cells with U18666A (2.5 μg/mL) in

150

Le´a P. Wilhelm et al.

normal culture medium for 16 h at 37  C, and then wash with PBS. 14. At this step, an immunofluorescence staining can be done to detect the protein of interest in addition to cholesterol. (a) Incubate the cells with 500 μL of blocking solution at room temperature for 30 min, under gentle agitation. (b) Add appropriate primary antibody diluted in 250 μL of blocking solution, and incubate for 3 h at room temperature, or overnight at 4  C, under gentle agitation. (c) Wash 3 times with 1 mL of PBS (5 min each time). (d) Add fluorescently conjugated secondary antibodies (Alexa Fluor 488), and incubate for 1 h at room temperature, under gentle agitation. (e) Wash three times with 1 mL of PBS at room temperature (5 min each time). 15. At this step, an immunofluorescence staining can be done to detect the protein of interest in addition to cholesterol. (a) Incubate the cells with 500 μL of blocking solution at room temperature for 30 min, under gentle agitation. (b) Add appropriate primary antibody diluted in 250 μL of blocking solution, and incubate for 3 h at room temperature, or overnight at 4  C, under gentle agitation. (c) Wash three times with 1 mL of PBS (5 min each time). (d) Add fluorescently conjugated secondary antibodies (Alexa Fluor 488), and incubate for 1 h at room temperature, under gentle agitation. (e) Wash three times with 1 mL of PBS at room temperature (5 min each time). (f) Incubate the cells with 500 μL of filipin solution at room temperature for 30 min, under gentle agitation. (g) Wash twice with 1 mL of PBS at room temperature (5 min each time). 16. Since filipin permeabilizes membranes, it allows antibody entry inside cells. Note that only some antibodies work in the presence of filipin; this has to be tested for your specific application. 17. The choice of the threshold value is somehow arbitrary. However, in order to allow a comparison between experiments, the threshold can be determined in control cells (no cholesterol accumulation) as the 99th percentile of pixel intensity values. Evaluate this value for all images of control cells and calculate a mean intensity which will be defined as the threshold value.

Intracellular and Plasma Membrane Cholesterol Labeling and Quantification

151

Acknowledgments We thank Thomas Di Mattia and Alastair McEwen for their critical reading of the manuscript. We thank the members of the Molecular and Cellular Biology of Breast Cancer team (IGBMC) for helpful advice and discussions, and the IGBMC imaging center. L.V. received an allocation from the Ministe`re de l’Enseignement Supe´rieur et de la Recherche (France; http://www. enseignementsup-recherche.gouv.fr/) and L.P.W. a fellowship from the Fondation pour la Recherche Me´dicale (https://www. frm.org/). This work was supported by grants from the Institut National Du Cancer INCA (INCA_9269; www.e-cancer.fr), the Ligue Contre le Cancer (Confe´rence de Coordination Interre´gionale du Grand Est; https://www.ligue-cancer.net), The Ara Parseghian Medical Research Fund (http://parseghianfund.nd.edu/), Vaincre les maladies lysosomales (http://www.vml-asso.org/) and the grant ANR-10-LABX-0030-INRT, a French State fund managed by the Agence Nationale de la Recherche under the frame program Investissements d’Avenir ANR-10-IDEX-0002-02. We also acknowledge funds from the Institut National de Sante´ et de Recherche Me´dicale (http://www.inserm.fr/), the Centre National de la Recherche Scientifique (http://www.cnrs.fr/), and the Universite´ de Strasbourg (http://www.unistra.fr). References 1. Maxfield FR, van Meer G (2010) Cholesterol, the central lipid of mammalian cells. Curr Opin Cell Biol 22:422–429 2. Mesmin B, Antonny B, Drin G (2013) Insights into the mechanisms of sterol transport between organelles. Cell Mol Life Sci 70:3405–3421 3. Maxfield FR, Menon AK (2006) Intracellular sterol transport and distribution. Curr Opin Cell Biol 18:379–385 4. van Meer G, Voelker DR, Feigenson GW (2008) Membrane lipids: where they are and how they behave. Nat Rev Mol Cell Biol 9:112–124 5. Wustner D, Herrmann A, Hao M et al (2002) Rapid nonvesicular transport of sterol between the plasma membrane domains of polarized hepatic cells. J Biol Chem 277:30325–30336 6. Chiapparino A, Maeda K, Turei D et al (2016) The orchestra of lipid-transfer proteins at the crossroads between metabolism and signaling. Prog Lipid Res 61:30–39 7. Holthuis JC, Levine TP (2005) Lipid traffic: floppy drives and a superhighway. Nat Rev Mol Cell Biol 6:209–220

8. Lev S (2010) Non-vesicular lipid transport by lipid-transfer proteins and beyond. Nat Rev Mol Cell Biol 11:739–750 9. Maxfield FR, Wustner D (2012) Analysis of cholesterol trafficking with fluorescent probes. Methods Cell Biol 108:367–393 10. Abe M, Makino A, Hullin-Matsuda F et al (2012) A role for sphingomyelin-rich lipid domains in the accumulation of phosphatidylinositol-4,5-bisphosphate to the cleavage furrow during cytokinesis. Mol Cell Biol 32:1396–1407 11. Ohno-Iwashita Y, Shimada Y, Waheed AA et al (2004) Perfringolysin O, a cholesterol-binding cytolysin, as a probe for lipid rafts. Anaerobe 10:125–134 12. Rossjohn J, Feil SC, McKinstry WJ et al (1997) Structure of a cholesterol-binding, thiol-activated cytolysin and a model of its membrane form. Cell 89:685–692 13. Shimada Y, Maruya M, Iwashita S et al (2002) The C-terminal domain of perfringolysin O is an essential cholesterol-binding unit targeting to cholesterol-rich microdomains. Eur J Biochem 269:6195–6203

152

Le´a P. Wilhelm et al.

14. Johnson BB, Moe PC, Wang D et al (2012) Modifications in perfringolysin O domain 4 alter the cholesterol concentration threshold required for binding. Biochemistry 51:3373–3382 15. Liu SL, Sheng R, Jung JH et al (2017) Orthogonal lipid sensors identify transbilayer asymmetry of plasma membrane cholesterol. Nat Chem Biol 13:268–274 16. Gimpl G, Gehrig-Burger K (2007) Cholesterol reporter molecules. Biosci Rep 27:335–358

17. Wilhelm LP, Wendling C, Vedie B et al (2017) STARD3 mediates endoplasmic reticulum-toendosome cholesterol transport at membrane contact sites. EMBO J 36:1412–1433 18. Zidovetzki R, Levitan I (2007) Use of cyclodextrins to manipulate plasma membrane cholesterol content: evidence, misconceptions and control strategies. Biochim Biophys Acta 1768:1311–1324

Chapter 12 Monitoring and Modulating Intracellular Cholesterol Trafficking Using ALOD4, a Cholesterol-Binding Protein Shreya Endapally, Rodney E. Infante, and Arun Radhakrishnan Abstract Mammalian cells carefully control their cholesterol levels by employing multiple feedback mechanisms to regulate synthesis of cholesterol and uptake of cholesterol from circulating lipoproteins. Most of a cell’s cholesterol (~80% of total) is in the plasma membrane (PM), but the protein machinery that regulates cellular cholesterol resides in the endoplasmic reticulum (ER) membrane, which contains a very small fraction (~1% of total) of a cell’s cholesterol. How does the ER communicate with PM to monitor cholesterol levels in that membrane? Here, we describe a tool, ALOD4, that helps us answer this question. ALOD4 traps cholesterol at the PM, leading to depletion of ER cholesterol without altering total cell cholesterol. The effects of ALOD4 are reversible. This tool has been used to show that the ER is able to continuously sample cholesterol from PM, providing ER with information about levels of PM cholesterol. Key words Cholesterol trafficking, Plasma membrane, Endoplasmic reticulum, Anthrolysin O

1

Introduction Cholesterol levels in mammalian cells are tightly regulated to lie within narrow limits. Regulation is achieved by controlling the two pathways by which cells acquire cholesterol. Cells can either synthesize cholesterol in the ER or uptake circulating cholesterol-rich low-density lipoprotein (LDL) via the LDL receptor [1]. When cholesterol levels in a cell rise above optimal levels, Scap, an ER membrane protein and cholesterol sensor, binds cholesterol and blocks proteolytic activation of a transcription factor, sterol regulatory element binding protein (SREBP). This leads to a decrease in expression of SREBP gene targets, which include cholesterol biosynthetic enzymes for endogenous production of cholesterol and the LDL receptor which uptakes lipoprotein particles containing exogenous cholesterol [2]. As a result, cholesterol levels decline and return to optimal levels. Cholesterol levels are highest in the PM, which contains ~80% of total cellular cholesterol [3]. To meet the needs of PM, most of the cholesterol synthesized in ER or

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_12, © Springer Science+Business Media, LLC, part of Springer Nature 2019

153

154

Shreya Endapally et al.

LDL-derived cholesterol from lysosomes must be ultimately transported to the PM. However, the cholesterol control machinery that regulates this cholesterol synthesis and uptake resides in the ER, which contains only ~1% of the cell’s total cholesterol [3, 4]. This raises the question: How does the ER monitor cholesterol levels in PM? Recent studies that focused on the organization of cholesterol in PM have provided some insights into this problem of organelle communication. These studies showed that cholesterol in PMs, which comprises 40–45 mol% of total PM lipids when cells are grown in lipoprotein-rich serum, is organized into three different pools [5, 6]. Two of these pools of cholesterol (~27 mol% of total PM lipids) are sequestered by sphingomyelin and other membrane factors, respectively, and are inaccessible to soluble proteins that bind membrane cholesterol [6]. PM cholesterol in excess of these two pools constitutes a third pool that is accessible to cholesterolbinding proteins and can be transported to ER to block Scapmediated proteolytic activation of SREBP, thereby inhibiting cholesterol synthesis and uptake. Thus, controlling the accessibility of PM cholesterol allows regulated communication between PM and ER, ensuring that cholesterol uptake and synthesis is not prematurely shut down before the cell’s cholesterol needs are met. Further understanding of the regulation of intracellular cholesterol trafficking requires new tools that selectively modulate transport pathways. Here, we provide a detailed description of one such method to monitor and modulate cholesterol transport from PM to ER [7]. The method uses a recently described non-lytic cholesterolbinding protein, designated as Anthrolysin O domain 4 (ALOD4), as a reversible inhibitor of cholesterol transport from PM to ER [7–9]. ALOD4 binds to accessible cholesterol in PMs of cells at 37
C without lysing cells or getting internalized. In addition to serving as an effective tool to monitor PM cholesterol accessibility, ALOD4 also modulates transport of cholesterol from PM to ER. This provides a useful tool to decouple cholesterol levels in ER from those in PM, and to study cholesterol transport between these membranes. Compared to traditional reagents such as β-cyclodextrin that lower ER cholesterol and trigger SREBP activation by decreasing total cellular cholesterol, ALOD4 traps PM cholesterol and triggers SREBP activation by lowering ER cholesterol without changing levels of cholesterol in the PM or the whole cell. The use of ALOD4 has shed light on how ER continuously monitors PM cholesterol levels and ensures cellular cholesterol homeostasis [7].

Purification and Use of ALOD4, a Cholesterol-Binding Protein

2

155

Materials

2.1 Plasmid Transformation

1. pALOD4—pRSET B expression vector encoding the domain 4 of Anthrolysin O (amino acids 404–512 with S404C and C472A mutations) along with an NH2-terminal hexahistidine tag followed by an enterokinase cleavage site (Addgene). 2. pALOD4(Mut)—pRSET B expression vector encoding the domain 4 of Anthrolysin O (amino acids 404–512 with S404C, C472A, G501A, T502A, T503A, L504A, Y505A, and P506A mutations) along with an NH2-terminal hexahistidine tag followed by an enterokinase cleavage site (Addgene). 3. BL21 (DE3) (Invitrogen).

pLysS

Escherichia

coli

competent

cells

4. SOC medium (Invitrogen). 5. LB-Agar (RPI) plates containing 100 μg/mL ampicillin (RPI). 2.2 ALOD4 Expression

1. LB broth (RPI) containing 100 μg/mL ampicillin (LB-Amp).

2.3 ALOD4 Purification

1. Buffer A: 50 mM Tris–HCl, pH 7.5, 1 mM Tris(2-carboxyethyl)phosphine (TCEP).

2. 1 M Isopropyl β-p-thiogalactopyranoside (IPTG) stock solution: Dissolve 2.38 g of IPTG in 10 mL of water.

2. Buffer B: Buffer A with 150 mM NaCl. 3. Buffer C: Buffer B with 500 mM imidazole. 4. Buffer D: Buffer A with 500 mM NaCl. 5. EDTA-free protease inhibitor tablets (Roche). 6. 20 mg/mL Phenyl methanesulfonyl fluoride (PMSF) stock solution: Dissolve 1 g of PMSF in 50 mL of ethanol. 7. Lysozyme from chicken egg white (Sigma-Aldrich). 8. Lysis buffer: Buffer B containing 1 mg/mL lysozyme, 400 μg/ mL PMSF, and one EDTA-free protease inhibitor tablet per 20 mL. 9. 100-mL Dounce homogenizer (Kimble). 10. Branson Digital Sonifier (S-250, Fisher Scientific). 11. 0.22 μm 250 mL Stericup filter unit (Millipore). 12. 5-mL HisTrap HP Ni column (GE Healthcare). 13. 1-mL HiTrap Q HP anion exchange column (GE healthcare). 14. Tricorn 10/300 (GE healthcare).

Superdex

200

Increase

column

15. Amicon Ultra-4 10-kDa cutoff centrifugal filters (Millipore). 16. 15% SDS-PAGE gels.

156

Shreya Endapally et al.

17. 5 gel loading buffer: 10% (w/v) SDS, 10 mM β-mercaptoethanol, 20% (v/v) glycerol, 0.2 M Tris–HCl, pH 6.8, 0.05% (w/v) bromophenol blue. 18. Precision Plus Protein Kaleidoscope Standards (BioRad). 19. Coomassie staining solution (Bio-Rad). 20. De-staining solution: Water/ethanol/acetic acid (5:4:1, v/v/v). 21. Spectrometer (e.g., NanoDrop, Thermo Fisher Scientific). 2.4

ALOD4 Labeling

1. 10 mM stocks of Alexa Fluor 488 C5 maleimide (Invitrogen) and Alexa Fluor 647 C2 maleimide (Invitrogen) in DMSO. 2. Nickel-nitrilotriacetic acid (Ni-NTA) agarose beads (Qiagen). 3. 1 M Dithiothreitol (DTT) stock solution: Dissolve 6.48 mg of DTT in 1 mL of water. 4. Glycerol for molecular biology (>99%).

2.5 PM Cholesterol Sequestration by ALOD4

1. CHO-K1 cells (American Type Culture Collection, CCL-61). 2. SV-589 cells (NIGMS Human Genetic Cell Repository). 3. 48-well plates (Fisher Scientific). 4. Medium A: 1:1 mixture of Ham’s F-12 and Dulbecco’s modified Eagle’s medium (Sigma-Aldrich) supplemented with 100 units/mL penicillin and 100 μg/mL streptomycin sulfate (Corning) and 5% (v/v) fetal bovine serum (FCS, SigmaAldrich). 5. Medium B: Dulbecco’s modified Eagle’s medium (low glucose) (Sigma-Aldrich) supplemented with 100 units/mL penicillin and 100 μg/mL streptomycin sulfate and 5% (v/v) FCS. 6. Medium C: 1:1 mixture of Ham’s F-12 and Dulbecco’s modified Eagle’s medium supplemented with 5% (v/v) FCS. 7. Medium D: Dulbecco’s modified Eagle’s medium (low glucose) supplemented with 5% (v/v) FCS. 8. Dulbecco’s Phosphate buffered saline solution (PBS) (SigmaAldrich). The formulation of this buffer is 8 g of NaCl, 0.2 g of potassium phosphate (monobasic), 1.15 g of sodium phosphate (dibasic), and 0.2 g of KCl, all in 1 L water. 9. Buffer E: 10 mM Tris–HCl, pH 6.8, 100 mM NaCl, 1% (w/v) SDS, 1 mM EDTA, 1 mM EGTA, 20 μg/mL PMSF, and EDTA-free protease inhibitor tablets (one tablet per 20 mL). 10. 10% and 15% SDS-PAGE gels. 11. Bio-Rad Trans Blot Turbo System and nitrocellulose transfer packs. 12. Antibodies: IgG-20B12 against hamster SREBP1 [10], IgG-7D4 against hamster SREBP2 [11], IgG-4H4 against

Purification and Use of ALOD4, a Cholesterol-Binding Protein

157

Scap [12], anti-His tagged antibody (Millipore), donkey antimouse secondary IgG antibody (Jackson ImmunoResearch), and goat anti-rabbit secondary IgG antibody (Jackson ImmunoResearch).

3

Methods

3.1 Plasmid Transformation

1. Add 100 ng of pALOD4 or pALOD4(Mut) plasmid to 50 μL of BL21 (DE3) pLysS cells and incubate the mixture on ice for 30 min. 2. Heat shock the cells by placing in a 42
C water bath for 45 s. 3. Immediately transfer cells to ice for 2 min. 4. Recover cells by adding 450 μL of SOC medium to cells and incubate for 1 h at 37
C. 5. Plate 100 μL of transformed and recovered cells on LB-Agar plates with ampicillin (100 μg/mL). Incubate plates for 16 h at 37
C in a bacterial incubator.

3.2 ALOD4 Expression

The protocol is the same for ALOD4 and ALOD4(Mut). 1. Pick a single colony from above plates, transfer into a 250 mL flask containing 160 mL of LB-Amp, and incubate the flask at 37
C in a shaker-incubator (225 rpm) until OD600 ¼ 0.8–1 (see Note 1). 2. Transfer 12 mL of the above starter culture to a 2 L flask containing 1 L of LB-Amp. Incubate flasks at 37
C in a shaker-incubator (225 rpm). 3. Once the OD600 ¼ 0.4–0.6 (see Note 2), reduce the temperature to 18
C and continue shaking for 1 h at 225 rpm. 4. Add IPTG to a final concentration of 1 mM to induce ALOD4 expression and continue shaking at 18
C for 16 h. 5. Harvest the cells by centrifugation at 3220  g for 10 min at 4
C. Pellets can be stored at 80
C after flash freezing in liquid nitrogen.

3.3 ALOD4 Purification

The protocol is the same for ALOD4 and ALOD4(Mut). 1. Resuspend cell pellet from 1 L of bacterial culture in 20 mL of lysis buffer. Homogenize the cell suspension using a Dounce homogenizer. We typically use cell pellets from 6 L of bacterial cultures which yields 1–2 mg of protein (see Note 3). 2. For lysozyme disruption, incubate the homogenized cell suspension at 4
C for 3 h while stirring on a rotator. 3. Lyse the lysozyme-disrupted cells using a tip sonicator (see Note 4).

158

Shreya Endapally et al.

4. Centrifuge the lysed cell suspension at 220,000  g for 1 h. 5. Filter the resultant supernatant using a 0.22 μm Millipore filter. 6. Connect a pre-packed 5-mL HisTrap-HP Ni column to an ¨ KTA, GE Healthcare) and pre-equilibrate colFPLC system (A umn with 10 column volumes (50 mL) of buffer B. Use a flow rate of 0.5 mL/min for all Ni chromatography operations. 7. Load the resultant filtered supernatant from step 5 onto the Ni column. 8. Wash column with 20 column volumes (100 mL) of a mixture of 90% buffer B and 10% buffer C (50 mM imidazole). Elute bound ALOD4 using a linear gradient starting from 90% buffer B/10% buffer C (50 mM imidazole) and ending at 40% buffer B/60% buffer C (300 mM imidazole) over 20 column volumes (100 mL) (see Note 5). 9. Subject an aliquot (~20 μL) of eluted fractions to 15% SDSPAGE followed by Coomassie staining to assay for fractions containing purified ALOD4 (MW ~16 kDa). Use Precision Plus Protein Kaleidoscope Standards as molecular weights. 10. Pool fractions containing ALOD4 and concentrate using an Amicon Ultra-4 10 kDa cutoff centrifugal filter to ~10 mL (see Note 6). 11. Add ~140 mL of NaCl-free Buffer A to the protein solution for decreasing NaCl concentration to ~15 mM. 12. Connect a pre-packed 1-mL HiTrap Q HP anion exchange column to an FPLC and pre-equilibrate with 10 column volumes (10 mL) of buffer A. Use a flow rate of 0.5 mL/min for all ion exchange chromatography operations. 13. Load ALOD4 in low-NaCl buffer on the HiTrap Q column. Wash the column with 10 column volumes (10 mL) of buffer A. 14. Elute bound ALOD4 using buffer D into a single 2 mL fraction. 15. Connect a Tricorn 10/300 Superdex 200 gel filtration column to an FPLC and pre-equilibrate with 2 column volumes (60 mL) of buffer B. Use a flow rate of 0.5 mL/min for all gel filtration chromatography operations. 16. Load the elution from step 14 on to the Superdex 200 column followed by 50 mL of buffer B. 17. Pool protein-rich fractions and concentrate using an Amicon Ultra-4 10 kDa cutoff centrifugal filter to a concentration of 1–2 mg/mL and store at 4
C for use over the next 4 weeks. Measure protein concentration using a NanoDrop instrument (see Note 7). For long-term storage, supplement concentrated protein with 20% (v/v) glycerol, flash-freeze in liquid nitrogen,

Purification and Use of ALOD4, a Cholesterol-Binding Protein

159

Fig. 1 Biochemical characterization of ALOD4. Gel filtration chromatography of purified proteins. Recombinant ALOD4 was purified and labeled with Alexa Fluor 488 (fALOD4-488) or Alexa Fluor 647 (fALOD4-647) fluorescent dyes as described in Subheading 3. Buffer B (1 mL) containing 0.8 mg of ALOD4, fALOD4-488, or fALOD4-647 was loaded onto a Tricorn 10/300 Superdex 200 column and chromatographed at a flow rate of 0.5 mL/min. Absorbance at 280 nm (A280) was monitored continuously to identify ALOD4(blue), fALOD4488(green), or fALOD4-647(red) proteins. Maximum A280 values for each protein (ALOD4: 390 mAU, fALOD4-488: 231 mAU, and fALOD4-647: 279 mAU) are normalized to 1. (Inset) 3 μg of each protein was subjected to 15% SDS/PAGE and stained with Coomassie (left) or imaged with the 600 nm filter (middle) or the 700 nm filter (right) on a LICOR instrument. Coom, Coomassie. Figure adapted from [7]

and store at 80
C (use within 6 months). See Fig. 1 for an example of the homogeneity and monodisperse nature of purified ALOD4 (lane 1 and blue curve). 3.4

ALOD4 Labeling

In many experiments, it is informative to quantify the amount of ALOD4 bound to plasma membranes of cells. Quantification can be easily carried out using ALOD4 labeled with fluorescent dyes. These reagents can also be used for cellular labeling studies using fluorescence microscopy. The labeling protocol is the same for ALOD4 and ALOD4(Mut). 1. Combine 20 nanomoles (320 μg) of purified ALOD4 with 200 nanomoles of Alexa Fluor maleimide in a total volume of 300 μL of buffer B in 1.5 mL Eppendorf tubes. Protect the tubes from light by wrapping with aluminum foil. 2. Place tubes on a rotator at 4
C for 16 h. 3. Quench labeling reactions by addition of DTT to a final concentration of 10 mM and then load the reaction onto a column containing 500 μL of Ni-NTA beads equilibrated in buffer B. 4. Wash column with 10 column volumes of buffer B containing 50 mM imidazole.

160

Shreya Endapally et al.

5. Elute bound labeled ALOD4 using buffer B containing 300 mM imidazole into a single 2 mL fraction. 6. Connect a Tricorn 10/300 Superdex 200 gel filtration column to an FPLC and pre-equilibrate with two column volumes (60 mL) of buffer B. Use a flow rate of 0.5 mL/min for all gel filtration chromatography operations. 7. Load fraction containing labeled ALOD4 from step 6 onto the Superdex 200 column followed by 50 mL of buffer B. Pool protein-rich fractions and concentrate using Amicon Ultra-4 10 kDa cutoff centrifugal filter to a concentration of 1–2 mg/ mL. Measure concentration and degree of labeling of ALOD4 using a spectrometer (see Note 8). 8. Subject an aliquot to 15% SDS-PAGE followed by Coomassie staining and fluorescence scanning (using an Odyssey FC Imager and LI-COR analysis) to verify protein purification and labeling. 9. For long-term storage, supplement concentrated and labeled ALOD4 with 20% (v/v) glycerol, flash-freeze in liquid nitrogen, and store at 80
C. See Fig. 1 for examples of the homogeneity and mono disperse nature of ALOD4 labeled with Alexa Fluor 488 or Alexa Fluor 647 (lanes 2, 3 and green, red curves). Either fluorescent reagent can be used for quantification and labeling studies. 3.5 PM Cholesterol Sequestration by ALOD4 3.5.1 PM Cholesterol Sequestration in CHO-K1 Cells

All experiments with CHO-K1 cells should be performed at 37
C in 8.8% CO2. 1. Set up CHO-K1 cells in medium A at a density of 6  104 cells/well in 48-well plates (day 0, see Note 9). 2. One day after, remove medium from each well and wash cells with 500 μL of PBS. 3. Add 200 μL of medium C containing varying amounts of purified ALOD4 or ALOD4(Mut) to each well. Either unlabeled or fluorescently labeled proteins can be used in this step. 4. Incubate 48-well plates for 1 h at 37
C. 5. Collect medium from each well. 6. Wash each well twice with 500 μL of PBS. 7. Add 200 μL of Buffer E to each well and place the 48-well plate on a shaker at room temperature for 20 min for lysis. 8. Mix equal aliquots of medium collected in step 5 and lysed cells from step 7 with 5 loading buffer in Eppendorf tubes and incubate tubes at 95
C for 10 min. 9. Subject above mixtures to either 10% or 15% SDS-PAGE (see Note 10).

Purification and Use of ALOD4, a Cholesterol-Binding Protein

161

10. Transfer electrophoresed proteins to nitrocellulose membranes using the Bio-Rad Trans Blot Turbo system. 11. Subject the nitrocellulose membranes to immunoblot analysis with an anti-His antibody (1:1000 dilution) for ALOD4, IgG-20B12 (2 μg/mL) for SREBP1, IgG-7D4 (10 μg/mL) for SREBP2, and IgG-4H4 (0.2 μg/mL) for Scap. 12. Visualize bound antibodies by chemiluminescence using a 1/5000 dilution of donkey anti-mouse IgG or a 1/2000 dilution of goat anti-rabbit IgG conjugated to horseradish peroxidase. 13. Expose membranes to Phoenix Blue X-Ray Film for 1–300 s or scan using an Odyssey FC Imager (Dual-Mode Imaging System; 2 min integration time) and analyze using Image Studio ver. 5.0. See Fig. 2 for an example of SREBP2 activation in CHO-K1 cells by ALOD4, but not by ALOD4(Mut). Increase in the cleaved nuclear form of SREBP-2 is a measure of activation of SREBP-2. 3.5.2 PM Cholesterol Sequestration in SV-589 cells

All experiments with SV-589 cells should be performed at 37
C in 5% CO2. 1. Day 0: Set up SV-589 cells in medium B at a density of 4  104 cells/well in 48-well plates.

Fig. 2 ALOD4 triggers activation of SREBP transcription factors. Immunoblot analysis of CHO-K1 cells after incubation with ALOD4 or ALOD4(Mut) proteins. On day 0, CHO-K1 cells were set up in medium A at a density of 3  104 cells/ well of 48-well plates. On day 2, medium was removed, cells were washed with 500 μL of PBS and then incubated with 200 μL of medium C with indicated concentrations of ALOD4 or ALOD4(Mut). After incubation for 1 h at 37
C, the medium was collected, and cells were harvested as described in Subheading 3. Equal aliquots of cells and media (10% of total) were subjected to immunoblot analysis or Coomassie staining as described in Subheading 3. P precursor form of SREBP2, N cleaved nuclear form of SREBP2. Figure adapted from [7]

162

Shreya Endapally et al.

2. Day 2: Remove medium from each well and wash the cells with 500 μL of PBS. 3. Add 200 μL of medium D containing varying amounts of purified ALOD4 or ALOD4(Mut) to each well. 4. Perform steps 3–8 as described in Subheading 3.5.1. 5. Subject the nitrocellulose membranes to immunoblot analysis with anti-His antibody (1/1000 dilution) for ALOD4, IgG-20B12 (2 μg/mL) for SREBP1, IgG-7D4 (10 μg/mL) for SREBP2, and IgG-4H4 (0.2 μg/mL) for Scap. 6. Perform steps 12 and 13 as described in Subheading 3.5.1. Typical results are similar to those observed for CHO-K1 cells in Fig. 2.

4

Notes 1. Care should be taken to avoid using a starter culture at OD600 > 0.8. Higher OD600 of starter cultures hinders growth rate of expression cultures and subsequent protein expression. Starter culture usually reaches OD600 ¼ 0.8 in 12 h. 2. OD600 ¼ 0.8 is optimal for protein expression. Care should be taken to not induce expression cultures at OD600 > 0.8 as this reduces protein yield. Culture usually reaches OD600 ¼ 0.5 in 4 h, at which point temperature can be reduced to 18
C. During an ~1 h cool-down period, the culture reaches OD600 ¼ 0.8. 3. 6 L of bacterial culture yields ~2 mg of protein. Number of liters of bacterial cultures used can be increased or reduced depending on the desired yield of protein. 4. We use a Branson Digital Sonifier (S-250, Fisher Scientific) to subject lysates to a 3 s on/3 s off sonication cycle at amplitude of 40% for 3 min. Following a 6 min cool-down period, repeat the cycle two more times. 5. Elution of ALOD4 with 300 mM imidazole (instead of a linear gradient) after washing with 50 mM imidazole results in protein yields with significant amounts of a nonspecific contaminant protein that migrates at ~50 kDa. 6. Concentrating the protein to lower volumes using Amicon Ultra-4 10 kDa cutoff centrifugal filter leads to protein aggregation. Employing the anion exchange column in the purification allows for further concentration without aggregation. 7. Extinction coefficients (ε280) for proteins can be obtained using “ProtParam tool” (expasy.org). Calculated ε280 values for ALOD4 and ALOD4(Mut) are 34,940 M1 cm1 and 33,630 M1 cm1, respectively.

Purification and Use of ALOD4, a Cholesterol-Binding Protein

163

8. Extinction coefficients of Alexa Fluor dyes can be found on Invitrogen website: Alexa Fluor 488 C5 maleimide (ε495 ¼ 71,000 M1 cm1), Alexa Fluor 647 C2 maleimide (ε651 ¼ 265,000 M1 cm1). 9. 48-well plates are a convenient format for setup of cells so that multiple variables can be tested at the same time. 10. Use 10% SDS-PAGE for analyzing SREBP and Scap and 15% SDS-PAGE for analyzing ALOD4. References 1. Brown MS, Goldstein JL (1997) The SREBP pathway: regulation of cholesterol metabolism by proteolysis of a membrane-bound transcription factor. Cell 89:331–340 2. Brown MS, Goldstein JL (2009) Cholesterol feedback: from Schoenheimer’s bottle to Scap’s MELADL. J Lipid Res 50(Suppl): S15–S27 3. Lange Y, Steck TL (2016) Active membrane cholesterol as a physiological effector. Chem Phys Lipids 199:74–93 4. Lange Y, Steck TL (1997) Quantitation of the pool of cholesterol associated with acyl-CoA: cholesterol acyltransferase in human fibroblasts. J Biol Chem 272:13103–13108 5. Das A, Goldstein JL, Anderson DD et al (2013) Use of mutant 125I-perfringolysin O to probe transport and organization of cholesterol in membranes of animal cells. Proc Natl Acad Sci U S A 110:10580–10585 6. Das A, Brown MS, Anderson DD et al (2014) Three pools of plasma membrane cholesterol and their relation to cholesterol homeostasis. elife 3:e02882 7. Infante RE, Radhakrishnan A (2017) Continuous transport of a small fraction of plasma membrane cholesterol to endoplasmic

reticulum regulates total cellular cholesterol. elife 6:e25466 8. Gay A, Rye D, Radhakrishnan A (2015) Switch-like responses of two cholesterol sensors do not require protein oligomerization in membranes. Biophys J 108:1459–1469 9. Chakrabarti RS, Ingham SA, Kozlitina J et al (2017) Variability of cholesterol accessibility in human red blood cells measured using a bacterial cholesterol-binding toxin. elife 6:e23355 10. Rong S, Cortes VA, Rashid S et al (2017) Expression of SREBP-1c requires SREBP-2mediated generation of a sterol ligand for LXR in livers of mice. elife 6:e25015 11. Yang J, Brown MS, Ho YK et al (1995) Three different rearrangements in a single intron truncate sterol regulatory element binding protein-2 and produce sterol-resistant phenotype in three cell lines. Role of introns in protein evolution. J Biol Chem 270:12152–12161 12. Ikeda Y, DeMartino GN, Brown MS et al (2009) Regulated endoplasmic reticulumassisted degradation of a polytopic protein: p97 recruits proteasomes to Insig-1 before extraction from membranes. J Biol Chem 284:34889–34900

Chapter 13 Measurement of Lysophospholipid Transport Across the Membrane Using Escherichia coli Spheroplasts Yibin Lin, Lei Zheng, and Mikhail Bogdanov Abstract In the inner membrane of Gram-negative bacteria lysophospholipid transporter (LplT) and the bifunctional acyl-acyl carrier protein (ACP) synthetase/2-acylglycerolphosphoethanolamine acyltransferase (Aas) form a glycerophospholipid remodeling system, which is capable of facilitating rapid retrograde translocation of lyso forms of phosphatidylethanolamine, phosphatidylglycerol, and cardiolipin across the cytoplasmic membrane. This coupled remodeling enzyme tandem provides an effective method for the measurement of substrate specificity of the lipid regeneration and lysophospholipid transport per se across the membrane. This chapter describes two distinct but complementary methods for the measurement of lysophospholipid transport across membrane using Escherichia coli spheroplasts. Key words Lysophospholipid flipping, Lysophospholipid transporter, Aas, Glycerophospholipid remodeling system, Thin-layer chromatography, Spheroplasts

1

Introduction Lysophospholipids (LPLs) are small bioactive lipid molecules characterized by a single carbon chain and a polar head group. Two subgroups can be distinguished: molecules containing the sphingoid base backbone (lysosphingolipids) and molecules containing the glycerol backbone (lysoglycerophospholipids). The LPL structure renders these lipids more hydrophilic and versatile than their corresponding phospholipids [1]. Distinct from their diacyl counterparts, these inverted cone-shaped molecules share physic characteristics of detergents, enabling modification of local membrane properties such as curvature and micellization [2]. In eukaryotic cells, it is now widely accepted that LPLs function as mediators through G-protein-coupled receptors (GPCRs) [3]. Intracellular or extracellular LPL transport across the membrane is required to activate corresponding GPCR, such as sphingosine-1-phosphate (S1P) [4], or for the uptake of bioactive substances, such as docosahexaenoic acid (DHA) [5]. The role of LPLs remains poorly

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_13, © Springer Science+Business Media, LLC, part of Springer Nature 2019

165

166

Yibin Lin et al.

characterized in bacteria. As a protective barrier, bacterial membranes are often stressed by harsh environments and challenged by external perturbations, caused by exposure to bile acids, different temperatures, or diverse phospholipases (PLAs). Bile acids and heat shock cause a dramatic increase in the lysophosphatidylethanolamine (LPE) level in the E. coli, Vibrio cholerae, and Yersinia pseudotuberculosis membranes, respectively [6–8]. Disrupting bacterial membranes by PLAs is one of the major host cell defense mechanisms used to kill an invading bacterium [9]. LPE can be taken by bacteria from exogenous sources [10] or a “silent” outer membrane phospholipase A1 (PldA) and lipid A palmitoyltransferase PagP with broad diacylated lipid specificity can be activated in stressed cells to form LPLs as products and by-products, respectively [11, 12]. LPL transport across the membrane and the phospholipid remodeling were reported to be important for maintaining bacterial membrane stability and bacterial survival [9], cell division and other functions [2]. Although the LPL turnover membrane is functionally important [2, 9, 13], the mechanism of LPL turnover and its physiological significance remain obscure and require further investigation. Development of a high-efficiency method for measurement of LPL transport across the membrane is required for addressing some of these outstanding questions. In the inner membrane of Gram-negative bacteria, LplT was identified as lysophospholipid transporter, which is capable of facilitating rapid retrograde translocation of LPE, lysophosphatidylglycerol (LPG), monoacyl-cardiolipin (MCL), diacyl-cardiolipin (DCL) across the cytoplasmic membrane [13]. Aas is a bifunctional enzyme catalyzing acyl transfer to LPL, generating diacyl form of the phospholipid [14, 15]. The LplT-Aas system has been reported to translocate and facilitate remodeling of all three bacterial major phospholipids including PE, PG, and CL with comparable translocation and remodeling efficiencies [13]. The LplT-Aas system in Gram-negative bacteria provides an efficient method for measuring LPL transport across the membrane (Fig. 1). Two methods have been developed and used to monitor LPL transport across the membrane based on the mechanism of functioning of LplT-Aas system in Gram-negative bacteria. The general design of the experiments to measure the LPL transport across the membrane is outlined in Fig. 2. 32P-labeled diacyl phospholipids are prepared by growing various recombinant E. coli strains (Table 1) in LB broth containing 5 μCi/mL 32P-orthophosphate. 32P-labeled PE is synthesized in E. coli strain UE54 (MG1655 lpp-2Δara714 rcsF:: mini-Tn10 cam pgsA::FRT-kan-FRT) containing up to 95% of PE [16]. 32P-labeled PG and CL are made in the E. coli AL95 strain (pss 93::kan lacY::Tn9), lacking the ability to synthesize PE. This strain contains only acidic phospholipids, such as PG and CL (45–50%, respectively) and PA which altogether build a negatively charged membrane consisting of 100% of negatively charged phospholipids

Lysophospholipid Flipping

167

Fig. 1 A topological diagram of the LPL biogenesis and remodeling in the Gram-negative bacterial inner membrane. Only lysophospholipids generated from diacylphospholipids by PLA-mediated deacylation or lipoprotein acyltransferase (Lnt)-catalyzed transacylation reactions within inner membrane of Escherichia coli are shown. LplT flips the resultant lysophospholipids across the inner membrane for reacylation by Aas to the cytoplasmic side. Figure and legend are freely reproduced with modification from reference [13]

Fig. 2 Two methods for the detection of LPL transport across membrane using E. coli spheroplasts. (a) TLCbased translocation assay. LPL is mixed with spheroplasts prepared from E. coli Δlplt cells expressing plasmid-borne LplT protein. E. coli Δlplt cells harboring an empty vector or Δaas cells are used as a negative control. The reactions are stopped by 0.5 M NaCl in 0.5 N HCl at the indicated times, and total lipids are extracted and then separated by TLC. The substrates and acylated products shown in TLC plate can be quantified using a phosphorimaging system. (b) Silicone oil-spin method. LPL is mixed with spheroplasts prepared from E. coli Δlplt/aas cells expressing LplT or Δlplt/aas cells harboring an empty vector (control). After incubation, samples are centrifuged through a layer of silicone oil to stop the reaction and separate spheroplasts from non-transported LPLs. After centrifugation, aliquots of the perchloric acid phase are removed and the radioactivity is counted

168

Yibin Lin et al.

