Optogenetics: light-driven actuators and light-emitting sensors in cell biology 978-1-78801-237-9, 1788012372, 978-1-78801-328-4, 978-1-78801-526-4

Citation preview

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-FP001

COMPREHENSIVE SERIES IN PHOTOCHEMICAL AND PHOTOBIOLOGICAL SCIENCE

Series editors:

Lesley Rhodes

University of Manchester, UK

Evelyne Sage

Institut Curie, France

Massimo Trotta

Istituto per i Processi Chimico Fisici – CNR, Italy

View Online

COMPREHENSIVE SERIES IN PHOTOCHEMISTRY AND PHOTOBIOLOGY

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-FP001

Series editors: Lesley Rhodes, Evelyne Sage and Massimo Trotta Titles in this Series: Volume 1

UV Effects in Aquatic Organisms and Ecosystems Edited by E.W. Helbling and H. Zagarese

Volume 2

Photodynamic Therapy Edited by T. Patrice

Volume 3

Photoreceptors and Light Signalling Edited by A. Batschauer

Volume 4

Lasers and Current Optical Techniques in Biology Edited by G. Palumbo and R. Pratesi

Volume 5 From DNA Photolesions to Mutations, Skin Cancer and Cell Death Edited by E. Sage, R. Drouin and M. Rouabhia Volume 6

Flavins: Photochemistry and Photobiology Edited by E. Silva and A.M. Edwards

Volume 7

Photodynamic Therapy with ALA: A Clinical Handbook Edited by R. Pottier, B. Krammer, R. Baumgartner and H. Stepp

Volume 8 Primary Processes of Photosynthesis, Part 1: Principles and Apparatus Edited by G. Renger Volume 9 Primary Processes of Photosynthesis, Part 2: Principles and Apparatus Edited by G. Renger Volume 10 Biophysical and Physiological Effects of Solar Radiation on Human Skin Edited by Paolo U. Giacomoni Volume 11 Photodynamic Inactivation of Microbial Pathogens: Medical and Environmental Applications Edited by Michael R. Hamblin and Giulio Jori

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-FP001

Volume 12 Surface Water Photochemistry Edited by Paola Calza and Davide Vione Volume 13 Singlet Oxygen: Applications in Biosciences and Nanosciences, Volume 1 Edited by Santi Nonell and Cristina Flors Volume 14 Singlet Oxygen: Applications in Biosciences and Nanosciences, Volume 2 Edited by Santi Nonell and Cristina Flors Volume 15 Photodynamic Medicine: From Bench to Clinic Edited by Herwig Kostron and Tayyaba Hasan Volume 16 Microalgal Hydrogen Production: Achievements and Perspectives Edited by Michael Seibert and Giuseppe Torzillo Volume 17 Light in Forensic Science: Issues and Applications Edited by Giorgia Miolo, Jacqueline L. Stair and Mire Zloh Volume 18 Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa Visit our website at www.rsc.org/books

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-FP001

View Online

     

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-FP001

Comprehensive Series in Photochemistry and Photobiology – Volume 18

Optogenetics

Light-driven Actuators and Light-emitting Sensors in Cell Biology

Editors Sophie Vriz

Paris Diderot University CNRS UMR 7241/INSERM U1050 Collège de France France Email: [email protected] and

Takeaki Ozawa

The University of Tokyo Japan Email: [email protected]

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-FP001

View Online

ISBN: 978-1-78801-237-9 PDF ISBN: 978-1-78801-328-4 EPUB ISBN: 978-1-78801-526-4 Print ISSN: 2041-9716 Electronic ISSN: 2041-9724 A catalogue record for this book is available from the British Library © European Society for Photobiology 2019 All rights reserved Apart from fair dealing for the purposes of research for non-commercial purposes or for private study, criticism or review, as permitted under the Copyright, Designs and Patents Act 1988 and the Copyright and Related Rights Regulations 2003, this publication may not be reproduced, stored or transmitted, in any form or by any means, without the prior permission in writing of The Royal Society of Chemistry or the copyright owner, or in the case of reproduction in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK, or in accordance with the terms of the licences issued by the appropriate Reproduction Rights Organization outside the UK. Enquiries concerning reproduction outside the terms stated here should be sent to The Royal Society of Chemistry at the address printed on this page. Whilst this material has been produced with all due care, The Royal Society of Chemistry cannot be held responsible or liable for its accuracy and completeness, nor for any consequences arising from any errors or the use of the information contained in this publication. The publication of advertisements does not constitute any endorsement by The Royal Society of Chemistry or Authors of any products advertised. The views and opinions advanced by contributors do not necessarily reflect those of The Royal Society of Chemistry which shall not be liable for any resulting loss or damage arising as a result of reliance upon this material. The Royal Society of Chemistry is a charity, registered in England and Wales, Number 207890, and a company incorporated in England by Royal Charter (Registered No. RC000524), registered office: Burlington House, Piccadilly, London W1J 0BA, UK, Telephone: +44 (0) 207 4378 6556. For further information see our website at www.rsc.org Printed in the United Kingdom by CPI Group (UK) Ltd, Croydon CR0 4YY, UK

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-FP007

Preface Nine years after the seminal review The optogenetic catechism by Gero Miesenböck1, optogenetic tools have expanded its application not only in neuroscience but also in cell physiology. It has allowed a spatiotemporal quantitative approach to interrogating biological processes such as cell trafficking, cell signalling and the discovery of new biological mechanisms. Owing to this surge of new applications of optogenetics in cell biology, we thought that it was the time for a survey of what has been done and what remains to be explored. Optogenetic development needs convergent efforts by chemists, physicists and biologists and these three disciplines are well represented amongst the authors. We decided to organize the book in three parts, which correspond to the major research territories in the optogenetic landscape: light taming, light-emitting sensors and light-driven actuators. Illumination and signal recording are the heart of optogenetics. Light needs to be tamed for sensitivity, faster recording and better resolution to improve our understanding of life mechanisms. Different approaches are used to achieve these goals and today living biological systems can be observed in three dimensions (3D), at subsecond resolution and at the diffraction limit or below. The first chapter lays the ground for illuminating methods and discusses light-sheet fluorescence microscopy (LSFM), which allows very fast recording of signals in a large volume. The combination of optogenetic tools with LSFM opens up the possibility of recording fluorescent signals from thousands of individual cells in less than 1 s. This allows the intricate study of cell signalling at the organ or animal scale with single-cell resolution and provides access to signalling patterns, spreading and feedback loop operation in living animals. This now leads to the need to develop computational methods to process and analyse massive datasets produced by such readout

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

vii

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-FP007

viii

Preface

techniques. The second chapter addresses the adjustment of super-resolution microscopy to optogenetics. Indeed, super-resolution microscopy of living systems remains a challenge, even though it is essential to stretch the limits for the analysis of signalling spreading inside the cell. Light-emitting sensors transform physiological signals into fluorescent signals. Their development revolutionises cell biology, and a large toolbox of fluorescent proteins able to reveal protein abundance, localization, dynamics and activity with unprecedented temporal and spatial resolution is now available. Each one is peculiar and specific to the signal to be recorded; however, two major families of fluorescent proteins are used: one is derived from canonical fluorescent proteins and the other from non-fluorescent probes (fluorogens) that become fluorescent only when complexed to a fluorogen-activating protein (FAP). Chapters 3 and 4 address major developments in these two families and give precise and practical examples for each of them. The spectrum of available wavelengths for absorption and emission is still expending, making multiplexing easier and signalling network recording possible. The last section is dedicated to actuators. The ultimate proof for the comprehension of a phenomenon is often the possibility of initiating and manipulating the signal(s) involved. Light-driven actuators achieve this goal. In order to turn a protein into a light-responsive entity, it is possible to rely either on light-sensing proteins already characterized in a specific organism or on the addition of a light-sensitive module to a protein of interest. In most cases, under illumination, the associated chromophore isomerizes and induces an overall conformational change of the protein, which can be rerouted to manipulate protein activity.2,3 Light-sensitive modules are continuously being improved for dynamic response and light sensitivity and adapted for the regulation of a wide range of proteins. The last part of this book presents specific examples of actuators dedicated to cell biology. Because of space constraints, it was not possible to be exhaustive, but we feel that the sampling we have made will give the flavour of what is possible with some examples presented in practical details. Chapter 6 gives an overview of the chromophore light inactivation (CALI) method to inactivate proteins. Chapters 5, 7 and 8 present different examples of signal transduction light manipulation and Chapter 10 outlines a general method to control the activity of receptor tyrosine kinases. Chapter 9 reports light control of transcription to discriminate oscillatory versus sustained gene expression during embryonic development. Finally, Chapter 11 gives practical examples of light-controlled mechanotransduction. In conclusion, we would like to express our strongest thanks to Evelyne Sage, Emeritus Research Director at the Curie Institute and co-Editor of the book series, who initiated this project and encouraged us throughout the preparation of this book. Sophie Vriz and Takeaki Ozawa

View Online

Preface

ix

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-FP007

References . G. Miesenböck, Science, 2009, 326, 395. 1 2. A. Gautier, C. Gauron, M. Volovitch, D. Bensimon, L. Jullien and S. Vriz, Nat. Chem. Biol., 2014, 10, 533. 3. M. Endo and T. Ozawa, J. Photochem. Photobiol., C, 2017, 30, 10.

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-FP011

Contents Part I: llumination Strategies and Imaging Chapter 1    Fast Volumetric Imaging Using Light-sheet   Microscopy. Principles and Applications  Sébastien Wolf and Georges Debrégeas Chapter 2    Super-resolution Microscopy  Xiyu Yi, Tal-Zvi Markus, Xavier Michalet, Shimon Weiss and David Bensimon

3 25

Part II: Light-emitting Sensors Chapter 3    The Glowing Panoply of Fluorogen-based   Markers for Advanced Bioimaging  Arnaud Gautier

43

Chapter 4    Optogenetic Reporters for Cell Biology and   Neuroscience  Wei Zhang and Robert E. Campbell

63

Part III: Light-driven Actuators Chapter 5    Light-driven Actuators: Spatiotemporal Dynamics   of Cellular Signaling Processes  Yoshibumi Ueda and Moritoshi Sato

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

xi

101

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-FP011

xii

Contents

Chapter 6    Optogenetic Control of the Generation of Reactive   Oxygen Species for Photoinducible Protein   Inactivation and Cell Ablation  Takeharu Nagai and Yemima Dani Riani Chapter 7    Optogenetic Tools for Quantitative Biology:   The Genetically Encoded PhyB–PIF Light-inducible Dimerization System and Its Application for   Controlling Signal Transduction  S. Oda, Y. Uda, Y. Goto, H. Miura and K. Aoki Chapter 8    Quantitative Control of Kinase Activity with a   Mathematical Model  Genki Kawamura and Takeaki Ozawa Chapter 9    Light Control of Transcription in Cells  Akihiro Isomura and Ryoichiro Kageyama Chapter 10    Building Light-inducible Receptor Tyrosine   Kinases  Nury Kim, Hyerim Park, Doyeon Woo and Won Do Heo

117

137

149 169

181

Chapter 11    Mechanotransduction and Optogenetics  Adèle Kerjouan and Olivier Destaing

197

Subject Index 

221

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

Part I

Illumination strategies and imaging

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

View Online

         

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

View Online

Chapter 1

Fast Volumetric Imaging Using Light-sheet Microscopy. Principles and Applications Sébastien Wolf and Georges Debrégeas* Sorbonne Université, CNRS, Laboratoire Jean Perrin, UMR 8237, 75005 Paris, France *E-mail: [email protected]

Table of Contents 1.1.  I ntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.  From Classical Approaches to LSFM: the Question   of Optical Sectioning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.1.  Confocal Microscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.2.  Two-photon Microscopy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2.3.  Light-sheet Fluorescent Microscopy. . . . . . . . . . . . . . . . . . . . 1.3.  Optical Principles of Light-sheet Microscopy. . . . . . . . . . . . . . . . . . . 1.3.1.  The Origins of LSFM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3.2.  Spatial Resolution in LSFM. . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.  Building the Improved LSFM: Pushing the Limits. . . . . . . . . . . . . . . 1.4.1.  Increasing the Spatial Resolution. . . . . . . . . . . . . . . . . . . . . . 1.4.2.  High-speed Volumetric Imaging. . . . . . . . . . . . . . . . . . . . . . . 1.4.3.  Contrast Enhancement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4.4.  Imaging Deeper in Semi-transparent Samples. . . . . . . . . . . 1.4.5.  Light-sheet Microscopy with a Single Objective. . . . . . . . . .

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

3

5 5 7 7 8 10 10 11 13 13 16 16 17 18

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

4

Sébastien Wolf and Georges Debrégeas

1.5.  A  pplication: Light-sheet Imaging of Zebrafish Brain. . . . . . . . . . . . . 19 1.5.1.  Light-sheet-based Whole-brain Functional   Imaging in Zebrafish Larvae . . . . . . . . . . . . . . . . . . . . . . . . . . .   19 1.5.2.  Whole-brain LSFM-based Functional Imaging   to Study Sensorimotor Integration in the   Vertebrate Brain. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   20 1.6.  LSFM in the Next Decade. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

View Online

Fast Volumetric Imaging Using Light-sheet Microscopy

5

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

1.1. Introduction The parallel development of genetic engineering and microscopy techniques has propelled biology into a new era. It is now possible to genetically encode fluorescent proteins in targeted cell populations and then to monitor optically their dynamics and activity in vivo with exquisite precision. Such approaches have become a central tool to probe the structure, development and function of biological tissues. In recent decades, great efforts have been made to improve the performances of fluorescence imaging systems, the challenge being how to record the fluorescence signal of a specific section/volume within a living specimen at high spatial and temporal resolution while minimally damaging the tissue. As we will show, light-sheet fluorescence microscopy (LSFM) surpasses classical optical methods for each of these criteria. Combining massive parallelization of data acquisition and efficient optical sectioning, LSFM allows long-term and fast volumetric recordings at high spatiotemporal resolution with minimal photodamage. Since the publication of the seminal article by Stelzer's group in 2004,1 the impact of LSFM has been significant in numerous fields of biology such as neuroscience, anatomy, physiology, developmental biology and cell biology (Figure 1.1). This technique has been used to image biological tissues at scales ranging from subcellular compartments2 to entire organs;3 it has enabled the long-term monitoring of entire developing embryos over days,4,5 and has also provided video-rate four-dimensional (4D) movies of rapid physiological processes such as heart beating in zebrafish.59 The power of LSFM lies in its unique amenability to diverse biological questions. However, each of these questions generally requires specifically adapting the microscope to the sample and to the type of information that one wishes to obtain. Although a standard LSFM instrument is relatively inexpensive to build and is now commercially available, it can also be combined with more advanced imaging techniques, such as super-resolution,6 two-photon excitation7 and structured illumination.8 In this chapter, we show how LSFM has emerged as a novel and powerful alternative to confocal and two-photon microscopy. After having established the principles underlying this new imaging method, we will review some of the most significant improvements that have been developed in recent years to address specific biological questions. Finally, we present in detail one particular application of LSFM, namely the functional imaging of the entire zebrafish larva brain.

1.2. From Classical Approaches to LSFM: the Question of Optical Sectioning Probing the physiology of living tissues in a minimally invasive way is arguably the central goal of optical microscopy in biology. Starting with the early microscopes designed in the seventeenth century, microscopy techniques

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

6

Sébastien Wolf and Georges Debrégeas

Figure 1.1.  Examples of LSFM applications. (a) LSFM recording of Drosophila embryonic development. Projections for dorsal (top) and ventral (bottom) views are shown. The Drosophila embryo is recorded at 30 s intervals with a frame period of 15 s from 3 to 18.5 h post-fertilization. PC, pole cells; VF, ventral furrow. (b) LSFM recording of a section of the brain volume reconstructed from a complete 3D stack of a Danio rerio (zebrafish) brain at 5 days post-fertilization. (c) Evolution of the Golgi apparatus (magenta) during mitosis of a live LLC-PK1 cell, with views parallel (top) and perpendicular (bottom) to the mitotic plane, showing partial fragmentation in metaphase and anaphase and eventual recondensation in telophase. The Golgi and chromosomes (green) are visualized via mEmerald–Mann II and mEmerald–histone H2B fluorescence. Scale bars, 5 µm. (d) Purkinje cell micron-scale neuroanatomy in the whole cerebellum. Left: coronal digital sections. Scale bars, 1 mm. Right: 10× magnification of the regions highlighted by the yellow boxes in the left panel. (e) Electrically tunable lens (ETL)-LSFM captures intracardiac blood flow. Front and side views of a 48 h post-fertilization zebrafish embryo [Tg(myl7:GFP, gata1a:DsRed)] imaged with movie-stack synchronization (left) and ETL-LSFM (right), showing myocardium in cyan and red blood cells in red. Solid vertical lines indicate position of cross-section. Scale bars, 30 µm. Part (a) reprinted with permission from Springer Nature: R. Tomer, K. Khairy, F. Amat and P. J. Keller, Quantitative high-speed imaging of entire developing embryos with simultaneous multiview light-sheet microscopy, Nat. Methods, 9, Copyright © 2012 Springer Nature. Part (c) reprinted with permission from Springer Nature: T. A. Planchon, L. Gao, D. E. Milkie, M. W. Davidson, J. A. Galbraith, C. G. Galbraith and E. Betzig, Rapid three-dimensional isotropic imaging of living cells using Bessel beam plane illumination, Nat. Methods, 8, Copyright © 2011 Springer Nature. Part (d) reprinted with permission from L. Silvestri, A. Bria, L. Sacconi, G. Iannello and F. S. Pavone, Confocal light sheet microscopy: micron-scale neuroanatomy of the entire mouse brain, Opt. Express, 20, 20582–20598. Copyright 2012 Optical Society of America. Part (e) reprinted with permission from Springer Nature: M. Mickoleit, B. Schmid, M. Weber, F. O. Fahrbach, S. Hombach, S. Reischauer and J. Huisken, High-resolution reconstruction of the beating zebrafish heart, Nat. Methods, 11, Copyright © 2014 Springer Nature.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

Fast Volumetric Imaging Using Light-sheet Microscopy

7

have always been an important source of new discoveries for biologists. In the twentieth century, the fluorescent labeling of specific cellular structures was developed, giving rise to fluorescence microscopy. However, one important issue remained – how to record the fluorescence signal of a specific section/volume within a volumetric biological tissue with large contrast. In standard epifluorescence microscopy, the entire specimen is illuminated, hence the collected fluorescent photons originate from both the in-focus and out-of-focus regions. These out-of-focus photons inevitably blur the imaged plane such that, for thick samples, the contrast vanishes. During the second half of the twentieth century, optical techniques were therefore developed in order to selectively image a thin section within a specimen, a process called optical sectioning. Two different approaches were successively proposed: confocal imaging and two-photon microscopy. 1.2.1. Confocal Microscopy In 1955, Marvin Minsky, who later worked on artificial intelligence, laid the foundation of the confocal microscope when he was a young professor: One day it occurred to me that the way to avoid all that scattered light was to never allow any unnecessary light to enter in the first place. […] There is no way to eliminate every possible such ray, because of multiple scattering, but it is easy to remove all rays not initially aimed at the focal point; just use a second microscope (instead of a condenser lens) to image a pinhole aperture on a single point of the specimen.9 The confocal microscope thus uses a pinhole conjugated to the focal point of the objective to block all photons originating from outside the focal volume. This simple method allows for the recording of the fluorescence signal from only one small focal volume of the sample. To build a complete image, one sequentially scans this elementary volume throughout the entire sample by moving the observation (pinhole) and the excitation arm together. Confocal microcopy has several limitations: (i) the need to scan throughout the sample severely limits the acquisition frame rate; (ii) the photoefficiency is intrinsically low since the entire specimen is continuously exposed to the excitation beam even if only one voxel is imaged at a given time; and (iii) confocal microscopy generally uses visible light and therefore has a limited penetration depth in heterogeneous (light-scattering) samples. 1.2.2. Two-photon Microscopy Fluorescence can be induced by the absorption of one photon of a given energy or by the simultaneous absorption of two photons of half the energy (and thus twice the wavelength). In the latter mode, the fluorescence signal varies quadratically (rather than linearly) with the intensity of the excitation light, and it is extremely low under standard illumination conditions. Hence a molecule of the natural chromophore rhodamine B excited by sunlight

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

8

Sébastien Wolf and Georges Debrégeas

will experience a two-photon absorption transition every 107 years.10 Observing this phenomenon therefore requires the use of spatially and temporally focused light. In two-photon microscopes, a near-infrared laser beam with 100 fs long high-energy pulses at a repetition rate of 80 MHz is focused through a high numerical aperture objective. The quadratic dependence of the fluorescence emission with the intensity yields a natural optical sectioning mechanism: only at the focal point of the objective does the spatial and temporal focusing yield a sufficient peak intensity to evoke significant fluorescence. The use of infrared light further minimizes scattering in the tissue, which allows for the imaging of thicker samples, and also limits photodamage in comparison with confocal microscopy. Since the invention of the two-photon microscope in 1990,11 numerous applications have been developed in biology, especially in neuroscience,12–14 in immunology15 and in developmental biology.16,17 1.2.3. Light-sheet Fluorescent Microscopy Although they use different approaches for optical sectioning, both confocal and the two-photon microscopes share the need for faster scanning across the specimen in order to acquire a complete image. The laser dwell time τdwell at each voxel needs to be large enough to allow for a significant number of fluorescent photons to be emitted (typically τdwell = 0.1–1 µs). This in turn imposes a compromise between the size of the imaged region (number of pixels N) and the acquisition frame rate facq, such that Nfacq = 1/τdwell. This drastic limitation in terms of the data throughput of point-scanning imaging motivated the development of light-sheet fluorescence microscopy (LSFM). In LSFM, the illumination is provided by a micrometer-thick sheet of light projected across the sample while the fluorescent photons are collected at 90° with a camera. This approach allows for a massive parallelization of the data collection, as all pixels from a single plane are now imaged simultaneously (see Figure 1.2). The maximum frame rate becomes facq = 1/τdwell and is in practice limited by the camera dynamics (typically 100 Hz for 106 pixels). Beyond this approximately 100-fold increase in data throughput, LSFM also offers unique performance in terms of photodamage. Indeed, in this particular geometry, only the imaged plane is illuminated, which maximizes the photonic efficiency. This contrasts with confocal microscopy, in which the sample is evenly illuminated while only photons originating from the focal plane are used. As photodamage is mostly associated with non-linear processes, spreading the light dose across time and space, thus reducing the peak intensity, provides a further advantage compared with raster scanning techniques (see Figure 1.3). In the first modern LSFM instruments, described by Huisken et al. in 2004,1 a cylindrical lens was used to focus a laser beam and to form a static light sheet. The sample was then sequentially moved along the vertical axis

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

Fast Volumetric Imaging Using Light-sheet Microscopy

9

Figure 1.2.  Imaging scheme in light-sheet microscopy. Two objectives, oriented  perpendicularly, are used for the illumination and the detection.

Figure 1.3.  Advantages of light-sheet microscopy compared to confocal microscopy. To illustrate the difference between laser scanning confocal microscopy and lightsheet microscopy, the processes of illumination and detection are split. In confocal microscopy, a tightly focused laser beam is scanned across the sample, thereby exposing the specimen to high-intensity illumination, not only in the plane of interest but also above and below. A pinhole rejects much of the excited fluorescence and confines the image to the plane of interest. In light-sheet microscopy, the sample is illuminated side-on by a thin sheet of light. As the entire fluorescence signal is collected and imaged onto a CCD camera, this method minimizes the photonic load and thus considerably limits phototoxicity.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

10

Sébastien Wolf and Georges Debrégeas

to create a volumetric image of the tissue. It was then proposed to form the light sheet by rapidly scanning a laser beam across the sample.18 This so-called digitally scanned light-sheet fluorescence microscope (DSLM) offers several advantages: each line in the specimen is illuminated with the same intensity, which can be beneficial for quantitative imaging of large specimens; the microscope does not rely on apertures to shape the beam profile, which reduces optical aberrations; and the DSLM mitigates striping artifacts induced by scattering/absorption objects in the tissue and thus provides a better image quality. However, the use of a DSLM results in a higher laser peak power for a similar average illumination and thus greater photodamage. Before going into the technical details of the performance of light-sheet microscopy, let us take a brief historical look at its origins, because the foundation of this new imaging technique in fact goes back a long time.

1.3. Optical Principles of Light-sheet Microscopy 1.3.1. The Origins of LSFM In 1902, Siedentopf and Zsigmondy published an article in Annalen der Physik19 describing a new method for the optical measurement of the size of gold particles. They projected sunlight through a slit aperture in the focal plane of an observation objective. The light in the sample was thus projected orthogonally to the observation objective (Figure 1.4), and they collected the scattered photons. They called this technique ultramicroscopy owing to its ability to image objects smaller than the diffraction limit.

Figure 1.4.  Part of the precursor of the light-sheet microscope developed by Siedentopf and Zsigmondy with an upright microscope containing a specimen holder that appears to be mounted to its objective lens and orthogonal illumination at 90° from what appears to be an illuminating objective. Reproduced from H. Siedentopf and R. Zsigmondy, Uber Sichtbarmachung und Größenbestimmung ultramikoskopischer Teilchen, mit besonderer Anwendung auf Goldrubingläser, Ann. Phys., 1902, 315(1), 1–39. Copyright 1903 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

Fast Volumetric Imaging Using Light-sheet Microscopy

11

Although in 1925 Zsigmondy received the Nobel Prize for Chemistry in part for this invention, it remained confined to the domain of colloidal physics for a long time. In 1993, Voie, Burns and Spelman20 reintroduced this optical method, albeit in a completely different context, renaming it orthogonal-plane fluorescence optical sectioning (OPFOS). Spelman's group developed the OPFOS device and used it to optically section, for the first time, whole fluorophore-stained and cleared cochleas,20–22 stating: Another approach to optical sectioning involves the use of a planar illumination beam. Planar illumination has been used in other imaging modalities such as flow cytometry and flow visualization. The illumination beam is focused into a plane with a cylindrical lens, and aligned to be co-planar with the depth of field of the imaging detector. The beam is thus orthogonal to the imaging axis. The region where both the imaging system and laser illumination focus coincide is the system focal zone. In 1994, Stelzer's laboratory was trying to improve the axial resolution of confocal microscopy. They developed an oblique illuminating confocal microscope called a confocal theta microscope.23,24 Their 1995 paper24 cited Voie et al.’s work on OPFOS, and confocal theta microscopy appeared to lay the foundation for their subsequent version of a light-sheet microscope device called a selective- or single-plane illumination microscope (SPIM). In 2004, a paper by Stelzer's group1 demonstrated the usefulness of lightsheet fluorescence microscopy for investigating embryonic development. This paper also showed images of developing embryos of medaka (the small fish Oryzias latipes) and Drosophila melanogaster embryos and ganglion cells monitored for 17 h. The spatial resolution was 6 µm with a field of view of 1.5 × 0.9 mm. Although Voie and colleagues published several articles20–22 claiming that light-sheet microscopy was very efficient, it was Huisken et al.’s paper1 that triggered the rapid development of light-sheet microscopy among biologists. 1.3.2. Spatial Resolution in LSFM In this section, we provide a simple evaluation of the spatial resolution in LSFM and the way in which it depends on the optical characteristics of the illumination and detection optics. 1.3.2.1.  Basics of Gaussian Optics.  In light-sheet microscopy, a cylindrical or scanned spherical Gaussian beam is used to illuminate the sample (see Figure 1.5). In the focal zone, the laser spot exhibits a Gaussian intensity distribution. In the x–z plane, the intensity profile can be expressed as    

   

 2z 2  i( x, z ) i ( x, z 0)exp  2   w( x ) 

(1.1)

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

12

Sébastien Wolf and Georges Debrégeas

Figure 1.5.  Geometry of a Gaussian beam. Owing to this divergent profile of the illumination beam, light-sheet microscopy imposes a compromise between the axial resolution W and the field of view l.

where the beam width w(x) follows the equation of a hyperbole:     2



 x  w  ( x ) w0 1     x0 

(1.2)

   

At the waist x = 0, the intensity is maximum and the beam radius is w0 =  λ/πNA, where λ is the excitation wavelength and NA the numerical aperture of the illumination arm. Far from the waist, the beam diverges with an angle θ such that NA = n sin θ, where n is the refractive index. The focal region of the beam, called the Rayleigh range, is the region around the waist within which the beam radius is less than 2w0 . This region is bounded by ±x0, where x0 = πnw02/λ. Hence large NA values yield small waist radii but also a small Rayleigh range. 1.3.2.2.  Resolution of the Light Sheet.  In standard bright-field microscopy, the spatial resolution reflects the shape of the PSF (point-spread function) of the objective and the detector, i.e. the response of the optical system to a point-like object. The lateral resolution (in the focus plane), is given by rlat ∼ λ/NA, where λ is the emission wavelength and NA is the numerical aperture of the objective. Its axial resolution (along the optical axis) is raxial ∼ λ/NA2, and is thus always larger than the lateral resolution. In DSLM, the lateral resolution is set by the detection objective and is therefore similar to that of a classical bright-field microscope: rlat ∼ λ/NA det. Note, however, that this expression is true only as long as the pixel size of the detection camera remains smaller than rlat/2. In contrast to two-photon and confocal microscopy, in light sheet microscopy the

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

Fast Volumetric Imaging Using Light-sheet Microscopy

13

lateral and axial resolutions are decoupled: the latter is essentially controlled by the thickness of the light sheet. Using a high-NA illumination objective, it is possible to produce a very thin light sheet, whose axial resolution raxial = λ/NA ill is comparable to the lateral resolution. DSLM thus offers the unique possibility of producing isotropic resolution. However, such configurations have very limited fields of view. Indeed, as shown in the preceding section, a very thin Gaussian beam has a small Rayleigh length, i.e. the region over which the beam is highly focused is extremely narrow. 1.3.2.3.  Trade-off Between the Field of View and the Axial Resolution.  In lightsheet microscopy, any configuration therefore reflects a trade-off between the axial resolution and the accessible field of view. A small calculation can be done in order to evaluate the best accessible axial resolution within a certain field of view, i.e. up to a given distance l from the waist. This axial resolution, denoted W, is the width of the Gaussian beam used to form the light sheet at a distance l from the waist, and it reads [see eqn (1.2) and Figure 1.5].    



 W l tan 

   

 NA



(1.3)

where θ is the beam divergence angle, such that NA = n sin θ. In the small angle limit, we have    

W l

NA   n NA

(1.4)

   

Therefore, θ is minimal for NA  n / l and its value is then W  2 l / n .  More precisely, the exact minimization of W leads to W  2l /πn . This trade-off is illustrated in Figure 1.6, which shows, for each choice of NA, the associated axial resolution and accessible field of view.

1.4. Building the Improved LSFM: Pushing the Limits In this section, we present the most significant improvements that have been implemented to increase the spatial or temporal resolution, enhance the penetration depth, limit photobleaching or adapt the geometry to accommodate a specific specimen. 1.4.1. Increasing the Spatial Resolution Increasing the resolution is fundamental in order to probe biological structures at the subcellular scale. Many studies have demonstrated that LSFM permits the imaging of small objects such as mitochondria, microtubules,25

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

14

Sébastien Wolf and Georges Debrégeas

Figure 1.6.  Spatial resolution in light-sheet fluorescence microscopy. (a) The lateral and axial resolutions are decoupled. (b) Dependence of the axial resolution (raxial) and field of view (x0) with the NA of the illumination arm. A large NA offers better axial resolution, but limited field of view. Adapted with permission from Springer Nature: R. M. Power and J. Huisken, A guide to light-sheet fluorescence microscopy for multiscale imaging, Nat. Methods, 2017, 14, 360–373. Copyright © 2018 Macmillan Publishers Limited, part of Springer Nature.60

single fluorescently labeled DNA binding proteins,26 single molecules27 and chromosomes.2 As mentioned in Section 1.3, by combining high-NA objectives in both the illumination and detection arms, it is in principle possible to obtain a very high and isotropic spatial resolution, albeit within a rather limited field of view. In practice, however, such configurations are prohibited by the steric constraints associated with high-NA (and thus massive) objectives. Two different strategies have been proposed to circumvent this issue. Galland et al.27 used a single high-NA objective with a 45° micromirror (Figure 1.7a). Illumination and detection are thus performed through a unique objective. In a study by Gebhardt et al.,26 two vertically opposed objectives (NA 1.4 and 1.35) faced each other and a thin light sheet was obtained by a small 45° mirror (Figure 1.7b). Because of the divergence of Gaussian beams, it is impossible to maintain a sub-micrometer axial resolution over a field of view of tens of micrometers (see the preceding section). This limitation motivated the use of non-diffracting beams. Bessel beams display an invariant profile over the direction of propagation. Their cross-section is well described by a Bessel function, i.e. it contains a central spot and a series of annular rings of decreasing intensity. These extra rings carry a large fraction of the illumination power and are therefore highly detrimental in one-photon imaging, making the use of Bessel beams impracticable. However, in the two-photon regime, owing to the quadratic dependence of the fluorescence emission on the illumination intensity, their contribution to the fluorescence signal becomes negligible,

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

View Online

Figure 1.7.  (a) Scheme of the reflected light-sheet principle developed by Gebhardt et al. in 2013.26 A laser beam is focused by an objective to form a vertical light sheet that is reflected by a 45° atomic force microscopy cantilever next to a cell in a Petri dish. Fluorescence is detected by a second high-NA objective. (b) Schematic representation of the single-objective light sheet-microscope developed by Galland et al. in 2015.27 A light sheet is created by reflection from a 45° micromirror. The light sheet is projected from the detection objective. (c) Comparison of methods in LSFM. (A) Traditional Gaussian beam LSFM. (B) Bessel beam LSFM has a much narrower core. However, Bessel beams exhibit concentric side lobes that tend to degrade the axial resolution. (C and D) Bound optical lattices create periodic patterns of high modulation depth across the plane, greatly reducing the peak intensity and the phototoxicity in live-cell imaging. The square lattice in (C) optimizes the confinement of the excitation to the central plane, and the hexagonal lattice in (D) optimizes the axial resolution. Scale bars are 1.0 mm except for the xz cross-section of the overall PSF of the microscope (scale bar, 200 nm). Part (c) reprinted by permission from AAAS: B. C. Chen, W. R. Legant, K. Wang, L. Shao, D. E. Milkie, M. W. Davidson and E. Betzig. Lattice light-sheet microscopy: imaging molecules to embryos at high spatiotemporal resolution. Science, 2014, 346, 1257998. Copyright © 2014, Science.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

16

Sébastien Wolf and Georges Debrégeas

as demonstrated by Betzig's group.2 This approach allowed the authors to image mitotic chromosomes with a 0.3 µm axial resolution over a 40 µm field of view. However, certainly the most impressive use of Bessel beams was reported in 2014, when the same group described the development of lattice light-sheet microscopy.28 The trick here was to produce a linear array of visible Bessel beams, the period of which was adjusted to induce destructive interferences between the rings. This method yielded an ultrathin 2D optical lattice, which, after dithering, formed a uniformly thin light sheet and allowed diffraction-limited axial resolution over a field of view of  80 µm (see Figure 1.7c). They also demonstrated that with the lattice lightsheet microscope, photobleaching and phototoxicity were reduced by one to two orders of magnitude compared to those seen with a 1D scanned Bessel beam. Rather than using non-diffracting beams, other groups proposed the use of electroacoustic lenses in the illumination arm in order to sweep the location of the waist rapidly along the propagation axis. This approach allowed for the production of uniform, divergence-free light sheets with 465 nm resolution over 50 µm.29,30 1.4.2. High-speed Volumetric Imaging LSFM largely outperforms other fluorescence techniques in terms of the imaging speed for a given specimen section. To perform fast 3D imaging, one further needs to displace rapidly both the light sheet and the focal plane of the objective along the z-axis. This is generally done by simultaneously moving the objective with a piezo device and the light sheet with a scanning mirror.31 The motion of the objective is in practice limited in terms of speed and range by the piezoelectric devices. To circumvent this issue, Fahrbach et al.32 used a remote focusing approach in which an electrically tunable lens, placed along the detection path, allowed them to displace the focal plane optically without any mechanical motion, allowing them to record up 30 volumetric images per second (Figure 1.8). 1.4.3. Contrast Enhancement The parallelization of data collection provided by LSFM comes at a cost: non-ballistic fluorescent photons, i.e. those scattered by the tissue and collected by the camera, induce a diffuse background signal, which tends to degrade the image contrast. This issue may become detrimental for thick and strongly scattering specimens. Two strategies have therefore been developed with the aim of rejecting these photons: line confocal detection and structured illumination. Line confocal LSFM relies on the same principle as confocal imaging. The idea is to use an optical slit (the 2D equivalent of a pinhole) conjugated with the digitally scanned laser beam to block any non-ballistic photons

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

Fast Volumetric Imaging Using Light-sheet Microscopy

17

Figure 1.8.  Two different approaches to volumetric imaging in LSFM. Left: Volume image data can be obtained by scanning the light sheet through the sample and synchronizing the focal plane by repositioning the detection objective. Right: The synchronization of the focal plane can also be obtained by remote focusing, using an electrically tunable lens.

from impinging on the sensor. Silvestri et al.3 demonstrated how this  method allows for the recovery of high-contrast subcellular resolution in thick tissue, such as an entire clarified mouse cerebellum. One very clever implementation of the line confocal method has recently been introduced that exploits the rolling shutter mode of SCMOS cameras to produce a virtual slit. In this mode, a few lines of the camera are exposed at each instant, and this region “rolls” across the sensor at constant speed to eventually  a full frame. By continuously conjugating the beam and rolling shutter, the confocal effect can be obtained without the need for extra optical components. Another approach to reject background signals is to implement structured illumination, a standard technique of contrast enhancement in fluorescence microscopy. This technique consists of illuminating the sample with a sinusoidal grating and recording several images for different spatial phase shifts of the projected pattern. One can then computationally recover the in-focus image using these different recorded frames, while eliminating out-of-focus signals. Keller et al.18 produced striped illumination patterns by rapid temporal modulation of the laser intensity during scanning. This approach allowed for a 2–3-fold contrast enhancement in developing Drosophila embryos. However, this technique requires at least three frames to be recorded for each time step, and thus reduce the accessible frame rate by the same factor. 1.4.4. Imaging Deeper in Semi-transparent Samples The ability to image deep down in tissue is of paramount importance to a number of bioimaging applications. As discussed in the previous section, the scattering of fluorescent photons degrades the image contrast by adding a diffuse background to the in-focus image. Another issue lies in the

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

18

Sébastien Wolf and Georges Debrégeas

Figure 1.9.  (a) Comparison of imaging depth using two-photon LSFM, one-photon LSFM and two-photon scanning microscopy of a Drosophila embryo at 50 µm from embryo surface. (b) Quantitative analysis of the z depth penetration performance of the three imaging modalities. Two-photon digitally scanned light-sheet microscope (2P-DSLM), one-photon digitally scanned light-sheet microscope (1P-SPIM), two-photon point-scanning microscope (2P-PSM). Reprinted with permission from Springer Nature: T. V. Truong, W. Supatto, D. S. Koos, J. M. Choi and S. E. Fraser, Deep and fast live imaging with two-photon scanned light-sheet microscopy, Nat. Methods, 2011, 8, 757–760. Copyright © 2011, Springer Nature.

blurring of the light sheet itself, which effectively reduces the axial resolution. This problem motivated the recent development of two-photon lightsheet imaging, which uses a near-infrared pulsed laser. This alternative offers two advantages compared to standard one-photon DLSM. First, light scattering is significantly reduced in the near-infrared compared with the visible range, thus enhancing the penetration depth of the light sheet. Second, in this non-linear regime of excitation, the relative contribution of the scattered photons to the overall fluorescence signal is highly reduced, so that the axial resolution is preserved. This approach, pioneered by Fraser's group in 2011,33 was shown to provide a significant gain in axial resolution when imaging, e.g. Drosophila embryos (Figure 1.9). In the specific context of functional imaging, the use of a near-infrared source can also be beneficial as this wavelength range is outside most animals' visible spectrum, and thus does not interfere with the visual system.7 One important limitation of this technique, however, needs to be mentioned: inducing two-photon fluorescence in the LSFM geometry requires the use of high laser power, which results in higher photodamage compared with one-photon LSFM for similar signal-to-noise ratios.34 1.4.5. Light-sheet Microscopy with a Single Objective One of the major practical limitations of LSFM is that it necessitates optical access of the specimen along two orthogonal axes. This requirement restricts its use to relatively small systems: for instance, it is incompatible

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

Fast Volumetric Imaging Using Light-sheet Microscopy

19

with functional brain imaging in the mouse cortex. To circumvent this limitation, single-objective LSFM has been proposed in the last few years. Building on the highly inclined laminated optical sheet microscope invented by Tokunaga et al.,35 Dunsby36 developed a light-sheet fluorescence method, called oblique plane microscopy, that used a single high-NA objective both to produce an oblique light sheet in the sample and to collect the fluorescent photons. Because the light sheet does not align with the focal plane of the imaging system, another objective is used to rotate the imaging plane. This complex optical path results in a low efficiency in photon collection, and a subsequent limited accessible frame rate. Recently, Bouchard et al.37 proposed a novel single-objective light-sheet design called the swept, confocally aligned planar excitation microscope, in which the second objective is replaced by a scanning mirror. This new LSFM instrument offers a better acquisition rate, and was shown to permit the recording of neuronal activity in behaving mice.

1.5. Application: Light-sheet Imaging of Zebrafish Brain This section focuses on the application of LSFM in zebrafish neuroscience. It aims to illustrate how the unique performance of this new imaging method may shed a new light on brain-scale neuronal processing in the vertebrate brain. 1.5.1. Light-sheet-based Whole-brain Functional Imaging in Zebrafish Larvae In the last two decades, the zebrafish (Danio rerio) has emerged as an important model in circuit neuroscience. Native to Asia, this small vertebrate displays important physiological and genetic homologies with mammals. Its genome is fully sequenced and a large panoply of genetic tools is now available. The zebrafish was originally used in developmental biology because its main organs develop within the first day post-fertilization and this process can be easily monitored. At this early stage, the larva is indeed transluscent—only a few pigments are present on the skin—which permits exceptional optical access to the entire specimen. The identification of a mutation controlling pigment cell formation38 led to the design of the so-called nacre zebrafish line that lacks skin pigments, thus improving the advantages of zebrafish larvae for in vivo imaging. With the development of genetically encoded calcium indicators, this asset was further exploited to perform in vivo functional imaging.39–41 The zebrafish larval brain being quasi-transparent, small and compact [the 105 neurons of a 6 dpf (days post-fertilization) larva are contained within a volume of 2 × 0.5 × 0.3 mm], the entire brain can be optically monitored in a minimally invasive way.39–41 To increase the number of neurons that could

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

20

Sébastien Wolf and Georges Debrégeas

be simultaneously recorded, several strategies were developed, including high-speed random-access imaging using acousto-optical deflectors42–44 and simultaneous multi-point excitation.45 These methods still proved to be too slow to record the entire brain volume of zebrafish larvae with sufficient dynamics. A whole-brain map of neural activity therefore required the sequential recording of the various brain areas from different individuals and then the patching together of these different regions.14,46 Such a sequential approach was tedious and only yielded mean activity patterns: it did not allow for the probing of large-scale concerted neural activities spanning distant brain regions. This imaging limitation precluded the study of complex neural processing that involved extended neural circuits. In 2013, the first LSFM-based functional recordings were reported on zebrafish larvae expressing the genetically encoded calcium reporter GCaMP pan-neurally, generating activity time traces of almost 80 000 neurons at 0.7 Hz or 25 000 neurons at 4 Hz.47,48 This new technique offered the first wholebrain recording at cellular resolution in a vertebrate, using activity correlation analysis. With these new sets of data, rich structures in space and time were now accessible, which could be revealed in a straightforward way using correlation analysis (Figure 1.10). It also paved the way to the study of brainscale processing of complex sensorimotor tasks, as illustrated in the following section. 1.5.2. Whole-brain LSFM-based Functional Imaging to Study Sensorimotor Integration in the Vertebrate Brain Although it possesses ∼10 times fewer neurons that a human retina, a 6-day-old zebrafish larva exhibits a rich behavioral repertoire, including goal-directed navigation (spontaneous swimming along illumination or thermal gradients), optomotor response (swimming in response to a global visual flux in order to maintain a constant position) and hunting (paramecia capture). These behaviors involve the integration and processing of various sensory stimuli to trigger adequate motor responses. LSFM allows the monitoring of the entire brain, from visual to integrative and motor centers, and thus offers a unique window into the neural substrates of such complex sensorimotor behaviors. Although the animal needs to be tethered in agarose in order to perform brain imaging, it is possible to record motor outputs either by video monitoring tail or eye motion (after partially removing the agarose around the tail or eyes) or by extracellular recording of fictive swim bouts in paralyzed preparations. These recordings can then be used to produce, using a real-time feedback loop, a sensory environment with which the animal can virtually interact.49 These approaches have recently been used, in two independent studies, to reveal specialized hindbrain circuits that control spontaneous and phototactic (towards a light source) navigation. Dunn et al.50 used whole-brain LSFM to identify two bilaterally distributed neural assemblies whose activity drives the (left/right) orientation of successive swim bouts, and thus orchestrates

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

Fast Volumetric Imaging Using Light-sheet Microscopy

21

Figure 1.10.  Light-sheet microscopy in zebrafish neuroscience. (a) LSFM recording of a section of the brain volume of a zebrafish larva brain at 5 days post-fertilization. (b) Neural control of eye movements in larval zebrafish brain. Dorsoventral projection view of a 3D functional map showing neuronal populations whose activity is tuned to the orientation of the eyes (blue and red) and to the angular velocity of the eyes (green and yellow). Te, telencephalon; OT, optic tectum; Cb, cerebellum; Hb, hindbrain; RH, rhombomere. (c) Identifying populations of neurons correlated with the swim direction of zebrafish larvae. Dorsal, sagittal and coronal sections from a whole brain map. Scale bar, 100 µm. Part (a) reprinted from T. Panier, S. A. Romano, R. Olive, T. Pietri, G. Sumbre, R. Candelier and G. Debrégeas, Fast functional imaging of multiple brain regions in intact zebrafish larvae using selective plane illumination microscopy, Front. Neural Circuits, 2013, 7, 65. Copyright 2013 The Authors, published under the terms of the Creative Commons Attribution License.47 Part (b) reprinted with permission from Springer Nature: S. Wolf, A. M. Dubreuil, T. Bertoni, U. L. Böhm, V. Bormuth, R. Candelier, S. Karpenko, D. G. C. Hildebrand, I. H. Bianco, R. Monasson and G. Debrégeas, Sensorimotor computation underlying phototaxis in zebrafish, Nat. Commun., 2017, 8, 651. Copyright © 2017, Springer Nature. Published under the terms of the Creative Commons CC BY License. Part (c) reprinted with permission from T. W. Dunn, Y. Mu, S. Narayan, O. Randlett, E. A. Naumann, C.-T. Yang, A. F. Schier, J. Freeman, F. Engert and M. B. Ahrens, Brain-wide mapping of neural activity controlling zebrafish exploratory locomotion, eLife, 2016, 5, e12741.

the spatiotemporal pattern of spatial exploration (see Figure 1.10). Wolf et al.51 used two-photon LSFM to record brain activity when the zebrafish was exposed to visual stimulation. They were able to show that the same circuit is under partial control of visual stimuli, in such a way that the animal's trajectory is biased towards a light source. One of the strong assets of LSFM is to provide a straightforward way to map the entire circuit involved in a particular task. This can be achieved by simply correlating the activity of the entire brain with the motor or sensory signals. Such functional mapping approaches do not rely on pre-existing hypotheses regarding the involved brain regions, as the entire brain can be analyzed at once.

View Online

22

Sébastien Wolf and Georges Debrégeas

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

1.6. LSFM in the Next Decade Light-sheet fluorescent microscopy is a powerful technique that allows fast volumetric imaging at high spatial resolution with minimal phototoxicity. Since its “re-invention” about 10 years ago, LSFM has rapidly spread through various fields of biology, reflecting its extreme versatility. Various imaging methods, such as confocal filtering, two-photon excitation, structured illumination and super-resolution, have been combined with LSFM to increase its performance and expand the scope of its applications. This evolution is likely to continue in the coming years. Hence adaptive optics technology, which is becoming standard in epifluorescence microscopy,52–54 will likely be used to improve the quality of the light sheet by compensating for tissue-induced optical aberration and scattering. Similarly, dynamic control of the light intensity during recordings, using computational methods developed for confocal microscopes,55 might further limit photobleaching in LSFM. Efforts to miniaturize LSFM,56,58 using optical fibers,57 may eventually allow the design of light-sheet endoscopes. Finally, the impact of LSFM may be reinforced when combined with other techniques, such as optogenetics or laser ablation, in order to probe the response of a biological system to a controlled perturbation. Beyond the technical issues associated with the optical design, LSFM also raises important computational and theoretical challenges: how to process, analyze and navigate through the massive datasets produced by such a technique. This problem already constitutes a practical bottleneck in many applications, such as whole-brain calcium recording or high-resolution imaging of intact clarified brains. Quantifying, standardizing and automating these processing methods will likely become crucial requirements in the near future.

References 1. J. Huisken, J. Swoger, F. Del Bene, J. Wittbrodt and E. H. Stelzer, Science, 2004, 305(5686), 1007. 2. T. A. Planchon, L. Gao, D. E. Milkie, M. W. Davidson, J. A. Galbraith, C. G. Galbraith and E. Betzig, Nat. Methods, 2011, 8(5), 417. 3. L. Silvestri, A. Bria, L. Sacconi, G. Iannello and F. S. Pavone, Opt. Express, 2012, 20(18), 20582. 4. R. Tomer, K. Khairy, F. Amat and P. J. Keller, Nat. Methods, 2012, 9(7), 755. 5. M. Rauzi, U. Krzic, T. E. Saunders, M. Krajnc, P. Ziherl, L. Hufnagel and M. Leptin, Nat. Commun., 2015, 6, 8677. 6. F. C. Zanacchi, Z. Lavagnino, M. P. Donnorso, A. Del Bue, L. Furia, M. Faretta and A. Diaspro, Nat. Methods, 2011, 8(12), 1047. 7. S. Wolf, W. Supatto, G. Debrégeas, P. Mahou, S. G. Kruglik, J. M. Sintes, E. Beaurepaire and R. Candelier, Nat. Methods, 2015, 12(5), 379. 8. P. J. Keller, A. D. Schmidt, A. Santella, K. Khairy, Z. Bao, J. Wittbrodt and E. H. Stelzer, Nat. Methods, 2010, 7(8), 637. 9. M. Minsky, Scanning, 1988, 10(4), 128.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

Fast Volumetric Imaging Using Light-sheet Microscopy

23

10. P. Mahou, PhD thesis, Ecole Polytechnique X, 2012. 11. W. Denk, J. H. Strickler and W. W. Webb, Science, 1990, 248(4951), 73. 12. K. Svoboda, D. W. Tank and W. Denk, Science, 1996, 272(5262), 716. 13. J. T. Trachtenberg, B. E. Chen, G. W. Knott, G. Feng, J. R. Sanes, E. Welker and  K. Svoboda, Nature, 2002, 420(6917), 788. 14. M. B. Ahrens, J. M. Li, M. B. Orger, D. N. Robson, A. F. Schier, F. Engert and  R. Portugues, Nature, 2012, 485(7399), 471. 15. T. A. Schwickert, R. L. Lindquist, G. Shakhar, G. Livshits, D. Skokos, M. H. Kosco-  Vilbois, M. L. Dustin and M. C. Nussenzweig, Nature, 2007, 446(7131), 83. 16. J. M. Squirrell, D. L. Wokosin, J. G. White and B. D. Bavister, Nat. Biotechnol., 1999, 17(8), 763. 17. A. McMahon, W. Supatto, S. E. Fraser and A. Stathopoulos, Science, 2008, 322(5907), 1546. 18. P. J. Keller, A. D. Schmidt, J. Wittbrodt and E. H. Stelzer, Science, 2008, 322(5904), 1065. 19. H. Siedentopf and R. Zsigmondy, Ann. Phys., 1902, 315(1), 1. 20. A. H. Voie, D. H. Burns and F. A. Spelman, J. Microsc., 1993, 170(3), 229. 21. A. H. Voie and F. A. Spelman, Comput. Med. Imaging Graph., 1995, 19(5), 377. 22. A. H. Voie, Hear. Res., 2002, 171(1), 119. 23. S. Lindek, R. Pick and E. H. Stelzer, Rev. Sci. Instrum., 1994, 65(11), 3367. 24. E. H. K. Stelzer, S. Lindek, S. Albrecht, R. Pick, G. Ritter, N. J. Salmon and R. Stricker, J. Microsc., 1995, 179(1), 1. 25. B. Huang, S. A. Jones, B. Brandenburg and X. Zhuang, Nat. Methods, 2008, 5(12), 1047. 26. J. C. M. Gebhardt, D. M. Suter, R. Roy, Z. W. Zhao, A. R. Chapman, S. Basu, T. Maniatis and X. S. Xie, Nat. Methods, 2013, 10(5), 421. 27. R. Galland, G. Grenci, A. Aravind, V. Viasnoff, V. Studer and J. B. Sibarita, Nat. Methods, 2015, 12(7), 641. 28. B. C. Chen, W. R. Legant, K. Wang, L. Shao, D. E. Milkie, M. W. Davidson, C. Janetopoulos, X. S. Wu, J. A. Hammer 3rd, Z. Liu and B. P. English, et al., Science, 2014, 346(6208), 1257998. 29. K. M. Dean and R. Fiolka, Opt. Express, 2014, 22(21), 26141. 30. W. Zong, J. Zhao, X. Chen, Y. Lin, H. Ren, Y. Zhang, M. Fan, Z. Zhou, H. Cheng, Y. Sun and L. Chen, Cell Res., 2015, 25(2), 254. 31. T. F. Holekamp, D. Turaga and T. E. Holy, Neuron, 2008, 57(5), 661. 32. F. O. Fahrbach, V. Gurchenkov, K. Alessandri, P. Nassoy and A. Rohrbach, Opt. Express, 2013, 21(9), 11425. 33. T. V. Truong, W. Supatto, D. S. Koos, J. M. Choi and S. E. Fraser, Nat. Methods, 2011, 8(9), 757. 34. W. C. Lemon, S. R. Pulver, B. Höckendorf, K. McDole, K. Branson, J. Freeman and P. J. Keller, Nat. Commun., 2015, 6, 7924. 35. M. Tokunaga, N. Imamoto and K. Sakata-Sogawa, Nat. Methods, 2008, 5(2), 159. 36. C. Dunsby, Opt. Express, 2008, 16(25), 20306. 37. M. B. Bouchard, V. Voleti, C. S. Mendes, C. Lacefield, W. B. Grueber, R. S. Mann, R. M. Bruno and E. M. Hillman, Nat. Photonics, 2015, 9(2), 113. 38. J. A. Lister, C. P. Robertson, T. Lepage, S. L. Johnson and D. W. Raible, Development, 1999, 126(17), 3757. 39. M. S. Siegel and E. Y. Isacoff, Neuron, 1997, 19(4), 735. 40. T. W. Chen, T. J. Wardill, Y. Sun, S. R. Pulver, S. L. Renninger, A. Baohan, E. R. Schreiter, R. A. Kerr, M. B. Orger, V. Jayaraman and L. L. Looger, et al., Nature, 2013, 499(7458), 295.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00001

24

Sébastien Wolf and Georges Debrégeas

41. G. Cao, J. Platisa, V. A. Pieribone, D. Raccuglia, M. Kunst and M. N. Nitabach, Cell, 2013, 154(4), 904. 42. R. Salome, Y. Kremer, S. Dieudonne, J. F. Léger, O. Krichevsky, C. Wyart, D. Chatenay and L. Bourdieu, J. Neurosci. Methods, 2006, 154(1), 161. 43. G. Katona, G. Szalay, P. Maák, A. Kaszás, M. Veress, D. Hillier, B. Chiovini, E. S. Vizi, B. Roska and B. Rózsa, Nat. Methods, 2012, 9(2), 201. 44. B. F. Grewe, D. Langer, H. Kasper, B. M. Kampa and F. Helmchen, Nat. Methods, 2010, 7(5), 399. 45. S. Quirin, J. Jackson, D. S. Peterka and R. Yuste, Front. Neural Circuits, 2014, 8, 29. 46. R. Portugues, C. E. Feierstein, F. Engert and M. B. Orger, Neuron, 2014, 81(6), 1328. 47. T. Panier, S. A. Romano, R. Olive, T. Pietri, G. Sumbre, R. Candelier and G. Debrégeas, Front. Neural Circuits, 2013, 7, 65. 48. M. B. Ahrens, M. B. Orger, D. N. Robson, J. M. Li and P. J. Keller, Nat. Methods, 2013, 10(5), 413. 49. N. Vladimirov, Y. Mu, T. Kawashima, D. V. Bennett, C. T. Yang, L. L. Looger, P. J. Keller, J. Freeman and M. B. Ahrens, Nat. Methods, 2014, 11(9), 883. 50. T. W. Dunn, Y. Mu, S. Narayan, O. Randlett, E. A. Naumann, C. T. Yang, A. F. Schier, J. Freeman, F. Engert and M. B. Ahrens, eLife, 2016, 5, e12741. 51. S. Wolf, A. M. Dubreuil, T. Bertoni, U. L. Böhm, V. Bormuth, R. Candelier, S. Karpenko, D. G. C. Hildebrand, I. H. Bianco, R. Monasson and G. Debrégeas, Nat. Commun., 2017, 8(1), 651. 52. N. Ji, D. E. Milkie and E. Betzig, Nat. Methods, 2010, 7(2), 141. 53. N. Ji, Nat. Methods, 2017, 14(4), 374. 54. Y. Kremer, J. F. Léger, R. Lapole, N. Honnorat, Y. Candela, S. Dieudonné and L. Bourdieu, Opt. Express, 2008, 16(14), 10066. 55. C. Conrad, A. Wünsche, T. H. Tan, J. Bulkescher, F. Sieckmann, F. Verissimo, A. Edelstein, T. Walter, U. Liebel, R. Pepperkok and J. Ellenberg, Nat. Methods, 2011, 8(3), 246. 56. P. Paiè, F. Bragheri, A. Bassi and R. Osellame, Lab Chip, 2016, 16(9), 1556. 57. M. Plöschner, V. Kollárová, Z. Dostál, J. Nylk, T. Barton-Owen, D. E. Ferrier, R. Chmelík, K. Dholakia and T. Čižmár, Sci. Rep., 2015, 5, 18050. 58. C. J. Engelbrecht, F. Voigt and F. Helmchen, Opt. Lett., 2010, 35(9), 1413. 59. M. Mickoleit, B. Schmid, M. Weber, F. O. Fahrbach, S. Hombach, S. Reischauer and J. Huisken, Nat. Methods, 2014, 11(9), 919. 60. R. M. Power and J. Huisken, A guide to light-sheet fluorescence microscopy for multiscale imaging, Nat. Methods, 2017, 14(4), 360.

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

Chapter 2

Super-resolution Microscopy Xiyu Yia, Tal-Zvi Markusb, Xavier Michaleta, Shimon Weissa and David Bensimon*a,b a

Department of Chemistry and Biochemistry, UCLA, Los Angeles, CA, 90095, USA; bEcole Normale Supérieure, Laboratoire de Physique   Statistique, UMR 8550 CNRS, 24 rue Lhomond, 75231, Paris, France *E-mail: [email protected]

Table of Contents 2.1.  The Diffraction Limit. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.  Super-resolution via Localization Microscopy. . . . . . . . . . . . . . . . . . 2.3.  Single-molecule Localization in 3D. . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.  Super-resolution via Correlation of Fluorescence Fluctuations. . . . 2.5.  Super-resolution via Optical Non-linear Effects. . . . . . . . . . . . . . . . . 2.6.  Super-resolution Based on Structured Illumination. . . . . . . . . . . . . 2.7.  Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

25

27 27 31 32 34 36 38 38 39

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

View Online

         

View Online

Super-resolution Microscopy

27

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

2.1. The Diffraction Limit In a regular microscope, the fluorescence of molecules is excited by light of wavelength λex focused by an objective lens characterized by its numerical aperture (NA). Light emitted by these fluorophores (at wavelength λem > λex) is collected by the same objective lens and imaged by the tube lens as a spot (the point spread function (PSF) of the microscope) of radius rAiry = 0.61λem/ NA (see Figure 2.1). Two point sources separated by a distance d significantly smaller than rAiry cannot be resolved easily,1 as their PSFs merge into a single spot (the so-called Rayleigh criterion). This diffraction effect has defined the resolution limit of microscopes for more than a century and has been one of the motivations for the development of electron microscopy. According to the Rayleigh criterion, the classical resolution limit of a good microscope objective (i.e. with NA = 1.4) is 0.44λem. Thus, when observing cellular structures labeled with fluorophores such as the green fluorescent protein (GFP), emitting at λem = 510 nm, details below 200 nm cannot be resolved. In the past 10 years, methods have been developed that overcome this constraint and allow for almost unlimited resolution (i.e. super-resolution) in optical microscopy.

2.2. Super-resolution via Localization Microscopy While photons emitted by a single fluorophore are spread by diffraction over a spot of radius rAiry, the determination of the spot's intensity maximum (which is also its center due to symmetry) can be determined with unlimited precision. Indeed, if N photons are detected [at positions r⃑i = (xi,yi); i = 1, …, N], the central limit theorem tells us that the variance σr̄  2 of their mean position r̄ = =∑Nr⃑i/N is equal to the variance of their position σr  2 divided by N†. Since σr = 0.21λem/NA,2 we have    



r 

   

0.21em NA N

(2.1)

Therefore, to a first approximation, the localization of an isolated source of photons can be determined with arbitrary precision (σr̄) provided that enough photons are detected.3 In far-field fluorescence imaging, this property was first used for single-molecule tracking in membranes4 and gels,5 with resolution between 30 and 100 nm, slightly greater than that expected for immobile molecules, but as expected due to Brownian motion.6 Unsurprisingly, better resolution can be obtained for immobile molecules, which can be used to measure sub-diffraction distances between distinct molecules, provided that each molecule's fluorescence can be separated from that of others, as first suggested by Betzig.7 This was first demonstrated using molecules †

 ote that if one approximates the PSF by a Gaussian of variance σr2, the Rayleigh criterion for N resolving two point sources becomes d > 2.76σr ≈ rAiry.

Published on 11 September 2018 on https://pubs.rsc.org |

28 Xiyu Yi, Tal-Zvi Markus, Xavier Michalet, SHIMON WEISS

Figure 2.1.  Image of point sources in an optical microscope. (a) Image (top row) and intensity profile (bottom row), in linear scale (left) or logarithmic scale (right), of a point source emitting light at lem = 690 nm, observed with an objective lens with NA = 1.4 (magnification M = 1). Log scale, four decades; image size, 1 µm. The distance from the center of the image to the first intensity minimum defines the radius of the so-called Airy disk (rAiry = 0.61λem/NA), while the full width at half maximum (FWHM) = 0.84rAiry. (b) Two point sources separated by a distance dRayleigh = rAiry (Rayleigh criterion) can still (barely) be resolved, thanks to a dip (Δ = 23%) of the intensity between the sources. (c) Two point sources separated by a distance d = 0.7dRayleigh cannot be resolved. Adapted from Figure 1 in X. Michalet and S. Weiss, Using photon statistics to boost microscopy resolution, Proc. Natl. Acad. Sci. U. S. A., 2006, 103(13), 4797–4798. Copyright 2006 National Academy of Sciences, USA.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

Super-resolution Microscopy

29

with different absorption spectra,8 and later using molecules with different emission spectra.9 Because the accurate distance resolution equation also involves pixel size and camera readout noise,3,10 achieving true nanometer resolution took a few more years, and was demonstrated, under the acronym FIONA11 (fluorescence imaging with one nanometer accuracy), by Selvin's group, who observed the movement of molecular motors labeled with a single fluorophore. They were therefore able to study the mechanism and step size of molecular motors such as myosin V11 and kinesin12 in vitro and later extended the approach in live cells, studying the dynamics and step size of kinesin and dynein motors moving along microtubules,13 while transporting small, dense vesicles (appearing as diffraction-limited dark spots in regular, non-fluorescent phase contrast microscopy). A similar type of single-particle tracking (SPT) with nanometer resolution has been used to monitor the diffusion on the cell membrane of proteins labeled with quantum dots.14,15 This approach was generalized to allow the imaging of samples consisting of more than a few isolated molecules, using fluorophores that can be stochastically turned on and off. They are switched on by a low-intensity illumination, so that only a very small number of molecules fluoresce, guaranteeing that the observed fluorescent spots are due to isolated single molecules. The fluorophores are then rapidly bleached (or turned off), and this on–off cycle is repeated with the same or other fluorophores, eventually to generate a pointillist image of the features they label (see Figure 2.2). Since the 2D localization of each fluorophore (observed to emit N photons) is determined with a standard deviation σr̄ given approximately by eqn (2.1), the 2D resolution of the features imaged in this manner is improved by a factor N compared with standard diffraction-limited microscopy (3D localization schemes have also been developed; see later). This technique, commonly referred to as single-molecule localization microscopy (SMLM), is illustrated in Figure 2.2. It has been implemented by different groups under different acronyms: PALM (photoactivated localization microscopy16), F-PALM (fluorescence photoactivation localization microscopy17) and STORM (stochastic optical reconstruction microscopy18). The concept of SMLM was initially proposed by Betzig7 in 1995, and demonstrated by Betzig et al.19 in 2006. The technique took advantage of the development of photoactivatable fluorescent proteins (PA-FPs), such as Dronpa, Kaede and Eos, which can be switched from a dark, non-fluorescent, metastable-state into a fluorescent state by illumination at an appropriate wavelength (405 nm). Cellular features (e.g. filaments of actin or microtubules) were labeled with these PA-FPs, which were then switched to the fluorescent state by illumination with low-intensity UV light (the activation beam), in order to achieve stochastic activation with a low probability (i.e. to obtain much less than one activated fluorophore per diffraction-limited spot covering a few pixels). A second strong illumination source (the excitation beam) was then used to excite the fluorescence of the activated PA-FPs and bleach them rapidly. The position of these fluorophores was recorded with a precision set approximately by eqn (2.1) (with N = the mean number of photons

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

30

Xiyu Yi, Tal-Zvi Markus, Xavier Michalet, SHIMON WEISS

Figure 2.2.  Principle of single molecule localization microscopy (SMLM). The feature of interest shown in (a) (e.g. microtubules) is first fluorescently labeled with a high enough density of fluorophores. Owing to diffraction, the image of individual fluorophores will be spread over a spot of size rAiry to yield the conventional diffraction-limited image shown in (b). In SMLM, fluorophores are turned on with low probability so that only a small proportion of the fluorophores are active at any given time. Different emitters will be on at different times and localized in different frames of the acquired movie, as shown in the upper panel of (c). Each fluorophore is identified and its location computed by a Gaussian fit of the intensity profile detected by a CCD camera, as illustrated in the lower panel in (c). The locations determined for the various fluorophores in successive frames are superimposed to generate the final pointillist image shown in (d), which displays a resolution improved by a factor N .  The SMLM approach allows the resolution of features that are too small to be distinguished by conventional microscopy, as illustrated in the area marked with a red circle in panels (b) and (d).

detected before the proteins bleached) and the procedure was repeated with another set of activated PA-FPs. A pointillist image of the labeled feature was obtained after many hundreds of such cycles of activation followed by excitation and bleaching. A resolution of a few tens of nanometers (set by the density of labeling and the resolution of individual spots) could be achieved at the expense of a long acquisition time on fixed samples19 (a few hours to obtain about 106 localized spots). STORM was developed by Zhuang and co-workers20 at about the same time. Rather than labeling cellular features with PA-FPs, they labeled them with photoswitchable dyes (Cy5, Cy7, Alexa Fluor 647), which could be reversibly switched between fluorescent and non-fluorescent states upon appropriate illumination, and provided they were in the vicinity of a cyanine dye emitting at lower wavelength (Cy3). The photoswitchable dye is turned on by illuminating the sample with a low-intensity activation beam at 532 nm to achieve a low density of active fluorescent dyes (as in PALM). These are then illuminated with an excitation beam at 633 nm, which has the dual purpose of exciting the fluorescent emission of the dye and later turning it off (after an average of 6000 detected photons). The same dyes (or others) can then be reactivated by illumination at the lower wavelength. These photoswitchable

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

Super-resolution Microscopy

31

dyes can be turned on and off hundreds of times before they bleach, thus yielding many thousands of detected photons, which is a significant advantage over PALM (which relies on photobleaching of PA-FPs, and therefore yields only a few hundred detected photons per molecule). On the other hand, it requires high-density labeling of cellular structures with two dyes in close proximity (for example, Cy5 and Cy3). This can only be achieved on permeabilized, fixed samples, whereas PALM can be implemented in live cells (by genetically fusing the PA-FPs with the desired protein). A significant improvement to STORM was made by Sauer and co-workers,21 who observed that many fluorophores can be switched reversibly between a fluorescent and a dark state without the requirement for a nearby emitting dye. In this direct STORM (dSTORM) implementation, as in regular STORM, an emitting dye (such as Alexa Fluor 647) stochastically switches from an emitting state into a dark state. However, using a laser intensity 200-fold stronger than that used for an activator dye in STORM, the emitting dye can be switched back from its dark state into the active state. The switching rates into and from the dark state depend linearly on the intensity of the laser beams and can therefore be easily controlled with intensities that are not detrimental to the cell. The advantage of dSTORM is that regular labeling with a single type of fluorophore only is needed to achieve super-resolution images by SMLM.

2.3. Single-molecule Localization in 3D In a good, aberration-free microscope, the PSF is azimuthally symmetric and symmetric under reflection (z → −z) about the focal plane (z = 0). Thus, although the previous single-molecule localization methods are capable of improving the lateral (xy) resolution, they did not provide any improvement in axial (z) resolution, which for a confocal microscope is dz = 2nλem/ NA2 (≈1.4λem for NA = 1.4), where n is the refractive index in the object space (generally water: n = 1.33). To improve on that limit, controlled aberrations were introduced in a microscope to generate a PSF that varies with the axial distance z. One of the simplest ways to achieve this is to introduce a cylindrical lens in the emission path,18 so that the image of a single fluorophore in any focal plane is not a circular but an elliptical spot. From the spot's aspect ratio, the axial distance z of the source can thus be accurately determined. One of the recent improvements to the 3D localization of single fluorophores came from engineering a double-helix PSF.22 By introducing an appropriate phase mask generated by a spatial light modulator (SLM) in the emission path, the image of a point source can be split into two spots whose orientation in the image plane varies with the distance of the object to the focal plane23 (see Figure 2.3). Using this engineered PSF, a single Alexa Fluor 647 fluorophore coupled with an Alexa Fluor 488 activator dye was imaged in a STORM setup. Its position (x, y, z) detected with 5000 photons could be determined with a precision (σx, σy, σz) = (7, 11, 22) nm23.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

32

Xiyu Yi, Tal-Zvi Markus, Xavier Michalet, SHIMON WEISS

Figure 2.3.  An engineered 3D point spread function (SPINDLE). (a) The phase mask shown here encoding a double-helix PSF is positioned in a plane conjugated to the pupil plane of the objective using 4f relay optics.23 (b) The image of a point source is split and rotated in the image plane as the focal plane is moved along the optical axis (z) by the amounts shown.

2.4. Super-resolution via Correlation of Fluorescence Fluctuations Another super-resolution method based on the existence (and development) of blinking dyes was proposed by Enderlein and co-workers24 and is known under the acronym SOFI (stochastic optical fluctuation imaging, see Figure 2.4). It consists in computing the nth-order cumulant of the intensity fluctuations over time, δI(x, y). The idea is that, for an ensemble of N fluctuating fluorophores (with intensity fluctuations δIi(t), i = 1, …, N) located at positions {r⃗i}, the intensity fluctuation of the image is    



 I  r ,t 



 

N

U  r  r  I (t ) i

i 1

   

i

(2.2)

where U(r⃑) is the PSF of the microscope [which, as discussed in section 2.1, spreads the point source over a spot of radius rAiry, usually modeled by a Gaussian: U(r⃑) ∼ exp(−r⃑2/2σr2)]. Since the emission of the fluorescent sources are uncorrelated, the two-point cumulant    

  C2  r 







 d  I  r ,t   r ,t    0





N    U 2  r  ri   d  Ii (t ) Ii (t   ) i 1

   

(2.3)

0

The second-order cumulant varies as the square of the PSF. Similar considerations show that the nth-order cumulant varies as the nth power of the PSF and therefore has a standard deviation that is n times smaller than the PSF.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

Super-resolution Microscopy

33

Figure 2.4.  SOFI principle. This approach can improve the resolution of the image of a sample containing multiple independent fluctuating emitters, such as the linear array of blinking fluorophores at positions {ri} simulated and shown in (a). If a movie of these blinking emitters is acquired as shown in (b), each pixel exhibits time-dependent intensity fluctuations as shown in (c). These intensity trajectories result from the fluorescent signals contributed to a given pixel by multiple emitters. A time average of these trajectories results in the low-resolution image shown in (d), where two adjacent emitters are not always resolved. A SOFI algorithm computes the nth-order cumulant of the time intensity trajectories at each pixel, yielding SOFI images of order n. As shown in (e), where AC2, AC3 and AC4 displays the reconstructed images of second-, third- and fourth-order SOFI, the resolution increases as the cumulant order increases. The resolution enhancement is illustrated in (f), where a cross-section plot at each order of SOFI is shown (the curves have been shifted vertically for legibility).

Moreover, since the optical transfer function (OTF) (the Fourier transform ⃗ is the convolution of n OTFs of the of the PSF) of a SOFI image [OTFSOFI(k)] microscope's PSF [Ũ(k), with a finite support k < k0 = 2π/rAiry]:    



    OTFSOFI k  U k  U k  U k ( n terms)

 

 

 

 

(2.4)

   

the OTF of a SOFI image is defined over a larger support of order O(nk0). Consequently, deconvolution of a SOFI image with OTFSOFI yields an image with a resolution increased by a factor n.24 Originally, to support this increased resolution, cross-cumulants were used to virtually increase the number of pixels.24 More recently, a simpler implementation known as f-SOFI25 has been proposed – it consists in interpolating the image between pixels by increasing the number of Fourier modes by a factor of (at least) n2. Since the Fourier spectrum of the image has

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

34

Xiyu Yi, Tal-Zvi Markus, Xavier Michalet, SHIMON WEISS

a finite support (it cannot contain modes higher than k0), one can pad the higher Fourier modes with zeros, up to values of (at least) nk0, without altering the information content of the image. Transforming back this spectrally increased image generates an image with n times smaller pixels. Computing the nth-order time cumulant of the resulting movie yields a movie in which each frame has a resolution increased by a factor of n , or of n after deconvolution by the SOFI OTF. The great advantage of SOFI over other super-resolution schemes is that it is a purely numerical scheme, which handles movies of blinking fluorophores obtained by any fluorescent microscope. It does not require sophisticated optical setups [in contrast to stimulated emission depletion (STED) or structured illumination microscopy (SIM); see later] and, even at order 2, significantly increases contrast by eliminating the contribution of out-of-focus fluorescent sources (uncorrelated background noise). Because it treats the whole image simultaneously, it is also advantageous over SMLM schemes (such as PALM and STORM).26 Its main drawbacks are artifacts due to the presence of fluorescent sources with various intensities (an imbalance that becomes amplified at higher orders) and the increased time required for computing high-order cumulants with reasonable accuracy.

2.5. Super-resolution via Optical Non-linear Effects Before these stochastic localization methods were put into practice, a revolutionary modification of confocal imaging microscopy was proposed by Hell and co-workers27 in order to achieve optical imaging with arbitrary resolution.27,28 In this approach, known as stimulated emission depletion (STED) microscopy, a doughnut-shaped beam (the STED beam), aligned with the Gaussian excitation beam, but using a slightly larger wavelength, is used to deplete the excited state via stimulated emission. This emission has a strong non-linear dependence on the intensity of the depletion beam. As a result, fluorescence emission is limited to a narrow region near the center of the depletion beam where its intensity is close to zero. However, to deplete the excited state by stimulated emission, a high-intensity STED beam (∼106 kW cm−2) is required, which can be detrimental to live cell imaging. To overcome this drawback, a variation on the method, known as reversible saturable optical fluorescence transitions (RESOLFT), was proposed.29 It uses the STED beam to reversibly convert switchable fluorescent proteins (RS-FPs) from their excited fluorescent state into their dark state.29 The dependence of the emitted fluorescence can be modeled after that of a saturable  absorber:    

   

I em 

max I em

1  I STED / I sat



(2.5)

View Online

Super-resolution Microscopy

35

with ISTED ≫ Isat. Hence, except at the center of the doughnut-shaped beam, fluorescence emission is negligibly low. However, near the beam center and 0 2 to leading order, I STED  I STED . Hence the extent of the emission spot: r 2 / rAiry

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

   



0 r  rAiry I sat / I STED  rAiry

(2.6)

   

can be set to dimensions that are much smaller than the diffraction limit (a few tens of nanometers) (see Figure 2.5). With the RESOLFT implementation of Hell and co-workers' original idea, intensities as low as 1 kW cm−2 were used to image live dendritic spines30 labeled with an RS-FP (Dronpa-M159T). The images were characterized by a lateral resolution of 65 nm, an axial resolution of 150 nm (using a 3D doughnut beam obtained with an appropriate phase mask) and a dwell time per pixel of only 0.5 ms.

Figure 2.5.  (a) Stimulated emission depletion (STED) microscopy is a super-resolution microscopy approach that breaks the diffraction limit by relying on the strong non-linearity of stimulated emission (or saturable absorption29). Two colinear beams combined with appropriate dichroic mirrors (BS) are used to illuminate the sample via the objective lens (O): a Gaussian-shaped excitation beam (the intensity of which is maximum at the center) and a doughnut-shaped stimulation (STED) beam (the intensity of which is zero at the center and which can be obtained from a Gaussian beam with an appropriate phase plate). Owing to the non-linearity of the stimulated emission, the stronger the STED beam, the more the excited electrons are driven to the ground state S0 by the stimulated beam, except in the narrower region where the intensity of the doughnut beam is zero, see (b). Hence the region where spontaneous (fluorescence) emission is possible becomes progressively smaller as the intensity of the STED beam is increased, allowing for theoretically unlimited optical resolution (in practice, resolution down to a few tens of nanometers has been achieved). Adapted from Figure 1 in A. Honigmann, C. Eggeling, M. Schulze and A. Lepert, BIOPHOTONICS: Super-resolution STED microscopy advances with yellow CW OPSL, Laser Focus World, 2012, 48, 75.31

View Online

36

Xiyu Yi, Tal-Zvi Markus, Xavier Michalet, SHIMON WEISS

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

2.6. Super-resolution Based on Structured Illumination Another approach for super-resolution is based on the very old observation of moiré patterns. These patterns were first observed when two fine silk sheets were brought into contact, creating the visual effect of a ripple structure. These visual effects are created by a combination of textures with high spatial frequency (not resolvable by the naked eye; in the case of silk sheets, the high density of threads) that generate a pattern with lower observable spatial frequency. The same effect is observed in movies, when the speed of the film (the number of frames per second) combines with the rotation speed of a recorded wheel to create the illusion that the spokes rotate slowly or even in the opposite direction to the movement of the vehicle. This effect is simply due to the fact that the beat between two high-frequency modes (k1, k2) creates a low-frequency mode (k1 − k2):    



sin k1 x sin k 2x

1 cos  k1  k2  x   cos  k1  k2  x  2





(2.7)

   

Based on this common experience, in 2000 Gustafsson32 suggested a way to increase the resolution of a regular microscope by a factor of two. The idea is to excite the fluorescent sample with an illumination modulated at a spatial frequency k0 = 2π/rAiry. For example, I = I0(1 + cos k0x) (higher frequency modulations are not possible owing to the finite extent of the PSF, which acts as a low-pass filter with a cut-off k0). The observed fluorescent image will be a product of the density of fluorophores fsource(x, y) and of excitation intensity low-pass filtered by the PSF. Low-frequency (k < k0) modes will now contain information from higher frequencies. Let the spatial frequency distribution (or Fourier spectrum) of the fluorescent sources be    

   

1  ik y dxdyfsource ( x, y )eikx x e y fsource  kx , k y   2 0

(2.8)

Then the frequency spectrum of the image is the convolution of f̃source(kx,ky) with the spectrum of the illumination (see Figure 2.6e):

   

  fsource  k x  k0 , k y    1       (2.9) fimage  k x , k y  I 0  fsource  k x , k y     2    f k  k , k     source x 0 y      

where, because of the filtering effect of the microscope PSF, only spatial frequencies (k′x, k′y) below k0 can be observed in the image plane, i.e. kx 2  k y 2  k02.  Hence the image frequency spectrum contains modes with frequencies kx2 + ky2 ≤ k02 (as in regular fluorescence microscopy), but also modes with frequencies (kx − k0)2 + ky2 ≤ k02, which may result from spatial frequencies of the source kx, which are as large as kx = 2k0. These modes can be computed and,

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

Super-resolution Microscopy

37

Figure 2.6.  Structural illumination microscopy. When the fluorescent sample shown in (a) is illuminated with the sinusoidal pattern shown in (b), the resulting image is the product of the two patterns, as shown in (c). In the spatial frequency domain, the spectrum of the observation (f) is the convolution of the frequency spectrum of the sample (d) with the spectrum of the illumination pattern (e). Because the spectrum of the image is low-pass filtered by the aperture of the detection optics (with cut-off frequency k0 = 2π/rAiry), the observation is limited to frequencies |k| ≤ k0 inside the yellow circle. However, some high-frequency components (inside the blue circles) are shifted to, and overlaid with, the low-frequency region (f), thus encoding information with higher resolution. Multiple patterns with different orientation of the modulation (g) can be captured, from which one can extract information on the high-frequency modes (h) with k0 < |k| ≤ 2k0, thus allowing for an increase in the resolution of the reconstructed image (i) by a factor of two.

from images obtained with illumination modulated in different directions, an image with twice the resolution (i.e. modes with k = 2k0) can be reconstructed (see Figure 2.6i). A 3D version of that scheme (3D SIM) has been developed by Gustafsson and co-workers,33 in which a 3D pattern of interference is created in the focal plane of the objective using three beams with appropriate phases generated

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

38

Xiyu Yi, Tal-Zvi Markus, Xavier Michalet, SHIMON WEISS

by an SLM.33 From any such patterns (3 lateral orientations × 5 phases = 15 patterns), a 3D image of the sample can be constructed with twice the lateral and axial resolution of a regular microscope (and a much better rejection of out-of-focus light, i.e. increase in contrast). Three-dimensional super-resolved images of a live HeLa cell (about 50 × 50 × 8 µm) were obtained in 21 s and dynamic phenomena occurring within the cell (e.g. mitochondrial movements) could be monitored over 37 min.

2.7. Conclusion The various super-resolution methods reviewed here and their variants have rapidly moved from the development and proof-of-concept stage to maturity and prime-time applications. As with the introduction of any novel imaging technology, a wave of new discoveries followed, as had been originally the case for the optical microscope: the discovery of microorganisms, bacteria and cells followed its invention in the seventeenth century by van Leeuwenhoek and Hooke. Similarly, the discovery of the structure of viruses and the nature of chromosomes, to name just a few, followed the development of the electron microscope in the 1930s. Indeed, super-resolution methods have already contributed many new discoveries, such as the different structure of actin filaments in axons (where they form equidistant rings) and in dendrites (where they form long filaments)34 and the dynamics of synaptic vesicles.35 We expect super-resolution to contribute similarly to many more new discoveries highlighting the inner working of the living cell. At the same time, the development of super-resolution techniques continues to evolve and improved methods are constantly being introduced. Interesting areas to monitor include the following:     ●● The development of cryogenic optical super-resolution methods that overcome fixation artifacts and allow for a significant reduction of photobleaching. Such methods therefore allow molecular-level (1 nm) resolution with chemical specificity and co-localization of multiple species (not afforded by electron microscopy). ●● The development of non-fluorescent-based optical super-resolution methods for applications in chemistry, materials science, etc. ●● The combination of optical super-resolution with scanning probe microscopy, electron microscopy, or X-ray microscopy techniques with various correlative capabilities.    

Acknowledgements D.B. and S.W. acknowledge support by the Partner University Fund, a program of the French American Culture Exchange. D.B. acknowledges the support of an ITMO Cancer grant “PhotoCancer” within the Plan Cancer 2014–2019

View Online

Super-resolution Microscopy

39

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

and the support of Structuration de la Recherche PSL grants: Microgut and SuperLine (ANR-10-IDEX-0001-02 PSL).

References 1. X. Michalet and S. Weiss, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 4797. 2. B. Zhang, J. Zerubia and J. C. Olivo-Marin, Appl. Opt., 2007, 46, 1819. 3. R. E. Thompson, D. R. Larson and W. W. Webb, Biophys. J., 2002, 82, 2775. 4. T. Schmidt, G. J. Schutz, W. Baumgartner, H. J. Gruber and H. Schindler, Proc. Natl. Acad. Sci. U. S. A., 1996, 93, 2926. 5. R. M. Dickson, D. J. Norris, Y. L. Tzeng and W. E. Moerner, Science, 1996, 274, 966. 6. T. Savin and P. S. Doyle, Biophys. J., 2005, 88, 623. 7. E. Betzig, Opt. Lett., 1995, 20, 237. 8. A. M. van Oijen, J. Kohler, J. Schmidt, M. Muller and G. Brakenhoff, J. Chem. Phys. Lett., 1998, 292, 183. 9. T. D. Lacoste, X. Michalet, F. Pinaud, D. S. Chemla, A. P. Alivisatos and S. Weiss, Proc. Natl. Acad. Sci. U. S. A., 2000, 97, 9461. 10. S. Ram, E. S. Ward and R. J. Ober, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 4457. 11. A. Yildiz, J. N. Forkey, S. A. McKinney, T. Ha, Y. E. Goldman and P. R. Selvin, Science, 2003, 300, 2061. 12. A. Yildiz, M. Tomishige, R. D. Vale and P. R. Selvin, Science, 2004, 303, 676. 13. C. Kural, H. Kim, S. Syed, G. Goshima, V. I. Gelfand and P. R. Selvin, Science, 2005, 308, 1469. 14. M. Dahan, S. Lévi, C. Luccardini, P. Rostaing, B. Riveau and A. Triller, Science, 2003, 302, 442. 15. F. Pinaud, X. Michalet, G. Iyer, E. Margeat, H.-P. Moore and S. Weiss, Traffic, 2009, 10, 691. 16. H. Shroff, C. G. Galbraith, J. A. Galbraith and E. Betzig, Nat. Methods, 2008, 5, 417. 17. S. T. Hess, T. P. Girirajan and M. D. Mason, Biophys. J., 2006, 91, 4258. 18. B. Huang, W. Wang, M. Bates and X. Zhuang, Science, 2008, 319, 810. 19. E. Betzig, G. H. Patterson, R. Sougrat, O. W. Lindwasser, S. Olenych and J. S. Bonifacino, et al., Science, 2006, 313, 1642. 20. M. J. Rust, M. Bates and X. Zhuang, Nat. Methods, 2006, 3, 793. 21. M. Heilemann, S. van de Linde, M. Schüttpelz, R. Kasper, B. Seefeldt and A. Mukherjee, et al., Angew. Chem. Int. Ed., 2008, 47, 6172. 22. S. R. P. Pavani, M. A. Thompson, J. S. Biteen, S. J. Lord, N. Liu and R. J. Twieg,  et al., Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 2995. 23. G. Grover, K. DeLuca, S. Quirin, J. DeLuca and R. Piestun, Opt. Express, 2012, 20, 26681. 24. T. Dertinger, R. Colyer, G. Iyer, S. Weiss and J. Enderlein, Proc. Natl. Acad. Sci. U. S. A.,  2009, 106, 22287. 25. S. C. Stein, A. Huss, D. Haehnel, I. Gregor and J. Enderlein, Opt. Express, 2015, 23, 16154. 26. S. Geissbuehler, C. Dellagiacoma and T. Lasser, Opt. Express, 2011, 2, 408. 27. S. W. Hell and J. Wichmann, Opt. Lett., 1994, 19, 780. 28. (a) T. A. Klar and S. W. Hell, Opt. Lett., 1999, 24, 954; (b) T. A. Klar, S. Jakobs, M. Dyba, A. Egner and S. W. Hell, Proc. Natl. Acad. Sci. U. S. A., 2000, 97, 8206. 29. M. Hofmann, C. Eggeling, S. Jakobs and S. W. Hell, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 17565.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00025

40

Xiyu Yi, Tal-Zvi Markus, Xavier Michalet, SHIMON WEISS

30. I. Testa, N. T. Urban, S. Jakobs, C. Eggeling, K. I. Willig and S. W. Hell, Neuron, 2012, 75, 992. 31. A. Honigmann, C. Eggeling, M. Schulze and A. Lepert, Laser Focus World, 2012, 48, 75. 32. M. G. Gustafsson, J. Microsc., 2000, 198, 82. 33. L. Shao, P. Kner, E. H. Rego and M. G. Gustafsson, Nat. Methods, 2011, 8, 1044. 34. K. Xu, G. Zhong and X. Zhuang, Science, 2013, 339, 452. 35. K. I. Willig, S. O. Rizzoli, V. Westphal, R. Jahn and S. W. Hell, Nature, 2006, 440, 935.

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

Part II

Light-emitting Sensors

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

View Online

         

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

View Online

Chapter 3

The Glowing Panoply of Fluorogen-based Markers for Advanced Bioimaging Arnaud Gautier* PASTEUR, Département de Chimie, École Normale Supérieure, PSL   University, Sorbonne Université, CNRS, 75005 Paris, France *E-mail: [email protected]

Table of Contents 3.1.  I ntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 3.2.  Fluorogen-based Markers Engineered from Natural   Photoreceptors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 3.2.1.  Flavin-binding Cyan–Green Fluorescent Proteins. . . . . . . . . 48 3.2.2.  Biliverdin-binding Far-red and Infrared Fluorescent   Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   49 3.2.3.  Bilirubin-binding Green Fluorescent Proteins . . . . . . . . . . . 51 3.3.  Semi-synthetic Fluorogen-based Markers. . . . . . . . . . . . . . . . . . . . . . 52 3.3.1.  Semi-synthetic Fluorogen-based Protein Markers . . . . . . . . 52 3.3.2.  Semi-synthetic Fluorogen-based RNA Markers. . . . . . . . . . . 57 3.4.  Concluding Remarks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

43

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

View Online

         

View Online

The Glowing Panoply of Fluorogen-based Markers

45

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

3.1. Introduction Living systems are controlled by dynamic biological events tightly orchestrated in space and time. Our understanding of the inner workings of these complex machines relies deeply on our ability to observe how their constituents organize and interact. Nowadays, researchers can use a wide range of imaging modalities (e.g. optical microscopy, electron microscopy, mass spectrometry imaging and optoacoustic imaging) to dissect the behavior of biological systems at various temporal and spatial scales. Among them, optical microscopy has spread into most biology laboratories, and has become invaluable for addressing major questions in e.g. cancer biology and immunology, and for deciphering e.g. embryo development or brain function. The widespread adoption of optical microscopy has resulted from the major efforts made to engineer microscopes that permit imaging with high speed, sensitivity and resolution while minimizing the side-effects of the applied light; the most advanced fluorescence microscopes today allow the observation of dynamic systems in three dimensions, at subsecond resolution and at the diffraction limit or below. Optical microscopy has achieved great success because of the concomitant advances in imaging probes able to label tissues, cells and molecules fluorescently with high selectivity. The main breakthrough in research imaging has been without question the discovery and development of the green fluorescent protein (GFP).1 For the first time, one could see and observe a broad range of specific proteins and cells in live specimens through straightforward genetic tagging techniques. The development of various color variants further allowed effective multicolor imaging and the design of various classes of biosensors.2–4 One remarkable application of fluorescent proteins is in the mapping of the brain connectivity using Brainbow technology, which allows one to mark (and track) individual neurons with several hundred hues using stochastic expression of multiple color variants.5 Fluorescent proteins have also been essential for the development of localization-based super-resolution microscopy techniques: the discovery that some fluorescent proteins can switch from a dark to a bright state under light excitation has provided a unique way to generate subsets of isolated emitting fluorophores whose positions can be determined with subdiffraction accuracy.6 The significance and importance of GFP-like fluorescent proteins for research in the life sciences was acknowledged by the award of the Nobel Prize in Chemistry to Osamu Shimomura, Martin Chalfie and Roger Tsien in 2008 for “the discovery and development of the green fluorescent protein, GFP”. By pushing the boundaries to more and more sophisticated and challenging observations, investigators have observed some limitations of GFP-like fluorescent proteins.4,7,8 First, they are weakly fluorescent in low-oxygen environments; the full maturation of their chromophore includes cyclization,

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

46

Arnaud Gautier

dehydration and oxidation of a triplet of amino acids (Ser–Tyr–Gly at positions 65–67 in GFP), and depends strictly on molecular oxygen as cofactor. Second, they fluoresce tens of minutes (up to several hours) after folding because of the slow maturation of their chromophore, preventing e.g. realtime monitoring of protein synthesis. Third, they are rather large proteins (25–30 kDa) and some of them have a tendency to oligomerize, which may lead to dysfunctional fusion proteins. Last, some GFP-like fluorescent proteins display confounding photophysics such as photoswitching, kindling or dark state conversion, which may complicate the quantitative analysis of some experiments. During the last decade, alternative markers have been developed in order to push the limits of what is seeable. This chapter focuses on key developments relying on genetically encoded protein (or nucleic acid) tags forming fluorescent complexes with small organic fluorogenic chromophores (also called fluorogens). Because of their fluorogenic properties, fluorogens are fluorescent only when bound to their cognate complementary tag, and are otherwise dark when free. Such fluorogenic labeling allows selective background-free imaging even in the presence of an excess of free fluorogen, opening up great prospects for imaging in complex samples such as tissues and whole organisms. A fluorogenic response usually results from changes in fluorescence quantum yield, spectral position or chromophore absorption coefficient induced by the change of environment undergone by the fluorogen upon binding. This chapter is divided into two parts depending on whether the bipartite fluorescent markers incorporate a natural or a synthetic fluorogenic chromophore. This natural versus synthetic distinction is explained by the distinct pros and cons of the two approaches. Hijacking natural fluorogenic chromophores solves the issue of the delivery, as the chromophores are endogenously present in cells. Moreover, several classes of natural chromophore-binding proteins can serve as the starting point to design fluorescent markers, thus facilitating engineering. These advantages are counterbalanced, however, first by the small number of natural chromophores that display fluorogenic properties, which can limit the potential engineering space, and second by the fact that the diversion of endogenous chromophores from their natural functions may engender cellular and physiological stress. On the other hand, synthetic fluorogenic chromophores have the clear advantage of being tailored with various spectral and photophysical properties by molecular engineering, enabling biological questions to be addressed with the molecular diversity offered by modern chemistry. Moreover, relying on the addition of an exogenous synthetic chromophore allows additional labeling refinement: as fluorescence is fully controlled by the applied concentration of fluorogen, semi-synthetic fluorogen-based markers permit on-demand applications in which fluorescence is desired only at a specific time or at a given density, opening up great prospects for the design of innovative labeling protocols for advanced multiplexed and super-resolution imaging.

View Online

The Glowing Panoply of Fluorogen-based Markers

47

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

3.2. Fluorogen-based Markers Engineered from Natural Photoreceptors In introduction of his Nobel Lecture, Roger Tsien told how, back in the 1980s before the first use of GFP as a fluorescent marker, he considered phycobili­ proteins as potential genetically encoded fluorescent markers.9 Phycobili­ proteins are a class of photoreceptors containing a phycocyanobilin (PCB) chromophore that act as light-harvesting antennae in the photosynthetic system of blue–green algae and cyanobacteria. Phycobiliproteins were known to fluoresce under visible light, making them attractive for imaging applications. Tsien recounted, however, how he rapidly realized that the complex biosynthesis and assembly of the phycocyanobilin chromophore would limit the general applicability of such systems, and therefore abandoned the idea. Progress in structural biology in the last three decades enabled the structures of a large collection of photoreceptors to be solved, and developments in molecular biology provided scientists with various techniques to modify protein sequences easily. These developments allowed the idea of turning chromophore-binding proteins into fluorescent proteins to be pushed further. Nature provides a large collection of chromophore-binding proteins, mainly photoreceptors, in which an endogenous chromophore (e.g. bilins, flavin, retinal, coumaric acid) is bound to the protein matrix.10 In natural photoreceptors, the chromophore reacts to light illumination by e.g. photoreduction or photoisomerization, which induces a conformational change and initiates a signaling cascade. Most photoreceptors evolved to maximize the efficacy of these photocycles, and are therefore weakly fluorescent. As fluorescence is a competing mechanism for dissipating light energy, introducing variations within the backbone of photoreceptors can impair their photocycle and thus increase their fluorescence properties. Figure 3.1 and Table 3.1 present the natural fluorogens and engineered reporters described in this section.

Figure 3.1.  Natural fluorogens found in fluorogen-based markers. Physicochemical properties of the corresponding fluorescent markers are given in Table 3.1.

View Online

48

Arnaud Gautier

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

Table 3.1.  Fluorogen-based markers engineered from natural photoreceptors. Structures of the fluorogens are shown in Figure 3.1 .a Tag

Oligomeric λem/ Fluorogen Binding mode state λabs/nm nm

ε/L mol−1 cm−1 ϕ/%

BsFbFP EcFbFP PpFbFP iLOV phiLOV2.1 MiniSOG IFP1.4 iRFP iRFP670 iRFP682 iRFP702 iRFP720 IFP2.0 mIFP miRFP670 miRFP703 miRFP709 smURFP UnaG

FMN FMN FMN FMN FMN FMN Biliverdin Biliverdin Biliverdin Biliverdin Biliverdin Biliverdin Biliverdin Biliverdin Biliverdin Biliverdin Biliverdin Biliverdin Bilirubin

13 900 14 500 13 900

Non-covalent Non-covalent Non-covalent Non-covalent Non-covalent Non-covalent Covalent Covalent Covalent Covalent Covalent Covalent Covalent Covalent Covalent Covalent Covalent Covalent Non-covalent

Dimer Dimer Dimer Monomer Monomer Monomer Dimer Dimer Dimer Dimer Dimer Dimer Dimer Monomer Monomer Monomer Monomer Dimer Monomer

449 448 450 447 450 447 684 692 643 663 673 702 690 683 642 674 683 642 498

495 496 496 497 497 497 708 713 670 682 702 720 711 705 670 703 709 670 527

14 200 92 000 105 000 114 000 90 000 93 000 96 000 98 000 82 000 87 400 90 900 78 400 180 000 77 300

39 44 27 44 20 41 7.0 5.9 11.1 11.3 8.2 6.0 8.1 8.4 14 8.6 5.4 18 51

Ref. 13 13 13 15 13 13 30 32 33 33 33 33 31 34 35 35 35 36 40

a

λ abs, wavelength of maximal absorption; λem, wavelength of maximal emission; ε, molar absorption coefficient at λabs; ϕ, fluorescence quantum yield.

3.2.1. Flavin-binding Cyan–Green Fluorescent Proteins The idea of reformatting natural photoreceptors was used to transform small light-, oxygen- and voltage-sensing (LOV) domains—a class of blue-light photoreceptors found in plants, algae and bacteria—into cyan–green fluorescent proteins.11–13 LOV proteins associate with the ubiquitous cofactor flavin mononucleotide (FMN). Blue-light photoreception is believed to involve the reversible formation of a covalent adduct between FMN and a conserved cysteine within the FMN-binding pocket upon blue-light illumination. The conformational change associated with the adduct formation then activates downstream signaling. Flavin-based fluorescent proteins (FbFPs) were engineered by replacing the conserved cysteine of bacterial LOV domains with an alanine in order to inhibit the natural photocycle and reduce the fluorescence quenching of FMN.14 Similarly, the fluorescent protein iLOV was generated from the LOV2 domain of Arabidopsis thaliana phototropin 2.15,16 FbFPs gained particular attention because of their potential use as alternatives to GFP in anaerobic conditions. Because fluorescence results only from the binding of FMN, an abundant cofactor in cells, FbFPs are fluorescent regardless of the level of oxygen. FbFPs proved to surpass GFP-like fluorescent proteins for reporting on protein expression in the absence of oxygen in bacteria,14,17,18 fungi19 and mammalian cells.20 FbFPs allowed in particular the study of host–pathogen interactions under physiologically relevant anaerobic conditions.21,22 FbFP was moreover used to design a ratiometric

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

The Glowing Panoply of Fluorogen-based Markers

49

Förster resonance energy transfer (FRET) oxygen sensor by fusing it to yellow fluorescent protein (YFP).23 Oxygen is required for the maturation of YFP (playing the role of FRET acceptor), but not for that of FbFP (the FRET donor). Consequently, the FRET efficiency directly gives the oxygen level: a high FRET efficiency indicates a high oxygen level, whereas a low FRET efficiency means a low oxygen level. FbFPs also proved to surpass GFP-like fluorescent proteins because of their smaller size: 12–16 kDa on average compared with 25–30 kDa for the latter. iLOV was shown, for instance, to be better suited than GFP to label viruses and monitor plant infection because the smaller genetic load maintains higher infectivity.15 Engineering of FbFPs further allowed the development of reporters that not only fluoresce but also generate singlet oxygen. The LOV2 domain of Arabidopsis thaliana phototropin 2 was transformed into MiniSOG (mini singlet oxygen generator), a fluorescent reporter generating high levels of singlet oxygen and other reactive oxygen species (ROS) upon blue-light illumination.24 Because of this singular property, MiniSOG was used to promote local photooxidation of diaminobenzidine into osmiophilic insoluble polymer, which can be then stained with osmium for high-resolution imaging by electron microscopy. As MiniSOG also fluoresces green light upon light excitation, MiniSOG fusions can be observed in both fluorescence and electron microscopy, thus allowing correlative imaging. The ability of MiniSOG to generate ROS locally upon light illumination was also used to (i) promote cell death in cancer cells25 and in neuronal cells in Caenorhabditis elegans,26 (ii) silence genetically specified synapses by chromophore-assisted light inactivation (CALI) of synaptic proteins27 and (iii) map protein proximity in large protein complexes.28 3.2.2. Biliverdin-binding Far-red and Infrared Fluorescent Proteins Infrared fluorescent proteins have long been a goal in the field of probe design because of their great potential for deep-tissue and whole-body imaging. Infrared light scatters much less than visible light into tissue, and the window 650–900 nm is transparent (hemoglobin, water and lipids do not absorb at these wavelengths),29 enabling tissues to be imaged at an unprecedented depth. The highest maximal emission wavelength found among red GFP-like fluorescent proteins, however, is only 650 nm (for mPlum and mNeptune) and is intrinsically limited by the structure of the chromophore. The first infrared fluorescent protein, IFP1.4 (with maximal emission at 708 nm), was engineered from the N-terminal PAS and GAF domains of Deinococcus radiodurans bacteriophytochrome.30 This biliprotein photoreceptor, which incorporates biliverdin IXα (hereafter called biliverdin) as cofactor, regulates pigment synthesis to protect the bacterium from intense visible light. Biliverdin is covalently attached to the apo-receptor via a thioether bond with a nearby cysteine side-chain through a self-catalytic process. Naturally, bacteriophytochromes sense light by reversible cis–trans photoisomerization of the C15=C16 double bond of biliverdin. Restricting the

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

50

Arnaud Gautier

conformational freedom of biliverdin by introducing mutations that prevent photoisomerization from occurring enabled fluorescence to be significantly increased. Although biliverdin is ubiquitously present in mammals, optimal labeling of IFP1.4 in mammalian cells and mice required an exogenous supply of biliverdin. Coexpression of heme oxygenase (HO1) in charge of biliverdin synthesis proved to boost biliverdin levels, allowing the imaging of the optimized IFP2.0 in Drosophila neurons (which contains low levels of biliverdin) and in mouse brain tumors without exogenous addition of biliverdin.31 The first infrared fluorescent protein that does not require addition of exogenous biliverdin was developed using a truncated version of the bacteriophytochrome RpBphP2 from Rhodopseudomonas palustris.32 The higher affinity for biliverdin explains the higher performance of this improved near-infrared fluorescent protein (iRFP). Engineering of iRFP variants with emission wavelengths extending from 670 to 720 nm further expanded the color range available, opening up new opportunities for multiplexed wholebody imaging.33 A limitation of the above-mentioned proteins, however, is that they are either dimeric (iRFP) or form dimers at high concentrations (IFP1.4 and IFP2.0).34 Fully monomeric IFP (mIFP) was engineering from a monomeric truncated bacteriophytochrome from Bradyrhizobium34. mIFP proved to label proteins correctly in mammalian cells, flies and zebrafish. Brighter monomeric iRFPs (miRFPs) with emission wavelengths extending from 670 to 710 nm were next engineered from bacteriophytochrome RpBphP1 35. miRFPs proved to be well suited for wide-field and structured illumination microscopy (SIM). The family of near-infrared fluorescent proteins was further extended by engineering the allophycocyanin α-subunit from a cyanobacterial phycobiliprotein instead of a bacterial phytochrome.36 Cyanobacterial phycobiliproteins are normally functionalized with phycocyanobilin (PCB) by a specific lyase. To develop a useful fluorescent protein, the protein was evolved to be self-sufficient (i.e. to not require any lyase) and to bind covalently biliverdin instead of PCB. The engineering process ultimately gave a bright protein, designated small ultra-red fluorescent protein (smURFP). smURFP is a homodimer of 15 kDa subunits. It has an exceptional absorption coefficient (ε = 180 000 L mol−1 cm−1) and a modest fluorescence quantum yield (ϕ = 0.18), making it the brightest far-red/near-infrared fluorescent protein. Although the brightest fluorescence in cells is obtained with addition of exogenous biliverdin (or the more cell-permeant biliverdin methyl ester), expression of smURFP was efficiently imaged in HT1080 tumor rodent xenographs in mice even without the addition of exogenous biliverdin.36 The great potential of infrared fluorescent proteins for in vivo imaging led to the development of various biosensors to sense specific substances/activities or protein–protein interactions. Most phytochrome-based biosensors are split- or insertion-based biosensors that rely on the reconstitution of the PAS–GAF assembly. A near-infrared split reporter for the detection of protein–protein interactions in vivo was designed by separating the PAS and GAF

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

The Glowing Panoply of Fluorogen-based Markers

51

domains of iRFP.37 The two domains complement efficiently only when in close proximity. Proximity-induced complementation then promotes biliverdin binding and thus fluorescence. Split iRFP allowed the visualization of known interactions in mice; however, (i) as with all biliverdin-based fluorescent proteins, the covalent attachment of biliverdin was slow, which led to a slow fluorescence maturation (several hours) in cells, and (ii) the assembly was irreversible and dimeric, which prevented the study of dynamic processes or complex stoichiometry. A reversible infrared split system was obtained from IFP1.4. Split IFP1.4 allowed the monitoring in real time of the disruption of protein–protein interactions in mammalian cells and yeast.38 Split IFP1.4 is, however, less bright than split iRFP, which may limit its use for whole-body imaging. Recently, Verkhusha and co-workers obtained truly monomeric split reporters using miRFPs, which should allow the efficient screening of novel protein–protein interactions.35 By designing split systems with distinct colors but sharing one split fragment, they were furthermore able to distinguish interactions of one given protein with two alternative partners.35 Apart from split biosensors, infrared fluorescent proteins were also used to design protease biosensors.39 IFP1.4 was modified so that protease activity promotes biliverdin attachment and therefore fluorescence. The uncleaved sensor (iProtease) is non-fluorescent because the cysteine involved in biliverdin attachment cannot react with biliverdin because of physical displacement. Proteolytic cleavage frees the cysteine, which returns in the biliverdin binding site, thus promoting fluorescence. This approach allowed the develop­ ment of a caspase-3 sensor (iCasper) for the visualization of apoptosis.39 iCasper allowed the study of the spatiotemporal coordination between cell apoptosis and embryonic morphogenesis in Drosophila, and revealed the dynamics of apoptosis during tumorigenesis in the brain of Drosophila. Such design opens up exciting prospects for in vivo biosensing, and demonstrates how infrared fluorescent proteins can allow the construction of fluorogenic biosensors for visualizing the spatiotemporal dynamics of cell signaling  in vivo. 3.2.3. Bilirubin-binding Green Fluorescent Proteins Although flavin- and biliverdin-based fluorescent proteins are synthetic reporters obtained by engineering natural photoreceptors, natural evolution has also generated fluorescent proteins incorporating natural fluorogenic chromophores. A natural fluorogen-based green fluorescent protein was discovered in Japanese eel muscles.40 The fluorescence properties of this small monomeric protein, called UnaG, result from the non-covalent, high-affinity and specific binding of the fluorogenic bilirubin. Fluorescent holo-UnaG forms efficiently in mammalian cells because bilirubin, an endogenous catabolic product of heme, is present at high concentrations in animals. Exogenous supply of bilirubin allowed the use of UnaG in organisms such as bacteria that do not produce bilirubin.40

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

52

Arnaud Gautier

UnaG displays the same advantages as FbFPs: UnaG is (i) twofold smaller than GFP-like fluorescent proteins, (ii) almost instantaneously fluorescent upon chromophore binding and (iii) well suited to visualize fusion proteins under anaerobic conditions (because the fluorescence maturation process is fully independent of molecular oxygen).40 Furthermore, UnaG is fivefold brighter than FbFPs, making it one of the brightest alternatives to GFP. The oxygen independence of UnaG fluorescence was used to design genetically encoded hypoxia sensors for light microscopy.41 In these sensors, the expression of destabilized versions of UnaG was under the control of hypoxia-responsive promoters. These sensors proved to be highly effective in visualizing hypoxia in tumors, and allowed strong heterogeneity in tumor hypoxia to be revealed at the cellular level.