Table 1 Bacterial strains Stain

Genotype

Grow condition

Application

References

UE54

MG1655 lpp-2Δara714 rcsF::miniTn10 cam pgsA::FRT-kan-FRT

LB

95% PE

[16]

AL95

pss 93::kan lacY::Tn9

LB + 50 mM MgCl2 40% PG, 45% CL [17]

AL95

pss 93::kan lacY::Tn9; pAC-PCSlp-SpGm

LB + 50 mM MgCl2 PC

[18]

LB

[19]

SM2-1 plsC(Ts)

20% PLA

[17]. Strain AL95 carrying plasmid pAC-PCSlp-Sp-Gm is used to prepare 32P-labeled phosphatidylcholine (PC) from cells synthesizing this foreign lipid in an amount equivalent to that of PE (75%) [18]. Radiolabeled LPE, LPC, LPG, MCL, and DCL are prepared by digestion of the corresponding purified 32P-labeled phospholipids with phospholipase A2 (PLA2). Lysophosphatidic acid (LPA) is made in the temperature-sensitive E. coli strain SM2-1 (plsC(Ts)) accumulating this LPL (up to 20%) after a shift to growth at nonpermissive temperature [19]. Lysophospholipid transporters are expressed in the appropriate E. coli strains, and the spheroplasts are made using the basic lysozyme-EDTA method [20]. For thinlayer chromatography (TLC)-based LPL translocation assay, radioactive LPLs are mixed with cold counterparts to fit an apparent μM binding affinity [13] and resuspended in ethanol. Transport assays are initiated by adding substrates into spheroplast solutions. At the indicated time, reactions are terminated by adding a chloroform/ methanol solvent system. The total lipids are extracted and separated by TLC. The Aas-dependent formation of diacyl form of lipids from corresponding lyso form can be quantified from TLC plate, which reflects those LPLs that has been translocated across the membrane (Fig. 3a, b). This is an indirect approach to measure LPL transport across the membrane. However, this technique, which requires coupled functioning of lysophospholipid transporter as well as acyltransferase/acyl-ACP synthetase has its own limits due to two facts [13]: (1) Substrates flipped into the spheroplasts are indistinguishable from non-transported substrates monitored by TLC because to maintain stability the spheroplasts were not collected by centrifugation before lipid extraction; (2) Whether LplT transport activity is facilitated by downstream Aas-mediated continuous reacylation of transported LPLs driving their downhill uptake is unknown. Although this assay can be used to test substrate specificity of LplT/Aas transport/acylation regeneration system in order to characterize LplT transport activity per se, the transport reaction should be separated from assay mixture by

Lysophospholipid Flipping

169

Fig. 3 LPE transport across the E. coli inner membrane measured by TLC-based translocation assay and silicone oil-spin method. (a) TLC for separating substrates and acylated products. (b) Quantification of the substrates and acylated products shown in (a). (c) Transport assay of LPE by LplT protein using silicone oil-spin method. For detailed experimental procedure see Fig. 2 and the main text. Figure and legend are freely reproduced with modification from [13]

centrifugation through a layer of silicon oil. Due to their different sedimentation rates and miscibility with oil, non-transported lipids are retained in silicon oil and the upper layer, whereas the pelleted spheroplasts containing transported substrates are assessed by scintillation counting to measure LplT transport kinetics (Fig. 3c). To eliminate any coupling effect of Aas, the transport assay should be performed in Δaas mutant expressing LplT from the plasmid. In this transport assay, LPLs (premixed “cold” and “hot”) were added to the spheroplasts to a desired final concentration as indicated and incubated at 37
C for 30 min. After incubation, samples were centrifuged through a layer of silicone oil [21] to stop the reaction and separate spheroplasts from non-transported LPLs. After centrifugation, samples of the perchloric acid phase were removed and the radioactivity was measured by a liquid scintillation counter. The silicon oil-spin method does not require functional acyltransferase/ acyl-ACP synthetase. These two methods each have strengths and limitations, yet combining them make it possible to accurately measure LPL transport across membrane. In this chapter, we describe these two methods. Although this protocol was developed based on a Gram-negative bacteria glycerophospholipid remodeling system, it can be extended to any other lysophospholipid transporters that can be expressed in Gram-negative bacteria inner

170

Yibin Lin et al.

membrane. TLC-less silicone oil-spin method can be adapted to any membrane protein and expression system to study kinetics of LPL translocation.

2 2.1

Materials Lipid Extraction

1. 0.5 M NaCl in 0.5 N HCl solution. 2. Methanol/chloroform solvent system (2:1, v/v). 3. Vortex equipped with microtube foam rack for multiple polyallomer tubes. 4. Tabletop centrifuge. 5. Savant Speed Vacuum Concentrator.

2.2

TLC

1. EMD Millipore TLC Silica Gel 60 F254. 2. TLC developing tanks with lids for use with TLC plates up to 20  20 cm. 3. Plain Whatman 3 M filter paper sheets. 4. Oven for activation of TLC. 5. Tank with regenerable indicating desiccant. 6. 1.2% (w/v) boric acid in ethanol/water (1:1, v/v). 7. Developing solvent system: Chloroform/methanol/water/ ammonia (39% w/v, 8.56 N), (60:37.5:3:1, v/v/v/v). 8. X-Ray developing cassettes. 9. Kodak HR GP plate. 10. Imaging system such as a Fluor-S Max MultiImager (Bio-Rad Laboratories) or similar.

2.3 Isolation of Lipids from TLC

1. X-Ray Film such as Thermo Scientific CL-XPosure Film.

2.4

1. Luria–Bertani (LB) liquid medium: 10 g/L tryptone, 5 g/L yeast extract, and 10 g/L NaCl.

Cell Culture

2. X-ray Film Auto Processor.

2. Luria–Bertani (LB) agar plate: 10 g/L tryptone, 5 g/L yeast extract, 10 g/L NaCl, and 16 g/L agar. 3. 2 M MgCl2 stock solution sterilized by passing through a 0.22 μm syringe filter. 4. 10% arabinose stock solution sterilized by passing through a 0.22 μm syringe filter and stocked fresh. 5. 1 M isopropyl β-D-1-thiogalactopyranoside (IPTG) stock solution sterilized by passing through a 0.22 μm syringe filter.

Lysophospholipid Flipping

171

6. 100 mM choline chloride stock solution sterilized by passing through a 0.22 μm syringe filter. 2.5 Preparation of 32P-Labeling LPLs

1. E. coli strain UE54 (MG1655 lpp-2Δara714 rcsF::mini-Tn10 cam pgsA::FRT-kan-FRT) is used to make 32P-labeled PE. E. coli AL95 strain (pss 93::kan lacY::Tn9), lacking the ability to synthesize PE, is used to make 32P-labeled PG and CL. Strain AL95 carrying plasmid pAC-PCSlp-Sp-Gm is used to prepare 32P-labeled phosphatidylcholine (PC). The temperature-sensitive E. coli strain SM2-1 (plsC(Ts)) is used to make lysophosphatidic acid (LPA) (see Table 1). 2. Phosphorus-32 radionuclide orthophosphoric acid. 3. Phospholipase A2 (PLA2) from porcine pancreas or venom. 4. Digestion solution: 0.1 M HEPES-NaOH, pH 7.5, 0.1 M KCl, 10 mM CaCl2 and 1% DDM (n-Dodecyl-β-D-maltoside).

2.6 Preparation of E. coli Spheroplasts

1. Spheroplast solution A: Add 125 μL of 0.1 M Tris–HCl, pH 8, 114 μL of 2 M sucrose, 21 μL of 1% EDTA, pH 7, and 21 μL of 0.5 mg/mL lysozyme into a 1.5 mL Eppendorf tube, and complete with ddH2O to 0.5 mL. 2. Spheroplast solution B: Add 125 μL of 0.1 M Tris–HCl, pH 8 and 114 μL of 2 M sucrose into a 1.5 mL Eppendorf tube, and complete with ddH2O to 0.5 mL. 3. 10 mM MgCl2, 0.75 M sucrose solution.

2.7 TLC-based LPL Translocation Assay

1. Lipids: 1-oleoyl-2-hydroxy-sn-glycero-3-phosphoethanolamine (18:1 Lyso PE), 1-oleoyl-2-hydroxy-sn-glycero-3-phospho-(1’-rac-glycerol) (sodium salt) (18:1 Lyso PG), 1-oleoyl2-hydroxy-sn-glycero-3-phosphocholine (18:1 Lyso PC), 1-oleoyl-2-hydroxy-sn-glycero-3-phosphate (sodium salt) (18:1 Lyso PA), monolyso heart CA, dilysocardiolipin (heart, bovine) (sodium salt). 2. Imaging system such as a Fluor-S Max MultiImager (Bio-Rad Laboratories) or similar. 3. Quantity One 1-D Analysis Software (Bio-Rad Laboratories). 4. Water bath.

2.8 Silicone Oil-Spin Method

1. Silicone oil (d ¼ 1.05). 2. 22% perchloric acid. 3. LKB Wallac Liquid Scintillation Counter (Model 2109) or similar. 4. E. coli dry weight: 2.8  1013 g per cell (adopted from Bio-Rad Laboratories).

172

3 3.1

Yibin Lin et al.

Methods Lipid Extraction

1. Resuspend 1 g of cell pellet, 0.5 mL of reaction solution, or 1 g of powder in 1.5 mL of 0.5 M NaCl in 0.5 N HCl solution and transfer it to a 15 mL falcon tube (see Note 1). 2. Add 4.5 mL of methanol/chloroform (2:1, v/v) mixture, and vortex for 20 min. 3. After vortexing, add another 1.5 mL of 0.5 M NaCl in 0.5 N HCl solution and continue to vortex for 5 min. 4. Centrifuge at 2000  g for 5 min. 5. Discard upper water-methanol phase, carefully transfer lower chloroform lipid phase to a new Eppendorf tube. 6. Dry the lipid extract in a Savant Speed Vacuum Concentrator.

3.2

TLC

1. Line the inner walls of a developing tank with Whatman 3 M paper, which absorbs the solvent and helps to saturate a tank by providing more surface area for evaporation of vapor rising from both the sides of tank in an even manner. Use a bench but not a hood to saturate a tank (see Note 2). 2. Prepare developing solvent system immediately before use. Mix it in a tightly closed 150–200 mL bottle and then transfer it into the tank. Seal the tank and wait about 2 h, which allows the tank atmosphere to be thoroughly saturated with solvent vapor before using. 3. Impregnate Silica Gel G thin-layer plates for 1 min in 1.2% (w/v) boric acid in ethanol/water (1:1, v/v) and let the plates standing on the absorbent paper towels for 5 min. Dry the plates 5 min on air, and then put the plates into the oven (90
C) for 60 min (see Note 2). Remove the plates from the oven and desiccate them for 15 min (see Note 2). 4. Subject 50 μL of lipid samples onto Silica Gel G thin-layer plates in their predetermined positions (pencil marks) along with a line drawn 2 cm above the bottom of the plate with TLC spotting capillary tubes. Leave at least 1 cm between sample positions and the edge of the silica plate. 5. Once fully loaded, position the plate in the solvent in the tank, and run the TLC until the solvent front reaches 5 mm below the top of the TLC plate. 6. Quickly remove the plate and allow them to dry in the air (under a fume hood) before exposing to an X-ray or Bio-Rad developing screens for several hours to several days.

Lysophospholipid Flipping

173

7. The phospholipid bands were visualized by developing the film, and bands corresponding to 32P-labeled phospholipids are quantified using the Quantity One software (Bio-Rad Laboratories). 3.3 Purification of Lipids from TLC

1. Perform TLC for 32P-labeled lipids mixture following the method as shown in Subheading 3.2. 2. Expose air-dried TLC plate to an X-ray film for 2 h. 3. Develop the film, and use this film to localize/map the bands on the TLC plate that correspond to the desired lyso- or diacyl phospholipid. 4. Wet the phospholipid spot with water, and then scrape it gently and carefully transfer the powder containing the wanted lipid into a 15 mL falcon tube. 5. Perform lipid extraction following the method as shown in Subheading 3.1 (see Note 1).

3.4 Preparation of 32P-Labeling LPLs 3.4.1 Preparation of 32PLabeled LPE

1. To recover E. coli strain UE54 from glycerol stock, open the tube and use a sterile loop, toothpick, or pipette tip to scrape some of the frozen bacteria off of the top. Do not let the glycerol stock unthaw! Streak the bacteria onto an LB agar plate. 2. Grow bacteria overnight at 37
C. 3. Next day, pick a single colony and grow bacteria overnight in 20 mL of LB broth containing 5 μCi/mL 32P-orthophosphate (final) at 37
C in a 50 mL falcon tube (see Note 3). 4. Overnight growth cells were harvested by centrifugation at 4000  g for 10 min. 5. Perform lipid extractions following the method described in Subheading 3.1. 6. Dry the lipid extract in a Savant Speed Vacuum Concentrator. 7. Generate LPE from PE: Redissolve the dried lipid extract with 0.5 mL of digestion solution. Sonicate in ice by 59-s bursts (amplitude 16%). Add 10 U of PLA2 from porcine pancreas and incubate at 37
C for overnight with shaking. 8. Extract lipids following the same method as in step 5. 9. Dry the lipid extract in a Savant Speed Vacuum Concentrator. Redissolve the dried lipid extract with 50 μL of chloroform. 10. Perform TLC to analyze results of PLA2 digest reactions (see Note 4) following the method as shown in Subheading 3.2. 11. Purify LPE from TLC following the method as shown in Subheading 3.3.

174

Yibin Lin et al.

3.4.2 Generation of LPG, D-CL, and M-CL

1. Revive an E. coli strain AL95 from glycerol stock on LB-agar supplemented with 50 mM MgCl2 (from 2.5 M sterile stock) since these cells strictly require a millimolar amount of magnesium for growth (see Note 3). 2. Grow bacteria overnight at 37
C. 3. Next day, pick a single colony and grow bacteria overnight in 20 mL of LB broth containing 5 μCi/mL 32P-orthophosphate (final concentration) and 50 mM MgCl2 at 37
C in a 50 mL falcon tube. 4. Overnight growth cells were harvested by centrifugation at 4000  g for 10 min. 5. Perform lipid extractions following the method as shown in Subheading 3.1. 6. Dry the lipid extract in a Savant Speed Vacuum Concentrator. Redissolve the dried lipid extract with 50 μL of chloroform. 7. Purify PG and CL from the TLC plate following the method as shown in Subheading 3.3. 8. Generation of LPG from PG: Redissolve the dried PG with 0.5 mL of digestion solution. Sonicate in ice by 59-s bursts (amplitude 16%). Add 10 U of PLA2 from porcine pancreas and incubate at 37
C overnight under shaking. 9. Generate D-CL and M-CL from CL: Redissolve the dried CL with 0.5 mL of digestion solution. Sonicate in ice by 59-s bursts (amplitude 16%). Add 10 U of PLA2 from porcine pancreas. To make D-CL, incubate the reaction mixture at 37
C for 2 h under shaking. To make M-CL, keep the reaction mixture at 37
C overnight under shaking. 10. Extract lipids following the same method as shown in step 5. 11. Dry the lipid extract in a Savant Speed Vacuum Concentrator. Redissolve the dried lipid extract with 50 μL of chloroform. 12. Perform TLC to analyze results of PLA2 digest reactions following the method as shown in Subheading 3.2 (see Note 4). 13. Purify LPLs from TLC following the method as shown in Subheading 3.3.

3.4.3 Generation of LPC

1. Revive an E. coli strain AL95 carrying plasmid pAC-PCSlp-SpGm from glycerol stock on LB-agar plates supplemented with 50 mM MgCl2 as described in step 1 of Subheading 3.4.2 (see Note 3). 2. Grow bacteria overnight. 3. Next day, pick a single colony and grow bacteria overnight in 4 mL of LB broth supplemented with 50 mM MgCl2 at 37
C in a 10 mL plastic culture tube.

Lysophospholipid Flipping

175

4. Next day, dilute overnight culture 1/100 into 20 mL of LB broth supplemented with 0.2% arabinose, 2 mM choline, and 5 μCi/mL 32P-orthophosphate and culture cells at 37
C in a 50 mL falcon tube (see Note 3). 5. When cell culture reaches an OD600 of 0.6–0.8, harvest the cells by centrifugation at 4000  g for 10 min. 6. Perform lipid extractions following the method as shown in Subheading 3.1. 7. Dry the lipid extract in a Savant Speed Vacuum Concentrator. 8. Generate LPC from PC: Redissolve the dried lipid extract with 0.5 mL of digestion solution. Sonicate in ice by 59-s bursts (amplitude 16%). Add 10 U of PLA2 from porcine pancreas and incubate at 37
C overnight under shaking. 9. Extract lipids following the same method as in step 6. 10. Dry the lipid extract in a Savant Speed Vacuum Concentrator. Redissolve the dried lipid extract with 50 μL of chloroform. 11. Perform TLC to analyze the result of PLA2 digest reactions following the method as shown in Subheading 3.2 (see Note 4). 12. Purify LPC from TLC following the method as shown in Subheading 3.3. 3.4.4 Generation of LPA

1. Recover temperature-sensitive E. coli strain SM2-1 from glycerol stock. 2. Grow bacteria overnight at 30
C. 3. Next day, pick a single colony and grow bacteria overnight in 4 mL of LB broth at 30
C in a 10 mL plastic culture tube. 4. Next day, dilute overnight culture to 1/100 into 100 mL of LB broth and culture cells at 30
C in a 250 mL glass Erlenmeyer Flask. 5. When cell culture reaches an OD600 of 0.6–0.8, add 5 μCi/mL 32 P-orthophosphate to the culture, and culture cells at 42
C for 2 h (see Note 5). 6. Harvest cells by centrifugation at 4000  g. 7. Perform lipid extractions following the method as described in Subheading 3.1. 8. Dry the lipid extract in a Savant Speed Vacuum Concentrator. Redissolve the dried lipid extract with 50 μL of chloroform. 9. Perform TLC to analyze the yield of LPA following the method as described in Subheading 3.2. 10. Purify LPA from TLC following the method as described in Subheading 3.3.

176

Yibin Lin et al.

3.5 Preparation of E. coli Spheroplasts

1. Revive E. coli WT or ΔlplT, Δaas, ΔlplT/aas mutant strains from glycerol stock as described above. 2. Grow bacteria overnight at 37
C. 3. Next day, pick a single colony and grow bacteria overnight in 4 mL of LB broth supplemented with antibiotic if needed at 37
C in a 10 mL plastic culture tube. 4. Next day, dilute overnight culture to 1/100 into 20 mL of LB broth supplemented with 0.1 mM IPTG and antibiotic if needed, and culture cells at 37
C in a 50 mL falcon tube for 2 h (see Note 6). 5. Harvest the cells by centrifugation at 4000  g. 6. For 3  109 cells, resuspend the pellet in a total of 500 μL of spheroplast solution A and incubate it at 25
C. 7. Monitor the formation of spheroplasts: every 5 min withdraw and inject 100 μL of reaction mixture (step 6) into 2 mL of water or a solution of 10 mM MgCl2, 0.75 M sucrose, and compare the OD600 of each other. The spheroplasts are stable in the solution of 10 mM MgCl2, 0.75 M sucrose (OD600 does not change), but will rupture (break) in plain water (OD600 drops immediately) (see Note 7). 8. Pellet osmotically unstable spheroplasts by centrifugation at 6000  g. 9. Wash spheroplasts three times with spheroplast solution B by resuspending the pellet in the same solution and re-pelleting the sample. 10. Resuspend spheroplasts in solution B at 10 mg/mL of total protein.

3.6 TLC-based LPL Translocation Assay

1. Weight “cold” LPLs and dissolve them in chloroform. 2. Dissolve radioactive LPLs with chloroform and mix them with their cold counterparts. 3. Dry the lipid mixture in a Savant Speed Vacuum Concentrator and resuspend it in ethanol to a final concentration of 200 μM by sonication (59-s bursts, amplitude 16%). 4. Add 10 μM (final concentration) of substrates into spheroplast solutions and incubate in a 37
C water bath (see Note 8). 5. At the indicated time of 0, 5 min, 10 min, 30 min, and 60 min, transfer 0.5 mL of reaction solution into a 15 mL falcon tube containing 0.5 M NaCl in 0.5 N HCl to stop the reaction. Extract lipids following the method as described in Subheading 3.1 (see Note 8). 6. Perform TLC as shown as described in Subheading 3.2 for LPL translocation analysis.

Lysophospholipid Flipping

177

7. Expose dried TLC to the imaging screen for appropriate time and store the image using a Molecular Imager FX (Bio-Rad Laboratories) (Fig. 3a). 8. Process and quantify the stored images using Quantity One software (Bio-Rad Laboratories) (Fig. 3b). 3.7 Silicone Oil-Spin Method

1. Prepare “cold” and “hot” lysophospholipid mixture following the method as shown in Subheading 3.5. 2. Add different concentration of substrates into spheroplast solutions and incubate in a 37
C water bath for 30 min. 3. After incubation, transfer 0.5 mL samples from the reaction mixtures and layer them onto 0.15 mL of 22% perchloric acid solution and 0.50 mL of silicone oil (d ¼ 1.05) in microcentrifuge tubes. Centrifuge through the silicone oil in an Eppendorf microcentrifuge at 14,000  g for 5 min at room temperature (see Note 9). 4. After centrifugation, discard the upper phase, carefully transfer the lower phase into a new Eppendorf tube. Use 1 mL and 200 μL Gilson pipetmans to remove upper and withdraw lower phases, respectively. 5. Measure the radioactivity of the lower phase using a liquid scintillation counter. 6. Normalize the data to nmol/g total protein/h based on the specific radioactivity of the lipids and considering a E. coli dry weight of 2.8  1013 g per cell (Fig. 3c).

4

Notes 1. To extract lipids from 1 g of cell pellet, 0.5 mL of reaction solution, or 1 g of powder, 1.5 mL of 0.5 N NaCl in 0.1 N HCl, 4.5 mL of methanol/chloroform (2:1, v/v), and 1.5 mL of 0.5 N NaCl in 0.1 N HCl (1:3:1, v/v/v) solution should be added sequentially. Dry material, such as a silica gel powder scraped from TLC plate, should be first saturated with 0.5 N NaCl in 0.1 N HCl alone by vigorous vortexing. Indeed, if silica flakes/powder are dry and chloroform/methanol is added first, the efficiency of extraction is weaker. Usually, all volumes can be scaled up or down depending on demand (number of cells or silica gel powder). Lower phase should be collected carefully without contamination from the upper phase. To extract lipids from wet or liquid solution (cell pellet or suspension), vortexing for 20 min should be enough. 2. The tank should be prepared 2 h before use to allow the chamber air to be thoroughly impregnated with solvent vapor. TLC plate should be heated for at least 1 h, but should

178

Yibin Lin et al.

not be heated for too long, such as overnight. Activated TLC adsorbs water vapor from the air and becomes hydrated. Therefore, after heating in an oven, a TLC plate should be desiccated for 10 min in the tank with regenerable indicating desiccant. An activated TLC can be kept in a cabinet with continuous air flow dehumidified by passing through a polycarbonate drying column with activated regenerable indicating desiccant. The column should be connected in-line between a drying cabinet and compressed air line or gas cylinder. Developing solvent for TLC should be freshly prepared. Only 28–30% ammonia hydroxide (e.g., highest concentration possible) should be used to prepare a solvent system. The concentrations of ammonia and boric acid solution can change after a long-time use, which may affect the mobility and separation of individual phospholipid TLC. The solutions should to be properly renewed. Before exposing to an X-ray screen, TLC plate should be dried completely; otherwise, chloroform may damage the Bio-Rad imaging screen. 3. E. coli UE54 should be always grown overnight in order to receive the cells with the maximal amount of PE (95–97%). PE-deficient cells should be always grown on LB-agar supplemented with 50 mM MgCl2. Although in AL95 strain carrying plasmid pAC-PCSlp-Sp-Gm zwitterionic PE is substituted by the same amount of net neutral PC, these cells still require millimolar Mg for growth. These two strains should be checked out routinely by growth on LB-agar plates with and without 50 mM MgCl2 (no growth without Mg should be observed). To fully express pcsA gene and get the cells with 75% PC, the cells should be outgrown first overnight in the presence of 50 mM MgCl2 and then diluted in the morning to 1/100 or more (to start growth approximately from an OD600 of 0.025). 4. PLA2 from porcine pancreas has a higher activity against phospholipids. Under the experimental condition described in this protocol, PLA2 can completely hydrolyze PE, PG, and PC substrate into the corresponding LPLs. As PLA2 can use CL, triacyl-CL, diacyl-CL, and monoacyl-CL as substrates, TLC, as shown in Subheading 3.2, was used to monitor the duration of reaction. Triacyl-CL, diacyl-CL, and monoacyl-CL can be purified from TLC plates as shown in Subheading 3.3. 5. The temperature-sensitive E. coli strain SM2-1 can be used to generate LPA, but the yield is relatively low (about 10% of total lipids). To get more LPA, the number of cells should be scaled up, and the volumes of solvent for extracting lipids from cell pellet and TLC spot should be adjusted accordingly. 6. The expression of plasmid-borne membrane proteins (LplT, Aas) for functional assays should be used with caution because

Lysophospholipid Flipping

179

the overexpression of integral membrane proteins often causes cell death and physiological interpretations problematic by overloading the cytoplasmic membranes, by disrupting the membrane integrity, or “jamming” a translocon and triggering the accumulation of membrane protein into newly made intracytoplasmic membranes [22]. To overcome these limitations, the use of lower-copy number plasmids, the induction of protein expression for short time, and the omission of inducer are recommended for the expression of target membrane proteins for functional assays. In this protocol, 0.1 mM IPTG was used to produce the desired membrane proteins in the cytoplasmic membrane rather than in intracytoplasmic membranes and inclusion bodies in a misfolded state. 7. To make spheroplasts, cells should be collected from middle log phase. Usually, the volumes of spheroplast solution A and B can be scaled up or down on demand. As spheroplasts are unstable, they should be always freshly prepared. All operations must be gentle to avoid spheroplasts rupture. Spheroplast formation and stability should be thoroughly monitored nephelometrically by comparing the OD600 of a 100 μL spheroplast solution with 2 mL of either plain water or a solution of 10 mM MgCl2, 0.75 M sucrose, respectively. 8. For TLC-based translocation assay, the final concentration of ethanol should not be higher than 5%. As spheroplasts are fragile structures, we do not recommend incubating the spheroplasts at 37
C for more than 1 h. 9. For silicone oil-spin method, the density of silicone oil is a critical factor for ensuring the success and reproducibility. It must be greater than that of the incubation medium used, but lower than that of the perchloric acid or the cells themselves. Silicone oil (d ¼ 1.05) and 22% perchloric acid are well suited for experiments with Gram-negative bacteria cells. Different cell types with unknown densities can be determined empirically by appropriate centrifugation conditions [23].

Acknowledgments This work was supported by NIH grants R01GM097290 and R01GM098572 (to L. Z.) and European Union Marie Skłodowska-Curie Grant H2020-MSCA-RISE-2015-690853 and NATO Science for Peace and Security Programme-SPS 985291 and Program of Competitive Growth of Kazan Federal University (to M. B.).

180

Yibin Lin et al.

References 1. Offermanns S, Rosenthal W (eds) (2008) Encyclopedia of molecular pharmacology, 2nd edn. Springer, Berlin; New York 2. Zheng L, Lin Y, Lu S et al (2017) Biogenesis, transport and remodeling of lysophospholipids in Gram-negative bacteria. Biochim Biophys Acta 1862:1404–1413 3. Makide K, Kitamura H, Sato Y et al (2009) Emerging lysophospholipid mediators, lysophosphatidylserine, lysophosphatidylthreonine, lysophosphatidylethanolamine and lysophosphatidylglycerol. Prostaglandins Other Lipid Mediat 89:135–139 4. Kawahara A, Nishi T, Hisano Y et al (2009) The sphingolipid transporter spns2 functions in migration of zebrafish myocardial precursors. Science 323:524–527 5. Nguyen LN, Ma D, Shui G et al (2014) Mfsd2a is a transporter for the essential omega-3 fatty acid docosahexaenoic acid. Nature 509:503–506 6. Kern R, Joseleau-Petit D, Chattopadhyay MK et al (2001) Chaperone-like properties of lysophospholipids. Biochem Biophys Res Commun 289:1268–1274 7. Giles DK, Hankins JV, Guan Z et al (2011) Remodelling of the Vibrio cholerae membrane by incorporation of exogenous fatty acids from host and aquatic environments. Mol Microbiol 79:716–728 8. Davydova L, Bakholdina S, Barkina M et al (2016) Effects of elevated growth temperature and heat shock on the lipid composition of the inner and outer membranes of Yersinia pseudotuberculosis. Biochimie 123:103–109 9. Lin Y, Bogdanov M, Lu S et al (2018) The phospholipid-repair system LplT/Aas in gram-negative bacteria protects the bacterial membrane envelope from host phospholipase A2 attack. J Biol Chem 293:3386–3398 10. Kol MA, Kuster DW, Boumann HA et al (2004) Uptake and remodeling of exogenous phosphatidylethanolamine in E. coli. Biochim Biophys Acta 1636:205–212 11. Malinverni JC, Silhavy TJ (2009) An ABC transport system that maintains lipid asymmetry in the gram-negative outer membrane. Proc Natl Acad Sci U S A 106:8009–8014 12. Bishop RE (2008) Structural biology of membrane-intrinsic beta-barrel enzymes: sentinels of the bacterial outer membrane. Biochim Biophys Acta 1778:1881–1896

13. Lin Y, Bogdanov M, Tong S et al (2016) Substrate selectivity of lysophospholipid transporter LplT involved in membrane phospholipid remodeling in Escherichia coli. J Biol Chem 291:2136–2149 14. Hsu L, Jackowski S, Rock CO (1991) Isolation and characterization of Escherichia coli K-12 mutants lacking both 2-acyl-glycerophosphoethanolamine acyltransferase and acyl-acyl carrier protein synthetase activity. J Biol Chem 266:13783–13788 15. Harvat EM, Zhang YM, Tran CV et al (2005) Lysophospholipid flipping across the Escherichia coli inner membrane catalyzed by a transporter (LplT) belonging to the major facilitator superfamily. J Biol Chem 280:12028–12034 16. Shiba Y, Yokoyama Y, Aono Y et al (2004) Activation of the Rcs signal transduction system is responsible for the thermosensitive growth defect of an Escherichia coli mutant lacking phosphatidylglycerol and cardiolipin. J Bacteriol 186:6526–6535 17. Bogdanov M, Heacock PN, Dowhan W (2002) A polytopic membrane protein displays a reversible topology dependent on membrane lipid composition. EMBO J 21:2107–2116 18. Bogdanov M, Heacock P, Guan Z et al (2010) Plasticity of lipid-protein interactions in the function and topogenesis of the membrane protein lactose permease from Escherichia coli. Proc Natl Acad Sci U S A 107:15057–15062 19. Coleman J (1990) Characterization of Escherichia coli cells deficient in 1-acyl-sn-glycerol-3phosphate acyltransferase activity. J Biol Chem 265:17215–17221 20. HR K (1971) Preparation and characterization of bacterial membranes. Methods Enzymol XXII:99–120 21. Bogdanov M, Dowhan W (1995) Phosphatidylethanolamine is required for in vivo function of the membrane-associated lactose permease of Escherichia coli. J Biol Chem 270:732–739 22. Bogdanov M (2017) Mapping of membrane protein topology by substituted cysteine accessibility method (SCAM). Methods Mol Biol 1615:105–128 23. Tyson CA, Frazier JM (eds) (1994) In vitro toxicity indicators. Methods in toxicology. Academic, San Diego

Chapter 14 Preparation of Proteoliposomes with Purified TMEM16 Protein for Accurate Measures of Lipid Scramblase Activity Janine Denise Brunner and Stephan Schenck Abstract The distribution of different lipid species between the two leaflets is tightly regulated and underlies the concerted action of distinct catalytic entities. While flippases and floppases establish membrane asymmetry, scramblases randomize the lipid distribution and play pivotal roles during blood clotting, apoptosis, and in processes such as N-linked glycosylation of proteins. The recent discovery of TMEM16 family members acting as scramblases has led to an increasing demand for developing protocols tailored for TMEM16 proteins to enable functional investigations of their scrambling activity. Here we describe a protocol for the expression, purification, and functional reconstitution of TMEM16 proteins into preformed liposomes and measurement of their scrambling activity using fluorescence-labeled lipid derivatives. The reconstitution involves extrusion of liposomes through a membrane, destabilization of liposomes using Triton X-100, and stepwise detergent removal by adsorption on styryl-beads. The scrambling assay is based on the selective bleaching of nitrobenzoxadiazol fluorescent lipids on the outer leaflet of liposomes by the membraneimpermeant reducing agent sodium dithionite. The assay allows conclusions on the substrate specificity and on the kinetics of the transported lipids as shown with the example of a Ca2+-activated TMEM16 scramblase from the fungus Nectria haematococca (nhTMEM16). Key words Sodium dithionite, Scrambling assay, TMEM16/Anoctamin, Reconstitution, NBD

1

Introduction The plasma membrane of typical animal cells shows a pronounced asymmetry in the lipid composition of the inner and outer leaflet [1–3]. This asymmetry plays a pivotal role for the functioning of the cell [4, 5]. The majority of phosphatidylserine (PS) and phosphatidylethanolamine (PE) locate to the cytosolic face, whereas the major lipids phosphatidylcholine (PC) and sphingomyelin (SM) are abundant in the outer leaflet [6, 7]. This gradient is utilized by the cell for immediate signaling cues by rapid equilibration of the asymmetry. The externalization of PS by the activation of a lipid scramblase is usually the effective component of the dissipation of the gradients [8, 9]. The identities of scramblase proteins have only recently begun to emerge. The Transmembrane Family of Proteins

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_14, © Springer Science+Business Media, LLC, part of Springer Nature 2019

181

182

Janine Denise Brunner and Stephan Schenck

16 (TMEM16) was demonstrated to underlie scrambling processes found during blood clotting, membrane repair, spermatogenesis, and bone mineralization, where exposure of PS is mediated in a Ca2+ -dependent manner [10–16]. The recent elucidation of the architecture of TMEM16 proteins provides a basis for future structure-function studies [9, 17], and increasing interest in this protein family and in scrambling processes asks for specific protocols enabling purification, reconstitution into liposomes, and measurement of scrambling activity of TMEM16 scramblases. Today’s diversity of protocols and progress in the expression and purification of membrane proteins has made it straightforward to investigate purified membrane proteins in reconstituted liposomes. Among liposome-based scrambling assays, one method became very common. In this method, which will be outlined here, the liposomes are prepared with small amounts of fluorescent lipids. The fluorescent moiety in the lipid is constituted by 7-nitrobenz-2-1,3 benzoxadiazol (NBD) usually coupled to the lipid headgroup. The scrambling assay described here is based on the bleaching of this fluorescent lipid probe [15, 17, 18]. Key is a membrane-impermeant reducing agent such that only fluorophores in the outer leaflet of the liposomes are bleached. The fluorophore, NBD, is reduced upon addition of sodium dithionite (alternative name is sodium hydrosulfite, Na2S2O4) resulting in a nonfluorescent 7-amino-2,1,3-benzoxadiazol derivative (Fig. 1). To date, within the scrambling assays using NBD and sodium dithionite, a few different protocols exist. These protocols differ with regard to the preparation of liposomes, which might result in liposomes of different size and composition. Further, varying amounts of purified protein or membrane extracts are used for reconstitution and variable concentrations of NBD-lipids and sodium dithionite are applied for scrambling measurements. Here we outline a well-elaborated protocol that, due to the step-by-step description, is easy to reproduce and allows for reliable analysis. This scrambling assay is adapted from different liposome-based scrambling assays [15, 18, 19] that originated from a protocol developed by McIntyre et al. in 1991 [20]. McIntyre and colleagues initially developed the underlying protocol to produce asymmetrically labeled liposomes, though they already recognized its potential application to detect phospholipid translocase activity in biological as well as reconstituted systems. Since then, the protocol was extended by numerous groups to generate a number of variations [18, 19, 21–25]. Malvezzi and colleagues refined the protocol to explore the scramblase function of a fungal TMEM16 protein, which is amenable to purification and reconstitution [15]. We have further optimized the protocol of Malvezzi and colleagues and measured the scramblase activity of a closely related TMEM16 protein family member from the fungus Nectria haematococca [17].

Scrambling Assay Using Purified TMEM16 Protein

183

Fig. 1 Scrambling assay and chemical reaction of the reduction of NBD by sodium dithionite. Left: Schematic representation of the liposomal scrambling assay illustrating the bleaching of fluorescent NBD-conjugated lipids in the outer leaflet of artificial liposomes upon addition of sodium dithionite. Red stars mark the fluorescent NBD-moiety, while empty gray stars represent the reduced and nonfluorescent form of the lipids after dithionite bleaching. The scramblase is indicated by a green box. Right: Reduction of fluorescent 7-nitro2,1,3-benzoxadiazol-4-yl (NBD) by sodium dithionite (Na2S2O4) to produce nonfluorescent 7-amino-2,1,3benzoxadiazol-4-yl

In principle, every method for the reconstitution of membrane proteins into liposomes can be adopted, as long as certain key parameters are considered. We made the best experience with a protocol which was pioneered by Rigaud et al. [26] and which is well described by Geertsma et al. [27]. This method uses preformed liposomes and subsequent stepwise destabilization with Triton X-100. Other long-chain detergents might be suitable as well [28]. Incorporation of Triton X-100 into the membrane is monitored with a spectrophotometer due to a characteristic swelling of the liposomes. The optimal time point to add the protein is between the onset of solubilization and complete solubilization (Fig. 2). The membrane-orientation can be neglected here, since scramblases mediate bidirectional transport of their substrates. Detergents are slowly and stepwise removed afterward by adsorption to styryl-beads. The procedure is very gentle due to the use of long-chain detergents and is therefore recommended for labile proteins. Alternative methods like dialysis-based protocols might also be adopted. These methods do usually involve shortchain detergents like octyl-glucoside or decyl-ß-D-maltopyranoside [29, 30] and thus it has to be tested if the protein can withstand the procedure. The scrambling assay requires homogeneous liposome populations with a low degree of multilamellar vesicles. Therefore, it is crucial to avoid extensive formation of multilamellar vesicles, because the outer leaflet of encapsulated liposomes cannot be

184

Janine Denise Brunner and Stephan Schenck

Fig. 2 Destabilization of preformed liposomes with Triton X-100. Liposomes composed of 3:1 E. coli polar lipids and egg PC were extruded through a 400 nm pore size polycarbonate membrane and diluted to 4 mg/mL. Subsequent destabilization by Triton X-100 in steps of 0.02% (10–14 μL aliquots of 10% solution) was monitored using a spectrophotometer at 570 nm. The onset of solubilization is marked in red. The recommended point in the titration curve at which the protein can be added to the liposomes is indicated by green dots and is slightly below the initial starting point

bleached by dithionite. A liposome extruder and polycarbonate membranes with 400 nm pores are thus essential components during the reconstitution. A final extrusion through polycarbonate membranes is advised for all reconstitution methods as a last step before measurement. The scramblase assay relies on fluorescence and requires the liposomes to contain small amounts of symmetrically incorporated NBD-labeled lipids. To achieve symmetrical incorporation, we add the fluorescent lipids at very early steps of liposome preparation when preparing the lipid mixture. Ideally, trace amounts of NBD-conjugated lipids such as maximally 0.5% of total lipid (w/w) or less are incorporated to prevent selfquenching of the fluorophore which might occur at higher NBD-conjugate concentration [31]. The desired amount of Ca2+ needed for activation of TMEM16 proteins is included during the

Scrambling Assay Using Purified TMEM16 Protein

185

preparation of the preformed liposomes before reconstitution of the liposomes with the protein. The scrambling assay enables the researcher to analyze the kinetics of the bleaching reaction. For analysis of the timedependence of the fluorescence decay recorded in the lipid scrambling assay after application of sodium dithionite, the traces can be fitted to exponential functions to obtain the corresponding time constants of the reaction. The calculated time constants can then be related to draw conclusions on the scrambling rates [15, 17].

2 2.1

Materials Post-purification

1. Mild detergent such as n-dodecyl-ß-D-maltopyranoside. 2. Buffer E: Any elution buffer (e.g., 10 mM HEPES, pH 7.4, 150 mM NaCl, 10% glycerol, eluent, containing ndodecyl-ß-D-maltopyranoside or any other appropriate detergent). The choice of the eluent depends on the affinity tag used for the purification of the scramblase. It would be imidazole, biotin, or D-Desthiobiotin to elute scramblase with an His-tag, an SBP-tag, or a Strep-tag II, respectively. 3. Buffer G: Any gel filtration buffer (e.g., 10 mM HEPES, pH 7.4, 150 mM NaCl, containing n-dodecyl-ß-D-maltopyranoside or any other appropriate detergent, see Note 1). 4. Purified scramblase (see Note 2).

2.2 Preparation of Liposomes

1. Glass pipette and solvent-compatible pipetting aid. 2. Glass vial for storage of lipid stocks. 3. Round bottom flask compatible with a rotary evaporator. 4. E. coli polar lipid extract in chloroform (Avanti Polar Lipids). 5. L-α-phosphatidylcholine from egg yolk in chloroform (SigmaAldrich). 6. NBD-lipids in chloroform (e.g., 1,2-dimyristoyl-sn-glycero-3phosphoethanolamine-N-(7-nitro-2-1,3-benzoxadiazol-4-yl), Avanti Polar Lipids). 7. Rotary evaporator (Bu¨chi Rotavapor R-200 or similar). 8. Diethyl ether. 9. Desiccator. 10. Buffer A: 20 mM HEPES-KOH, pH 7.4, 300 mM KCl. 11. Sonicator. 12. Eppendorf safe-lock microcentrifuge tubes.

186

Janine Denise Brunner and Stephan Schenck

2.3 Preparation of Proteoliposomes

1. Syringe and needle (Braun, Inject F, 1 mL and Braun, Sterican 0.7
30 mm). 2. Extruder and polycarbonate filter with 400-nm pore size (Avestin). 3. 15 and 50 mL centrifuge tubes. 4. Buffer A containing the desired amount of free Ca2+. Add the amount of Ca2+, e.g., as Ca(NO3)2, and the appropriate amount of EGTA to obtain the desired concentration of free Ca2+ (see Note 3). 5. Magnetic stirrer and stirrer bar. 6. 10% (v/v) Triton X-100 (Sigma-Aldrich) in water. 7. Spectrophotometer. 8. Protein at 1 mg/mL. 9. Rotation device. 10. Econo-Column chromatography column (Bio-Rad). 11. Bio-Beads SM-2 polystyrene beads (Bio-Rad) equilibrated in buffer A: In a gravity-flow column such as an Econo-Column chromatography column wash Bio-Beads with five volumes of methanol, then wash with 5
10 volumes of H2O, followed by washing the Bio-Beads three times with 10 volumes of buffer A. Store Bio-Beads in buffer A. 12. Table ultracentrifuge and rotor. 13. Disposable polyprep columns (Bio-Rad). 14. Liquid nitrogen (N2).

2.4 Scrambling Assay

1. Buffer B: 80 mM HEPES-KOH, pH 7.4, 300 mM KCl containing the desired amount of free Ca2+ titrated as Ca(NO3)2 and EGTA (see Notes 3 and 4). 2. 1.5 M sodium dithionite: Prepare a stock solution in water, immediately make 50 μL aliquots and flash-freeze them in liquid N2. Always freshly prepare the sodium dithionite stock solution and use the frozen aliquots within the same day. Five minutes before the addition of sodium dithionite, thaw an aliquot (see Note 5). 3. Proteoliposomes. 4. Disposable plastic standard cuvettes for fluorescence (12.5
12.5
45 mm) with an optical pathlength of 10 mm. 5. Magnetic stirrer bar. 6. Fluoromax-4 spectrofluorometer (Horiba, Jobin Yvon) or similar.

Scrambling Assay Using Purified TMEM16 Protein

3 3.1

187

Methods Post-purification

3.2 Preparation of Liposomes

1. The scramblase (e.g., nhTMEM16) is prepared at a concentration of 1 mg/mL in Buffer E or Buffer G. We recommend using a high-affinity tag such as Streptavidin-binding peptide tag (SBP-tag) or Strep-tag-II for purification to obtain protein of highest purity. In general, protein purity used is >95% (see Note 6). 1. Mix stock solutions of E. coli polar lipids and L-α-phosphatidylcholine in chloroform in a 3:1 (mol/mol) ratio in a glass vial. Add 0.25–0.5% (w/w) of NBD-labeled lipids (see Note 7). The mixture turns yellow (see Notes 8–10). For two reconstitutions (e.g., protein vs. negative control) transfer 20 mg in a round bottom flask. Dry lipids in a rotary evaporator at 35  C for 1 h. Dissolve lipids in 2 mL of diethylether and dry again for another hour to remove traces of organic solvents. Further dry lipids in the round bottom flask using a desiccator overnight and protect the fluorescent lipids with an aluminum foil from light. 2. Suspend lipids in buffer A to a final concentration of 20 mg/ mL. Use a bath-sonicator to prepare a homogenous suspension. Cool down on ice every 30–60 s to prevent overheating of the lipid mixture. Repeat sonication until all lipid clumps have dissolved. After sonication, small unilamellar vesicles (SUVs) are formed. 3. Make aliquots of suitable size, e.g., 1 mL. Flash-freeze in liquid N2 using Eppendorf safe-lock microcentrifuge tubes (see Notes 11 and 12).

3.3 Reconstitution of Scramblase in Liposomes

1. Thaw an aliquot of 20 mg of SUVs in a water bath at room temperature. 2. Add the amount of Ca2+ or any other divalent ion as X(NO3)2, and the appropriated amount of EGTA to obtain the desired concentration of free Ca2+ after thawing the SUVs (see Note 3). Flash-freeze the sample again in liquid N2. Do a total of three freeze-thawing cycles. By this procedure, the SUVs fuse with each other to form large multilamellar vesicles (LMVs). 3. Load the extruder with the LMVs suspension using a syringe and a needle. Extrude the liposomes 11 times through a 400 nm pore size polycarbonate filter to form large unilamellar vesicles (LUVs) (see Note 13). 4. Dilute the LUVs to 4 mg/mL into a 15 mL centrifugation tube using buffer A containing the appropriate concentration of free Ca2+. Add a magnetic stirrer bar of suitable size.

188

Janine Denise Brunner and Stephan Schenck

5. For titration of the LUVs (20 mg of lipid in 5 mL of buffer A containing the desired amount of free Ca2+), add Triton X-100 (10%) in aliquots of 10 μL (steps of 0.02%). To allow for sufficient equilibration, incubate for 1–2 min under decent stirring on a magnetic stirrer between each addition of Triton X-100. The progression of the destabilization procedure can be monitored by measuring the optical density of the liposomes at a wavelength of 540 nm using a spectrophotometer. First an increase in optical density is observed (onset of solubilization), followed by a decrease. Usually, 10–14 aliquots are needed to arrive at an optical density slightly below the initial starting point. The green data points in Fig. 2 correspond to the ideal section to add the protein. Remove the magnetic stirrer bar. 6. Add 50 μg of purified scramblase (or equal amount/ volume of negative control) per 10 mg of lipids to get a 1:200 (w/w) protein-to-lipid ratio (time point 0 min). Place the tube with the liposomes on a rotator at 4–8  C for 15 min. Add 100 mg (net weight) of Bio-Beads SM-2 per 10 mg of lipids to gradually remove the detergent at time points 15 min, 45 min, and 1 h 45 min under gentle rotation at 4–8  C. Add another 200 mg (net weight) of Bio-Beads SM-2 per 10 mg of lipids after 24 h (see Notes 1 and 14). During detergent removal using the Bio-Beads method the length and degree of saturation of the NBD-conjugated lipids needs to be considered. Try to avoid working with NBD-(head-group) labeled lipids with a fatty acid chain length > 16 and use NBD-conjugated lipids with saturated fatty acids to reduce adsorption of the NBD-lipids to the Bio-Beads (see Note 7). 7. Remove the Bio-Beads SM-2 from the proteoliposomes after 40 h using disposable polyprep columns and collect the proteoliposomes in the eluate using suitable ultracentrifugation tubes. For quantitative recovery of the proteoliposomes, wash the Bio-Beads with an additional volume of 1–2 mL of buffer A containing the appropriate concentration of free Ca2+. 8. Pellet the proteoliposomes at roughly 180,000
g (gmax) for 30 min at 16  C. Resuspend the proteoliposomes in buffer A containing the appropriate concentration of free Ca2+ at a concentration of 20 mg/mL. 9. Flash-freeze the proteoliposomes in liquid N2 and store them at 80  C (see Notes 11 and 12). 3.4 Scrambling Assay

1. Do 3 cycles of freeze-thawing of the proteoliposomes in a water bath at room temperature. 2. Extrude proteoliposomes 11 times through a 400 nm pore size polycarbonate filter (see Note 13). Keep at room temperature.

Scrambling Assay Using Purified TMEM16 Protein

189

See Note 15 for an alternative method (compatible with this protocol) to measure scramblase activity using back-extraction of phospholipids by fatty acid-free bovine serum albumin. 3. Set up the spectrofluorometer as follows: Excitation wavelength at 470 nm and emission wavelength at 530 nm. Record the fluorescence at 23  C every 2–3 s using the shutter to minimize bleaching of the sample. Dilute 20 μL of the proteoliposome suspension into a cuvette filled with 1.98 mL of buffer B (see Note 4) containing the appropriate concentrations of EGTA and Ca(NO3)2 to get the desired concentration of free Ca2+ (see Note 3). Include a magnetic stirrer bar for gentle mixing. 4. To visualize the scrambling process, add sodium dithionite after a recording time of 1 min (base line) to a final concentration of 30 mM (see Note 5). This will bleach the fluorophore of the NBD-conjugated lipids in the outer leaflet of the liposomes. Continue the fluorescence measurement for another 10 min until the reaction has completed (stable line) (see Note 16 for setup and explanation of a typical experiment). 5. Plot the data as F/Fmax (y-axis) as a function of time t(s) (x-axis) (see Note 17 for analysis of a typical experiment, Note 18 for technical limitations of the assay and Note 19 for troubleshooting).

4

Notes 1. We recommend to include an appropriate negative control in every experiment. Empty liposomes, which are treated exactly the same way as proteoliposomes, but are reconstituted with detergent-containing buffer (buffer E or buffer G) in the place of protein, or proteoliposomes that contain a protein without scrambling activity, are suitable negative controls. Since it is conceivable that a reconstituted scramblase provides also a path for dithionite, it might be necessary to test for such events. Accordingly, permeating dithionite would also bleach NBD on the internal leaflet. The encapsulation of an NBD-fluorophore attached to a membrane-impermeant molecule, such as dextran or glucose, into the liposomes, could serve as a control [15, 17]. 2. Expression of TMEM16 proteins for the purpose of purification is well documented. Fungal TMEM16 scramblases are routinely expressed in Saccharomyces cerevisiae under the control of the GAL1 promoter using uracil selection [15, 17]. Mammalian TMEM16 proteins were successfully produced using either Sf9 cells [32] or human embryonal

190

Janine Denise Brunner and Stephan Schenck

kidney (HEK) cells by various protocols, including transient expressions [17, 33] and stable cell lines such as the Flp-in T-Rex 293 system [34–36] which allows gentle overexpression of the protein of interest using hygromycin selection and tetracycline induction. Furthermore, Ba/F3 cells stably transfected with a plasmid carrying mTMEM16F under the elongation factor-1α promoter with puromycin selection was used for expression of TMEM16F [37]. For suitable purification protocols refer to [15, 17, 32–38]. TMEM16 proteins were routinely purified using a His-tag, a Strep-tag-II, or an SBP-tag. As a detergent the protocols mostly involved n-dodecyl-ß-D-maltopyranoside, but alternative detergents such as digitonin and lauryl maltose neopentyl glycol (LMNG) were applied as well. 3. For calculation of the appropriate amount of Ca(NO3)2 and EGTA use the Ca-EGTA Calculator v1.3 available under http://maxchelator.stanford.edu/CaEGTA-TS.htm (maxchelator.stanford.edu) or a similar program. A solution of 2 mM EGTA is used for Ca2+-free conditions, while for Ca2+ -containing conditions 2 mM EGTA and the appropriate amount of Ca(NO3)2 are mixed to get the desired concentration of free Ca2+. Note that the Ca2+ concentration is set in the steps of reconstitution, and cannot be changed during the scrambling assay since the orientation of the scramblase in the liposomes will be random (and the inside of the liposomes will not be accessible anymore for a change of the Ca2+ concentration once the liposomes have been formed). 4. For measurement of the scramblase activity in step 4 of Subheading 3.4, a sufficiently high buffer capacity is needed. Since sodium dithionite acts as an acid in aqueous solutions (due to the following reaction: 2S2O42 + H2O ! S2O32 + 2HSO3) we have made the best experience during our scrambling measurements if the sample is sufficiently buffered. For this reason the buffer concentration is raised from 20 mM HEPES (pH 7.4) during the reconstitution steps to 80 mM HEPES (pH 7.4) for the scrambling measurements to provide sufficient buffering for the application of 30 mM sodium dithionite (see also [17]). 5. Sodium dithionite has a short shelf life and slowly deteriorates in aqueous solutions, so it cannot be stored for a long period. 6. For demonstration of the scramblase assay we used a Ca2+activated scramblase of the TMEM16 family from the fungus Nectria haematococca (nhTMEM16, uniprot accession number C7Z7K1) [17]. To obtain highest purity of nhTMEM16 or any other TMEM16 protein we use a high-affinity tag such as streptavidin-binding peptide (SBP) tag or Strep-tag II [17]. nhTMEM16 is used for reconstitution after elution

Scrambling Assay Using Purified TMEM16 Protein

191

Fig. 3 PE and possible positions for NBD conjugation. 1,2-dimyristoyl-sn-glycero-3-phosphoethanolamine, headgroup-labeled 1,2-dimyristoyl-sn-glycero-3-phosphoethanolamine-N-(7-nitro-2-1,3-benzoxadiazol-4-yl) and fatty acid-labeled 1-myristoyl-2-{6-[(7-nitro-2-1,3-benzoxadiazol-4-yl)amino]hexanoyl}-sn-glycero-3phosphoethanolamine from top to bottom

from a size exclusion column (Superdex 200 10/300 GL) or directly after elution from affinity chromatography with similar results (the latter is only possible if a monodisperse peak upon size exclusion can be achieved). For stabilization of the detergent-solubilized TMEM16 protein, all purification steps are routinely performed at 4  C, in the presence of 10% glycerol and using a mild detergent such as n-dodecyl-ß-D-maltopyranoside. nhTMEM16 was used for reconstitution at a concentration of 1 mg/mL, aiming for a protein-to-lipid ratio of 1:200 (w/w). 7. The NBD label can be attached to a fatty acid moiety (usually on shorter lipid tails like hexyl groups) or at the headgroup, where it is most often attached to an amine group due to the ease of the coupling chemistry (Fig. 3). A scramblase is by definition nonspecific and the attached label should not interfere with the scrambling process. This is probably true for the protein class described in this assay and has been evaluated in detail for NBD-PE versus PE with a fungal member of the TMEM16 family [15]. However, the modification of the headgroup could well interfere with the scrambling process since the NBD label is not so small, and the characteristics of a given headgroup might be altered. If this factor is critical for the study, it is recommended to use a tail-labeled lipid. The tail-

192

Janine Denise Brunner and Stephan Schenck

label does reduce the hydrophobicity of the lipid to some extent. Since the NBD-label on the fatty acid can be effectively bleached in liposomes [24, 39], one can assume that the fluorophore locates to the surface of the leaflet into the region of the lipid headgroups. This implies that the labeled fatty acid lies probably more horizontal in the leaflet. In order to preserve the lipid character of such substrates the remaining unlabeled fatty acid chain should be relatively long, e.g., a myristoyl acid moiety in minimum. However, for headgroup-labeled lipids, we observed relatively strong adsorption on the surface of Bio-Beads, if di-oleoyl versions of phospholipids were used, which were not saturated [17]. This can lead to big loss of signal and makes the data more noisy and difficult to interpret. Intermediate lengths of the fatty acids like di-myristoyl phospholipids circumvented this problem, and are still hydrophobic enough to be fully incorporated into the bilayer after detergent removal. Alterations of the physical properties of the lipids by introduction of the fluorophore have thus to be taken into account when planning the experiments. Some lipid classes have an intrinsically high flip-flop rate, for instance ceramides [40, 41]. For such lipids (bearing only a very small headgroup) increased bleaching due to a higher spontaneous flip-flop rate in the absence of a scramblase might be observable. Different types of NBD-conjugated lipids are commercially available (Avanti Polar Lipids or Echelon Biosciences). In addition, many substrates, which are not commercially available, can be synthesized following described protocols [25, 39, 42–44]. Accordingly, scrambling processes can be investigated using NBD conjugated to PS, PE, PC, SM, phosphatidylglycerol, phosphatidylinositol, glycosylceramides, galactosylceramides, and probably many more. 8. Use glassware while working with chloroform and diethyl ether to avoid extracting of polymer from the plastic. Avoid using plastic ware for storage of lipids. 9. Do not exceed a total amount of 0.5% NBD-conjugated lipids, since the NBD fluorophore might self-quench at higher concentrations. 10. Lipid mixtures can be stored at 20  C for at least 6 months. 11. With a hypodermic needle, prepare the safe-lock Eppendorf tubes in such a way that they contain a small hole in the lid. This will prevent the tubes from bursting when thawing them after freezing in liquid N2. 12. Liposomes and proteoliposomes can be stored at 80  C for up to 6 months. 13. Pre-equilibrate the extruder with buffer A containing the appropriate amounts of EGTA and concentrations of Ca

Scrambling Assay Using Purified TMEM16 Protein

193

(NO3)2 before adding the lipid mixture. Use a syringe with needle to fill the extruder. Make sure that the extruder is tight and the filter intact before proceeding to the liposome sample. After the first passages through the filter, the sample will turn from turbid and opaque to transparent. The odd number of passages ensures that further work will be done with the filtered sample. 14. The assay requires only small quantities of protein. In a typical experiment it is sufficient to provide 50 μg of purified scramblase ideally at 1 mg/mL (much lower concentrations dilute the mixture to undesirable levels) to reconstitute 10 mg of preformed liposomes. A protein-to-lipid ratio of 1:200 (w/w) or less has turned out to be ideally suited using the reconstitution protocol described herein. We made the experience that higher protein-to-lipid ratios (like 1:20 (w/w)) are unfavorable if working with TMEM16 scramblases. Since TMEM16 scramblases probably possess a basal activity in the absence of Ca2+ [15, 17], higher protein-to-lipid ratios impede that the different activities (e.g., in the absence and presence of Ca2+) can be resolved properly. This is due to the slow reaction of the bleaching process of the NBD-fluorophore by sodium dithionite, which becomes rate-limiting in this case (see Note 18). 15. Among liposome-based lipid scrambling assays, an alternative method exists which, due to its lower sensitivity, is less often used. However, it is a very useful addition to the method described in this chapter. In the back-extraction assay, fluorescent NBD-conjugated lipids are incorporated into the liposomes in a similar way as for the herein described protocol. In this assay, fatty acid-free bovine serum albumin (BSA) is used to extract the NBD-fluorescent lipids from the outer liposomal leaflet and thereby quenches the fluorescence [15, 18, 23, 45–47]. The maximal fluorescence decay that can be achieved with a fully activated scramblase is only 50% due to quenching of the fluorophore by fatty acid-free BSA of 50%. 16. A typical scrambling experiment is shown in Fig. 4. In the absence of a reconstituted scramblase, the flip-flop of phospholipid headgroups is very slow (t½ of several hours) [48]. Since the NBD-conjugated lipids are distributed symmetrically, the fluorescence will drop to 50% of the initial value upon addition of dithionite because NBD labels residing in the inner leaflet are shielded from dithionite (black trace). In proteoliposomes containing an activated scramblase (as described in this protocol), virtually all NBD-lipids will transit through the external leaflet and are thereby exposed to sodium dithionite. Therefore, an almost complete loss in fluorescence is observed. In these experiments, we diluted 20 μL of liposomes into 1.98 mL

194

Janine Denise Brunner and Stephan Schenck

Fig. 4 Scrambling of NBD-PE by nhTMEM16. Scrambling activity of proteoliposomes containing nhTMEM16 was investigated in the presence and absence of Ca2+ and compared to protein-free liposomes. The fluorescence decay is shown as F/Fmax, where F means the measured fluorescence at t(s) and Fmax denotes the average fluorescence before addition of 30 mM sodium dithionite. An asterisk marks addition of sodium dithionite at t ¼ 60 s. Three to four measurements were performed per experiment and standard deviations are indicated by error bars

of buffer B containing 2 mM EGTA (green trace) and the calculated concentration of EGTA and Ca(NO3)2 to get 300 μM of free Ca2+ (blue trace). Scrambling processes were visualized by the addition of 30 mM sodium dithionite after 60 s for a total time of 500 s. Fluorescence was recorded every 3 s (excitation wavelength at 470 nm, emission wavelength at 530 nm) to minimize bleaching. We have measured the bleaching by the spectrophotometer setup, which was negligible with a loss of 0.03% after a 500 s recording. It is nevertheless recommended to use the shutter between the measurements, if available. Usually, the fluorescence does not drop below 10%, although theoretically all NBD-lipids should be bleached when a flip-flop activity is present. The remaining fluorescence is due to liposomes that are devoid of scramblases and/or due to a certain percentage of multilamellar vesicles [15, 17]. 17. Plot the data as F/Fmax (y-axis) as a function of time t(s) (xaxis). F/Fmax illustrates the fluorescence decay and F is the fluorescence at time t(s) while the initial fluorescence (Fmax) is read out before addition of sodium dithionite. For analysis of the time-dependence of the fluorescence decay, the traces can be fitted to exponential functions to obtain the corresponding time constants of the reaction. To extract the time constants we used the program Grace (http://plasma-gate.weizmann.ac.il/ Grace/), but alternative software for data analysis can be used [15, 49]. The fluorescence decay in protein-free liposomes

Scrambling Assay Using Purified TMEM16 Protein

195

Fig. 5 Scrambling of NBD-PE by nhTMEM16 at increasing Ca2+ concentrations and analysis. Left: Traces are shown for WT nhTMEM16 with 0 μM (green), 0.25 μM (yellow), 2.5 μM (magenta), and 300 μM Ca2+ (blue) and a Ca2+-binding site mutant with 0 μM and 300 μM Ca2+ (both red). Several measurements were performed per experiment and standard deviations are indicated by error bars. Right: The corresponding time constants of the second reaction are indicated and related. Time constants were calculated using the program Grace (http://plasma-gate.weizmann.ac.il/Grace/)

settles down at 50% and is well described by a single exponential function with a time constant of 15 s (see left panel in Fig. 5, black trace). For proteoliposomes containing nhTMEM16, the fluorescence decrease soon reaches a minimal value around 100 s after sodium dithionite application in the presence of 300 μM Ca2+ (blue trace). This reaction is also described by a single exponential function with a time constant of 22 s. However, in scrambling reactions of proteoliposomes containing nhTMEM16 in the absence of Ca2+, we observe a biphasic decay that can be characterized by a sum of two exponential functions with time constants of 25 and 175 s. In the presence of 0.25 and 2.5 μM Ca2+ the second time constants decrease to 75 and 25 s, respectively (Fig. 5, right panel). In proteoliposomes reconstituted with a Ca2+-binding site mutant the second time constants increases to 803 s in the absence and presence of Ca2+. These second time constants can be related to each other to draw conclusions on the scrambling rates. The activity of nhTMEM16 increases by a factor of 2.3 and 7 in the presence of 0.25 and 2.5 μM free Ca2+, respectively, compared to the basal activity in the absence of Ca2+. Moreover, the mutant is 4.5 times less active than the WT scramblase in the absence of Ca2+ and 32 times less active as the WT protein in the presence of 2.5 μM Ca2+. In scrambling reactions with a biphasic decay we can discriminate between two kinetic elements: the first one coming from the reduction of fluorophore by dithionite, and the second one can be related to the flipping of the NBD-lipids from inside to outside. 18. The scrambling assay enables precise conclusion on the kinetics of scrambling processes; however, it is important to note the limitations: only traces with a biphasic decay can be analyzed in

196

Janine Denise Brunner and Stephan Schenck

Table 1 Troubleshooting. Summary of the most frequently occurring problems with possible explanations and proposed solutions Problem at step 5 in Subheading 3.4 Possible reason

Solution

Fluorescence reduction of >50% in the negative control

Chloroform or diethylether was not fully removed from lipids

Fluorescence reduction of >50% in the negative control

Detergent was not completely Add more Bio-Beads during the last addition (step 6 in Subheading removed during detergent 3.3) removal using Bio-Beads SM-2

Fluorescence reduction of >50% in the negative control

Bursting of liposomes due to insufficient buffering in the assay buffer (Buffer B)

Do the scrambling measurements in the presence of e.g., 80 mM HEPES-KOH, pH 7.4 while bleaching with 30 mM sodium dithionite (step 3 in Subheading 3.4)

Fluorescence reduction of >50% in the negative control

Liposomes might be too old

Prepare fresh liposomes (step 1 in Subheading 3.2)

Total fluorescence is low before addition of sodium dithionite, therefore the fluorescence recordings become very shaky

The total amount of NBD-lipids added during lipid mixture preparation was NBD; BODIPY>AV>NBD.

brightness:

16. The diameter of small unilamellar vesicles prepared by rapid ethanol injection is ~30nm [90, 91]. 17. The efficiency of probe sonication for generating small unilamellar vesicles results in a final concentration of ~40 mM POPC because of lipid loss to the pellet during the ultracentrifugation step. 18. A drawback of the signal response is similarity with that generated by fusion of donor and acceptor vesicles. Ruling out fusion is easily accomplished by monitoring for significant

In-Vitro Intermembrane Transfer of Sphingolipids

251

changes in light scattering that accompany vesicle fusion (e.g., [55]). The recent development of fluorescently labeled membrane-spanning lipids also provides a new means to assess the issue [92]. 19. Verification that the FRET signal changes reflects lipid transfer rather than donor and acceptor vesicle fusion or aggregation can be accomplished by monitoring light scattering changes during the assay (e.g., [55]). Light scattering is extremely sensitive to changes in vesicle aggregation or size due to fusion. Electron microscopy can also be used effectively for monitoring changes in lipid vesicle size (e.g., [91, 93, 94]). 20. Use of Tween-20 rather than Triton X-100 is preferred due to photophysical artifacts produced by the aromatic ring of Triton X-100. 21. Good software programs for fitting and analyzing the initial transfer rates are ORIGIN (Origin Lab, Northampton, MA) or Prism GraphPad (La Jolla CA). 22. Expression of the transfer rate in terms of lipid mass rather than relative fluorescence intensity can be accomplished in the following way. The emission signal counts recorded after Tween20 addition provide insights into signal counts representing the total amount of fluorescent SL (400 pmol/assay). Assuming that the fluorescent SL is mass distributed in the small donor vesicles, the accessible pool size available for transfer on the external surface of the donor vesicles is ~60–65% of the total fluorescent SL or ~240–260 pmol. 23. Variations to the competition assay approach include addition of a second set of vesicles containing the “competitor” lipid after the normal FRET assay is underway [58, 95]. 24. BSA (25 μg in 50 μL) can be included in assays (added prior to LTP) as a control when low LTP amounts (0.05–0.1 μg) are being measured to avoid the possibility of LTP adsorption to the reaction tube walls. 25. [14C]-tripalmitin serves as a nontransferable marker in the donor lipid vesicles and is initially present at similar cpm levels as [3H]-glycolipid. Thus, when only low cpm levels of [14C]tripalmitin (typically 1–2%) are observed in the recovered acceptor vesicle eluants, this verifies that the large cpm levels of [3H]-glycolipid (50–60%) reflects transfer rather than donor and acceptor vesicle fusion (e.g., [71]). 26. We have observed high spontaneous transfer of sulfatide by GLTP and C1P by CPTP when using donor lipid vesicles containing 10 mol% negatively charged lipid (e.g., DPPA).

252

Roopa Kenoth et al.

This complication results in high background signal relative to the LTP-mediated signal for lipid transfer. 27. The lower transfer rates obtained with large planar membranes (e.g., liposomes) necessitate use of higher amounts of transfer protein. Other natural membranes sometimes employed as acceptor membranes that also can be easily pelleted by low-speed centrifugation are red blood cell ghosts and mitochondria [52, 96].

Acknowledgments We are grateful to many individuals who contributed to the development of the experimental approaches used routinely in the REB lab for many years and detailed here. They include J.G. Molotkovsky, H.M. Pike, X Zhai, I.A. Boldyrev, Y-G Gao, and P. Mattjus. We also are grateful for support from Dept. of Science and Technology, Science and Engineering Research Board (SERB), Govt. of India to RKK (YSS/2014/000021) and to RK (YSS/2015/000783) as well as support by NIH/NIGMSGM45928, NIH/NCI-CA121493, and NIH/NHLBIHL125353 (to REB) and the Hormel Foundation. We thank VIT for research facilities and infrastructure. References 1. Holthuis JCM, Menon AK (2014) Lipid landscapes and pipelines in membrane homeostasis. Nature 510:48–57 2. Drin G (2014) Topological regulation of lipid balance in cells. Annu Rev Biochem 83:51–77 3. Apodaca G, Brown WJ (2014) Membrane traffic research: challenges for the next decade. Front Cell Dev Biol 2:e52 4. Tatsuta T, Scharwey M, Langer T (2014) Mitochondrial lipid trafficking. Trends Cell Biol 24:44–52 5. Hurlock AK, Roston RL, Wang K et al (2014) Lipid trafficking in plant cells. Traffic 15:915–932 6. Malinina L, Simanshu DK, Zhai X et al (2015) Sphingolipid transfer proteins defined by the GLTP-fold. Q Rev Biophys 48:281–322 7. Yamaji T, Hanada K (2015) Sphingolipid metabolism and interorganellar transport: localization of sphingolipid enzymes and lipid transfer proteins. Traffic 16:101–122 ˇ opicˇ A, Delfosse V et al 8. Moser von Filseck J, C (2015) Phosphatidylserine transport by ORP/Osh proteins is driven by phosphatidylinositol 4-phosphate. Science 349:432–436

9. Hotamisligil GS, Bernlohr DA (2015) Metabolic functions of FABPs—mechanisms and therapeutic implications. Nat Rev Endocrinol 11:592–605 10. Grabon A, Khan D, Bankaitis VA (2015) Phosphatidylinositol transfer proteins and instructive regulation of lipid kinase biology. Biochim Biophys Acta 1851:724–735 11. Wakana Y, Kotake R, Oyama N et al (2015) CARTS biogenesis requires VAP–lipid transfer protein complexes functioning at the endoplasmic reticulum–Golgi interface. Mol Biol Cell 26:4686–4699 12. Mattjus P (2016) Specificity of the mammalian glycolipid transfer proteins. Chem Phys Lipids 194:72–78 13. Gallo A, Vannier C, Galli T (2016) Endoplasmic reticulum–plasma membrane associations: structures and functions. Annu Rev Cell Dev Biol 32:279–301 14. Tong J, Manik MK, Yang H et al (2016) Structural insights into nonvesicular lipid transport by the oxysterol binding protein homologue family. Biochim Biophys Acta 1861:928–939

In-Vitro Intermembrane Transfer of Sphingolipids 15. Kentala H, Weber-Boyvat M, Olkkonen VM (2016) OSBP-related protein family: mediators of lipid transport and signaling at membrane contact sites. Int Rev Cell Mol Biol 321:299–340 16. Huang J, Ghosh R, Bankaitis VA (2016) Sec14-like phosphatidylinositol transfer proteins and the biological landscape of phosphoinositide signaling in plants. Biochim Biophys Acta 1861:1352–1364 17. Salminen TA, Blomqvist K, Edqvist J (2016) Lipid transfer proteins: classification, nomenclature, structure, and function. Planta 244:971–997 18. Wong LH, Levine TP (2016) Lipid transfer proteins do their thing anchored at membrane contact sites . . . but what is their thing? Biochem Soc Trans 44:517–527 19. de Campos MKF, Schaaf G (2017) The regulation of cell polarity by lipid transfer proteins of the SEC14 family. Curr Opin Plant Biol 40:158–168 ˇ opicˇ A, Levine TP (2017) 20. Wong LH, C Advances on the transfer of lipids by lipid transfer proteins. Trends Biochem Sci 42:516–530 21. Malinina L, Patel DJ, Brown RE (2017) How α-helical motifs form functionally diverse lipidbinding compartments. Annu Rev Biochem 86:609–636 22. Huang J, Mousley CJ, Dacquay L et al (2018) A lipid transfer protein signaling axis exerts dual control of cell-cycle and membrane trafficking systems. Dev Cell 44:378–391 23. Mishra SK, Gao Y-G, Deng Y et al (2018) CPTP: A sphingolipid transfer protein that regulates autophagy and inflammasome activation. Autophagy 14(5):862–879 24. Malinina L, Malakhova ML, Teplov A et al (2004) Structural basis for glycosphingolipid transfer specificity. Nature 430:1048–1053 25. Kenoth R, Simanshu DK, Kamlekar RK et al (2010) Structural determination and tryptophan fluorescence of heterokaryon incompatibility C2 protein (HET-C2), a fungal glycolipid transfer protein (GLTP), provide novel insights into glycolipid specificity and membrane interaction by the GLTP-fold. J Biol Chem 285:13066–13078 26. Simanshu DK, Kamlekar RK, Wijesinghe DS et al (2013) Nonvesicular trafficking of a ceramide-1-phosphate that regulates eicosanoids. Nature 500:463–468 27. Simanshu DK, Zhai X, Munch D et al (2014) Arabidopsis accelerated cell death 11, ACD11, is a ceramide-1-phosphate transfer protein and

253

intermediary regulator of phytoceramide levels. Cell Rep 6:388–399 28. Ochoa-Lizarralde B, Gao Y-G, Popov AN et al (2018) How FAPP2 selects simple glycosphingolipids using the GLTP-fold: Structural insights into specificity. J Biol Chem (in press) 29. D’Angelo G, Polishchuk E, Di Tullio G et al (2007) Glycosphingolipid synthesis requires FAPP2 transfer of glucosylceramide. Nature 449:62–67 30. Kamlekar RK, Simanshu DK, Gao YG et al (2013) The glycolipid transfer protein (GLTP) domain of phosphoinositol 4-phosphate adaptor protein-2 (FAPP2): structure drives preference for simple neutral glycosphingolipids. Biochim Biophys Acta 1831:417–427 31. Bergelson LD, Molotkovsky JG, Manevich YM (1985) Lipid specific fluorescent probes in studies of biological membranes. Chem Phys Lipids 37:165–195 32. Molotkovsky JG, Mikhalyov II, Imbs AB et al (1991) Synthesis and characterization of new fluorescent glycolipid probes: molecular organization of glycosphingolipids in mixed composition lipid bilayer. Chem Phys Lipids 58:199–212 33. Boldyrev IA, Zhai X, Momsen MM et al (2007) New BODIPY lipid probes for fluorescence studies of membranes. J Lipid Res 48:1518–1532 34. Boldyrev IA, Brown RE, Molotkovsky JG (2013) An expedient synthesis of fluorescent labeled ceramide-1-phosphate analogues. Russ J Bioorgan Chem 39:539–542 35. Radin NS, Evangelatos GP (1981) The use of galactose oxidase in lipid labeling. J Lipid Res 22:536–541 36. Brown RE, Sugar IP, Thompson TE (1985) Spontaneous transfer of gangliotetraosylceramide between phospholipid vesicles. Biochemistry 24:4082–4091 37. Brown RE, Thompson TE (1987) Spontaneous transfer of ganglioside GM1 between phospholipid vesicles. Biochemistry 26:5454–5460 38. Sonnino S, Nicolini M, Chigorno V (1996) Preparation of radiolabeled gangliosides. Glycobiology 6:479–487 39. Sonnino S, Chigorno V, Tettamanti G (2000) Preparation of radioactive gangliosides, 3H or 14C isotopically labeled at oligosaccharide or ceramide moieties. Methods Enzymol 311:639–656

254

Roopa Kenoth et al.