3.3. Semi-synthetic Fluorogen-based Markers The modular nature of fluorogen-based markers enables a priori the tuning of the chromophore by molecular engineering, and biological questions with the molecular diversity offered by modern chemistry to be addressed. Synthetic fluorogenic chromophores with various spectral and physicochemical properties have been mobilized for engineering new imaging probes.42 Semi-synthetic fluorogen-based markers usually use design principles based on conformational locking (of e.g. molecular rotors) or ground-state isomerization (of e.g. silicon rhodamines) to achieve a fluorogenic response. In the case of molecular rotors, internal rotation disrupts the dye planarity and is a source of non-radiative relaxation; when bound to complementary receptors, the rotation is blocked and the fluorescence of these fluorogens is strongly enhanced. In the case of silicon rhodamines, the dye adopts a non-fluorescent spirolactone form in polar solvents, and a fluorescent zwitterionic open form in the vicinity of proteins because of the local decrease in polarity. Compared with the fluorogen-based markers described in Section 3.2, which rely on endogenous natural fluorogens, semi-synthetic markers allow additional labeling refinement since fluorescence can be fully controlled by the applied concentration of the exogenous fluorogen. This feature allows the development of on-demand applications in which fluorescence is desired only at a specific time or at a given density, opening up great prospects for the design of innovative labeling protocols for advanced multiplexed and super-resolution imaging. Figure 3.2 and Table 3.2 present the synthetic fluorogens and engineered reporters described in this section. 3.3.1. Semi-synthetic Fluorogen-based Protein Markers 3.3.1.1.  Fluorogen-activating Proteins.  Activating the fluorescence of a fluoro­gen by molecular recognition requires the development of complementary protein receptors. Fluorogen-activating proteins (FAPs) that generate fluorescence through immobilization of fluorogenic molecular

View Online

53

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

The Glowing Panoply of Fluorogen-based Markers

Figure 3.2.  Synthetic fluorogens found in fluorogen-based markers. Physicochemical properties of the corresponding fluorescent markers are given in Table 3.2.

rotors were first engineered from 25–30 kDa single-chain antibodies (scFvs). The screening of yeast-displayed libraries of human scFvs by fluorescenceactivated cell sorting (FACS) allowed the isolation of FAPs that bind noncovalently with high affinity modified variants of Malachite Green (MG) and Thiazole Orange (TO), two well-known fluorogenic molecular rotors that fluoresce strongly in constrained environments.43 The generated systems fluoresce far-red light (for the MG complex) and green–yellow light (for the TO complex), with brightness levels as good as those encountered in GFP-like fluorescent proteins. To be fully active, however, the original FAPs needed nonreducing environments to allow the formation of internal disulfide bonds, thus limiting their use to the cell surface and secretory pathway.43 Recently, disulfide-free FAPs were engineered for protein labeling in various reducing compartments, including the cytosol, using cell-permeant MG-ester.44,45

View Online

54

Arnaud Gautier

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

Table 3.2.  Semi-synthetic fluorogen-based protein markers. Structures of the fluorogens are shown in Figure 3.2 .a Tag

Oligomeric λabs/ Fluorogen Binding mode state nm

FAP HL1.01 Thiazole orange FAP H6 Malachite green Self-  SiR650 labeling proteins Self-  SiR700 labeling proteins Self-  JF646 labeling proteins Self-  JF585 labeling proteins Self-  JF635 labeling proteins FAST HMBR FAST HBR-3,5DM FAST HBR3,5DOM Spinach 2 DFHBI Spinach 2 DFHBI-1T Broccoli DFHBI-1T Corn DFHO Orange DFHO Broccoli Red DFHO Broccoli

ε/L mol−1 λem/nm cm−1 ϕ/%

Ref.

Non-covalent

Monomer

509

530

60 000

47

43

Non-covalent

Monomer

635

656

105 000

25

43

Covalent

Monomer

645

661

100 000

39

62

Covalent

Monomer

687

716

100 000

Covalent

Monomer

646

664

152 000

54

67

Covalent

Monomer

585

609

156 000

78

67

Covalent

Monomer

635

652

167 000

56

67

Non-covalent Non-covalent Non-covalent

Monomer Monomer Monomer

481 499 518

540 562 600

45 000 48 000 39 000

23 49 31

70 70 70

Non-covalent Non-covalent Non-covalent Non-covalent Non-covalent

447 482 472 505 513

501 505 507 545 562

22 000 31 000 29 600 29 000 34 000

72 94 94 25 28

83 83 85 92 92

Non-covalent

518

582

35 000

34

92

65

a

λ abs, wavelength of maximal absorption; λem, wavelength of maximal emission; ε, molar absorption coefficient at λabs; ϕ, fluorescence quantum yield.

Additional engineering efforts enabled the available chromatic palette to be expanded from the blue to the far-red edge of the visible spectrum.46,47 Interestingly, FAPs' fluorogens can be rendered poorly membrane permeant by adding electronic charge. This property was used to selectively label cellsurface proteins.48,49 Add-and-read protocols were developed to study the endocytosis and recycling of FAP-tagged receptors.50 Pulse chase with two fluorogens of different colors and different permeability properties further allowed the quantitative study of receptor recycling upon agonist activation.51 FAPs proved to display great potential for super-resolution microscopy and single-molecule tracking. Far-red MG-based FAPs are highly photostable, which allowed live-cell imaging with stimulated emission depletion (STED) nanoscopy in mammalian cells and bacteria.52,53 Furthermore, because the investigator can control the fluorogen concentration at will, it is possible to

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

The Glowing Panoply of Fluorogen-based Markers

55

label only a subset of proteins independently of their expression level. This property was used for tracking single receptors on the cell surface54 and for generating stochastic binding-based blinking in order to generate sparse subsets of emitters for the reconstruction of images of FAP-tagged proteins with sub-diffraction resolution.55 FAP technology also allowed the design of photosensitizer molecules able to produce ROS upon illumination for applications such as inactivation through CALI of directly linked proteins, targeted cell damage or cellular ablation. A genetically encoded FAP that binds a heavy atom-containing fluorogenic MG dye forms an active photosensitizer that produces efficiently singlet oxygen when activated by near-infrared light.56 Unlike MiniSOG discussed in Section 3.2.1 that is constitutively active, this FAP-based photosensitizer can be activated on-demand by addition of the fluorogenic dye. Interestingly, FAP-based photosensitizers display near-infrared excitation and emission, which provides a new spectral range for photosensitization and thus opens up great prospects for imaging, protein and cell manipulation and cellular ablation in whole organisms. Beyond the use for protein inactivation, FAP-based photosensitizers allowed the CALI of proteins, targeted cell killing and targeted lineage ablation in zebrafish.56 3.3.1.2.  Self-labeling Tags.  Semi-synthetic fluorogen-based markers were also obtained by exploiting site-specific labeling systems such as SNAPtag/CLIP-tag and Halo-tag, which react covalently with specific substrates bearing chemical probes. SNAP-tag is a 20 kDa protein evolved from the human DNA repair protein O6-alkylguanine-DNA alkyltransferase (AGT).57,58 SNAP-tag transfers the functionalized benzyl group of O6-benzylguanine (BG) derivatives to its active-site cysteine, thus allowing irreversible covalent labeling of fusion proteins. SNAP-tag accepts a broad variety of chemical functionalities on BG, making it one of the most versatile tags currently available.59 CLIP-tag, an engineered variant of SNAP-tag, reacts selectively with O2-benzylcytosine (BC) substrates instead of BG.60 Halo-tag is a 33 kDa protein engineered from a bacterial haloalkane dehydrogenase that covalently binds chloroalkane ligands.61 The use of fluorogenic chromophores instead of permanent fluorophores allowed the development of labeling protocols with no-washing steps, as free unreacted substrates do not fluoresce. Beyond increasing contrast, the use of fluorogenic substrates allowed the need for extensive washing steps to be removed, thus increasing the temporal resolution. Silicon–rhodamine (SiR) derivatives were used to design fluorogenic substrates for SNAP-tag, CLIP-tag and Halo-tag.62 The fluorogenic response of SiR relies on ground-state isomerization that breaks the dye conjugation: in aqueous solution, SiR adopts mainly a closed UV-absorbing spirolactone form, whereas it undergoes ring opening in less polar environments such as the vicinity of proteins. The open zwitterionic form absorbs at 640–650 nm and fluoresces in the far-red region at 660–670 nm. SiR-based substrates proved to be highly efficient for labeling

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

56

Arnaud Gautier

SNAP-, CLIP- and Halo-tagged proteins in various organelles of living cells within 30–60 min with no significant background.62 Because of their excellent spectroscopic properties, SiR substrates allowed live-cell super-resolution microscopy of biological structures using GSDIL (ground-state depletion followed by individual molecule return)63 and dSTORM (direct stochastic optical reconstruction microscopy)64 by relying on the stochastic blinking of these conventional fluorophores. Recently, near-infrared SiR analogs with emission and excitation wavelength maxima at 680 and 715 nm were reported, allowing multicolor super-resolution imaging.65 Moreover, SiR brightness and photostability were improved by incorporating azetidine four-membered rings. The resulting dye, Janelia Fluor 646 (JF646), was shown to be an efficient label for imaging SNAP- and Halo-tagged proteins by conventional and super-resolution microscopy.66 Refinement and extension of this strategy allowed the development of fluorogenic labels with excitation ranging from orange (JF585) to red (JF635).67 These dyes are cell permeant and exhibit very high fluorogenicity, which allowed protein labeling in neural tissues in explants and in Drosophila larvae,67 and open up great prospects for imaging deep structures in whole animals. 3.3.1.3.  Fluorescence-activating and Absorption-shifting Tag.  Fluorescenceactivating and absorption-shifting tag (FAST) is a small protein tag of 14 kDa that binds and switches on the fluorescence of hydroxybenzylidene rhodanine (HBR) analogs through conformational locking.68 FAST was evolved from the apo photoactive yellow protein (apo-PYP) by directed evolution using yeast display and FACS sorting. The fluorogenicity of HBR analogs originates from their push–pull structure composed of an electron-donating phenol ring conjugated with an electron-withdrawing rhodanine heterocycle. In solution, these fluorogens dissipate light energy non-radiatively through internal rotation or cis–trans isomerization. Binding to FAST locks the fluorogen, which slows non-radiative decay and strongly increases fluorescence. Upon binding, the fluorogen undergoes also a 80 nm red shift in absorption due to a selective deprotonation reaction. Consequently, the free fluorogen barely absorbs at the wavelength used for exciting the bipartite complex, further enhancing the fluorogenic response. FAST proved to be highly effective in fluorescently labeling proteins in living cells (bacteria, yeast, mammalian cells) in a large number of organelles and subcellular localizations.68 Full labeling is achieved within a few seconds after fluorogen addition in living cells. Provided that the fluorogen is present, FAST is fluorescent instantaneously after folding because of fast binding kinetics, which allows fast processes to be followed in near real time. Because HBR analogs are highly cell permeant, efficient labeling of FAST was observed in multicellular organisms such as zebrafish embryos, opening up great prospects for in vivo imaging. FAST is distinguished from other fluorogen-based markers because fluorogen binding is non-covalent, highly dynamic and fully reversible (because of

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

The Glowing Panoply of Fluorogen-based Markers

57

a high dissociation rate constant). Fluorogen washing allows one to reverse labeling and switch off fluorescence within a few seconds in cells, making FAST a fluorescence switch that can be switched on or off efficiently at will by the addition or removal of fluorogen. The rapid exchange dynamics further allow efficient fluorogen renewal, which reduces the apparent photobleaching rate.69 Finally, because of its fast exchange dynamics, FAST behaves as a blinking fluorophore at the single-molecule level. Such spontaneous stochastic blinking might find very interesting applications for super-resolution microscopy in live cells. Ongoing efforts are further testing the ability to expand the spectral properties of FAST. Recently, FAST emission color was extended to the orange and red regions by modifying the structure of its complementary fluorogen.70 The ability to make FAST fluoresce green–yellow, orange or red by a simple change of fluorogen enables one to adapt the color of FAST to the experimental spectral constraints without the need for recloning the tag, providing an experimental versatility not encountered with GFP-like fluorescent proteins. The ability to swap color dynamically by exchanging fluorogens was furthermore used as a unique kinetic signature to image FAST selectively in spectrally crowded environments.70 By evaluating the degree of temporal anticorrelation of the green and red fluorescence signals upon color swapping, FAST-tagged proteins could be selectively detected in cells already tagged with green and red reporters, illustrating the general potential of non-covalent fluorogenic reporters for the development of new innovative imaging methods for advanced biological imaging. 3.3.2. Semi-synthetic Fluorogen-based RNA Markers The idea of forming a fluorescent marker by association of a genetically encoded module and a fluorogenic chromophore is a very general idea that goes well beyond the labeling of proteins. Fluorogenic labeling was used to expand fluorescent labeling to more diverse cellular molecules such as RNA. Various studies showed that fluorescence could be generated by engineered RNA aptamers binding selectively fluorogenic chromophores,71–78 paving the way towards new approaches to image RNA in living cells.79–81 Efficient RNA imaging in live cells was made possible with the development of an RNA aptamer mimic of GFP named Spinach82,83 and its optimized versions Spinach 2 84 and Broccoli.85 These engineered RNA aptamers form fluorescent complexes with analogs of the fluorogenic 3,5-difluoro-4-hydroxybenzylideneimidazolidinone (DFHBI). DFHBI is related to the chromophore of GFP known to be fluorescent only when encased inside the GFP barrel. Likewise, DFHBI fluoresces only when bound to Spinach RNA aptamers,86 allowing Spinach-tagged RNA in living cells to be imaged with high contrast. Spinach aptamers proved to be highly efficient in designing biosensors by coupling them with aptameric sensing units. Sensors in which the binding of a given analyte (e.g. metabolite or protein) promotes fluorogen binding and activation through conformational coupling enabled the visualization

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

58

Arnaud Gautier

of the dynamics of the analyte levels in bacteria,87–90 further demonstrating the great potential of such fluorogen-based markers for the design of sensors able to sense the abundance, distribution and flux of intracellular molecules. Broccoli was also used to design a fluorimetric assay to measure the activity of RNA-modifying enzymes in cells.91 Broccoli was modified to contain N6-methyladenosine, a prevalent mRNA base modification. Methylation renders Broccoli non-fluorescent; the fluorescence can be recovered by the action of RNA demethylases. This approach allowed the development of high-throughput screens for inhibitors of the RNA demethylase fat mass and obesity-associated protein (FTO). RNA mimics of red fluorescent proteins (RFPs) were obtained by similar engineering strategies. These fluorogen-based RNA markers, named Corn, Orange Broccoli and Red Broccoli, bind 3,5-difluoro-4-hydroxybenzylideneimidazolinone-2-oxime (DFHO), a fluorogen resembling the chromophore found in RFP.92,93 Corn, Orange Broccoli and Red Broccoli form fluorescent complexes with DFHO displaying red-shifted emissions (545, 562 and 582 nm, respectively) with respect to Spinach. Corn : DFHO showed high photostability, unlike Spinach and Broccoli, which undergo fast reversible photobleaching,94,95 allowing quantitative fluorescence imaging of mTOR-dependent Pol III transcription.

3.4. Concluding Remarks Fluorogen-based markers allow more and more sophisticated and challenging observations in living cells and organisms. They are very attractive alternatives to canonical fluorescent proteins because they open up exciting new possibilities for whole-body imaging, biosensor design, multiplex imaging and high-resolution imaging. Fluorogen-based markers have not yet shown their full potential, and it is a safe bet to say that the coming years will see further exciting and unpredictable advances.

Acknowledgements The European Research Council (ERC-2016-CoG-724705 FLUOSWITCH), France BioImaging (ANR-10-INBS-04) and the Equipex Morphoscope 2 (ANR11-EQPX-0029) supported this work.

References 1. R. Y. Tsien, Annu. Rev. Biochem., 1998, 67, 509–544. 2. N. Shaner, P. Steinbach and R. Tsien, Nat. Methods, 2005, 2, 905–909. 3. D. M. Chudakov, M. V. Matz, S. Lukyanov and K. A. Lukyanov, Physiol. Rev., 2010, 90, 1103–1163. 4. E. A. Rodriguez, R. E. Campbell, J. Y. Lin, M. Z. Lin, A. Miyawaki, A. E. Palmer, X. Shu, J. Zhang and R. Y. Tsien, Trends Biochem. Sci., 2017, 42, 111–129.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

The Glowing Panoply of Fluorogen-based Markers

59

5. J. Livet, T. A. Weissman, H. Kang, R. W. Draft, J. Lu, R. A. Bennis, J. R. Sanes and J. W. Lichtman, Nature, 2007, 450, 56–62. 6. E. Betzig, G. H. Patterson, R. Sougrat, O. W. Lindwasser, S. Olenych, J. S. Bonifacino, M. W. Davidson, J. Lippincott-Schwartz and H. F. Hess, Science, 2006, 313, 1642–1645. 7. S. J. Remington, Curr. Opin. Struct. Biol., 2006, 16, 714–721. 8. M. Zimmer, Chem. Rev., 2002, 102, 759–782. 9. R. Y. Tsien, Angew. Chem. Int. Ed., 2009, 48, 5612–5626. 10. M. A. van der Horst and K. J. Hellingwerf, Acc. Chem. Res., 2004, 37, 13–20. 11. A. Mukherjee and C. M. Schroeder, Curr. Opin. Biotechnol., 2015, 31, 16–23. 12. T. Drepper, T. Gensch and M. Pohl, Photochem. Photobiol. Sci., 2013, 12, 1125–1134. 13. M. Wingen, J. Potzkei, S. Endres, G. Casini, C. Rupprecht, C. Fahlke, U. Krauß, K.E. Jaeger, T. Drepper and T. Gensch, Photochem. Photobiol. Sci., 2014, 13, 875–879. 14. T. Drepper, T. Eggert, F. Circolone, A. Heck, U. Krauß, J.-K. Guterl, M. Wendorff, A. Losi, W. Gärtner and K.-E. Jaeger, Nat. Biotechnol., 2007, 25, 443–445. 15. S. Chapman, C. Faulkner, E. Kaiserli, C. Garcia-Mata, E. I. Savenkov, A. G. Roberts, K. J. Oparka and J. M. Christie, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 20038–20043. 16. J. M. Christie, K. Hitomi, A. S. Arvai, K. A. Hartfield, M. Mettlen, A. J. Pratt, J. A. Tainer and E. D. Getzoff, J. Biol. Chem., 2012, 287, 22295–22304. 17. T. Drepper, R. Huber, A. Heck, F. Circolone, A. K. Hillmer, J. Buchs and K. E. Jaeger, Appl. Environ. Microbiol., 2010, 76, 5990–5994. 18. A. M. Buckley, C. Jukes, D. Candlish, J. J. Irvine, J. Spencer, R. P. Fagan, A. J. Roe, J. M. Christie, N. F. Fairweather and G. R. Douce, Sci. Rep., 2016, 6, 23463. 19. D. Tielker, I. Eichhof, K. E. Jaeger and J. F. Ernst, Eukaryotic Cell, 2009, 8, 913–915. 20. J. Walter, S. Hausmann, T. Drepper, M. Puls, T. Eggert and M. Dihné, PLoS One, 2012, 7, e43921. 21. L. A. Lobo, C. J. Smith and E. R. Rocha, FEMS Microbiol. Lett., 2011, 317, 67–74. 22. C. H. Choi, J. V. DeGuzman, R. J. Lamont and Ö. Yilmaz, PLoS One, 2011, 6, e18499. 23. J. Potzkei, M. Kunze, T. Drepper, T. Gensch, K.-E. Jaeger and J. Büchs, BMC Biol., 2012, 10, 28. 24. X. Shu, V. Lev-Ram, T. J. Deerinck, Y. Qi, E. B. Ramko, M. W. Davidson, Y. Jin, M. H. Ellisman and R. Y. Tsien, PLoS Biol., 2011, 9, e1001041. 25. A. P. Ryumina, E. O. Serebrovskaya, M. V. Shirmanova, L. B. Snopova, M. M. Kuznetsova, I. V. Turchin, N. I. Ignatova, N. V. Klementieva, A. F. Fradkov, B. E. Shakhov, E. V. Zagaynova, K. A. Lukyanov and S. A. Lukyanov, Biochim. Biophys. Acta, Gen. Subj., 2013, 1830, 5059–5067. 26. Y. B. Qi, E. J. Garren, X. Shu, R. Y. Tsien and Y. Jin, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 7499–7504. 27. J. Y. Lin, S. B. Sann, K. Zhou, S. Nabavi, C. D. Proulx, R. Malinow, Y. Jin and R. Y. Tsien, Neuron, 2013, 79, 241–253. 28. T.-L. To, M. J. Fadul and X. Shu, Nat. Commun., 2014, 5, 4072. 29. V. Marx, Nat. Methods, 2014, 11, 717–720. 30. X. Shu, A. Royant, M. Z. Lin, T. A. Aguilera, V. Lev-Ram, P. A. Steinbach and R. Y. Tsien, Science, 2009, 324, 804–807. 31. D. Yu, W. C. Gustafson, C. Han, C. E. L. Lafaye, M. Noirclerc-Savoye, W.-P. Ge, D. A. Thayer, H. Huang, T. B. Kornberg, A. Royant, L. Y. Jan, Y. N. Jan, W. A. Weiss and X. Shu, Nat. Commun., 2014, 5, 1–7. 32. G. S. Filonov, K. D. Piatkevich, L.-M. Ting, J. Zhang, K. Kim and V. V. Verkhusha, Nat. Biotechnol., 2011, 29, 759–763.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

60

Arnaud Gautier

33. D. M. Shcherbakova and V. V. Verkhusha, Nat. Methods, 2013, 10, 751–754. 34. D. Yu, M. A. Baird, J. R. Allen, E. S. Howe, M. P. Klassen, A. Reade, K. Makhijani, Y. Song, S. Liu, Z. Murthy, S.-Q. Zhang, O. D. Weiner, T. B. Kornberg, Y. N. Jan, M. W. Davidson and X. Shu, Nat. Methods, 2015, 12, 763–765. 35. D. M. Shcherbakova, M. Baloban, A. V. Emelyanov, M. Brenowitz, P. Guo and V. V. Verkhusha, Nat. Commun., 2016, 7, 12405. 36. E. A. Rodriguez, G. N. Tran, L. A. Gross, J. L. Crisp, X. Shu, J. Y. Lin and R. Y. Tsien, Nat. Methods, 2016, 13, 763–769. 37. G. S. Filonov and V. V. Verkhusha, Chem. Biol., 2013, 20, 1078–1086. 38. E. Tchekanda, D. Sivanesan and S. W. Michnick, Nat. Methods, 2014, 11, 641–644. 39. T.-L. To, B. J. Piggott, K. Makhijani, D. Yu, Y. N. Jan and X. Shu, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, 3338–3343. 40. A. Kumagai, R. Ando, H. Miyatake, P. Greimel, T. Kobayashi, Y. Hirabayashi, T. Shimogori and A. Miyawaki, Cell, 2013, 153, 1602–1611. 41. R. Erapaneedi, V. V. Belousov, M. Scha fers and F. Kiefer, EMBO J., 2016, 35, 102–113. 42. A. S. Klymchenko, Acc. Chem. Res., 2017, 50, 366–375. 43. C. Szent-Gyorgyi, B. A. Schmidt, Y. Creeger, G. W. Fisher, K. L. Zakel, S. Adler, J. A. J. Fitzpatrick, C. A. Woolford, Q. Yan, K. V. Vasilev, P. B. Berget, M. P. Bruchez, J. W. Jarvik and A. Waggoner, Nat. Biotechnol., 2008, 26, 235–240. 44. B. P. Yates, M. A. Peck and P. B. Berget, Mol. Biotechnol., 2012, 54, 829–841. 45. C. A. Telmer, R. Verma, H. Teng, S. Andreko, L. Law and M. P. Bruchez, ACS Chem. Biol., 2015, 10, 1239–1246. 46. H. Ozhalici-Unal, C. L. Pow, S. A. Marks, L. D. Jesper, G. L. Silva, N. I. Shank, E. W. Jones, J. M. Burnette, P. B. Berget and B. A. Armitage, J. Am. Chem. Soc., 2008, 130, 12620–12621. 47. K. J. Zanotti, G. L. Silva, Y. Creeger, K. L. Robertson, A. S. Waggoner, P. B. Berget and B. A. Armitage, Org. Biomol. Chem., 2011, 9, 1012–1020. 48. J. Holleran, D. Brown, M. H. Fuhrman, S. A. Adler, G. W. Fisher and J. W. Jarvik, Cytometry A, 2010, 77, 776–782. 49. G. W. Fisher, S. A. Adler, M. H. Fuhrman, A. S. Waggoner, M. P. Bruchez and J. W. Jarvik, J. Biomol. Screening, 2010, 15, 703–709. 50. Q. Yan, B. F. Schmidt, L. A. Perkins, M. Naganbabu, S. Saurabh, S. K. Andreko and M. P. Bruchez, Org. Biomol. Chem., 2015, 13, 2078–2086. 51. C. P. Pratt, J. He, Y. Wang, A. L. Barth and M. P. Bruchez, Bioconjugate Chem., 2015, 26, 1963–1971. 52. J. A. J. Fitzpatrick, Q. Yan, J. J. Sieber, M. Dyba, U. Schwarz, C. Szent-Gyorgyi, C. A. Woolford, P. B. Berget, A. S. Waggoner and M. P. Bruchez, Bioconjugate Chem., 2009, 20, 1843–1847. 53. S. Saurabh, A. M. Perez, C. J. Comerci, L. Shapiro and W. E. Moerner, J. Am. Chem. Soc., 2016, 138, 10398–10401. 54. S. L. Schwartz, Q. Yan, C. A. Telmer, K. A. Lidke, M. P. Bruchez and D. S. Lidke, ACS Chem. Biol., 2015, 10, 539–546. 55. Q. Yan, S. L. Schwartz, S. Maji, F. Huang, C. Szent-Gyorgyi, D. S. Lidke, K. A. Lidke and M. P. Bruchez, ChemPhysChem, 2014, 15, 687–695. 56. J. He, Y. Wang, M. A. Missinato, E. Onuoha, L. A. Perkins, S. C. Watkins, C. M. St Croix, M. Tsang and M. P. Bruchez, Nat. Methods, 2016, 13, 263–268. 57. A. Keppler, S. Gendreizig, T. Gronemeyer, H. Pick, H. Vogel and K. Johnsson, Nat. Biotechnol., 2003, 21, 86–89. 58. A. Keppler, H. Pick, C. Arrivoli, H. Vogel and K. Johnsson, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 9955–9959.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

The Glowing Panoply of Fluorogen-based Markers

61

59. A. Gautier, K. Johnsson and H. O'Hare, AGT/SNAP-Tag: A Versatile Tag for Covalent Protein Labeling, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany, 2008. 60. A. Gautier, A. Juillerat, C. Heinis, I. R. Corrêa, Jr, M. Kindermann, F. Beaufils and K. Johnsson, Chem. Biol., 2008, 15, 128–136. 61. G. Los, L. Encell, M. McDougall, D. Hartzell, N. Karassina, C. Zimprich, M. Wood, R. Learish, R. Ohana, M. Urh, D. Simpson, J. Mendez, K. Zimmerman, P. Otto, G. Vidugiris, J. Zhu, A. Darzins, D. Klaubert, R. Bulleit and K. Wood, ACS Chem. Biol., 2008, 3, 373–382. 62. G. Lukinavicius, K. Umezawa, N. Olivier, A. Honigmann, G. Yang, T. Plass, V. Mueller, L. Reymond, I. R. Corrêa, Jr., Z.-G. Luo, C. Schultz, E. A. Lemke, P. Heppenstall, C. Eggeling, S. Manley and K. Johnsson, Nat. Chem., 2013, 5, 132–139. 63. J. Fölling, M. Bossi, H. Bock, R. Medda, C. A. Wurm, B. Hein, S. Jakobs, C. Eggeling and S. W. Hell, Nat. Methods, 2008, 5, 943–945. 64. M. Heilemann, S. van de Linde, M. Schüttpelz, R. Kasper, B. Seefeldt, A. Mukherjee, P. Tinnefeld and M. Sauer, Angew. Chem. Int. Ed., 2008, 47, 6172–6176. 65. G. Lukinavicius, L. Reymond, K. Umezawa, O. Sallin, E. D'Este, F. Göttfert, H. Ta, S. W. Hell, Y. Urano and K. Johnsson, J. Am. Chem. Soc., 2016, 138, 9365–9368. 66. J. B. Grimm, B. P. English, J. Chen, J. P. Slaughter, Z. Zhang, A. Revyakin, R. Patel, J. J. Macklin, D. Normanno, R. H. Singer, T. Lionnet and L. D. Lavis, Nat. Methods, 2015, 12, 244–250. 67. J. B. Grimm, A. K. Muthusamy, Y. Liang, T. A. Brown, W. C. Lemon, R. Patel, R. Lu, J. J. Macklin, P. J. Keller, N. Ji and L. D. Lavis, Nat. Methods, 2017, 14, 987–994. 68. M.-A. Plamont, E. Billon-Denis, S. Maurin, C. Gauron, F. M. Pimenta, C. G. Specht, J. Shi, J. Querard, B. Pan, J. Rossignol, K. Moncoq, N. Morellet, M. Volovitch, E. Lescop, Y. Chen, A. Triller, S. Vriz, T. Le Saux, L. Jullien and A. Gautier, Proc. Natl. Acad. Sci. U. S. A., 2016, 113, 497–502. 69. F. M. Pimenta, G. Chiappetta, T. Le Saux, J. Vinh, L. Jullien and A. Gautier, Sci. Rep., 2017, 7, 12316. 70. C. Li, M.-A. Plamont, H. L. Sladitschek, V. Rodrigues, I. Aujard, P. Neveu, T. Le Saux, L. Jullien and A. Gautier, Chem. Sci., 2017, 8, 5598–5605. 71. J. R. Babendure, S. R. Adams and R. Y. Tsien, J. Am. Chem. Soc., 2003, 125, 14716–14717. 72. M. N. Stojanovic and D. M. Kolpashchikov, J. Am. Chem. Soc., 2004, 126, 9266–9270. 73. B. A. Sparano and K. Koide, J. Am. Chem. Soc., 2005, 127, 14954–14955. 74. S. Sando, A. Narita, M. Hayami and Y. Aoyama, Chem. Commun., 2008, 3858–3860. 75. T. P. Constantin, G. L. Silva, K. L. Robertson, T. P. Hamilton, K. Fague, A. S. Waggoner and B. A. Armitage, Org. Lett., 2008, 10, 1561–1564. 76. R. Pei, J. Rothman, Y. Xie and M. N. Stojanovic, Nucleic Acids Res., 2009, 37, e59. 77. J. Lee, K. H. Lee, J. Jeon, A. Dragulescu-Andrasi, F. Xiao and J. Rao, ACS Chem. Biol., 2010, 5, 1065–1074. 78. E. V. Dolgosheina, S. C. Y. Jeng, S. S. S. Panchapakesan, R. Cojocaru, P. S. K. Chen, P. D. Wilson, N. Hawkins, P. A. Wiggins and P. J. Unrau, ACS Chem. Biol., 2014, 9, 2412–2420. 79. S. Tyagi, Nat. Methods, 2009, 6, 331–338. 80. B. A. Armitage, Curr. Opin. Chem. Biol., 2011, 15, 806–812. 81. J. Ouellet, Front. Chem., 2016, 4, e144. 82. J. S. Paige, K. Y. Wu and S. R. Jaffrey, Science, 2011, 333, 642–646. 83. W. Song, R. L. Strack, N. Svensen and S. R. Jaffrey, J. Am. Chem. Soc., 2014, 136, 1198–1201.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00041

62

Arnaud Gautier

84. R. L. Strack, M. D. Disney and S. R. Jaffrey, Nat. Methods, 2013, 10, 1219–1224. 85. G. S. Filonov, J. D. Moon, N. Svensen and S. R. Jaffrey, J. Am. Chem. Soc., 2014, 136, 16299–16308. 86. K. D. Warner, M. C. Chen, W. Song, R. L. Strack, A. Thorn, S. R. Jaffrey and A. R. Ferré-D'Amaré, Nat. Struct. Mol. Biol., 2014, 21, 658–663. 87. J. S. Paige, T. Nguyen-Duc, W. Song and S. R. Jaffrey, Science, 2012, 335, 1194. 88. W. Song, R. L. Strack and S. R. Jaffrey, Nat. Methods, 2013, 10, 873–875. 89. M. You, J. L. Litke and S. R. Jaffrey, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, E2756–E2765. 90. C. A. Kellenberger, S. C. Wilson, J. Sales-Lee and M. C. Hammond, J. Am. Chem. Soc., 2013, 135, 4906–4909. 91. N. Svensen and S. R. Jaffrey, Cell Chem. Biol., 2016, 23, 415–425. 92. W. Song, G. S. Filonov, H. Kim, M. Hirsch, X. Li, J. D. Moon and S. R. Jaffrey, Nat. Chem. Biol., 2017, 13, 1187–1194. 93. K. D. Warner, L. Sjekloća, W. Song, G. S. Filonov, S. R. Jaffrey and A. R. Ferré-  D'Amaré, Nat. Chem. Biol., 2017, 13, 1195–1201. 94. P. Wang, J. Querard, S. Maurin, S. S. Nath, T. Le Saux, A. Gautier and L. Jullien, Chem. Sci., 2013, 4, 2865–2873. 95. K. Y. Han, B. J. Leslie, J. Fei, J. Zhang and T. Ha, J. Am. Chem. Soc., 2013, 135, 19033–19038.

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Chapter 4

Optogenetic Reporters for Cell Biology and Neuroscience Wei Zhang and Robert E. Campbell* Department of Chemistry, University of Alberta, Edmonton, Alberta,   T6G 2G2, Canada *E-mail: [email protected]

Table of Contents 4.1.  I ntroduction to Optogenetic Reporters. . . . . . . . . . . . . . . . . . . . . . . . 4.2.  Design Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1.  BiFC. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.2.  Dimerization-dependent FPs. . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.3.  FRET. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.4.  Engineered Allosteric Effects. . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.5.  Enhancement of Intrinsic FP Sensitivities. . . . . . . . . . . . . . . 4.3.  Applications of Optogenetic Reporters. . . . . . . . . . . . . . . . . . . . . . . . 4.3.1.  Cell Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.2.  pH Sensing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.3.  Programmed Cell Death. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.4.  Messenger Molecules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.5.  Protein Kinases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.6.  Membrane Potential. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4.  Conclusions and Future Directions. . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

63

65 66 66 66 68 69 71 72 72 74 75 78 83 86 88 89 89

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

View Online

         

View Online

Optogenetic Reporters for Cell Biology and Neuroscience

65

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

4.1. Introduction to Optogenetic Reporters Optogenetic techniques have become widespread throughout the life-science community since first rising to prominence about a decade ago.1 Compared with conventional pharmacological and electrophysiological techniques, optical methods are relatively non-invasive and provide high spatiotemporal resolution. Furthermore, genetic introduction of optogenetic reporters, although requiring greater up-front investment of time and resources, permits precise targeting of biological events at the molecular level and can greatly facilitate experimental investigations in the long term. The combination of optical and genetic methods, for the first time in history, allows us to interrogate specific biological processes under close to native conditions.2,3 Originally, the term “optogenetics” referred only to the optical control of electrical activity in neurons with microbial opsins expressed heterologously.1 The repertoire of optogenetic control has since been expanded with the development and use of a variety of optogenetic actuators.3–6 In contrast to optogenetic actuators that transduce light into biological stimuli, optogenetic reporters reveal biological events by producing detectable optical readouts due to changes in light absorbance or emission.7–9 Historically, the discovery and extensive engineering of genetically encoded light-emitting proteins, including fluorescent proteins (FPs)10,11 and luciferases,12 have triggered the development of genetically encoded reporters and now still provide the basis for constructing new optogenetic reporters. Advances in FP and bioluminescence technologies are covered in Chapter 3. This chapter focuses on the engineering and applications of FP-based optogenetic reporters. FPs, which generally refer to Aequorea jellyfish green FP (GFP)-like FPs, are a family of proteins that exhibit fluorescence due to an intrinsic chromophore that forms from a tripeptide sequence within the native protein.13 As chromophore formation is a series of self-sufficient reactions,14 FPs are fully genetically encoded fluorophores that exhibit bright fluorescence when expressed in cells.15 One of the most important applications of FPs is to image the localization and dynamics of a particular protein of interest (POI) in real time, which is achieved by simple genetic fusion of the genes encoding FP and POI to create a recombinant chimera. This kind of chimeric protein can be considered the simplest type of optogenetic reporter. To shed light on a wider range of biological processes, researchers have engineered FP-based indicators that can change their inherent fluorescence intensity (i.e. an intensiometric change) or color (i.e. a ratiometric change) in response to a biochemical stimulus. Generally, to generate readout that is responsive to the change of a certain biological parameter, an optogenetic reporter is made of a sensing moiety that responds to a specific biological event and a reporter moiety that is based on an engineered FP variant. The main challenge in designing an optogenetic reporter is to couple the sensing moiety with the reporter moiety, using one or more chimeric polypeptide chains, such that the fluorescence could be tuned intensiometrically or ratiometrically by the change that occurs in the sensing moiety. The following

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

66

Wei Zhang and Robert E. Campbell

section introduces the major design strategies for optogenetic reporters, including bimolecular fluorescence complementation (BiFC),16 dimerization-dependent FPs (ddFPs),17 Förster resonance energy transfer (FRET),18 engineered allosteric effects and enhancement of intrinsic FP sensing abilities.

4.2. Design Strategies 4.2.1. BiFC Protein-fragment complementation is a strategy that has been used to design various optogenetic tools, including both actuators and reporters. In this strategy, a protein is genetically split into two non-functional fragments, which can assemble non-covalently when brought into sufficiently close proximity by the interaction of the genetically fused proteins. Accordingly, this approach has been applied to identify protein interactions by using split protein reporters that could produce proximity-dependent readouts. Some of the reporter proteins that have been successfully split and reconstituted include antibiotic-resistant proteins,19,20 chromogenic or fluorogenic enzymes,20–22 luciferases23–26 and transcription factors.27 Split Aequorea victoria GFP (avGFP) variants were first reported in 2000.28 To use split FPs to detect protein interactions in living cells, the non-fluorescent N- and C-terminal FP fragments are separately fused to candidate POIs that are suspected to interact (Figure 4.1a).16,29,30 FP-based protein complementation is also known as bimolecular fluorescence complementation (BiFC)16 and has been applied to FPs with hues ranging from blue to far red.31–35 The availability of BiFC color variants has allowed simultaneous multicolor imaging of multiple protein interactions.36 BiFC systems must be employed with particular caution and care, as the fragments of most split FPs suffer from poor folding and aggregation and some degree of background complementation is expected. In 2013, Cabantous et al. reported a tripartite BiFC system in which “superfolder” GFP (sfGFP) was split into two small peptides (GFP10 and GFP11) and one large fragment (GFP1–9).37 The complete FP was reconstituted when GFP10 and GFP11 were brought together by interaction of their fused protein partners and subsequently bound to GFP1–9 (Figure 4.1b). The tripartite BiFC system has been applied to detect protein interactions in bacteria and mammalian cells and exhibited larger response and improved protein solubility relative to bipartite systems. 4.2.2. Dimerization-dependent FPs Despite the substantial efforts invested in engineering of split FPs for BiFC, there are some fundamental disadvantages of this strategy, including POI interaction-independent self-assembly, slow chromophore formation and

View Online

67

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

Figure 4.1.  Schematic representations of BiFC and ddFP strategies. (a) Split GFP; (b) tripartite split GFP; (c) ddFP; (d) FPX for intermolecular interaction; (e) FPX for intramolecular conformational changes.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

68

Wei Zhang and Robert E. Campbell

irreversible FP reconstitution. In an effort to overcome some of these limitations, Alford and co-workers developed ddFP technology as a fast and fully reversible alternative.17,38 Instead of structurally splitting one FP into fragments, they engineered a palette of low-affinity dimeric FPs, including green, yellow and red variants, that become brighter upon heterodimerization. Each heterodimeric pair consists of one version designated ddFP-A and one version designated ddFP-B. The ddFP-A is dimly fluorescent in its monomeric state, but becomes significantly brighter upon binding with its non-fluorescent partner ddFP-B (Figure 4.1c). In 2016, Chen et al. engineered an orange ddFP from the red ddFP template by using an ultra-high-throughput screening technology, termed microcapillary single-cell analysis and laser extraction (µSCALE), which allows the measurement of millions of protein variants in parallel.39 Since the dimerization of ddFP-A and ddFP-B is determined by the effective concentration (increased by closer molecular distance), ddFPs can be used as genetically encoded protein tags to report protein interactions. Unlike BiFC, the ddFP approach is particularly suitable for probing dynamic biological processes in real time. Additionally, by combining two different hued copies (e.g. green and red) of ddFP-A and a communal ddFP-B, Ding et al. invented a biosensor design strategy, termed the fluorescent protein exchange (FPX), that permits ratiometric visualization of intermolecular protein interactions (Figure 4.1d) and intramolecular protein conformational changes (Figure 4.1e).40 Another major application of ddFPs is to report the activity of proteases. Examples include ddFP-based biosensors for caspases17,40 and matriptase.41 One concern regarding both BiFC and ddFP is the substantial affinity between the main players of these systems (i.e. split FP fragments or ddFP-A and -B), which leads to background fluorescence and could potentially disturb the functions of targeted proteins. 4.2.3. FRET FRET is the non-radiative energy transfer from one chromophore (donor) to another (acceptor) via dipole–dipole coupling. This process requires that the donor emission spectrum and the acceptor absorbance spectrum are overlapped.18 The FRET efficiency, which is defined as the ratio of energy transfer events to donor excitation events, is determined by several parameters, including the donor fluorescence quantum yield, acceptor extinction coefficient, extent of spectral overlap and the distance and relative orientation between the two chromophores. For a FRET pair composed of two particular chromophores, the first three parameters are fixed, so the FRET efficiency depends only on the distance and relative orientation of the two chromophores. The acceptor can be either a fluorophore or a non-fluorescent chromophore. For FRET pairs with a fluorescent acceptor, a change in FRET efficiency would manifest as a change in the ratio between donor and acceptor fluorescence intensities (ratiometric fluorescence change). If

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

69

the acceptor is a non-fluorescent chromophore, there is only an intensiometric change of the donor fluorescence, which is inversely proportional to FRET efficiency. As a result of the extensive engineering of FP color variants,10,11,13 FRET has become an effective strategy for the development of optogenetic reporters. The commonly used FP-based FRET pairs include, but are not limited to, cyan and yellow fluorescent proteins (CFP–YFP),42–46 mAmetrine–dTomato,47,48 mClover–mRuby2 49 and mCyRFP–mMaroon.50 For most FP-based FRET pairs, the FRET efficiency is most sensitive at donor to acceptor distances between 3 and 9 nm.49,51 Accordingly, fusing FP FRET pairs with a POI or POIs allows the fluorescent detection of a wide variety of molecular behaviors that occur on this length scale. For example, FRET can be used to investigate protein conformational changes (when the donor and acceptor are fused to one POI) and relative protein proximity or interaction (when the donor and acceptor are fused to two different POIs). An important advantage of FRET, as an optogenetic reporter design strategy, is its versatility. This versatility stems from the fact that most biological processes, including analyte fluctuation, enzyme-mediated protein modification and even the change in membrane potential, ultimately involve a protein conformational change or a change in protein proximity or interaction. Furthermore, the ratiometric imaging enabled by FRET is exquisitely sensitive to even small changes in inter-chromophore distance. Ratiometric imaging also helps to correct for variabilities resulting from cell morphology changes, uneven illumination and drifting of focus and sample. Ratiometric FRET imaging also facilitates internal calibration to perform quantitative analysis.52 4.2.4. Engineered Allosteric Effects A third strategy for designing FP-based reporters is to engineer allosteric control of FP fluorescence properties. In this design strategy, protein domains, which change their conformation or oligomerization state in response to biochemical stimuli, are fused with FPs at sites that are structurally adjacent to the chromophore. This is done such that a change in the conformation or interactions of the sensing domain results in a modulation of the chromophore microenvironment and a concomitant change in the chromophore fluorescent intensity or hue. Because only a single FP is involved, reporters based on this strategy are also known as single FP-based reporters. Since the N- and C-termini of FPs are far from the chromophore (which is buried near the center of the β-barrel),53,54 simply fusing the interacting protein to one of the termini typically does not produce the desired allosteric control of fluorescence. Rather, interacting proteins can be inserted into the β-barrel or linked to the termini of a circularly permutated FP (cpFP), in which the original N- and C-termini are joined by a peptide

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

70

Wei Zhang and Robert E. Campbell

linker and new termini have been introduced close to the chromophore.55 In this way, the conformation and oligomerization changes occurring in a sensing domain can effectively modulate the microenvironment of the FP chromophore, leading to changes in quantum yield, extinction coefficient and protonation state. The strategy of allosteric control of FP fluorescence was first reported in 1999 by Baird et al., who created the “camgaroo” calcium ion (Ca2+) indicator by inserting calmodulin (CaM) into avGFP.55 Subsequently, this strategy, using either insertions into FPs or fusions to cpFPs, has been exploited to construct a variety of reporters, including Ca2+ indicators,56–58 indicators for membrane potential,59,60 adenosine triphosphate (ATP)–adenosine diphosphate (ADP) indicators61,62 and a glutamate indicator.63 In some cases, the sensing moiety of a single FP-based reporter can be engineered into the β-barrel of the FP itself by rationally modifying residues that are in proximity to the chromophore and have side-chains directed towards the exterior. Examples include redox reporters, such as rxYFP and reduction– oxidation-sensitive GFPs (roGFPs), with a disulfide bond formed by two close cysteine residues,64–66 and Ca2+ reporters with a Ca2+ binding site engineered on the surface of the FP.67 However, the engineered β-barrels generally suffer from poor specificity, sensitivity and affinity for ligand binding. For example, roGFPs are responsive to multiple oxidants.65 To engineer hydrogen peroxide (H2O2)-specific optogenetic reporters, Belousov et al. genetically inserted cpYFP into the regulatory domain of the prokaryotic H2O2-sensing protein OxyR, producing HyPer, which is exclusively sensitive to H2O2, rather than other common oxidants in cells (e.g. oxidized glutathione and NO).68 HyPer variants with expanded dynamic range and faster kinetics have also been developed by Belousov and co-workers.69,70 Similarly, Fan and co-workers designed TrxRFP, a selective reporter for thioredoxin redox dynamics, by fusing a cpRFP-based redox reporter (roRFP) with Trx1.71,72 The main advantage of single FP-based reporters, relative to FRET-based reporters, is their typically larger intensiometric fluorescence change that facilitate single-channel imaging. Notably, ongoing protein engineering efforts have led to a number of color variants of single FP-based reporters,58,60,73,74 which greatly facilitate multichannel and multiparameter fluorescence imaging. Although the majority of single FP-based reporters are intensiometric, ratiometric reporters (typically excitation ratiometric) have been reported.57,58,61,75,76 The mechanistic principle for ratiometric fluorescence changes is excited-state proton transfer (ESPT).77 The protonated neutral chromophore of an FP becomes more acidic upon excitation, leading to a proton transfer from the phenol moiety of the chromophore to a proximal acidic group. Because of ESPT, both the protonated and deprotonated states of some FP chromophores can be excited with different wavelengths of light, yet exhibit the same wavelength of emission from the deprotonated state. Since the protonated state is excited with a shorter wavelength of light than the deprotonated state, some engineered FPs with enhanced chromophore

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

71

protonation and ESPT can exhibit efficient long Stokes shift fluorescence.78–80 In ratiometric single FP-based reporters, the allosteric effect can modulate the chromophore microenvironment and change the ratio of normal to long Stokes shift fluorescence. Compared with the FRET-based strategy, the allosteric effect-based design strategy is somewhat less versatile, as the engineering of the allosteric effect between the sensing domain and FP has proven to be challenging and highly empirical. However, because of their advantages, single FP-based reporters are often the tools of choice for in vivo imaging applications in cell biology and neuroscience. 4.2.5. Enhancement of Intrinsic FP Sensitivities Although the chromophore of an FP is isolated from the environment due to the cage-like β-barrel, the fluorescence of some FPs is sensitive to certain small ions, such as H+, halides and nitrate, owing to the existence of solvent access channels or proton relay networks in the β-barrel.81–83 However, such intrinsically sensitive FPs tend not to be ideal optogenetic reporters owing to small response in the physiological concentration range. Accordingly, protein engineering efforts including directed evolution and rational design can be used to enhance the natural ion-sensing ability of FPs for biological applications. AvGFP and its analogs are intrinsically sensitive to H+ concentration (pH) and have a built-in sensing mechanism: the protonation and deprotonation of the chromophore phenolate group causes reversible changes in the fluorescence profile.81 In 1998, Miesenböck et al. explored this mechanism and engineered two classes of pH-sensitive GFPs (pHluorins) by substituting several key amino acid residues in the vicinity of the chromophore.84 Owing to their enhanced pH sensitivity and shifted pKas (to ∼7.0), pHluorins undergo large intensiometric (ecliptic) or ratiometric fluorescence changes in response to pH changes in physiologically relevant ranges. YFPs are red-shifted variants of GFPs containing the key substitution T203Y.53 Some pH-sensitive YFP variants82 are responsive to halides and nitrate, largely due to the interaction of these ions with the chromophore and Y203.83,85,86 As the transport of chloride ion (Cl−) across membrane compartments of cells (e.g. plasma membrane and vesicle membranes) occurs in many biophysical processes, the Cl− sensitivity of YFP has been most studied and exploited. At high concentrations, Cl− can bind with YFP at a site close to the chromophore and change the FP fluorescence by shifting its apparent pKa due to electrostatic interactions.86 These YFP variants have been used as FRET acceptors with a CFP donor to create the Clomeleon ratiometric Cl− reporter.87 The Cl−-sensitive YFPs and Clomeleon have been exploited to visualize the Cl− fluctuation in biological processes, such as cystic fibrosis transmembrane conductance regulator (CFTR)-mediated Cl− transport83 and GABA-induced Cl− influx for synaptic inhibition.87 However, the Cl− binding affinities of the original YFP variants are all higher than 100 mM at neutral

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

72

Wei Zhang and Robert E. Campbell

pH, which makes the imaging of Cl− changes in the physiological range (∼4–6 mM) challenging.88 To enhance the Cl− affinity, Grimley et al. performed intensive protein engineering with a high-throughput screening method and identified SuperClomeleon. which has a Cl− affinity of ∼8 mM.89 SuperClomeleon is more suitable than its progenitors for the detection of cellular Cl− fluctuations.