40. Schwarzmann G, Sandhoff K (1987) Lysogangliosides: synthesis and use in preparing labeled gangliosides. Methods Enzymol 138:319–341 41. Ito M, Mitsutake S, Tani M et al (2000) Enzymatic synthesis of [14C]ceramide, [14C]glycosphingolipids, and ω-aminoceramide. Methods Enzymol 311:682–689 42. Li X-M, Malakhova ML, Lin X et al (2004) Human glycolipid transfer protein: Probing conformation using fluorescence spectroscopy. Biochemistry 43:10285–10294 43. Malakhova ML, Malinina L, Pike HM et al (2005) Point mutational analysis of the liganding site in human glycolipid transfer protein: functionality of the complex. J Biol Chem 280:26312–26320 44. Szoka F, Demetrios Papahadjopoulos D (1980) Comparative properties and methods of preparation of lipid vesicles (liposomes). Annu Rev Biophys Bioeng 9:467–508 45. Patil YP, Jadhav S (2013) Novel methods for liposome preparation. Chem Phys Lipids 177:8–18 46. Akbarzadeh A, Rezaei-Sadabady R, Davaran S et al (2014) Liposome: classification, preparation, and applications. Nanoscale Res Lett 8: e102 47. Nichols JW, Pagano RE (1981) Kinetics of soluble lipid monomer diffusion between vesicles. Biochemistry 20:2783–2789 48. Nichols JW (1988) Kinetics of fluorescentlabeled phosphatidylcholine transfer between nonspecific lipid transfer protein and phospholipid vesicles. Biochemistry 27:1889–1896 49. Roseman M, Thompson TE (1980) Mechanism of the spontaneous transfer of phospholipids between bilayers. Biochemistry 19:439–444 50. Correa-Freire MC, Barenholz Y, Thompson TE (1982) Glucocerebroside transfer between phosphatidylcholine bilayers. Biochemistry 21:1244–1248 51. Abe A, Sakagami T, Yamada K et al (1985) A fluorimetric determination of the activity of glycolipid transfer protein and some properties of the protein purified from pig brain. Biochim Biophys Acta 778:239–244 52. Wong M, Brown RE, Barenholz Y et al (1984) Glycolipid transfer protein from bovine brain. Biochemistry 23:6498–6505 53. Bai J, Pagano RE (1997) Measurement of spontaneous transfer and transbilayer movement of BODIPY-labeled lipids in lipid vesicles. Biochemistry 36:8840–8848 54. Elvington SM, Nichols JW (2007) Spontaneous, intervesicular transfer rates of fluorescent,

acyl chain-labeled phosphatidylcholine analogs. Biochim Biophys Acta 1768:502–508 55. Mattjus P, Molotkovsky JG, Smaby JM et al (1999) A fluorescence resonance energy transfer approach for monitoring protein-mediated glycolipid transfer between vesicle membranes. Anal Biochem 268:297–304 56. Mattjus P, Pike HM, Molotkovsky JG et al (2000) Charged membrane surfaces impede the protein mediated transfer of glycosphingolipids between phospholipid bilayers. Biochemistry 39:1067–1075 57. Mattjus P, Kline A, Pike HM et al (2002) Probing for preferential interactions among sphingolipids in bilayer vesicles using the glycolipid transfer protein. Biochemistry 41:266–273 58. West G, Viitanen L, Alm C et al (2008) Identification of a glycosphingolipid transfer protein GLTP1 in Arabidopsis thaliana. FEBS J 275:3421–3437 59. Nylund M, Mattjus P (2005) Protein mediated glycolipid transfer is inhibited FROM sphingomyelin membranes but enhanced TO sphingomyelin containing raft like membranes. Biochim Biophys Acta 1669:87–94 60. Nylund M, Kjellberg MA, Molotkovsky JG et al (2006) Molecular features of phospholipids that affect glycolipid transfer proteinmediated galactosylceramide transfer between vesicles. Biochim Biophys Acta 1758:807–812 61. Nichols JW, Pagano RE (1983) Resonance energy transfer assay of protein-mediated lipid transfer between vesicles. J Biol Chem 258:5368–5371 62. Schwarzmann G, Wendeler M, Sandhoff K (2005) Synthesis of novel NBD-GM1 and NBD-GM2 for the transfer activity of GM2-activator protein by a FRET-based assay system. Glycobiology 15:1302–1311 63. Epand RF, Schlattner U, Wallimann T et al (2007) Novel lipid transfer property of two mitochondrial proteins that bridge the inner and outer membranes. Biophys J 92:126–137 64. Brown RE (1990) Spontaneous transfer of lipids between membranes. Subcell Biochem 16:333–363 65. Jones JD, Almeida PFF, Thompson TE (1990) Spontaneous interbilayer transfer of hexosylceramides between phospholipid bilayers. Biochemistry 29:3892–3897 66. Brown RE (1992) Spontaneous lipid transfer between organized lipid assemblies. Biochim Biophys Acta 1113:375–389 67. Brown RE, Mattjus P (2007) Glycolipid transfer proteins. Biochim Biophys Acta 1771:746–760

In-Vitro Intermembrane Transfer of Sphingolipids 68. Rao CS, Lin X, Pike HM et al (2004) Glycolipid transfer protein mediated transfer of glycosphingolipids between membranes: a model for action based on kinetics and thermodynamic analyses. Biochemistry 43:13805–13815 69. Samygina VR, Popov AN, Cabo-Bilbao A et al (2011) Structure 19:1644–1654 70. Brown RE, Stephenson FA, Markello T et al (1985) Properties of a specific glycolipid transfer protein from bovine brain. Chem Phys Lipids 38:79–93 71. Brown RE, Jarvis KL, Hyland KJ (1990) Purification and characterization of glycolipid transfer protein from bovine brain. Biochim Biophys Acta 1044:77–83 72. Zhai X, Gao Y-G, Mishra SK et al (2017) Phosphatidylserine stimulates ceramide 1-phosphate (C1P) intermembrane transfer by C1P transfer proteins. J Biol Chem 292:2531–2541 73. Brown RE, Hyland KJ (1992) Spontaneous transfer of ganglioside GM1 from its micelles to lipid vesicles of differing size. Biochemistry 31:10602–10609 74. Lev S (2010) Non-vesicular lipid transport by lipid-transfer proteins and beyond. Nat Rev Mol Cell Biol 11:739–750 75. Somerharju P (2015) Is spontaneous translocation of polar lipids between cellular organelles negligible? Lipid Insights 8(S1):87–93 76. Richens JL, Tyler AII, Barriga HMG et al (2017) Spontaneous charged lipid transfer between lipid vesicles. Sci Report 7:e12606 77. Shenkarev ZO, Melnikova DN, Finkina EI et al (2017) Ligand binding properties of the lentil lipid transfer protein: Molecular insight into the possible mechanism of lipid uptake. Biochemistry 56:1785–1796 78. Sumi M, Makino A, Inaba T et al (2017) Photoswitchable phospholipid FRET acceptor: detergent free intermembrane transfer assay of fluorescent lipid analogs. Sci Report 7:e2900 79. Ho¨ltt€a-Vuori M, Uronen R-L, Repakova J et al (2008) BODIPY-Cholesterol: A new tool to visualize sterol trafficking in living cells and organisms. Traffic 9:1839–1849 80. Locatelli-Hoops S, Remmel N, Klingenstein R et al (2006) Saposin a mobilizes lipids from low cholesterol and high bis(monoacylglycerol) phosphate-containing membranes: Patient variant saposin A lacks lipid extraction capacity. J Biol Chem 281:32451–32460 81. Kernstock RM, Girotti AW (2007) Lipid transfer protein binding of unmodified natural lipids as assessed by surface plasmon resonance methodology. Anal Biochem 365:111–121

255

82. Sugiki T, Takahashi H, Nagasu M et al (2010) Real-time assay method of lipid extraction activity. Anal Biochem 399:162–167 83. Ohvo-Rekil€a H, Mattjus P (2011) Monitoring glycolipid transfer protein activity and membrane interaction with the surface plasmon resonance technique. Biochim Biophys Acta 1808:47–54 84. Davison JM, Bankaitis VA, Ghosh R (2012) Devising powerful genetics, biochemical and structural tools in the functional analysis of phosphatidylinositol transfer proteins (PITPs) across diverse species. Methods Cell Biol 108:249–302 85. Hanada K, Sugiki T (2017) In vitro assay to extract specific lipid types from phospholipid membranes using lipid-transfer proteins: a lesson from the ceramide transport protein CERT. In: Wood P (ed) Lipidomics, Neuromethods, vol 125. Humana Press, New York, NY 86. Zhai X, Malakhova ML, Pike HP et al (2009) Glycolipid acquisition by human glycolipid transfer Protein dramatically alters intrinsic tryptophan fluorescence: Insights into glycolipid binding affinity. J Biol Chem 284:13620–13628 87. Kamlekar RK, Gao Y-G, Kenoth R et al (2010) Human GLTP: three distinct functions for the three tryptophans in a novel peripheral amphitropic fold. Biophys J 99:2626–2635 88. Kenoth R, Zou X, Simanshu DK et al (2018) Functional evaluation of tryptophans in glycolipid binding and membrane interaction by HET-C2, a fungal glycolipid transfer protein. BBA-Biomembranes 1860:1069–1076 89. Zhai X, Momsen WE, Malakhov DA et al (2013) GLTP-fold interaction with planar phosphatidylcholine surfaces is synergistically stimulated by phosphatidic acid and phosphatidylethanolamine. J Lipid Res 54:1103–1113 90. Kremer JM, Esker MW, Pathmamanoharan C, Wiersema PH (1977) Vesicles of variable diameter prepared by a modified injection method. Biochemistry 16:3932–3935 91. Nordlund JR, Schmidt CF, Dicken SN et al (1981) Transbilayer distribution of phosphatidylethanolamine in large and small unilamellar vesicles. Biochemistry 20:3237–3241 92. Schwarzmann G, Breiden B, Sandhoff K (2015) Membrane-spanning lipids for an uncompromised monitoring of membrane fusion and intermembrane lipid transfer. J Lipid Res 56:1861–1879 93. Coldren B, van Zanten R, Mackel MJ et al (2003) From vesicle size distributions to bilayer elasticity via cryo-transmission and

256

Roopa Kenoth et al.

freeze-fracture electron microscopy. Langmuir 19:5632–5639 94. Meister A, Blume A (2017) (Cryo)transmission electron microscopy of phospholipid model membranes interacting with amphiphilic and polyphilic molecules. Polymers 9:521

95. Edqvist J, Ro¨nnberg E, Rosenquist S et al (2004) Plants express a lipid transfer protein with high similarity to mammalian sterol carrier protein-2. J Biol Chem 279:53544–53553 96. Wetterau JR, Zilversmit DB (1984) Quantitation of lipid transfer activity. Methods Biochem Anal 30:199–226

Chapter 18 Purification and Characterization of Human Niemann–Pick C1 Protein Xin Gong and Hongwu Qian Abstract Niemann–Pick C1 (NPC1) is a membrane protein required for the transport of low-density lipoprotein (LDL)-derived cholesterol from endosomes and lysosomes to the other organelles. Here, we describe the recombinant protein expression, purification, and characterization of the human NPC1. The protein is transiently expressed in human embryonic kidney (HEK) cells. Our purification protocol describes the steps to obtain a pure and homogeneous NPC1 protein. Niemann–Pick C2 (NPC2) is a small soluble protein, which mediates cholesterol transport in tandem with NPC1. Finally, we also describe two biochemical approaches to characterize NPC1 function in vitro—a cholesterol transfer assay from purified NPC2 to NPC1 and a binding assay between NPC1 and NPC2. Key words Niemann–Pick disease type C, NPC1, NPC2, Cholesterol transport, Membrane protein, Transient expression, Protein purification, Cholesterol transfer

1

Introduction Niemann–Pick type C (NPC) is a lysosomal storage disease characterized by the accumulation of cholesterol, sphingomyelin, and other lipids in endosomes and lysosomes [1]. Approximately 95% and 5% of NPC-related disease mutations are found in the NPC1 and NPC2 genes, respectively [2, 3]. The integral membrane protein NPC1 and the soluble protein NPC2 collectively execute cholesterol transport from the endosome/lysosomes to other cellular membranes [4, 5]. The human NPC1 consists of 1278 amino acids that are organized into 13 transmembrane helices (TM) and 3 distinct lumenal domains: A, C, and I [6]. The N-terminal domain A is also designated NTD. The prevailing model suggests that cholesterol derived from endocytic LDL is extracted by NPC2, which may be recruited to NPC1 by domain C and deliver the bound cholesterol to NTD through a “hydrophobic handoff” mechanism [7–9].

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_18, © Springer Science+Business Media, LLC, part of Springer Nature 2019

257

258

Xin Gong and Hongwu Qian

Considerable progress toward understanding the structural insights of NPC1 function has been made since 2009. In particular, the cryo-EM structure of full-length human NPC1 and the crystal structure of human NPC1* (lacking NTD and TM1) are reported by us and another group in 2016 [10, 11]. Notwithstanding these exciting advances, considerable work still needs to be done to improve our understanding of the mechanism of NPC1-mediated cholesterol transport including not only structural biology but also various functional studies. Therefore, the development of an easy and efficient recombinant expression and purification protocol for this protein is important. Here, we provide a simple and fast protocol for the transient expression and purification of human NPC1 in HEK cells. The transient expression allows for rapid and reliable scale-up for milligram amounts of protein production, which makes it an attractive choice for recombinant protein overexpression. More importantly, we also describe two biochemical assays to investigate NPC1 function in vitro: (1) the examination of cholesterol transfer from purified NPC2 to NPC1 variants in a cholesterol transfer assay system and (2) the measurement of the binding affinity between NPC1 variants and NPC2 using isothermal titration calorimetry (ITC). Notably, these two biochemical assays are performed with purified full-length NPC1, whereas previously reported assays are conducted either with isolated domains or in vivo [5, 7, 8]. These assays can serve as powerful tools to establish structure–function relationships and elucidate the pathogenic mechanism of disease mutations.

2 2.1

Materials Reagents

1. E. coli strain: DH5α. 2. E. coli culture media: LB Broth, Miller. 3. LB agar plates containing 1.5% agar and 100 μg/mL ampicillin. 4. Phusion DNA Polymerase, NotI/XhoI restriction enzymes, and T4 DNA Ligase. 5. Plasmid Miniprep Kit, Gel Purification Kit, and DNA Purification Kit. 6. EndoFree Plasmid Maxi Kit. 7. Modified vectors from pCAG. 8. Human cell line: HEK293F (Invitrogen). 9. 100 Penicillin-Streptomycin solution. 10. HEK293F medium: SMM 293T-I (Sino Biological Inc). 11. Transfection reagent PEI25K: Polyethylenimine, Linear, MW 25K (Polysciences).

NPC1 Purification and Characterization

259

12. Ni-NTA Agarose affinity resin. 13. Anti-FLAG M2 affinity resin. 14. FLAG peptide (20 mg/mL stock solution dissolved in water). 15. Detergents: Digitonin (Sigma), Lauryl Maltose Neopentyl Glycol (LMNG, Anatrace), and Octaethylene Glycol Monododecyl Ether (C12E8, Anatrace). 16. Radioactive [3H]-cholesterol: 20 μM in ethanol, 49.0 Ci/ mmol (Perkin Elmer). 17. Scintillation cocktail (PerkinElmer). 2.2

Equipment

1. Glassware for cell growth: 250 mL conical baffled flasks and 2800 mL Fernbach flasks. 2. Shaker incubators with temperature, humidity, and CO2 control. 3. Centrifugation equipment. 4. Ultrasonic homogenization sonicator. 5. Polypropylene gravity-flow chromatography columns. ¨ KTA FPLC system. 6. Chromatography equipment: A 7. Size-exclusion chromatography column: Superose 6 10/300. 8. Centrifugal concentrators with a 100 kDa or 10 kDa cutoff. 9. Sartocon Slice Filtration System (Sartorius). 10. MicroBeta Microplate Counters (Perkin Elmer). 11. MicroCal™ ITC200 (GE). 12. ORIGIN 7 (Origin Lab).

2.3

Buffers

1. Cell lysis buffer: 25 mM Tris–HCl, pH 8, 150 mM NaCl, and a protease inhibitors cocktail (1 μg/mL aprotinin, 1 μg/mL pepstatin, and 1 μg/mL leupeptin). 2. Digitonin at 5% (w/v) in water (see Note 1). 3. Wash buffer: Cell lysis buffer plus 30 mM imidazole and 0.06% (w/v) digitonin. 4. Elution buffer: Wash buffer plus 250 mM imidazole. 5. Filtration buffer: 25 mM MES-NaOH, pH 6, 150 mM NaCl. 6. Transfer buffer A: 25 mM MES-NaOH, pH 5.5, 150 mM NaCl, 0.06% (w/v) digitonin. 7. Transfer buffer 150 mM NaCl.

B:

25

mM

MES-NaOH,

pH

5.5,

8. Transfer buffer C: 250 mM HEPES-NaOH, pH 7.5, 150 mM NaCl, 0.06% (w/v) digitonin. 9. Transfer buffer D: 25 mM HEPES-NaOH, pH 7.5, 150 mM NaCl, 0.06% (w/v) digitonin.

260

Xin Gong and Hongwu Qian

10. ITC buffer: 25 mM MES-NaOH, pH 6, 150 mM NaCl, 0.02% (w/v) C12E8, 0.1 mM TCEP (Tris(2-carboxyethyl) phosphine).

3

Methods The general steps required for large-scale transient expression and purification in mammalian cells comprise (1) cloning of the target gene into an expression vector, (2) generating milligram transfection quality plasmids, (3) transient expression in mammalian cells, and (4) recombinant protein purification. Transient mammalian expression systems can be improved by fine-tuning multiple parameters such as the expression vector, transfection reagent, cell line, culture medium, and transduction conditions. In this chapter, we describe our protocol using the pCAG expression vector, PEI 25K transfection reagent, and HEK293F cells (see Fig. 1). The whole procedure, from target cloning to protein purification, can be completed in 7 days. This timescale is significantly faster than the widely used baculovirus-derived BacMam (Baculovirus-mediated gene transduction of mammalian cells) system, which usually takes 3–4 weeks.

3.1 Generating the Recombinant pCAG Plasmid

The synthesized and codon-optimized cDNA of full-length human NPC1 (1–1278 aa, Uniprot: O15118) is subcloned into the NotI and XhoI sites of modified pCAG vector with a C-terminal His10tag. The cDNA of full-length human NPC2 (1–151 aa, Uniprot: P61916) is subcloned into the NotI and XhoI sites of modified pCAG vectors with a C-terminal FLAG-tag or His10-tag.

Fig. 1 Flow chart of transient mammalian expression system used in this chapter. The whole procedure, from target cloning to protein purification, can be completed in 1 week

NPC1 Purification and Characterization

261

1. The cDNAs are amplified using a standard PCR protocol. The PCR reaction mixture, 50 μL in total, is composed of 1 μL (200 ng) of DNA template, 50 pmol of each primer (Forward: AAATATGCGGCCGCATGACTGCTCGTGGACTG; Reverse: GTACGCCTCGAGGAAATTCAGCAGACGCTCG), 0.5 μL of Phusion enzyme, 1 μL of dNTP stock solution (10 mM), and 10 μL of 5 Phusion buffer. 2. The PCR products are purified with a Gel Purification Kit. 3. The purified PCR products and vectors are digested by NotI and XhoI for 3 h at 37  C. The restricted digestion reaction (50 μL) is composed of 43 μL of PCR products or 3 μg of vectors, 1 μL of NotI solution, 1 μL of XhoI solution, and 5 μL of 10 Cutsmart buffer. 4. The digested PCR products are purified with a DNA Purification Kit. The digested vectors are purified with a Gel Purification Kit. 5. The purified PCR products and vectors are ligated using T4 ligase for 30 min at room temperature. The ligation reaction (10 μL) is composed of 7 μL of digested DNA insert (~100 ng), 1 μL of digested vectors (~20 ng), 1 μL of T4 ligase, and 1 μL of 10 T4 ligase buffer. 6. The ligation mixture is transformed into DH5α by heat shock and selected for ampicillin-resistance on LB agar plates (amp+). 7. Colonies are analyzed by PCR screening with one primer from the insert and another primer from the vector. The positive colony is purified with a Plasmid Miniprep Kit and further checked with sequencing. 3.2 Generating Milligram Transfection Quality Plasmid DNA

1. The constructed pCAG plasmids containing NPC1 or NPC2 are transformed into DH5α competent cells by heat shock and inoculated into 200 mL of LB medium plus ampicillin. 2. The culture is incubated at 37  C, with shaking at 220 rpm for 12–16 h. 3. The bacterial cells are pelleted by centrifugation at 3800  g for 8 min at room temperature, and the supernatants are discarded. 4. The plasmid DNA is purified from the cell pellets using a Plasmid Maxiprep Kit that contains endotoxin removal steps. 5. The DNA concentration is determined using a UV-visible spectrophotometer at 260 nm. Approximately 1–2 mg of pCAG plasmids per preparation is obtained from each 200 mL culture.

3.3 Transient Expression in HEK293F Cells

1. HEK293F cells are cultured in SMM 293T-I medium with 1 penicillin-streptomycin solution at 37  C under 5% CO2 and 75% humidity in a 130-rpm shaker.

262

Xin Gong and Hongwu Qian

2. When the cell density reaches 2  106 cells per mL, the cells are ready to be transiently transfected with the expression plasmids. 3. Approximately 1 mg of endo-free plasmids are premixed with 3 mg of PEI 25 K (see Note 2) in 50 mL of fresh SMM 293T-I medium for 15–30 min before transfection. 4. For transfection, the 50 mL mixture is added to 1 L of HEK293F cell culture and incubated for 15–30 min. 5. Transfected cells are then cultured for 48 h as in the step 1. 3.4 Purification of the NPC1 Protein

1. After 48 h, the transfected cells are pelleted by centrifugation at 3800  g for 8 min at room temperature, and the supernatants are discarded. 2. One-liter cells are resuspended in 40 mL of cell lysis buffer. The collected cells can either be used immediately or frozen in liquid nitrogen and then stored at 80  C. 3. Cells on ice are disrupted using an ultrasonic homogenization sonicator for 1 min (per liter) with 65% amplitude (see Note 3). 4. The membrane is then solubilized at 4  C for more than 2 h with 1% (final concentration) (w/v) digitonin. 5. Centrifuge at 25,000  g for 1 h, and then collect the detergent-soluble supernatant. 6. The supernatant is incubated with nickel affinity resin (Ni-NTA) (all the affinity purifications are performed with polypropylene gravity-flow chromatography columns, ~2 mL resin per column) at 4  C for 30 min. 7. The resin is then rinsed 4 times with 10 mL of wash buffer. 8. The protein is eluted from the affinity resin with 4  2 mL of elution buffer. 9. The eluent is concentrated to 1 mL using a centrifugal concentrator with a 100 kDa cutoff. 10. The NPC1 protein is further purified by Superose 6 size-exclusion chromatography (0.5 mL/min, monitor the elution by UV absorption at 280 nm) in transfer buffer A (see Fig. 2). 11. The peak fractions are then pooled and concentrated for the cholesterol transfer assay (see Subheading 3.6). The protein can either be used immediately or frozen in liquid nitrogen and then stored at 80  C. The NPC1 protein used for the ITC binding assay is purified similarly, except that digitonin is replaced by 1% LMNG (w/v) and 0.02% C12E8 (w/v) in the membrane solubilization and purification procedure, respectively. The final size-exclusion chromatography is performed in the ITC buffer.

NPC1 Purification and Characterization

263

Fig. 2 The last step in purification of NPC1 through size-exclusion chromatography. The indicated fractions are applied to SDS-PAGE and visualized by Coomassie blue staining 3.5 Purification of the NPC2 Protein

The human NPC2 proteins used for the cholesterol transfer assay (with a C-terminal FLAG-tag) and ITC binding assay (with a C-terminal His10-tag) are purified from the medium of transiently transfected HEK293F cells. The following is the procedure for the purification of FLAG-tagged NPC2. 1. Transfect HEK293F cells with the pCAG vectors coding for FLAG-tagged NPC2 following steps similar to those described in Subheading 3.3. 2. After 48 h, the transfected cells are pelleted by centrifugation at 3800  g for 8 min at room temperature, and the supernatants are collected. The supernatants containing the secreted NPC2 are concentrated using a Sartocon Slice Filtration System in the filtration buffer. 3. After centrifugation at 25,000  g for 10 min, the clear supernatants are collected. 4. The supernatant is incubated with Anti-FLAG M2 resin (~2 mL resin per polypropylene gravity-flow chromatography column) at 4  C for 30 min. 5. The resin is then washed by 4  10 mL of buffer. 6. The protein is eluted from the affinity resin with 4  2 mL of filtration buffer containing 200 μg/mL FLAG peptide. 7. The eluent is concentrated to 1 mL using centrifugal concentrator with a 10 kDa cutoff. 8. The NPC2 protein is then further purified by Superose 6 sizeexclusion chromatography (0.5 mL/min) in transfer buffer B. 9. The peak fractions are pooled and concentrated for the cholesterol transfer assay. The protein can either be used immediately or frozen in liquid nitrogen and then stored at 80  C. The His-tagged NPC2 protein used for the ITC binding assay is purified by Ni-NTA. The procedure is similar to the FLAG-

264

Xin Gong and Hongwu Qian

Fig. 3 The flowchart and schematic diagram of the [3H]-cholesterol transfer assay using purified full-length NPC1 and NPC2. Please refer to Subheading 3.6 for details

tagged NPC2, except that the wash and elution buffers contained 30 mM imidazole and 250 mM imidazole, respectively. The final size-exclusion chromatography is performed in the ITC buffer. 3.6 [3H]-Cholesterol Transfer Assay

Modifying the established protocol for NPC1-NTD and NPC2 [5], we set up an in vitro assay to examine cholesterol transfer from NPC2 to full-length NPC1 (see Fig. 3). The NPC1 and NPC2 are fused with His10 and FLAG tags, respectively, and purified to homogeneity. 1. Approximately 10 μL of [3H]-cholesterol in ethanol is added to 300 μL of transfer buffer B containing approximately 80 μg of FLAG-tagged NPC2 (all the protein concentrations are determined by UV absorption at 280 nm) and then incubated for approximately 3 h at room temperature (see Note 4). 2. After incubation, the mixture is diluted with 700 μL of transfer buffer B and centrifuged at 15,000  g for 20 min to remove the insoluble cholesterol (see Note 5). 3. The suspension is then loaded onto the cold pre-equilibration Anti-FLAG M2 affinity column for more than 5 passes. 4. The Anti-FLAG M2 resin is washed 4 times with 1 mL of cold transfer buffer B. The [3H]-cholesterol loaded NPC2 protein is

NPC1 Purification and Characterization

265

then eluted with 1 mL of cold transfer buffer B plus 200 μg/ mL FLAG peptide (see Note 6). 5. The amount of [3H]-cholesterol in eluent is quantified with scintillation count using a MicroBeta Microplate Counters. Briefly, 25 μL of eluent is diluted to 1 mL with transfer buffer B, and then 200 μL of the dilution is mixed with 300 μL of scintillation cocktail. After incubation overnight in a dark environment, the scintillation count is regarded as Input (see Note 7). 6. His-tagged NPC1 prepared in 125 μL of cold transfer buffer A is mixed with 25 μL of the aforementioned NPC2 eluent. Meanwhile, 125 μL of cold transfer buffer A without any NPC1 is mixed with the same amount of NPC2 eluent as a negative control (see Note 8). 7. After incubation at 4  C for 30 min, each reaction is diluted with 875 μL of cold transfer buffer C to adjust the pH value to approximately 7.5 (see Note 9). 8. The dilution is then loaded onto the Ni-NTA resin immediately. 9. The Ni-NTA resin is washed 3 times with 1 mL of cold transfer buffer D. 10. The [3H]-cholesterol loaded NPC1 protein is eluted with 1 mL of cold transfer buffer D plus 250 mM imidazole. 11. Two hundred microliters of the elution are mixed with 300 μL of scintillation cocktails and incubated at room temperature overnight in a dark environment. The scintillation count for the [3H]-cholesterol loaded NPC1 protein minus the scintillation count for the negative control is regarded as the Output (see Note 10). 12. GraphPad Prism is used for data analysis. The cholesterol transfer activity from NPC2 to NPC1 is calculated as the percentage of Output to Input. The transfer assays are performed at three different NPC1 concentrations: 40 nM, 100 nM, and 250 nM. Each data point is the average of three independent experiments. Standard deviation (SD) is calculated for each point (see Fig. 4). 3.7 ITC Binding Assay

The ITC binding assays are performed with a MicroCal™ ITC200 at 18  C. 1. NPC1 proteins prepared in ITC buffer are quantified by UV280 and diluted to 1 μM with the same buffer. NPC2 proteins prepared in ITC buffer are also quantified by UV280 and diluted to 300 μM with the same buffer. 2. Approximately 200 μL of NPC1 is added into the chamber, and 40 μL of NPC2 is added into the syringe.

266

Xin Gong and Hongwu Qian

Fig. 4 Validation of the transfer assay system with reported conditions and mutations. (a) The purified NPC1 and NPC2 proteins used in the assay. (b) Consistent with previous reported observations, the transfer activity of cholesterol transfer from NPC2 to NPC1 at pH 7.5 (green) is only half of that at pH 5.5 (black) in our assay system. Two well-characterized NTD mutations, P202A/F203A (blue) and L175A/L176A (purple), which are deficient in cholesterol binding and transfer, respectively, showed marked reduction of cholesterol transfer at pH 5.5. For each data point, the reading of the negative control (when NPC1 is not included throughout the experiment) is subtracted from the Output before calculating the ratio of Output and Input. Each data point is the average of three independent experiments. Error bar represents SD

3. For each injection, 2 μL of NPC2 in the syringe is titrated into NPC1 in the chamber. 4. The data are recorded and fitted using the software ORIGIN 7.0. In our ITC system, the NPC1 binds to NPC2 with a Kd of ~16 μM at pH 6.

4

Notes 1. The digitonin is solubilized in hot water and then stored at 20  C. 2. The PEI 25 K is solubilized in hot water and then stored at 20  C as a 1 mg/mL stock. 3. The HEK293F cells can also be easily disrupted by a Dounce homogenizer. 4. To get the [3H]-cholesterol loaded NPC2, [3H]-cholesterol and NPC2 should be incubated at room temperature, which can increase the association rate between NPC2 and [3H]cholesterol. 5. Before purification of [3H]-cholesterol loaded NPC2, the insoluble [3H]-cholesterol in the mixture should be eliminated by centrifugation.

NPC1 Purification and Characterization

267

6. The purification of [3H]-cholesterol loaded NPC2 and the latter cholesterol transfer assay should be performed at a low temperature to reduce the dissociation rates between cholesterol and proteins. 7. The [3H]-cholesterol loaded NPC2 should be freshly prepared before each assay. The scintillation counts show that 25 μL of eluent contains approximately 0.7 pmol of [3H]-cholesterol. 8. The transfer assay at pH 7.5 is carried out similarly except that the pH of transfer buffer A is adjusted to 7.5 with 100 mM HEPES, pH 7.5. 9. Before the purification of NPC1 with Ni-NTA resin, the pH of the mixture should be adjusted to basic levels, which is required for the interaction between the His-tag and Ni-NTA resin. 10. Before scintillation count, the eluent and liquid scintillation cocktails should be incubated overnight to get a stable readout. References 1. Vanier MT (2015) Complex lipid trafficking in Niemann-Pick disease type C. J Inherit Metab Dis 38:187–199 2. Carstea ED, Morris JA, Coleman KG et al (1997) Niemann-Pick C1 disease gene: homology to mediators of cholesterol homeostasis. Science 277:228–231 3. Davies JP, Chen FW, Ioannou YA (2000) Transmembrane molecular pump activity of Niemann-Pick C1 protein. Science 290:2295–2298 4. Sleat DE, Wiseman JA, El-Banna M et al (2004) Genetic evidence for nonredundant functional cooperativity between NPC1 and NPC2 in lipid transport. Proc Natl Acad Sci U S A 101:5886–5891 5. Infante RE, Wang ML, Radhakrishnan A et al (2008) NPC2 facilitates bidirectional transfer of cholesterol between NPC1 and lipid bilayers, a step in cholesterol egress from lysosomes. Proc Natl Acad Sci U S A 105:15287–15292 6. Davies JP, Ioannou YA (2000) Topological analysis of Niemann-Pick C1 protein reveals that the membrane orientation of the putative sterol-sensing domain is identical to those of

3-hydroxy-3-methylglutaryl-CoA reductase and sterol regulatory element binding protein cleavage-activating protein. J Biol Chem 275:24367–24374 7. Kwon HJ, Abi-Mosleh L, Wang ML et al (2009) Structure of N-terminal domain of NPC1 reveals distinct subdomains for binding and transfer of cholesterol. Cell 137:1213–1224 8. Wang ML, Motamed M, Infante RE et al (2010) Identification of surface residues on Niemann-Pick C2 essential for hydrophobic handoff of cholesterol to NPC1 in lysosomes. Cell Metab 12:166–173 9. Deffieu MS, Pfeffer SR (2011) Niemann-Pick type C 1 function requires lumenal domain residues that mediate cholesterol-dependent NPC2 binding. Proc Natl Acad Sci U S A 108:18932–18936 10. Gong X, Qian HW, Zhou XH et al (2016) Structural insights into the Niemann-Pick C1 (NPC1)-mediated cholesterol transfer and ebola infection. Cell 165:1467–1478 11. Li XC, Wang JW, Coutavas E et al (2016) Structure of human Niemann-Pick C1 protein. Proc Natl Acad Sci U S A 113:8212–8217

Chapter 19 In Vitro Strategy to Measure Sterol/Phosphatidylinositol-4Phosphate Exchange Between Membranes Nicolas-Fre´de´ric Lipp and Guillaume Drin Abstract Recent findings unveiled that Oxysterol-binding protein-related proteins (ORP)/Oxysterol-binding homology (Osh) proteins, which constitute a major family of lipid transfer proteins (LTPs), conserved among eukaryotes, are not all mere sterol transporters or sensors. Indeed, some of them are able to exchange sterol for phosphatidylinositol-4-phosphate (PI4P) or phosphatidylserine (PS) for PI4P between membranes and thereby to use PI4P metabolism to generate sterol or PS gradients in the cell, respectively. Here, we describe a full strategy to measure in vitro a sterol/PI4P exchange process between artificial membranes using Fo¨rster resonance energy transfer (FRET)-based assays and a standard spectrofluorometer. Such an approach can serve to better characterize the activity of known sterol/PI4P exchangers, but also to reveal whether ill-defined ORP/Osh proteins or LTPs belonging to other families have such an exchange activity. Besides, this protocol is amenable to test whether molecules can act as Orphilins, which have been found to inhibit the sterol/PI4P exchange activity of certain ORPs. Last, our strategy to measure in real-time PI4P transport using a known lipid-binding domain can serve as a basis for the design of novel in vitro protocols aiming to detect other lipid species. Key words ORP/Osh proteins, Sterol, Dehydroergosterol, PI4P, FRET, Liposome, Recombinant protein

1

Introduction Lipids are precisely distributed within and between the membranes of eukaryotic cells [1, 2]. Such a distribution relies on lipid metabolism as well as lipid transport processes that are mostly mediated by cytosolic proteins called LTPs. These proteins belong to various families but share a common feature: they have a domain with a lipid-pocket, generally to host only one lipid molecule at a time. Decrypting the mode of action of these LTPs is an important issue in cell biology [3–5]. It is commonly assumed that a LTP is able to extract a lipid from a donor organelle membrane, shield this highly hydrophobic molecule from the aqueous medium, diffuse through the cytosol, and deposit this lipid in a target organelle

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_19, © Springer Science+Business Media, LLC, part of Springer Nature 2019

269

270

Nicolas-Fre´de´ric Lipp and Guillaume Drin

[6–9]. These LTPs work either across long distances or confined in membrane contact sites where two organelles are in close apposition. In vitro approaches are invaluable to decipher at the molecular level how these LTPs work (e.g., some recent references [10–14] and several chapters in this book). Recently, we devised protocols to establish that LTPs, Osh4p, and OSBP (Oxysterol-binding protein), belonging to the ORP/Osh family, have the ability to bind alternately PI4P and sterol as well as to exchange these lipids between membranes [15]. We demonstrated with OSBP that a sterol/PI4P exchanger can use PI4P metabolism to transport vectorially sterol from the endoplasmic reticulum (ER) to the Golgi [16]. In vitro, we showed that Osh4p is able to dissipate a preexisting PI4P gradient to generate in return a sterol gradient between two distinct membranes [17]. This mechanism of counterexchange fueled by PI4P metabolism is used by other ORP/Osh proteins to transport a lipid that is not sterol. Indeed, Osh6p/Osh7p in yeast and ORP5/ORP8 in humans use a PI4P gradient to transport PS from the ER to the plasma membrane [18, 19]. Along with this, it has been revealed that the sterol/PI4P exchange activity of OSBP has a pivotal role in the replication of many human viruses, including Hepatitis C virus [20] or in cancer [21]. Consequently, OSBP appears as a potential pharmaceutical target and small molecules called Orphilins have been found to inhibit its exchange activity [21–23]. In that respect, the availability of our in vitro approach to measure a sterol/PI4P exchange activity with high temporal resolution proved to be useful to characterize some of these compounds [22, 23]. Herein, we describe in detail how to set up this approach and kinetic measurements which simply require a standard L-format spectrofluorometer. Our strategy combines two FRET-based assays to accurately measure (1) sterol transport from LA to LB liposomes, which are present in an equimolar amount and mimic each an organelle membrane (ER and trans-Golgi, respectively, in reference to our studies on Osh4p and OSBP [16, 17]) and (2) PI4P transport in the opposite direction. It is noteworthy that these two experiments are performed under the same conditions (i.e., identical buffer, temperature, total lipid concentration, sterol, and PI4P concentration) such as to observe whether these two transport processes are coupled, and thus the occurrence of sterol/PI4P exchange (Fig. 1) [17]. In the first assay (Fig. 1a) we use LA-1 and LB-1 liposomes that are essentially made of phosphatidylcholine (see Note 1). LA-1 liposomes contain specifically both dehydroergosterol (DHE, 5 mol%), a natural fluorescent analogue of ergosterol and cholesterol (major sterol in yeast and human, respectively) and a second fluorescent lipid called 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-(5-dimethylamino-1-naphthalenesulfonyl) (DNS-PE, 2.5 mol%). DHE molecules excited at 310 nm transfer resonance energy to the dansyl group of DNS-PE [24]. The level of FRET

Fluorescence-Based Sterol and PI4P Transport Assays

271

Fig. 1 Description of the two transport assays. In the DHE transport assay, LA-1 liposomes (DOPC/DHE/DNS-PE, 92.5/5/2.5 mol/mol) are mixed with LB-1 liposomes (DOPC/PI4P 96/4 mol/mol). FRET between DHE and DNS-PE in the LA-1 liposomes diminishes if DHE is transported to the LB-1 liposomes. In the PI4P transport assay, DOPC/PI4P/Rhod-PE liposomes (94/4/2 mol/mol, LB-2) are incubated with 250 nM NBD-PHFAPP. DOPC liposomes (LA-2) doped with 5 mol % DHE are added. If PI4P transport occurs, this provokes a dequenching of the NBD signal corresponding to the translocation of NBD-PHFAPP from LB-2 to LA-2 liposomes

depends on the amount of DHE in the LA-1 membranes. If DHE is transferred from LA-1 to LB-1 liposomes, this elicits a decrease in the FRET signal because DHE is not anymore in close proximity to DNS-PE. The normalization of this signal allows quantifying the transport of DHE from LA-1 to LB-1 liposomes. Using this assay, Osh4p was found to efficiently transport DHE and this, in a proportional manner to the amount of PI4P initially present in the LB-1 liposomes [17]. Here, we describe a protocol in which LB-1 liposomes incorporate 4 mol% PI4P. To measure the transport of PI4P in the reverse direction, from LB to LA liposomes, we designed a PI4P-probe, derived from the Pleckstrin Homology (PH) domain of the human phosphoinositol4-phosphate adapter protein (FAPP-1, Uniprot: Q9HB20). This domain has a binding site for the PI4P headgroup and a hydrophobic wedge that inserts into the membrane [25]. We reengineered this domain to include a single solvent-exposed cysteine (C13), localized in the hydrophobic wedge, to which is covalently linked a polarity-sensitive NBD (7-nitrobenz-2-oxa-1,3-diazol) fluorophore [17]. Its ability to recognize PI4P at the liposome surface is unaffected by the level of sterol [17]. In the transport assay (Fig. 1b), NBD-PHFAPP is mixed with LB-2 liposomes, which contain 4 mol% PI4P and 2 mol% Rhod-PE, and LA-2 liposomes, which incorporate 5 mol% DHE. At time zero, the NBD fluorescence is quenched due to a FRET process with Rhod-PE. If PI4P is

272

Nicolas-Fre´de´ric Lipp and Guillaume Drin

transported from LB-2 to LA-2 liposomes (e.g., upon injecting Osh4p), a fast dequenching occurs, due to the translocation of NBD-PHFAPP onto LA-2 liposomes (Fig. 1b). In our assay, NBD-PHFAPP is completely bound to the membranes which contain an accessible amount of PI4P between 2 and 4 μM [17]. As a consequence, the NBD signal directly reflects the distribution of NBD-PHFAPP between LA and LB liposomes and can thus be easily normalized and converted into an amount of transported PI4P. This tracking method, albeit indirect, has at least two advantages. First it does not rely on a fluorescently labeled PI4P that is unlikely to be properly accommodated by the binding pocket of ORP/Osh proteins, as it bears an extra bulky moiety. Second, this approach offers a far higher time resolution than methods based on radiolabeled PI4P and liposome separation. In addition, such a PI4P is not commercially available and remains uneasy to produce. Along with this, we have indications that the ability of a protein to transport PI4P is not impacted by the presence of NBD-PHFAPP. Given the amount of NBD-PHFAPP and of accessible PI4P in our assays, only 6.25% of PI4P can be monopolized by the probe during the measurement. In addition, only 0.9% of the membrane surface is covered by the probe, considering the maximal membrane surface that one PH molecule can cover (
4.8 nm2, estimated from references [26, 27]) and the total surface of liposomes (LA + LB liposomes; 5.1  1016 nm2 with an area of 0.7 nm2 per lipid). Experimentally, we observed that doubling or dividing by two the amount of NBD-PHFAPP in the sample did not change the transport kinetics that we recorded (unpublished results). Our strategy, which is detailed step by step below, allows for measuring the transport of sterol along its concentration gradient between two membranes by Osh4p via the dissipation of a preexisting PI4P gradient. It can be modified to measure the transport of sterol against its concentration gradient simply by changing the initial amount of DHE in the LA and LB liposomes as done in [17]. Our strategy is also potent to further characterize known sterol/PI4P exchangers via structure–function analysis or to screen novel drugs for their ability to block an exchange activity. Likewise, it can help to define in the future whether other members of the ORP/Osh family are sterol/PI4P exchangers. Besides, it can serve to explore the ability of LTPs, belonging to other families, to transport sterol and/or PI4P, as done recently with START-like proteins [13]. NBD-PHFAPP has been used to follow the transport of phosphatidylinositol-4,5-bisphosphate (PIP2) between membranes, as it can recognize this phosphoinositide [13, 28]. We also developed a NBD-C2Lact domain to specifically follow PS transport between liposomes and thereby unveiled the capacity of Osh6p and Osh7p to serve as PS/PI4P exchangers [19]. This exemplifies that our FRET-based strategy can be adapted to measure in vitro other transport processes by using lipid-binding

Fluorescence-Based Sterol and PI4P Transport Assays

273

domains whose tridimensional structure and lipid-binding site are known.

2

Materials

2.1 Protein Purification and Labeling

1. pGEX-PHFAPP and pGEX-Osh4p plasmids (available on request from our lab). 2. Sterilized water. 3. Electro-competent bacteria: BL21 Gold Competent Cells (Agilent). 4. Ampicillin: Prepare a 50 mg/mL stock solution with filtered and sterilized water and store it at 20  C. 5. LB medium: Lennox LB Broth medium without glucose (10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl) prepared with deionized water and autoclaved. 6. LB/Ampicillin: Dilute ampicillin stock solution to 1/1000 in LB medium (50 μg/mL final concentration). 7. 1 M Isopropyl β-D-1-thiogalactopyranoside (IPTG) stock solution in water. 8. 1 M Dithiothreitol (DTT) stock solution in water. Prepare 1 mL aliquots and store them at 20  C. 9. PBS: 137 mM NaCl, 2.7 mM KCl, 10 mM NaH2PO4, 1.8 mM KH2PO4, autoclaved and stored at 4  C. 10. TN buffer: 50 mM Tris–HCl, pH 7.4, 150 mM NaCl, filtered with a Stericup filtration unit (Merck) or with a membrane vacuum filtration equipment. Store at 4  C. 11. TND buffer: Dilute 1 M DTT solution to 1/500 in TN buffer (final concentration 2 mM DTT). 12. 200 mM PMSF (Phenylmethylsulfonyl fluoride) stock solution in isopropanol. Store at 4  C. 13. Lysis buffer. To prepare 50 mL, dissolve one tablet of complete EDTA-free protease inhibitor cocktail (Roche) in TN buffer by vortexing (see Note 2). Add 250 μL of 200 mM PMSF stock solution. Complete with other antiproteases, i.e., 50 μL of 1 mM bestatin stock solution, 50 μL of 1 mg/mL pepstatin A in MeOH, and 50 μL of 1 mM phosphoramidon solution (these solutions are stored at 20  C). Add 100 μL of 1 M DTT stock solution to obtain a final concentration at 2 mM. 14. DNAse I Recombinant, RNAse free, in powder. 15. 2 M MgCl2 solution. Filter the solution using a 0.45 μm filter.

274

Nicolas-Fre´de´ric Lipp and Guillaume Drin

16. Thrombin stock solution: Dissolve 20 units in 1 mL of doubledeionized water and prepare 25 μL aliquots in 0.5 mL Eppendorf tubes. Then freeze and store at 80  C. 17. 10 mM CaCl2 stock solution in water. 18. Dimethylformamide (DMF), anhydrous, >99% pure. 19. IANBD Amide (N,N0 -Dimethyl-N-(Iodoacetyl)-N0 - (7-Nitrobenz-2-Oxa-1,3-Diazol-4-yl)Ethylenediamine). Dissolve 25 mg of IANBD in 2.5 mL of dimethylsulfoxide (DMSO) and prepare 25 aliquots of 100 μL in 1.5 mL screw cap tubes. Do not completely screw the cap. Then, remove DMSO in a freeze-dryer to obtain 1 mg of dry IANBD per tube. Tubes are closed and stored at 20  C in the dark. 20. 10 mM L-cysteine solution in degassed water stored at 20  C. 21. Glycerol (99% pure). 22. 50 mL polypropylene tube (e.g., Falcon). 23. Eppendorf tubes (0.5, 1 and 2 mL). 24. Electroporator (e.g., Eppendorf 2510) and electroporation cuvette (e.g., Cell Projects, 50  2 mm gap). 25. Glutathione Sepharose 4B beads stored in 20% (v/v) ethanol at 4  C. 26. French pressure cell press (French press). 27. Fixed-angle rotor and tubes for ultracentrifuge (e.g., Ti45, Beckman). 28. Ultracentrifuge (e.g., Beckman). 29. Amicon Ultra-4 and Ultra-15 centrifugal filter units, each with a molecular weight cut-off (MWCO) of 3 and 10 kDa (Millipore). 30. UV/Visible absorbance spectrometer with quartz cuvettes. 31. Poly-Prep® chromatography column (with a 0–2 mL bed volume and a 10 mL reservoir). 32. Econo-Pac® chromatography columns (1.5  12 cm). 33. Illustra NAP 10 desalting column (GE healthcare). 34. Thermomixer. 35. XK 16/70 column packed with Sephacryl S200HR for FPLC system (GE healthcare). ¨ KTA purifier (GE healthcare) or similar. 36. FPLC system: A 37. TG-SDS running buffer: 25 mM Tris, pH 8.6, 192 mM glycine, 0.1% SDS. 38. 4 highly denaturating Laemmli buffer: 62.5 mM Tris–HCl, pH 6.8, 10% (v/v) glycerol, 8% (w/v) SDS, 5% (v/v) β-mercaptoethanol, 0.005% (w/v) bromophenol blue.

Fluorescence-Based Sterol and PI4P Transport Assays

275

39. Precast 15% acrylamide gel SDS-PAGE. 40. SDS-PAGE standards, low range (Biorad). 2.2 Liposome Preparation and Fluorescent Assays

1. Chloroform (CHCl3) and methanol (MeOH). 2. Phospholipid stock solutions (usually in CHCl3) are aliquoted in 2 mL amber glass vials filled with argon and tightly sealed with a Teflon cap. Vials are stored at 20  C. DOPC (1,2-dioleoyl-sn-glycero-3-phosphocholine, 25 mg/mL in CHCl3), brain PI4P (L-α-phosphatidylinositol-4-phosphate, 5 mg/mL in CHCl3/MeOH/H2O (20:9:1)) and DNS-PE (1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-(5-dimethylamino-1-naphthalenesulfonyl), 1 mg/mL) and Rhod-PE (1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl), 1 mg/mL) are from Avanti Polar Lipids. 3. DHE stock solution: Dissolve 5 mg of DHE powder (Ergosta5,7,9(11),22-tetraen-3β-ol E2634, purity ~96%, SigmaAldrich) in 1 mL of MeOH. Dilute 10–20 μL of this solution to 1/30 in MeOH and record an absorbance spectrum of this solution against a blank using two quartz cells (e.g., Hellma 104-002B-QS with a chamber of 2 mm in width and 1 cm optical path) and a standard double beam UV/visible spectrometer. Determine the DHE concentration by measuring the DHE absorbance peak at 325 nm, considering a molar extinction coefficient of 13,000 M1 cm1. If necessary, adjust the volume of the stock solution with MeOH to set the DHE concentration at 2.5 mM. Store 500 μL aliquots in 2 mL glass vials at 20  C, as done with phospholipids. 4. C16:0/C16:0-PI4P stock solution: Dissolve 1 mg of C16:0/ C16:0-PI4P powder (Echelon Lipids) in 250 μL of MeOH and 250 μL of CHCl3. Then complete with methanol to 1 mL. The solution must become clear. 5. HK buffer: 50 mM HEPES-KOH, pH 7.4, 120 mM K-Acetate. Dissolve 23.8 g of HEPES and 23.6 g of K-Acetate in 2 L of deionized water at room temperature. Adjust the pH to 7.4. Prepare frozen aliquots of 50 mL. 6. HKM buffer: Add 25 μL of a 2 M MgCl2 stock solution to 50 mL of HK buffer. 7. Rotary evaporator (Buchi B-100 or similar). 8. Pear-shaped glass flasks (25 mL, 14/23, Duran). 9. Aluminum foil. 10. 4 mm-diameter glass beads (Sigma-Aldrich). 11. Avanti Polar Mini-Extruder with two 1 mL gas-tight Hamilton syringes.

276

Nicolas-Fre´de´ric Lipp and Guillaume Drin

12. Prefilters (10 mm in diameter, Avanti Polar Lipids). 13. Polycarbonate filters (19 mm in diameter, Avanti Polar Lipids) with pore size of 0.2 μm. 14. Hemolysis tubes with a cap. 15. Water bath. 16. Liquid nitrogen. 17. UV/visible spectrofluorometer with a temperature-controlled cell holder and stirring device. 18. Quartz cuvette (e.g., Hellma) for UV/visible fluorescence (minimum volume of 600 μL). 19. Small magnetic PFTE stirring bar (5  2 mm). 20. Hamilton syringes (10, 25, and 50 μL).

3

Methods

3.1 Purification and Fluorescent Labeling of NBDPHFAPP

3.1.1 Expression of PHFAPP in E. coli

The PH domain of the FAPP-1 protein was mutated to introduce a unique cysteine within its membrane-binding interface (at position 13 of the protein) and label it with a NBD fluorophore. PHFAPP (T13C/C45S/C57S/C94S) construct is cloned into a pGEX-4T3 vector (GE healthcare) to be expressed in fusion to GST. A thrombin cleavage site is located between the GST construct and the N-terminal end of the PH domain. We describe here a purification protocol based on the use of a French press and a final gel filtration ¨ KTA FPLC system. This protocol can be modified step using an A to include other methods for disrupting bacteria and other FPLC chromatography devices. It is important to use very freshly degassed and filtered lysis buffer supplemented with 2 mM DTT to avoid any oxidation of cysteine. In contrast, to label the protein, it is mandatory to perfectly remove DTT using a desalting column and a freshly degassed DTT-free buffer. All the steps must be done on ice at 4  C, to avoid any degradation of the protein. 1. Mix 2 μL of pGEX-PHFAPP plasmid (at ~65 ng/μL) with 20 μL of electro-competent bacteria and 18 μL of sterilized water. Transform the bacteria by electroporation with a 1500 V pulse for 5 ms. Then, resuspend the bacteria with 150 μL of LB medium and let them recover for 1 h at 37  C. 2. Dilute the 150 μL culture into 25 mL of LB/Amp medium in a 200 mL sterilized Erlenmeyer flask. 3. Incubate the flask in a shaker at 37  C at 185 rpm overnight. The preculture can be used the next day or stored one day long at 4  C before induction procedure.

Fluorescence-Based Sterol and PI4P Transport Assays

277

4. Inoculate 4  500 mL of LB/Amp medium with 5 mL of preculture suspension and let bacteria growth at 37  C under agitation at 220 rpm in 2 L Erlenmeyer flasks. 5. Once the optical density of the suspension, measured at a wavelength (λ) of 600 nm, reaches a value between 0.6 and 0.7, add 500 μL of 1 M IPTG stock solution in each flask to induce protein expression. Maintain the flasks for 4 h at 37  C under constant shaking at 185 rpm. 6. To harvest the bacteria, fill four polypropylene centrifuge bottles each with 500 mL of culture and centrifuge for 30 min at 3000  g at 4  C. 7. Remove the supernatant and resuspend each pellet with 25 mL of cold PBS buffer. Then, pool the bacterial suspension and fill two 50 mL falcon tubes with the suspension. 8. Centrifuge the tubes for 30 min at 3500  g at 4  C. Discard the supernatant, let drain the residual buffer upon absorbent paper, and store the pellets at 20  C (see Note 3). 3.1.2 Purification and Labeling of PHFAPP

To check the proper progress of the purification, it is requested to collect 30 μL aliquots at different steps of the protocol to perform an analysis on a 15% SDS-PAGE gel (Fig. 2). Add to each aliquot 30 μL of 4 highly denaturating Laemmli buffer, vortex, spin down, and heat the aliquot at 95  C in a dry bath heater. Store the tubes at 20  C until analysis. Generally, we run a first gel to check the first part of the purification process prior to the gel filtration (Fig. 2a). At step 28, for the SDS-PAGE analysis of the protein content of the collected fractions after gel filtration (Fig. 2b, c), we mix 25 μL aliquot of each interesting fraction with 15 μL of Laemmli buffer prior to heating and loading onto a gel. 1. On ice, prepare two 50 mL falcon tubes that each contain 50 mL of lysis buffer. 2. Add up to 30 mL of lysis buffer to the two tubes, prepared at step 8 of Subheading 3.1.1, which contain a bacterial pellet. Let defreeze the pellets on ice for 5–10 min. Crush the pellets with a stainless spatula, vortex, and resuspend well by pipetting back and forth the suspension with a 25 mL pipette until it is homogeneous. 3. Lysis is performed using a prechilled French press. Load 30 mL of sample inside the press cell and apply a pressure to reach 1000 psi. Collect the lysate in the same tube, keep the tube on ice, and continue this first round of lysis with the second sample. 4. Repeat step 3 one more time to obtain a smoothed lysate (see Note 4).

278

Nicolas-Fre´de´ric Lipp and Guillaume Drin

Fig. 2 NBD-PHFAPP purification (a) SDS-PAGE analysis is used to check the first part of the purification protocol before labeling. It is used to verify the presence of the protein at different steps of the procedure. The arrowheads indicate the position of the PHFAPP domain (black arrowhead), of the GST alone (open arrowhead) and of the GST-PH construct (gray arrowhead). An equivalent of 2.5 μL of each aliquot is loaded onto the gel (b) Gel filtration step. The chromatogram shows the evolution of the tryptophan and NBD absorbances monitored, respectively, at a wavelength of 280 and 480 nm as a function of the volume of buffer passed on the column. The peak corresponding to the protein (after 150–200 mL of elution) and the 2.5 mL fractions that are collected are framed in red. (c) SDS-PAGE analysis of collected fractions. The first picture is acquired under UV illumination without staining and reveals the presence of the NBD-PHFAPP construct since it emits fluorescence. The second picture shows the same gel after a Sypro staining procedure (Note that the gel was protected from light during migration to avoid any bleaching of the NBD fluorophore)

5. Use the remaining 40 mL of lysis buffer to rinse the French press cell immediately after lysate collection and to adjust each volume of lysate (~30 mL) to a final volume of 50 mL. 6. Add 125 μL of 2 M MgCl2 solution (final concentration 2 mM) and a pinch of DNAse I (using a spatula) to each 50 mL lysate to cut DNA in small fragments, thereby limiting the viscosity of the sample. Incubate on ice for 30 min. Take an aliquot for gel analysis.

Fluorescence-Based Sterol and PI4P Transport Assays

279

7. Transfer each 50 mL lysate in a precooled polycarbonate ultracentrifuge tube (two in total) and run a centrifugation for 1 h at 186,000  g at 4  C with an appropriate rotor and ultracentrifuge. 8. During centrifugation, dispense 1.4 mL of Glutathione Sepharose 4B bead slurry into two 50 mL tubes. Wash the beads twice by centrifugation at 500  g for 5 min with 20 mL of TND buffer. 9. Once the centrifugation is over, take an 30 μL aliquot from the supernatant and transfer the supernatant from each ultracentrifuge tube in the two 50 mL tubes containing washed beads. For gel analysis, resuspend the debris pellet in one of the ultracentrifuge tube with 50 mL of TND buffer and collect a 30 μL aliquot. 10. Place the tubes on a rotator to constantly homogenize the solution and the beads for 3–4 h at 4  C. 11. Pool the bead suspensions in a unique Econo-Pac® chromatography column. Let decant the beads by gravity flow to remove buffer and unbound proteins. Take an aliquot from the eluate for analysis. 12. Wash the beads by resuspending them with 20 mL of TND buffer followed by decantation. Repeat this step twice. Collect and pool the eluates and take an aliquot for analysis. After decantation, a volume of ~2 mL of a suspension of bead, to which GST-PHFAPP is bound, lies at the bottom of the column. 13. Fill two 2 mL snap-cap microcentrifuge tubes each with 1 mL of bead suspension. Complete each tube with TND buffer up to 1.970 mL. Take a 30 μL aliquot from one tube for analysis (aliquot B1). Add 10 μL of 10 mM CaCl2 solution and 25 μL of human thrombin protease to each tube for cleaving the PHFAPP construct off the GST domain. 14. Close and place the tubes overnight on a rotator at 4  C for an optimal cleavage. 15. Add 10 μL of 200 mM PMSF solution to the 2 mL bead suspension for irreversibly stopping the thrombin activity. 16. Centrifuge the tubes at 700  g for 3 min 30 s. Collect out of each tube up to 1 mL of supernatant, which contains soluble PHFAPP domain, pool, and transfer it in 2 mL snap-cap microcentrifuge tube (E1 eluate) which is maintained on ice. Avoid pipetting beads during this step. 17. Resuspend the beads with 1 mL of TN buffer to wash them and repeat three times step 16 to recover most of the protein. Each time, pool and transfer the supernatants into a new 2 mL tube (E2, E3, and E4 eluates). For SDS-PAGE analysis take an

280

Nicolas-Fre´de´ric Lipp and Guillaume Drin

aliquot from each eluate and from the bead suspension (aliquot B2) at the end of the wash. 18. Pool the supernatants (i.e., the content of 4 tubes, ~8 mL) collected during steps 16 and 17 into a 10 mL Poly-Prep® chromatography column. Retrieve by gravity flow the eluate that is free of any contaminating beads as those latter are retained on the bed support. 19. Concentrate the solution to ~1 mL in an Ultra-4 centrifugal filter unit with a MWCO of 3 kDa using a centrifugation speed of 1500  g. 20. Equilibrate an Illustra NAP 10 desalting column with TN buffer. Load 1 mL of the PHFAPP concentrated sample inside the column and let it fully penetrate into the Sephadex gel. Then, elute the sample by adding 1.5 mL of freshly degassed TN buffer devoid of DTT to the gel and collect the eluate by gravity flow into a 2 mL snap-cap microcentrifuge tube. 21. Dilute 50–100 μL of the eluate to 1/6 in TN buffer and record an absorbance spectrum from 230 to 450 nm against a blank using two quartz cuvettes. Determine the PHFAPP concentration using the maximal absorbance at 280 nm and considering an extinction coefficient ε ¼ 29,450 M1 cm1. 22. To guarantee a complete labeling of the PHFAPP domain, it is preferred to add IANBD to the protein with a ten-times molar excess. Dissolve 1 mg of IANBD powder in DMF considering that the final volume of DMF used for labeling must not exceed 5% (v/v) of PHFAPP sample volume. Thus adjust the volume of DMF (VDMF) to dissolve IANBD following these formulas. V DMF ¼ V max  ðm0 =mÞ with V max ¼ 0:05  V and m ¼ 10, 000  C  V  MW IANBD C is the deduced concentration of PHFAPP based on absorbance measurement as described in step 21, V is the volume of PHFAPP solution, MWIANBD is the molecular weight of IANBD (420 g/mol), m is the required amount of IANBD in mg, Vmax is the maximum volume of DMF usable for PHFAPP labeling, m0 corresponds to 1 mg of dry IANDB powder (see Note 5). 23. Add the IANBD solution to the PHFAPP protein sample and shake thoroughly the reaction mixture at 800–900 rpm for 30 min at 25  C using a thermomixer protected from light. Let the reaction continue for 90 min on ice to reach its completion.

Fluorescence-Based Sterol and PI4P Transport Assays

281

24. Add L-cysteine to the reaction mixture to quench unreacted IANBD. L-cysteine must be in a tenfold molar excess relative to IANBD. 25. Transfer the NBD-PHFAPP solution into an Ultra-4 centrifugal filter unit with a MWCO of 3 kDa. Add 5 mL of TN buffer and concentrate the sample until 2 mL to remove a large part of free NBD from the protein by centrifugation at 1500  g. Repeat this washing step twice more. 26. Store the 2 mL protein sample at 4  C before the gel filtration procedure. Storage shall not exceed a day long. 27. Prior to gel filtration, check that no orange deposit is present at the bottom of the tube. Aggregation may occur during the concentration step due to an excessive amount of protein. In that case, centrifuge the sample for 10 min at 540,000  g. Collect and inject the clarified supernatant on the column for gel filtration. 28. Gel filtration is performed on a Sephacryl S200HR column. Column is fully equilibrated with TND buffer. Flow rate is set at 1 mL/min. Tryptophan and NBD absorbances are followed at λ ¼ 280 and 480 nm, respectively. 29. Load the protein sample in a 2 mL clean injection loop. Inject the sample on the column and immediately collect the eluate in hemolysis tubes using a fractionation volume of 2.5 mL. A major peak, which is concomitantly detected at λ ¼ 280 and 480 nm, appears after an elution volume of ~150 mL (Fig. 2b). All the fractions that correspond to this peak are analyzed on a 15% SDS-PAGE gel (Fig. 2c). 30. Pool all the fractions that only contain NBD-PHFAPP protein (~13 kDa). Add glycerol to the sample at a final concentration of 10% (v/v) for the cryo-protection of the protein. Concentrate the sample using an Ultra-15 centrifugal filter unit with a MWCO of 3 kDa to a final volume of 2 mL by centrifugation at 1500  g. 31. Record an absorbance spectrum from 230 to 650 nm against a blank (see Note 6) using two quartz cuvettes. Determine the NBD-PHFAPP concentration using the maximal absorbance at 280 and 495 nm, and considering an extinction coefficient ε ¼ 29,450 M1 cm1 and 25,000 M1 cm1 for the protein and the NBD moiety, respectively. If the two concentration values are alike, this means that the protein is labeled at 1:1 ratio with the NBD group (see Note 7). 32. Make 50 μL aliquots of protein in 0.5 mL Eppendorf tubes. Flash-freeze the tubes in liquid nitrogen and store them at 80  C for several months.

282

Nicolas-Fre´de´ric Lipp and Guillaume Drin

3.2 Osh4p Purification

We express Osh4p in fusion in GST without any labeling step. The plasmid (pGEX-Osh4p) with the coding sequence of the fulllength Osh4p (1–434 residues) is used to transform electrocompetent bacteria. The purification protocol, except a few differences, is similar to the one used to obtain NBD-PHFAPP. Follow steps 1–8 described in Subheading 3.1.1 but let the bacteria grow overnight at 30  C after IPTG induction. Thereafter, follow all the steps described in Subheading 3.1.2, but at step 19, concentrate the protein sample to 2 mL for gel filtration and skip steps 20–27. Use ultracentrifugal filter units with a MWCO of 10 kDa. The protein is usually collected once 60–90 mL of buffer are passed on the gel filtration column.

3.3 Preparation of Liposomes

Perform all the steps at room temperature otherwise specified. Handle organic solvent, rotavapor, and liquid nitrogen with care. Use fresh, filtered, and degassed HK buffer. 1. To make the liposomes for the transport assays, mix volumes of lipid stock solutions in a pear-shaped glass according to Table 1 (also see Notes 8 and 9). Adjust the volume of each mixture to 1 mL with pure CHCl3. 2. Write down the name of each liposome type (LA, LB, . . .) on the neck of the respective flask. Wrap the flasks containing a mixture with fluorescent lipids (i.e., DHE, DNS-PE, or Rhod-PE) with aluminum foil. 3. Connect the flask to a rotary evaporator. Remove the solvent under vacuum while heating the flask at 25  C for at least 1 h using a rotation speed of 100 rpm (see Note 10). A lipid film appears on the glass surface. Fill the evaporator and the flask with argon. Remove the flask and place it in a vacuum chamber for 45 min to remove any solvent traces. 4. Hydrate the film with 2 mL of HK buffer. Add a half-dozen of 4 mm-diameter glass beads to the flask to optimize the resuspension. Then, gently vortex the flask for at least 2 min in order to obtain a suspension of multilamellar lipid vesicles (MLVs) with a 4 mM lipid concentration. Divide the suspension into 1 mL aliquots stored in screw cap micro tubes. Properly label each tube cap with the liposome name. 5. Freeze and thaw the tubes five times (using liquid nitrogen and a water bath at 37  C, respectively). Liposomes can then be directly extruded or stored at 20  C. 6. Wet two prefilters with HK buffer, and place each of them over the orifice, inside the O-ring inner diameter of each membrane support of the mini-extruder. Insert one of the membrane supports, with the prefilter, into the extruder outer casing with the O-ring facing up. Place one 19 mm polycarbonate filter over the prefilter support and O-ring. Carefully place the

Fluorescence-Based Sterol and PI4P Transport Assays

283

Table 1 Volume of lipid stock solutions to be mixed for liposome preparation Lipid stock solution Lipid composition Liposome (mol/mol) LA-1

DOPC DHE DNS-PE (25 mg/mL) (2.5 mM) (1 mg/mL) 233 μL

DOPC/DHE/ DNS-PE 92.5/5/2.5

LA-no DHE DOPC/DNS-PE 97.5/2.5

160 μL

245 μL

Brain PI4P (5 mg/mL)

Rhod-PE (1 mg/mL)

199 μL

199 μL

LB-1

DOPC/PI4P 96/4

241 μL

62 μLa

LA-2

DOPC/DHE 95/5

239 μL

LB-2

DOPC/PI4P/ Rhod-PE 94/4/2

236 μL

LA-Eq

DOPC/DHE/ PI4P 95.5/2.5/2

240 μL

80 μL

31 μLb

LB-Eq

DOPC/DHE/ PI4P/Rhod-PE 93.5/2.5/2/2

235 μL

80 μL

31 μLb

160 μL 62 μLa

200 μL

200 μL

Alternatively, add 306 μL of C16:0/C16:0-PI4P at 1 mg/mL Alternatively, add 153 μL of C16:0/C16:0-PI4P at 1 mg/mL

a

b

second membrane support into the casing (O-ring facing down). Place the retainer nut on the threaded end of the extruder outer casing and screw the retainer nut by hand just until it is sufficiently tight. 7. Prepare two hemolysis tubes: one filled with 1 mL of MLVs suspension and a second one to store the liposomes once they are extruded (see Note 11). 8. Load the MLVs sample into one of the 1 mL gas-tight Hamilton syringes and carefully connect the syringe at one end of the mini-extruder. 9. Connect the second syringe, which is empty, to the other end of the mini-extruder. The syringe plunger must be set to zero. 10. Gently push the plunger of the first syringe until the lipid solution is completely transferred to the second syringe. Gently push the plunger of the second syringe to transfer the solution back to the first syringe. Repeat these operations to pass the suspension 21 times through the filter and fragment the MLVs

284

Nicolas-Fre´de´ric Lipp and Guillaume Drin

into smaller and unilamellar liposomes of ~200 nm in diameter. Fill the second syringe with the final liposomes suspension and transfer this suspension into a hemolysis tube. The final suspension should be less milky and more opalescent than the suspension prior to extrusion. 11. Store extruded liposomes at 4  C and in the dark if they contain fluorescent lipids. Liposomes should be used within 2 days. 3.4 Real-Time Measurement of Sterol Transport

Measurements are performed using a standard fluorimeter (90 format; e.g., Shimadzu RF5301PC or JASCO FP-8300) with a temperature-controlled cell holder. To accurately measure kinetics, it is mandatory to continuously stir the sample at a constant temperature (30  C or 37  C depending whether it is a yeast or mammalian protein, respectively). The protocol described thereafter is for measuring transport in a 600 μL sample contained in a cylindrical quartz cell (see Note 12) and the use of a Shimadzu RF5301PC apparatus. Use fresh, filtered, and degassed HKM buffer (see Note 13). Maintain the tubes filled with the suspension of extruded liposomes at room temperature during the experiment. Wrap the tubes containing fluorescent liposomes with aluminum foil and/or store them in an opaque box to protect them from light. 1. To measure the DHE-to-dansyl FRET, tune the monochromator such as to excite the sample with a wavelength (λ) equal to 310 nanometer (nm) along with a short bandwidth (1–3 nm) to prevent any DHE photobleaching (see Note 14). Record the dansyl emission at λ ¼ 525 nm with a larger bandwidth (5 nm) to maximize the signal-to-noise ratio. Set up the acquisition time at 24 min and, if possible, with a time resolution ½ L   ½ P   2K þ ð½L T  þ ½P T  þ 2K d Þ2  4½P T ½L T  > T T d < =  ln 1 þ > > 2K d > > : ; ð10Þ where [LT] and [PT] are the total concentration of the ligand and protein, respectively. 3.4.2 Use of the First Approach to Determine the Ligand Binding Affinity of STARD6

1. Perform the temperature-induced denaturation curves at different protein-to-ligand molar ratios. 2. Transform raw mdeg values in mean residue molar ellipticity (Eq. 1) and compare spectra (Fig. 1a). For STARD6, the binding of testosterone induces a slight increase in helical content, judging by the minima at 222 nm, but overall, there are no substantial secondary structural changes upon ligand binding. Although changes at 222 nm and 218 nm may indicate an increase in α-helical or β-structure upon binding, more precise estimates of structural changes can be carried out on a complementary basis by spectral deconvolution using computer programs [18, 19]. 3. Determine from the temperature-induced denaturation curves, the apparent melting temperature (Tm), the apparent enthalpy of unfolding at Tm, (ΔHu(Tm)), and the apparent temperaturedependent Gibbs free energy of unfolding (ΔGu(Tm)) by the simulation of the temperature denaturation curves (Eqs. 5–7) with a model describing the equilibrium for a two-state

Ligand Binding to STARD6 by CD-TSA

301

Fig. 1 Far-UV CD spectroscopy of STARD6, experimental determination of the thermodynamic parameters, ΔHu(Tm), ΔHu,L(Tm), ΔCp,u,L and evaluation of the binding constant (Kb) for STARD6-testosterone complex. (a) Mean residue ellipticity of STARD6 (full line) and in complex with testosterone (dashed line). Each spectrum was baseline-corrected for buffer and ligand. (b) Thermal melting curve and temperature dependence of the population of unfolded state (Pu) for Apo-STARD6 (Black full line) and in complex at ratio of 1:0.5 (light gray), 1:1 (mid gray), 1:2 (dark gray) and 1:5 (dashed line) with testosterone. The curves for molar ratio of 1:1.5 and 1:3 are not represented for the sake of visual clarity. (c) Estimation of ΔCp,u,L by a linear fit of ΔHu,L(Tm) in function to Tm,L values determined experimentally for each STARD6:testosterone ratio. The value for the apo-STARD6 (□) falls under the fit line, reflecting the non-contribution of the binding enthalpy to the denaturation enthalpy. (d) Dependence of ΔΔGu,L(T) on testosterone concentration used for the determination of the Kb value at the Tm of the apo-state, using the model in which ligands binds selectively to the native state

unfolding mechanism and assuming an heat capacity of unfolding (ΔCp,u) value of 1 kcal mol1 K1 (as described in [14]) (Fig. 1b). 4. Estimate the change in the heat capacity of unfolding upon ligand binding ΔCp,u,L by plotting the resultant ΔHu(Tm) as a function of Tm at each ligand concentration (Tm,L). Since ΔCp, u,L is assumed to be temperature independent, its value is given by the slope of the linear curve (Fig. 1c). 5. Once ΔCp,u,L is known, one can determine ΔGu,L at all (Tm,L) and ligand concentrations (Eq. 9) and determine the ΔΔGu,L

302

Danny Le´tourneau et al.