4.3. Applications of Optogenetic Reporters In the past two decades, a tremendous number of FP-based optogenetic reporters have been designed based on the strategies discussed in the previous section. Invariably, the first-reported version of a new optogenetic reporter is not ideal by all criteria and properties and can be considered a prototype reporter. Key reporter properties can include molecular brightness, protein expression, sensitivity and response speed. These properties can often be improved or customized for specific cell-imaging applications, by using protein engineering methods including both rational design and directed evolution. In this way, a first-generation prototype reporter can be further developed into an optimized and widely applicable tool for use in cell biology or neuroscience. In this section is provided an overview of individual families of optogenetic reporters organized by their applications in cell biology. A particular emphasis is placed on optogenetic reporters that have been designed or optimized for neuronal imaging. 4.3.1. Cell Cycle The cell cycle is the series of cellular events that are necessary for cell division. It consists of four sequential phases: G1 phase (first gap phase), S phase (DNA replication), G2 phase (second gap phase) and M phase (mitosis).90 Accurate monitoring of the cell cycle is important in many areas of cell biology research, including studies of tissue maintenance,91 development,92 stem-cell cell renewal93 and tumor growth.94 Accordingly, optogenetic reporters of the cell cycle are of broad utility in cell biology research. The first generation of cell cycle reporter was designated “Fucci” (fluorescent, ubiquitination-based cell cycle indicator).95 This technology relies on the ubiquitylation-mediated, and cell cycle-dependent, proteolysis of certain proteins. In the Fucci system, the sensing domains are the truncated versions of two E3 ligase substrates, Cdt1 and geminin, whose levels oscillate inversely during the cell cycle. Specifically, Cdt1 is accumulated and maintained by cells in the G1 phase. In contrast, geminin increases in concentration in the S phase, stays highest through the S, G2 and M phases, and decreases to almost zero once the cell has entered G1 phase. As the heterologous overexpression of full-length geminin and Cdt1 proteins would disrupt the normal cell division, truncated versions were constructed and

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

73

Figure 4.2.  Schematic representations of Fucci technologies. (a) Fucci; (b) Fucci2; (c) Fucci with the capability to report G1–G0 transition; (d) Fucci4.

characterized in live cells to identify fragments that exhibited minimal interference to cell division and maintained protein oscillation during the cell cycle. The resulting truncation variants of Cdt1 and geminin were fused to orange FP (monomeric Kusabira Orange, mKO2) and green FP (monomeric Azami Green, mAG), respectively. The nucleus of a cell expressing the Fucci reporter system is green fluorescent in the G1 phase, orange fluorescent in the S, G2 and M phases, yellow (overlap of green and red) fluorescent during the G1-to-S phase transition and not fluorescent in the early G1 phase (Figure 4.2a). In the later Fucci2 system,96 mKO2 and mAG were replaced with a red FP (mCherry) and a yellow FP (mVenus) in order to achieve better spectral separation and brighter green fluorescence (Figure 4.2b). Both Fucci and Fucci2 have been widely applied to study cell division in many types of preparations ranging from cultured cells to live embryos and freely behaving mammals.95–97 A limitation of Fucci and Fucci2 is that they do not specifically mark transitions between the proliferative phases of the cell cycle (S, G2 and M) or the transition between the cycling phases and the resting (G0) phase (G1– G0 transition). In 2014, Oki et al. modified the Fucci system to permit the visualization of the G0–G1 transition.98 This was achieved by introducing a third sensing domain, a defective mutant of cyclin-dependent kinase (CDK) inhibitor p27, which has a high protein level in the G0 phase and is degraded in the G1, S, G2 and M phases (Figure 4.2c). Recently, an even more advanced

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

74

Wei Zhang and Robert E. Campbell

Fucci version, Fucci4, was developed for imaging of the four cycling phases (G1, S, G2 and M).99 Fucci4 uses mKO2 fused to truncated Cdt1 to label the G1 phase (as in the original Fucci system) and a green FP (Clover) fused to truncated geminin to label proliferative phases (S, G2 and M). In addition, the S phase is differentiated from other proliferative phases (G2 and M) with a cyan FP (mTurquoise2) fused to a truncated version of human stemloop binding protein (SLBP), which has a high protein level in the G1 and S phases and is degraded in the G2 and M phases. The G2 and M phases are further distinguished from each other through cell morphology changes. In all phases of cell cycle, the fusion of histone H1.0 and a monomeric far-red FP (mMaroon1) is expressed constitutively as an internal reference for fluorescence intensities from other channels (i.e. cyan, green and orange) (Figure 4.2d). As the newest generation of Fucci technology, Fucci4 should be applicable in a wide variety of studies that require monitoring of all four phase transitions in the cell cycle. 4.3.2. pH Sensing In eukaryotic cells, the pH levels of certain subcellular compartments are highly regulated and are distinct from that of the cytoplasm. Furthermore, pH levels can change spatially and temporally to facilitate biochemical reactions that occur in membrane-isolated areas.100 The subcellular pH level is tightly controlled in biological events including, but not limited to, endocytosis, exocytosis and autophagy.101–103 In addition, in neurons, synaptic neurotransmission relies on the secretion of the neurotransmitters from acidic pH presynaptic vesicles into the neutral pH synaptic cleft through exocytosis.104,105 Accordingly, pH transitions are an important phenomenon that can be monitored for studies of synaptic activity. Compared with fluorescent organic dyes, optogenetic reporters are ideal tools to sense pH levels in cell biology and neuroscience studies, because of their inherent compatibility with precise molecular targeting to subcellular components. As discussed in Section 4.2.5, FP-based pHluorin pH indicators, including ratiometric and ecliptic pHluorins, were engineered based on the intrinsic pH sensitivity of avGFP, by using structure-guided design.84 In 2000, ecliptic pHluorin was further optimized for improved fluorescence in live cells, producing a new variant named superecliptic pHluorin (SEP).106 To date, the pHluorin family remain the most extensively used pH indicators, owing to their high brightness, minimal interference with the tagged protein and high pH sensitivity. They have been targeted to various subcellular compartments, such as secretory vesicles,106–110 endocytic vesicles,109–113 autophagosome114 and mitochondria,115 for monitoring of local pH changes. In neuroscience, the primary use of pHluorins is to report synaptic transmission. To target the lumen of presynaptic vehicles, pHluorins are genetically fused with a vesicle-associated membrane protein such as VAMP2 and synaptophysin.84,116 These targeted chimeras have been successfully used for a broad range of applications ranging from the mechanistic study of single vesicle cycling117

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

75

to the in vivo chronic imaging of synaptic activities in selectively targeted subpopulations of cells.118,119 In addition to presynaptic vesicles, pHluorins have also been targeted to postsynaptic vesicles to study the postsynaptic response to neurotransmitters. For example, pHluorin was fused to the N-terminal extracellular domain of glutamate receptor 1 (GluR1) or GluR2, to visualize the trafficking of α-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) receptors (AMPARs), which are composed of four types of subunits (GluR1–GluR4).120 The synaptic insertion and endocytosis of AMPARs on the postsynaptic side plays an important role in the determination of synaptic strength that could regulate long-term potentiation (LTP) and long-term depression (LTD). The pHluorin–GluR1 and pHluorin–GluR2 chimeric proteins are brightly fluorescent at the extracellular surface and almost non-fluorescent in endosomes, allowing spatiotemporal visualization of AMPAR trafficking. In combination with non-optical strategies, pHluorin–GluR1 was applied to elucidate the role of the AAA + ATPase Thorase in the regulation of N-methyl-d-aspartate (NMDA)-stimulated AMPA receptor endocytosis and recycling of GluRs.121 Red FPs with pH dependence, such as mNectarine,122 pHTomato,123 pHRed124 and pHuji,125 have also been developed. When used in conjunction with blue light-activated optogenetic tools, such as pHluorins, green fluorescent Ca2+ indicators and channelrhodopsins (ChRs), they have allowed two-color imaging and all-optical assays.123,125 For example, pHTomato was fused with synaptophysin, producing a presynaptic vesicle-targeted red fluorescent pH indicator, sypHTomato.123 When expressed heterologously in combination with the GFP-based Ca2+ indicator GCaMP3,126 sypHTomato allowed the simultaneous imaging of Ca2+ and neurotransmitter release in nerve terminals. To date, pHuji is the most sensitive red fluorescent optogenetic pH indicator in the physiological pH range.125 It has permitted the detection of single exocytosis and endocytosis events in the red fluorescent channel. In combination with SEP, pHuji was used to perform two-color imaging of clathrin-mediated internalization of both the transferrin receptor (fused with pHuji) and the β2-adrenergic receptor (fused with SEP). 4.3.3. Programmed Cell Death Programmed cell death (PCD) is a set of regulated intracellular responses that lead to the death of cells following certain cellular stresses or external stimuli. Three primary types of cell death have been identified, namely apoptosis, necroptosis and autosis (autophagic cell death).127–129 These types of PCDs can be differentiated by specific morphology changes. However, the morphological characteristics are often variable and difficult to observe in multicellular biological preparations. Accordingly, optogenetic reporters based on the molecular mechanisms of different types of PCDs are valuable tools for relative studies.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

76

Wei Zhang and Robert E. Campbell

4.3.3.1.  Apoptosis.  Apoptosis is a highly regulated cell death process, which generally occurs to maintain or modulate cell populations in tissues.127,130 It plays an important role in various processes, including development, regeneration and functioning of the immune system and defensive responses to cell damage. Apoptosis was original defined based on its unique cell morphological characteristics, but many aspects of its molecular mechanism have now been elucidated.130 Cells can undergo apoptosis through three main pathways: the extrinsic pathway, which is initiated by the interaction between certain transmembrane receptors (death receptors) and the corresponding extracellular ligands; the intrinsic pathway, which is in response to intracellular stimuli (e.g. DNA damage, viral infections and free radicals); and the perforin/granzyme pathway, which is mediated by the transmembrane pore-forming molecule perforin and serine protease granzyme, which are secreted by immune cell cytotoxic T lymphocytes. Although apoptosis occurs through complex signaling cascades, all pathways end up with an execution phase that involves the activation of the executioner caspases. Executioner caspases are a family of cysteine proteases that cleave various proteins that are critical to cell survival and activate cytoplasmic endonucleases to degrade DNA in the nucleus.131 Caspase-3 is one of the most powerful executioner caspases and cuts the peptide sequence Asp–Glu–Val–Asp (DEVD).132 The first FP-based reporter for apoptosis was designed to detect the activation of caspase-3 by fusing GFP and BFP via a peptide linker containing DEVD.133 Prior to caspase-3 activation, there is efficient FRET from the BFP donor to the GFP acceptor. Upon activation of caspase-3, the peptide linker is proteolyzed and the free diffusion of the two FPs leads to a complete loss of FRET (Figure 4.3a). This is manifested as a ratiometric change with an enhancement of blue fluorescence and reduction of green fluorescence, when exciting BFP. Since then, a wide variety of FRET pairs, including CFP–YFP,134 mAmetrine–tdTomato, mTFP1–mCitrine,47 LSSmOrange–mKate2 135 and CFP–DsRed,136 have been used to replace GFP and BFP. In some cases, these new FRET pairs provide enhanced photophysical properties and in others they permit specialized applications such as multiparameter imaging. In addition to executioner caspases, such as caspase-3, caspase-6 and caspase-7, there are also initiator caspases (e.g. caspase-8, caspase-9 and caspase-10) that target the inactive proenzyme of executioner caspases and activate them through site-specific proteolysis.131 The substrate specificity of different caspases has been explored in depth132,137 and, based on these studies, FRET-based reporters have been developed to analyze the function of various caspases. Further, ddFPs have also been applied to monitor caspase activity intensiometrically (Figure 4.3b) and ratiometrically (Figure 4.3c and d).17,40 Contributing to the regulation of apoptosis are series of anti-apoptotic and pro-apoptotic proteins that suppress and trigger apoptosis, respectively.138 Since the function of an anti- or pro-apoptotic protein is often pathway specific, the use of FP-based reporters to visualize their activity can reveal the activation of certain pathway during apoptosis. For example, to investigate

View Online

77

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

Figure 4.3.  Schematic representations of optogenetic reporters for programmed cell death. (a) FRET-based reporters for caspases; (b) ddFP-based intensiometric reporter for caspases; (c, d) ddFP-based ratiometric reporters for caspases; (e) FRETbased reporters for the cleavage of Bid protein.

glutamate excitotoxicity-induced apoptosis in neurons, Ward et al. generated a reporter by fusing CFP and YFP to the N- and C-termini of Bcl-2 homology domain (BH) 3-only protein Bid (Figure 4.3e).139 Bid protein is a pro-apoptotic protein, which is cleaved by caspase-8 upon activation of apoptosis-related transmembrane receptors. Cleaved Bid translocates to the surface of

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

78

Wei Zhang and Robert E. Campbell

mitochondria, where it initiates the evolutionarily conserved mitochondrial apoptotic pathway common to both extrinsic and intrinsic pathways.140 The recombinant protein, CFP–Bid–YFP, allows the direct observation of Bid protein cleavage and translocation in neurons, in response to either tumor necrosis factor-α (TNFα) or glutamate exposure. The different behaviors of Bid protein in death receptor-induced and glutamate excitotoxicity-induced apoptosis were revealed. 4.3.3.2.  Non-apoptotic PCD.  Although apoptosis is the most thoroughly studied pathway of PCD, the mechanism and function of other types of PCD have also been attracting an increasing level of attention. However, there have been relatively few reports on the development of optogenetic reporters for non-apoptotic PCD. The only example is a reporter for pyroptosis, which is an inflammatory form of PCD that occurs only in immune cells.141 Immune cells infected with intracellular pathogens will initiate pyroptosis, release pro-inflammatory cytokines, burst and die. The released cytokines trigger recruitment of other immune cells to the site of infection. In this process, the proenzyme form of caspase-1 is recruited to the inflammasome, where it oligomerizes and becomes activated through self-cleavage.142 Caspase-1 induces a lytic form of cell death and activates the pro-inflammatory cytokines of IL-1β and IL-18 via site-specific cleavage. Liu et al. engineered a FRETbased indicator for caspase-1 activation by fusing ECFP and Venus (YFP) through a peptide linker containing the consensus peptide sequence cleaved by caspase-1, YVAD.143 As caspase-1 activation is not involved in other PCD pathways, the indicator was used to report specifically on the inflammasome signaling response during pyroptosis. Overall, there remain numerous opportunities for developing additional optogenetic reporters for other types of non-apoptotic PCD-like necroptosis and autosis. 4.3.4. Messenger Molecules Extracellular signaling molecules (first messengers) and intracellular signaling molecules (second messengers) are relatively small chemical entities (ions, small organic molecules or peptides) that play central roles in various signaling pathways. First messengers such as hormones and neurotransmitters transmit signals between cells via the interaction with specific protein receptors.144 Second messengers are produced during intracellular signaling cascades resulting from binding of a first messenger to a cell surface receptor.145 Examples of second messengers include cyclic nucleotides (e.g. cAMP and cGMP), inositol triphosphate (IP3), diacylglycerol (DAG) and Ca2+. A number of optogenetic reporters have been developed to monitor the level of various messenger molecules. The engineering of indicators for small chemical entities generally relies on protein domains that bind the entity at physiologically relevant concentrations. The conformational changes of the binding protein must be coupled with the fluorescence properties of a genetically fused FP(s) through one of the strategies discussed

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

79

in Section 4.2. Optogenetic reporters for important messenger molecules such as glutamate, cAMP, cGMP, IP3, DAG and Ca2+ have been developed and extensively used in cell biology and neuroscience.146–152 The following section focuses on the development and applications of optogenetic reporters for neurotransmitters and Ca2+, as representative illustrations of optogenetic reporter development. 4.3.4.1.  Neurotransmitters.  Neurotransmitters are the chemical messengers that transmit signals across a chemical synapse from a neuron to another neuron or non-neuronal cell. Neurotransmitters include various types of molecules, such as amino acids (e.g. glutamate, glycine and GABA), monoamines (e.g. dopamine and serotonin), peptides and gasotransmitters (e.g. NO and CO). Among them, glutamate is a major excitatory neurotransmitter in neural systems.153 In 2005, two genetically encoded reporters for glutamate, fluorescent indicator protein for glutamate (FLIPE)154 and glutamate-sensitive fluorescent reporters (GluSnFR),146 were reported. They share a similar design, composed of a CFP and a YFP fused to the N- and C-termini, respectively, of GltI (also known as ybeJ, a periplasmic glutamate-binding protein from Escherichia coli). Upon binding to glutamate, the recombinant proteins undergo conformational changes in the glutamate-binding domain, leading to a change in FRET efficiency. In 2008, Hires et al. reported the systematic optimization of the linkers between the FPs and GltI of GluSnFR and adjusted the glutamate affinity by rationally mutating GltI residues known to coordinate ligand binding.155 The resulting variant, SuperGluSnFR, allowed quantitative measurements of glutamate propagation over time after synaptic release and was used to study the regulation of glutamate spillover and reuptake in dissociated hippocampal neurons. In 2013, Marvin et al. reported an intensity-based glutamate-sensing fluorescent reporter (iGluSnFR) that had been engineered using the allosteric design strategy to modulate the fluorescence of cpGFP.63 Because of its high brightness, sensitivity, membrane trafficking and ligand specificity, it has been successfully applied for in vivo imaging of glutamate release in worms, zebrafish and mice. As described in Section 4.3.2, the release of neurotransmitters can be monitored using presynaptic vesicle-targeted pH indicators. However, to dissect the contributions of particular neurotransmitters to synaptic transmission, optogenetic reporters that respond directly to specific neurotransmitters are required. To date, FP-based indicators for the major neurotransmitters, with the important exception of glutamate, are still lacking. To create a more comprehensive toolbox of neurotransmitter indicators, protein engineers are now focused on two groups of sensing proteins. The first group are bacterial periplasmic binding proteins (PBPs). PBPs are known to bind a variety of molecules that serve as neurotransmitters, including GABA, acetylcholine and glycine.156 The second group are the native receptors of neurotransmitters, which necessarily bind specific neurotransmitters at the physiologically relevant concentration and undergo conformational changes to initiate the downstream signaling.157 By using the design principles discussed in this

View Online

80

Wei Zhang and Robert E. Campbell

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

chapter, genetically encoded indicators for various neurotransmitters will likely be reported in the near future. 4.3.4.2.  Calcium Ion.  Ca2+ is a ubiquitous signaling ion – it regulates the activity of various proteins allosterically, acts in a wide range of signaling pathways and, accordingly, is relevant to practically all fundamental cellular processes.158 Owing to the fundamental importance and ubiquity of Ca2+, genetically encoded Ca2+ indicators (GECIs) have been one of the most widely used families of optogenetic tools. GECIs have been used not only to monitor Ca2+ as a second messenger in signaling pathways, but also to detect neuronal action potentials, which are associated with an increase in cytosolic Ca2+ due to opening of voltage-gated Ca2+ channels.159,160 4.3.4.2.1  FRET-based GECIs.  The first class of GECIs, cameleons, were reported by Miyawaki and co-workers in the lab of Roger Y. Tsien in 1997.152 Cameleons are one of the first examples of the use of FRET between FPs as an optogenetic reporter design strategy. They consist of one donor FP (BFP or CFP) and one acceptor FP (GFP or YFP), which are fused to the N- and C-termini of a Ca2+-sensing domain containing a vertebrate CaM and the CaM-binding region of chicken myosin light-chain kinase (M13). In response to an increased Ca2+ concentration, these recombinant proteins act like the lizard chameleon by changing colors and moving a long tongue (M13) into and out of their mouth (CaM) (Figure 4.4a). When M13 and CaM are in the Ca2+ free state, the FRET efficiency between the donor and acceptor FPs is low. In the Ca2+-bound state, CaM and M13 bind and the compact form of CaM– M13 leads to an increased FRET efficiency. This manifests itself spectrally as stronger acceptor fluorescence upon excitation of the donor. Compared with the original cameleons (with a BFP–GFP FRET pair), yellow cameleons (YCs, with a CFP–YFP FRET pair) have a number of advantages, including improved brightness and reduced auto-fluorescence and phototoxicity due to the longer wavelength excitation light. However, the YFP variant used by the original YCs was pH sensitive in the physiological pH range. In 1999, the same group developed improved versions of YCs with decreased pH sensitivity, by introducing mutations in the YFP acceptor.161 Among the resulting variants, YC2.1, which has a higher affinity to Ca2+, was used to image Ca2+ dynamics in dissociated hippocampal neurons. A major limitation of the first generation of cameleons was a small ratiometric change. For example, there is only about a twofold emission ratio (yellow-to-cyan) increase on going from the Ca2+-free state to the Ca2+-bound state for the most sensitive variants, YC2.1 and YC3.1.161 In 2004, Nagai et al. described their efforts to optimize the relative orientation of donor and acceptor FPs in YC by replacing YFP with various cpYFPs.45 The best variant, YC3.6, adopted a YFP variant Venus, which is circularly permuted at position 173 and exhibited an increased ratiometric increase (about sixfold) upon Ca2+ binding. In 2010, Nagai's group introduced YC-nanos, which are a

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

81

Figure 4.4.  Schematic representations of three major classes of GECIs. (a) Yellow cameleon; (b) TnC-based GECI, Twitch; (c) single FP-based GECI.

series of YCs with further improved Ca2+ response (up to ∼15-fold ratiometric change) and high Ca2+ affinity.162 The dissociation constant (Kd) values for Ca2+ binding range from 15 to 140 nM, allowing more reliable Ca2+ imaging for biological events that involve subtle Ca2+ changes at low concentrations (e.g. neuronal firing). One concern about using GECIs is the possibility of disturbing normal cell physiology owing to the high levels of expression of a protein that contains the important signaling molecule CaM and contributes to buffering of Ca2+. Miyawaki et al. demonstrated that the overexpression of cameleons does not perturb CaM-dependent signaling.161 However, many researchers remain concerned about artifacts arising from potential cross-reactivity between endogenous proteins and cameleons. To address this concern, Heim and Griesbeck developed a troponin C (TnC)-based GECI, named TN-L15.163 Compared with CaM–M13, TnC should interact with fewer endogenous proteins. Like cameleons, TnC-based GECIs have been engineered extensively to tune the Ca2+ response in terms of magnitude, Ca2+ binding affinity and speed.164,165 The most recent versions are the “Twitch” family based on Opsanus TnC (Figure 4.4b). Compared with traditional TnC-based GECIs, the Twitch family have a reduced number of Ca2+ binding sites, enhanced ratiometric changes (up to 10-fold) and a wide range of Ca2+ affinities (Kd from 150 nM to 257 µM).166 The in vivo performance of Twitch has been demonstrated by Ca2+ imaging in the brain and lymph nodes of mice.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

82

Wei Zhang and Robert E. Campbell

4.3.4.2.2  Single FP-based GECIs.  Single FP-based GECIs are engineered by coupling the conformational change of a Ca2+ sensing domain to modification of the microenvironment of an FP chromophore through an allosteric effect. Baird et al. first explored this strategy and developed camgaroo1 by inserting CaM into the β-barrel of YFP variant at position 145.55 Owing to the CaM insertion, the chromophore of camgaroo1 was predominantly protonated in the Ca2+-free state. Upon binding to Ca2+, CaM transitions to a more compact conformation, allosterically leading to deprotonation of the chromophore. This change in protonation state manifests as an increase in blue light-induced fluorescence and a reduction in violet light-induced fluorescence (as explained in Section 4.2.4). Later, Griesbeck et al. optimized camgaroo1 by replacing YFP with Citrine, a YFP variant with fast maturation in cells and reduced sensitivity to Cl− and pH.43 In 2001, both Nakai et al.56 and Nagai et al.57 reported a similar new design of single FP-based GECIs (Figure 4.4c). This new design consisted of a cpFP with CaM binding peptide and CaM fused at the N- and C-termini, respectively. The resulting recombinant proteins are pericam, which used cpYFP and a Ca2+/CaM-binding peptide M13,57 and GCaMP1, which uses cpGFP and a Ca2+/CaM-binding peptide derived from the smooth muscle myosin light-chain kinase (RS20).56 Compared with camgaroo, both pericam and GCaMP1 exhibited a higher signal-to-noise ratio and higher affinity for Ca2+. In the years since, multiple generations of GCaMPs, from GCaMP1.6 to GCaMP6, have been engineered for enhanced brightness, improved dynamic range, faster response speed and optimal Ca2+ binding affinity.126,167–171 One important breakthrough that enabled further development of the GCaMP family was the determination of the X-ray crystal for GCaMP2 in the Ca2+-bound state.172,173 Based on structural studies, the Ca2+ sensing mechanism of GCaMP2 was elucidated. The X-ray structure of GCaMP2 revealed that the deprotonation of the chromophore upon Ca2+ binding is facilitated by the positively charged side-chain of residue R377 in CaM, which is positioned near the phenol group of the chromophore and stabilizes the deprotonated state. In 2009, Tian et al. reported GCaMP3 using rational design guided by the crystal structure of GCaMP2.126 Compared with GCaMP2, GCaMP3 exhibited threefold increased brightness and intensiometric change and higher Ca2+ binding affinity. These improvements proved critical for enabling reliable imaging of neural activities in intact animals, such as worms, flies and mice. The latest generation of GCaMP is GCaMP6, which is the result of extensive structure-based mutagenesis and neuron-based screening.171 GCaMP6 outperforms other reporters, including both genetically encoded indicators and synthetic dyes, for Ca2+ imaging in a very wide variety of biological preparations, ranging from cultured neurons to freely behaving animals. In 2011, Zhao et al. reported a series of GECIs, designated GECOs, which, for the first time, expanded the color palette of single FP-based GECIs beyond green.58 To perform effective directed evolution of the GECO prototypes, they designed a high-throughput screening method to detect

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

83

Ca2+ response in bacteria colonies. This was achieved by targeting protein into the periplasmic space of E. coli, where the Ca2+ concentration could be manipulated experimentally. The blue, green and red GECOs allowed simultaneous multicolor Ca2+ imaging in different subcellular compartments. Of these reporters, the RFP-based GECO R-GECO1 represented the most important advance. Red-shifted indicators should permit improved Ca2+ imaging in deep tissue (within hundreds of micrometers),174 as there is less scattering and autofluorescence at longer wavelengths. Later, an RFP mRuby-based GECI, RCaMP, was reported.74 The dynamic range of RCaMP is smaller than that of R-GECO1 and its descendants.58,73,74 However, an advantage of RCaMP is that it is preferred for use in combination with blue light-activatable optogenetic actuators to allow simultaneous optical control and recording. The R-GECO variants tend to be less suitable for such applications, as they can exhibit significant photochromism under strong blue light illumination.73,74 Like the GCaMP family, both R-GECO and RCaMP have been optimized for better performances in neurons, leading to jRGECO and jRCaMP, respectively.174 Beyond optimizing for reporting of neuronal activity, GECIs have also been engineered to have altered photophysical and biophysical properties to address a wider variety of problems in cell biology. For example, in 2014 Wu et al. reported an R-GECO-based long Stokes shift GECI, REX-GECO, which exhibits excitation ratiometric changes upon Ca2+ binding.76 Another important direction has been the development of low-affinity GECIs (Kd >1 µM) for Ca2+ imaging in subcellular compartments [e.g. endoplasmic reticulum (ER), sarcoplasmic reticulum (SR) and mitochondria] where Ca2+ is present at high concentration.175–177 Examples include, but are not limited to, FRET-based GECIs, such as cameleon-er,152 D1ER178 and D4cpv,179 and single FP-based GECIs, such as CEPIA,180 LAR-GECO181 and GCaMPer.182 One unique example of a Ca2+-sensitive FP is CaMPARI (Ca2+-modulated photoactivatable ratiometric integrator),183 which is based on the greento-red photoconvertible FP mEOS2.184 CaMPARI exhibits Ca2+-dependent fluorescence in both green and red states. More importantly, the photoconversion of CaMPARI is ∼20 times faster in the Ca2+-saturated state than in the Ca2+-free state, which allows the illumination-dependent integration of Ca2+ signaling activity. As a Ca2+ integrator, CaMPARI has been applied in vivo to mark nerve cells that are active over a time frame of seconds to minutes. 4.3.5. Protein Kinases Protein post-translational modification is another ubiquitous mechanism to propagate intracellular stimuli through signaling cascades. Protein phosphorylation is the most studied form of post-translational modification and is mediated by protein kinases that specifically transfer the γ-phosphate of ATP to certain amino acid side-chains of the protein substrates.185 Site-specific phosphorylation alters the behaviors of the modified proteins by changing

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

84

Wei Zhang and Robert E. Campbell

protein interactions, conformations or subcellular localizations. These three types of changes have been exploited to build optogenetic reporters for protein kinases. In 2001, Zhang et al. reported the first genetically encoded reporter for a kinase, AKAR (A-kinase activity reporter), which was designed to detect the activation of cAMP-dependent protein kinase A (PKA) in live cells.186 The kinase-sensing domain of AKAR is the fusion of a phosphoamino acid-binding domain (PAABD), 14–3–3τ, and a PKA-specific phosphorylatable peptide sequence, kemptide (LRRASLP). A FRET pair (CFP and YFP) were linked to the N- and C-termini of the kinase sensing domain separately, to report on the conformational change of the kinase-sensing domain. A number of other kinase reporters that use similar schemes have since been reported (Figure 4.5a).187,188 These include, but are not limited to, reporters for PKB,189 PKC,151 ERK,190 Src,191 JNK192 and FAK.193 Recently, Zhang's group reported a new generation of kinase activity reporters based on a phenomenon termed fluorescence fluctuation increase by contact (FLINC).194 Briefly, the fluorescence fluctuations (i.e. blinking) of TagRFP-T could be significantly increased by the proximity of a mutant of the photoswitchable FP Dronpa. Like FRET and ddFPs, FLINC could be used to reveal protein conformational changes and protein interactions reversibly in live cells. The main advantage of FLINC over FRET and ddFP is its compatibility with the existing super-resolution imaging procedure of photochromic stochastic optical fluctuation (pcSOFI).195 The authors constructed FLINC-based kinase activity reporters for PKA and ERK and demonstrated super-resolution imaging of protein kinase activities in live cells. Another strategy to design kinase reporters involves exploiting the conformational change that occurs in the domain of protein kinases upon activation. For example, to construct a reporter for the activation of Ca2+/ CaM-dependent protein kinase II (CaMKII), Takao fused CFP and YFP to the N- and C-termini of full-length CaMKII.196 During the activation of endogenous CaMKII, the recombinant protein, named Camuiα, undergoes a conformational change due to Ca2+/CaM binding and autophosphorylation, leading to a reduced FRET signal (Figure 4.5b). Along with activity-based kinase reporters, the conformational change-based kinase reporters have also been thoroughly reviewed by Zhang and co-workers.187,188 In 2004, Kawai et al. reported single-color fluorescent indicators for protein phosphorylation (sinphos), which were designed by allosteric coupling of the interaction between PAABD and substrate peptide with the fluorescence of a cpFP (Figure 4.5c).197 Beyond this example, the strategy of engineered allosteric effect has not been extensively exploited to construct single FP-based indicators for protein kinases. This has greatly limited progress in the area of simultaneous multicolor imaging for protein kinases. As an alternative approach, Regot et al. introduced the design of kinase translocation reporters (KTRs), which uses a generalizable strategy to convert phosphorylation into a nucleocytoplasmic shuttling event that can be easily tracked

View Online

85

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

Figure 4.5.  Schematic representations of optogenetic reporters for protein kinases. (a) Kinase activity reporters; (b) conformation-based kinase reporters – the example shown is Camuiα; (c) single FP-based kinase reporter, sinphos; (d) kinase translocation reporters (KTRs).

using a single color of FP (Figure 4.5d).198 The sensing domain of KTRs is a hybrid peptide composed of a nuclear localization sequence that is functionally inhibited by phosphorylation and a nuclear exclusion sequence that is functionally enhanced by phosphorylation. Upon kinase activation, the fluorescent KTR construct translocates from nucleus to cytoplasm. The specificity of KTR is determined by the substrate sequence of the hybrid peptide and a kinase docking domain that is fused to the N-terminus of KTR to assist substrate targeting. The authors developed KTRs for protein kinases including JNK, PKA, p38 and ERK and then measured the multiple kinase activities simultaneously.

View Online

86

Wei Zhang and Robert E. Campbell

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

4.3.6. Membrane Potential In neuroscience, direct measurement of the transmembrane potential, the most fundamental neural signal, is essential for studying the information processing in a complex neural circuit. However, it is particularly challenging to record action potentials from large numbers of individual cells, as electrophysiological methods are impractical for this application.199 As an alternative approach, genetically encoded voltage indicators (GEVIs) have been developed to allow high-resolution optical imaging of changes in membrane potential for many cells simultaneously. Compared with GECIs, the most popular optogenetic reporters for neuronal activities, GEVIs can, in principle, provide a direct and real-time measurement of action potentials and subthreshold membrane potential changes that are critical to neuronal functions, but may not lead to any Ca2+ transition. In 1997, Siegel and Isacoff reported the first GEVI, fluorescent Shaker (FlaSh), in which a GFP variant was inserted into the voltage-activated Shaker K+ channel at a site between the transmembrane domain and the C-terminal intracellular domain.200 The voltage-dependent structural rearrangements in the K+ channel modulated the optical properties of GFP, resulting in a reversible fluorescence decrease upon membrane depolarization. In 2001, a FRET-based GEVI, voltage-sensitive FP (VSFP), was described by Knöpfel and co-workers.201 VSFP consists of the voltage sensing domain (VSD) of the K+ channel Kv2.1 and a CFP–YFP FRET pair that is linked to the C-termini of the VSD. The distance and relative orientation of the two FPs could be modulated by the VSD in response to membrane potential changes, providing a voltage-dependent FRET signal. In addition to these early prototypes, a series of GEVIs were designed based on the voltage sensitivity of K+ or Na+ channels.202,203 However, they generally suffered from poor plasma membrane localization.204 In 2007, Dimitrov et al. described the development of a second-generation VSFP, VSFP2, by substituting the VSD of a K+ channel with the VSD of the Ciona intestinalis voltage-sensitive phosphatase (Ci-VSP).205 Compared with VSFP, VSFP2 exhibits improved membrane trafficking in mammalian cells, allowing reliable optical recording of electrical events in neurons. Following on from this work, the VSDs of voltage-sensitive phosphatases were used as the basis to engineer other FRET-based and single FP-based GEVIs with enhanced voltage response and signal-to-noise ratio.206–209 Despite this progress, most GEVI prototypes were impractical for widespread application owing to relatively small fluorescence changes in response to membrane potential changes. A breakthrough came in 2012, when Jin et al. reported a new Ci-VSP-based GEVI, Arclight, which exhibited a relatively large decrease in fluorescence intensity with membrane depolarization, which greatly facilitated the visualization of single-action potentials and subthreshold changes.210 Arclight uses an SEP mutant (A227D) as its reporter domain. A mechanistic study of Arclight function has provided evidence that the mechanism of voltage sensitivity involves FP dimerization.211 Recently, further

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

87

engineering of Arclight has produced a variant, designated Marina, that increases fluorescence with membrane depolarization.212 In 2014, St-Pierre et al. reported a fluorescent voltage sensor, accelerated sensor of action potentials 1 (ASAP1), with faster kinetics than Arclight.59 ASAP1 was engineered by insertion of a cpGFP variant in an extracellular loop of a VSD. To achieve higher expression in mammalian cells and better plasma membrane localization, ASAP1 uses a VSD from the chicken Gallus gallus. A variety of cpGFP mutants were screened to identify the recombinant protein with an optimal combination of brightness and voltage sensitivity. The main advantage of ASAP1 over Arclight is its rapid response with a time constant of ∼2 ms for both on and off kinetics, which allows high-fidelity recording of action potentials. ASAP2, a further improved version of ASAP1 that exhibits larger fluorescence changes and similar response kinetics, was used for two-photon subcellular imaging in vivo.213 Additionally, in 2016 Abdelfattah et al. reported FlicR1, a fast and red fluorescent GEVI with properties that are comparable to those of ASAP1, based on the Ci-VSP VSD.60 In contrast to the designs of Arclight and ASAP1, FlicR1 contains a cpRFP fused to the C-terminus of the VSD. To optimize the prototype of FlicR1, libraries of thousands of variants were screened in both bacteria and mammalian cells. As an alternative to GEVIS based on chimeras of VSDs and FPs, Kralj and co-workers demonstrated that microbial opsins can also serve as GEVIs for fluorescent imaging of voltage transients in neurons.214,215 They discovered that Arch-3 exhibited voltage-dependent fluorescence changes [excitation maximum (λex) at 560 nm and emission maximum (λem) at 690 nm], in addition to its long-known function as an outward proton pump. By mutating one key residue required for proton transfer, they generated Arch (D95N), a fluorescent opsin-based voltage indicator. However, the utility of Arch (D95N) was limited owing to its relatively low fluorescent brightness and slow off-time constant of 41 ms. Hochbaum et al. used directed evolution to improve the photophysical characteristics of Arch (D95N), leading to two variants named QuasAr1 and QuasAr2, which showed increased brightness and voltage sensitivity and responded to membrane potential changes on the sub-millisecond time scale.216 They combined the QuasAr reporters with a blue-shifted channelrhodopsin (λex at 460 nm) to permit all-optical control and detection of membrane voltage with minimal cross-talk. The combination was named Optopatch, because of its functional similarity with patch clamp electrophysiological techniques. To improve the brightness of Arch-based GEVIs further, in 2014 Zou et al. explored the use of voltage-dependent electrochromic FRET (eFRET) from an FP donor to a QuasAr acceptor.217 In this strategy, an FP is fused to the C-terminus of QuasAr. As the absorption of QuasAr increases upon membrane depolarization, the fluorescence of the FP is quenched by QuasAr in a voltage-dependent manner. eFRET successfully combined the fast speed of QuasAr2 with the brightness of FPs, allowing fast (1 kHz acquisition frequency) imaging of single action potentials at a relative low illumination intensity.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

88

Wei Zhang and Robert E. Campbell

Similarly, also in 2014, Gong et al., developed eFRET-based GEVIs by fusing yellow and orange FPs to a mutant of Leptosphaeria maculans (Mac) rhodopsin.218 In 2015 the same group described new eFRET-based GEVIs, in which mutants of Acetabularia acetabulum (Ace) rhodopsin were fused with the bright mNeonGreen GFP.219 The resulting recombinant proteins are brighter and faster (about five-fold) than Mac rhodopsin-based GEVIs and have been used to record neuron spikes at acquisition frequencies as high as 5 kHz in live animals, including mice and flies.

4.4. Conclusions and Future Directions This chapter has provided an overview of the design strategies for the engineering of optogenetic reporters. In addition, a number of application areas have been discussed in detail, including optogenetic visualization of the cell cycle, pH levels, programmed cell death, neurotransmitter release, Ca2+ dynamics, protein kinase activity and changes in membrane potential. The examples provided do not represent an exhaustive list of all optogenetic tools of relevance to cell biology and neuroscience. However, they do provide a clear view of the scope of optogenetic reporters currently available and illustrate general trends for the future of optogenetic imaging. Future trends will include the development of new classes of optogenetic reporters for biological events that have not previously been visualized using genetically encoded tools, exploitation of a wider range of optical phenomena to achieve optogenetic sensing and an increasing use of genetically encoded near-infrared (NIR) fluorophores. So far, optogenetic reporters have been mainly designed to allow the visualization of the concentration of biologically relevant molecules (e.g. second messengers and neurotransmitters) and the activation of particular enzymes (e.g. kinases). As discussed in this review, the corresponding design strategies are well established. Much less well established are design strategies for detecting biophysical parameters. Although there are a number of design strategies for reporters of membrane potential, reporters for imaging of most other biophysical parameters, such as tension, pressure, temperature, crowding, surface roughness and viscosity, are not nearly as advanced or are non-existent. Notably, some pioneering work has been carried out to engineer optogenetic reporters for protein tension,220 temperature221,222 and crowding.223 Optogenetic reporters that allow us to visualize these and other biophysical parameters with high spatiotemporal resolution will greatly facilitate our understanding of the mechanisms and roles of physical effects in biological systems. Recent years have seen the effective exploitation of new and unexpected FP optical phenomena, including FLlNC194 and primed conversion.224 These phenomena have been exploited to enable novel bioimaging techniques that were previously impossible or impractical. It is certain that these phenomena will be increasingly used for innovative imaging approaches and

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

89

it reasonable to expect that additional new phenomena, which will enable unprecedented new applications, have yet to be discovered. Finally, it is reasonable to expect that an important trend for the near future will be the increasing use of in vivo optogenetic imaging (with tissue depths ranging from ∼0.1 to 1 cm) with wavelengths of light in the NIR optical window (∼650–1000 nm). Light in this region exhibits lower scattering and phototoxicity than visible light, and also minimal absorption by hemoglobin, melanin and water within mammalian tissue.225 In recent years, various NIR FPs have been engineered from bacterial phytochrome photoreceptors226–229 and the allophycocyanin α-subunit of cyanobacteria.230 The NIR FPs have been used as simple optogenetic reporters to track protein location and dynamics in live cells and as genetically encoded contrast agents for deep-tissue fluorescence imaging. However, despite the ever-growing repertoire of NIR FPs and efforts to build reporters based on them,229,231–233 the development of NIR optogenetic reporters has lagged far behind the development of optogenetic reporters based on GFP homologs. In the coming years, there is very likely to be an increasing number of FRET-based strategies that incorporate NIR donors and acceptors and single NIR FP-based optogenetic reporters. Continuing advances in this direction, and the other two discussed above, will push forward the frontiers of optogenetic reporter technology.

Acknowledgements The authors acknowledge support from the Natural Sciences and Engineering Research Council of Canada (NSERC) (RGPIN 288338-2010), the Canadian Institutes for Health Research (CIHR) (MOP 123514 and FS 154310), a Brain Canada Platform Support Grant and Alberta Innovates Technology Futures (AITF).

References

1. K. Deisseroth, G. Feng, A. K. Majewska, G. Miesenbock, A. Ting and M. J. Schnitzer, J. Neurosci., 2006, 26, 10380–10386. 2. K. Deisseroth, Nat. Methods, 2011, 8, 26–29. 3. E. Pastrana, Nat. Methods, 2011, 8, 24–25. 4. D. Tischer and O. D. Weiner, Nat. Rev. Mol. Cell Biol., 2014, 15, 551–558. 5. X. X. Zhou, M. Pan and M. Z. Lin, Front. Mol. Neurosci., 2015, 8, 37. 6. K. Zhang and B. Cui, Trends Biotechnol., 2015, 33, 92–100. 7. S. C. Alford, J. Wu, Y. Zhao, R. E. Campbell and T. Knöpfel, Biol. Cell, 2013, 105, 14–29. 8. M. Tantama, Y. P. Hung and G. Yellen, in Progress in Brain Research, Elsevier, 2012, vol. 196, pp. 235–263. 9. Y. Zhao and R. E. Campbell, in New Techniques in Systems Neuroscience, ed. A. D. Douglass, Springer International Publishing, Cham, 2015, pp. 57–96. 10. M. W. Davidson and R. E. Campbell, Nat. Methods, 2009, 6, 713–717.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

90

Wei Zhang and Robert E. Campbell

11. E. A. Rodriguez, R. E. Campbell, J. Y. Lin, M. Z. Lin, A. Miyawaki, A. E. Palmer,  X. Shu, J. Zhang and R. Y. Tsien, Trends Biochem. Sci., 2017, 42, 111–129. 12. K. Suzuki and T. Nagai, Curr. Opin. Biotechnol., 2017, 48, 135–141. 13. R. Y. Tsien, Angew. Chem., Int. Ed., 2009, 48, 5612–5626. 14. R. Heim, D. C. Prasher and R. Y. Tsien, Proc. Natl. Acad. Sci. U. S. A., 1994, 91, 12501–12504. 15. M. Chalfie, Y. Tu, G. Euskirchen, W. Ward and D. Prasher, Science, 1994, 263, 802–805. 16. T. K. Kerppola, Annu. Rev. Biophys., 2008, 37, 465–487. 17. S. C. Alford, A. S. Abdelfattah, Y. Ding and R. E. Campbell, Chem. Biol., 2012, 19, 353–360. 18. T. Förster and T. Forster, Radiat. Res., Suppl., 1960, 2, 326. 19. J. N. Pelletier, F. X. Campbell-Valois and S. W. Michnick, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 12141–12146. 20. A. Galarneau, M. Primeau, L.-E. Trudeau and S. W. Michnick, Nat. Biotechnol., 2002, 20, 619–622. 21. F. Rossi, C. A. Charlton and H. M. Blau, Proc. Natl. Acad. Sci. U. S. A., 1997, 94, 8405–8410. 22. J. D. Martell, M. Yamagata, T. J. Deerinck, S. Phan, C. G. Kwa, M. H. Ellisman, J. R. Sanes and A. Y. Ting, Nat. Biotechnol., 2016, 34, 774–780. 23. T. Ozawa, A. Kaihara, M. Sato, K. Tachihara and Y. Umezawa, Anal. Chem., 2001, 73, 2516–2521. 24. R. Paulmurugan and S. S. Gambhir, Anal. Chem., 2003, 75, 1584–1589. 25. I. Remy and S. W. Michnick, Nat. Methods, 2006, 3, 977–979. 26. A. S. Dixon, M. K. Schwinn, M. P. Hall, K. Zimmerman, P. Otto, T. H. Lubben, B. L. Butler, B. F. Binkowski, T. Machleidt, T. A. Kirkland, M. G. Wood, C. T. Eggers, L. P. Encell and K. V. Wood, ACS Chem. Biol., 2016, 11, 400–408. 27. S. Fields and O. Song, Nature, 1989, 340, 245–246. 28. I. Ghosh, A. D. Hamilton and L. Regan, J. Am. Chem. Soc., 2000, 122, 5658–5659. 29. T. K. Kerppola, Nat. Rev. Mol. Cell Biol., 2006, 7, 449–456. 30. M. Morell, A. Espargaro, F. X. Aviles and S. Ventura, Nat. Protoc., 2008, 3, 22–33. 31. Y. J. Shyu, H. Liu, X. Deng and C.-D. Hu, BioTechniques, 2006, 40, 61–66. 32. K. Ohashi, T. Kiuchi, K. Shoji, K. Sampei and K. Mizuno, BioTechniques, 2012, 52, 45–50. 33. G. Jach, M. Pesch, K. Richter, S. Frings and J. F. Uhrig, Nat. Methods, 2006, 3, 597–600. 34. J.-Y. Fan, Z.-Q. Cui, H.-P. Wei, Z.-P. Zhang, Y.-F. Zhou, Y.-P. Wang and X.-E. Zhang, Biochem. Biophys. Res. Commun., 2008, 367, 47–53. 35. J. Chu, Z. Zhang, Y. Zheng, J. Yang, L. Qin, J. Lu, Z.-L. Huang, S. Zeng and Q. Luo, Biosens. Bioelectron., 2009, 25, 234–239. 36. C.-D. Hu and T. K. Kerppola, Nat. Biotechnol., 2003, 21, 539–545. 37. S. Cabantous, H. B. Nguyen, J.-D. Pedelacq, F. Koraïchi, A. Chaudhary, K. Ganguly, M. A. Lockard, G. Favre, T. C. Terwilliger and G. S. Waldo, Sci. Rep., 2013, 3, 2854. 38. S. C. Alford, Y. Ding, T. Simmen and R. E. Campbell, ACS Synth. Biol., 2012, 1, 569–575. 39. B. Chen, S. Lim, A. Kannan, S. C. Alford, F. Sunden, D. Herschlag, I. K. Dimov, T. M. Baer and J. R. Cochran, Nat. Chem. Biol., 2016, 12, 76–81. 40. Y. Ding, J. Li, J. R. Enterina, Y. Shen, I. Zhang, P. H. Tewson, G. C. H. Mo, J. Zhang, A. M. Quinn, T. E. Hughes, D. Maysinger, S. C. Alford, Y. Zhang and R. E. Campbell, Nat. Methods, 2015, 12, 195–198.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

91

41. A. C. Mitchell, S. C. Alford, S. A. Hunter, D. Kannan, R. A. Parra Sperberg, C. H. Chang and J. R. Cochran, ACS Chem. Biol., 2018, 13, 66–72. 42. R. Heim and R. Y. Tsien, Curr. Biol., 1996, 6, 178–182. 43. O. Griesbeck, G. S. Baird, R. E. Campbell, D. A. Zacharias and R. Y. Tsien, J. Biol. Chem., 2001, 276, 29188–29194. 44. T. Nagai, K. Ibata, E. S. Park, M. Kubota, K. Mikoshiba and A. Miyawaki, Nat. Biotechnol., 2002, 20, 87–90. 45. T. Nagai, S. Yamada, T. Tominaga, M. Ichikawa and A. Miyawaki, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 10554–10559. 46. G.-J. Kremers, J. Goedhart, E. B. van Munster and T. W. J. Gadella, Biochemistry, 2006, 45, 6570–6580. 47. H. Ai, K. L. Hazelwood, M. W. Davidson and R. E. Campbell, Nat. Methods, 2008, 5, 401–403. 48. Y. Ding, H. Ai, H. Hoi and R. E. Campbell, Anal. Chem., 2011, 83, 9687–9693. 49. A. J. Lam, F. St-Pierre, Y. Gong, J. D. Marshall, P. J. Cranfill, M. A. Baird, M. R. McKeown, J. Wiedenmann, M. W. Davidson, M. J. Schnitzer, R. Y. Tsien and M. Z. Lin, Nat. Methods, 2012, 9, 1005–1012. 50. T. Laviv, B. B. Kim, J. Chu, A. J. Lam, M. Z. Lin and R. Yasuda, Nat. Methods, 2016, 13, 989–992. 51. R. E. Campbell, Anal. Chem., 2009, 81, 5972–5979. 52. Y. Liu, Y. Song, L. Jiang and J. Liao, Front. Biol., 2012, 7, 57–64. 53. M. Ormö, A. B. Cubitt, K. Kallio, L. A. Gross, R. Y. Tsien and S. J. Remington, Science, 1996, 273, 1392–1395. 54. F. Yang, L. G. Moss and G. N. Phillips, Nat. Biotechnol., 1996, 14, 1246–1251. 55. G. S. Baird, D. A. Zacharias and R. Y. Tsien, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 11241–11246. 56. J. Nakai, M. Ohkura and K. Imoto, Nat. Biotechnol., 2001, 19, 137–141. 57. T. Nagai, A. Sawano, E. S. Park and A. Miyawaki, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 3197–3202. 58. Y. Zhao, S. Araki, J. Wu, T. Teramoto, Y.-F. Chang, M. Nakano, A. S. Abdelfattah, M. Fujiwara, T. Ishihara, T. Nagai and R. E. Campbell, Science, 2011, 333, 1888–1891. 59. F. St-Pierre, J. D. Marshall, Y. Yang, Y. Gong, M. J. Schnitzer and M. Z. Lin, Nat. Neurosci., 2014, 17, 884–889. 60. A. S. Abdelfattah, S. L. Farhi, Y. Zhao, D. Brinks, P. Zou, A. Ruangkittisakul, J. Platisa, V. A. Pieribone, K. Ballanyi, A. E. Cohen and R. E. Campbell, J. Neurosci., 2016, 36, 2458–2472. 61. J. Berg, Y. P. Hung and G. Yellen, Nat. Methods, 2009, 6, 161–166. 62. M. Tantama, J. R. Martínez-François, R. Mongeon and G. Yellen, Nat. Commun., 2013, 4, 2550. 63. J. S. Marvin, B. G. Borghuis, L. Tian, J. Cichon, M. T. Harnett, J. Akerboom, A. Gordus, S. L. Renninger, T.-W. Chen, C. I. Bargmann, M. B. Orger, E. R. Schreiter, J. B. Demb, W.-B. Gan, S. A. Hires and L. L. Looger, Nat. Methods, 2013, 10, 162–170. 64. H. Ostergaard, EMBO J., 2001, 20, 5853–5862. 65. C. T. Dooley, T. M. Dore, G. T. Hanson, W. C. Jackson, S. J. Remington and R. Y. Tsien, J. Biol. Chem., 2004, 279, 22284–22293. 66. G. T. Hanson, R. Aggeler, D. Oglesbee, M. Cannon, R. A. Capaldi, R. Y. Tsien and S. J. Remington, J. Biol. Chem., 2004, 279, 13044–13053. 67. S. Tang, H.-C. Wong, Z.-M. Wang, Y. Huang, J. Zou, Y. Zhuo, A. Pennati, G. Gadda, O. Delbono and J. J. Yang, Proc. Natl. Acad. Sci. U. S. A., 2011, 108, 16265–16270.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