Table 2 Thermodynamic stability of apo-STARD6 and STARD6-Steroids complexes obtained from the simulation of the temperature denaturation Tm (
C)

ΔGu(25
C) ΔGu(37
C) Kd at Tmd

ΔHu(T )a,c

STARD6 (apo) 54.0 (0.3)b 106.5 (2.3)b Testosterone a

58.2 (0.2)

117.9 (2.3)

Kd at Tme

9.4 (0.5) 4.6 (0.5) 16.6 (0.5) 9.5 (0.5) 3.08 (0.23) 3.01 (0.21)

1

Values given in kcal·mol Standard deviation of the fits c Calculated at 1:5 protein-to-ligand ratio d 106 M1 (first approach) e 106 M1 (second approach) b

values at the temperature of the melting of the apo-state (Tm,apo) for various ligand concentrations. 6. Finally, determine the affinity of the interaction by fitting the canonical equation for the free energy of ligand binding in the stoichiometric regime by plotting the ΔΔGu,L(Tm,L) as a function of ligand concentration (Eq. 10 and Fig. 1d). The thermodynamic parameters for the Apo-STARD6 and the STARD6-testosterone complexes are presented in Table 2. 3.4.3 Second Approach

It is also possible to evaluate the Kb(Tm) with an approximate and simpler approach. Schellman have demonstrated theoretically that stability enhancement in the presence of ligands can be used for determining substrate binding constants by the analysis of denaturation curves [20]. Hence, the analysis of denaturation curves determined in the presence of ligands can be used for determining ligands binding constants. This method was validated later by Pace and McGrath for the binding of lysozyme and its natural ligand [21]. Accordingly, if a ligand binds to a protein only in its folded form, the increase in stability of the complex compared to the free protein (apoprotein) can be used to evaluate the binding constant. When the changes in enthalpy of folding upon binding are small, Kb(Tm) values can be estimated using the following equation: ΔT m ¼

T m, L T m, apo R ln ð1 þ K b ½L free Þ ΔH u, L

ð11Þ

where ΔTm is the change in the midpoints (Ku ¼ 1) of the thermal denaturation curves, Tm,L and Tm are the temperatures where Ku ¼ 1 in the presence and absence of ligand, respectively. ΔHu,L is the enthalpy of unfolding of the protein in the presence of the ligand, Kb(Tm,L) is the equilibrium constant for the binding of the ligand to the folded protein at Tm, [Lfree] is the free concentration of ligand, and R is the gas constant [21–23].

Ligand Binding to STARD6 by CD-TSA

303

1. Perform temperature-induced denaturation curves in temperatures ranging from 10 to 90
C with a rate heating of 1
C/ min. 2. The raw mdeg values are transformed in mean residue molar ellipticity. 3. The apparent melting temperature (Tm), the apparent enthalpy of unfolding of the complex, (ΔHu,L(T)), and the apparent Gibbs free energy of unfolding (ΔGu(T)) are determined by the simulation of the temperature denaturation curves with a model describing the equilibrium for a two-state unfolding mechanism for a monomeric protein considering the enthalpy of unfolding at Tm(ΔHu(T)) and assuming a (ΔCp,u) value of 1 (Eq. 4) [24]. As for the first approach (vide supra), the temperature-dependent population of the unfolded state (Pu(T)) is obtained from the simple two-state model, where the folded state of the protein is in equilibrium with its unfolded state. 3.4.4 Use of the Second Approach to Determine the Ligand Binding Affinity of STARD6

4

Under conditions where a ligand binds only the folded state of a protein, the increase in stability (ΔΔGu(T)) of the complex relative to the free protein can be used to estimate the Kb(Tm,L) or 1/Kd(Tm,L). For small enthalpy (ΔHu(T)) changes, the Kd(T) can be estimated using Eq. 11. The only parameters required are the transition temperatures in the presence and absence of ligand (Tm,L and Tm,apo) and the enthalpy of unfolding of the protein in the presence of the ligand ΔHu,L. For STARD6, we used a 1:5 STARD6:Testosterone ratio and the free-ligand concentration (Lfree) at Tm,L is approximated by ([LTotal]  [PTotal]/2). The calculation gives a value of 3.01 ( 0.21) for the Kd at Tm,L using this approximation, which is a very good approximation compared to the first approach that gives a value of 3.08 ( 0.23) (see Table 2).

Notes 1. In CD spectroscopy, optically active substances should be avoided and buffers should be as transparent as possible. All reagents should be of the highest purity available and organic solvents should be of spectroscopic grade. Buffer and salt concentration must be kept as low as possible to avoid any excessive noise or other artifacts in the spectra. Sample solution should be degassed and filtered (e.g., 0.1–0.2 μm) or centrifuged before use. Reducing agent like TCEP for reducing disulfide bonds must be used with care (generally used in concentration equivalent to the number of cysteine residue inside the protein) and can be completely avoided if there are no cysteine or

304

Danny Le´tourneau et al.

disulfide bonds in the protein. For review on CD spectroscopy and buffer composition see [19, 25–29]. 2. Accurate sample (proteins) concentrations are essential for the analysis of CD spectra and TSA. Typical quantities are 300 μL of 0.1–0.15 mg/mL solution for a 1-mm pathlength cuvette. 3. The instrument, calibrated regularly, should always be purged with high-purity nitrogen for at least 2 min (or according to the manufacturer) before starting the light source and while making measurements. 4. Cuvettes should always be cleaned immediately after use, rinsed thoroughly with distilled water and then ethanol, and finally dried appropriately using a stream gaseous N2. 5. Typical parameters are a scan rate of 100 nm/min and a response of 2 s. Collecting multiple scans will improve the signal-to-noise (S/N) ratio which is proportional to the square root of the number of scans. The spectral bandwidth diminishes noise by increasing light throughput. The bandwidth should always be 2 nm or less to avoid distorting the spectrum. The wavelength range should generally be scanned from 250 nm to the lowest possible wavelength (ex. 190 nm) depending on the buffer being used. 6. In thermal unfolding experiments, the temperature should be increased slowly (no more than 1
/min). 7. Generally, proteins equilibrate slowly, and thus accurate measurements of thermodynamic parameters require an equilibration time long enough to allow full equilibration. Preliminary experiments must be performed in order to determine proper equilibration time and to assure complete unfolding at the highest temperature studied. It is also a good practice to take preliminary spectra at 5, 10, 25, 37, 50, 70, and 90
C to determine what is the best observation wavelength, for your specific experiment, that displays the greatest dynamic range and signal-to-noise ratio. Moreover, attention should be given to isodichroic points (if present), secondary structure spectral transition, and, finally, reversibility should always be tested for the unfolding and folding mechanism. 8. The baseline correction, with the same cuvette and buffer and using the same instrument settings, is mandatory: the solution may contain unusual components like dithiothreitol (DTT), PMSF, high concentrations of salt or common chelators (EDTA) or ligands that can distort the spectrum.


9. The parameters to be fit are ΔHu(T), ΔCp,u, T , [θ]F, and [θ]U. To reduce the number of parameters, ΔCp,u may be set to an arbitrary value (e.g., 1 kcal mol1 K1) and the values of [θ]F

Ligand Binding to STARD6 by CD-TSA

305

and [θ]U may be set as constants rather than parameters to be minimized if they are known. References 1. Clark BJ, Wells J, King SR et al (1994) The purification, cloning, and expression of a novel luteinizing hormone-induced mitochondrial protein in MA-10 mouse Leydig tumor cells. Characterization of the steroidogenic acute regulatory protein (StAR). J Biol Chem 269:28314–28322 2. Lin D, Sugawara T, Strauss JF 3rd et al (1995) Role of steroidogenic acute regulatory protein in adrenal and gonadal steroidogenesis. Science 267:1828–1831 3. Mathieu AP, Fleury A, Ducharme L et al (2002) Insights into steroidogenic acute regulatory protein (StAR)-dependent cholesterol transfer in mitochondria: evidence from molecular modeling and structure-based thermodynamics supporting the existence of partially unfolded states of StAR. J Mol Endocrinol 29:327–345 4. Lehoux JG, Mathieu A, Lavigne P et al (2003) Adrenocorticotropin regulation of steroidogenic acute regulatory protein. Microsc Res Tech 6:288–299 5. Alpy F, Tomasetto C (2005) Give lipids a START: the StAR-related lipid transfer (START) domain in mammals. J Cell Sci 118:2791–2801 6. Kudo N, Kumagai K, Matsubara R et al (2010) Crystal structures of the CERT START domain with inhibitors provide insights into the mechanism of ceramide transfer. J Mol Biol 396:245–251 7. Lavigne P, Najmanivich R, Lehoux JG (2010) Mammalian StAR-related lipid transfer (START) domains with specificity for cholesterol: structural conservation and mechanism of reversible binding. Subcell Biochem 51:425–437 8. Roderick SL, Chan WW, Agate DS et al (2002) Structure of human phosphatidylcholine transfer protein in complex with its ligand. Nat Struct Biol 9:507–511 9. Romanowski MJ, Soccio RE, Breslow JL et al (2002) Crystal structure of the Mus musculus cholesterol-regulated START protein 4 (StarD4) containing a StAR-related lipid transfer domain. Proc Natl Acad Sci U S A 99:6949–6954 10. Thorsell AG, Lee WH, Persson C et al (2011) Comparative structural analysis of lipid binding START domains. PLoS One 6(6):e19521

11. Tsujishita Y, Hurley JH (2000) Structure and lipid transport mechanism of a StAR-related domain. Nat Struct Biol 7:408–414 12. Letourneau D, Bedard M, Cabana J et al (2016) STARD6 on steroids: solution structure, multiple timescale backbone dynamics and ligand binding mechanism. Sci Rep 6:28486 13. Privalov PL, Khechinashvili NN (1974) A thermodynamic approach to the problem of stabilization of globular protein structure: a calorimetric study. J Mol Biol 86:665–684 14. Roostaee A, Barbar E, Lehoux JG et al (2008) Cholesterol binding is a prerequisite for the activity of the steroidogenic acute regulatory protein (StAR). Biochem J 412:553–562 15. Santoro MM, Bolen DW (1988) Unfolding free energy changes determined by the linear extrapolation method. 1. Unfolding of phenylmethanesulfonyl alpha-chymotrypsin using different denaturants. Biochemistry 27:8063–8068 16. Bolen DW, Santoro MM (1988) Unfolding free energy changes determined by the linear extrapolation method. 2. Incorporation of delta G degrees N-U values in a thermodynamic cycle. Biochemistry 27:8069–8074 17. Layton CJ, Hellinga HW (2010) Thermodynamic analysis of ligand-induced changes in protein thermal unfolding applied to highthroughput determination of ligand affinities with extrinsic fluorescent dyes. Biochemistry 49:10831–10841 18. Whitmore L, Wallace BA (2008) Protein secondary structure analyses from circular dichroism spectroscopy: methods and reference databases. Biopolymers 89:392–394 19. Greenfield NJ (2006) Using circular dichroism spectra to estimate protein secondary structure. Nat Protoc 1:2876–2890 20. Schellman JA (1975) Macromolecular binding. Biopolymers 14:999–1018 21. Pace CN, McGrath T (1980) Substrate stabilization of lysozyme to thermal and guanidine hydrochloride denaturation. J Biol Chem 255:3862–3865 22. Elwell M, Schellman J (1975) Phage T4 lysozyme. Physical properties and reversible unfolding. Biochim Biophys Acta 386:309–323

306

Danny Le´tourneau et al.

23. Elwell ML, Schellman JA (1977) Stability of phage T4 lysozymes. I. Native properties and thermal stability of wild type and two mutant lysozymes. Biochim Biophys Acta 494:367–383 24. Letourneau D, Lorin A, Lefebvre A et al (2012) StAR-related lipid transfer domain protein 5 binds primary bile acids. J Lipid Res 53:2677–2689 25. Greenfield NJ (2004) Circular dichroism analysis for protein-protein interactions. Methods Mol Biol 261:55–78

26. Greenfield NJ (2006) Analysis of the kinetics of folding of proteins and peptides using circular dichroism. Nat Protoc 1:2891–2899 27. Greenfield NJ (2015) Circular dichroism (CD) analyses of protein-protein interactions. Methods Mol Biol 1278:239–265 28. Kelly SM, Price NC (2000) The use of circular dichroism in the investigation of protein structure and function. Curr Protein Pept Sci 1:349–384 29. Martin SR, Bayley PM (2002) Absorption and circular dichroism spectroscopy. Methods Mol Biol 173:43–55

Chapter 21 Synthesis of Fluorescent Membrane-Spanning Lipids for Studies of Lipid Transfer and Membrane Fusion Gu¨nter Schwarzmann Abstract For uncompromised in vitro assays for intermembrane lipid transfer and membrane fusion fluorescent membrane-spanning lipids have proved to be invaluable tools. These lipids in contrast to phosphoglycerolipids and sphingolipids are resistant to spontaneous as well as protein-mediated intermembrane transfer. Here I describe the synthesis of some homo-substituted fluorescent bipolar membrane-spanning lipids that bear a fluorescent tag either directly or via a phosphoethanolamine spacer to the lipid core. For the synthesis the lipid core of the bipolar membrane-spanning lipids, i.e., the tetraether lipid caldarchaeol, is prepared from cultures of the archaea Thermoplasma acidophilum. Key words Caldarchaeol, Membrane-spanning lipids, Tetraetherlipids, Fluorescent sn-caldarchaeylbis-phosphoethanolamine, Fluorescent sn-caldarchaeyl-bis amine, Fluorescence resonance energy transfer, Intermembrane lipid transfer, Membrane fusion

1

Introduction For in vitro studies of membrane fusion and intermembrane lipid transfer, the use of liposomes as model biological membranes has become the method of choice. The procedures to detect fusion of lipid vesicles or intermembrane lipid transfer must be highly sensitive and also reliable. In the past, in transfer and fusion assays, biotinylated donor liposomes were separated from fluorescent acceptor liposomes by streptavidin-coated magnetic beads applying N-Biotin-phosphatidylethanolamine (Biotin-PE) in donor liposomes for their separation from acceptor membranes. N-NBDphosphatidylethanolamine (NBD-PE) served for controlling loss of acceptor membranes [1, 2]. However, this procedure does not allow real-time measurements and is also hampered by the fact that a drastic shift in pH is necessary for the binding of biotin residues to streptavidin, which may mess up the results for binding and transfer. Previously, Fo¨rster resonance energy transfer (FRET) [3] has been used to measure membrane fusion [4] as well as

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_21, © Springer Science+Business Media, LLC, part of Springer Nature 2019

307

308

Gu¨nter Schwarzmann

intermembrane lipid transfer [5, 6]. This procedure allows realtime measurements and does not require separation of donor from acceptor membranes. It turned out, however, that the fluorescent vesicle markers NBD-PE and N-Rhodamine-phosphatidylethanolamine (Rh-PE) used as energy donor and acceptor, respectively, were also transferred by some lysosomal transfer proteins. This was also demonstrated for a dansylated PE [7]. That is why these hitherto applied lipids are not well suited as membrane markers to study fusion and lipid transfer mediated by lysosomal lipid transfer and/or fusion proteins. I tried to overcome this problem by using the respective fluorescent reporter lipids that cannot be extracted from lipid bilayer membranes. I was aware that archaea, e.g., Thermoplasma acidophilum, can survive in the most extreme and arcane of environments with respect to heat and acidity because they contain bipolar tetraether lipids that stabilize their plasma membrane [8]. Thus, I made use of their basic tetraether lipid caldarchaeol to synthesize various membrane-spanning phosphoethanolamine (MSPE) and membrane-spanning lipid (MSL) as shown in Fig. 1 and in reference [9]. Since caldarchaeol is so far not commercially available, this compound is prepared from archaea. The syntheses are based on well-known chemical procedures and do not require sophisticated devices.

Fig. 1 Structures of homo-substituted MSPE (a) and MSL (b) derivatives with fluorescent tags linked via a phosphoethanolamine spacer (MSPE derivatives) or directly tied to the lipid core (MSL derivatives). MSPE membrane-spanning phosphoethanolamine, MSL membrane-spanning lipid

Fluorescent Membrane-Spanning Lipids

2

309

Materials

2.1 Solvents and Chemicals

1. Solvents: 2-Propanol, Benzene, Chloroform (CHCl3), Dichloromethane (DCM), Dimethylformamide (DMF), Ethanol (EtOH), Ethyl acetate (EtOAc), Methanol (MeOH), nHexane, Tetrahydrofurane (THF), Toluene, Trichloroethylene (TCE). Dry solvents are obtained immediately prior to use using filtration tubes equipped with a polyethylene frit and filled with either basic, neutral, or acidic aluminum oxide to remove traces of water and peroxides from ethers, and amines from DMF, respectively, similarly as reported in [10]. 2. Fluorescent dyes: N-7-Nitrobenz-2-oxa-1,3-diazol-4-yl fluoride (NBD-F), ATTO647N succinimidyl ester (ATTO647N SE) (ATTO-TEC), (5-(and 6-)-caboxytetramethylrhodamine succinimidyl ester, mixed isomers (TAMRA-SE). 3. 1,4,7,10,13-Pentaoxacyclopentadecane (15-crown-5). 4. 4-Toluenesulfonyl chloride (TsCl). 5. Ammonium acetate (NH4OAc). 6. 40% (v/v) acetic acid in water. 7. Ethanolamine. 8. N,N0 -Dicyclohexylcarbodiimide (DCC). 9. N,N-Diisopropylethylamine (DIPEA). 10. N-Hydroxysuccinimide (NHSI). 11. Phosphorus oxychloride (POCl3). 12. Platinum oxide (PtO2). 13. Sodium azide (NaN3) in powder. 14. Sodium hydrogencarbonate (NaHCO3). 15. Sodium sulfate (Na2SO4). 16. Triethylamine (NEt3). 17. Gaz: Hydrogen, Argon, Nitrogen. 18. Liquid nitrogen.

2.2

Chromatography

1. DEAE-Sephadex A25. 2. Glass column. 3. LiChroprep Si 60 (40–63 μm) (Merck). 4. HPTLC Silica gel 60 F254 plates (Merck). 5. CuSO4 spray. Dissolve 100 g of CuSO4∙5H2O and 80 g of H3PO4 in 800 mL of H2O and adjust to 1 L with H2O. 6. CHCl3/EtOAc (9:1 to 1:1, v/v). 7. CHCl3/MeOH/1 M NH3 (65:25:4, v/v/v).

310

Gu¨nter Schwarzmann

8. CHCl3/MeOH/2 M NH3 (60:35:8, v/v/v). 9. CHCl3/MeOH/2 M NH3 (65:25:4, v/v/v). 10. CHCl3/MeOH/H2O (20:10:1, v/v/v). 11. CHCl3/MeOH/H2O (3:7:1, v/v/v). 12. CHCl3/MeOH/H2O (60:35:8, v/v/v). 13. CHCl3/MeOH/H2O (65:25:4, v/v/v). 14. DCM/THF (1:1, v/v). 15. n-Hexane/EtOAc (1:1, v/v). 16. n-Hexane/EtOAc (4:1, v/v). 17. n-Hexane/EtOAc (7:3, v/v). 18. n-Hexane/EtOAc (9:1, v/v). 19. TCE/MeOH (4:1, v/v). 20. TCE/THF (3:1, v/v). 2.3 Isolation of Caldarchaeol from Archaea Thermoplasma acidophilum

1. Archaea Thermoplasma acidophilum (DSM 1728, German Collection of Microorganisms and Cell cultures, Braunschweig, Germany). 2. Culture medium: Dissolve 0.372 g of KH2PO4, 1.32 g of (NH4)2SO4, 0.247 g of MgSO4∙7H2O, 0.074 g of CaCl2∙2H2O, 19.3 mg of FeCl3∙6H2O, 1.8 mg of MnCl2∙4H2O, 4.5 mg of Na2B4O7∙10H2O, 22 mg of ZnSO4∙7H2O, 0.05 mg of CuCl2∙2H2O, 0.03 mg of Na2MoO4∙2H2O, 0.038 mg of VOSO4∙5H2O, and 0.02 mg of CoSO4∙7H2O in 1.8 L of water. Add 10 g of glucose and 6 g of yeast extract under stirring. Adjust the pH to 1.42 by addition of about 4 mL of 50% (v/v) H2SO4. Filter this medium to remove flocculants that may have formed after standing for 4 h at room temperature. 3. Cellulose thimble (e.g., 43 mm  123 mm). 4. Lipid extraction solution: CHCl3/MeOH (2:1, v/v). 5. Soxleth device. 6. Rotary evaporator. 7. 10% (w/v) Na2CO3. 8. CHCl3/EtOAc (9:1, v/v). 9. Dilute H2SO4: Add 1 mL of H2SO4 in 1 L of H2O. 10. 1 M HCl in MeOH.

2.4 Liposome Preparation and FRET Experiments

1. Lipids: 1,2-Dioleoyl-sn-glycero-3-phosphocholine (DOPC), 2-NBD-(glyco)sphingolipid (e.g., 2-NBD-GM1, GM2,GM3, -GlcCer, -LacCer, and -Cer) [11], cholesterol, N-(Lissamine rhodamine B sulfonyl)-1,2-sn-dipalmitoylglycerophosphoethanolamine (Rh-PE), N-(N-7-Nitrobenz-2-oxa-1,3-

Fluorescent Membrane-Spanning Lipids

311

diazol-4-yl)-sn-1,2-dipalmitoylglycerophosphoethanolamine (NBD-PE), Bis(monooctadecenoylglycero)phosphate (BMP). 2. 20 mM Citrate buffer, pH 4.3, with 150 mM NaCl. Dissolve 2.395 g of sodium citrate dehydrate, 2.277 g of citric acid, and 8.76 g of NaCl in 900 mL of H2O. Adjust with either NaOH or HCl to pH 4.3 and add H2O to 1 L. 3. Sonifier bath. 4. Mini extruder (Avestin). 5. Polycarbonate filters with pore size of 100 nm. 6. Quartz cuvette with a path length of 5 mm.

3

Methods

3.1 Synthesis of Fluorescent Membrane-Spanning Lipids from Caldarchaeols with a Phosphoethanolamine Spacer at Both Ends of the Lipid Core (MSPE) 3.1.1 Isolation of Caldarchaeol from Archaea Thermoplasma acidophilum

Tetraether lipid caldarchaeol is, to my knowledge, currently not commercially available and must, therefore, be isolated from archaea. Caldarchaeol is a generic term of tetraether lipids containing various amounts of cyclopentane rings that depend on growth conditions [12, 13]. This isolation involves culturing of the archaea in acidic medium under aerobic condition at elevated temperature, collecting cells and extracting their total lipid by organic solvents prior to acid hydrolysis for destroying ester lipids and cleaving the polar head groups of di- and tetraether lipids archaeol and caldarchaeol (including a little isocaldarchaeol). Purification is performed by chromatographic means [9]. 1. Culture archaea in culture medium in a 3 L Erlenmeyer flask at 59  C on a rotary shaker at 140–150 rpm. 2. Determine cell growth by measuring optical density (OD) at 600 nm. 3. When OD reaches 1.2–1.5, harvest archaea by centrifugation at 7000  g. 4. Wash the cell pellet several times by successive centrifugation at 7000  g with 30 mL of dilute H2SO4 until the supernatant appears clear and colorless. 5. Finally wash the pellet with water to remove the acid and store frozen until use. 6. Place up to 10 g of wet pellet of archaea into a cellulose thimble. 7. Extract total lipids with 800 mL of lipid extraction solution in a Soxleth device for 60 h under reflux. 8. Evaporate the extracted lipids in a 2 L round bottom flask to dryness in a rotary evaporator. To the residue add 1 L of 1 M HCl and boil under reflux for 36 h. Thereafter evaporate the solvent and dissolve the residue in 700 mL of CHCl3.

312

Gu¨nter Schwarzmann

9. Wash the organic solution in a separatory funnel twice with 100 mL of 10% Na2CO3 and then three times with 100 mL of H2O. Then evaporate the organic solvent to obtain a dark brown viscous syrup. 10. Dissolve the viscous residue in 20 mL of CHCl3 and apply 5 mL of the solution to preparative silica gel column chromatography using a glass column containing about 250–300 g of silica gel LiChroprep Si 60 and being pre-equilibrated with CHCl3/EtOAc (9:1, v/v). Elute the diether and tetraether lipids by step gradients with different mixtures of CHCl3/ EtOAc. Start with 100 mL of CHCl3/EtOAc (9:1, v/v) and continue with 100 mL of CHCl3/EtOAc (8:2, v/v) followed by 200 mL of CHCl3/EtOAc (7:3, v/v) and 700 mL of CHCl3/EtOAc (1:1, v/v). Collect fractions of about 50 mL (see Note 1). 11. Analyze collected fractions for caldarchaeols by TLC with CHCl3/EtOAc (8:2, v/v) and visualize compounds with CuSO4 spray and heating according to reference [14]. Collect those with Rf ¼ 0.38–0.26 (see Note 2). 12. Repeat this chromatographic procedure with the remaining 15 mL of crude lipid extraction. 13. Purify the respective fraction from above by rechromatography using a similar step gradient (see Note 3). 14. Analytical data: Caldarchaeols were characterized by ESI-QTOF mass spectrometry in positive ion mode. They yield ions [M + H]+ at m/z 1302.3, 1300.0, 1298.3, 1296.3, and 1294.3 for caldarchaeols with zero, one, two, three, and four five-membered rings, respectively. Mass spectrometry by addition of NH4OAc yield the respective ions [M + NH4]+ at 1319.3, 1317.3, 1315.3, 1313.3, and 1311.3. 3.1.2 Caldarchaeyl-BisPhosphoethanolamine (MSPE)

Since the synthesis of PE as described previously [15] has proved efficacious, the synthesis of MSPE follows, with slight modifications, the same steps. Because the synthetic reactions are moisturesensitive, the reactants have to be dissolved in freshly dried solvents and processed in small polyethylene reaction tubes (Eppendorf) that may be cooled in a metal block with frequent droppings of liquid nitrogen to keep the temperature slightly below 5  C. The synthetic steps are outlined in Fig. 2. For stereochemical orientation of the methyl groups in the tetraether lipid core see [9, 16–18]. 1. Dissolve 97 mg of caldarchaeol (75 μmol corresponding to 150 μmol OH groups) in 600 μL of TCE and take it up in a gas-tight syringe. 2. Slowly inject this solution within 10 min via a needle pierced through the cap of a 2 mL polyethylene reaction tube into a

Fluorescent Membrane-Spanning Lipids

313

Fig. 2 Synthetic route to a membrane-spanning phosphoethanolamine derivative MSPE. Caldarchaeol is converted into caldarchaeo-bis-phosphoric ester dichloride using phosphoroxychloride in the presence of triethylamine. Treatment with ethanolamine produces caldarchaeo-bis-oxazaphospholane, which upon mild acid hydrolysis yields MSPE as a precursor for membrane-spanning spacer-linked fluorophores (MSPE derivatives) as shown in Fig. 1

precooled solution of 22 μL (225 μmol) of POCl3 dissolved in a mixture of 75 μL of n-hexane, 32 μL (225 μmol) of NEt3, and 150 μL of TCE. 3. Withdraw the needle and close the pierced hole with selfadhesive tape. Keep the mixture at about 5  C for further 10 min. 4. Centrifuge the mixture at 20  C in a table centrifuge for 3 min at 8600  g and aspirate the resulting clear solution beneath the floating NEt3Cl. Dilute with 100 μL of toluene and then dry the mixture under an argon or nitrogen jet to remove solvents and the excess of POCl3.

314

Gu¨nter Schwarzmann

5. Dissolve the oily residue representing caldarchaeo-bis-phosphoric acid ester dichloride in 375 μL of THF at 10  C. To this solution add dropwise under gentle stirring at 10  C a mixture of 9.7 μL (9.8 mg, 160 μmol) ethanolamine dissolved in 600 μL of NEt3 and 375 μL of TFH. 6. Centrifuge at 20  C for 3 min at 8600  g to sediment NEt3Cl, wash the pellet twice with 100 μL of THF, and combine with the supernatant. Remove solvents with an argon or nitrogen jet. 7. Dissolve the residue containing caldarchaeo-bis-oxazaphospholane in a mixture made of 500 μL of THF, 300 μL of MeOH, and 200 μL of 2-propanol. Then add 500 μL of 40% acetic acid solution and keep the solution at room temperature overnight. 8. Centrifuge the mixture at 20  C for 3 min at ~9000  g. Dry the supernatant with a nitrogen jet and redissolve together with the sediment in 3 mL of CHCl3/MeOH (1:1, v/v). 9. Subject this solution to Folch partitioning [19]. 10. Analyze the lower phase by TLC with CHCl3/MeOH/H2O (65:25:4, v/v/v) and visualize compounds with CuSO4 spray and heating according to ref. [14]. The main spot with Rf ¼ 0.25 is MSPE. Small amounts of side products may also be detected such as MSPE methyl ester (Rf ¼ 0.52), and phosphoryl-caldarchaeo-phosphoethanolamine (Rf ¼ 0.13) and its methyl ester (Rf ¼ 0.30). 11. Dry the lower phase and dissolve the residue in 5 mL of CHCl3/MeOH/H2O (3:7:1, v/v/v). Filter this solution through a small column containing 4 mL of DEAE-Sephadex A25 in the acetate form and wash the column with 5 mL of CHCl3/MeOH/H2O (3:7:1, v/v/v). Dry the filtrate and washings to obtain non-retained MSPE and its methyl ester. 12. Purify MSPE and separate from its methyl ester by column chromatography on about 10 g LiChroprep Si 60 using CHCl3/MeOH/H2O (65:25:4, v/v/v) as eluent. Collect fractions with pure MSPE as detected by TLC. The yield is in the range of 50% (see Note 4). 13. Analytical data: MSPE is characterized by ESI-QTOF mass spectrometry in positive ion mode. MSPE with two fivemembered rings yield prominent ions [M + H]+ at m/z 1544.3 and 1545.3. Ions [M + H]+ at m/z 1546.3 and 1542.3 for MSPE with one or three cyclopentane rings, respectively, may also be found.

Fluorescent Membrane-Spanning Lipids 3.1.3 Caldarchaeyl-BisN-NBDPhosphoethanolamine (NBD-MSPE)

315

1. Dissolve 3.4 mg (2.2 μmol) of MSPE in 200 μL of TCE/MeOH (4:1, v/v) and add 10 μL of DIPEA. 2. Add to this solution, in increments of 20 μL, 1.5 mg (8.1 μmol) of NBD-F dissolved in 340 μL of TCE/THF (3:1, v/v) over a period of 2 h at 20  C. After this time check the reaction by TLC with CHCl3/MeOH/H2O (65:25:4, v/v/v). A yellow band with stark green fluorescence upon excitation with UV light (366 nm) should be visible with Rf ¼ 0.28 and represents NBD-MSPE. 3. Purify NBD-MSPE from side products and excess NBD-F by column chromatography on 10 g of LiChroprep Si 60 using CHCl3/MeOH/H2O (65:25:4, v/v/v) as eluent. Check fractions by TLC and collect pure NBD-MSPE. Yield is 3.7 mg (1.96 μmol, 89%). 4. Analytical data. NBD-MSPE (containing molecules with one, two, and three cyclopentane rings in the lipid core) is characterized by ESI-QTOF mass spectrometry in positive ion mode. NBD-MSPE yield prominent ions [M + 2NH4]2+ at m/z 952.6, 953.6, and 951.6 for NBD-MSPE with two, one, or three cyclopentane rings, respectively.

3.1.4 Caldarchaeyl-BisN-5-[and-6]-CarboxytetramethylrhodaminylPhosphoethanolamine (TAMRA-MSPE)

1. Dissolve 3.1 mg (2 μmol) of MSPE in 200 μL of CHCl3/ MeOH/H2O (65:25:4, v/v/v) and add 30 μL of DIPEA. 2. To this solution add 3.2 mg (6 μmol) of solid TAMRA-SE under stirring. The solid should have dissolved after a few minutes. After a period of 6 h at 20  C the reaction is checked by TLC with CHCl3/MeOH/2 M NH3 (60:35:8, v/v/v). A new dark red fluorescent compound of TAMRA-MSPE should have formed (Rf ¼ 0.63) at the expense of MSPE (Rf ¼ 0.51) (see Note 5). 3. Purify TAMRA-MSPE from side products and excess TAMRASE by column chromatography on 10 g of LiChroprep Si 60 using CHCl3/MeOH/H2O (60:35:8, v/v/v) as eluent. Check fractions by TLC and collect pure TAMRA-MSPE. Yield is 2.9 mg (1.2 μmol, 60%). 4. Analytical data: TAMRA-MSPE (containing molecules with one, two, and three cyclopentane rings in the lipid core) is easily characterized by ESI-QTOF mass spectrometry in positive ion mode. TAMRA-MSPE yield prominent ions [M + 2H]2+ at m/z 1183.8, 1184.8, and 1185.8 for TAMRA-MSPE with three, two, and one cyclopentane rings, respectively.

316

Gu¨nter Schwarzmann

3.1.5 Caldarchaeyl-BisN-ATTO647NPhosphoethanolamine (ATTO-MSPE)

1. Dissolve 1.6 mg (1 μmol) of MSPE in 100 μL of CHCl3/ MeOH/H2O (60:35:8, v/v/v) and add 3 μL of DIPEA. 2. To this solution add 3.1 mg (4 μmol) of solid ATTO647N SE under stirring. The solid dissolves instantly. Keep the reaction for 2 h at 20  C and check product formation by TLC with CHCl3/MeOH/H2O (20:10:1, v/v/v). A new dark blue compound of ATTO-MSPE should have formed (Rf ¼ 0.8) at the expense of MSPE. 3. Purify ATTO-MSPE from side products and excess ATTO647N SE by column chromatography on 10 g of LiChroprep Si 60 using a step gradient of CHCl3/MeOH/ H2O (20:10:1, v/v/v) and CHCl3/MeOH/H2O (60:35:8, v/v/v) as eluent. Check fractions by TLC and collect pure ATTO-MSPE. Yield is about 2 mg (0.75 μmol, 75%). 4. Analytical date: ATTO-MSPE (containing molecules with one, two, and three cyclopentane rings) yield prominent ions [M + 2H]2+ at m/z 1401.0, 1400.0, and 1399.0 characteristic for ATTO-MSPE with one, two, and three cyclopentane rings, respectively (see Note 6).

3.2 Synthesis of MembraneSpanning Lipids with the Fluorophore Directly Linked to the Lipid Core (MSL)

The synthesis of this type of membrane-spanning lipids came from the idea to substitute a highly reactive NH2 group for a less reactive OH group in as much as most commercially available fluorophores are amine reactive. The synthesis of MSL is very effective due to the use of crown ether 15-crown-5 [20] in the conversion of the intermediate bis-tosylate to the bis-azide, especially since a tosylate is an excellent leaving group in nucleophilic substitution reactions. For structures see Fig. 1 and for the synthetic steps see Fig. 3.

3.2.1 Caldarchaeyl-BisTosylate

1. Dissolve 108 mg (83 μmol) of caldarchaeol in 1 mL of benzene in a polyethylene reaction tube and dry this solution with an argon jet to remove traces of H2O, if any. 2. Redissolve the residue in 1.4 mL of DCM and 70 μL of NEt3. Then add 40 mg (209 μmol) of TsCl and vortex. 3. After 4 h at 20  C check the reaction on HPTLC plates with nhexane/EtOAc (7:3, v/v). Fluorescence quench results from caldarchaeyl-bis-tosylate and also from the little remainder of TsCl. Visualization with CuSO4 spray and heating according to [14] should confirm that almost all caldarchaeol has been transformed into its bis-tosylate. 4. Wash the organic solution several times with 0.1 M NaHCO3 and dry with solid Na2SO4. Purify caldarchaeyl-bis-tosylate by column chromatography on 10 g LiChroprep Si 60 using nhexane/EtOAc (4:1, v/v) as eluent. Check fractions by TLC with n-hexane/EtOAc (7:3, v/v) and collect pure

Fluorescent Membrane-Spanning Lipids

317

Fig. 3 Synthetic route to membrane-spanning lipids with fluorescent tags directly tied to the tetraether lipid core. Caldarchaeol reacts with tosyl chloride in the presence of triethylamine to give caldarchaeyl-bis-tosylate, which in the presence of the crown ether 15-crown-5 is transformed into caldarchaeyl-bisazide. This azide is subjected to hydrogenolysis in the presence of platinum to yield caldarchaeyl-bis-amine as a precursor for fluorescent MSL derivatives without a phosphoethanolamine spacer as shown in Fig. 1

caldarchaeyl-bis-tosylate with Rf ¼ 0.76. Freeze-dry from benzene. Yield is 130 mg (97%). 5. Analytical data. The bis-tosylate yields ions [M + H]+ at m/z 1608.3, 1606.3, and 1604.4 for the derivatives with one, two, and three cyclopentane rings, respectively, in the lipid core.

318

Gu¨nter Schwarzmann

3.2.2 Caldarchaeyl-BisAzide

1. Dissolve 65 mg (40 μmol) of caldarchaeyl-bis-tosylate and 17.6 mg (80 μmol) of 15-crown-5 in 200 μL of THF and 400 μL of DMF. To the clear solution add 65 mg (1 mmol) of finely powdered NaN3. Stir the mixture at 50  C overnight. 2. Cool the reaction mixture to room temperature and add 6 mL of DCM. Wash the organic phase several times with 2 mL of H2O to remove salts. Dry the organic solution over solid Na2SO4 and reduce to volume to 2 mL. 3. Purify caldarchaeyl-bis-azide by column chromatography on 10 g of LiChroprep Si 60 using n-hexane/EtOAc (9:1, v/v) as eluent. Check fractions by TLC with n-hexane/EtOAc (9:3, v/v) and collect pure caldarchaeyl-bis-azide with Rf ¼ 0.7. The product is obtained as a colorless viscous oil. Yield is 52 mg (96%) (see Note 7). 4. Analytical data. The bis-azide yields ions [M + H]+ at m/z 1350.3, 1348.3, and 1346.3 for the derivatives with one, two, and three cyclopentane rings, respectively, in the lipid core. With NH4OAc the mass spectrum showed the respective ions [M + NH4]+ at m/z 1367.3, 1365.3, and 1363.3. The IR spectrum displays a characteristic band for azide oscillation at 2090 cm1.