92

Wei Zhang and Robert E. Campbell

68. V. V. Belousov, A. F. Fradkov, K. A. Lukyanov, D. B. Staroverov, K. S. Shakhbazov, A. V. Terskikh and S. Lukyanov, Nat. Methods, 2006, 3, 281–286. 69. K. N. Markvicheva, D. S. Bilan, N. M. Mishina, A. Y. Gorokhovatsky, L. M. Vinokurov, S. Lukyanov and V. V. Belousov, Bioorg. Med. Chem., 2011, 19, 1079–1084. 70. D. S. Bilan, L. Pase, L. Joosen, A. Y. Gorokhovatsky, Y. G. Ermakova, T. W. J. Gadella, C. Grabher, C. Schultz, S. Lukyanov and V. V. Belousov, ACS Chem. Biol., 2013, 8, 535–542. 71. Y. Fan, Z. Chen and H. Ai, Anal. Chem., 2015, 87, 2802–2810. 72. Y. Fan, M. Makar, M. X. Wang and H. Ai, Nat. Chem. Biol., 2017, 13, 1045–1052. 73. J. Wu, L. Liu, T. Matsuda, Y. Zhao, A. Rebane, M. Drobizhev, Y.-F. Chang, S. Araki, Y. Arai, K. March, T. E. Hughes, K. Sagou, T. Miyata, T. Nagai, W. Li and R. E. Campbell, ACS Chem. Neurosci., 2013, 4, 963–972. 74. J. Akerboom, N. Carreras Calderón, L. Tian, S. Wabnig, M. Prigge, J. Tolö, A. Gordus, M. B. Orger, K. E. Severi, J. J. Macklin, R. Patel, S. R. Pulver, T. J. Wardill, E. Fischer, C. Schüler, T.-W. Chen, K. S. Sarkisyan, J. S. Marvin, C. I. Bargmann, D. S. Kim, S. Kügler, L. Lagnado, P. Hegemann, A. Gottschalk, E. R. Schreiter and L. L. Looger, Front. Mol. Neurosci., 2013, 6, 2. 75. Y. Zhao, A. S. Abdelfattah, Y. Zhao, A. Ruangkittisakul, K. Ballanyi, R. E. Campbell and D. J. Harrison, Integr. Biol., 2014, 6, 714–725. 76. J. Wu, A. S. Abdelfattah, L. S. Miraucourt, E. Kutsarova, A. Ruangkittisakul, H. Zhou, K. Ballanyi, G. Wicks, M. Drobizhev, A. Rebane, E. S. Ruthazer and R. E. Campbell, Nat. Commun., 2014, 5, 5262. 77. M. Chattoraj, B. A. King, G. U. Bublitz and S. G. Boxer, Proc. Natl. Acad. Sci. U. S. A., 1996, 93, 8362–8367. 78. T. Kogure, S. Karasawa, T. Araki, K. Saito, M. Kinjo and A. Miyawaki, Nat. Biotechnol., 2006, 24, 577–581. 79. K. D. Piatkevich, J. Hulit, O. M. Subach, B. Wu, A. Abdulla, J. E. Segall and V. V. Verkhusha, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 5369–5374. 80. Y. Shen, Y. Chen, J. Wu, N. C. Shaner and R. E. Campbell, PLoS One, 2017, 12, e0171257. 81. W. W. Ward, H. J. Prentice, A. F. Roth, C. W. Cody and S. C. Reeves, Photochem. Photobiol., 1982, 35, 803–808. 82. R. M. Wachter, M.-A. Elsliger, K. Kallio, G. T. Hanson and S. J. Remington, Structure, 1998, 6, 1267–1277. 83. S. Jayaraman, J. Biol. Chem., 2000, 275, 6047–6050. 84. G. Miesenböck, D. A. De Angelis and J. E. Rothman, Nature, 1998, 394, 192–195. 85. R. M. Wachter and S. James Remington, Curr. Biol., 1999, 9, R628–R629. 86. R. M. Wachter, D. Yarbrough, K. Kallio and S. J. Remington, J. Mol. Biol., 2000, 301, 157–171. 87. T. Kuner and G. J. Augustine, Neuron, 2000, 27, 447–459. 88. K. Berglund, W. Schleich, P. Krieger, L. S. Loo, D. Wang, N. B. Cant, G. Feng, G. J. Augustine and T. Kuner, Brain Cell Biol., 2006, 35, 207–228. 89. J. S. Grimley, L. Li, W. Wang, L. Wen, L. S. Beese, H. W. Hellinga and G. J. Augustine, J. Neurosci., 2013, 33, 16297–16309. 90. K. A. Schafer, Vet. Pathol., 1998, 35, 461–478. 91. S. C. Pruitt, A. Freeland and A. Kudla, Stem Cells, 2010, 28, 1250–1259. 92. J. Artus and M. Cohen-Tannoudji, Mol. Cell. Endocrinol., 2008, 282, 78–86. 93. K. W. Orford and D. T. Scadden, Nat. Rev. Genet., 2008, 9, 115–128. 94. M. Malumbres and M. Barbacid, Nat. Rev. Cancer, 2009, 9, 153–166.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

93

95. A. Sakaue-Sawano, H. Kurokawa, T. Morimura, A. Hanyu, H. Hama, H. Osawa, S. Kashiwagi, K. Fukami, T. Miyata, H. Miyoshi, T. Imamura, M. Ogawa, H. Masai and A. Miyawaki, Cell, 2008, 132, 487–498. 96. A. Sakaue-Sawano, T. Kobayashi, K. Ohtawa and A. Miyawaki, BMC Cell Biol., 2011, 12, 2. 97. T. Abe, A. Sakaue-Sawano, H. Kiyonari, G. Shioi, K. -i. Inoue, T. Horiuchi, K. Nakao, A. Miyawaki, S. Aizawa and T. Fujimori, Development, 2013, 140, 237–246. 98. T. Oki, K. Nishimura, J. Kitaura, K. Togami, A. Maehara, K. Izawa, A. SakaueSawano, A. Niida, S. Miyano, H. Aburatani, H. Kiyonari, A. Miyawaki and T. Kitamura, Sci. Rep., 2014, 4, 4012. 99. B. T. Bajar, A. J. Lam, R. K. Badiee, Y.-H. Oh, J. Chu, X. X. Zhou, N. Kim, B. B. Kim, M. Chung, A. L. Yablonovitch, B. F. Cruz, K. Kulalert, J. J. Tao, T. Meyer, X.-D. Su and M. Z. Lin, Nat. Methods, 2016, 13, 993–996. 100. J. R. Casey, S. Grinstein and J. Orlowski, Nat. Rev. Mol. Cell Biol., 2010, 11, 50–61. 101. F. R. Maxfield and D. J. Yamashiro, Adv. Exp. Med. Biol., 1987, 225, 189–198. 102. L.-G. Wu, E. Hamid, W. Shin and H.-C. Chiang, Annu. Rev. Physiol., 2014, 76, 301–331. 103. N. Mizushima, Genes Dev., 2007, 21, 2861–2873. 104. T. C. Sudhof and J. Rizo, Cold Spring Harbor Perspect. Biol., 2011, 3, a005637. 105. S. O. Rizzoli, EMBO J., 2014, 33, 788–822. 106. S. Sankaranarayanan, D. De Angelis, J. E. Rothman and T. A. Ryan, Biophys. J., 2000, 79, 2199–2208. 107. T. Tsuboi and G. A. Rutter, Curr. Biol., 2003, 13, 563–567. 108. S. J. An, C. P. Grabner and D. Zenisek, Nat. Neurosci., 2010, 13, 577–583. 109. J. Balaji and T. A. Ryan, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 20576–20581. 110. D. Jullie, D. Choquet and D. Perrais, J. Neurosci., 2014, 34, 11106–11118. 111. C. J. Merrifield, D. Perrais and D. Zenisek, Cell, 2005, 121, 593–606. 112. M. J. Taylor, D. Perrais and C. J. Merrifield, PLoS Biol., 2011, 9, e1000604. 113. G. A. Yudowski, M. A. Puthenveedu and M. von Zastrow, Nat. Neurosci., 2006, 9, 622–627. 114. I. Tanida, T. Ueno and Y. Uchiyama, PLoS One, 2014, 9, e110600. 115. R. Orij, J. Postmus, A. Ter Beek, S. Brul and G. J. Smits, Microbiology, 2009, 155, 268–278. 116. B. Granseth, B. Odermatt, S. J. Royle and L. Lagnado, Neuron, 2006, 51, 773–786. 117. S. P. Gandhi and C. F. Stevens, Nature, 2003, 423, 607–613. 118. M. Ng, R. D. Roorda, S. Q. Lima, B. V. Zemelman, P. Morcillo and G. Miesenböck, Neuron, 2002, 36, 463–474. 119. T. Bozza, J. P. McGann, P. Mombaerts and M. Wachowiak, Neuron, 2004, 42, 9–21. 120. D.-T. Lin and R. L. Huganir, J. Neurosci., 2007, 27, 13903–13908. 121. J. Zhang, Y. Wang, Z. Chi, M. J. Keuss, Y.-M. E. Pai, H. C. Kang, J. Shin, A. Bugayenko, H. Wang, Y. Xiong, M. V. Pletnikov, M. P. Mattson, T. M. Dawson and V. L. Dawson, Cell, 2011, 145, 284–299. 122. D. E. Johnson, H. -w. Ai, P. Wong, J. D. Young, R. E. Campbell and J. R. Casey, J. Biol. Chem., 2009, 284, 20499–20511. 123. Y. Li and R. W. Tsien, Nat. Neurosci., 2012, 15, 1047–1053. 124. M. Tantama, Y. P. Hung and G. Yellen, J. Am. Chem. Soc., 2011, 133, 10034–10037. 125. Y. Shen, M. Rosendale, R. E. Campbell and D. Perrais, J. Cell Biol., 2014, 207, 419–432.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

94

Wei Zhang and Robert E. Campbell

126. L. Tian, S. A. Hires, T. Mao, D. Huber, M. E. Chiappe, S. H. Chalasani, L. Petreanu, J. Akerboom, S. A. McKinney, E. R. Schreiter, C. I. Bargmann, V. Jayaraman, K. Svoboda and L. L. Looger, Nat. Methods, 2009, 6, 875–881. 127. R. A. Lockshin and Z. Zakeri, Nat. Rev. Mol. Cell Biol., 2001, 2, 545–550. 128. J. Belizário, L. Vieira-Cordeiro and S. Enns, Mediators Inflammation, 2015, 2015, 1–15. 129. Y. Liu and B. Levine, Cell Death Differ., 2015, 22, 367–376. 130. S. Elmore, Toxicol. Pathol., 2007, 35, 495–516. 131. D. R. McIlwain, T. Berger and T. W. Mak, Cold Spring Harbor Perspect. Biol., 2013, 5, a008656. 132. R. V. Talanian, C. Quinlan, S. Trautz, M. C. Hackett, J. A. Mankovich, D. Banach, T. Ghayur, K. D. Brady and W. W. Wong, J. Biol. Chem., 1997, 272, 9677–9682. 133. X. Xu, A. L. Gerard, B. C. Huang, D. C. Anderson, D. G. Payan and Y. Luo, Nucleic Acids Res., 1998, 26, 2034–2035. 134. L. Tyas, V. A. Brophy, A. Pope, A. J. Rivett and J. M. Tavaré, EMBO Rep., 2000, 1, 266–270. 135. D. M. Shcherbakova, M. A. Hink, L. Joosen, T. W. J. Gadella and V. V. Verkhusha, J. Am. Chem. Soc., 2012, 134, 7913–7923. 136. H. Kawai, T. Suzuki, T. Kobayashi, H. Sakurai, H. Ohata, K. Honda, K. Momose, I. Namekata, H. Tanaka, K. Shigenobu, R. Nakamura, T. Hayakawa and T. Kawanishi, J. Pharmacol. Sci., 2005, 97, 361–368. 137. J. Song, H. Tan, H. Shen, K. Mahmood, S. E. Boyd, G. I. Webb, T. Akutsu and J. C. Whisstock, Bioinformatics, 2010, 26, 752–760. 138. M. B. Jarpe, C. Widmann, C. Knall, T. K. Schlesinger, S. Gibson, T. Yujiri, G. R. Fanger, E. W. Gelfand and G. L. Johnson, Oncogene, 1998, 17, 1475–1482. 139. M. W. Ward, M. Rehm, H. Duessmann, S. Kacmar, C. G. Concannon and J. H. M. Prehn, J. Biol. Chem., 2006, 281, 5837–5844. 140. H. Li, H. Zhu, C. Xu and J. Yuan, Cell, 1998, 94, 491–501. 141. T. Bergsbaken, S. L. Fink and B. T. Cookson, Nat. Rev. Microbiol., 2009, 7, 99–109. 142. L. Franchi, T. Eigenbrod, R. Muñoz-Planillo and G. Nuñez, Nat. Immunol., 2009, 10, 241–247. 143. T. Liu, Y. Yamaguchi, Y. Shirasaki, K. Shikada, M. Yamagishi, K. Hoshino, T. Kaisho, K. Takemoto, T. Suzuki, E. Kuranaga, O. Ohara and M. Miura, Cell Rep., 2014, 8, 974–982. 144. E. J. Kodis, R. J. Smindak, J. M. Kefauver, D. L. Heffner, K. L. Aschenbach, E. R. Brennan, K. Chan, K. K. Gamage, P. S. Lambeth, J. R. Lawler, A. K. Sikora, N. R. Vercruysse and C. D. Deppmann, in eLS, ed. John Wiley & Sons, Ltd, John Wiley & Sons, Ltd, Chichester, UK, 2012. 145. A. C. Newton, M. D. Bootman and J. D. Scott, Cold Spring Harbor Perspect. Biol., 2016, 8, a005926. 146. R. Y. Tsien, FEBS Lett., 2005, 579, 927–932. 147. M. Zaccolo, F. De Giorgi, C. Y. Cho, L. Feng, T. Knapp, P. A. Negulescu, S. S. Taylor, R. Y. Tsien and T. Pozzan, Nat. Cell Biol., 2000, 2, 25–29. 148. A. Tanimura, A. Nezu, T. Morita, R. J. Turner and Y. Tojyo, J. Biol. Chem., 2004, 279, 38095–38098. 149. M. Sato, Y. Ueda, M. Shibuya and Y. Umezawa, Anal. Chem., 2005, 77, 4751–4758. 150. A. Honda, S. R. Adams, C. L. Sawyer, V. Lev-Ram, R. Y. Tsien and W. R. G. Dostmann, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 2437–2442.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

95

151. J. D. Violin, J. Zhang, R. Y. Tsien and A. C. Newton, J. Cell Biol., 2003, 161, 899–909. 152. A. Miyawaki, J. Llopis, R. Heim, J. M. McCaffery, J. A. Adams, M. Ikura and R. Y. Tsien, Nature, 1997, 388, 882–887. 153. B. S. Meldrum, J. Nutr., 2000, 130, 1007S–1015S. 154. S. Okumoto, L. L. Looger, K. D. Micheva, R. J. Reimer, S. J. Smith and W. B. Frommer, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 8740–8745. 155. S. A. Hires, Y. Zhu and R. Y. Tsien, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 4411–4416. 156. L. L. Looger and O. Griesbeck, Curr. Opin. Neurobiol., 2012, 22, 18–23. 157. T. G. Smart and P. Paoletti, Cold Spring Harbor Perspect. Biol., 2012, 4, a009662. 158. M. J. Berridge, P. Lipp and M. D. Bootman, Nat. Rev. Mol. Cell Biol., 2000, 1, 11–21. 159. M. J. Berridge, Neuron, 1998, 21, 13–26. 160. R. Kerr, V. Lev-Ram, G. Baird, P. Vincent, R. Y. Tsien and W. R. Schafer, Neuron, 2000, 26, 583–594. 161. A. Miyawaki, O. Griesbeck, R. Heim and R. Y. Tsien, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 2135–2140. 162. K. Horikawa, Y. Yamada, T. Matsuda, K. Kobayashi, M. Hashimoto, T. Matsu-ura, A. Miyawaki, T. Michikawa, K. Mikoshiba and T. Nagai, Nat. Methods, 2010, 7, 729–732. 163. N. Heim and O. Griesbeck, J. Biol. Chem., 2004, 279, 14280–14286. 164. M. Mank, D. F. Reiff, N. Heim, M. W. Friedrich, A. Borst and O. Griesbeck, Biophys. J., 2006, 90, 1790–1796. 165. M. Mank, A. F. Santos, S. Direnberger, T. D. Mrsic-Flogel, S. B. Hofer, V. Stein, T. Hendel, D. F. Reiff, C. Levelt, A. Borst, T. Bonhoeffer, M. Hübener and O. Griesbeck, Nat. Methods, 2008, 5, 805–811. 166. T. Thestrup, J. Litzlbauer, I. Bartholomäus, M. Mues, L. Russo, H. Dana, Y. Kovalchuk, Y. Liang, G. Kalamakis, Y. Laukat, S. Becker, G. Witte, A. Geiger, T. Allen, L. C. Rome, T.-W. Chen, D. S. Kim, O. Garaschuk, C. Griesinger and O. Griesbeck, Nat. Methods, 2014, 11, 175–182. 167. D. F. Reiff, J. Neurosci., 2005, 25, 4766–4778. 168. J. Díez-García, S. Matsushita, H. Mutoh, J. Nakai, M. Ohkura, J. Yokoyama, D. Dimitrov and T. Knöpfel, Eur. J. Neurosci, 2005, 22, 627–635. 169. J. Akerboom, T.-W. Chen, T. J. Wardill, L. Tian, J. S. Marvin, S. Mutlu, N. C. Calderon, F. Esposti, B. G. Borghuis, X. R. Sun, A. Gordus, M. B. Orger, R. Portugues, F. Engert, J. J. Macklin, A. Filosa, A. Aggarwal, R. A. Kerr, R. Takagi, S. Kracun, E. Shigetomi, B. S. Khakh, H. Baier, L. Lagnado, S. S.-H. Wang, C. I. Bargmann, B. E. Kimmel, V. Jayaraman, K. Svoboda, D. S. Kim, E. R. Schreiter and L. L. Looger, J. Neurosci., 2012, 32, 13819–13840. 170. X. R. Sun, A. Badura, D. A. Pacheco, L. A. Lynch, E. R. Schneider, M. P. Taylor, I. B. Hogue, L. W. Enquist, M. Murthy and S. S.-H. Wang, Nat. Commun., 2013, 4, 2170. 171. T.-W. Chen, T. J. Wardill, Y. Sun, S. R. Pulver, S. L. Renninger, A. Baohan, E. R. Schreiter, R. A. Kerr, M. B. Orger, V. Jayaraman, L. L. Looger, K. Svoboda and D. S. Kim, Nature, 2013, 499, 295–300. 172. Q. Wang, B. Shui, M. I. Kotlikoff and H. Sondermann, Structure, 2008, 16, 1817–1827. 173. J. Akerboom, J. D. V. Rivera, M. M. R. Guilbe, E. C. A. Malavé, H. H. Hernandez, L. Tian, S. A. Hires, J. S. Marvin, L. L. Looger and E. R. Schreiter, J. Biol. Chem., 2009, 284, 6455–6464.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

96

Wei Zhang and Robert E. Campbell

174. H. Dana, B. Mohar, Y. Sun, S. Narayan, A. Gordus, J. P. Hasseman, G. Tsegaye, G. T. Holt, A. Hu, D. Walpita, R. Patel, J. J. Macklin, C. I. Bargmann, M. B. Ahrens, E. R. Schreiter, V. Jayaraman, L. L. Looger, K. Svoboda and D. S. Kim, eLife, 2016, 5, e12727. 175. D. Burdakov, O. H. Petersen and A. Verkhratsky, Cell Calcium, 2005, 38, 303–310. 176. E. A. Sobie and W. J. Lederer, J. Mol. Cell. Cardiol., 2012, 52, 304–311. 177. S. Arnaudeau, W. L. Kelley, J. V. Walsh and N. Demaurex, J. Biol. Chem., 2001, 276, 29430–29439. 178. A. E. Palmer, C. Jin, J. C. Reed and R. Y. Tsien, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 17404–17409. 179. A. E. Palmer, M. Giacomello, T. Kortemme, S. A. Hires, V. Lev-Ram, D. Baker and R. Y. Tsien, Chem. Biol., 2006, 13, 521–530. 180. J. Suzuki, K. Kanemaru, K. Ishii, M. Ohkura, Y. Okubo and M. Iino, Nat. Commun., 2014, 5, 4153. 181. J. Wu, D. L. Prole, Y. Shen, Z. Lin, A. Gnanasekaran, Y. Liu, L. Chen, H. Zhou, S. R. W. Chen, Y. M. Usachev, C. W. Taylor and R. E. Campbell, Biochem. J., 2014, 464, 13–22. 182. M. J. Henderson, H. A. Baldwin, C. A. Werley, S. Boccardo, L. R. Whitaker, X. Yan, G. T. Holt, E. R. Schreiter, L. L. Looger, A. E. Cohen, D. S. Kim and B. K. Harvey, PLoS One, 2015, 10, e0139273. 183. B. F. Fosque, Y. Sun, H. Dana, C.-T. Yang, T. Ohyama, M. R. Tadross, R. Patel, M. Zlatic, D. S. Kim, M. B. Ahrens, V. Jayaraman, L. L. Looger and E. R. Schreiter, Science, 2015, 347, 755–760. 184. S. A. McKinney, C. S. Murphy, K. L. Hazelwood, M. W. Davidson and L. L. Looger, Nat. Methods, 2009, 6, 131–133. 185. B. M. Sefton, in Current Protocols in Cell Biology, ed. J. S. Bonifacino, M. Dasso, J. B. Harford, J. Lippincott-Schwartz and K. M. Yamada, John Wiley & Sons, Inc., Hoboken, NJ, USA, 2001. 186. J. Zhang, Y. Ma, S. S. Taylor and R. Y. Tsien, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 14997–15002. 187. J. Zhang and M. D. Allen, Mol. Biosyst., 2007, 3, 759. 188. L. Oldach and J. Zhang, Chem. Biol., 2014, 21, 186–197. 189. M. T. Kunkel, Q. Ni, R. Y. Tsien, J. Zhang and A. C. Newton, J. Biol. Chem., 2005, 280, 5581–5587. 190. M. Sato, Y. Kawai and Y. Umezawa, Anal. Chem., 2007, 79, 2570–2575. 191. A. Y. Ting, K. H. Kain, R. L. Klemke and R. Y. Tsien, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 15003–15008. 192. M. Fosbrink, N.-N. Aye-Han, R. Cheong, A. Levchenko and J. Zhang, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 5459–5464. 193. J. Seong, M. Ouyang, T. Kim, J. Sun, P.-C. Wen, S. Lu, Y. Zhuo, N. M. Llewellyn, D. D. Schlaepfer, J.-L. Guan, S. Chien and Y. Wang, Nat. Commun., 2011, 2, 406. 194. G. C. H. Mo, B. Ross, F. Hertel, P. Manna, X. Yang, E. Greenwald, C. Booth, A. M. Plummer, B. Tenner, Z. Chen, Y. Wang, E. J. Kennedy, P. A. Cole, K. G. Fleming, A. Palmer, R. Jimenez, J. Xiao, P. Dedecker and J. Zhang, Nat. Methods, 2017, 14, 427–434. 195. P. Dedecker, G. C. H. Mo, T. Dertinger and J. Zhang, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 10909–10914. 196. K. Takao, J. Neurosci., 2005, 25, 3107–3112. 197. Y. Kawai, M. Sato and Y. Umezawa, Anal. Chem., 2004, 76, 6144–6149. 198. S. Regot, J. J. Hughey, B. T. Bajar, S. Carrasco and M. W. Covert, Cell, 2014, 157, 1724–1734.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

Optogenetic Reporters for Cell Biology and Neuroscience

97

199. S. B. Kodandaramaiah, G. T. Franzesi, B. Y. Chow, E. S. Boyden and C. R. Forest, Nat. Methods, 2012, 9, 585–587. 200. M. S. Siegel and E. Y. Isacoff, Neuron, 1997, 19, 735–741. 201. R. Sakai, V. Repunte-Canonigo, C. D. Raj and T. Knöpfel, Eur. J. Neurosci., 2001, 13, 2314–2318. 202. G. Guerrero, M. S. Siegel, B. Roska, E. Loots and E. Y. Isacoff, Biophys. J., 2002, 83, 3607–3618. 203. K. Ataka and V. A. Pieribone, Biophys. J., 2002, 82, 509–516. 204. B. J. Baker, H. Lee, V. A. Pieribone, L. B. Cohen, E. Y. Isacoff, T. Knopfel and E. K. Kosmidis, J. Neurosci. Methods, 2007, 161, 32–38. 205. D. Dimitrov, Y. He, H. Mutoh, B. J. Baker, L. Cohen, W. Akemann and T. Knöpfel, PLoS One, 2007, 2, e440. 206. A. Lundby, H. Mutoh, D. Dimitrov, W. Akemann and T. Knöpfel, PLoS One, 2008, 3, e2514. 207. H. Tsutsui, S. Karasawa, Y. Okamura and A. Miyawaki, Nat. Methods, 2008, 5, 683–685. 208. W. Akemann, H. Mutoh, A. Perron, Y. K. Park, Y. Iwamoto and T. Knöpfel, J. Neurophysiol., 2012, 108, 2323–2337. 209. L. Barnett, J. Platisa, M. Popovic, V. A. Pieribone and T. Hughes, PLoS One, 2012, 7, e43454. 210. L. Jin, Z. Han, J. Platisa, J. R. A. Wooltorton, L. B. Cohen and V. A. Pieribone, Neuron, 2012, 75, 779–785. 211. Z. Han, L. Jin, F. Chen, J. J. Loturco, L. B. Cohen, A. Bondar, J. Lazar and V. A. Pieribone, PLoS One, 2014, 9, e113873. 212. J. Platisa, G. Vasan, A. Yang and V. A. Pieribone, ACS Chem. Neurosci., 2017, 8, 513–523. 213. H. H. Yang, F. St-Pierre, X. Sun, X. Ding, M. Z. Lin and T. R. Clandinin, Cell, 2016, 166, 245–257. 214. J. M. Kralj, D. R. Hochbaum, A. D. Douglass and A. E. Cohen, Science, 2011, 333, 345–348. 215. J. M. Kralj, A. D. Douglass, D. R. Hochbaum, D. Maclaurin and A. E. Cohen, Nat. Methods, 2011, 9, 90–95. 216. D. R. Hochbaum, Y. Zhao, S. L. Farhi, N. Klapoetke, C. A. Werley, V. Kapoor, P. Zou, J. M. Kralj, D. Maclaurin, N. Smedemark-Margulies, J. L. Saulnier, G. L. Boulting, C. Straub, Y. K. Cho, M. Melkonian, G. K.-S. Wong, D. J. Harrison, V. N. Murthy, B. L. Sabatini, E. S. Boyden, R. E. Campbell and A. E. Cohen, Nat. Methods, 2014, 11, 825–833. 217. P. Zou, Y. Zhao, A. D. Douglass, D. R. Hochbaum, D. Brinks, C. A. Werley, D. J. Harrison, R. E. Campbell and A. E. Cohen, Nat. Commun., 2014, 5, 4625. 218. Y. Gong, M. J. Wagner, J. Zhong Li and M. J. Schnitzer, Nat. Commun., 2014, 5, 3674. 219. Y. Gong, C. Huang, J. Z. Li, B. F. Grewe, Y. Zhang, S. Eismann and M. J. Schnitzer, Science, 2015, 350, 1361–1366. 220. A. S. LaCroix, K. E. Rothenberg, M. E. Berginski, A. N. Urs and B. D. Hoffman, Methods in Cell Biology, Elsevier, 2015, vol. 125, pp. 161–186. 221. S. Kiyonaka, T. Kajimoto, R. Sakaguchi, D. Shinmi, M. Omatsu-Kanbe, H. Matsuura, H. Imamura, T. Yoshizaki, I. Hamachi, T. Morii and Y. Mori, Nat. Methods, 2013, 10, 1232–1238. 222. M. Nakano, Y. Arai, I. Kotera, K. Okabe, Y. Kamei and T. Nagai, PLoS One, 2017, 12, e0172344. 223. A. J. Boersma, I. S. Zuhorn and B. Poolman, Nat. Methods, 2015, 12, 227–229.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00063

98

Wei Zhang and Robert E. Campbell

224. W. P. Dempsey, L. Georgieva, P. M. Helbling, A. Y. Sonay, T. V. Truong, M. Haffner and P. Pantazis, Nat. Methods, 2015, 12, 645–648. 225. R. Weissleder, Nat. Biotechnol., 2001, 19, 316–317. 226. X. Shu, A. Royant, M. Z. Lin, T. A. Aguilera, V. Lev-Ram, P. A. Steinbach and R. Y. Tsien, Science, 2009, 324, 804–807. 227. D. M. Shcherbakova, M. Baloban and V. V. Verkhusha, Curr. Opin. Chem. Biol., 2015, 27, 52–63. 228. D. Yu, M. A. Baird, J. R. Allen, E. S. Howe, M. P. Klassen, A. Reade, K. Makhijani, Y. Song, S. Liu, Z. Murthy, S.-Q. Zhang, O. D. Weiner, T. B. Kornberg, Y.-N. Jan, M. W. Davidson and X. Shu, Nat. Methods, 2015, 12, 763–765. 229. D. M. Shcherbakova, M. Baloban, A. V. Emelyanov, M. Brenowitz, P. Guo and V. V. Verkhusha, Nat. Commun., 2016, 7, 12405. 230. E. A. Rodriguez, G. N. Tran, L. A. Gross, J. L. Crisp, X. Shu, J. Y. Lin and R. Y. Tsien, Nat. Methods, 2016, 13, 763–769. 231. G. S. Filonov and V. V. Verkhusha, Chem. Biol., 2013, 20, 1078–1086. 232. E. Tchekanda, D. Sivanesan and S. W. Michnick, Nat. Methods, 2014, 11, 641–644. 233. T.-L. To, B. J. Piggott, K. Makhijani, D. Yu, Y. N. Jan and X. Shu, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, 3338–3343.

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

Part III

Light-driven Actuators

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

View Online

         

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

View Online

Chapter 5

Light-driven Actuators: Spatiotemporal Dynamics of Cellular Signaling Processes Yoshibumi Uedaa,b and Moritoshi Sato*a a

Graduate School of Arts and Sciences, The University of Tokyo, 3-8-1   Komaba, Meguro-ku, Tokyo, 153-8902, Japan; bAMED-PRIME, Japan Agency for Medical Research and Development, 1-7-1 Otemachi, Chiyoda-ku,   Tokyo, 100-0004, Japan *E-mail: [email protected]

Table of Contents 5.1.  I ntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 5.2.  Experimental Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 5.2.1.  Plasmid Construction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 5.2.2.  Cell Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 5.2.3.  Bioluminescence Assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 5.2.4.  Dish Coating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 5.2.5.  Confocal Laser Scanning Microscopy Imaging. . . . . . . . . . 105 5.2.6.  Half-life Evaluation of Off Kinetics. . . . . . . . . . . . . . . . . . . . 106 5.3.  Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 5.3.1.  Tuning Switch-off Kinetics and Dimerization   Efficiencies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ��  108 5.3.2.  Assembly Method to Enhance the Performances   of the Magnet System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ��  109 5.3.3.  Kinetic Study of the CAD–Magnet System. . . . . . . . . . . . . . 110

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

101

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

102

Yoshibumi Ueda and Moritoshi Sato

5.3.4.  O  ptical Control of Membrane Morphology by   the CAD–Magnet System . . . . . . . . . . . . . . . . . . . . . . . . . . . ��  112 5.3.5.  The Induction of Cell Membrane Dynamics Using   CAD–Magnet. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ��  112 5.4.  Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

View Online

Light-driven Actuators: Spatiotemporal Dynamics

103

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

5.1. Introduction The first application of channelrhodopsin for the activation of neuronal networks in 2005 resulted in huge interest in optogenetics because this approach harnesses the advantages of light for activating proteins of interest with high spatial and temporal resolution. The landmark study for non-channelrhodopsin-type optogenetic tools was the development of photoactivatable Rac1 (PA-Rac1),1 in which the flavin-binding LOV2 (LOV, light-, oxygen- and voltage-sensing) domain derived from Avena sativa phototropin-1 was utilized to regulate Rac1 activity. The LOV2 domain was fused to the N-terminus of Rac1. Under dark conditions, Rac1 is caged by the LOV2 domain with a Jα rigid helical structure, hence Rac1 does not exert its activity. However, upon irradiation with blue light, Jα turns into a flexible form, and Rac1 downstream effectors such as P21-activated kinase (PAK) can freely bind to Rac1. Using this optogenetic tool, Hahn's group was able to induced lamellipodial protrusions and membrane ruffles around the cell edges after irradiation with blue light. The development of PA-Rac1 opened up a new era for non-channelrhodopsin photoswitches, including the development of a variety of photoreceptors based on Vivid, phytochrome B (PhyB) and cryptochrome 2 (CRY2). Vivid (VVD) is a small, 21 kDa flavin adenine dinucleotide (FAD)-binding protein consisting of a LOV domain and an N-terminal cap derived from Neurospora crassa.2 VVD switches its state from a monomer to a homodimer in a blue light-dependent manner.2–4 Although this property has the potential to allow control of the interaction between proteins of interest, aspects of VVD require improvement prior to its application. For example, when VVD is attached to two different proteins (protein A and protein B) to act as a photoswitch to bring these proteins close together, the photoreceptor induces not only the A–B interaction but also unwanted A–A and B–B interactions (Figure 5.1A), which diminish the level of signal transduction. We set out to overcome this problem by improving the photoswitch.5,6

5.2. Experimental Procedures 5.2.1. Plasmid Construction The mutagenesis of VVD was performed using an overlap extension technique and a multisite-directed mutagenesis kit (MBL) according to the manufacturer's instructions (Agilent Technologies Japan, Tokyo, Japan).7,8 The following obtained mutants were sequenced using an ABI310 genetic analyzer (Applied Biosystems, Foster City, CA, USA): pMag, I52R/M55R; pMagFast1, I52R/M55R/I85V; pMagFast2, I52R/M55R/I74V/I85V; pMagHigh1, I52R/M55R/M135I/M165I; nMag, I52D/M55G; nMagFast1, I52D/M55G/I85V; nMagFast2, I52D/M55G/I74V/I85V; and nMagHigh1, I52D/M55G/M135I/ M165I. cDNAs encoding the mouse CaMKIIα association domain (315–478) and full-length human H ferritin (1–183) were synthesized by GenScript

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

104

Yoshibumi Ueda and Moritoshi Sato

Figure 5.1.  The concept for developing the Magnet system. (A) The VVD as shown in orange rectangles not only induces dimerization between protein A and protein B but also the undesirable A–A and B–B association upon irradiation with blue light. The Magnet system identifies two VVD variants (red and blue rectangles) based on electrostatic charges. The two variants prevent homodimerization between A–A and B–B interactions because of electrostatic repulsion, and induce only heterodimerization. (B) The VVD structure and its key amino acid residues. Upon irradiation with blue light, the flavin–cysteine adduct between Cys108 (green) and flavin adenine dinucleotide (FAD) (red) occurs, which induces a conformational change in the N-terminal α-helix, termed Ncap (yellow), resulting in the homodimerization of VVDs.

(Piscataway, NJ, USA). Both cDNA sequences were humanized. Mouse Tiam1 cDNA (Plasmid #22277) was obtained from Addgene (Cambridge, MA, USA). mCherry, DsRed-Express and DsRed-Express2 were purchased from Clontech (Mountain View, CA, USA). A DH/PH domain (1033–1406) of Tiam1 was prepared by polymerase chain reaction (PCR). All constructs were cloned into pcDNA3.1 (Invitrogen, Carlsbad, CA, USA). 5.2.2. Cell Culture COS-7 and HEK293T cells [American Type Culture Collection (ATCC)] were cultured at 37 °C under 5% CO2 in Dulbecco's Modified Eagle Medium (DMEM) (Invitrogen) supplemented with 10% fetal bovine serum (GIBCO, Carlsbad, CA, USA), 100 units mL−1 of penicillin and 100 mg mL−1 of streptomycin (GIBCO). All cell cultures were performed using these conditions, unless indicated otherwise. 5.2.3. Bioluminescence Assay To screen VVD variants, COS-7 cells were plated at ∼2.0 × 104 cells per well in 96-well black-walled plates (Thermo Fisher Scientific, Waltham, MA, USA) and cultured for 24 h at 37 °C in 5% CO2. The cells were then transfected with Lipofectamine 2000 reagent (Invitrogen) according to the manufac­ turer's protocol. cDNAs encoding Nfluc-X and X(Y)-Cfluc (X and Y stand for VVD variants) were transfected at a 1 : 1 ratio. The total amount of DNA was 0.4 µg per well. Twenty-four hours after transfection, the culture medium was

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

Light-driven Actuators: Spatiotemporal Dynamics

105

replaced with 100 µL of a luciferase assay reagent (Promega, Madison, WI, USA) containing d-luciferin as a substrate. The luciferase assay reagent was prepared by adding 5 mL of Hanks' Balanced Salt Solution (HBSS) (Grand Island Biological, Grand Island, NY, USA) to one vial of lyophilized luciferase assay substrate (Promega). After the cells had been treated with the luciferase assay reagent for 1 h at room temperature, bioluminescence measurements were performed for 10 s at room temperature using a Centro XS3 LB 960 plate-reading luminometer (Berthold Technologies, Bad Wildbad, Germany). Irradiation with blue light was performed for 30 s using an LED light source (470 ± 20 nm) (CCS, Kyoto, Japan) at an intensity of 3 mW cm−2. 5.2.4. Dish Coating Glass-bottomed dishes (Matsunami Glass, Osaka, Japan) were coated with 25 µg mL−1 of human fibronectin (BD Biosciences; Franklin Lakes, NJ, USA) at room temperature for 1 h, then washed twice with 2 mL of Milli-Q water. 5.2.5. Confocal Laser Scanning Microscopy Imaging To obtain confocal images of plasma membrane translocation of assembled pMag activators, HEK293T cells were plated at 3.0 × 104 cells per well and COS-7 cells were plated at 1.5 × 104 cells per well in 35 mm glass-bottomed dishes coated with fibronectin, then cultured for 24 h at 37 °C in 5% CO2. The cells were transfected with cDNAs encoding each assembled pMag activator and nMagHigh1-EGFP-CAAX at a 1 : 1 ratio using X-tremeGENE 9 reagent (Roche, Basel, Switzerland) at 37 °C. The total amount of DNA was 0.15 µg per dish. One day after transfection, the medium was replaced with DMEM culture medium supplemented with 10% FBS and the cells were maintained for 24 h at 28 °C. Before imaging, the culture medium was replaced with HBSS. Imaging was performed using a confocal laser scanning microscope (FV1200) (Olympus, Tokyo, Japan) with a 60× oil objective lens. During the experiment, the temperature of the sample was maintained at 37 °C by a stage-top incubator (Tokai Hit, Fujinomiya, Japan). EGFP, mCherry and DsRed were excited using laser diodes at 473 nm (Olympus) and at 559 nm (NTT Electronics, Yokohama, Japan) and using a solid-state laser at 515 nm (Coherent, Santa Clara, CA, USA). Irradiation with blue light was performed with a laser diode at 473 nm at 78 mW cm−2. To observe morphological changes in the plasma membrane using assembled pMag activators, COS-7 cells were plated at 0.3 × 104 cells per dish on glass-bottomed dishes coated with fibronectin. The cells were transfected with both Lifeact-mCherry-P2A-Tiam1-pMagFast2-CAD and nMagHigh1EGFP-CAAX plasmids at a 1 : 9 ratio using X-tremeGENE 9 reagent. The total amount of DNA was 0.15 µg per dish. Twenty-four hours after transfection, the medium was replaced with DMEM culture medium supplemented with 10% FBS and the cells were maintained for 24 h at 28 °C. Imaging was performed

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

106

Yoshibumi Ueda and Moritoshi Sato

at 37 °C using a confocal laser scanning microscope (LSM710) (Carl Zeiss, Oberkochen, Germany) and a 63× oil objective lens. The temperature of the sample was maintained at 37 °C during the experiment by a stage-top incubator. EGFP, mCherry and DsRed-Express2 were excited using a multi-line argon laser (514 nm) and a helium–neon laser (543 nm). Irradiation with blue light was performed with a multi-line argon laser (488 nm) at 0.25 mW. The 4D image shown later in Figure 5.6 was taken at 37 °C using a white light laser scanning confocal microscope (TCS SP8 X) (Leica Microsystems, Wetzlar, Germany) and a 63× oil objective lens. Each image was composed by stacking 17 slices of X–Y images to cover an area of 12.8 µm2. 5.2.6. Half-life Evaluation of Off Kinetics The half-life value was evaluated by quantifying the extent of pMagFast2 variant recruitment to the plasma membrane upon irradiation with blue light and based on the intensity of pMagFast2-mCh-CAD at the plasma membrane. The time course of the change in mCherry intensity was fitted to an exponential decay equation [y = A exp(–kt) + B] using the non-linear leastsquares method, where y represents fluorescence intensity at time t, A and B are parameters and k is the rate constant. The half-life value, t½, was determined using the equation t½ = ln 2/k.

5.3. Results and Discussion Crystal structures of dark- and light-state VVD reveal a light, oxygen or voltage Per-Arnt-Sim domain with an unusual N-terminal cap (Ncap) and a loop insertion that accommodates the flavin cofactor. Photoinduced formation of a cysteine–flavin adduct drives flavin protonation to induce an N-terminal conformational change. We first engineered the VVD homodimer interface (Figure 5.1B). The Ncap sequence from Ile47 to Asn56 consists of neutral amino acids (Figure 5.1B). To monitor the blue light-dependent dimerization of VVD and its variants, a bioluminescence assay system based on the complementation of split-firefly luciferase fragments was developed. Using this system, we systematically searched for key amino acid residues in the Ncap of VVD that would lead to the expected electrostatic discrimination following substitution. We found that I52R and I52D as an electropositive and an electronegative substitution, respectively, efficiently induced heterodimerization and suppressed homodimerization of the VVD variants. Additionally, the I52R/M55R and I52D/M55G variants were selected as the best pair of distinct VVD variants that recognize each other based on electrostatic interactions. Given that electrostatic interactions drive their recognition, these VVD variants were collectively named Magnets. Specifically, the I52R/M55R variant has a positively charged arginine at residue 52 and was named positive Magnet (pMag) and I52D/M55G has a negatively charged aspartic acid at residue 52 and was named negative Magnet (nMag) (Figure 5.2B).

Published on 11 September 2018 on https://pubs.rsc.org |

Light-driven Actuators: Spatiotemporal Dynamics 107

Figure 5.2.  Tuning of the dissociation kinetics and dimerization efficiency of the Magnet system. (A) Dissociation kinetics of the variants of the Magnet system: pMag–nMag (black), pMagFast1–nMagFast1 (green), pMagFast2–nMagFast2 (red) and pMagHigh1–nMagHigh1 (blue). Dissociation kinetics were measured in COS-7 cells expressing Nfluc-X and Y-Cfluc (X and Y stand for pMag variants and nMag variants, respectively) after 30 s of irradiation with blue light to induce complex formation of the Magnet system. (B) Schematic representation of the variants with kinetic mutations within Ncap and the Per-Arnt-Sim (PAS) core. (C) Evaluation of the binding ability of the variants of the Magnet system by a bioluminescence assay system by the complementation of split-firefly luciferase complementation. The results are shown as the mean ± SD of three independent measurements.

View Online

108

Yoshibumi Ueda and Moritoshi Sato

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

5.3.1. Tuning Switch-off Kinetics and Dimerization Efficiencies pMag and nMag quickly form a heterodimer upon irradiation with blue light but they dissociate slowly over a period of hours after illumination is terminated. To tune the switch-off kinetics, we mutated Ile74, Ile85, Met135 and Met165 within the Per-Arnt-Sim (PAS) core of both pMag and nMag because mutation of these positions was reported to shorten or lengthen the photocycle of VVD.9 The substitution of Ile85 with valine in both pMag and nMag caused the complex of the two variants to dissociate with a t½ of 4.2 min (rate constant 2.7 × 10−3), which is 25-fold faster than that of the pMag–nMag complex (Figure 5.2A and B). We named these two mutant variants pMagFast1 and nMagFast1, respectively. Variants in which both Ile74 and Ile85 were substituted with valine exhibited even more dramatically accelerated dissociation (t½ 25 s; rate constant 2.8 × 10−2), which is 250-fold faster than that of the pMag–nMag complex (Figure 5.2A and B). These two variants were named pMagFast2 and nMagFast2, respectively. Despite our successful tuning of the switch-off kinetics of the pMag– nMag complex, we noticed that the bioluminescence signals following the dimerization of pMagFast1 with nMagFast1 and pMagFast2 with nMagFast2 decreased to 47 and 51%, respectively, of that of pMag and nMag. This result showed that a change in the photocycle of pMag and nMag substantially affected the extent of dimerization and revealed an undesirable functional correlation between fast dissociation kinetics and dimerization efficiency. This unwanted trade-off is an unacceptable limitation in the development of useful photoswitches. To overcome this problem, we focused on other photocycle-related substitutions. We found that the M135I and M165I substitutions lengthened the photocycle of VVD and substantially decelerated the dissociation of the pMag–nMag complex (t½ 4.7 h; rate constant 4.1 × 10−5) (Figure 5.2A and C) and furthermore dramatically increased the extent of light-dependent dimerization of pMag and nMag by up to 1700% (Figure 5.2C). These pMag and nMag variants were named pMagHigh1 and nMagHigh1, respectively. We examined the blue light-dependent heterodimerization between pMag variants and nMag variants with different switch-off kinetics. The extent of heterodimerization of pMagFast1 and nMagHigh1, which exhibited fast and slow switch-off kinetics, respectively, was enhanced by up to 1400% compared with the heterodimerization of pMagFast1 and nMagFast1 (Figure 5.2C). In addition, the combination of pMagFast2 and nMagHigh1 showed an enhancement of up to 760% compared with the heterodimerization of pMagFast2 and nMagFast2 (Figure 5.2C). These results indicate that pairing with nMagHigh1 substantially enhances the dimerization efficiencies of the pMag variants with fast dissociation kinetics. Furthermore, nMagHigh1 does not slow the switch-off kinetics of its heterodimers with the fast pMag variants. Also, pairing the fast pMag variants with nMagHigh1 did not prevent their reversible dimerization properties. Indeed, the heterodimerization of pMag with nMagHigh1 was enhanced by up to 1300% compared with pMag and nMag, without affecting its switch-off kinetics.

View Online

Light-driven Actuators: Spatiotemporal Dynamics

109

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

5.3.2. Assembly Method to Enhance the Performances of the Magnet System To improve the Magnet system further, we developed the constructs using two approaches: a tandem domain method and an assembly domain method. The tandem domain method allows the introduction of several pMagFast2 modules in a tandem manner into a single construct, such as in pMagFast2(3x) (Figure 5.3B). In contrast, the assembly domain method uses assembly domains (ADs) to tag pMags in order to form an oligomeric complex that possesses multivalency that promotes improved avidity (Figure 5.3C). We initially expressed repeated pMags in a tandem manner to create an mCherry-tagged version of pMag (pMag-mCh), pMag(2x) [pMag(2x)-mCh] and pMag(3x) [pMag(3x)-mCh] (Figure 5.3A) and expressed them in COS-7 cells. However, we found that the expression level of our construct decreased drastically with increasing number of pMagFast2 integrated into a single construct. To overcome the trade-off between expression level and nMag binding ability, we assessed the possibility of incorporating several pMag repeats into a single oligomer through ADs (Figure 5.3C). We anticipated that this method would increase the apparent binding affinity (avidity) of pMags for its partner (nMag) through multivalency arising from pMag

Figure 5.3.  Schematic representative for improvement of the Magnet system. (A) A single positive Magnet (pMag) tagged with fluorescent protein (FP) binds with a single negative Magnet (nMag) tethered to the plasma membrane (PM) through a CAAX motif from K-ras in response to blue light. (B) To gain binding ability, several pMag molecules were repeated in a tandem manner. The apparent binding ability (avidity) with nMag is increased. (C) pMag oligomerization through assembly domains (ADs).

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

110

Yoshibumi Ueda and Moritoshi Sato

assembly, while at the same time keeping the molecular weight of the pMag construct low. Currently, several ADs are available, such as the CaMKIIα association domain (CAD), which forms a donut-like dodecamer,10 the ADs from the red fluorescent proteins DsRed-Express (DsRedEx) and DsRed-Express2  (DsRedEx2), which form homotetramers11 and ferritin (FTN), which forms a spherical 24-mer.12 We fused these ADs to pMag, creating the constructs named pMag-DsRed-Ex, pMag-DsRed-Ex2, pMag-CAD and pMag-FTN. We checked the expression of pMag-CAD by transfecting COS-7 cells with an mCherry-tagged version of this construct (pMag-mCh-CAD). The expression level of pMag-mCh-CAD was comparable to those of pMag-mCh, pMag-DsRedEx2 and pMag-mCh-FTN, indicating that our CAD approach solved the low protein expression problem observed with conventional tandem fusions. Next, we compared the association of different pMag variants to nMag using fluorescence confocal microscope imaging. We transfected pMag-mCh into HEK293T cells together with nMagHigh1-EGFP-CAAX, anchored to the plasma membrane via its CAAX motif.13–15 Two days after transfection, the mCherry signal from pMag-mCherry exhibited an even cytosolic distribution, while we observed nMagHigh1-EGFP-CAAX at the plasma membrane. Irradiation of the whole cell with blue light for 5 s resulted in no significant translocation of pMag-mCh at the plasma membrane. Next, we examined the plasma membrane translocation of pMag(3x)-mCh and pMag-DsRedEx2. Both systems accumulated pMag at the plasma membrane after irradiation with blue light, but some cytosolic fluorescence remained. In contrast, the mCherry signal of pMag-mCherryCAD clearly translocated to the plasma membrane after irradiation with blue light. These results indicate that pMag-mCh-CAD has a far superior avidity to nMags compared with other pMags. 5.3.3. Kinetic Study of the CAD–Magnet System Next, in order to achieve quick photoswitching, we used the CAD–Magnet system with pMagFast2, a pMag variant with fast switch-off kinetics.5 We expressed pMagFast2-mCh-CAD (F2C) and nMagHigh1-EGFP-CAAX together in HEK293T cells. F2C exhibited even cytosolic distribution in the absence of blue light stimulation whereas upon irradiation with blue light, F2C was promptly translocated to the plasma membrane. After turning off the blue light, F2C returned to the cytosol. F2C was repeatedly recruited to the plasma membrane by switching the blue light irradiation on and off (Figure 5.4A). Time-lapse imaging showed that the translocation of F2C to the plasma membrane was complete within a few seconds. Kymograph analysis showed that the half-life of the switch-on kinetics of F2C was 0.82 s (Figure 5.4B and C). Additionally, the half-life of the switch-off kinetics was 8.6 s (Figure 5.4D and E), comparable to that of pMagFast2(3x)-iRFP (6.8 s), indicating that CAD did not interfere with the fast switch-off kinetics of pMagFast2.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

Light-driven Actuators: Spatiotemporal Dynamics

111

Figure 5.4.  Characterization of pMagFast2-mCherry-CAD. (A) Repetitive translocation of pMagFast2-mCherry-CAD (F2C) to the plasma membrane of HEK293T cells expressing nMagHigh1-EGFP-CAAX upon consecutive irradiations with blue light. The cells were consecutively irradiated with blue light of 473 nm for 20 s at intervals of 90 s. The white bar indicates 5 µm. (B) The change in F2C fluorescence intensity at the cross-section of the yellow line shown in (A). The white bar indicates 30 s. (C) Evaluation of the kinetics of F2C and nMagHigh1 binding. Normalized intensity was analyzed based on the F2C intensity at the plasma membrane. The half-life t½ was calculated by curve fitting as a monoexponential function to the time course of the normalized intensity of F2C. The error bars indicate the standard error of the mean (SEM) from three independent measurements. (D) Magnified images of the recovery of F2C to the cytosol after the blue light was turned off in the white square shown in Figure 5.4A. Time point 0 was set at the time when the blue light was turned off. The white bar indicates 5 µm. (E) Evaluation of the kinetics of F2C dissociation from nMagHigh1. Normalized intensity was analyzed based on the F2C intensity at the plasma membrane. t½ was calculated by curve fitting as a monoexponential function to the time course of the normalized intensity of F2C. The error bars indicate the SEM from three independent measurements.