3.2.3 Caldarchaeyl-BisAmine

1. Dissolve 5.4 mg (4 μmol) of caldarchaeyl-bis-azide in 200 μL of n-hexane and 400 μL of EtOH in a hydrogenation tube. Add 3 mg of PtO2 and displace the air first by argon and then by hydrogen. Perform hydrogenolysis of the azide overnight at 20  C and 1 bar hydrogen pressure. 2. Check for completion of hydrogenation by TLC with CHCl3/ MeOH/H2O (65:25:4, v/v/v). Detect caldarchaeyl-bisamine by CuSO4 spray and heating [14]. The educt should have been completely transformed into the bis-amine (Rf ¼ 0.6). 3. Remove the finely dispersed platinum by centrifugation and dry the clear supernatant in a nitrogen jet. Yield is about 4.9 mg (3.8 μmol, 95%). This caldarchaeyl-bis-amine is used in the synthesis of fluorescent membrane-spanning lipids without further purification (see Note 8).

3.2.4 Caldarchaeyl-BisN-NBD (NBD-MSL)

1. Treat a solution of 4 mg (3 μmol) of caldarchaeyl-bis-amine in 140 μL of THF and 3 μL of DIPEA with 1.65 mg (9 μmol) of NBD-F at 20  C for 2 h. 2. Control the reaction by TLC with CHCl3/MeOH/H2O (65:25:4, v/v/v). The educt (Rf ¼ 0.6) should have disappeared and a new yellow product should be visible at the front. A further TLC with n-hexane/EtOAc (7:3, v/v) should show a main product (NBD-MSL) and sometimes a trace of side product with Rf of about 0.33 and 0.06, respectively.

Fluorescent Membrane-Spanning Lipids

319

3. Purify NBD-MSL by column chromatography on 10 g of LiChroprep Si 60 using n-hexane/EtOAc (1:1, v/v) as eluent. Check fractions by TLC with n-hexane/EtOAc (7:3, v/v) and collect pure NBD-MSL with Rf ¼ 0.33. The product is obtained as a yellow solid. Yield is 3.3 mg (67%) (see Note 9). 4. Analytical data. For NBD-MSL (a mixture containing one, two, and three cyclopentane rings in the lipid core) prominent ions [M + NH4]+ are found at m/z 1641.3, 1639.3, and 1637.3 for the bis-N-NBD derivative with one, two, and three cyclopentane rings, respectively. 3.2.5 Caldarchaeyl-BisN-TAMRA (TAMRA-MSL)

1. Dissolve about 4 mg (3 μmol) of caldarchaeyl-bis-amine in 600 μL of DCM-THF (1:1, v/v) and add 5 μL of DIPEA. 2. Treat this solution with 5.2 mg (10 μmol) of solid TAMRA-SE (5 and 6 isomers) at 30  C for 20 h. Check the reaction by TLC with CHCl3/MeOH/2 M NH3 (65:25:4, v/v/v) for the conversion of the bis-amine to TAMRA-MSL (Rf ¼ 0.67 and 0.70). 3. Dilute the reaction mixture with 4 mL of DCM and wash four times with 3 mL of H2O. Dry the organic solution over solid Na2SO4 and purify TAMRA-MSL by column chromatography on 10 g of LiChroprep Si 60 using CHCl3/MeOH/2 M NH3 (65:25:4, v/v/v) as eluent. Check fractions by TLC with CHCl3/MeOH/2 M NH3 (65:25:4, v/v/v) and collect pure TAMRA-MSL with Rf ¼ 0.67 and Rf ¼ 0.70 (see Note 10). The product is obtained as a dark red solid. Yield is about 3.8 mg (1.8 μmol, 60%). 4. Analytical data: Prominent ions [M + 2H]2+ are found at m/z 1061.8, 1060.8, and 1059.8 for TAMRA-MSL with one, two, and three cyclopentane rings, respectively. In addition respective ions [M + H] + are found at m/z 2122.5, 2120.5, and 2118.5.

3.3 Use of Fluorescent Membrane-Spanning Lipids for In Vitro Studies of Intermembrane Lipid Transfer and Membrane Fusion by FRET 3.3.1 Liposome Preparation for In Vitro Lipid Transfer Studies

1. Prepare large unilamellar donor liposomes in citrate buffer [21] with the following composition of lipids: 72 mol% DOPC, 5 mol% cholesterol, 20 mol% BMP, 1 mol% TAMRA-MSPE, and 2 mol% 2-NBD-(glyco)sphingolipid (see [5, 9, 11, 22]) or N-NBD-PE in a total of 200 nmol lipid per mL of buffer. For a reference of 100% dequenching, prepare liposomes without the quencher TAMRA-MSPE. Dry the appropriate amounts of lipid in a polyethylene reaction tube under an argon jet (see Note 11). 2. Then add 1 mL of citrate buffer and vortex to obtain a suspension of lipids in buffer. 3. Sonify the suspension in a sonifier bath for 15 min at 20  C and subject to eight freeze-thaw cycles to obtain a uniform distribution of buffer and solutes across the bilayers. This is best performed by placing the capped vial into liquid nitrogen prior to thawing in a metal block at 40  C.

320

Gu¨nter Schwarzmann

4. Then pass the suspension 21 times through polycarbonate filters with a pore size of 100 nm in a mini extruder. 5. Repeat this procedure with a new batch of lipids and use this one for measurements (see Note 12). 6. Prepare acceptor liposomes with the following lipid composition: 75 mol% DOPC, 5 mol% cholesterol, and 20 mol% BMP in a total of 1 μmol per mL of buffer but otherwise according to the steps described above. 3.3.2 Liposome Preparation for In Vitro Membrane Fusion Studies

3.3.3 In Vitro FRETBased Intermembrane Lipid Transfer Assays

1. Prepare large unilamellar donor liposomes with a total of 200 nmol of lipid per mL of citrate buffer with 73 mol% DOPC, 5 mol% cholesterol, 20 mol% BMP, 1 mol% NBD-MSPE (or NBD-MSL), and 1 mol% TAMRA-MSPE (or TAMRA-MSL) following the procedure as described above for the preparation of liposomes for transfer studies (see Notes 12 and 13). For transfer studies of sphingolipids their derivatives with a NBD group in position 2 of their acyl chain is used in the donor liposomes [5, 9, 11, 22]. 1. In a quartz cuvette with an optical path length of 5 mm, mix 40 μL of the respective donor liposomes with 40 μL of acceptor liposomes and 320 μL of citrate buffer. 2. Excite this mixture with UV light at 468 nm and measure fluorescence intensity at 522 nm. The fluorescence of the mixture containing donor liposomes without TAMRA-MSPE corresponds to 100% dequenching. 3. To the mixture containing donor liposomes with TAMRAMSPE add a small amount of the transfer protein under study at time zero (see Note 14). 4. Measure the time course of NBD fluorescence increase until a stable plateau is reached. 5. Evaluate results as outlined in [9]. Briefly, the NBD fluorescence of liposomes without quencher represents 100% dequenching. The NBD fluorescence of liposomes with quencher is taken as 0% dequenching. From the NBD fluorescence measured in the presence of a transfer protein at a defined time is deducted the value for 0% dequenching. The resulting value is multiplied by 100 and divided by the value for 100% dequenching. The result expresses the percentage of transfer of the NBD-lipid evoked by the transfer protein and should come close to 41.7%. This is usually the case when a plateau is reached, i.e., when the maximum of NBD-lipid of the outer membrane of donor liposomes is transferred to the outer membrane of acceptor liposomes. With unilamellar liposomes of

Fluorescent Membrane-Spanning Lipids

321

~100 nm in diameter, approximatively 5:6 of 50% of NBD-lipid should become unquenched in this assay. 3.3.4 In Vitro FRETBased Membrane Fusion Assays

1. In a quartz cuvette with an optical path length of 5 mm, mix 40 μL of donor liposomes containing the respective membrane-spanning NBD- and TAMRA-lipids with 40 μL of acceptor liposomes and 320 μL of citrate buffer. 2. Measure the unquenched NBD fluorescence in the mixture with donor liposomes. In addition, determine the NBD fluorescence in a mixture containing donor liposomes lacking TAMRA-MSPE. This fluorescence corresponds to 100% dequenching. 3. At time zero add the putative fusion protein (e.g., recombinant Sap C with a His6-tag [9, 23, 24]) and measure the increase of NBD-fluorescence intensity with time. Perform calculations as described in [9]. Briefly, from the measured value of NBD fluorescence, deduct the value of NBD fluorescence representing 0% dequenching. Multiply the resulting value by 100 and divide by the value of NBD fluorescence representing the 100% dequenching to obtain the percentage of NBD fluorescence unquenched by the fusion protein, which is a measure of reduction of the concentration of NBD-MSPE and TAMRAMSPE in membranes and thus of membrane fusion.

4

Notes 1. Under these culture conditions (see Subheading 3.1.1) caldarchaeols with one to three cyclopentane rings are predominant. 2. Fractions of caldarchaeols differing in their content of cyclopentane rings besides caldarchaeol with no cyclopentane ring or caldarchaeol with one open chain are obtained. Fractions with Rf ¼ 0.38 contain mostly caldarchaeols with zero and one cyclopentane ring besides little with 2 rings or one open chain. Fractions with Rf ¼ 0.36 contain mostly caldarchaeols with 2 cyclopentane rings. Fractions with Rf ¼ 0.31 to 0.26 contain caldarchaeols with mainly 3 and 4 cyclopentane rings. 3. Caldarchaeols and isocaldarchaeols with antiparallel and parallel arrangement, respectively, of their glycerol moieties but otherwise isobaric cannot be separated from each other under this chromatographic condition. Owing to the small mass difference of minus two mass units per additional cyclopentane ring a complete separation into single species of the tetraether lipids is difficult to achieve. For the purpose of synthesis of membrane-spanning lipids and their use in membrane lipid

322

Gu¨nter Schwarzmann

transport studies this is also not necessary. It is sufficient to use caldarchaeols with a narrow range of mass, e.g., those with mostly two five-membered rings. 4. The yield may be smaller or higher depending on the presence or absence of water in the reaction and on the formation of side products during mild acid hydrolysis of caldarchaeo-bisoxazaphospholane. 5. A predominant double band observed by TLC is likely due to the mixture of 5- and 6-isomers of the TAMRA residue rather than to differing contents of five-membered ring in the tetra ether lipid core. 6. The mass of ATTO-MSPE is taken from the zwitterionic form with a formula weight of 2798.1 for ATTO-MSPE with two five-membered rings in the lipid core. The respective ions [M + 3H]3+ at m/z 934.4, 933.7, and 933.1 for ATTOMSPE with one, two, and three cyclopentane rings are also prominent in the mass spectrum. Most pronounced are ions [M + 2H + NH4]3+ at m/z 940.0, 939.4, and 938.7 for the compounds with one, two, and three cyclopentane rings, respectively. 7. A side product is sometimes obtained in about 4% yield that according to mass spectrometry is a tetraether lipid with a tosyl group at one end and an azide group at the other end of the lipid core. The respective ions [M + NH4]+ are found at m/z 1496.3, 1494.3, and 1492.3 for this heterogenic derivative with one, two, and three cyclopentane rings in the lipid core. 8. For further synthesis it is advisable to prepare a fresh batch of caldarchaeyl-bis-amine by hydrogenation of the bis-azide since the latter is much more stable than the bis-amine on storage. 9. The side product (Rf ¼ 0.06) may also be isolated and is the mono-NBD derivative of caldarchaeyl-bis-amine. This product can be further reacted to NBD-MSL or can be used to synthesize a membrane-spanning lipid with a NBD group at one and another label at the other end of the lipid core. 10. On TLC TAMRA-MSL separates into a double band most likely due to the 5 and 6 isomers of the TAMRA residue. A further separation is not necessary for the use of this compound as a non-extractable and membrane-spanning energy quencher for an excited NBD fluorophore in bilayer membranes. 11. This low pH value is chosen because of the fact that in lysosomes a low pH prevails and that the interaction of sphingolipids, their degrading hydrolytic enzymes, and the respective saposines is optimal in a narrow realm around pH 4.3 [25, 26].

Fluorescent Membrane-Spanning Lipids

323

12. Some lipid, especially membrane-spanning lipids, tend to adhere to the polycarbonate filter, thus reducing their content in liposomal membranes. It is, therefore, recommended to prepare liposomes twice whereby the first extrusion process serves to saturate the polycarbonate filter and is discarded. The liposome preparation after the second extrusion process with the desired lipid composition contains mostly unilamellar besides a little bilamellar vesicles as revealed by less than 50% reduction of the NBD group by dithionite [27]. Alternatively, the extrusion process can be omitted and liposomes can be prepared by ultrasonication in a Branson sonifier cup horn at around 15  C for 6 min. Though not always yielding liposomes as uniform in size as those obtained after extrusion, the ultrasonication is significantly more material- and time-saving. In addition, transfer and fusion studies do not differ significantly in our hands when extruded or sonicated vesicles are used (personal communication). 13. For FRET-based fusion studies the energy donor-quencher pair TAMRA-MSPE and ATTO-MSPE can also be employed. 14. The pair NBD-MSPE and TAMRA-MSPE can be replaced by the pair NBD-MSL and TAMRA-MSL. Measurements should be performed at a constant temperature in the range of 20–30  C.

Acknowledgments The author’s work was supported by the Deutsche Forschungsgemeinschaft, grants Schw 143/5-4 and SFB 284/B5, which hereby is gratefully acknowledged. References 1. Babalola JO, Wendeler M, Breiden B et al (2007) Development of an assay for the intermembrane transfer of cholesterol by NiemannPick C2 protein. Biol Chem 388:617–626 2. Abdul-Hammed M, Breiden B, Adebayo MA et al (2010) Role of endosomal membrane lipids and NPC2 in cholesterol transfer and membrane fusion. J Lipid Res 51:1747–1760 3. Fo¨rster T (1949) Experimentelle und theoretische Untersuchung des zwischenmolekularen € Ubergangs von Elektronenanregungsenergie. Z Naturforsch 4a:321–327 4. Struck DK, Hoekstra D, Pagano RE (1981) Use of resonance energy transfer to monitor membrane fusion. Biochemistry 20:4093–4099

5. Schwarzmann G, Wendeler M, Sandhoff K (2005) Synthesis of novel NBD-GM1 and NBD-GM2 for the transfer activity of GM2-activator protein by a FRET-based assay system. Glycobiology 15:1302–1311 6. Wendeler M, Werth N, Maier T et al (2006) The enzyme-binding region of human GM2-activator protein. FEBS J 273:982–991 7. Ran Y, Fanucci GE (2009) Ligand extraction properties of the GM2 activator protein and its interactions with lipid vesicles. Biophys J 97:257–266 8. Langworthy TA (1978) Membranes and lipids of extremely thermoacidophilic microorganisms. In: Friedman SM (ed) Biochemistry of thermophily. Academic Press, Cambridge, pp 11–30

324

Gu¨nter Schwarzmann

9. Schwarzmann G, Breiden B, Sandhoff K (2015) Membrane-spanning lipids for an uncompromised monitoring of membrane fusion and intermembrane lipid transfer. J Lipid Res 56:1861–1879 10. Pangborn AB, Giardello MA, Grubbs RH et al (1996) Safe and convenient procedure for solvent purification. Organometallics 15:1518–1520 11. Schwarzmann G (2018) Synthesis of fluorescent gangliosides. In: Sonnino S (ed) Gangliosides. Methods Mol Biol, Springer, Berlin, p 323-356, Chapter 16 12. Uda I, Sugai A, Itoh YH et al (2001) Variation in molecular species of polar lipids from thermoplasma acidophilum depends on growth temperature. Lipids 36:103–105 13. Yasuda M, Oyaizu H, Yamagishi A et al (1995) Morphological variation of new Thermoplasma acidophilum isolates from Japanese hot springs. Appl Environ Microbiol 61:3482–3485 14. Touchstone JC, Levin SS, Dobbins MF et al (1981) Differentiation of saturated and unsaturated phospholipids on thin layer chromatograms. J High Resolut Chromatogr 4:423–424 15. Eibl H (1978) Phospholipid synthesis: oxazaphospholanes and dioxaphospholanes as intermediates. Proc Natl Acad Sci U S A 75:4074–4077 16. Gr€a ther O, Arigoni D (1995) Detection of regioisomeric macrocyclic tetraethers in the lipids of methanobacteriumthermoautotrophicum and other archaeal organisms. J Chem Soc Chem Comm:405–406 17. Heathcock CH, Finkelstein BL, Aoki T et al (1985) Stereostructure of the Archaebacterial C-40 Diol. Science 229:862–864

18. Montenegro E, Gabler B, Paradies G et al (2003) Determination of the configuration of an archaea membrane lipid containing cyclopentane rings by total synthesis. Angew Chem 42:2419–2421 19. Folch J, Lees M, Sloane Stanley GH (1957) A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem 226:497–509 20. Pedersen CJ (1967) Cyclic polyethers and their complexes with metal salts. J Am Chem Soc 89:7017–7036 21. MacDonald RC, MacDonald RI, Menco BP et al (1991) Small-volume extrusion apparatus for preparation of large, unilamellar vesicles. Biochim Biophys Acta 1061:297–303 22. Schwarzmann G, Arenz C, Sandhoff K (2014) Labeled chemical biology tools for investigating sphingolipid metabolism, trafficking and interaction with lipids and proteins. Biochim Biophys Acta 1841:1161–1173 23. Qi X, Chu Z (2004) Fusogenic domain and lysines in saposin C. Arch Biochem Biophys 424:210–218 24. Vaccaro AM, Tatti M, Ciaffoni F et al (1994) Saposin C induces pH-dependent destabilization and fusion of phosphatidylserinecontaining vesicles. FEBS Lett 349:181–186 25. Bierfreund U, Kolter T, Sandhoff K (2000) Sphingolipid hydrolases and activator proteins. Methods Enzymol 311:255–276 26. Sandhoff K, Harzer K, Wassle W et al (1971) Enzyme alterations and lipid storage in three variants of Tay-Sachs disease. J Neurochem 18:2469–2489 27. McIntyre JC, Sleight RG (1991) Fluorescence assay for phospholipid membrane asymmetry. Biochemistry 30:11819–11827

Chapter 22 Setting Up All-Atom Molecular Dynamics Simulations to Study the Interactions of Peripheral Membrane Proteins with Model Lipid Bilayers Viviana Monje-Galvan, Linnea Warburton, and Jeffery B. Klauda Abstract All-atom molecular dynamics (MD) simulations enable the study of biological systems at atomic detail, complement the understanding gained from experiment, and can also motivate experimental techniques to further examine a given biological process. This method is based on statistical mechanics; it predicts the trajectory of atoms over time by solving Newton’s Laws of motion taking into account all forces. Here, we describe the use of this methodology to study the interaction between peripheral membrane proteins and a lipid bilayer. Specifically, we provide step-by-step instructions to set up MD simulations to study the binding and interaction of the amphipathic helix of Osh4, a lipid transport protein, and Thanatin, an antimicrobial peptide (AMP), with model lipid bilayers using both fully detailed lipid tails and the highly mobile membrane-mimetic (HMMM) method to enhance conformational sampling. Key words Peripheral membrane proteins, Lipid bilayers, Amphipathic helices (AHs), Antimicrobial peptides (AMPs), All-atom molecular dynamics

1

Introduction Molecular dynamics (MD) simulations are used to understand the interactions and forces of a given system at the molecular level and their influence on macroscopic properties. This method facilitates the study of processes that take place over short timescales, predicts trends or mechanisms, and can be helpful in elucidating experimental hypothesis as well as in the selection of experimental targets. MD is a deterministic technique, based on statistical mechanics and empirical energy functions known as force fields (FFs). Bonded and nonbonded interactions within a system are modeled based on the forces acting on each atom. A FF is as complex as the molecules of study; complex models use a continuous energy potential to describe the forces acting on an atom every time it changes position. At the beginning of a simulation trajectory, initial velocities are assigned to every atom in a random manner.

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5_22, © Springer Science+Business Media, LLC, part of Springer Nature 2019

325

326

Viviana Monje-Galvan et al.

The trajectory progresses as the position and velocity of each atom in the system are predicted by solving Newton’s laws of motion numerically; the user can select among different numerical integrators. To avoid errors in the integration of these equations, the integration time step is usually set to 1–2 fs for biological systems using all atoms with constrained hydrogens—2 fs is roughly the timescale of the fastest molecule motion, bond vibration [1]. Proteins are essential in all biological organisms; they perform a variety of roles in a cell, from catalyzing reactions to transporting cargo, serving as mechanical support, or altering signaling pathways [2]. Membrane interactions of peripheral membrane proteins or peptides have gained interest in recent years [3, 4]. However, the transient nature of theses interactions makes them challenging to study only through experimental techniques. MD simulations are a useful technique to perform a close-up study of these molecules and gain understanding of specific forces and contacts that promote protein-membrane interaction [4]. Depending on the specific research question, one may choose all-atom or coarse-grain models to represent the molecules in a system. Furthermore, enhanced sampling simulation techniques such as umbrella sampling, replica exchange, or metadynamics [3, 5, 6] can be employed to facilitate conformational sampling of a system over large energy barriers. Current computational power and simulation techniques allow us to study protein-membrane and protein-protein interactions in terms of their energy landscape as well as the natural (unbiased) evolution of the system. Our particular research interests are on the study of amphipathic helices (AHs) and antimicrobial peptides (AMPs) using classical all-atom MD simulations. Both AHs and AMPs are ubiquitous in biology; understanding their interaction with various cell membranes serves to understand fundamental cell homeostasis and membrane remodeling processes, and could contribute to the development of novel drug therapies [7–9]. AHs are linked to membrane stress response, cell signaling, curvature, and lipid packing sensing, and are common domains in lipid transport proteins [10, 11]. AMPs have the ability to kill microorganisms and clear infections from viruses, fungi, and bacteria by either permeating their cell membrane or simply attaching to it [9]. However, none of these processes are well understood, and are excellent candidates for the use of MD to examine the specific steps, membrane properties, and explicit residues involved in protein-membrane interactions. Here we describe in detail how to set up and run all-atom MD simulations to study peripheral membrane proteins and their interaction with membranes. We consider two methods for these studies; namely, classical all-atom MD (AAMD) with fully detailed lipid tails in the membrane models to examine the natural timescale and dynamics of interaction, and the highly mobile membrane-mimetic

All-Atom MD Simulations for PMPs

327

Fig. 1 (a) The ALPS peptide of Osh4 (PDB ID: 1ZHZ), hydrophobic residues are shown in white and basic ones in blue. (b) Thanatin peptide (PDB ID: 8TFV), showing the hydrogen bonds between the two beta sheets in red. Each peptide is shown using the New-Cartoon representation and ResType (residue type) coloring method on VMD; polar residues are shown in green, nonpolar residues in white, basic residues in blue, and charged residues in red

(HMMM) model [12] to expedite the timescale of this interaction without sacrificing the atomic-level detail of the peptide-membrane interface. We also discuss routine analyses to determine if the simulation has reached thermal equilibrium. Finally, we included precise notes about potential challenges and key considerations during system preparation, simulation, and analysis to facilitate the usage of these techniques. Specific examples are given in the context of studying the amphipathic-lipid-packing-sensor (ALPS) motif of Osh4, a member of a seven-lipid-transport-protein family in yeast S. cerevisiae [4, 10], and Thanatin, an AMP with high activity against Gram positive and negative bacteria as well as fungi [13]—both shown in Fig. 1.

2

Materials To run a MD simulation one needs the starting coordinates of a system, the topology files for each molecule (i.e., the connectivity and angle lists), and the bonded and nonbonded parameters for the system (the force field, FF). Coordinates for resolved protein structures can be found in the Protein Data Bank (PDB, www.rcsb.org), and CHARMM-GUI Membrane Builder [14–17] can be used to generate membrane systems as discussed in the next section (see Note 1 for advice on protein models from the PDB). The CHARMM36 FF for lipids [18–21] and CHARMM36m for proteins [22] are the most accurate to reproduce experimental observables; however, the user may choose other common FFs such as

328

Viviana Monje-Galvan et al.

GROMOS [23], OPLS [24], or AMBER [25] depending on the sets of properties to be reproduced or the simulation software. AAMD and HMMM are both run with explicit solvent; a typical water model used for biological simulations is a three-site model, TIP3P [26]. Common simulation packages for these systems are CHARMM [27], NAMD [28], GROMACS [29], and AMBER [30]. In this work we describe the steps to set up and run a simulation on NAMD, and use CHARMM for analysis or to manipulate the coordinates of the system in the case of the AAMD simulations. CHARMM-GUI also has the ability to faithfully develop inputs for running in a wide array of simulations packages (beyond those mentioned above) if the user so desires [15]. One of the most common visualization software packages is Visual Molecular Dynamics (VMD) [31]. The user can choose from several display and coloring styles to render or edit a given system; all the protein and membrane snapshots included in this publication where generated using this software, which can additionally be used to run post-process analysis from the simulation trajectories. Additionally, Chimera [32] was used to edit some structures along with Packmol [33] as detailed in the next section.

3

Methods

3.1 System Preparation

To prepare the starting configuration for both the AAMD and HMMM simulations we use CHARMM-GUI Membrane Builder [14–17] (www.charmm-gui.org), a user-friendly interface to build membrane and membrane-protein systems, among other options. A detailed description on the use of this software as it pertains to the building of protein-membrane or membrane-only systems to run AAMD is found in ref. [34]. Similarly, the capabilities of CHARMM-GUI for the preparation of an HMMM system are found in ref. [16]. Key advice to build MD simulation systems is listed in Notes 1–6 and steps to build these systems discussed below.

3.1.1 AAMD Systems

In this case, the membrane needs to be built and equilibrated separately before introducing the peptide in the solvent phase, or at another preferred location, to allow lipids to relax into a more energetically favored conformation. To this end, we follow the membrane building steps as presented in [34] and other membrane-only studies we have performed in the past [35–37]. In short: 1. In the CHARMM-GUI Membrane Builder, input the desired lipid species and their respective amounts per leaflet (see Notes 7 and 8). Set the amount of water molecules per lipid (typically

All-Atom MD Simulations for PMPs

329

40–45 for a fully hydrated system). Select the temperature in Kelvin, and choose between adding neutralizing ions or a salt at preferred concentration (to match experimental conditions) to render the system neutral—a requirement in long-range electrostatic methods to converge. 2. One may even select different number of lipids for the top and bottom leaflets of the membrane, given asymmetry is an important feature of some cell membranes [38, 39]. However, doing so introduces complexity to the simulation both in terms of determining when equilibrium is reached (simulation length) and the characterization of binding (see Note 9). In this publication, we only describe the preparation of systems with symmetric bilayers. 3. To generate the membrane-protein systems, introduce the peptide into the solvent of an equilibrated membrane system using CHARMM and remove overlapping waters. The peptide can be translated and rotated to the desired orientation using CHARMM at the time when coordinates are merged; the user may choose other programs like VMD to manipulate and merge coordinate files (see Notes 1 and 5). The ALPS peptide was positioned vertically and horizontally above the bilayer at least ˚ above the bilayer with enough water to prevent its interac8A tion with the image atoms of the membrane, as indicated in reference [11] to make sure that the final bound conformation was not biased by the initial orientation of the peptide. 4. It is a good practice to use CHARMM-GUI Quick Solvator to ensure the coordinate file for the peptide/protein has the correct termini. This tool runs a quick minimization and equilibration of the protein in water, which can then be merged into the membrane-only system coordinate file. For example, for the ALPS peptide of Osh4, the coordinates were obtained reading the first 29 amino acids of PDB ID: 1ZHZ that contained the coordinates for the full protein. The Quick Solvator allows the user to select the first and last residues of the final peptide such that they are an N- and a C-terminus, respectively (or other termini, even adding a lipidated residue at the end). This tool also allows the user to mutate protein residues or protonate them as desired or needed due to the pKa. 5. If the solvent region in the main simulation box of the membrane-only system is too narrow to introduce the peptide, or a larger protein, add water merging the coordinate files of the membrane-water system with a file that contains a box of similar dimensions but packed with water molecules only (see Fig. 2 and Note 3). We use Packmol [33] to build water-only boxes when needed. This software only needs the coordinate file in pdb format of one molecule of the solvent of choice,

330

Viviana Monje-Galvan et al.

Fig. 2 (a) Membrane-ALPS system viewed on VMD to determine whether additional water is needed to run the simulation. (b) Added water box to the membrane-protein system (top) or to the membrane-only system (bottom). DOPS lipids are shown in red and DOPC lipids in cyan. The peptide is shown using the same representation as specified in Fig. 1, and the lipids are displayed using the Licorice representation (omitting the hydrogen atoms); the oxygen atom of water is shown explicitly using the van der Waals representation

˚ , and the number of TIP3P in this case, the box size in A molecules to be packed in the box. One must simply be careful to estimate enough water molecules to result in a system in the liquid state (density of 1 g/cc). 6. Merging the membrane and water coordinate files is readily done on CHARMM, or another software of choice. As shown in Fig. 2b, one can either: (a) Add the water box to the membrane-protein system, remove overlapping waters, and then run a short minimization to relax the system before proceeding to run the simulation trajectory, or (b) Combine the membrane-only and water-only coordinate files, and run a short minimization to allow for all solvent molecules to relax and “merge” before adding the peptide and removing overlapping water molecules. 3.1.2 HMMM Systems

In this case, the system is prepared with an organic solvent in between the all-atom representations of the lipid head-groups of choice with truncated tails to accelerate the lateral diffusion of lipids in the membrane model [12]. By doing so, the conformational sampling at the membrane interface is enhanced. To prepare these type of simulations: 1. Use the CHARMM-GUI HMMM Builder [16] to load the PDB structure to be used in the simulation. The PDB structure 8TFV was used for Thanatin, a 21-amino acid peptide. The

All-Atom MD Simulations for PMPs

331

Fig. 3 Thanatin at the beginning (a) and the end of a simulation (b). The peptide is displayed using the same settings as described in Fig. 1 and the lipids of the binding leaflet are shown using the Licorice representation for all their atoms (including hydrogens)

user can choose between the original orientation of the peptide from the PDB file or adjust it. For example, align the principal axis of the peptide to the z-axis, or rotate the molecule with respect to the x- or y-axes and then translate it along the z-axis. The peptide should be placed at least 10–12 A˚ above the membrane surface (including hydrogen atoms, not heavyatoms only) to prevent a forced interaction with the membrane (Fig. 3a). 2. If a more complicated peptide orientation is needed such as at a specific angle, or a specific layout of several peptides is desired, the PDB file can be edited in Chimera (as discussed below in steps 3 and 4), CHARMM, or VMD before uploading it to CHARMM-GUI to be positioned above the bilayer. 3. To manipulate the PDB file in Chimera, load the PDB file and use the structure measurement tool to define a center of mass. Using the reply log, identify the current center of mass of the protein, and then translate it to the new center of mass. Save the coordinates to a new PDB file when all repositioning is done. 4. To position multiple peptides near the membrane surface, create a membrane-only system on CHARMM-GUI as explained in the previous section and use that coordinate file as a template. It does not need to be the final membrane model or have the actual lipid species, only the approximate dimensions of your final system. Position the peptides as desired on Chimera and then save only the peptides coordinates to a new PDB file. Load the resulting PDB to CHARMM-GUI when building the protein-membrane system as in Subheading 3.1.1. 5. The next step to generate an HMMM membrane is critical as it determines the size of the system. The lipid scaling factor and terminal acyl carbon number will also determine the lipid mobility in the HMMM system. 6. Lipid scaling factor: it controls the average area per lipid of the membrane, which is equivalent to imposing a surface tension

332

Viviana Monje-Galvan et al.

on the membrane. Increasing this parameter increases lipid mobility. A value between the default of 1.1 and 1.3 is commonly used [16, 40] (see Note 10). The closer this parameter is to 1, the closer the surface area per lipid resembles that of an AAMD simulation and lipid diffusion will be reduced as well as a reduction in enhanced sampling. 7. Terminal acyl carbon number: HMMM systems have lipids with shorter acyl-tail lengths and a nonpolar solvent as membrane core [16]. The shorter the tail lengths, the more exposed the hydrophobic core and the more prone the lipids to lateral diffusion. The default setting in CHARMM-GUI is to use an acyl carbon length of 6, but one may choose 4 or 8. 8. Finally, similar to the AAMD systems, one selects the FF for the simulation. New systems should use the CHARMM36m FF, but an earlier version may be selected for consistency if other simulations in the study were run with previous FF parameters. 3.2 Simulation Settings

Typical AAMD simulation settings for a membrane-only system built and downloaded from CHARMM-GUI are: 1. 2 fs time-step to integrate the simulation trajectory using NPT dynamics (constant number of molecules, pressure, and temperature, see Note 11). 2. The SHAKE algorithm to constraint hydrogen atoms [41]. 3. Nose´-Hoover thermostat [42] and Langevin dynamics [1, 43] to keep the temperature constant; our simulations run at 303.15
K to ensure a fluid phase membrane (see Note 12 for advice on simulation temperature settings). 4. Semi-isotropic pressure control using a Langevin piston [28, 44, 45] with constant pressure of 1 bar, i.e., the cell dimensions vary with X ¼ Y and separately in the Z-dimension. van der Waals (VDW) and electrostatics computed using a Lennard-Jones force-switching function over ˚ [46]. 10–12 A 5. Periodic boundary conditions (PBC) to simulate an infinite membrane slab in the membrane plane. 6. Particle Mesh Ewald (PME) method to evaluate long-range electrostatic interactions [47]. 7. Standard CHARMM-GUI six-step protocol over 225 ps for the initial equilibration of the membrane-only systems [34, 48]. These parameters are maintained after adding the peptide to the solvent phase; except for the six-step membrane equilibration protocol, the same parameters are used in an HMMM system.

All-Atom MD Simulations for PMPs

333

Fig. 4 (a) SA of a binary lipid membrane used in an AAMD run to determine equilibrium prior to peptide insertion (first 50 ns of the membrane-only trajectory) (b) Distance time series for Thanatin-membrane interactions in an HMMM run 3.3 Reaching Equilibrium

1. A typical membrane property to determine if a membrane-only system has reached thermal equilibrium is the surface area ˚ 2/lipid. To compute this value, obtain the (SA) per lipid in A time series for the X and Y dimensions of the main simulation box during the simulation, multiply the values (X∙Y), and divide over the total number of lipids. Fig. 4a shows a time series of the SA for one of the AAMD membranes equilibrated before introducing the ALPS peptide. 2. To calculate the SA of the membrane after peptide binding would require the use of other techniques, such as Voronoi triangulation [49], to account for the volume and area of the peptide on the membrane surface and deduct it from the overall SA before obtaining the SA per lipid. Another option is to use GridMAT-MD [50], a software that calculates the SA of a system with or without an embedded protein, and also provides estimates of the bilayer thickness one can plot as a 2D projection of the bilayer in the X-Y plane (the membrane surface). 3. When examining the peptide-membrane interactions, stable binding is determined by examining the time series of different quantities to see when, and if, they reach steady state. For example, the distance between the centers of mass of the peptide and the bilayer core, the tilt angle between the main axis of the protein and the membrane surface, or the interaction energy between peptide and membrane. A sample time series of the distance between Thanatin and the bilayer from an HMMM run is shown in Fig. 4b. 4. Final measurements of peptide orientation, bilayer orientation, hydrogen bonding, and strength of interaction should be blocked averaged for the equilibrated portion of the trajectory as determined from the time series curves (see Notes 13 and 14). Blocking will enable more robust statistical analysis, and allow for noise reduction in the data set and average computation of the quantity of interest. For example, the steady state

334

Viviana Monje-Galvan et al.

Table 1 System size comparison between AAMD and HMMM simulations System ALPS ALPS

#Residues Method 29 29

Thanatin 21

AAMD

#Atoms

80

11,210

64

2000

b

80

15,681

64

200

65,919 80

15,979

28

150

54,922

HMMM 66,103 HMMM

Lipids/leaflet #Waters #Ions Trajectory length (ns)

a

a

The number of atoms for this system is for the binary lipid mixture used in study [11] b This system is for a binary lipid mixture used in reference [40]

(final) tilt angle of Thanatin with respect to the bilayer and the stable angle between the beta sheets of the peptide should be computed over the last 125 ns of equilibrated trajectory. The average tilt angle should be the average of blocked angle averages (typically) every 10–20 ns. 5. The AAMD simulations require nearly an order of magnitude longer simulation time compared to HMMM. In our studies, ALPS-membrane interactions were stable after approximately 1 μs of simulation [11], and trajectories were 2 μs long per replicate, attainable on the Anton Machine [51], whereas HMMM runs of this peptide required at most 200 ns of simulation [40]. Comparably, HMMM simulations of Thanatin required only trajectories of 150 ns. Table 1 summarizes system size, peptide length, and trajectory length for these simulations as reference (see Notes 15–17).