View Online

112

Yoshibumi Ueda and Moritoshi Sato

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

5.3.4. Optical Control of Membrane Morphology by the CAD–Magnet System Previous reports indicated that the plasma membrane translocation of Tiam1, a guanine nucleotide exchange factor (GEF) that regulates Rac1 activation, through its DH/PH domain induces the development of membrane ruffles and lamellipodia.16,17 In order to control lamellipodia formation precisely, we fused the Tiam1 DH/PH domain to pMagFast2(3x) (TiamF2x3) (Figure 5.5A), and to visualize actin dynamics, we fused Lifeact, a 17 amino acid actin-binding peptide,18 to mCherry (LifemCh). Both constructs flanked a P2A peptide (LifemCh-P2A-TiamF2x3), allowing equal expression of both proteins.19 We expressed LifemCh-P2A-TiamF2x3 and nMagHigh1-EGFP-CAAX in COS-7 cells; following whole-cell irradiation with blue light, we observed that LifemCh labeled the periphery of the cell, indicating the generation of lamellipodia at the plasma membrane (Figure 5.5C). Next, we examined pMagFast2-CAD. As in the case of TiamF2x3, we fused the Tiam1 DH/PH domain to pMagFast2-CAD (TiamF2C) (Figure 5.5B), linked it to Lifeact-mCherry through P2A (LifemChP2A-TiamF2C) and expressed the construct together with nMagHigh1-EGFPCAAX in COS-7 cells. After irradiation with blue light, these cells exhibited stronger plasma membrane expansion and lamellipodia formation compared with cells expressing LifemCh-P2A-TiamF2x3 (Figure 5.5C). To analyze the extent of cell expansion as a measure of lamellipodia formation, we measured the cell area before and after irradiation. LifemCh-P2A-TiamF2C induced an expansion 2.6 times greater than that induced by LifemCh-P2A-TiamF2x3. 5.3.5. The Induction of Cell Membrane Dynamics Using CAD–Magnet Finally, we used 4D imaging to observe the plasma membrane after optogenetic perturbation by LifemCh-P2A-TiamF2C. LifemCh-P2A-TiamF2C and nMagHigh1-EGFP-CAAX in COS-7 cells were expressed and a local region of the cell was irradiated with blue light (Figure 5.6A). After irradiation, ruffles rose from the apical plasma membrane, resulting in cell membrane expansion at the edge closest to the irradiated region (Figure 5.6A). These ruffles were spatially and temporally confined. Magnification of the region revealed that the vertical ruffle walls were 5 µm in height and that they formed actin rings (Figure 5.6B). Time course analysis of the z-cross-section (shown with a blue rectangle in Figure 5.6B) showed that actin ring formation at the apical membrane was complete within 2 min (Figure 5.6C, top panels). The x-cross-section (shown with a green rectangle in Figure 5.6B) showed that the ruffles grew vertically (Figure 5.6C, middle panels). The ruffle overhung the apical plasma membrane for 2 min and finally formed a dome structure. The y-cross-section (shown with a red rectangle in Figure 5.6B) showed that the bottom and rim parts of the ruffle, indicated with white arrowheads on the 65.3 s frame, appeared earlier than the top part (Figure 5.6C, lower panels). Our 4D imaging analysis allowed us to detail the entire apical ruffle formation process (Figure 5.6D). In summary, these results demonstrated that our CAD–Magnet system displayed strong dimer binding, fast photoswitching kinetics and the ability to drive the localization of multiple effectors. The

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

Light-driven Actuators: Spatiotemporal Dynamics

113

Figure 5.5.  The Tiam1-CAD–Magnet system induced cell membrane expansion through the reorganization of actin filaments. (A) Schematic representative of Tiam1-pMagFast2(3x). The DH/PH domain of Tiam1 was fused with pMagFast2(3x). The activator translocates to the plasma membrane where nMagHigh1-EGFP-CAAX was tethered upon irradiation with blue light, and induces ruffles and lamellipodia through Rac1 activation. (B) Schematic representative of Tiam1-pMagFast2-CAD (TiamF2C). The 12 Tiam1 molecules are incorporated into an F2C complex. The TiamF2C is expected to lead efficiently to ruffles and lamellipodia through Rac1 activation on the plasma membrane in a blue light-dependent manner. (C) Effect of the Tiam1–Magnet system on cell membrane expansion in COS-7 cells upon irradiation with blue light for 10 min of the whole cell. Actin dynamics were visualized by Lifeact-mCherry. (D) The cell area change was analyzed by the change in kymograph of the cell shape. Data represent averages and SEMs. Asterisks indicate significant differences by an unpaired two-tailed Student's t-test: ***p < 0.001.

CAD–Magnet system in combination with 4D imaging enabled us to describe the process of ruffle formation in detail.

5.4. Conclusion Our goal is to develop rational design-based photoswitches to induce cellular signaling in living cells reliably. First, we eliminated the unwanted heterodimerization of VVD and introduced electropositive and electronegative

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

114

Yoshibumi Ueda and Moritoshi Sato

Figure 5.6.  4D image of local ruffles in COS-7 cells upon local irradiation with blue light. (A) 4D image of local ruffles visualized using Lifeact-mCh in COS-7 cells expressing nMagHigh1-EGFP-CAAX and TiamF2C. The region surrounded by the blue rectangle was irradiated with blue light for 20 min. The white bar indicates 10 µm. (B) Cross-section analysis of a ruffle on the apical plasma membrane. (C) Time course of the cross-sections of the ruffle specified with a white arrowhead in (B). The upper, middle and bottom cross-sections were acquired as shown with blue, green and red rectangles, respectively, in (B). The white bars indicate 2 µm. (D) Schematic representative of ruffle formation to ruffle dome. Blue lines show filamentous actin at the plasma membrane.

mutations in N-cap in VVD, resulting in establishment of the Magnet system. Next, we addressed the slow dissociation rate of the original VVD. Application of our rational design approach allowed us to manipulate the Magnet system with high temporal resolution using the pMagfast2 and nMagHigh1 pair. The combination of pMagFast2 and nMagHigh1 resulted in fast switch-on and switch-off kinetics, but the binding ability remained relatively low. We solved this issue by adopting ADs. Our CAD-fused pMag oligomerized after translation, providing a construct with a lower molecular weight compared with constructs with tandem repeated pMags and hence greatly improved expression. Additionally, the oligomerization of 12 pMags by CAD enhanced their binding to nMag compared with the original pMag and pMag(3x). These modifications resulted in greatly improved performance compared with our previous Magnet system. Furthermore, the CAD–Magnet system allows the specific localization of effector proteins such as Tiam1 in regions irradiated with blue light. Relocalization of the Tiam1 dodecamer effectively created membrane ruffles and lamellipodia through Rac. Therefore, the CAD–Magnet-  Tiam1 system induces cell membrane expansion and retraction more robustly than previously described Magnet systems.

View Online

Light-driven Actuators: Spatiotemporal Dynamics

115

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00099

References 1. Y. I. Wu, et al., A genetically encoded photoactivatable Rac controls the motility of living cells, Nature, 2009, 461(7260), 104–108. 2. B. D. Zoltowski, et al., Conformational switching in the fungal light sensor Vivid, Science, 2007, 316(5827), 1054–1057. 3. A. T. Vaidya, et al., Structure of a light-activated LOV protein dimer that regulates transcription, Sci. Signaling, 2011, 4(184), ra50. 4. B. D. Zoltowski and B. R. Crane, Light activation of the LOV protein vivid generates a rapidly exchanging dimer, Biochemistry, 2008, 47(27), 7012–7019. 5. F. Kawano, et al., Engineered pairs of distinct photoswitches for optogenetic control of cellular proteins, Nat. Commun., 2015, 6, 6256. 6. A. Furuya, et al., Assembly domain-based optogenetic system for the efficient control of cellular signaling, ACS Synth. Biol., 2017, 6(6), 1086–1095. 7. S. N. Ho, et al., Site-directed mutagenesis by overlap extension using the polymerase chain reaction, Gene, 1989, 77(1), 51–59. 8. A. Sawano and A. Miyawaki, Directed evolution of green fluorescent protein by a new versatile PCR strategy for site-directed and semi-random mutagenesis, Nucleic Acids Res., 2000, 28(16), E78. 9. B. D. Zoltowski, B. Vaccaro and B. R. Crane, Mechanism-based tuning of a LOV domain photoreceptor, Nat. Chem. Biol., 2009, 5(11), 827–834. 10. C. Thaler, et al., Structural rearrangement of CaMKIIalpha catalytic domains encodes activation, Proc. Natl. Acad. Sci. U. S. A., 2009, 106(15), 6369–6374. 11. R. L. Strack, et al., A noncytotoxic DsRed variant for whole-cell labeling, Nat. Methods, 2008, 5(11), 955–957. 12. J. Trikha, et al., Crystallization and structural analysis of bullfrog red cell L-subunit ferritins, Proteins, 1994, 18(2), 107–118. 13. M. Sato, et al., Production of PtdInsP3 at endomembranes is triggered by receptor endocytosis, Nat. Cell Biol., 2003, 5(11), 1016–1022. 14. M. Sato, Y. Ueda and Y. Umezawa, Imaging diacylglycerol dynamics at organelle membranes, Nat. Methods, 2006, 3(10), 797–799. 15. W. D. Heo, et al., PI(3,4,5)P3 and PI(4,5)P2 lipids target proteins with polybasic clusters to the plasma membrane, Science, 2006, 314(5804), 1458–1461. 16. A. Levskaya, et al., Spatiotemporal control of cell signalling using a light-switchable protein interaction, Nature, 2009, 461(7266), 997–1001. 17. T. Inoue, et al., An inducible translocation strategy to rapidly activate and inhibit small GTPase signaling pathways, Nat. Methods, 2005, 2(6), 415–418. 18. J. Riedl, et al., Lifeact: a versatile marker to visualize F-actin, Nat. Methods, 2008, 5(7), 605–607. 19. J. H. Kim, et al., High cleavage efficiency of a 2A peptide derived from porcine teschovirus-1 in human cell lines, zebrafish and mice, PLoS One, 2011, 6(4), e18556.

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

Chapter 6

Optogenetic Control of the Generation of Reactive Oxygen Species for Photoinducible Protein Inactivation and Cell Ablation Takeharu Nagai*a,b and Yemima Dani Riania a

Graduate School of Engineering, Osaka University, 1-3 Yamadaoka Suita, Osaka, 565-0871, Japan; bThe Institute of Scientific and Industrial Research, Osaka University, 8-1 Mihogaoka, Ibaraki, Osaka, 567-0047, Japan *E-mail: [email protected]

Table of Contents 6.1.  I ntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 6.2.  Molecular Mechanism of CALI. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 6.2.1.  Photosensitization Mechanism. . . . . . . . . . . . . . . . . . . . . . . 120 6.2.2.  ROS Effects on Intracellular Molecules . . . . . . . . . . . . . . . . 121 6.2.3.  How Specific is CALI?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 6.3.  Development of CALI Agents and Their Application in   Cell Biology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 6.3.1.  Chemical-based Photosensitizers. . . . . . . . . . . . . . . . . . . . . 122 6.3.2.  Genetically Encoded Photosensitizers. . . . . . . . . . . . . . . . . 125 6.3.3.  Genetically Encoded Photosensitizers for   Photodynamic Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . ��  128 6.4.  Future Perspectives for CALI. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

117

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

View Online

         

View Online

Optogenetic Control of the Generation

119

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

6.1. Introduction Understanding the protein function at the molecular, cellular, or even tissue level has allowed the advancement of wide-ranging fields, such as neurobiology, developmental biology and biomedical engineering. Observing loss-of-function phenotypes is one method for understanding the functions of particular proteins. Well-known methods for generating loss-of-function phenotypes include gene knockouts via homologous recombination, RNAi and the use of inhibitory drugs or function-blocking antibodies. However, there are some drawbacks to these methods, which can hamper the elucidation of particular protein functions. Gene knockout methods that are normally applied to embryonic stem cells, which delete or edit target genes to produce knockout (KO) organisms, cannot be applied to essential genes for developmental processes or house-keeping genes, since the loss of those genes is lethal. RNAi can specifically inactivate essential proteins, but the time-lag required to achieve the full loss-of-function phenotype depends on the target protein turnover rate. The use of drugs and function-blocking antibodies may generate artefacts or side-effects, depending on their specificity for the target protein. To overcome these problems, an optogenetic tool that can control protein inactivation in a spatiotemporal manner is desirable. In 1988, Daniel G. Jay first introduced the concept of chromophore-assisted light inactivation (CALI), which utilizes a photosensitizer and a chromophore that generates reactive oxygen species (ROS) when irradiated with excitation light, to inactivate proteins in an area- and time-specific manner. 1 ROS are known to be highly reactive molecules, causing damage to lipids, DNA and proteins. ROS generated by chromophores may diffuse over a range of less than 6 nm, hence localizing chromophores close enough to target proteins could damage them specifically and produce loss-of function phenotypes.2 The applications and achievements of CALI for revealing protein functions have extended beyond the limitations of the conventional methods mentioned earlier. Photoinduced ROS generation is not limited to protein inactivation but has also been demonstrated to induce specific cellular ablations, by targeting photosensitizers to particular cells. This method is therefore also useful for studying cellular interactions, cell functions within tissues/organisms and photodynamic therapy (targeted cancer treatment utilizing light irradiation). This chapter discusses details of the CALI mechanism, the development of ROS-generating optogenetic tools for protein and cellular inactivation and future perspectives regarding this method.

6.2. Molecular Mechanism of CALI In 1979, Miller and Selverston successfully killed a single neuron by injecting it with the fluorescent dye Lucifer Yellow and irradiating it with high-intensity laser light at the maximum absorption wavelength of the fluorophore.3

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

120

Takeharu Nagai and Yemima Dani Riani

The mechanism underlying this phenomenon was not well understood at the time. Two speculations emerged: the first hypothesis was that the cells were killed by heat, converted from excitation energy; the second was that the dye was converted into a toxic substance that killed the cells.3 Later, in 1988, Jay used this photoabsorption method to inactivate protein.1 He chose Malachite Green as the fluorescent dye because its absorption maximum is 620 nm, a wavelength that is not absorbed by most intracellular molecules. He successfully used Malachite Green-conjugated antibodies to specifically inactivate alkaline phosphatase and β-galactosidase with 25 mW cm−2, 10 Hz pulsed laser irradiation in 10 s, separated by rests of 100 ms, for 5 min. It was proposed that the primary factor of CALI effectiveness was the heat energy that denatured the target protein.1 In 1994, Jay and co-workers discovered that it was impossible for CALI to be mediated by photothermal effects, since beyond a distance of 40 Å from the dye (the distance between the dye and the target protein, separated by an antibody), the temperature increase caused by the laser pulse was supposedly just 0.35 °C, which was insufficient for denaturing protein targets.4 They then experimented using continuous-wave lasers, or pulsed lasers with lower power density than in previous experiments, yet CALI did not occur. This suggested that the chromophore must obtain enough energy to show the apparent CALI result. A high pulsed laser was effective in inducing CALI because after excitation by the first laser pulse, the excited dye absorbed a second photon, sending it into a higher excited state. The sequential two-photon process would result in the generation of higher energies, that is sufficient to form hydroxyl radicals from water. Based on this result, they speculated that ROS generation was the underlying CALI mechanism. To confirm their hypothesis, they experimented using several ROS quenchers, such as sodium azide, mannitol, butylated hydroxytoluene and vitamin E, to measure the effects of acetylcholinesterase and galactosidase inactivation by Malachite Green. The results showed that free radical species, specifically hydroxyl radicals, were the main player underlying Malachite Green-mediated CALI.4 6.2.1. Photosensitization Mechanism By the time that ROS-induced CALI had been proposed, the photosensitization mechanism had already been well studied. When photosensitizers absorb light, light energy excites chromophores, promoting them to enter excited states, which can be several levels higher than the lowest excited state. This is followed by a non-radiative transition that mostly generates heat and then, the photosensitizer will remain in the lowest excited single states for several nanoseconds. In this state, the photosensitizer may emit fluorescence (hvF), transfer its energy to the solvent without emission, enter a triplet state from which an emission called phosphorescence is generated, or cause a one-electron reduction of ground-state oxygen (3O2), to produce superoxide anion (O2•−). Mostly, effective photosensitizers undergo intersystem crossing to produce an excited triplet state, which is relatively long lived. In the presence of oxygen, this energy will be transferred to a nearby oxygen

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

Optogenetic Control of the Generation

121

Figure 6.1.  ROS production illustrated in a Jablonski diagram. A chromophore in the ground state (S0) will enter the excited state (S1), or several higher states, upon absorption of a photon. It will undergo non-radiative transition (black dashed arrow) and remain in the lowest excited single state. In this state, it can emit fluorescence or release its energy without emission (wavy arrow). In the presence of oxygen, chromophores in this state can stimulate electron transfer, generating superoxide anion species (Type I mechanism). On entering a triplet state through intersystem crossing, chromophores will transfer energy to oxygen, generating singlet oxygen (Type II mechanism).

molecule and generate singlet oxygen, which has paired electrons with opposite spins. The sensitization mechanisms that generate superoxide anion and singlet oxygen are called the Type I and Type II mechanism, respectively (see Figure 6.1).2,5 6.2.2. ROS Effects on Intracellular Molecules Following ROS generation, free ROS will diffuse in the solvent and attack neighbouring molecules because of their high reactivity. Intracellular ROS attacks DNA bonds, causes lipid peroxidation and oxidizes the side-chains of amino acid residues, including histidine, tyrosine, tryptophan, methionine and cysteine. Attacks of ROS on proteins result in changes to intramolecular protein structure and function. In addition to intramolecular dysfunction, affected proteins may undergo protein–protein crosslinking and aggregation, leading to intermolecular protein inactivation. ROS can also directly cause fragmentation of peptide bonds.2

View Online

122

Takeharu Nagai and Yemima Dani Riani

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

6.2.3. How Specific is CALI? The answer to this question is dependent on the diffusion range and lifetime of ROS in solution. In fact, different species of reactive oxygen will have different lifetimes and diffusion rates. Singlet oxygen has a lifetime of 3 µs in water, giving approximate diffusion distances of 130 nm, assuming that the singlet oxygen diffusion coefficient is ∼1000 µm2 s−1.2 However, the diffusion range of superoxide anion has not been assessed, except by those of its derivatives, hydrogen peroxide and hydroxyl radical. Hydrogen peroxide is speculated to diffuse at a rate of 1000–2000 µm2 s−1 and has a long lifetime, hence it may diffuse several microns within the cellular environment if no scavenging enzyme is present.6 Since the lifetime of the hydroxyl radical is very short (1 ns), damage caused by hydroxyl radicals would need to occur within a range of 10 Å.4 Theoretically, hydroxyl radical is preferable for inactivation of a target protein without causing collateral damage to the neighbouring proteins, since the average intermolecular distance inside cells is 80 Å.7 In contrast, singlet oxygen is preferable for causing universal damage, such as killing cells. Based on the data collected on the currently available CALI tools, exact measurements of their half radius of damage is well described only for Malachite Green and fluorescein. Utilizing Malachite Green, the half maximum radius is 15 Å from the dye moiety and damage will not be generated at further than 60 Å.7 For another assessed CALI tool, fluorescein, which generates singlet oxygen, the half damage radius was measured as ∼4 nm, which  is different from the expected diffusion range of singlet oxygen in water. Despite the possibility of non-specific inactivation caused by singlet oxygen, CALI by singlet oxygen has been widely performed in vitro and in vivo, without specific collateral damage to unintended targets. This suggests that the effect might be caused by effective ROS projection to the target and that the abundance of scavenging molecules present in the area surrounding the chromophores prevents the further diffusion of ROS to a non-specific target. Owing to the different types of ROS generated from photosensitizers, it is necessary to choose the right photosensitizer for the experiment, although precise computation of a CALI experiment is sometimes neglected. Some factors to be considered are (1) type of ROS, (2) range of action and (3) effectiveness of ROS generation. Later, we will discuss the development of different kinds of tools available for CALI, together with their properties and the achievements that can be made by utilizing those properties.

6.3. Development of CALI Agents and Their Application in Cell Biology 6.3.1. Chemical-based Photosensitizers As mentioned before, the first CALI was demonstrated using Malachite Green as the photosensitizing agent. Malachite Green was able to inactivate alkaline phosphatase and β-galactosidase in vitro, after irradiation with 

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

Optogenetic Control of the Generation

123

25 mW cm−2 of 620 nm pulsed laser light for 5 min.1 Subsequently, CALI utilizing Malachite Green has been very useful for elucidating protein function in vivo. The first application of this was the elucidation of the specific function of fasciclin I in grasshopper embryos. Although it was known that fasciclin I mediated homophilic adhesion in the Drosophila S2 cell line, its function in vivo remained unknown. Fasciclin I is expressed on the surface of Ti1 pioneer neurons in limb bud development. These cells remain in close contact with each other and form growth cones, extending their axons towards the central nervous system by crossing through epithelia, in a process called fasciculation. Inactivation of fasciclin I by CALI during this axonal projection process resulted in the separation of two axons, indicating defasciculation. This result revealed that fasciclin I was essential for the fasciculation of Ti1 pioneers neurons.8 A more specific and targetable CALI was performed years later, using a technique named micro-CALI, which targeted specific laser irradiation at a particular area, causing fast and efficient bleaching and inactivation of the chromophore. Micro-CALI was used to target fasciclin II, which is similar to fasciclin I and known to play a role in the adhesion of Drosophila S2 cell lines, although its in vitro functions were still unknown. Fasciclin II is expressed primarily on the longitudinal fascicles of the central nervous system and also in the peripheral neurons of grasshoppers. Micro-CALI performed at a different stage of axonogenesis showed that fasciclin II inhibited axon outgrowth only after the Ti1 neurons had completely emerged from the epithelium, but before they became tear shaped. When micro-CALI of fasciclin II was performed on post-axonogenic Ti1 neurons, the length of axon growth was unaffected. These results suggested that micro-CALI could elucidate the importance of fasciclin II requirement in a narrow temporal window during Ti1 development that could not be achieved using other conventional methods. Another advantage of micro-CALI is that since it targets only specific cells within the embryo, the non-irradiated cells in the embryo could serve as internal controls.9 CALI could also be used to elucidate the cellular mechanisms of cancer, enabling the identification and validation of protein targets for drug discovery. Malachite Green has been used to reveal proteins that might be involved in cancer metastasis, such as ezrin and pp60 c-src. In cultures of rat fibroblasts, micro-CALI of ezrin caused loss of membrane ruffling and pseudopodial retraction, suggesting that ezrin is required for pseudopodia-based extension, which is involved in cancer metastasis.10 While the function of proto-oncogene pp60 c-src as a tyrosine kinase has been established previously, its in situ function is not well characterized. Using a knockout-based method to observe the loss of function effects in mice did not reveal any obvious effects. Conversely, implementing micro-CALI of pp60 c-src in developing chick neurons showed growth cone motility and significant neurite extension, suggesting that CALI is useful in the study of proto-oncogenes. Moreover, implementing CALI of pp60 c-src has also shown other advantages over gene knockouts. The use of gene knockouts

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

124

Takeharu Nagai and Yemima Dani Riani

is often associated with compensated expressions of other proteins to overcome the loss of target protein, thus producing no phenotypic changes.11 In addition, the superiority of CALI over gene knockout methods has also been demonstrated when CALI was applied to the Drosophila segment-polarity gene patched (Ptc). This gene is essential for segmentation in the early embryo and at later stages is important for fate determination of optic lobe primordia precursors. Using gene knockout methods to target the protein Ptc causes abnormal patterns of segmentation, resulting in an inability to observe its specific function at later developmental stages. Implementing CALI has revealed that Ptc has specific functions after the segmentation period, to discriminate between neural and epithelial cells in optic-lobe precursor cells.12 However, this first generation of the CALI method, utilizing Malachite Green targeted via antibody, was considered laborious. The antibody itself is required to avoid blocking the active site of the protein target. Moreover, another test such as immunoblotting with immunocytochemistry should confirm that the antibody recognizes the antigen and binds to the intended target. The delivery of antibody-conjugated dyes usually requires microinjection for a small number of target cells. But, if a large number of cells are required, then bulk loading techniques such as trituration, scrape loading or electroporation are required.13 Moreover, using a laser as the light source means that the power density requirements are high and may cause phototoxicity to the cells. The development of CALI tools after Malachite Green has provided us with less laborious techniques, allowing more efficient ROS generation, with lower illumination power. For example, replacing Malachite Green with fluorescein (excitation at 488 nm), in a method known as FALI, produces a 50-fold higher inactivation efficiency. Unlike Malachite Green, which generates hydroxyl radicals, fluorescein is known to generate singlet oxygen.14 Several attempts have been made to replace antibody-targeting techniques with transgenically encoded tag-based techniques. A membrane-permeable fluorescein derivative, 4′,5′-bis(1,3,2-dithioarsolan-2-yl)fluorescein (FlAsH), could bind to the tetracysteine motif, which could be easily encoded and fused with protein targets. FlAsH was fused to synaptotagmin I (SytI), which functions in a post-docking step of vesicle fusion by acting as the major calcium sensor for transmitter release. Illumination of FlAsH–SytI decreased transmitter release in vivo in Drosophila, simply using visible light from a mercury arc lamp for illumination.15 A more effective fluorescein derivative, ReAsH, a red biarsenical dye, was established later and possesses similar properties to FlAsH, in that it is membrane permeable and binds to the tetracysteine motif. Targeting connexin43 with ReAsH and illuminating with excitation at 17 W cm−2 for 25 s has been shown to inactivate gap junctions.16 However, utilizing FlAsH or ReAsH has also been criticized, owing to their non-specific binding to any cysteine residue of endogenous proteins and high cytotoxicity.17–19

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

Optogenetic Control of the Generation

125

As an alternative to antibody and tetracysteine motif-based targeting, commercially available tags such as SNAP-tag and HaloTag, have also been used. CALI of SNAP-tag–fluorescein fused to the C-terminal region of γ-tubulin caused a failure in microtubule nucleation and growth, thus leading to metaphase arrest.20 HaloTag protein, a modified haloalkane dehalogenase, was designed to bind covalently to a variety of haloalkane-containing compounds. Thus, specific CALI could be achieved by fusing target proteins with photosensitizer-labelled HaloTag. Haloalkyl derivatives of fluorescein and Ru(ii) tris(bipyridyl) were tested for use in combination with HaloTag. The Ru(ii) tris(bipyridyl) dication was demonstrated to have superior performance, compared with fluorescein, for generating singlet oxygen and was shown to inactivate Renilla luciferase effectively.21 However, ruthenium excitation light, which is at 380 nm, is known to cause phototoxicity to living cells. Years later, eosin was found to be stronger than ruthenium, with 517 nm excitation light (almost the same as that of fluorescein), which is less harmful than UV excitation light. A membrane-permeable eosin, diAc-eosin, was synthesized to be linked with HaloTag7, a new HaloTag mutant, with superior ability to form covalent bonds with ligands. Using HaloTag7– diAc-eosin to target specifically Aurora B has been shown to stop cell division, resulting in multinuclear structure formation, after mitosis.22 To date, eosin is considered to be the most powerful photosensitizing agent that has been applied for CALI purposes in mammalian cells and in vivo. Recently, CALI using eosin was demonstrated to target synaptic AMPA (α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) receptors and successfully manipulated mouse behaviour by erasing fear memory.23 6.3.2. Genetically Encoded Photosensitizers Although CALI utilizing chemical photosensitizers has been improved, the exogenous addition of labelling reagents is still needed, thus it is problematic to penetrate deep tissue targets. Moreover, incomplete washing of unbound chromophore may cause non-specific target inactivation. Genetically encoded proteins, such as EGFP (enhanced green fluorescent protein), are well known to be easily fused with protein targets or subcellular localization tags. The first photosensitization experiment using genetically encoded photosensitizer was reported by Surrey et al. in 1998,14 the same time as the introduction of fluorescein as a Malachite Green alternative. EGFP was targeted to inactivate β-galactosidase. They showed that EGFP could inactivate the protein specifically, with the energy doses required being almost same as for Malachite Green. Later, EGFP was used to inactivate α-actinin, resulting in the detachment of stress fibres from focal adhesions and causing retraction of stress fibres. The irradiation time was 100 ms with a 40 mW laser, focused to a 2.2 µm diameter spot. The approximate light dose used for CALI utilizing EGFP was at least 106 times higher than that for typical confocal imaging, which caused phototoxicity to living cells.24 An attempt to avoid irradiation with blue light was made using near-infrared (NIR) excitation

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

126

Takeharu Nagai and Yemima Dani Riani

light. Multiphoton excitation was demonstrated for EGFP-mediated CALI of connexin43. Multiphoton excitation at 850 nm with 2.7 mW cm−2 illumination was applied for 380 ms to separate the gap junctions of HeLa cells.25 Despite the success of achieving the expected results, non-specific photodamage to mammalian cells by using multiphoton excitation could not be avoided, and EGFP remains a non-effective CALI tool owing to the very highpower densities required for ROS generation. Similar properties were tested in colour variants of EGFP, such as EYFP and ECFP, which showed even less effective ROS production. It was suggested that the effectiveness of CALI of EGFP variants was dependent on the ratio of escaped ROS to those trapped in the beta-barrel, where chromophore bleaching occurred. As ROS escape the beta-barrel, chromophores will be conserved and ROS become more likely to reach the protein target.26 In 2006, KillerRed, a genetically encoded photosensitizer derived from anm2CP protein, was developed.27 KillerRed is activated by both orange and green light. Initially, anm2CP protein did not generate ROS. ROS generation could be detected only after several mutations were introduced into the protein. A unique feature in the structure of KillerRed is the presence of a water channel that connected the chromophore to the opening of the beta-barrel. Several hypotheses were made regarding the mechanism underlying the effective generation of ROS by KillerRed. The first was that the channel provides access to oxygen molecules, which diffuse inside the beta-barrel. When a chromophore becomes excited, it will enter the triplet state and transfer energy to oxygen, producing singlet oxygen. However, based on experiments with D2O, it was suggested that KillerRed actually produces superoxide rather than singlet oxygen. Hence the electron transfer reaction may occur when the chromophore is in an excited state, resulting in the donation of an electron from the chromophore to oxygen, allowing superoxide to diffuse out of the protein via the channel. It was also suggested that these water molecules may serve as electron wires, conducting electron transfer generated by excited chromophores to external molecules, resulting in ROS generation near the opening of the beta-barrel.28,29 ROS generation by KillerRed was examined and it was finally introduced as the first effective genetically encoded photosensitizer. When fused to β-galactosidase, KillerRed resulted in 100% protein inactivation after 25 min of irradiation with 1 W cm −2 white light, whereas EGFP resulted in only 40% inactivation upon irradiation with sevenfold greater light power. Compared with ReAsH, KillerRed is only threefold less efficient. After the development of KillerRed, employment of the CALI tool has shifted from chemical-based photosensitizers towards genetically encoded photosensitizers. They have been proven to constitute an effective, specific and non-laborious method compared with the antibody labelling method. 27 For CALI purposes, KillerRed has been demonstrated to successfully inactivate water transport and protein–protein interactions of aquaporin, a protein

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

Optogenetic Control of the Generation

127

that is essential for cell survival.30 KillerRed also has been used to study the different roles of GRASP65 and GRASP55 in Golgi compartmentalization. Coupled with live imaging, the rapid and acute inactivation of GRASP65 and GRASP55 revealed the true function of those proteins in a dynamic environment of Golgi compartmentalization.31 Another live observation on the spatiotemporal dynamics of proteins was applied to cofilin, supporting the hypothesis that the principal role of cofilin in lamellipodia is to break down F-actin, control filament turnover and regulate the rate of retrograde flow.32 In addition to inactivating protein function, the employment of genetically encoded photosensitizers has been extended to study chromosomal architecture and genotyping. Since the expression of several genes that are located near the nuclear lamina is highly regulated and any changes might result in diseases, study of the localization of genes inside the nucleus would provide useful information for biomedical applications. KillerRed targeted to laminB1, which is located near the nuclear envelope, and KillerRed targeted to histone 2A, a protein localized ubiquitously throughout chromatin, showed different damaged genes after irradiation with light. This indicates that KillerRed is useful for obtaining information about gene distribution within the chromatin.33 In addition, KillerRed was used to induce DNA double-strand breaks spatiotemporally, to study the distinct roles of several factors involved in DNA repair, such as DNA glycosylase, PARP1, FEN1 and PCNA, in heterochromatin and euchromatin.34 Moreover, targeted DNA damage by KillerRed to histone 2B and telomeres triggered cellular senescence and death, respectively.35,36 An in vivo application of KillerRed was demonstrated in Xenopus laevis by targeting KillerRed to the plasma membranes of Xenopus embryos by mRNA microinjection. ROS generation was induced with 5 min of exposure to green light at 2.5 W cm−2, to target several organs and tissues. Apoptosis was observed specifically in targeted developing eye and pronephric kidney.37 Another demonstration of the use of KillerRed in vivo was the killing of cells in zebrafish kidneys using plasma membrane-targeted KillerRed and muscle tissue, to induce heart-failure animal models using mitochondria-targeted KillerRed.38,39 The use of KillerRed was also demonstrated in Caenorhabditis elegans to induce cell ablation of AWA sensory neurons, allowing manipulation of C. elegans chemotaxis behaviour towards AWA-sensitive attractants. These experiments showed that KillerRed is useful for inducing cell ablation in vivo, thus allowing the in situ study of developmental biology, neurobiology and animal behaviour.40 However, KillerRed was reported to have a tendency to dimerize when expressed at high concentrations in cells. This dimeric property might hamper its localization or targeting purposes.27 In 2013, a monomeric version of KillerRed, SuperNova, was established and was shown to possess superior properties for targeting, while showing minimal aggregation in cells.41   SuperNova has also been demonstrated to promote CALI of cofilin and F-actin  and cell ablation in vitro and in vivo.41,42

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

128

Takeharu Nagai and Yemima Dani Riani

In 2011, a genetically encoded photosensitizer, miniSOG, was developed from the light-, oxygen- and voltage-sensing (LOV) domain of Arabidopsis phototropin 2. miniSOG contains flavin mononucleotide (FMN) as its chromophore, with its toxicity being dependent on FMN concentration. MiniSOG was identified as a generator of both superoxide and singlet oxygen, which means that FMN undergoes a photoinitiated electron-transfer reaction to produce superoxide as well as energy transfer to ground-state oxygen to produce singlet oxygen when irradiated with blue light (∼440 nm).43–45 As is the case for KillerRed, miniSOG has also been extensively applied to the inactivation of many proteins, cell ablation and photodynamic therapy. Furthermore, miniSOG was reported to be more powerful than KillerRed.46 Fusions of miniSOG to VAMP2 and synaptophysin allowed the disruption of presynaptic vesicular release upon illumination with blue light.47 miniSOG was also successfully applied to inactivate acutely subunits of respiratory complex II in mitochondria of C. elegans, permitting the detailed study of the subunit functions. These observations were impossible using the knockout approach, owing to their lethality.48 Although the use of the current membrane-targeted miniSOG method has been shown to promote cell ablation in C. elegans effectively, there is still a demand to develop stronger genetically encoded photosensitizers for promoting optogenetic cell ablation in more complex organisms. Some improvements to miniSOG have been made to suppress the photoinitiated electron transfer producing superoxide, which competes with singlet oxygen production. The first improved version was called SOPP (singlet oxygen photosensitizing protein), which contains a Q102L mutation to reduce hydrogen bonding between FMN and its surrounding amino acids.49 In vivo application has also been performed in C. elegans to promote neuron and muscle cell ablation effectively.50 A second point mutation was made to produce miniSOG2, which has been shown to promote cellular ablation in developing Drosophila melanogaster neurons and the wing imaginal disc.51 Table 6.1 summarizes in vitro and in vivo applications of genetically encoded photosensitizers. 6.3.3. Genetically Encoded Photosensitizers for Photodynamic Therapy After the generation of effective genetically encoded photosensitizers, some attempts have been made to use this tool extensively for other applications. Although chemical photosensitizers have been proven to effectively kill cancer and bacterial cells,52–54 some groups have researched the best strategy for targeting genetically encoded photosensitizers to kill cancer cells. For example, targeting cancer cells using an anti-receptor antibody–KillerRed/miniSOG fusion protein as a genetically encoded immunophotosensitizer has demonstrated fine targeting properties and efficiently killed p185(HER-2-ECD)-expressing cancer cells upon light irradiation in vitro.46,55 Other methods, such as targeting to the inner and outer membranes of mitochondria, nuclei, plasma membranes, lysosomes and peroxisomes, have been attempted and shown to achieve cell

Published on 11 September 2018 on https://pubs.rsc.org |

In vitro Photosensitizing protein Target KillerRed

PH domain of PLC δ1 27 AQP1, AQP4 isoforms30 Cofilin32 GRASP 65 GRASP 5531 DNA damage33

HeLa

Cell ablation

Light irradiation time

Cell line

Irradiation power

Irradiation time Target

HEK293

7 W cm−2

10 s

Heart cells39

D. rerio

FRT U87MG CAD

18 mW cm−2 to   175 µm area 100% laser   power, 9 mW 15.8 mW, 45 µs   per pixel 32 W bulb,   20 000 lux 6–9 mJ µm−2 4 W cm−2

60 s

Muscle cells65

C. elegans

100 mW cm−2

60 min

2 min

Multiple classes of neuron39,40,67 Tumour xenografts59

C. elegans D. rerio M. musculus

0.2–0.3 W cm−2

X. laevis

2.5 W cm−2

1–2 h 20 min 30 min 7 days 5 min

7 W cm−2 9.6 W cm−2 7 mW, 1000 nm two-photon   laser 0.98 W cm−2

10 s 1 min 4s C. elegans

70–540 mW cm−2 5–25 min

65 mW cm−2

15 min

C. elegans

1.02 W cm−2

3–5 min

12 mW cm−2 52 mW cm−2 55 mW cm−2

90 min 10 min

C. elegans

57 mW cm−2

30 min

DU145 U2OS HEK 293T HeLa65 SKOV-3 55 HeLa Neuron slice culture

SuperNova

Cofilin41 CaMKII–  F-actin42

MiniSOG

VAMP2 and syn- Hippocampal aptophysin47 neuron DNA damage35 HeLa Cell ablation56

In vivo

HeLa SK-BR-3

30, 60 min Eye and kidney tissue37

Organism

Power density

20 min

150 mW cm−2

30 s

2 min

VAMP2 and synaptophysin47 Mitochondria electron transport chain complex II48 Multiple classes of neurons71

129

(continued)

Optogenetic Control of the Generation

Table 6.1.  In vitro and in vivo applications of genetically encoded photosensitizers.

Published on 11 September 2018 on https://pubs.rsc.org |

130

Table 6.1.  (continued) In vitro

Cell line

Irradiation power

Irradiation time Target Neuron cells72 Muscle cells72 Neurons51

MiniSOG Q103L MiniSOG2

Wing imaginal disc51 KillerOrange

Cell ablation68

HEK293

60 mW for 447 nm and 20 mW for 590 nm

Organism

Power density

Light irradiation time

C. elegans C. elegans D. melanogaster D. melanogaster

200 mW cm−2 200 mW cm−2 1.8 W cm−2

2–4 min 2–4 min 3 min

0.5 W cm−2

3 min

Takeharu Nagai and Yemima Dani Riani

Photosensitizing protein Target

In vivo

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

Optogenetic Control of the Generation

131

Figure 6.2.  Timeline of CALI tools and targeting method development.

ablation both in vitro and in vivo experiments.56–61 However, obstacles to utilizing genetically encoded photosensitizers for photodynamic therapy lie in the delivery of the protein to cancer cells and in the penetration of light irradiation into deep-tissue targets. Many efforts have been made to overcome these problems. One, for example, is the use of polymeric micelle-encapsulating quantum dots for delivering KillerRed plasmids to cancer cells. Qdot itself will absorb light at lower wavelengths and emit light at higher wavelengths in the cell cytoplasm, activating the expressed KillerRed.62 Another effort to mediate deep-penetrating photodynamic therapies involved the use of upconversion nanoparticles, which are able to convert deep-penetrating NIR light to green light, thus activating KillerRed locally and resulting in tumour cell ablation under centimetre-thick tissue.63 Since infrared radiation is more useful for targeting deep tissue, a transgenically encoded NIR photosensitizer has also been developed. An NIR targeted and activated photosensitizer (TAP) derived from Malachite Green only produces singlet oxygen and fluorescence when bound to genetically encoded fluorogen-activating protein (FAP). This genetically encoded FAP–TAP approach allowed protein inactivation, targeted cell killing and rapid targeted cell ablation in living larval and adult zebrafish.64 In Figure 6.2, the timeline of CALI tools and targeting method development is summarized.

6.4. Future Perspectives for CALI For all applications that have been described here, utilizing light for optogenetic control of ROS generation has brought great advantages to many fields of study. Another prospective use of photosensitizers as optogenetic tools is to examine protein–protein interactions. To manipulate protein interactions by utilizing photoreceptors (e.g. LOV domain and CRY2/CIB1) that change conformation upon irradiation with blue light, splitting or fusing the target protein with its inhibitor or activator is needed. Compared with that approach, targeting the protein of interest with a genetically encoded

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

132

Takeharu Nagai and Yemima Dani Riani

photosensitizer would be less laborious since only direct fusion to target proteins is needed. Moreover, to see longer term loss of function effects, genetically encoded photosensitizers would be preferable compared to LOV domain-derived methods when the dissociation rate of the LOV domain is faster than protein turnover rate.66 Since ROS generation by the photosensitizers mentioned earlier are activated by specific wavelengths, similarly to other fluorescent proteins that are useful for two- or three-colour imaging, using different colours of photosensitizer also can be applied to inactivate two or three different kinds of target proteins or cells in a spatiotemporal manner. Proof of concept has been demonstrated for KillerRed and miniSOG to ablate cholinergic motor neurons in C. elegans. Blue and green light specifically activated miniSOG and KillerRed, respectively, with no cross-talk. This indicates that expressing KillerRed and miniSOG in a particular cell or organism to inactivate two different kinds of proteins or cells in a spatiotemporal manner is achievable.67 For multiple CALI purposes, several colour variants of genetically encoded photosensitizers have been developed. Orange variants of KillerRed and SuperNova are activated upon irradiation with 440 and 510 nm light.68 A green variant of KillerRed was also reported, but is suspected not to generate ROS.69 Recently, a new green variant of SuperNova, SuperNova Green, was established70 and has been demonstrated to promote cell death and generate ROS specifically upon irradiation with blue light. The SuperNova Green chromophore is a spontaneously formed chromophore like GFP, thus eliminating dependency on other molecules, such as FMN for miniSOG. In conclusion, there are currently three colour variants of KillerRed-based photosensitizers, namely red, orange and green. By using these three colour variants, three consecutive inactivations can be achieved by irradiating samples with light of 560, 510 and 440 nm respectively.

References 1. D. G. Jay, Proc. Natl. Acad. Sci. U. S. A., 1988, 85, 5454–5458. 2. K. Jacobson, Z. Rajfur, E. Vitriol and K. Hahn, Trends Cell Biol., 2008, 18, 443–450. 3. J. P. Miller and A. I. Selverston, Science, 1979, 206, 702–704. 4. J. C. Liao, J. Roider and D. G. Jay, Proc. Natl. Acad. Sci. U. S. A., 1994, 91, 2659–2663. 5. G. Laustriat, Biochimie, 1986, 68, 771–778. 6. C. C. Winterbourn, Nat. Chem. Biol., 2008, 4, 278–286. 7. K. G. Linden, J. C. Liao and D. G. Jay, Biophys. J., 1992, 61, 956–962. 8. D. G. Jay and H. Keshishian, Nature, 1990, 348, 548–550. 9. P. Diamond, A. Mallavarapu, J. Schnipper, J. Booth, L. Park, T. P. O'Connor and  D. G. Jay, Neuron, 1993, 11, 409–421. 10. R. F. Lamb, B. W. Ozanne, C. Roy, L. McGarry, C. Stipp, P. Mangeat and D. G. Jay, Curr. Biol., 1997, 7, 682–688. 11. D. Hoffman-Kim, J. A. Kerner, A. Chen, A. Xu, T. F. Wang and D. G. Jay, Mol. Cell. Neurosci., 2002, 21, 81–93. 12. D. Schmucker, A. L. Su, A. Beermann, H. Jäckle and D. G. Jay, Proc. Natl. Acad. Sci. U. S. A., 1994, 91, 2664–2668.



View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

Optogenetic Control of the Generation

133

13. W. Li, N. Stuurman and G. Ou, Neurosci. Bull., 2012, 28, 333–341. 14. T. Surrey, M. B. Elowitz, P. E. Wolf, F. Yang, F. Nédélec, K. Shokat and S. Leibler, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 4293–4298. 15. K. W. Marek and G. W. Davis, Neuron, 2002, 36, 805–813. 16. O. Tour, R. M. Meijer, D. A. Zacharias, S. R. Adams and R. Y. Tsien, Nat. Biotechnol., 2003, 21, 1505–1508. 17. A. C. Hearps, M. J. Pryor, H. V. Kuusisto, S. M. Rawlinson, S. C. Piller and D. A. Jans, J. Fluoresc., 2007, 17, 593–597. 18. B. R. Martin, B. N. G. Giepmans, S. R. Adams and R. Y. Tsien, Nat. Biotechnol., 2005, 23, 1308–1314. 19. K. Stroffekova, C. Proenza and K. G. Beam, Pflugers Arch. Eur. J. Physiol., 2001, 442, 859–866. 20. A. Keppler and J. Ellenberg, ACS Chem. Biol., 2009, 4, 127–138. 21. J. Lee, P. Yu, X. Xiao and T. Kodadek, Mol. BioSyst., 2008, 4, 59–65. 22. K. Takemoto, T. Matsuda, M. McDougall, D. H. Klaubert, A. Hasegawa, G. V. Los, K. V. Wood, A. Miyawaki and T. Nagai, ACS Chem. Biol., 2011, 6, 401–406. 23. K. Takemoto, H. Iwanari, H. Tada, K. Suyama, A. Sano, T. Nagai, T. Hamakubo and T. Takahashi, Nat. Biotechnol., 2016, 35, 38–47. 24. Z. Rajfur, P. Roy, C. Otey, L. Romer and K. Jacobson, Nat. Cell Biol., 2002, 4, 286–293. 25. T. Tanabe, M. Oyamada, K. Fujita, P. Dai, H. Tanaka and T. Takamatsu, Nat. Methods, 2005, 2, 503–505. 26. M. A. McLean, Z. Rajfur, Z. Chen, D. Humphrey, B. Yang, S. G. Sligar and K. Jacobson, Anal. Chem., 2009, 81, 1755–1761. 27. M. E. Bulina, D. M. Chudakov, O. V. Britanova, Y. G. Yanushevich, D. B. Staroverov, T. V. Chepurnykh, E. M. Merzlyak, M. A. Shkrob, S. Lukyanov and K. A. Lukyanov, Nat. Biotechnol., 2006, 24, 95–99. 28. S. Pletnev, N. G. Gurskaya, N. V. Pletneva, K. A. Lukyanov, D. M. Chudakov, V. I. Martynov, V. O. Popov, M. V. Kovalchuk, A. Wlodawer, Z. Dauter and V. Pletnev, J. Biol. Chem., 2009, 284, 32028–32039. 29. P. Carpentier, S. Violot, L. Blanchoin and D. Bourgeois, FEBS Lett., 2009, 583, 2839–2842. 30. F. Baumgart, A. Rossi and A. S. Verkman, J. Gen. Physiol., 2012, 139, 83–91. 31. T. Jarvela and A. D. Linstedt, Mol. Biol. Cell, 2014, 25, 133–144. 32. E. A. Vitriol, A. L. Wise, M. E. Berginski, J. R. Bamburg and J. Q. Zheng, Mol. Biol. Cell, 2013, 24, 2238–2247. 33. W. Waldeck, G. Mueller, K. H. Glatting, A. Hotz-Wagenblatt, N. Diessl, S. Chotewutmonti, J. Langowski, W. Semmler, M. Wiessler and K. Braun, Int. J. Med. Sci., 2013, 10, 1136–1148. 34. L. Lan, S. Nakajima, L. Wei, L. Sun, C. L. Hsieh, R. W. Sobol, M. Bruchez, B. Van Houten, A. Yasui and A. S. Levine, Nucleic Acids Res., 2014, 42, 2330–2345. 35. N. V. Petrova, A. V. Luzhin, E. O. Serebrovskaya, A. P. Ryumina, A. K. Velichko, S. V. Razin and O. L. Kantidze, Aging, 2016, 8, 2449–2462. 36. L. Sun, R. Tan, J. Xu, J. LaFace, Y. Gao, Y. Xiao, M. Attar, C. Neumann, G. M. Li, B. Su, Y. Liu, S. Nakajima, A. S. Levine and L. Lan, Nucleic Acids Res., 2015, 43, 6334–6347. 37. K. Jewhurst, M. Levin and K. A. McLaughlin, J. Cell Death, 2014, 7, 25–31. 38. C. Buckley, M. T. Carvalho, L. K. Young, S. A. Rider, C. McFadden, C. Berlage, R. F. Verdon, J. M. Taylor, J. M. Girkin and J. J. Mullins, Sci. Rep., 2017, 7, 5096. 39. C. Teh, D. M. Chudakov, K. L. Poon, J. Y. Sek, K. Shidlovsky, S. Lukyanov, V. Korzh, BMC Dev. Biol., 2010, 10, 110.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

134

Takeharu Nagai and Yemima Dani Riani

40. J. Kobayashi, H. Shidara, Y. Morisawa, M. Kawakami, Y. Tanahashi, K. Hotta and K. Oka, Neurosci. Lett., 2013, 548, 261–264. 41. K. Takemoto, T. Matsuda, N. Sakai, D. Fu, M. Noda, S. Uchiyama, I. Kotera, Y. Arai, M. Horiuchi, K. Fukui, T. Ayabe, F. Inagaki, H. Suzuki and T. Nagai, Sci. Rep., 2013, 3, 2629. 42. K. Kim, G. Lakhanpal, H. E. Lu, M. Khan, A. Suzuki, M. Kato-Hayashi, R. Narayanan, T. T. Luyben, T. Matsuda, T. Nagai, T. A. Blanpied, Y. Hayashi and K. Okamoto, Neuron, 2015, 87, 813–826. 43. X. Shu, V. Lev-Ram, T. J. Deerinck, Y. Qi, E. B. Ramko, M. W. Davidson, Y. Jin, M. H. Ellisman and R. Y. Tsien, PLoS Biol., 2011, 9, e1001041. 44. F. M. Pimenta, R. L. Jensen, T. Breitenbach, M. Etzerodt and P. R. Ogilby, Photochem. Photobiol., 2013, 89, 1116–1126. 45. M. E. Barnett, T. M. Baran, T. H. Foster and A. P. Wojtovich, Free Radical Biol. Med., 2018, 116, 134–140. 46. K. E. Mironova, G. M. Proshkina, A. V. Ryabova, O. A. Stremovskiy, S. A. Lukyanov, R. V. Petrov and S. M. Deyev, Theranostics, 2013, 3, 831–840. 47. J. Y. Lin, S. B. Sann, K. Zhou, S. Nabavi, C. D. Proulx, R. Malinow, Y. Jin and R. Y. Tsien, Neuron, 2013, 79, 241–253. 48. A. P. Wojtovich, A. Y. Wei, T. A. Sherman, T. H. Foster and K. Nehrke, Sci. Rep., 2016, 6, 29695. 49. M. Westberg, L. Holmegaard, F. M. Pimenta, M. Etzerodt and P. R. Ogilby, J. Am. Chem. Soc., 2015, 137, 1632–1642. 50. Y. Qi and S. Xu, Bio-Protoc., 2013, 3, 3–6. 51. K. Makhijani, T. L. To, R. Ruiz-González, C. Lafaye, A. Royant and X. Shu, Cell Chem. Biol., 2017, 24, 110–119. 52. C. A. Robertson, D. H. Evans and H. Abrahamse, J. Photochem. Photobiol., B, 2009, 96, 1–8. 53. J. C. Junqueira, M. a. Ribeiro, R. D. Rossoni, J. O. Barbosa, S. M. R. Querido and O. C. Jorge, Photomed. Laser Surg., 2010, 28, 67–72. 54. L. Ryskova, V. Buchta and R. Slezak, Cent. Eur. J. Biol., 2010, 5, 400–406. 55. E. O. Serebrovskaya, E. F. Edelweiss, O. A. Stremovskiy, K. A. Lukyanov, D. M. Chudakov and S. M. Deyev, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 9221–9225. 56. A. P. Ryumina, E. O. Serebrovskaya, M. V. Shirmanova, L. B. Snopova, M. M. Kuznetsova, I. V. Turchin, N. I. Ignatova, N. V. Klementieva, A. F. Fradkov, B. E. Shakhov, E. V. Zagaynova, K. A. Lukyanov and S. A. Lukyanov, Biochim. Biophys. Acta, 2013, 1830, 5059–5067. 57. M. V. Shirmanova, E. O. Serebrovskaya, K. A. Lukyanov, L. B. Snopova, M. A. Sirotkina, N. N. Prodanetz, M. L. Bugrova, E. A. Minakova, I. V. Turchin, V. A. Kamensky, S. A. Lukyanov and E. V. Zagaynova, J. Biophotonics, 2013, 6, 283–290. 58. E. O. Serebrovskaya, A. P. Ryumina, M. E. Boulina, M. V. Shirmanova, E. V. Zagaynova, E. A. Bogdanova, S. A. Lukyanov and K. A. Lukyanov, J. Biomed. Opt., 2013, 19, 71403. 59. M. Shirmanova, D. Yuzhakova, L. Snopova, G. Perelman, E. Serebrovskaya, K. Lukyanov, I. Turchin, P. Subochev, S. Lukyanov, V. Kamensky and E. Zagaynova, PLoS One, 2015, 10, e0144617. 60. Z. X. Liao, Y. C. Li, H. M. Lu and H. W. Sung, Biomaterials, 2014, 35, 500–508. 61. A. P. Ryumina, E. O. Serebrovskaya, D. B. Staroverov, O. A. Zlobovskaya, A. S. Shcheglov, S. A. Lukyanov and K. A. Lukyanov, BioTechniques, 2016, 61, 92–94. 62. M. Muthiah, S. H. Park, M. Nurunnabi, J. Lee, Y. Kyu Lee, H. Park, B. Il Lee, J. J. Min and I. K. Park, Colloids Surf., B, 2014, 116, 284–294.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00117

Optogenetic Control of the Generation

135

63. L. Liang, Y. Lu, R. Zhang, A. Care, T. A. Ortega, S. M. Deyev, Y. Qian and A. V.   Zvyagin, Acta Biomater., 2017, 51, 461–470. 64. J. He, Y. Wang, M. A. Missinato, E. Onuoha, L. A. Perkins, S. C. Watkins, C. M. St Croix, M. Tsang and M. P. Bruchez, Nat. Methods, 2016, 13, 263–268. 65. T. Shibuya, Y. Tsujimoto, J. Photochem. Photobiol. B., 2012, 117, 1–12. 66. A. Pudasaini, K. K. El-Arab and B. D. Zoltowski, Front. Mol. Biosci., 2015, 2, 18. 67. D. C. Williams, R. ElBejjani, P. Ramirez, S. Coakley, S. A. Kim, H. Lee, Q. Wen, A. Samuel, H. Lu, M. A. Hilliard and M. Hammarlund, Cell Rep., 2013, 5, 553–563. 68. K. S. Sarkisyan, O. A. Zlobovskaya, D. A. Gorbachev, N. G. Bozhanova, G. V. Sharonov, D. B. Staroverov, E. S. Egorov, A. V. Ryabova, K. M. Solntsev, A. S. Mishin and K. A. Lukyanov, PLoS One, 2015, 10, e0145287. 69. E. De Rosny and P. Carpentier, J. Am. Chem. Soc., 2012, 134, 18015–18021. 70. Y. D. Riani, T. Matsuda, K. Takemoto, T. Nagai, BMC Biol., 2018, 16, 50. 71. Y. B. Qi, E. J. Garren, X. Shu, R. Y. Tsien and Y. Jin, Proc. Natl. Acad. Sci., 2012, 109, 7499–7504. 72. S. Xu, A. D. Chisholm, Sci. Rep., 2016, 6, 21271.

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00137

Chapter 7

Optogenetic Tools for Quantitative Biology: The Genetically Encoded PhyB–PIF Light-inducible Dimerization System and Its Application for Controlling Signal Transduction S. Odaa, Y. Udaa,b, Y. Gotoa, H. Miuraa,c and K. Aoki*a,d a

Exploratory Research Center on Life and Living Systems, National Institute   for Basic Biology, National Institutes of Natural Sciences, Okazaki, Aichi,   444-8787, Japan; bDepartment of Pathology and Biology of Diseases, Graduate School of Medicine, Kyoto University, Sakyo-ku, Kyoto, 606-8501, Japan;   c Laboratory of Bioimaging and Cell Signaling, Graduate School of Biostudies, Kyoto University, Sakyo-ku, Kyoto, 606-8501, Japan; dDepartment of Basic Biology, Faculty of Life Science, SOKENDAI (Graduate University for Advanced Studies), Myodaiji, Okazaki, Aichi, 444-8787, Japan *E-mail: [email protected]

Table of Contents 7.1.  I ntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.  Light-induced Dimerization (LID) Systems for Controlling   Cell Signaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.  Synthesis of the Chromophore of Phytochrome . . . . . . . . . . . . . . . 7.4.  PhyB–PIF LID system. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5.  Quantitative Manipulation of Cell Signaling by the   PhyB–PIF System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

137

139 139 141 143 143

View Online

138

S. Oda, Y. Uda, Y. Goto, H. Miura and K. Aoki

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00137

7.6.  Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

View Online

Optogenetic Tools for Quantitative Biology

139

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00137

7.1. Introduction Signal transduction is a mechanism by which the cell responds to an extracellular input such as growth factor and finally outputs a phenotype to adapt to the constantly changing environment. Signal transduction is comprised of a diverse array of chemical reactions that take place within the cell. Several decades of extensive investigations have unveiled the complicated network of signal transduction pathways, including feedback/feedforward regulations.1,2 For instance, it has been demonstrated that the epidermal growth factor (EGF) receptor–Ras–ERK MAP kinase pathway includes a series of negative feedback regulations to fine tune the amplitude and duration of ERK signaling.3,4 To understand the regulatory reaction quantitatively, a direct approach involves a combination of perturbation and sampling with a frequency higher than that of the regulatory system, which is known as the Nyquist frequency.5 The recent advent of fluorescence imaging techniques now allows us to observe the dynamics of signal transduction with sufficient temporal and spatial resolution.6,7 However, perturbation methods of signal transduction are still under development. To perturb signal transduction, small molecules including organic compounds have been used in cell biology. In particular, kinase inhibitors are commonly used, although the number of available inhibitors is limited. Chemically induced dimerization (CID) is a technique in which a small molecule (i.e. dimerizer) induces homo- or heterodimerization of two proteins.8,9 The rapamycin-inducible dimerization of FKBP and FKBP–rapamycin binding (FRB) domain is a representative model of the CID system, successfully demonstrating small GTPase activation,10,11 receptor tyrosine kinase activation12 and ERK activation.13 Although the CID system has difficulty in subcellular control, the recent development of caged compounds allows control of spatially confined signal transduction.14,15 Nonetheless, the high affinity of dimerizers to the target protein hinders repetitive and reversible regulation of signal transduction.

7.2. Light-induced Dimerization (LID) Systems for Controlling Cell Signaling Optogenetics is an emerging technique in which a light-responsive protein is used for the perturbation of biological function by light, and is extensively applied in research in neuroscience.16 Furthermore, the concept of optogenetics expands the idea of light-induced manipulation of cell signaling in space and time. The photoresponsive proteins derived from fungi, cyanobacteria and plants and modified fluorescent proteins are utilized. The principle is mainly based on their authentic functions; irradiation with light induces conformational changes in the photoresponsive proteins, followed by binding to their target domains or proteins (Figure 7.1A). The particular

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00137

140

S. Oda, Y. Uda, Y. Goto, H. Miura and K. Aoki

Figure 7.1.  Light-induced dimerization (LID) system. (A) The spectra of previously reported LID systems range from the UV to the infrared region. (B) Apo-PhyB covalently attaches its chromophore, which is phycocyanobilin (PCB) or phytochromobilin (PΦB), to produce holo-PhyB. There are two holo-PhyB forms, PhyB (Pr) and PhyB (Pfr), which interchange in a manner dependent on the chromophore status. Only PhyB (Pfr) associates with PIF. (C) Schematic representation of the absorbance spectra of PhyB (Pr) and PhyB (Pfr).

wavelength of the light depends on the chromophore of the photoresponsive proteins17 (Figure 7.1A). The idea of light-induced dimerization (LID) has expanded rapidly in recent years as a system capable of manipulating signal transduction, because it works in a manner akin to the CID system. For example, UV-B resistance 8 (UVR8), light-, oxygen- and voltage-sensing (LOV) domain, cryptochrome 2 (CRY2) and Dronpa have been reported so far.17,18 Most of the LID systems reported previously are responsive to ultraviolet–blue light (Figure 7.1A), thereby preventing GFP-based imaging because of the cross-activation of the photoresponsive protein by the excitation light of GFP. Further, shorter wavelength light has a disadvantage in tissue penetration compared with longer wavelength light such as red and infrared light. Phytochrome B (PhyB)–phytochrome interacting factor (PIF) has been reported as a LID system using red and infrared light19,20 (Figure 7.1B). PhyB is a member of the phytochromes, which are a widespread family of red/ infrared-responsive photoreceptors that regulate photomorphogenesis in plants.21 As a LID system, PhyB binds to PIF upon irradiation with red light (∼670 nm), whereas the PhyB–PIF complex is dissociated by infrared light (∼730 nm) (Figure 7.1B and C). The reversible control of association and dissociation by light is a unique feature in the PhyB–PIF system, because either

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00137

Optogenetic Tools for Quantitative Biology

141

association or dissociation is regulated by light in other LID systems. Covalent attachment of a chromophore, phycocyanobilin (PCB) or phytochromobilin (PΦB), is required for the light-responsive function of PhyB–PIF (Figure 7.1B). These chromophores are not present in vertebrates, hence it is necessary to add PCB purified from cyanobacteria to the cells externally. The purified PCB was introduced into budding yeast (Saccharomyces cerevisiae),19,22 fission yeast (Schizosaccharomyces pombe),23 mammalian cultured cells20,24 and zebrafish (Danio rerio)25 and the light-responsive function of PhyB–PIF in their respective systems was confirmed. However, this was not the case in the fruit fly.26 Furthermore, the PhyB–PIF system has not yet been introduced into the nematode Caenorhabditis elegans or the mouse. Interestingly, administered PCB does not have any toxic, but rather beneficial, effects on mice.27

7.3. Synthesis of the Chromophore of Phytochrome It is difficult to administer PCB to animals in a non-invasive manner. One way to overcome the issue of PCB introduction is to synthesize the chromophore within the cell by introducing enzymes that synthesize the chromophore. In photosynthetic organisms, phycocyanobilin:ferredoxin oxidoreductase (PcyA) and phytochromobilin:ferredoxin oxidoreductase (HY2) catalyze biliverdin to generate PCB and PΦB, respectively28,29(Figure 7.2A). Biliverdin is produced from heme, which is metabolized by heme oxygenase 1 (HO1 or HY1). These reactions require ferredoxin (Fd) and ferredoxin:NADP+ oxidoreductase (Fnr). It has been reported that PCB and PΦB are heterologously synthesized in Escherichia coli by coexpression of HO1 with PcyA and HY2, respectively,21,30 indicating that Fd- and Fnr-like proteins of E. coli complement the reactions of PCB and PΦB synthesis. Furthermore, the methylotrophic yeast Pichia pastoris successfully synthesizes PCB and PΦB with the coexpression of HO1–PcyA and HO1–HY2.31 Müller et al. reported for the first time the synthesis of PCB in mammalian cells by expressing both HO1 and PcyA.32 Both PcyA and HO1 were targeted to mitochondria where heme was synthesized.32–34 Using this system, they demonstrated the induction of gene expression by the PhyB–PIF LID system in mammalian cells and zebrafish.32,35 However, in our hands, the amount of PCB synthesized in this system was not sufficient to visualize PhyB–PIF LID in cultured mammalian cells. Recently, we developed a method for the efficient synthesis of PCB by introduction of the gene products of HO1, PcyA, Fd and Fnr into mitochondria in mammalian cells (Figure 7.2B and C) and fission yeast (S. pombe).23 The PCB was visualized by red fluorescence emanating from PCB bound to PhyBY276H mutant20,36 (Figure 7.2B). The amount of PCB synthesized was comparable to that of externally delivered PCB (Figure 7.2C). When we expressed only HO1 and PcyA, PCB fluorescence was not detectable (data not shown), suggesting that mammalian homologs of Fd and Fnr hardly function for PCB

View Online

S. Oda, Y. Uda, Y. Goto, H. Miura and K. Aoki

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00137

142

Figure 7.2.  Synthesis of phycocyanobilin (PCB) in mammalian cells. (A) The metabolic pathway of phytochrome chromophores. (B) Scheme for measurement of PCB abundance. The PhyB mutant PhyB-Y276H emits fluorescence when it binds to PCB. (C) Quantification of PCB in HeLa cells. Average values of PCB fluorescence are shown as a bar graph with the standard deviation. MTS is the mitochondrial targeting sequence. PHFF is a polycistronic vector expressing MTS-PcyA, MTS-HO1, MTS-Fd and MTS-Fnr.

synthesis. To facilitate the gene transfer and achieve efficient gene expression, we constructed the polycistronic vector pPHFF,23 which enables PCB to be produced by introducing a single plasmid DNA (Figure 7.2C).

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00137

Optogenetic Tools for Quantitative Biology

143

Biliverdin and PCB are known to be catalyzed by biliverdin reductase A (BVRA), generating bilirubin and phycocyanorubin (PCR) (Figure 7.2A). It has been reported that the knockdown (KD) of BVRA leads to an increase in PCB production in mammalian cells.32 We also confirmed that the knockout (KO) and KD of BVRA by CRISPR/Cas9 and RNAi, respectively, potentiated PCB production several-fold in HeLa cells and mouse embryonic stem cells.23 In our study, we did not observe any detectable adverse effects of PHFF expression and/or BVRA KO in cultured cells, possibly because biliverdin and bilirubin are provided from the serum in the culture medium. However, KO or KD of BVRA could affect the physiology of animals, because it is suggested that heme and its metabolites, biliverdin and bilirubin, have diverse physiological functions, including electron transfer, antioxidant and cell signaling, respectively.33,37,38 In addition, liver-specific BVRA KO mice show a broad spectrum of non-alcoholic fatty liver disease (NAFLD) phenotypes, indicating that BVRA protects against lipid accumulation and NAFLD.39

7.4. PhyB–PIF LID system It is expected that PCB-bound PhyB will respond to red and infrared light, followed by association and dissociation with PIF, respectively. We assessed the interaction of PhyB with PIF by the plasma membrane recruitment upon exposure to light (Figure 7.3A). HeLa cells without PCB treatment or PHFF expression showed no membrane recruitment when exposed to red light (data not shown), indicating the requirement of the chromophore for PhyB–PIF binding. Interestingly, the addition of PCB that was subjected to high-performance liquid chromatographic (HPLC) purification demonstrated clear light-induced membrane recruitment of PIF–EGFP (enhanced green fluorescent protein), whereas the crude PCB extract that was not purified by HPLC was incapable of responding to light (Figure 7.3B and C). Both PCBs emitted infrared fluorescence when they bound to PhyB-Y276H (data not shown), suggesting the contamination of non-functional PCB in the crude extract. Even under this condition, the introduction of pPHFF induced more robust photoswitching of PhyB–PIF binding (Figure 7.3B and C). Our data also indicate that PhyB– PIF3 has features of faster kinetics and better reversibility compared with PhyB–PIF6.23

7.5. Quantitative Manipulation of Cell Signaling by the PhyB– PIF System LID of PhyB–PIF has been applied to the manipulation of signal transduction such as Rac1 20, PI3K40, ERK MAP kinase24 and GPCR signaling41 and gene expression.19,42 We also demonstrated light-induced ERK activation with the

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00137

View Online

Figure 7.3.  Photoswitching of PhyB–PIF3 binding. (A) Schematic representation of membrane translocation of PIF3–mEGFP (monomeric enhanced green fluorescent protein) upon exposure to light through binding to PhyB–mCherry-HRasCT, which is localized at the plasma membrane. (B) Representative images of PIF3–mEGFP in HeLa cells treated with crude PCB extract (top row), HPLC-purified PCB extract (middle row) or HeLa cells expressing PHFF (bottom row). Crude PCB extract and HPLC-purified PCB extract were purchased from Santa Cruz and SiChem, respectively. (C) Quantification of membrane translocation of the PIF3–mEGFP in (B). Average values of membrane-translocated PIF3–mEGFP are plotted as a function of time with the standard deviation.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00137

Optogenetic Tools for Quantitative Biology

145

Figure 7.4.  Synthetic ERK activation by the PhyB–PIF LID system. (A) Scheme of ERK activation by the PhyB–PIF LID system. ERK activity is monitored by a FRET biosensor for ERK, EKAREV-NLS. (B) Representative images of ERK activity at the nucleus in BVRA KO HeLa cells expressing EKAREV-NLS, PhyB–mCherry-HRasCT, PIF3–CRaf and PHFF. Images of ERK activity are represented in the intensity-modulated display (IMD) mode. (C) The average ERK activity (normalized FRET/CFP ratio) is plotted as a function of time with the standard deviation.

PhyB–PIF LID system combined with endogenously synthesized PCB, showing the artificial oscillation of ERK activation23,43 (Figure 7.4). In this system, CRaf, MAP kinase kinase kinase (MAP3K) for ERK, which is fused with PIF3, is recruited to the plasma membrane through binding to PhyB upon illumination with red light (Figure 7.4A). Our results were consistent with previous reports, which confirmed that the membrane recruitment of CRaf suffices to activate ERK.44 Toettcher's group has played a leading role in the quantitative manipulation of signal transduction with the PhyB–PIF system. Taking full advantage of the PhyB–PIF system, they succeeded in quantitatively controlling the output of signal transduction by feedback control of red/infrared light.40 More recently, they visualized the information transfer from ERK to target gene mRNAs to their proteins at the single-cell level by combining the PhyB–PIF system with the CRISPR/Cas9 genome editing technique, providing new insights into how the activation pattern of ERK dynamics encodes information on gene and protein expression.45

7.6. Conclusion The LID system can be applied to the control of other signaling pathways through not only membrane recruitment17 but also aggregation,46 sequestration to the other subcellular space22,47 and nuclear import/export.48,49 A

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00137

146

S. Oda, Y. Uda, Y. Goto, H. Miura and K. Aoki

single-chain, intramolecular photoswitchable protein kinase is also a good target for the application of the LID system.50 Moreover, multiplexed optogenetic control of signal transduction has already been demonstrated in combination with UV- and blue light-responsive LID systems and the PhyB–PIF LID system.51 For the application of in vivo optogenetic manipulation of cell signaling, the LID system controlled by longer wavelength light such as red and infrared light is preferable because light of those wavelengths has two advantageous properties compared with shorter wavelength light, namely the higher tissue penetrance and lower phototoxicity of the light. Hence PhyB–PIF is an ideal LID system for in vivo optogenetic manipulation of cell signaling. More recently, Verkhusha and co-workers reported a red/infrared light-inducible LID system with the bacterial phytochrome BphP1 and its natural partner PpsR2 from Rhodopseudomonas palustris bacteria.52 In this system, endogenous biliverdin is used as a chromophore of BphP1, thereby lowering the requirement for chromophore addition or its synthesis. Although, at the moment, the BphP1–PpsR2 LID system needs the external addition of biliverdin for adequate functioning and demonstrates slower association and dissociation kinetics than the PhyB–PIF system, these drawbacks may possibly be improved upon. In summary, as technology progresses, optogenetics is opening up a new window for the optical control of biological systems, permitting quantitative assessment of causality between cell signaling and biological functions.

Acknowledgements We thank Michiyuki Matsuda and the members of the Aoki laboratory for their helpful input. K. Onoda and E. Ebine are also to be thanked for their technical assistance. S. O. and K. A. were supported by JST CREST (JPMJCR1654). K. A. was funded by JSPS KAKENHI Grant No. 16H01425 “Resonance Bio,” 16H01447 and 16KT0069, the Hori Sciences and Arts Foundation, Takeda Science Foundation and the Nakajima Foundation. S. O. was funded by JSPS KAKENHI Grant No. 17H07406 and a grant received from the Sumitomo Foundation.

References 1. K. Oda, Y. Matsuoka, A. Funahashi and H. Kitano, Mol. Syst. Biol., 2005, 1, E1–E17. 2. M. C. Mendoza, E. E. Er and J. Blenis, Trends Biochem. Sci., 2011, 36, 320–328. 3. Y. Fujita, N. Komatsu, M. Matsuda and K. Aoki, FEBS J., 2014, 281, 3177–3192. 4. O. Brandman and T. Meyer, Science, 2008, 322, 390–395. 5. T. Inoue and T. Meyer, PLoS One, 2008, 3, e3068. 6. A. Miyawaki, Dev. Cell, 2003, 4, 295–305. 7. K. Aoki, E. Kiyokawa, T. Nakamura and M. Matsuda, Philos. Trans. R. Soc., B, 2008, 363, 2143–2151.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00137

Optogenetic Tools for Quantitative Biology

147

8. R. Derose, T. Miyamoto and T. Inoue, Pflugers Arch., 2013, 465, 409–417. 9. S. J. Kopytek, R. F. Standaert, J. C. D. Dyer and J. C. Hu, Chem. Biol., 2000, 7, 313–321. 10. T. Inoue, W. Do Heo, J. S. Grimley, T. J. Wandless and T. Meyer, Nat. Methods, 2005, 2, 415–418. 11. M. Fivaz, S. Bandara, T. Inoue and T. Meyer, Curr. Biol., 2008, 18, 44–50. 12. I. A. Graef, L. J. Holsinger, S. Diver, S. L. Schreiber and G. R. Crabtree, EMBO J., 1997, 16, 5618–5628. 13. K. Aoki, M. Yamada, K. Kunida, S. Yasuda and M. Matsuda, Proc. Natl. Acad. Sci. U. S. A., 2011, 108, 12675–12680. 14. N. Umeda, T. Ueno, C. Pohlmeyer, T. Nagano and T. Inoue, J. Am. Chem. Soc., 2011, 133, 12–14. 15. E. R. Ballister, C. Aonbangkhen, A. M. Mayo, M. A. Lampson and D. M. Chenoweth, Nat. Commun., 2014, 5, 5475. 16. L. Fenno, O. Yizhar and K. Deisseroth, Annu. Rev. Neurosci., 2011, 34, 389–412. 17. M. Endo and T. Ozawa, J. Photochem. Photobiol., C, 2017, 30, 10–23. 18. W. W. D. Mühlhäuser, A. Fischer, W. Weber and G. Radziwill, Biochim. Biophys. Acta, Mol. Cell Res., 2017, 1864, 280–292. 19. S. Shimizu-Sato, E. Huq, J. M. Tepperman and P. H. Quail, Nat. Biotechnol., 2002, 20, 1041–1044. 20. A. Levskaya, O. D. Weiner, W. A. Lim and C. A. Voigt, Nature, 2009, 461, 997–1001. 21. K. Mukougawa, H. Kanamoto, T. Kobayashi, A. Yokota and T. Kohchi, FEBS Lett., 2006, 580, 1333–1338. 22. X. Yang, A. P.-T. Jost, O. D. Weiner and C. Tang, Mol. Biol. Cell, 2013, 24, 2419–2430. 23. Y. Uda, Y. Goto, S. Oda, T. Kohchi, M. Matsuda and K. Aoki, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, 11962–11967. 24. J. E. Toettcher, O. D. Weiner and W. A. Lim, Cell, 2013, 155, 1422–1434. 25. C. E. Buckley, R. E. Moore, A. Reade, A. R. Goldberg, O. D. Weiner and J. D. W. Clarke, Dev. Cell, 2016, 36, 117–126. 26. H. E. Johnson, Y. Goyal, N. L. Pannucci, T. Schüpbach, S. Y. Shvartsman and J. E. Toettcher, Dev. Cell, 2017, 40, 185–192. 27. J. Zheng, T. Inoguchi, S. Sasaki, Y. Maeda, M. F. McCarty, M. Fujii, N. Ikeda, K. Kobayashi, N. Sonoda and R. Takayanagi, Am. J. Physiol.: Regul., Integr. Comp. Physiol., 2013, 304, R110–R120. 28. T. Kohchi, K. Mukougawa, N. Frankenberg, M. Masuda, A. Yokota and J. C. Lagarias, Plant Cell, 2001, 13, 425–436. 29. N. Frankenberg, Plant Cell Online, 2001, 13, 965–978. 30. G. A. Gambetta and J. C. Lagarias, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 10566–10571. 31. A. Y. Shin, Y. J. Han, P. S. Song and J. Il Kim, FEBS Lett., 2014, 588, 2964–2970. 32. K. Müller, R. Engesser, J. Timmer, F. Nagy, M. D. Zurbriggen and W. Weber, Chem. Commun., 2013, 49, 8970. 33. P. Ponka, Am. J. Med. Sci., 1999, 318, 241–256. 34. R. S. Ajioka, J. D. Phillips and J. P. Kushner, Biochim. Biophys. Acta, Mol. Cell Res., 2006, 1763, 723–736. 35. K. Müller, R. Engesser, S. Metzger, S. Schulz, M. M. Kämpf, M. Busacker, T. Steinberg, P. Tomakidi, M. Ehrbar, F. Nagy, J. Timmer, M. D. Zubriggen and W. Weber, Nucleic Acids Res., 2013, 41, e77. 36. A. J. Fischer, N. C. Rockwell, A. Y. Jang, L. A. Ernst, A. S. Waggoner, Y. Duan, H. Lei and J. C. Lagarias, Biochemistry, 2005, 44, 15203–15215.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00137

148

S. Oda, Y. Uda, Y. Goto, H. Miura and K. Aoki

37. N. Lerner-Marmarosh, T. Miralem, P. E. M. Gibbs and M. D. Maines, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 6870–6875. 38. G. J. Maghzal, M. C. Leck, E. Collinson, C. Li and R. Stocker, J. Biol. Chem., 2009, 284, 29251–29259. 39. T. D. Hinds, K. A. Burns, P. A. Hosick, L. McBeth, A. Nestor-Kalinoski, H. A. Drummond, A. A. Alamodi, M. W. Hankins, J. P. Vanden Heuvel and D. E. Stec, J. Biol. Chem., 2016, 291, 25179–25191. 40. J. E. Toettcher, D. Gong, W. A. Lim and O. D. Weiner, Nat. Methods, 2011, 8, 837–839. 41. G. Yu, H. Onodera, Y. Aono, F. Kawano, Y. Ueda, A. Furuya, H. Suzuki and M. Sato, Sci. Rep., 2016, 6, 35777. 42. K. Müller, R. Engesser, S. Metzger, S. Schulz, M. M. Kämpf, M. Busacker, T. Steinberg, P. Tomakidi, M. Ehrbar, F. Nagy, J. Timmer, M. D. Zubriggen and W. Weber, Nucleic Acids Res., 2013, 41, e77. 43. N. Komatsu, K. Aoki, M. Yamada, H. Yukinaga, Y. Fujita, Y. Kamioka and M. Matsuda, Mol. Biol. Cell, 2011, 22, 4647–4656. 44. D. Stokoe, S. G. Macdonald, K. Cadwallader, M. Symons and J. F. Hancock, Science, 1994, 264, 1463–1467. 45. M. Z. Wilson, P. T. Ravindran, W. A. Lim and J. E. Toettcher, Mol. Cell, 2017, 67, 757–769.e5. 46. L. J. Bugaj, A. T. Choksi, C. K. Mesuda, R. S. Kane and D. V. Schaffer, Nat. Methods, 2013, 10, 249–252. 47. H. Park, S. Lee and W. Do Heo, Methods Mol. Biol., 2016, 1408, 363–376. 48. D. Niopek, D. Benzinger, J. Roensch, T. Draebing, P. Wehler, R. Eils and B. Di Ventura, Nat. Commun., 2014, 5, 4404. 49. H. Yumerefendi, D. J. Dickinson, H. Wang, S. P. Zimmerman, J. E. Bear, B. Goldstein, K. Hahn and B. Kuhlman, PLoS One, 2015, 10, 1–19. 50. D. I. Schlipalius, N. Valmas, A. G. Tuck, R. Jagadeesan, L. Ma, R. Kaur, A. Goldinger, C. Anderson, J. Kuang, S. Zuryn, Y. S. Mau, Q. Cheng, P. J. Collins, M. K. Nayak, H. J. Schirra, M. A. Hilliard and P. R. Ebert, Science, 2012, 338, 807–810. 51. K. Müller, R. Engesser, S. Schulz, T. Steinberg, P. Tomakidi, C. C. Weber, R. Ulm,  J. Timmer, M. D. Zurbriggen and W. Weber, Nucleic Acids Res., 2013, 41, e124. 52. A. Kaberniuk, A. A. Shemetov and V. V. Verkhusha, Nat. Methods, 2016, 13, 591–597.

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

Chapter 8

Quantitative Control of Kinase Activity with a Mathematical Model Genki Kawamura and Takeaki Ozawa* Department of Chemistry, School of Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan *E-mail: [email protected]

Table of Contents 8.1.  I ntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 8.2.  Experimental Design of the PA-Akt System. . . . . . . . . . . . . . . . . . . . 153 8.2.1.  Principle of the PA-Akt System. . . . . . . . . . . . . . . . . . . . . . . . 153 8.2.2.  Design of the Construction of the PA-Akt   System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ��  154 8.2.3.  Notes on Light Illumination Wavelength and   Strength. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ��  155 8.3.  Application of the PA-Akt System for Cellular   Signaling Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 8.3.1.  Optogenetic Control of the PA-Akt System. . . . . . . . . . . . . . 158 8.3.2.  Dissecting the Signaling Pathway by the   PA-Akt System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ��  159 8.3.3.  Spatial Regulation of Actin Remodeling by   Localized Activation of PA-Akt. . . . . . . . . . . . . . . . . . . . . . . ��  159 8.3.4.  Prediction of Light-induced Akt Activation   Using a Mathematical Model. . . . . . . . . . . . . . . . . . . . . . . . ��  161 8.4.  Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

149

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

View Online

         

View Online

Quantitative Control of Kinase Activity

151

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

8.1. Introduction Optogenetics is a method that uses light to control cellular events. Applications of optogenetics include manipulation of an action potential,1 translocation of a protein of interest to cellular compartments,2 oligomerization-mediated activation of cellular signaling3 and regulation of protein function by the uncaging of activation domains.4 Photosensory proteins are used in the optogenetic system, where they absorb light and undergo a conformational change. Protein–protein interactions serve as a fundamental motif in cellular signal transduction. Therefore, light-inducible dimers are useful for controlling the protein function in a signaling pathway. Representatives of light-inducible dimers include the cryptochrome 2 (Cry2) and cryptochrome-interacting basic helix–loop–helix 1 (Cib1) pair, the iLID system,5 the Magnets6 for blue light-absorbing photosensory proteins and phytochrome B (PhyB) and phytochrome-interacting factor (Pif) pair for a red light-absorbing system.7 Recently, an infrared optogenetic dimerization system using bacterial phytochrome was developed,8,9 further expanding the available spectra for optogenetic tool development. Each system has a unique property, especially in the temporal kinetics of association and dissociation, and which is used in the perturbation of signaling pathways. Because of the high spatiotemporal delivery of incident light, optogenetic tools have benefits for revealing cellular functions that require analysis of spatial and temporal regulation.10 The applicability of spatial and temporal regulation by optogenetics is highlighted in a study that applied an optogenetic son of sevenless (optoSOS) system to elucidate intercellular propagation of the Erk signaling by single-cell specific activation and also to show activation frequency-dependent regulation of downstream gene expression.11 The capabilities of optogenetics to manipulate widely diverse cellular events with high spatiotemporal resolution serve as effective tools for the analysis of cellular signaling pathways that conventional techniques such as chemical stimulation or microfluidic activation cannot achieve. Akt is a kinase involved in various cellular functions, including cellular metabolism, apoptosis and cell proliferation and differentiation. Akt, which is crucially important for insulin and growth factor signaling, functions as a hub molecule for the regulation of diverse downstream pathways.12 A fairly recent study demonstrated that selective activation patterns of Akt downstream molecules are determined from the temporal kinetics of Akt, in which different downstream pathways of Akt are induced selectively by the distinct activation pattern of Akt, resulting in specific control of the signaling pathway by a single hub regulator.13,14 However, such studies require a controlled administration pattern and concentration of insulin to change the temporal activity of Akt. Thereby, non-Akt-dependent pathways that insulin activates, such as the Ras–Raf–MEK–Erk pathway and protein kinase C (PKC) pathway, are also upregulated.15 This upregulation makes it difficult to discern the causative relation between the activity pattern of Akt and the outputs, which are potentially affected by other signaling pathways. Another study described

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

152

Genki Kawamura and Takeaki Ozawa

the importance of the activation pattern of Akt and another hub protein in growth factor signaling, Erk, for cell fate determination and demonstrated that the degree of Akt and Erk activation determines the cell fate in proliferation and differentiation.16 Nonetheless, it remains elusive whether the activation rate of Akt and Erk is sufficient for cell fate decisions because of a lack of a method to activate these signaling pathways selectively and exclusively. These limitations underscore the importance of tools to separate the activation patterns of individual pathways to elucidate the contribution of each signaling pathway to the downstream effects. The combinatorial use of inhibitors and activators potentially interferes with other pathways. Therefore, a method that can manipulate specific proteins without perturbing other signaling pathways can be anticipated as a desirable tool for dissecting signaling. Moreover, for the understanding of causative relations between temporal activation patterns of signaling molecules to the downstream outputs, some methods must be found that  are capable of manipulating protein functions with precise timing. Considering these requirements, optogenetics is an ideal technique for analyzing individual signaling pathways. Recent developments in optogenetic tools for growth factor-mediated signaling permit such specific activation and dissection of the signaling pathway at different node proteins. Figure 8.1 presents schematic illustrations of the effectors in growth factor signaling pathways and the availability of optogenetic tools for the respective molecules. Activation of receptor tyrosine kinase (RTK) such as fibroblast growth factor receptor (FGFR) and epidermal growth factor receptor (EGFR) was accomplished using light-inducible dimerization of the light-, oxygen- and voltage-sensing (LOV) domain of aureochrome-1,17 which activates both the PI3K–Akt and Ras–Raf–MEK–Erk pathways. Optogenetic dissection

Figure 8.1.  Schematic diagram of the molecules involved in the growth factor signaling pathway, including the Ras–Raf–MEK–Erk pathway (red circles) and PI3K–Akt pathway (blue circles). The availability of reported optogenetic tools for each molecule is indicated by a blue flash next to the protein name.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

Quantitative Control of Kinase Activity

153

of the Ras–Raf–MEK–Erk signaling pathway can be achieved at different nodes in the signaling pathway, including SOS,11 Raf18,19 and MEK.20 For the PI3K–Akt signaling pathway, activation at PI3K21,22 or Akt22,23 is possible. Activation at different nodes results in different actions of the output, as demonstrated for the distinct role of PI3K and Akt in insulin action in adipocytes by comparing the optogenetic activation of PI3K and Akt.22 Using these tools, analysis of specific node proteins in growth factor signaling can be achieved. This chapter specifically describes an optogenetic method to control the function of protein kinase Akt. This system, photoactivatable Akt (PA-Akt), consists of the light-inducible dimerization pair Cry2 and Cib1.23 Selective manipulation of Akt function by light is used as a spatial and temporal regulator of Akt signaling, which allows the identification of the dynamic regulation of Akt signaling. Furthermore, computational modeling of the PA-Akt system predicts the Akt activation pattern as a function of light, which is useful as a generator of the Akt activation pattern for decoding temporal codes of the Akt signaling. We describe the basic design and use of the PA-Akt system for the selective and spatiotemporal perturbation of the Akt signaling pathway.

8.2. Experimental Design of the PA-Akt System 8.2.1. Principle of the PA-Akt System Phosphorylation of Akt, a kinase activated upon stimulation by phosphorylation of threonine at the 308th residue (Thr308) and serine at the 473rd residue (Ser473), takes place at the plasma membrane by upstream kinases such as phosphoinositide-dependent kinase-1 (PDK1) and mammalian target of rapamycin complex 2 (mTORC2). At the onset of activation, Akt translocates to the plasma membrane, where it is phosphorylated by these kinases. Plasma membrane anchoring of Akt protein by a myristoyl sequence (myr) is known to cause constitutive activation of Akt.24 The PA-Akt system is designed to mimic this physiological activation mechanism of Akt; the PA-Akt system causes translocation of Akt to the plasma membrane using a light-inducible dimerization system to activate Akt. For selection of the dimerization system, photoreceptor proteins from Arabidopsis thaliana, Cry2 and Cib1, were used. This photoreceptor pair forms a dimer upon a conformational change of Cry2 induced by illumination with blue light (420–490 nm). Light-inducible dimerization of Cry2–Cib1 is advantageous for the large dynamic range of photoinduced association, with rapid dissociation on the order of minutes. This rapid dissociation makes the Cry2–Cib1 pair ideal for a system that requires a low background of signal leakage with temporally reversible interactions. Accordingly, PA-Akt comprises Cry2 fused to Akt (Cry2–Akt) and plasma membrane-anchored Cib1 with a myr sequence (myr–Cib1). Upon illumination with blue light, Cry2 changes conformation. Subsequently, Cry2–Akt translocates to the plasma membrane by interaction

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

154

Genki Kawamura and Takeaki Ozawa

Figure 8.2.  Schematic diagram depicting light-induced activation of Cry2–Akt. The PA-Akt system consists of Cry2-fused Akt (Cry2–Akt) and plasma membrane-  anchored Cib1 (myr–Cib1). Before light stimulation, Cry2–Akt is inactive and located in the cytoplasm. Upon illumination with blue light (420–490 nm), Cry2 changes conformation and binds to Cib1 located at the plasma membrane. Subsequent activation of Cry2–Akt occurs upon phosphorylation from upstream kinases at the plasma membrane.

with myr–Cib1, resulting in the activation of Cry2–Akt by phosphorylation from PDK1 and mTORC2 (Figure 8.2). Formation of the Cry2–Cib1 dimer is a reversible process. Therefore, a temporal activity pattern of Akt can be generated using this system by changing the light illumination pattern. 8.2.2. Design of the Construction of the PA-Akt System In the probe design of an optogenetic tool, the orientation or linker length between a photoreceptor and the target protein sometimes leads to difficulties in the system functionality. Here, based on the design strategy of the PA-Akt system, we describe a basic strategy for constructing an optogenetic module. The PA-Akt system consists of Cry2 fused to Akt and plasma membrane-anchored Cib1 protein. Therefore, we particularly examine the construction design using a Cry2–Cib1 photoreceptor pair. Cry2 is a photoreceptor protein that forms a heterodimer with Cib1 upon illumination with blue light. Reportedly, Cry2 also forms a homodimer or oligomer,25 and mutation in the photolyase homology region enhances oligomer formation.3 Because of this property, Cry2 has been engineered as a molecular tool for light-dependent cluster formation with addition of a short peptide26 or low-complexity sequence domain (LC domain) of an

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

Quantitative Control of Kinase Activity

155

RNA-binding protein, which results in clustering of Cry2 in a liquid-like droplet.27 When using Cry2 for the induction of a light-dependent hetero-complex, efficient formation of the Cry2–Cib1 interaction is required. Reportedly, the N-terminal region of Cry2 is important for Cry2 association with Cib1.28 Therefore, in the PA-Akt system, Akt was fused to the C-terminus of Cry2 to minimize the interference of the fusion protein in Cry2–Cib1 interaction. In addition, anchoring Cib1 protein to the cellular compartment and Cry2 diffusion in the cytoplasm is expected to enhance the translocation efficiency. This enhanced efficiency occurs because Cry2 tends to form a homodimer or an oligomer, which enhances the stoichiometric probability of association between Cry2 and Cib1.29 The native Cib1 carries the nuclear localization signal (NLS), which hampers plasma membrane targeting of Cib1 fusion protein. Consequently, a point mutation to original NLS in Cib1 was introduced for improved membrane targeting. As the membrane targeting signal for the PA-Akt system, we selected a myristoylation and palmitoylation (myr) signal originating from the Fyn kinase (MGCVQCKDKEATKLTE). However, optogenetic activation of Akt by other-  group targeted Akt to the plasma membrane by a prenylation motif, CaaX box,22 indicated that the selection of the plasma membrane targeting signal for the Akt activation is flexible. The efficiency of the light dimerization system using Cry2 and Cib1 was compared using a truncated version of the system, the photolyase homology region of Cry2 (Cry2PHR; amino acids 1–498) and N-terminal fragment of Cib1 (Cibn), which lacks the conserved basic helix–loop–helix (bHLH) domain (amino acids 1–170).30 Generally, reducing the length of DNA of a probe by truncation has benefits for the effective transduction of external DNA to living cells, especially when using viruses that have a limitation in the DNA size for producing viruses with an effective titer. Considering all these points, the final construction of the PA-Akt system consists of Cry2PHR and Cibn with mutated NLS as a photoinduced dimerization system. Hereafter, we refer to Cry2–Akt as Cry2PHR fused with Akt and myr–Cib1 as NLS-mutated Cibn fused with the myristoyl sequence from the Fyn kinase. 8.2.3. Notes on Light Illumination Wavelength and Strength For activation of an optogenetic tool, the choice of the incident light source is important for the acquisition of intended data. For spatial analysis using an optogenetic tool, high spatial resolution is necessary for illumination. For that reason, a laser light source or a simulated illumination pattern using a digital micromirror device (DMD) is helpful for spatial control of the photoactivation.31 For temporal analysis using an optogenetic tool, especially in a biochemical assay, uniform illumination over the samples is necessary. Consequently, selection of the light source depending on the experiment design is an important topic that must be addressed in optogenetic experiments. Here, we discuss the light illumination used for opto­ genetic experiments.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

156

Genki Kawamura and Takeaki Ozawa

In imaging analysis, multiple wavelengths of excitation light are usually used to observe multiple biological phenomena. In the case of an optogenetic system using a blue light-absorbing photoreceptor, yellow to infrared fluorescence is compatible for observations. However, since the absorption spectrum of the Cry2 photoreceptor is broad, a careful experimental design is necessary for the assays. It is noteworthy that observation using a high laser power might unintentionally activate an optogenetic system, especially when illuminated for a longer period. To avoid such interference, the observation conditions must be checked before imaging analysis. The other commonly available incident light source for optogenetic activation is blue light-emitting diodes (LEDs). For biochemical assays requiring uniform photoactivation for treating cells with the same illumination pattern, a trans-illuminator panel (i.e. 470 nm LEDB-SBOXH; Optocode, Tokyo, Japan) is useful (Figure 8.3A). The LED panel allows for reproducible activation of an optogenetic system because of uniform light illumination intensity over the panel. Another type of light illumination source is LED arrays, which allow the application of different light pulse stimulation with various light illumination strengths.32 These are commercially available. They can also be constructed manually with a combination of a microcontroller (e.g. Arduino UNO; Arduino) and blue light-emitting LEDs, as shown

Figure 8.3.  Blue light-emitting LED illuminators. (A) Trans-illuminator panel that allows uniform light stimulation for multiple dishes. (B) Microcontroller-programmable LED illuminator designed for illuminating cells seeded on multi-well plates. A blue light-emitting pattern of LED placed underneath the wells is programmable using an Arduino microcontroller.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

Quantitative Control of Kinase Activity

157

in Figure 8.3B. Custom-built LED arrays feature benefits for fine tuning of the illuminator design at a low cost. Simple programming is necessary for controlling the duration, frequency and intensity of the LED light illumination applied to each sample. It is worth noting that the illumination used in the photoactivation of an optogenetic probe in time-lapse imaging assays and that used in biochemical assays are typically different because of limitations in the observation system setup. Although both light sources are capable of activating the optogenetic probe, careful interpretation of the data is generally necessary when directly comparing results derived from different light sources that have different peak emission wavelengths and intensities: the photoactivation kinetics of the photoreceptor might differ among stimulations. Regarding this point, quantitative comparison of the results obtained from different light sources might be avoided, especially when discussing the kinetics of the photoactivatable system. Another feature in an optogenetic experiment is the effect of light stimulation on the cell sample conditions. High-intensity blue light illumination damages the cellular condition, presumably because of blue light-mediated phototoxicity.33 Therefore, we suggest trying the lowest limit of light intensity when activating an optogenetic system. Furthermore, because Cry2– Cib1 dissociation has a half-time on the order of minutes, continuous light illumination would not be necessary for inducing sustained interactions. Consequently, a pulsed light illumination protocol is recommended to minimize the effects of light illumination to cell samples.

8.3. Application of the PA-Akt System for Cellular Signaling Analysis One challenge in signaling pathway analysis is to address the contribution of each pathway to the global outputs.34,35 In addition to evaluation of the influence of the node protein on its responsible signaling output, a method that specifically activates the pathway is required. In the case of the Akt signaling pathway, Akt is well known to regulate divergent outputs. However, the mechanism of Akt selectivity is yet to be fully characterized. The concept of “temporal coding,” by which a temporal activation pattern of the effector molecule determines the downstream activation modes, has succeeded in clarifying the mechanistics of the selectivity of the signaling pathway sharing the same channels.36,37 Temporal coding in the cellular system is observed in many signaling pathways, including p53, NF-κB and Ras–Raf–MEK–Erk.38–41 In addition, in the case of Akt signaling, selective control of the Akt downstream outputs is regulated by the temporal action of insulin, which is transmitted through the temporal phosphorylation pattern of Akt.13,14,42,43 Optogenetic manipulation of the temporal pattern of Akt is expected to contribute to the understanding of the influence of the Akt selectivity on the downstream effects because selective activation of Akt

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

158

Genki Kawamura and Takeaki Ozawa

eliminates effects from other signaling pathways. In this section, we introduce the PA-Akt system as a powerful tool for analyzing the Akt signaling pathway, presenting some examples of experimentally obtained results for optogenetic activation. 8.3.1. Optogenetic Control of the PA-Akt System This section describes evaluation of the functionality of the PA-Akt system in living cells. The PA-Akt system regulates Cry2–Akt activity by promoting translocation of Cry2–Akt to the plasma membrane, thereby causing Cry2– Akt to be phosphorylated by the upstream kinases. Therefore, one effective means of evaluating the functionality of the PA-Akt system in living cells is to observe translocation of Cry2–Akt to a plasma membrane upon illumination with blue light. Observation of Cry2–Akt localization by tagging a fluorescent protein is beneficial for the visualization of Cry2–Akt location at high spatial and temporal resolution.44 Figure 8.4 depicts fluorescence images of myr–Cib1 labeled with ECFP (myr–ECFP–Cib1) and Cry2–Akt

Figure 8.4.  Translocation of Cry2–Akt upon illumination with blue light. Images of C2C12 cells expressing Cry2–Venus–Akt and myr–ECFP–Cib1 before and after illumination are shown. The temporal localization change of Cry2–Venus–Akt upon illumination with blue light was quantified by measuring the cytosol fluorescence intensity during the observation. Each line corresponds to the intensity profile of a single cell. Dotted lines represent times when cells were illuminated. An acute decrease in the cytosol intensity after illumination is indicative of light-induced membrane translocation of Cry2–Akt. Scale bar: 10 µm.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

Quantitative Control of Kinase Activity

159

labeled with yellow fluorescent protein Venus (Cry2–Venus–Akt). For evaluation of the time course of the PA-Akt activation, changes in the fluorescence intensity in the cytosol region are used, based on the assumption that the fluorescence intensity in the cytosol region is more robust against noise or fluctuations in the fluorescence signal. Briefly, raw images of Cry2–Akt were processed using background subtraction, which then takes a threshold to produce binary images of the fluorescent cells. Subsequently, the binary images were eroded to construct mask images for the cytosol region. After image calculation using processed images and the masks to obtain images of a cytosol, temporal changes in the fluorescence intensity were quantified. From imaging analysis, the estimated half-life of Cry2–Akt fluorescence recovery in cytosol after illumination with light can be deduced. Moreover, it can be applied further to validate the kinetic parameters in the mathematical modeling, which will be discussed in a later section. 8.3.2. Dissecting the Signaling Pathway by the PA-Akt System Akt is known to be fully activated upon phosphorylation of Thr308 and Ser473 residues.45 Of the two sites, Thr308 is known to be phosphorylated by PDK1 in a phosphatidylinositol 3,4,5-trisphosphate (PIP3)-dependent manner upon phosphoinositide 3-kinase (PI3K) activation.46 When the effects of optogenetic activation of Akt and insulin stimulation are compared, light-mediated selective phosphorylation of Thr308 in Cry2–Akt is induced by the PA-Akt system (Figure 8.5A). Photoinduced activation of Akt downstream signaling by the PA-Akt system was demonstrated by measuring the phosphorylation level of typical Akt substrate molecules, including Gsk3 and FoxO1. Increased phosphorylation levels of both Gsk3 and FoxO1 upon light stimulation indicated enhanced kinase activity of Akt (Figure 8.5B). The specificity of the PA-Akt system to the Akt signaling pathway activation was demonstrated by the quantification of the phosphorylation level of Erk, which is a downstream kinase of the Ras–Raf–MEK–Erk pathway that insulin and growth factor stimulation activates, but is not a direct substrate of Akt. In contrast to the increased phosphorylation of Akt signaling pathway molecules, the phosphorylation level of Erk did not change upon light stimulation. The reduced phosphorylation of Erk is consistent with an earlier report that Akt activation lowers Erk phosphorylation.47 These results demonstrate that the PA-Akt system can activate the Akt signaling pathway selectively. 8.3.3. Spatial Regulation of Actin Remodeling by Localized Activation of PA-Akt Spatially controlled signaling is another fundamental aspect of cellular signaling.48 Because of the character of the optogenetic tools to be activated by light, they allow control of the signaling effector at high spatial

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

160

Genki Kawamura and Takeaki Ozawa

Figure 8.5.  Quantification of phosphorylation of Akt signaling molecules after illumination with blue light. (A) Selective activation of Cry2–Akt after illumination. Representative blot images of phosphorylated Akt (Thr308) after light or insulin stimulation. Cry2–Akt bands show phosphorylation of the probe and Akt bands show phosphorylation of endogenous Akt. (B) The PA-Akt system is capable of inducing selective activation of the Akt signaling pathway. Gsk3 and FoxO1 are representative substrates of Akt. Erk is a protein that is activated by insulin stimulation, but not a direct substrate of Akt. An increase in the band intensity of phosphorylated Gsk3 and FoxO1 shows light-induced activation of the Akt signaling  pathway.

resolution.49 This section describes the subcellular activation of the PA-Akt system and the localized effect of Akt signaling activation on actin remodeling. One dynamic event in a living cell is migration, which is regulated by remodeling of the actin. Actin remodeling is controlled by the PI3K–Akt signaling pathway; Akt is known to be a main regulator of the remodeling.50 Figure 8.6 presents an example of the spatial regulation of actin remodeling by the PA-Akt system. Localized activation of Akt by the PA-Akt system is achieved by illumination with blue light of a limited area within a cell. Plasma membrane translocation of Cry2–Akt is expected to occur only in the area around the illuminated region (Figure 8.6A), which indicates spatially restricted activation of Akt in the illuminated area. The effects of spatial control of the Akt signaling pathway on actin remodeling can be visualized using an F-actin indicator, Lifeact,51 as shown in Figure 8.6B. Fluorescent spots of Lifeact upon subcellular activation of Akt suggest actin filament remodeling. These results demonstrate that the PA-Akt system can serve as a tool for the analysis of dynamic events downstream

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

Quantitative Control of Kinase Activity

161

Figure 8.6.  Subcellular activation of the Akt signaling pathway by the PA-Akt system. (A) Image of a C2C12 cell expressing the PA-Akt system, illuminated with blue light in a restricted region, indicated as a blue circle. Positional fluorescence intensities at the cross-section of the non-illuminated area (A–B) and illuminated area (A'–B') before and after illumination are shown. Scale bar: 10 µm. (B) Subcellular control of the actin remodeling by the PA-Akt system. A merged image of a C2C12 cell expressing Cry2–Akt (green) and F-actin indicator shows Lifeact (magenta) illuminated at the circled region. Representative Lifeact signal around the illuminated area is enlarged with arrows indicating F-actin-accumulated spots. Scale bar: 10 µm.

of Akt at subcellular resolution by spatially controlling the area of illumination with blue light. 8.3.4. Prediction of Light-induced Akt Activation Using a Mathematical Model Mathematical modeling in cellular signaling research is undertaken to characterize biomolecular transformations in terms of equations using prior knowledge related to the biomolecules.52 Using appropriate mathematical models, signaling molecules can be represented in a realistic manner. That knowledge is then useful for elucidating the functions of signaling pathway components or for predicting biomolecule behavior.34 For example, mathematical modeling has successfully identified signaling pathways and dynamics with non-linear behaviors including Ras–Raf–MEK–Erk, p38, p53, NF-κB and Akt pathways.40,41,53–55 Consequently, mathematical modeling serves as a powerful tool to investigate the principles of biological systems. The input stimulation of optogenetic tools can be delivered in a high

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

162

Genki Kawamura and Takeaki Ozawa

Figure 8.7.  Schematics of mathematical modeling in the prediction of the output of optogenetic activation. (A) Procedure of mathematical modeling to predict the activity output as a function of light input: F(Light(t)). This procedure comprises three steps. First, the model is constructed and parameters are estimated based on the law of mass action. Second, simulation of the outputs using the model and light inputs: Light(t). Third, a model is validated by comparing simulated results and experimentally obtained results. Usually multiple rounds of this cycle are necessary to construct a good predictive model. (B) Predictive model as a function generator of the output. Once the model has been built, simulations can be conducted to generate an activation pattern of an interest. These patterns can be applied further to biological assays to analyze the signaling pathway.

temporal manner with a controllable strength of the light illumination. Therefore, mathematical modeling has a strong potential to contribute to optogenetic techniques.56–59 For instance, using a mathematical model, an optogenetic feedback controller using the Phy–PIF system for regulating the signaling pathways has been developed.60 This system allows constant activation of the cellular system because of precise control of the signaling by light stimulation. Mathematical modeling consists of three steps: (1) construction of the model and estimation of parameters using experimentally obtained results; (2) simulation of the outputs using the model; and (3) validation of the model by comparing simulated results and experimentally obtained “real” results (Figure 8.7A). Usually, multiple rounds of this cycle are necessary to construct a good predictive model. In the PA-Akt system, the mathematical modeling approach succeeded in the quantitative prediction of Akt activation as a function of light input by these steps.23 By modeling the activation pattern of Cry2–Akt as a function of light, numerous patterns of Cry2–Akt activation can be generated, which are then further applied to the biological assays to analyze the temporal codes of the Akt signaling (Figure 8.7B). In this section, we follow the mathematical model

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

Quantitative Control of Kinase Activity

163

Figure 8.8.  Schematics of the model for predicting Akt activity as an output. (A) Schematics of the activation mechanism for the PA-Akt system. Translocation of phosphorylated and dephosphorylated Cry2–Akt to the plasma membrane is triggered by illumination with light. Phosphorylation occurs at the plasma membrane, whereas dephosphorylation occurs in the cytosol. (B) Parameterized activation mechanism of the PA-Akt system: c_pAkt, phosphorylated cytosolic Cry2–Akt; m_pAkt, phosphorylated membrane-bound Cry2–Akt; c_Akt, non-phosphorylated cytosolic Cry2–Akt; m_Akt, non-phosphorylated membrane-bound Cry2–Akt. k1, k2, k3, k4, kinetic parameters; k5, constraints for total amount of Akt, which is assumed to be constant. Light takes a value of 0 or 1 corresponding to the ON time or OFF time of the illumination.

construction for predicting the Akt activity output by the PA-Akt system to demonstrate the power of the modeling in combination with an optogenetic tool. For the construction of the mathematical model for predicting light-induced Cry2–Akt phosphorylation, four states of Cry2–Akt are considered: phosphorylated cytosolic Cry2–Akt (c_pAkt), phosphorylated membrane-bound Cry2–Akt (m_pAkt), non-phosphorylated cytosolic Cry2–Akt (c_Akt) and non-phosphorylated membrane-bound Cry2–Akt (m_Akt) (Figure 8.8A). This assumption is based on the inference that Cry2–Akt in different cellular compartments has different functionality states and that Cry2–Akt translocation occurs only between the plasma membrane and the cytosol. This model also includes three assumptions: (1) both phosphorylated and dephosphorylated cytosolic Cry2–Akt translocate to the plasma membrane upon illumination with light (c_Akt → m_Akt and c_pAkt → m_pAkt); (2) both phosphorylated and dephosphorylated membrane-bound Cry2–Akt dissociate from the plasma membrane (m_Akt → c_Akt and m_pAkt → c_pAkt) with the same kinetic parameters; and (3) phosphorylation of Akt occurs only at the plasma membrane (m_Akt → m_pAkt), whereas dephosphorylation occurs only in the cytosol (c_pAkt → c_Akt). Based on these settings, a light-dependent Akt activation model was constructed based on the law of mass action, with

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

164

Genki Kawamura and Takeaki Ozawa

constrains that total amount of Cry2–Akt does not change over time. The parameter for light can be either 0 or 1, corresponding to the ON time or OFF time of the illumination (Figure 8.8B). Ordinary differential equations (ODE) for the model are presented as follows:    



d  c_pAkt  dt

  k1  c_pAkt  Light  k3  c_pAkt  k4  m_pAkt (8.1)

   



d  m_pAkt  dt

 k2  m_Akt  Light  k3  c_pAkt  k4  m_pAkt (8.2)

   



d  c_Akt   k1  c_pAkt  Light  k3  c_Akt  k4  m_Akt dt

(8.3)

d  m_Akt   k2  m_Akt  Light  k3  c_Akt  k4  m_Akt dt

(8.4)

   

   

Parameters in the model are estimated using experimental data for phosphorylated Cry2–Akt. The phosphorylated band intensities of Cry2–Akt are interpreted as the total amount of phosphorylated Cry2–Akt, which is a summation of “m_pAkt” and “c_pAkt” with respect to the number of light pulses applied to the sample. These data were subsequently used to estimate the parameters with two methods in series. First, a meta-evolutionary programming method was used to approach the neighborhood of the local minimum. In evolutionary programming, a group of parameters are mutated randomly as generation proceeds. Parameters with better estimates are selected for each generation. The estimation for the values was assessed by the objective function value (OFV), which is defined as the residual sum of the squares between the experimentally obtained results and the simulations, as in the following equation:    



 OFV

  Sim i

1

i

 Expi  2

(8.5)

   

where Simi are data from a simulation and Expi are data from the experiment. This equation shows that lower values of OFV represent a better fit to the experimental data. After approaching the neighborhood of the local minimum, the Nelder–Mead method was used to reach the local minimum.61 The initial model is sometimes not the best for predicting biological systems because the model is usually too abstract to reproduce the real

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

Quantitative Control of Kinase Activity

165

biochemical pathway. In the case of the PA-Akt system, in addition to the originally assumed model, a positive feedback factor that accelerates phosphorylation of Akt was introduced to reproduce the experimentally obtained Cry2–Akt phosphorylation pattern.23 Comparison of the fitness of the model can be evaluated using Akaike's information criterion (AIC) value, which is an estimator of the relative quality of statistical models, as shown in the following equation:62    

   

RSS  2m AIC ln     N  N

(8.6)

where N is the number of data points, m is the number of estimating parameters and RSS denotes the residual sum of squares. The AIC value includes m as the number of estimating parameters, hence a model comparison using AIC values can avoid overfitting of the model with more numerous parameters. The model with feedback has a lower AIC value than the non-feedback model, so the feedback model was selected for predicting Akt activity.23 Using these processes, the mathematical model for Cry2–Akt activation was obtained. Finally, the model was validated using a new experimental dataset to ascertain whether simulated data with different stimulation patterns can predict the Cry2–Akt phosphorylation pattern or not. The mathematical model for predicting the PA-Akt activation pattern is constructed by making a simple inference that Cry2–Akt is phosphorylated upon light-induced membrane translocation. Despite its simplicity, the model-predicted Cry2–Akt activity shows a good fit to the experimentally obtained data. Predicting the output of an optogenetic system is an important feature of an optogenetic tool because a non-linear relation might exist for the frequency or duration of the input light pulses and the output. The modeling approach is applicable to other optogenetic tools, which can enhance the ability of precise control of cellular signaling by these tools.