4

Notes 1. Check the models and chains available for your protein on the coordinate file from the PDB using VMD to select the appropriate/desired one(s). This selection will later be entered on CHARMM-GUI after loading the PDB structure. 2. In both cases, AAMD and HMMM, it is very important to select the number of lipid per leaflet wisely such that there is enough membrane surface area for the peptide/protein to survey the bilayer without interacting with itself due to PBC. It is recommended to have a protein coverage of at most 40% for surface only-binding proteins/peptides. Additionally, and in particular for multicomponent lipid membranes, finite size effects must be taken into consideration to ensure the membrane patch represented in the main simulation box is relevant for the biological system [52]. 3. There must be enough solvent to prevent peptide/protein interaction with both leaflets at the same time due to the use

All-Atom MD Simulations for PMPs

335

˚ of water of PBC. A good rule of thumb is to have at least 10 A in-between the protein and each leaflet, using the longest axis of the protein for this measurement. For example, using VMD, for a folded protein that measures from end-to-end 25 A˚, the solvent region in the main simulation box should be at least ˚ thick in the z-axis. Solvent interactions do slow down the 45 A simulation, but failing to account for enough solvent may result in a protein interacting with one leaflet and the image atoms of the opposite leaflet at the same time. 4. When building an HMMM system, the “Protein Surface Penetration” error message indicates that lipid tails are overlapping with the protein surface. It may be necessary to change the protein orientation or place it farther away from the membrane. HMMM should be used to sampling binding to the membrane and not imposing a bound state. 5. Always visualize your built system before starting the simulation to ensure the major components are placed correctly. For example, the peptide is placed above the bilayer and with enough solvent to prevent it from interacting with both leaflets of the bilayer at the same time when using PBC. Similarly, the membrane is large enough to prevent the interaction of the peptide/protein atoms with their images when running the simulation. 6. In both AAMD and HMMM, one can choose the number of initial peptide orientations to be examined. Computational resources will determine to large extend the number of trajectories that can be run. Ideally, at least three replicate trajectories should be run per system given the stochastic nature of binding events and to perform statistical analysis on the results. 7. Membrane properties are highly influenced by lipid composition [35], which also determines what molecules interact with them [4, 53]. Choose the lipid composition of your membrane model according to the system of study; i.e., if a protein is known to interact with charged lipids, include a charged lipid species in your model. For example, to simulate Thanatin, a peptide that interacts with Gram positive and Gram negative bacteria, a reasonable choice for lipids is 1-palmitoyl-2-oleoylsn-glycero-3-phosphoethanolamine (POPE) and 1-palmitoyl2-oleoyl-sn-glycero-3-phosphoglycerol (POPG) lipids in a 4:1 ratio [54]. 8. There is a trade-off between lipid diversity, system size, and simulation length. The larger the membrane, the longer the simulation will take—mainly because there is much more water needed to maintain full hydration conditions during the simulation and there are many more calculations during the integration of the trajectory equations.

336

Viviana Monje-Galvan et al.

9. Common lipid species to mimic a membrane environment are 1,2-dioleoyl-sn-glycero-3-phospatidylserine (DOPS, negatively charged) or 1-palmitoyl-2-oleoyl-sn-glycero-2-phosphocoline (POPC, neutral), but to study protein-membrane interactions at the membrane interface as in the case of peripheral membrane proteins, one must select lipid species characteristic of the cell membrane that is being modeled. For the ALPS studies, three membrane models were used: a binary lipid mixture of 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC)/DOPS in a 3:2 ratio to model a charged bilayer, a 9-lipid mixture modeling the endoplasmic reticulum of yeast S. cerevisiae, and a 10-lipid mixture modeling the trans-Golgi Network of yeast [35]. 10. When running HMMM simulations decreasing the lipid packing with a lipid scaling factor greater than 1.0 will increase packing defects and short-chain lipid diffusion. Shorter chains will also increase the exposure of the hydrophobic interior. These can be adjusted to provide increased sampling of binding events. 11. One may choose to perform simulations using different thermodynamic ensembles when running AAMD. The constant number of molecules, pressure, and temperature (NPT) ensemble is by far the most common for simulations of biological systems. Variations such as NPAT or NPγT, with an added constraint to keep the lateral area or incorporate surface tension, respectively, can also be selected depending on the properties to be measured or replicated during the simulation [55]. For example, the NPγT ensemble was used in the AAMD simulations of ALPS to increase the lipid packing defects on the membrane surface to accelerate peptide binding [11] and because ALPS is a lipid packing sensor [56], a result that is achieved to a greater degree using the HMMM method. 12. The default temperature in CHARMM-GUI HMMM builder is 303.5
K, which can be changed to accelerate lipid diffusion. However, this should be done with caution to prevent undesired phase transitions in the membrane since the temperature range in which membrane lipids are stable is small [16]. 13. The simulation time for membrane-only systems is determined to allow for at least 50 ns of equilibrated trajectory. In the systems used to study ALPS binding mechanism, 100 ns were enough for the DOPS-DOPC membrane model, but at least 200 ns were required for complex 9-lipid-species models for the yeast membrane models as determined by the computed SA of each membrane-only trajectory (see Fig. 4a). 14. Triplicate runs should be run for each system to perform relevant statistical analysis and avoid bias to initial lipid and

All-Atom MD Simulations for PMPs

337

protein coordinates. Initial lipid configuration on the membrane should ideally be different in each run to ensure each replicate is independent from the other (i.e., build each membrane model separately, even if the lipid composition will remain constant). Initial protein coordinates can vary in orientation or distance to the membrane surface as determined by the user (see also Note 6). 15. A hybrid approach may be desired, where HMMM is used to sample binding conformations and then AAMD are used to run for at least 200 ns to examine specific protein interactions with the full-length lipid tails and a more natural evolution of the system. The conversion from HMMM to AAMD can be done directly on CHARMM-GUI. 16. MD simulation speed depends on the resources available for research. CPU-only resources provide an option to increase speed (nanoseconds per day) if there is access to a highperformance computing cluster with multiple nodes preferable connected with high-speed communication. Benchmarking should be performed (testing number of compute nodes vs. speed in ns/day) to determine the optimal number of nodes. 17. Recent advances in Graphical Processing Units (GPUs) have allowed improved speedup of MD simulations. Most simulation codes have methods to compile or running binaries to use with GPUs, including NAMD. Depending on the system size and dimensions, running on a single node with two or more GPUs can be equivalent to running on 10 CPU-only nodes. This has allowed simulations with AAMD to obtain μs long simulations in a reasonable time. As with CPU-only, benchmarks should be done to determine optimal number of nodes for each system setup.

Acknowledgments The AAMD and HMMM method discussed here were used to study the binding mechanism of ALPS to model membranes, results are presented in [11, 40], respectively; the Thanatin system is currently under study. These simulation trajectories were partially supported by NSF grant DBI-1145652 and MCB-1149187 and the High Performance Deepthought & Deepthought 2 Computing Clusters at the University of Maryland, College Park administered by the Division of Information Technology. The AAMD runs were possible thanks to time on the Anton Computer provided by the Pittsburgh Supercomputing Center (PSC) through Grant R01GM116961 from the National Institutes of Health and our specific time associated with the grant PSCA14030P. The Anton machine at PSC was generously made available by D.E. Shaw Research.

338

Viviana Monje-Galvan et al.

References 1. Leach AR (ed) (2001) Molecular modeling principles and applications, 2nd edn. Great Britain, Pearson Education 2. Stryer L (ed) (1989) Molecular design of life. W.H. Freeman and Company, New York, USA 3. Kalli AC, Sansom MSP (2014) Interactions of peripheral proteins with model membranes as viewed by molecular dynamics simulations. Biochem Soc Trans 42:1418–1424 4. Monje-Galvan V, Klauda JB (2016) Peripheral membrane proteins: Tying the knot between experiment and computation. Biochim Biophys Acta 1858:1584–1593 5. Mori T, Miyashita N, Im W et al (2016) Molecular dynamics simulations of biological membranes and membrane proteins using enhanced conformational sampling algorithms. Biochim Biophys Acta 1858:1635–1651 6. Barducci A, Bonomi M, Parrinello M (2016) Metadynamics. Wiley Interdiscip Rev Comput Mol Sci 1:826–843 7. Doucet CM, Esmery N, de Saint-Jean M et al (2015) Membrane curvature sensing by amphipathic helices is modulated by the surrounding protein backbone. PLoS One 10: e0137965 8. Cui H, Mim C, Vazquez FX et al (2013) Understanding the role of amphipathic helices in N-BAR domain driven membrane remodeling. Biophys J 104:404–411 9. Sinha S, Zheng L, Mu Y et al (2017) Structure and interactions of a host defense antimicrobial peptide thanatin in lipopolysaccharide micelles reveal mechanism of bacterial cell agglutination. Nat Sci Rep 7:17795 10. Drin G, Casella J-F, Gautier R et al (2007) A general amphipathic alpha-helical motif for sensing membrane curvature. Nat Struct Mol Biol 14:138–146 11. Monje-Galvan V, Klauda JB (2018) Preferred binding mechanism of Osh4’s amphipathic lipid-packing sensor motif, insights from molecular dynamics. J Phys Chem B 122:9713–9723 12. Ohkubo YZ, Pogorelov TV, Arcario MJ et al (2012) Accelerating membrane insertion of peripheral proteins with a novel membrane mimetic model. Biophys J 102:2130–2139 13. Mandard N, Sodano P, Labbe H et al (1998) Solution structure of thanatin, a potent bactericidal and fungicidal insect peptide, determined from protein two-dimensional nuclear magnetic resonance data. Eur J Biochem 256:404–410

14. Jo S, Kim T, Iyer VG et al (2008) CHARMMGUI: a web-based graphical user interface for CHARMM. J Comput Chem 29:1859–1865 15. Lee J, Cheng X, Swails JM et al (2016) CHARMM-GUI input generator for NAMD, GROMACS, AMBER, OpenMM, and CHARMM/OpenMM simulations using the CHARMM36 additive force field. J Chem Theory Comput 12:405–413 16. Qi YF, Cheng X, Lee J et al (2015) CHARMM-GUI HMMM builder for membrane simulations with the highly mobile membrane-mimetic model. Biophys J 109:2012–2022 17. Wu EL, Cheng X, Jo S et al (2014) CHARMM-GUI membrane builder toward realistic biological membrane simulations. J Comput Chem 35:1997–2004 18. Klauda JB, Venable RM, Freites JA et al (2010) Update of the CHARMM all-atom additive force field for lipids: validation on six lipid types. J Phys Chem B 114:7830–7843 19. Lim JB, Rogaski B, Klauda JB (2012) Update of the cholesterol force field parameters in CHARMM. J Phys Chem B 116:203–210 20. Venable RM, Sodt AJ, Rogaski B et al (2014) CHARMM all-atom additive force field for sphingomyelin: elucidation of hydrogen bonding and of positive curvature. Biophys J 107:134–145 21. Li Z, Venable RM, Rogers LA et al (2009) Molecular dynamics simulations of PIP2 and PIP3 in lipid bilayers: determination of ring orientation, and the effects of surface roughness on a Poisson-Boltzmann description. Biophys J 97:155–163 22. Huang J, Rauscher S, Nawrocki G et al (2016) CHARMM36m: an improved force field for folded and intrinsically disordered proteins. Nat Methods 14:71–73 23. Reif MM, Winger M, Oostenbrink C (2013) Testing of the GROMOS force-field parameter set 54A8: structural properties of electrolyte solutions, lipid bilayers, and proteins. J Chem Theory Comput 9:1247–1264 24. Harder E, Damm W, Maple J et al (2016) OPLS3: a force field providing broad coverage of drug-like small molecules and proteins. J Chem Theory Comput 12:281–296 ˚ A et al (2014) 25. Dickson CJ, Madej BD, Skjevik A Lipid14: the amber lipid force field. J Chem Theory Comput 10:865–879 26. Jorgensen WL, Chandrasekhar J, Madura JD et al (1983) Comparison of simple potential functions for simulating liquid water. J Chem Phys 79:926–935

All-Atom MD Simulations for PMPs 27. Brooks BR, Brooks CL, Mackerell JAD et al (2009) CHARMM: the biomolecular simulation program. J Comput Chem 30:1545–1614 28. Phillips JC, Braun R, Wang W et al (2005) Scalable molecular dynamics with NAMD. J Comput Chem 26:1781–1802 29. Pronk S, Pall S, Schulz R et al (2013) GROMACS 4.5: a high-throughput and highly parallel open source molecular simulation toolkit. Bioinformatics 29:845–854 30. Case DA, Betz RM, Botello-Smith W et al (2016) AMBER16. University of California, San Francisco 31. Humphrey W, Dalke A, Schulten K (1996) VMD: Visual molecular dynamics. J Mol Graph 14:33–38 32. Pettersen EF, Goddard TD, Huang CC et al (2004) Chimera - a visualization system for exploratory research and analysis. J Comput Chem 25:1605–1612 33. Martinez L, Andrade R, Birgin EG et al (2009) PACKMOL: a package for building initial configurations for molecular dynamics simulations. J Comput Chem 30:2157–2164 34. Jo S, Lim JB, Klauda JB et al (2009) CHARMM-GUI Membrane Builder for mixed bilayers and its application to yeast membranes. Biophys J 97:50–58 35. Monje-Galvan V, Klauda JB (2015) Modelling yeast organelle membranes and how lipid diversity influences bilayer properties. Biochemistry 54:6852–6861 36. Boughter CT, Monje-Galvan V, Im W et al (2016) Influence of cholesterol on phospholipid bilayer structure and dynamics. J Phys Chem B 120:11761–11772 37. Khakbaz P, Monje-Galvan V, Zhuang X et al (2017) Modeling lipid membranes. In: Geiger O (ed) Biogenesis of fatty acids, lipids and membranes. Springer International Publishing, Cham, pp 1–19 38. Marquardt D, Geier B, Pabst G (2015) Asymmetric lipid membranes: towards more realistic model systems. Membranes 5:180–196 39. Ingo´lfsson HI, Melo MN, van Eerden FJ et al (2014) Lipid organization of the plasma membrane. J Am Chem Soc 136:14554–14559 40. Wildermuth KD, Monje-Galvan V, Warburton LM et al (2018) Effect of membrane lipid packing on stable binding of the ALPS peptide. J Chem Theory Comput. Accepted 41. Ryckaert J-P, Ciccotti G, Berendsen HJC (1977) Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes. J Comput Phys 23:327–341

339

42. Hoover WG (1985) Canonical dynamics: equilibrium phase-space distributions. Phys Rev A 31:1695–1697 43. Frenkel D, Smit B (2001) Understanding molecular simulation from algorithms to applications, 2nd edn. Academic Press, Cambridge 44. Feller SE, Zhang Y, Pastor RW et al (1995) Constant pressure molecular dynamics simulation: the Langevin piston method. J Chem Phys 103:4613–4621 45. Martyna GJ, Tobias DJ, Klein ML (1994) Constant pressure molecular dynamics algorithms. J Chem Phys 101:4177–4189 46. Steinbach PJ, Brooks BR (1994) New spherical-cutoff methods for long-range forces in macromolecular simulation. J Comput Chem 15:667–683 47. Darden T, York D, Pedersen L (1993) Particle mesh Ewald: an N·log(N) method for Ewald sums in large systems. J Chem Phys 98:10089–10092 48. Jo S, Kim T, Im W (2007) Automated builder and database of protein/membrane complexes for molecular dynamics simulations. PLoS One 2:e880 49. Barber CB, Dobkin DP, Huhdanpaa H (1996) The quickhull algorithm for convex hulls. ACM Trans Math Softw 22:469–483 50. Allen WJ, Lemkul JA, Bevan DR (2009) GridMAT-MD: a grid-based membrane analysis tool for use with molecular dynamics. J Comput Chem 30:1952–1958 51. Shaw DE, Deneroff MM, Dror RO et al (2008) Anton, a special-purpose machine for molecular dynamics simulation. Commun ACM 51:91–97 52. Castro-Roma´n F, Benz RW, White SH et al (2006) Investigation of finite system-size effects in molecular dynamics simulations of lipid bilayers. J Phys Chem B 110:24157–24161 53. Vanni S, Hirose H, Barelli H et al (2014) A sub-nanometre view of how membrane curvature and composition modulate lipid packing and protein recruitment. Nat Commun 5:4916 54. Robert E, Lefe`vre T, Fillion M et al (2015) Mimicking and understanding the agglutination effect ofthe antimicrobial peptide thanatin using model phospholipid vesicles. Biochemistry 54:3932–3941 55. Ikeguchi M (2004) Partial rigid-body dynamics in NPT, NPAT, and NPγT ensembles for proteins and membranes. J Comput Chem 25:529–541 56. Vanni S, Vamparys L, Gautier R et al (2013) Amphipathic lipid packing sensor motifs: probing bilayer defects with hydrophobic residues. Biophys J 104:575–584

INDEX A Acyl-acyl carrier protein (ACP) synthetase/2acylglycerolphosphoethanolamine acyltransferase (Aas) ..................................................166–169, 178 Acyl-CoA-sterol acyltransferase (ACAT) ............ 119, 126 Affinity chromatography anti-FLAG M2 affinity resin ................................... 264 Bio-Beads SM-2 polystyrene .................................. 186 CaptoQ HiTrap columns........................................ 217 Glutathione Sepharose 4B beads .................. 274, 279 His60 Ni Superflow Resin ...................................... 205 HisPur Cobalt ................................................ 216, 220 HisPur Nickel .......................................................... 222 HisTrap column ...................................................... 223 nickel-nitrilotriacetic acid (Ni-NTA) agarose ............................................................... 262 All-atom molecular dynamics .............................. 325–337 Amphipathic helices (AHs)........................................... 326 Amphitropic proteins .................................................... 237 Anthrolysin O....................................................... 154, 155 Antibody ..................................................... 49, 50, 52, 54, 55, 108, 113, 121, 128, 140, 150, 157, 161, 162 Antimicrobial peptides (AMPs)........................... 326, 327 Arabidopsis thaliana ..................................................70, 71 Autoradiography ............................................................. 62

B Binding affinity....................................168, 258, 293–305 Bioluminescence resonance energy transfer (BRET)...... v, 13–21, 23–32 Blue native polyacrylamide gel electrophoresis (BN-PAGE) ............................219, 221, 226–228, 231, 233

C Calcium (Ca2+).............................................................. 213 Caldarchaeols................................................308, 310–321 C1α domain........................................................ 50, 52, 54 C2 domain.............................................................. 36, 201 Cell lines CHO .......................................................................... 27 CHO-K1................................................ 156, 160, 161 COS-7........................................................................ 27

Expi293F ........................................................ 205, 207 HEK293 .................................................................... 15 HEK-AT1 .................................................................. 15 HEK293T.................................................................. 27 HeLa ......................................................2–5, 7–10, 27, 50–52, 54, 99, 100, 110, 143, 146, 147 Neuro 2A................................................................... 27 SV-589................................................... 156, 161, 162 Cell plate.................................................. 98, 99, 102, 142 Cellular signaling.................................................. 213, 326 Chromatography ........................................ 23, 60, 70, 75, 96–99, 102, 106, 111, 119, 124, 125, 140, 141, 158–160, 168, 186, 191, 205, 206, 208, 210, 217, 218, 220–223, 244, 248, 259, 262–264, 274, 276, 279, 280, 296, 309, 310, 312, 314–316, 318, 319 Circular dichroism (CD) ..................................... 293–304 Clear native polyacrylamide gel (CN-PAGE)........ 70, 72, 74, 77, 79–83, 85, 227 Click mix.................................................. 97–99, 101, 102 Click reaction ..........................................................99–101 Coelenterazine h ............................ 16, 18, 20, 26, 27, 29 Confocal microscopy ................................. 3, 4, 9, 14, 24, 50–55, 138, 141, 142, 145–147 Crosslinking...................................... 96, 98, 99, 101, 102 Culture bacteria................................................... 157, 162, 222 cells...................................................... 4, 5, 27, 64, 71, 102, 139, 170, 171, 175, 176, 262, 310 yeast ............................................................42, 64, 123

D D4 fragment, see Perfringolysin O Detergent digitonin .................................................................. 259 lauryl maltose neopentyl glycol (LMNG) ................................................... 190, 259 n-dodecyl-ß-D-maltopyranoside .......... 185, 190, 191 n-dodecyl-ß-D-maltoside (DDM) ..........73, 171, 206 octaethylene glycol monododecyl ether (C12E8)............................................................. 259 Triton X-100 .................................121, 130, 183, 249 Detergent-resistant membranes (DRMs) ...................116, 121, 127, 130

Guillaume Drin (ed.), Intracellular Lipid Transport: Methods and Protocols, Methods in Molecular Biology, vol. 1949, https://doi.org/10.1007/978-1-4939-9136-5, © Springer Science+Business Media, LLC, part of Springer Nature 2019

341

INTRACELLULAR LIPID TRANSPORT: METHODS

342 Index

AND

PROTOCOLS

E Endoplasmic reticulum (ER)........................1–10, 14, 15, 36–38, 43–45, 48–50, 57–66, 115–118, 120–127, 153, 154, 201, 202, 213, 215, 270 Endoplasmic reticulum-mitochondrion encounter structure (ERMES) complex ..................... 58, 214 Endosomes ................................................. 115, 137, 146, 148, 149, 257 ER-PM junctions ........................................................ 1–10 Escherichia coli (E. coli) strains AL95 (pss 93::kan lacY::Tn9) ...................... 166, 168, 171, 174, 178 BL21 (DE3) ..................................139, 141, 295, 296 BL21 (DE3) pLysS ................................................. 155 BL21 Gold............................................................... 273 C43 (DE3) .............................................................. 216 DH5α.............................................................. 139, 258 SM2-1 (plsC(Ts)) ........................................... 168, 171 Extruder...................................................... 121, 131, 184, 186, 187, 192, 219, 274, 282, 289, 311, 320 Extrusion .................................................... 132, 184, 196, 219, 229, 284, 323

F Fast protein liquid chromatography (FPLC)............................158, 160, 259, 274, 276 Fatty acids ..........................................................39, 74, 75, 82–85, 92, 96, 109, 110, 113, 120, 123, 126, 188, 189, 191–193, 196, 225, 240 Filipin .................................................................... 137–148 FK506-binding protein12 (FKBP12) module ........15, 16 Flippases......................................................................... v, vi Fluorescence ................................................. 2, 15, 27, 43, 50, 99, 118, 144, 160, 184, 193, 203, 219, 242, 271, 315 Fluorescence microscopy ...................................... 2, 9, 10, 36, 50, 117, 118, 122, 123, 158 Fluorescent and functionalized lipid 1,2-diacyl-sn-glycero-3-phosphoethanolamine-N(5-dimethylamino-1-naphthalenesulfonyl) (DNS-PE or Dansyl-PE) ......................... 270, 274 1,2-diacyl-sn-glycero-3-phosphoethanolamineN-(7-nitro-2-1,3-benzoxadiazol-4-yl) (NBD-PE) ...............................185, 191, 206, 219 1,2-diacyl-sn-glycero-3-phosphoethanolamine-N(lissamine rhodamine B sulfonyl) (Rhod-PE or Rh-PE) ............................................. 206, 274, 311 2-NBD-(glyco)sphingolipid .......................... 310, 319 bifunctional sphingosine (pacSph) ........................... 97 dehydroergosterol (DHE)............................ 116, 118, 123, 124, 270, 274, 290 trifunctional lipids ..................................................... 96

Fo¨rster resonance energy transfer (FRET) ................... 19, 24, 50, 122, 130, 131, 133, 203, 208, 209, 238, 241–246, 248–251, 270–272, 283, 285, 286, 290, 307, 310, 319–321, 323 FRB ............................................................................15–17

G Gas chromatography-flame ionisation detector (GC-FID) ........................................ 75, 83, 86, 90 Genetically-encoded fluorescent lipid sensors .................vi Glycerophospholipid remodeling system .................... 169 Golgi complex ..............................................47–52, 54, 55

H Highly-mobile-membrane-mimetic (HMMM) model ................................................326, 328–337 High-performance thin-layer chromatography (HPTLC) .............................................97, 99, 106, 108, 111–113, 215, 217, 222–225, 228–231, 309, 316 Histamine ...................................................................... 2–8

I Image analysis.................. 6, 8, 40, 42, 43, 141, 145–148 Immunoblotting .......................... 65, 112, 128, 161, 162 Immunofluorescence ..........................50, 54, 55, 99, 150 Intermembrane lipid transfer ......................307, 319–321 Intracellular lipid localization ......................................... 14 Ionic exchange chromatography DEAE-Sephacel.............................................. 246, 247 DEAE-Sephadex A25 .................................... 309, 314 HiTrap Q HP anion exchange column ................. 158 Isopropyl β-D-1-thiogalactopyranoside (IPTG) ............................................ 139, 141, 155, 157, 170, 176, 179, 216, 273, 277, 282, 295, 296 Isothermal titration calorimetry (ITC).............. 258, 260, 262, 263, 265, 293

K Kinetics analysis .................................................... 283, 287

L Labeling ............................................................... 5, 23, 35, 70, 77, 80, 91, 97–99, 101, 102, 105–108, 110–111, 116, 120, 127, 128, 130, 137–148, 238, 248, 250, 273–282 Ligand specificity.................................................. 293–305 Lipid bis(monooctadecenoylglycero)phosphate (BMP) .............................................. 311, 319, 320 cardiolipin (CL)........................................58, 217, 225

INTRACELLULAR LIPID TRANSPORT: METHODS ceramide.................................... 48, 99, 110, 192, 238 cholesterol ...............................................49, 120, 125, 126, 154, 270, 319, 320 cholesteryl phosphocholine (CholPC) ......... 107, 110 D-erythro-sphingosine ........................................... 110 diacyl-cardiolipin (DCL) ............................... 166, 168 1,2-diacyl-sn-glycerol (DAG) ................................. 206 digalactosyldiacylglycerol (DGDG) ..... 70, 72, 83, 88 1,2-dimyristoyl-sn-glycero-3phosphatidylethanolamine (DMPE) ................ 108 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) ................. 121, 206, 217, 274, 310, 336 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE) .................................................... 121, 217 1,2-dioleoyl-sn-glycero-3-phospho-L-serine (DOPS) 217 E. coli total lipid extract ................................. 217, 225 ergosterol............................................... 116, 128, 270 galactosylceramide................................................... 108 ganglioside............................................................... 239 globoside ................................................................. 108 glucosylceramide .............................................. 48, 112 glycolipid ............................................... 106, 238, 244 glycosphingolipid ...............................................48, 49, 105–113, 238, 239, 246, 249 lactosylceramide ............................................. 108, 112 L-α-phosphatidic acid (PA) .................................... 217 L-α-phosphatidylglycerol (PG) .............................. 217 L-α-phosphatidylinositol (PI) ....................... 121, 217 L-α-phosphatidylinositol-4,5-bisphosphate (PI(4,5)P2) ........................................................ 206 L-α-phosphatidylinositol-4-phosphate (PI4P)...... 274 lyso-PS ....................................................................... 36 membrane-spanning lipid (MSL).................. 307–323 membrane-spanning phosphoethanolamine (MSPE) .............................................308, 311–316 monoacyl-cardiolipin (MCL) ........................ 166, 168 1-oleoyl-2-hydroxy-sn-glycero-3-phosphate (18:1 Lyso PA) .................................................. 171 1-oleoyl-2-hydroxy-sn-glycero-3-phospho(1’-rac-glycerol) (18:1 Lyso PG) ..................... 171 1-oleoyl-2-hydroxy-sn-glycero-3-phosphocholine (18:1 Lyso PC) ................................................. 171 1-oleoyl-2-hydroxy-sn-glycero-3phosphoethanolamine (18:1 Lyso PE) ............ 171 1-palmitoyl-2-oleoyl-sn-glycero-2-phosphocholine (POPC).............................................................. 336 1-palmitoyl-2-oleoyl-sn-glycero-3phosphoethanolamine (POPE) ........................ 335 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol (POPG).............................................................. 335 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine (POPS)............................................................... 206

AND

PROTOCOLS Index 343

phosphatidylcholine (PC)........................... 50, 58, 83, 99, 168, 181, 225, 270 phosphatidyldimethylethanolamine (PDME) ......... 63 phosphatidylethanolamine (PE)........................57, 58, 83, 88, 181, 225 phosphatidylserine (PS) ........... 57, 58, 181, 202, 225 phosphoinositide ........................................23, 47, 272 phospholipid..........................................57–59, 61–66, 70, 92, 108, 111, 122, 132, 133, 135, 165, 166, 172, 173, 178, 182, 192, 193, 202, 214, 215, 217, 219, 223–229, 231, 233, 274 sphinganine....................................................... 48, 110 sphingolipid ..............................................v, 47–50, 52, 95, 105, 106, 110, 237–252, 310, 319, 320, 322 sphingomyelin ..................48, 99, 108, 154, 181, 257 tetraether lipid ............. 308, 311, 312, 317, 321, 322 yeast polar lipid extract .................................. 217, 228 yeast total lipid extract .......................... 217, 228, 229 Lipid analysis ........................... vi, 70, 107, 108, 111–113 Lipid bilayers .......................................288, 308, 325–337 Lipid-binding domains (LBD) ................. 24–26, 30, 272 Lipid displacement assay ... 215, 219, 225–228, 230, 231 Lipid exchange ...............................................36, 116, 229 Lipid extraction ........................................ 71, 74, 81, 108, 111, 118, 121, 123, 125, 128, 130, 135, 168, 170, 172, 174, 175, 217, 223–226, 244, 250, 310–312 Lipid metabolism ....................................... 47–52, 54, 55, 96, 97, 102, 105, 269 Lipidomic analysis .............................................. 77, 82, 83 Lipid synthesis ...................................................v, 307–323 Lipid trafficking .................................................. 69–72, 80 Lipid transfer protein (LTP) extended synaptotagmine .............................. 201–211 glycolipid transfer protein (GLTP) .............. 106, 110, 237–252 Mdm1 ............................................................. 213–233 Nir2....................................................................2–8, 10 NPC1 ...........................................................v, 257–267 NPC2 .............................................................. 257–267 ORP5 ......................................................................... 14 Osh4p ............................................ 270, 272, 285–290 Osh proteins ................................................... 270, 272 oxysterol-binding protein related protein....................................................... 270, 272 phosphatidylinositol transfer protein ................... 1–10 STARD6 ......................................................... 295–303 synaptotagmin-like mitochondrial lipid binding protein (SMP) .......................................... 201, 223 Lipid vesicle, see Liposome Liposome ......................................................vi, 14, 45, 58, 70, 117, 182, 202, 215, 241, 270, 307 Liposome flotation.......................................220, 228–233

INTRACELLULAR LIPID TRANSPORT: METHODS

344 Index

AND

PROTOCOLS

Liquid chromatography-mass spectrometry (LC-MS) .............................................................. 75 Low-density lipoprotein (LDL) .......................... 153, 257 Luciferase...........................................................16, 18, 20, 24–28, 30–32 Lysis bacteria................................................... 157, 276, 277 cell ...................................................41, 141, 157, 207, 220, 222, 262, 277, 278 yeast ................................................................ 217, 222 Lysophospholipid flipping .......................... 165, 167, 169 Lysophospholipid transporter (LplT).........166–169, 178

M Mass spectrometry (MS)...................................70, 75, 82, 83, 85, 91, 95, 312, 314, 315, 322 Membrane contact sites (MCSs) .........................v, 13, 49, 69–71, 215, 270 Membrane fusion ................................................. 307–323 Membrane proteins...................................... 77, 153, 170, 178, 182, 183, 214, 257, 325, 327, 330, 331, 334 Membrane tethering ............................................ 201–211 Metabolisms ............................................... 47–52, 54, 55, 95, 97, 99, 101, 102, 105, 269, 270 Methanolysis.......................................... 74, 75, 82–84, 90 Methyl-beta-cyclodextrin (MβCD).................... 119–121, 128, 130, 134, 135, 138, 139, 143, 145, 146, 149 Microfluidics.......................... 36, 37, 40, 42, 44, 45, 250 Microsomes .......................................................... 119, 126 Mitochondria...............................................57–66, 69–92, 137, 213, 252 Mitochondria-associated membranes (MAMs) ........... 213 Mitochondrial Transmembrane Lipoprotein (MTL) complex .............................................69–92 Model system .................................................................... 4

N NanoDrop ...........................................140, 156, 158, 206 Niemann-Pick disease type C ...........................v, 257–267

O Organelle membranes ................................v, 24, 269, 270

P Palmitic acid ......................................................... 107, 110 Perfringolysin O ............................................................ 138 Peripheral membrane proteins (PMP)................ 325–337 Phospholipase A2 (PLA2) .................................. 168, 171, 172, 174, 175, 178 Phospholipid transport ................................58–62, 64, 66 Photoactivation ............................................................... 96 PI(4,5)P2 replenishment ................................. 1–3, 5–7, 9

Plasma membrane (PM) ..................................v, 1, 23–32, 36, 49, 115, 116, 137–148, 158, 181, 213, 237, 270, 308 Plasmid E-Syt1cyto ........................................................ 202–210 FYVE domain ............................................................ 52 GFP-PLCδ-PH........................................... 2, 3, 5–7, 9 GST-Osh4p ............................................................. 282 GST-PHFAPP ......................................................... 279 His6-EGFP-D4 .............................................. 139, 141 histidine-tagged yeast Mdm12............................... 216 H1R ...............................................................2–5, 7–10 human NPC1 ........................................ 257, 258, 260 human NPC2 ................................................. 260, 263 human STARD6...................................................... 295 L10-mVenus-T2A-sLuc-Lact-C2.......................15, 17 L10-mVenus-T2A-sLuc-P4M(2) ....................15, 17 LactC2-GFP .............................................................. 39 Lyn-targeted FRB ..................................................... 16 MAPPER ...................................................... 3–5, 8, 10 mCherry-tagged FKBP-ORP5................................. 16 Nir2-mCherry ..................................................... 3–5, 8 ORP5-PH-GFP...................................................51, 52 ORP8L-PH-GFP ................................................51, 52 pAC-PCSlp-Sp-Gm.......................168, 171, 174, 178 pALOD4......................................................... 155, 157 Pleckstrin homology (PH) domain................2, 9, 14–16, 26, 49, 50, 271, 276 Polymerase chain reaction (PCR) ............ 39–41, 43, 261 Protein labeling ............................................................. 276 Protein purification .................................... 160, 216–218, 220–222, 257–267, 273–275 Proteoliposomes................................................... 181–197 Pulse-chase ............................ 97, 98, 100, 102, 128, 130

Q Quantification of phospholipids................. 122, 132, 133 Quartz cuvettes ................................. 122, 133, 240, 244, 274, 276, 280, 281, 285, 311, 320, 321

R Radioactivity counting ............................... 119, 127, 167, 169, 207, 247 Radiolabelled lipid and precursor [3H]-cholesterol ...................................................... 119 D-erythro-[3-3H]-sphingosine ..................... 107, 109 1,2-dioleoyl [9,10-3H] rac-glycerol (3H-DAG) ......................................................... 206 [9,10-3H]-palmitic acid .......................................... 107 UDP-[14C]-galactose................................................ 72 Rapamycin .................................................................15–20 Real-time assay ....................................................... 36, 250 Recombinant protein ................... 14, 215, 258, 260, 296

INTRACELLULAR LIPID TRANSPORT: METHODS Reconstitution ....................................182, 184, 187–188, 190, 193, 197, 219 Reverse phase .............................................. 118, 124, 128 Rotary evaporator ...................................... 185, 187, 196, 274, 282, 310, 311

S Saccharomyces cerevisiae.............................. 117, 122, 123, 126, 134, 189, 214, 222–223, 225, 228, 327, 336 Scrambling assays ...............................182, 183, 188–190, 193, 195 Silicone oil-spin method .....................169–171, 177, 179 Size-exclusion chromatography Sephacryl S200HR.................................................. 274 Superdex 200 .......................................................... 208 Superose 6 ............................................. 259, 262, 263 Sodium dithionite (Na2S2O4) ............................ 182, 183, 185, 189, 190, 193, 194, 196, 197, 204, 206 Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE).............................65, 128, 155, 156, 158, 160, 163, 208, 218, 221–223, 231, 233, 263, 275, 277, 278, 281, 297 Solvent system ....................................107, 108, 110–112, 119, 124, 125, 168, 170, 172, 178, 239, 249 Sonication .............................................44, 111, 162, 176, 187, 240, 241, 246, 249, 250 Spectrofluorometer .................................... 186, 189, 240, 270, 276, 283, 285, 289 Spectrometer .............................................. 75, 85, 86, 92, 122, 131, 142, 156, 160, 274 Spheroplasts..............................59, 60, 64, 165, 167, 169 Sterol regulatory element binding protein (SREBP)...................................153, 154, 161, 163 Subcellular fractionation.......................65, 116, 127, 128

T Testosterone ........................................295, 298, 300–303 Thermal shift assays (TSA) .................................. 293–304 Thermoplasma acidophilum .........................308, 310–312 Thin layer chromatography (TLC) .................. 60, 62–63, 65, 70, 96–99, 102, 106–108, 110–113, 119, 124–126, 167–172, 174, 175, 177, 178, 312, 314–316, 318, 319, 322

AND

PROTOCOLS Index 345

TLC revelation CuSO4 ................................................... 312, 314, 318 iodine ..................................................... 108, 112, 113 orsinol ...................................................................... 108 TMEM16/Anoctamin......................................... 181–197 Total internal reflection fluorescence (TIRF) microscopy ...................................................2–7, 10 Transfection effectene...............................................................16, 17 ExpiFectamine 293 ........................................ 205, 207 GeneCellin...........................................................27, 28 Lipofectamine 2000............................................27, 28 PEI25K .................................................................... 258 TransIT-LT1...............................................4, 5, 52, 54 Transient expressions ..........................190, 258, 260–262 Transport rate........................................66, 133, 135, 287

V Venus protein ............................................................30, 32

W Web resources and software CellASIC ONIX ........................................................ 40 CellProfiler ..................................................... 141, 146 Chimera .......................................................... 328, 331 Excel.....................................................................40, 45 Fiji ............................................. 6, 8, 40, 42, 141, 146 Grace............................................................... 194, 195 GraphPad Prism ............................................. 265, 295 ImageJ..............................................45, 52, 54, 55, 63 ORIGIN ................................................ 251, 259, 266 Packmol .......................................................... 328, 329 protein data bank (PDB) ........................................ 327 ProtParam................................................................ 162 visual molecular dynamics (VMD)................ 328, 331

Y Yeasts........................................................... 13, 14, 35, 36, 38–45, 57–60, 63–66, 115, 116, 128, 131, 170, 201, 213, 215–217, 221, 222, 224, 225, 227, 228, 230, 231, 270, 273, 283, 289, 310, 327, 336