8.4. Conclusion Optogenetics is a system that can manipulate a specific target in time and space, which engenders investigation of the dynamic nature of cellular signaling. The PA-Akt system, an optogenetic system to activate Akt spatiotemporally, consisting of a pair of light-inducible dimers, Cry2–Akt and myr–Cib1, exhibits light-dependent selective activation of the Akt signaling pathway. In addition, the PA-Akt system manipulates Akt activity at subcellular resolution, which is applicable to the spatial signaling dynamics of the Akt signaling pathway. Furthermore, a mathematical modeling approach is used to predict the activation pattern of Cry2–Akt as a function of the light input, serving as a generator of the Akt temporal activity pattern. These features and concepts of the PA-Akt system also apply to other optogenetic tools. Future studies using optogenetics involving the

View Online

166

Genki Kawamura and Takeaki Ozawa

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

analysis of the temporal dynamics of the signaling pathway are expected to pave the way to unveiling the diverse dynamics of the outputs of their signaling pathways.

References 1. E. S. Boyden, F. Zhang, E. Bamberg, G. Nagel and K. Deisseroth, Nat. Neurosci., 2005, 8, 1263–1268. 2. P. Van Bergeijk, M. Adrian, C. C. Hoogenraad and L. C. Kapitein, Nature, 2015, 518, 111–114. 3. A. Taslimi, J. D. Vrana, D. Chen, S. Borinskaya, B. J. Mayer, M. J. Kennedy and C. L. Tucker, Nat. Commun., 2014, 5, 1–9. 4. Y. I. Wu, D. Frey, O. I. Lungu, A. Jaehrig, I. Schlichting, B. Kuhlman and K. M. Hahn, Nature, 2009, 461, 104–108. 5. G. Guntas, R. A. Hallett, S. P. Zimmerman, T. Williams, H. Yumerefendi, J. E. Bear and B. Kuhlman, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, 112–117. 6. F. Kawano, H. Suzuki, A. Furuya and M. Sato, Nat. Commun., 2015, 6, 1–8. 7. A. Levskaya, O. D. Weiner, W. A. Lim and C. A. Voigt, Nature, 2009, 461, 997–1001. 8. A. Kaberniuk, A. A. Shemetov and V. V. Verkhusha, Nat. Methods, 2016, 13, 591–597. 9. T. A. Redchuk, E. S. Omelina, K. G. Chernov and V. V. Verkhusha, Nat. Chem. Biol., 2017, 13, 633–639. 10. N. A. Repina, A. Rosenbloom, A. Mukherjee, D. V. Schaffer and R. S. Kane, Annu. Rev. Chem. Biomol. Eng., 2017, 8, 13–39. 11. J. E. Toettcher, O. D. Weiner and W. A. Lim, Cell, 2013, 155, 1422–1434. 12. B. D. Manning and A. Toker, Cell, 2017, 169, 381–405. 13. H. Kubota, R. Noguchi, Y. Toyoshima, Y. Ozaki, S. Uda, K. Watanabe, W. Ogawa and S. Kuroda, Mol. Cell, 2012, 46, 820–832. 14. R. Noguchi, H. Kubota, K. Yugi, Y. Toyoshima, Y. Komori, T. Soga and S. Kuroda, Mol. Syst. Biol., 2013, 9, 1–12. 15. V. T. Samuel and G. I. Shulman, Cell, 2012, 148, 852–871. 16. J. Y. Chen, J. R. Lin, K. A. Cimprich and T. Meyer, Mol. Cell, 2012, 45, 196–209. 17. M. Grusch, K. Schelch, R. Riedler, E. Reichhart, C. Differ, W. Berger, A. Ingles-  Prieto and H. Janovjak, EMBO J., 2014, 33, 1713–1726. 18. K. Aoki, Y. Kumagai, A. Sakurai, N. Komatsu, Y. Fujita, C. Shionyu and M.   Matsuda, Mol. Cell, 2013, 52, 529–540. 19. K. Zhang, L. Duan, Q. Ong, Z. Lin, P. M. Varman, K. Sung and B. Cui, PLoS One, 2014, 9, e92917. 20. X. X. Zhou, L. Z. Fan, P. Li, K. Shen and M. Z. Lin, Science, 2017, 355, 836–842. 21. O. Idevall-Hagren, E. J. Dickson, B. Hille, D. K. Toomre and P. De Camilli, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, E2316–E2323. 22. Y. Xu, D. Nan, J. Fan, J. S. Bogan and D. Toomre, J. Cell Sci., 2016, 129, 2085–2095. 23. Y. Katsura, H. Kubota, K. Kunida, A. Kanno, S. Kuroda and T. Ozawa, Sci. Rep., 2015, 5, 1–10. 24. A. D. Kohn, F. Takeuchi and R. A. Roth, J. Biol. Chem., 1996, 271, 21920–21926. 25. X. Yu, R. Sayegh, M. Maymon, K. Warpeha, J. Klejnot, H. Yang, J. Huang, J. Lee, L. Kaufman and C. Lin, Plant Cell, 2009, 21, 118–130.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

Quantitative Control of Kinase Activity

167

26. H. Park, N. Y. Kim, S. Lee, N. Kim, J. Kim and W. Do Heo, Nat. Commun., 2017, 8, 1–7. 27. Y. Shin, J. Berry, N. Pannucci, M. P. Haataja, J. E. Toettcher and C. P. Brangwynne, Cell, 2017, 168, 159–171.e14. 28. L. Duan, J. Hope, Q. Ong, H. Y. Lou, N. Kim, C. McCarthy, V. Acero, M. Z. Lin and B. Cui, Nat. Commun., 2017, 8, 4–13. 29. D. L. Che, L. Duan, K. Zhang and B. Cui, ACS Synth. Biol., 2015, 4, 1124–1135. 30. M. J. Kennedy, R. M. Hughes, L. A. Peteya, J. W. Schwartz, M. D. Ehlers and C. L. Tucker, Nat. Methods, 2010, 7, 973–975. 31. H. E. Johnson, Y. Goyal, N. L. Pannucci, T. Schüpbach, S. Y. Shvartsman and J. E. Toettcher, Dev. Cell, 2017, 40, 185–192. 32. K. P. Gerhardt, E. J. Olson, S. M. Castillo-Hair, L. A. Hartsough, B. P. Landry, F. Ekness, R. Yokoo, E. J. Gomez, P. Ramakrishnan, J. Suh, D. F. Savage and J. J. Tabor, Sci. Rep., 2016, 6, 1–13. 33. J. H. Stockley, K. Evans, M. Matthey, K. Volbracht, S. Agathou, J. Mukanowa,  J. Burrone and R. T. Káradóttir, Sci. Rep., 2017, 7, 1–11. 34. B. Kholodenko, M. B. Yaffe and W. Kolch, Sci. Signaling, 2012, 5, re1. 35. W. A. Lim, Nat. Rev. Mol. Cell Biol., 2010, 11, 393–403. 36. M. Behar and A. Hoffmann, Curr. Opin. Genet. Dev., 2010, 20, 684–693. 37. M. Behar, H. G. Dohlman and T. C. Elston, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 16146–16151. 38. A. Hoffmann, A. Levchenko, M. L. Scott and D. Baltimore, Science, 2002, 298, 1241–1245. 39. S. Tay, J. J. Hughey, T. K. Lee, T. Lipniacki, S. R. Quake and M. W. Covert, Nature, 2010, 466, 267–271. 40. J. E. Purvis, K. W. Karhohs, C. Mock, E. Batchelor, A. Loewer and G. Lahav, Science, 2012, 336, 13–16. 41. S. Sasagawa, Y. Ozaki, K. Fujita and S. Kuroda, Nat. Cell Biol., 2005, 7, 365–373. 42. T. Sano, K. Kawata, S. Ohno, K. Yugi, H. Kakuda, H. Kubota, S. Uda, M. Fujii, K. Kunida, D. Hoshino, A. Hatano, Y. Ito, M. Sato, Y. Suzuki and S. Kuroda, Sci. Signaling, 2016, 9, 1–11. 43. S. J. Humphrey, G. Yang, P. Yang, D. J. Fazakerley, J. Stöckli, J. Y. Yang and D. E. James, Cell Metab., 2013, 17, 1009–1020. 44. E. A. Rodriguez, R. E. Campbell, J. Y. Lin, M. Z. Lin, A. Miyawaki, A. E. Palmer, X. Shu, J. Zhang and R. Y. Tsien, Trends Biochem. Sci., 2017, 42, 111–129. 45. L. Bozulic and B. A. Hemmings, Curr. Opin. Cell Biol., 2009, 21, 256–261. 46. P. Liu, H. Cheng, T. M. Roberts and J. J. Zhao, Nat. Rev. Drug Discovery, 2009, 8, 627–644. 47. S. Zimmermann and K. Moelling, Science, 1999, 286, 1741–1744. 48. B. N. Kholodenko, J. F. Hancock and W. Kolch, Nat. Rev. Mol. Cell Biol., 2010, 11, 414–426. 49. W. K. A. Karunarathne, P. R. O'Neill and N. Gautam, J. Cell Sci., 2015, 128, 15–25. 50. Q. Yong, C. Linda, M. Qiao, B. John, Z. Z. Jenny, S. Xianglin, C. F. Daniel and J. Bing-Hua, Am. J. Physiol.: Cell Physiol., 2003, 286, 153C–163C. 51. J. Riedl, A. H. Crevenna, K. Kessenbrock, J. H. Yu, D. Neukirchen, M. Bista,  F. Bradke, D. Jenne, T. A. Holak, Z. Werb, M. Sixt and R. Wedlich-Soldner, Nat. Methods, 2008, 5, 605–607. 52. B. B. Aldridge, J. M. Burke, D. A. Lauffenburger and P. K. Sorger, Nat. Cell Biol., 2006, 8, 1195–1203.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00149

168

Genki Kawamura and Takeaki Ozawa

53. T. Tomida, M. Takekawa and H. Saito, Nat. Commun., 2015, 6, 1–9. 54. R. Cheong, A. Hoffmann and A. Levchenko, Mol. Syst. Biol., 2008, 4, 1–11. 55. K. A. Fujita, Y. Toyoshima, S. Uda, Y. Ozaki, H. Kubota and S. Kuroda, Sci. Signaling, 2010, 3, 1–11. 56. P. Hannanta-Anan and B. Y. Chow, Cell Syst., 2016, 2, 283–288. 57. L. Valon, F. Etoc, A. Remorino, F. Di Pietro, X. Morin, M. Dahan and M. Coppey, Biophys. J., 2015, 109, 1785–1797. 58. E. J. Olson, L. A. Hartsough, B. P. Landry, R. Shroff and J. J. Tabor, Nat. Methods, 2014, 11, 449–455. 59. A. Milias-Argeitis, M. Rullan, S. K. Aoki, P. Buchmann and M. Khammash, Nat. Commun., 2016, 7, 1–11. 60. J. E. Toettcher, D. Gong, W. A. Lim and O. D. Weiner, Nat. Methods, 2011, 8, 837–839. 61. J. A. Nelder and R. Mead, Comput. J., 1965, 7, 308–313. 62. H. Akaike, IEEE Trans. Autom. Control, 1974, 19, 716–723.

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00169

Chapter 9

Light Control of Transcription in Cells Akihiro Isomuraa,b and Ryoichiro Kageyama*a,b,c,d,e a

Institute for Frontier Life and Medical Sciences, Kyoto University, Kyoto, 606-8507, Japan; bJapan Science and Technology Agency, PRESTO, Saitama, 332-0012, Japan; cGraduate School of Biostudies, Kyoto University, Kyoto, 606-8501, Japan; dGraduate School of Medicine, Kyoto University, Kyoto, 606-8501, Japan; eInstitute for Integrated Cell-Material Sciences, Kyoto   University, Kyoto, 606-8501, Japan *E-mail: [email protected]

Table of Contents 9.1.  I ntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171 9.2.  Light-inducible Transcription Systems . . . . . . . . . . . . . . . . . . . . . . . 171 9.3.  Light-induced Oscillatory Versus Sustained   Expression of Ascl1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 9.4.  Light-induced Oscillatory Expression of Dll1. . . . . . . . . . . . . . . . . . 175 9.4.1.  Dll1 Oscillations in Neurogenesis and   Somitogenesis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ��  175 9.4.2.  Cell-to-Cell Transfer of Oscillatory Information   via Dll1 Oscillations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ��  177 9.5.  Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

169

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00169

View Online

         

View Online

Light Control of Transcription in Cells

171

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00169

9.1. Introduction During embryonic development, gene expression patterns are tightly controlled in space and time, which gives rise to the rich structures of multicellular organisms. As genes are activated or deactivated in a restricted spatiotemporal window, it has been hypothesized that the position and timing of gene activity encode embryonic body plans. To test this idea, various genetic tools have been developed, including tissue-specific promoters, recombinase systems and drug-inducible gene expression systems, which permit gene-specific perturbations to be introduced in developing embryos. For example, a drug-inducible Cre-recombinase driven by tissue-specific promoters is a standard tool for loss-of-function analysis in the field of developmental biology. Combinations of these technologies have allowed the artificial control of gene activity in developing tissues; however, spatiotemporal accuracy remains poor. These drawbacks hamper the dissection of highly dynamic events in embryos, such as oscillatory gene expression of transcription factors with 2–3 h periodicity. Recently, the evolution of optogenetic tools, which harness genetically encoded photosensitive proteins to control cellular functions, has been changing this situation. As light is highly tunable and its timing and power can be controlled easily, optogenetic technology is likely a promising approach to reveal the functional roles of biomolecules, especially the dynamic patterns of transcription factors in embryonic tissues. Here, we overview the recent advances in the optogenetic toolbox for gene expression control and introduce representative studies that analyzed the functional capabilities of transcription factors in cellular decision-making processes and signaling ligands in dynamic cell-to-cell communications.

9.2. Light-inducible Transcription Systems Since the first report of a light-induced gene expression system, the variety of optogenetic modules has been growing and expanding. The first artificial system for the optical control of gene transcription was reported by Shimizu-Sato et al., who developed a heterodimerizer system responsive to illumination with red light in yeast cells.1 In this system, illumination with red light reconstructs a complex of the DNA-binding protein (PhyB–GBD) and transactivation domain (PIF3–GAD), which subsequently triggers the transcription of arbitrary genes under the control of UAS (upstream activation sequence) promoters (Figure 9.1A). By contrast, illumination with farred light dissociates the complex, thereby inactivating the gene expression (Figure 9.1A). This optogenetic switch was optimized for mammalian cells;2 however, PhyB proteins require an external supply of a chromophore molecule, phycocyanobilin (PCB), for photosensitivity, which mammalian cells do not produce endogenously. To overcome this issue, Yazawa et al. reported a blue light-inducible heterodimerizer system, GIGANTEA-FKF1, which was

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00169

172

Akihiro Isomura and Ryoichiro Kageyama

Figure 9.1.  Light-inducible transcription systems. (A) The PhyB–PIF3 system; (B) the GIGANTEA–FKF1 system; (C) the LightOn system. GOI, gene of interest.

optimized for mammalian cells (Figure 9.1B).3 This system utilizes genes derived from Arabidopsis thaliana, i.e. flavin-binding, Kelch repeat, F-box 1 (FKF1) and GIGANTEA (GI) proteins, which regulate gene transcription. FKF1 contains a light-, oxygen- and voltage-sensing (LOV) domain, which is an amino acid motif that can bind to blue light-sensitive flavin molecules. As mammalian cells endogenously synthesize flavin molecules, this optogenetic system does not require the burden of additional chemicals. It was demonstrated that the GI–Gal4 and FKF1–VP16 pair can activate the UAS promoter upon illumination with blue light (Figure 9.1B).3 Although photosensitive heterodimerizer systems have been utilized successfully for the optical control of gene expression, simpler systems could be advantageous to minimize complexity in experimental design. To this end, a photoinducible homodimerizer system, LightOn, which harnesses blue light-sensitive VIVID proteins derived from Neurospora crassa, was developed (Figure 9.1C).4 VIVID is known as the smallest LOV domain-containing protein and binds to flavin adenine dinucleotide (FAD), which is produced

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00169

Light Control of Transcription in Cells

173

endogenously by mammalian cells. Illumination with blue light can trigger the homodimerization of VIVID proteins, whereas the homodimers dissociate into monomers in darkness (Figure 9.1C). Hence the photoswitching of VIVID proteins is a reversible process. The LightOn system uses a synthetic photosensitive transcriptional factor, termed GAVPO, which is a fusion protein consisting of a Gal4 DNA-binding domain, a photosensitive VIVID protein and a p65 transactivation domain. This system can induce the transcription of arbitrary genes under the control of UAS promoters upon illumination with blue light. It was found that the induction level of photoinduced gene expression is proportional to the intensity and duration of light illumination.4 It was also shown that transient illumination can induce pulsatile patterns of reporter gene expression with 6 h rising phases and 18 h falling phases.4 However, when the induced genes encode unstable/destabilized mRNAs and proteins, the LightOn system is able to induce oscillatory expression with 2–3 h periodicity.5,6 These results suggest that the LightOn system is a tunable photosensitive controller of gene expression with various dynamics, which is suitable for the analyses of many developmental processes, as discussed in this chapter.

9.3. Light-induced Oscillatory Versus Sustained Expression of Ascl1 In the developing nervous system, neural stem cells proliferate intensively, while giving rise to various types of neurons first and astrocytes later (Figure 9.2A). It has been shown that basic helix–loop–helix (bHLH) transcription factors such as Hes1 and Ascl1 regulate these processes. For example, Ascl1, a proneural gene, induces cell cycle exit and neuronal fate determination, and Hes1 regulates astrocyte differentiation.7,8 Interestingly, these factors also play an important role in the maintenance and proliferation of neural stem cells,9,10 indicating that these bHLH factors have opposing functions. Time-lapse imaging analyses revealed that the expression of these factors oscillates in neural stem cells (Figure 9.2B).5,11 Hes1, a transcriptional repressor, represses its own expression by binding directly to its own promoter (negative feedback) (Figure 9.2C).12 Upon activation of Notch signaling, the transmembrane protein Notch is processed, releasing the Notch intracellular domain (NICD), which then upregulates Hes1 expression. However, upregulated Hes1 represses its own expression by negative feedback, while this downregulation relieves negative feedback, allowing the next round of Hes1 expression to begin. In this way, Hes1 expression oscillates autonomously with a period of approximately 2–3 h in neural stem cells (Figure 9.2C). Hes1 also represses the expression of the proneural gene Ascl1 by binding directly to its promoter. Hes1 oscillations periodically repress Ascl1 expression, leading to Ascl1 oscillations with a period of approximately 2–3 h in neural stem cells (Figure 9.2B).5 By contrast, when neural stem cells differentiate into

View Online

Akihiro Isomura and Ryoichiro Kageyama

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00169

174

Figure 9.2.  The expression dynamics of Hes1 and Ascl1 in the developing nervous system. (A) In the developing nervous system, neural stem cells proliferate intensively, while giving rise to various types of neurons first and astrocytes later. (B) In neural stem cells, Hes1 oscillations induce Ascl1 oscillations by periodic repression. When neural stem cells differentiate into neurons, Hes1 expression disappears, whereas Ascl1 is expressed in a sustained manner. Similarly, when neural stem cells differentiate into astrocytes, Ascl1 expression disappears, whereas Hes1 expression becomes sustained. (C) Hes1 expression oscillates autonomously by negative feedback.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00169

Light Control of Transcription in Cells

175

neurons, Hes1 expression disappears, whereas Ascl1 is expressed in a sustained manner (Figure 9.2B).6 Similarly, when neural stem cells differentiate into astrocytes, Ascl1 expression disappears, whereas Hes1 is expressed in a sustained manner (Figure 9.2B).5 These results indicate that the expression dynamics of bHLH factors are different between neural stem cells (oscillatory expression) and differentiating cells (sustained expression). To examine the functional significance of oscillatory versus sustained expression dynamics, we undertook an optogenetic approach with the LightOn system using hGAVPO (in which the codon usage of GAVPO is humanized). The LightOn–hGAVPO system, which induces Ascl1 expression upon illumination with blue light, was introduced into Ascl1-null neural stem cells.5 Exposure to blue light at 3 h intervals induced the oscillatory expression of Ascl1 with a 3 h period, whereas that with 30 min intervals induced the sustained expression of Ascl1. In the dark condition, these neural stem cells were negative for Ascl1 expression and Ascl1-negative neural stem cells proliferated very slowly. By contrast, light-induced Ascl1 oscillations activated the proliferation of neural stem cells, whereas the light-induced sustained expression of Ascl1 induced cell cycle exit and neuronal differentiation (Figure 9.3). At least 6–8 h of sustained Ascl1 expression was required for the activation of neuronal differentiation.5 These results showed that expression dynamics are very important for the activity of Ascl1. It was a puzzle how Ascl1 is able to activate the proliferation of neural stem cells and also induce cell cycle exit and neuronal differentiation,9 and the precise molecular mechanism for such opposing functions was unknown. However, our optogenetic study clearly revealed that Ascl1 exerts such opposing functions depending on its expression dynamics: oscillatory versus sustained Ascl1 expression regulates stem cell versus neuronal fates, respectively, although the detailed mechanism of how these different dynamics lead to differential gene regulation remains to be analyzed.

9.4. Light-induced Oscillatory Expression of Dll1 9.4.1. Dll1 Oscillations in Neurogenesis and Somitogenesis As described above, the Notch signaling pathway is important for the maintenance of proliferating neural stem cells. This pathway induces Hes1 oscillations, which drive Ascl1 oscillations, thereby maintaining neural stem cells in a proliferative state. Interestingly, these oscillations also lead to the oscillatory expression of Delta-like1 (Dll1), a ligand for Notch signaling, which is controlled positively by Ascl1 and negatively by Hes1.11,13 By contrast, in differentiating neurons, which are negative for Hes1 and express Ascl1 in a sustained manner, Dll1 expression is also sustained.11 It has been suggested that the different expression dynamics of Dll1 may result in different outcomes. During neurogenesis, differentiating neurons express Dll1 in a sustained manner and activate Notch signaling in neighboring cells, thereby inhibiting these cells from differentiating into neurons, a process called lateral

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00169

176

Akihiro Isomura and Ryoichiro Kageyama

Figure 9.3.  The significance of expression dynamics of Ascl1. The optogenetic induction of Ascl1 oscillations activated the proliferation of neural stem cells. By contrast, the optogenetic induction of sustained Ascl1 expression induced cell cycle exit and neuronal differentiation. Hence the expression dynamics are very important for the Ascl1 activity. NSC, neural stem cell.

inhibition. In this case, differentiating neurons maintain neural stem cells via the Dll1–Notch pathway. However, during the early stages before neurogenesis starts, there are no Dll1-expressing neurons and neural stem cells themselves express Dll1. In these cells, Hes1 and Ascl1 oscillations drive Dll1 oscillations, which lead to the mutual activation of Notch signaling. Indeed, by using the LightOn–hGAVPO system, it was shown that the optogenetic induction of sustained Dll1 expression made more cells prone to differentiate into neurons, whereas the optogenetic induction of oscillatory Dll1 expression made more cells prone to remain as neural stem cells.13 These results indicate that Dll1 expression dynamics are important for the activity of neural stem cells. According to mathematical modeling,13 the time required for Dll1–Notch signaling transmission between neighboring cells is very important for oscillatory expression dynamics, and if this time becomes longer or shorter than is appropriate, the oscillations would be damped or quenched. This mathematical prediction was tested by genetic modifications of the Dll1 locus.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00169

Light Control of Transcription in Cells

177

The timing of Dll1 expression was accelerated or slowed by decreasing or increasing the size of the Dll1 gene, respectively. It was found that when Dll1 expression was accelerated or slowed, Dll1 oscillations in neural stem cells were severely damped.13 Furthermore, Hes1 oscillations were also severely damped in both cases.13 As a result, neural stem cell proliferation was severely impaired, resulting in microcephaly.13 Therefore, the time required for Dll1– Notch signaling transmission between neighboring cells is very important for oscillatory expression and neurogenesis. During somitogenesis, the presomitic mesoderm (PSM) periodically generates a bilateral pair of somites, which later differentiate into vertebrae, ribs, skeletal muscles and subcutaneous tissues. Many factors, including Dll1 and the Notch effector Hes7, are expressed in an oscillatory manner in the PSM, similarly to Hes1 oscillations in neural stem cells, and regulate the periodic formation of somites.14,15 When Dll1 expression was accelerated or slowed, Dll1 oscillations in the PSM were severely damped.13 Furthermore, Hes7 oscillations were also severely damped in both cases.13 As a result, the somites were fused and the somite-derived tissues such as the vertebrae and ribs were also fused.13 Hence the time required for Dll1–Notch signaling transmission between neighboring cells is very important for the periodic formation of somites. These results together indicate that oscillatory expression dynamics are important for tissue morphogenesis, such as neurogenesis and somitogenesis. 9.4.2. Cell-to-Cell Transfer of Oscillatory Information via Dll1 Oscillations During neurogenesis, oscillations seem to be anti-phase or out-of-phase between neighboring neural stem cells, whereas during somitogenesis, oscillations are in-phase between neighboring PSM cells. When Notch signaling is inhibited, such anti-phase and in-phase oscillations are inhibited. These results suggest that Dll1 oscillations may be involved in oscillatory expression dynamics and that the transmission of oscillatory information between neighboring cells is key to Dll1 oscillations. However, methods to generate oscillatory gene expression at multiple nodes and monitor the responses in real time at single-cell resolution were not available. Recently, we established a new method combining optogenetic perturbation and bioluminescence imaging, which solved the above problem and offers a very powerful approach to control and monitor the dynamic cellto-cell transfer of oscillatory information.6 With this new method, we successfully showed that optogenetic induction (using the LightOn–hGAVPO system) of Dll1 oscillations in sender cells entrained Hes1 oscillations (monitored using a destabilized luciferase reporter under the control of the Hes1 promoter) in neighboring cells via the periodic activation of Notch signaling (Figure 9.4A). Furthermore, the optogenetic induction of the oscillatory expression of NICD, an active form of Notch, was also able to entrain Hes1 oscillations (Figure 9.4A).6 However, in this case, the peaks of Hes1 oscillations appeared at earlier timings, compared with Dll1, suggesting that this

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00169

178

Akihiro Isomura and Ryoichiro Kageyama

Figure 9.4.  Schematic of the node-specific perturbations in the Delta–Notch pathway. (A) Dll1 triggers the cleavage of Notch receptors in neighboring cells, giving rise to the Notch intracellular domain (NICD), which induces Hes1 oscillations. Node-specific optogenetic perturbation directly controls either Dll1 or NICD production. (B) By Dll1 and NICD perturbations, Hes1 oscillations were entrained, but the peaks of Hes1 oscillations appeared at different timings, suggesting that this difference reflects the time required for Dll1 to induce NICD formation in neighboring cells.

difference reflects the time required for Dll1 to induce NICD formation in neighboring cells (Figure 9.4B). These results indicate that Dll1 oscillations are able to convey oscillatory information to neighboring cells via the periodic formation of NICD, thereby coordinating gene expression patterns at the population level.

9.5. Conclusion The LightOn–hGAVPO system is very powerful when the precise spatiotemporal control of gene expression is required. It has been shown that the activity of many genes is oscillatory and that not only the amplitude but also the frequency and phase of their expression convey information for cellular activity.16–18 Furthermore, such oscillatory information seems to be transmitted from cell to cell to coordinate gene activity at the population level. However, the dynamics and precise mechanisms for such cell–cell communications are mostly unknown for oscillatory factors. The LightOn–hGAVPO system can be applied to many such factors without any obstacles and will be useful in deciphering the detailed programs underlying the transfer of intracellular and intercellular information.

View Online

Light Control of Transcription in Cells

179

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00169

References 1. S. Shimizu-Sato, E. Huq, J. M. Tepperman and P. H. Quail, Nat. Biotechnol., 2002, 20, 1041. 2. A. Levskaya, O. D. Weiner, W. A. Lim and C. A. Voigt, Nature, 2009, 209(461), 997. 3. M. Yazawa, A. M. Sadaghiani, B. Hsueh and R. E. Dolmetsch, Nat. Biotechnol., 2009, 27, 941. 4. X. Wang, X. Chen and Y. Yang, Nat. Methods, 2012, 9, 266. 5. I. Imayoshi, A. Isomura, Y. Harima, K. Kawaguchi, H. Kori, H. Miyachi, T. Fujiwara, F. Ishidate and R. Kageyama, Science, 2013, 342(6163), 1203. 6. A. Isomura, F. Ogushi, H. Kori and R. Kageyama, Genes Dev., 2017, 31(5), 524. 7. F. Guillemot, L. C. Lo, J. E. Johnson, A. Auerbach, D. J. Anderson and A. L. Joyner, Cell, 1993, 75(3), 463. 8. T. Ohtsuka, M. Sakamoto, F. Guillemot and R. Kageyama, J. Biol. Chem., 2001, 276(25), 30467. 9. D. S. Castro, B. Martynoga, C. Parras, V. Ramesh, E. Pacary, C. Johnston, D. Drechsel, M. Lebel-Potter, L. G. Garcia, C. Hunt, D. Dolle, A. Bithell, L. Ettwiller, N. Buckley and F. Guillemot, Genes Dev., 2011, 69(6), 1069. 10. M. Ishibashi, S. L. Ang, K. Shiota, S. Nakanishi, R. Kageyama and F. Guillemot, Genes Dev., 1995, 9(24), 3136. 11. H. Shimojo, T. Ohtsuka and R. Kageyama, Neuron, 2008, 58(1), 52. 12. H. Hirata, S. Yoshiura, T. Ohtsuka, Y. Bessho, T. Harada, K. Yoshikawa and R. Kageyama, Science, 2002, 298(5594), 840. 13. H. Shimojo, A. Isomura, T. Ohtsuka, H. Kori, H. Miyachi and R. Kageyama, Genes Dev., 2016, 30(1), 102. 14. O. Pourquié, Cell, 2011, 145, 650. 15. R. Kageyama, Y. Niwa, A. Isomura, A. González and Y. Harima, WIREs Dev. Biol., 2012, 1, 629. 16. J. H. Levine, Y. Lin and M. B. Elowitz, Science, 2013, 342, 1193. 17. J. E. Purvis and G. Lahav, Cell, 2013, 152, 945. 18. A. Isomura and R. Kageyama, Development, 2014, 141, 3627.

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

Chapter 10

Building Light-inducible Receptor Tyrosine Kinases Nury Kima, Hyerim Parkb, Doyeon Wooa and Won Do Heo*a,b a

Center for Cognition and Sociality, Institute for Basic Science, Daejeon, Republic of Korea; bDepartment of Life Sciences, Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea *E-mail: [email protected]

Table of Contents 10.1.  I ntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2.  Experimental. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2.1.  Materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2.2.  Sample Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2.3.  Microscope Setup. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2.4.  Image Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.3.  Results and Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.3.1.  Screening OptoRTKs with PHR. . . . . . . . . . . . . . . . . . . . . 10.3.2.  Activation of Canonical Signaling Pathways. . . . . . . . . . 10.3.3.  Functional Validation of Downstream Activation. . . . . . 10.3.4.  Responses to Different Light Conditions. . . . . . . . . . . . . 10.3.5.  Application of Diverse Actuators. . . . . . . . . . . . . . . . . . . . 10.4.  Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

181



183 183 183 184 184 184 185 185 186 189 191 192 194 194 194

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

View Online

         

View Online

Building Light-inducible Receptor Tyrosine Kinases

183

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

10.1. Introduction Receptor tyrosine kinases (RTKs) are members of a superfamily of surface receptors consisting of 20 subfamilies that transmit environmental signaling cues into cells.1 Binding of signaling cues (i.e. ligands) to RTKs usually results in dimerization and leads to the activation of signals through auto-phosphorylation and phosphorylation of downstream substrates. This dimerization-dependent activation led to the idea of synthetic regulation of RTKs using chemical dimerizers. However, recent developments in optogenetic actuators, especially light-responsive, homo-interacting proteins such as the PHR domain of cryptochrome 2 (CRY2) from Arabidopsis thaliana (PHR hereafter) and light-, oxygen- and voltage-sensing (LOV) domains, have established light as the inducer and the activator of signaling molecules.2 The era of optogenetics started with the introduction of channelrhodopsins and their variants, which are capable of modulating membrane potential following stimulation with light. Initially, these optogenetic tools were mainly used by neuroscientists to activate or suppress neuronal activity. However, with the development of photosensory proteins, especially those with light-dependent homo- and hetero-interaction properties, the use of light as an actuator has become more generalized.3 One of the best-studied of such models is PHR, which can be used alone (homo-interaction) or paired with its partner, the N-terminus of CIB1 (CIBN) (hetero-interaction).4,5 A number of research groups have developed innovative and diverse strategies to exploit this optogenetic module for the regulation of biological behaviors such as gene expression, activation and sequestration-based inhibition of protein activity through the introduction of new protein–protein interactions; additional lines of research have also led to an increased understanding of the tool itself.6 The focus of this chapter is on explaining how PHR can be used as an actuator of homo-interactions to activate RTK signaling pathways. This approach is exemplified by fibroblast growth factor receptor-1 (optoFGFR1) and TrkB (optoTrkB) that have been remodeled so that their activity can be regulated by light, that is, optogenetically.7,8 We illustrate the light-responsive properties of these optogenetic RTKs (optoRTKs) by characterizing the activities of downstream signaling pathways in response to stimulation by light with different wavelengths and intensity. We also review recent advances in PHR by protein engineering and changes in the activation efficiency of optoRTKs with the evolved PHRs.

10.2. Experimental 10.2.1. Materials The coding sequences of RTKs were cloned from a cDNA library (Dharmacon). OptoFGFR1, Lyn-R-GECO1, optoTrkB, mCherry-CRY2clust9 and mCherry-CRY2olig5 plasmids were used. R-GECO1 was a gift from Robert Campbell (#32444, Addgene).10 dTomato-PHAkt1, composed of the PH

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

184

Nury Kim, Hyerim Park, Doyeon Woo and Won Do Heo

domain of mouse Akt1 attached to the C-terminus of dTomato (amino acids 2–147, NM_009652),11 and Erk1-dTomato, composed of full-length mouse Erk1 attached to N-terminus of dTomato (NM_011952), were used to analyze intracellular signaling pathways. FusionRed-NLS, composed of the nuclear localization signal (NLS) from pDsRed2-Nuc fused to pmCherry-C1 (Clontech), and iRFP682-Lifeact (Lifeact in iRFP670-N1 12), a gift from Vladislav Verkhusha (#45457, Addgene), were used to reveal the changes in intracellular signaling and cell morphology. An optical power meter (Cat. No. 8230E, ADCMT) was used to measure the intensity of blue light. 10.2.2. Sample Preparation HeLa cells were maintained in Dulbecco's Modified Eagle's Medium containing 10% fetal bovine serum at 37 °C in a humidified 10% CO2 incubator. Human umbilical vein endothelial cells (HUVECs; C-003-5C, Gibco) were maintained in Medium 200 (M-200-500, Gibco) supplemented with 2% Low Serum Growth Supplement (S-003-10, Gibco) at 37 °C in a humidified 5% CO2 environment. Lipofectamine LTX (15338-100, Invitrogen) and the Neon transfection system (MPK5000, Invitrogen) were used for transfection of HeLa cells and HUVECs, respectively. Experimental procedures have been described in detail in previous publications.13,14 10.2.3. Microscope Setup A Nikon A1R laser scanning confocal microscope equipped with CFI Apo 60X Oil λS [numerical aperture (NA) 1.4, working distance (WD) 0.14 mm] and Nikon CFI Plan Apochromat λ 20X (NA 0.75, WD 1.00 mm) lenses and a Nikon multi-line Ar laser (488 nm/40 mW), Coherent sapphire solid laser (561 nm/20 mW) and Coherent CUBE diode laser (640 nm/40 mW) was used. Photoactivation was achieved using a galvano scanner incorporated into a hybrid confocal scan head of the A1R containing a high-speed hyper selector. The stage was equipped with a CO2 supply (gas cylinder), for live cell imaging, and a Chamlide WP system, which includes a stage-top incubator and cover, lens heater, temperature controller, gas flow rate controller for CO2 and humidifier (Live Cell Instrument). 10.2.4. Image Analysis MetaMorph (Molecular Devices), NIS (Nikon) and Microsoft Excel with Solver (Microsoft) were used for analyses. For kinetic analyses, the following four-parameter logistic curve was used:

   

   

F (t ) A 

B A   1   t / T1  C 2  

(10.1)

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

Building Light-inducible Receptor Tyrosine Kinases

185

where F is the measurement (intensity, for example) at time t and A, B, C and T½ denote the minimum asymptote, maximum asymptote, steepness of the curve and inflection point (the time at which the intensity reaches half of its final asymptotic value), respectively.7 Statistical analyses were performed using an unpaired, two-tailed Student's t-test (Microsoft Excel).

10.3. Results and Discussion 10.3.1. Screening OptoRTKs with PHR OptoTrkB was constructed as full-length TrkB with C-terminal PHR and FP tags (Figure 10.1a).8 After our serendipitous success in creating optoTrkB, we first tried to apply the same synthetic gene-reconstruction scheme to other RTKs and generalize the manner of activation. Among the many available RTKs, we selected EphB2, EGFR, ErbB2, FGFR1, IGF1R, INSR, PdgfrB, RET and TrkB as targets for preliminary tests. Unfortunately, none of these constructs showed any evidence of activation and most failed to localize to the plasma membrane (Figure 10.1i). Therefore, we varied the reconstruction scheme by introducing CIBN as a mediator or replacing the extracellular (EC) and transmembrane (TM) regions with a myristoylation (Myr) signal sequence (Figure 10.1b–h). To our surprise, only TrkB, FGFR1 and RET were efficiently activated following application of any of these strategies. All three RTKs were optogenetically functional after applying the scheme shown in Figure 10.1d, in which the cytoplasmic domain of RTKs was fused to PHR and FP at the C-terminus and the N-terminal region,

Figure 10.1.  Generalized strategies for screening optoRTKs. (a–h) Scheme for cloning optoRTKs containing PHR and CIBN. (i) Examples of expression patterns of RTKs in screens. HeLa cells expressing optoRTKs in scheme (a) are shown. Scale bars: 20 µm.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

186

Nury Kim, Hyerim Park, Doyeon Woo and Won Do Heo

including EC and TM domains, were replaced by a Myr signal sequence, whereas TrkB and FGFR1 were also functional without the Myr replacement (Figure 10.1h). Why only a handful of restricted forms of RTKs were responsive to light remains to be elucidated. In the case of the insulin receptor (INSR) and related RTKs, which are intrinsic dimers,15 a different strategy will be required to achieve light responsiveness. The diversity of RTK activation mechanisms is very likely responsible for the failure of a single, general reconstruction strategy, but with clever combinations of RTKs and optogenetic modules it should be possible to create more and more RTK variants that can be activated with light. 10.3.2. Activation of Canonical Signaling Pathways Activated RTKs relay signals through three canonical signaling cascades – RAS/ERK (extracellular signal-regulated kinase), phosphoinositide 3-kinase (PI3K)/AKT and phospholipase C (PLC)/Ca2+ pathways – to regulate cell behavior. To assess changes in the status of these pathways, we monitored the translocation of mCherry-tagged ERK into the nucleus (to assess RAS/ ERK activity), translocation of mCherry-tagged AKT1 (to assess the PI3K/AKT pathway) and changes in the intensity of R-GECO1, which indicates the level of intracellular Ca2+ ions (to assess the PLCγ pathway) (Figure 10.1a).7 The activation pattern of optoTrkB was different from that obtained by treatment with ligands. Activation by a light pulse produced a transient increase in downstream signals, whereas treatment with ligand produced a prolonged activated state (Figure 10.2b and c).8 Because fluorescence microscopes use light, the status of each signaling pathway can be examined simultaneously with excitation; simple snapshot imaging with a GFP (green fluorescent protein) channel, which uses blue light near 488 nm for activation, can be used to evoke RTK activation and the activation strength can be modulated by changing the light intensity or exposure time. When combined with laser scanning confocal microscopy, it is possible to achieve spatially restricted activation of a defined area in a defined shape. This spatial resolution is demonstrated by activation of optoTrkB or optoFGFR1 within a subcellular region of transfected cells by a scanning laser, observed by monitoring R-GECO1 fluorescence. Spatially restricted activation of optoTrkB causes Ca2+ elevations and flows within a cell; this spatial and temporal activation can be used to activate distinct cells reversibly and repeatedly at distinct times (Figure 10.2d and e).7,8 These results imply that a cell or group of cells can be selectively and asymmetrically activated to induce specialized cell behavior, for example, cell migration. Because light travels through the whole system at the speed of light, the activation switch can be turned on and off in much more efficient way than is possible with previous methods. Compared with synthetic activation using chemically induced dimerization (CID), which requires time for diffusion during activation and washout during deactivation by virtue of

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

View Online

Figure 10.2.  Activation of signaling downstream of optoRTKs. (a) Changes in three pathways downstream of FGFR1 in optoFGFR1-expressing HeLa cells after a single pulse of light illumination (488 nm, 7 µW). (b) Translocation kinetics of ERK-mCherry and mCherry-PHAKT1 after stimulation (arrow) of optoTrkB-cotransfected HeLa cells with BDNF or blue light. The graph shows quantification of translocation, calculated as the ratio of nuclear to cytosolic ERK-mCherry (ERKnuc/ERKcyt) and changes in cytosolic mCh-PHAKT1, both normalized to the baseline at 1 min (dotted lines). Colored numbers on each axis of the graph display different time scales and fold-change scales for ERK-mCherry (red) and mCh-PHAKT1 (blue). (c) Time course of changes in R-GECO1 fluorescence in HeLa cells expressing optoTrkB constructs, before and after exposure to BDNF (100 ng mL−1) or light. Cells were photostimulated with or without a 30 min pre-incubation with 100 µM K252a or 2.5 µM U73122. (d) A HeLa cell coexpressing optoTrkB and R-GECO1 was exposed to whole-cell ① or local ②, ③ light stimulation. The stimulated areas are outlined with white dashed lines. (e) HeLa cells coexpressing optoFGFR1 and R-GECO1 were repeatedly exposed to a series of light stimulations, indicated by blue bars. (f) Quantified intensities of time-lapse images showing relative cytoplasmic intensities of R-GECO1 and dTomato-PHAKT1 and relative nuclear intensity of ERK-dTomato in HeLa cells expressing optoFGFR1 (blue) or ihFGFR1 (brown) following stimulation with light (488 nm, 7 µW) or AP20187 (100 nM), respectively. Error bars: ± SEM. Scale bars: 20 µm. Parts (a), (e) and (f) reprinted from Chemistry & Biology, 21 (7), N. Kim, J. M. Kim, M. Lee, C. Y. Kim, K.-Y. Chang and W. D. Heo, Spatiotemporal control of fibroblast growth factor receptor signals by blue light, 903–912, Copyright 2014, with permission from Elsevier. Parts (b), (c) and (d) reprinted with permission from Macmillan Publishers Ltd: Nature Communications, 5, K.-Y. Chang, D. Woo, H. Jung, S. Lee, S. Kim, J. Won, T. Kyung, H. Park, N. Kim, H. W. Yang, J.-Y. Park, E. M. Hwang, D. Kim and W. D. Heo, Light-inducible receptor tyrosine kinases that regulate neurotrophin signaling, 4057, Copyright 2014.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

188

Nury Kim, Hyerim Park, Doyeon Woo and Won Do Heo

Figure 10.3.  Light-induced changes in cell behavior. (a) Polarity change induced by partial light stimulation. Repeated partial stimulation (488 nm, 1 µW, 30 s intervals) was applied to the peripheral region of a HUVEC expressing optoFGFR1. Two different sites (light-blue region, S1 and S2) were illuminated sequentially and adjacent linear regions (white dotted line) were drawn with a kymograph. The light-induced change in cell morphology is presented as a protrusion/retraction graph in the lower panel. Red, protrusion; blue, retraction; white, overlapping area; light blue, illuminated point. (b) OptoFGFR1-mediated phototaxis model. Circular blue light (488 nm, 30 µW) with a 160 µm radius (blue dotted circle) was repeatedly applied (30 s intervals) to the center of the imaging field containing sparsely plated, serum-starved (6 h) HUVECs expressing optoFGFR1 and mCherry-Lifeact. Cell movement was monitored for 20 h and compared with that of cells without light illumination (white dotted circle). (c) Changes in the directionality of HUVECs during light-induced directed migration, produced by stimulation at different locations in the cell. The average speed (µm h−1) and directionality (cos θ) was measured every 30 µm from the centroid of the imaging field and are presented from 60 to 300 µm. (d) Pseudo-colored images of Lyn-R-GECO1 (membrane-targeted Ca2+ sensor) showing puffs (arrowheads) and sparklets (arrows). Kymographs scanning the white-squared region are presented in the lower panels. (e) Frequency of Ca2+ sparklets in HUVECs treated with various inhibitors. Nimo, nimodipine; STR, streptomycin; 2-APB, 2-aminoethoxydiphenyl

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

Building Light-inducible Receptor Tyrosine Kinases

189

the fact that it is a small molecule, the use of light activation as a switch yields faster kinetics of signal regulation (Figure 10.2f).7 Using light as a regulatory stimulus also allows pulsatile activation by varying the stimulation frequency and strength, which could be used to study the properties of downstream receptor signaling, as demonstrated by seminal work on the RAS/ERK module using optoSOS, showing that this pathway acts as a low-pass filter.16 10.3.3. Functional Validation of Downstream Activation One of the cellular phenotypes induced by activation of RTK downstream signaling pathways is cell protrusion and migration. A chemoattractant gradient can create restricted activation of signaling pathways and eventually cause cells to migrate along the gradient.17 Although other methods such as microfluidics can produce a gradient of attractants that induces polarization and directed migration of cells, optoFGFR1 provides a precise and convenient tool for studying this phenomenon with high spatiotemporal resolution. Light illumination of one part of a cell expressing optoFGFR1 induces polarization and protrusions (lamellipodia) at the illuminated region of a cell edge and retraction of the other edge. The direction of polarization can be repeatedly reversed by switching the position of light illumination (Figure 10.3a).7 This polarization ultimately induces cell migration in the direction of light illumination (Figure 10.3b and c).7 Directed cell migration is a complex behavior in which multiple signaling pathways are involved.18 Through the power of optogenetics, we were able to demonstrate the involvement of a previously unrecognized pathway

borate. One outlier was excluded. (f) Representative images of optoTrkB-transfected (upper panel) and phalloidin-stained (lower panel) PC12 cells. Serum-starved (24 h) optoTrkB-transfected PC12 cells were unstimulated (no stim.) or stimulated with BDNF (50 ng mL−1) or with a 5 s ON/5 min OFF illumination protocol using a blue LED array for an additional 24 h. (g) Quantified results of neurite outgrowth from non-transfected (mock) and optoTrkB-transfected (optoTrkB) PC12 cells treated as described in (f). Neurite-presenting cells are defined as those with at least one neurite more than one cell-body diameter in length. Results are expressed as the percentage of neurite-bearing cells among the total number of counted cells. Error bars: ± SEM. Scale bars: 50 µm. Parts (a), (b) and (c) reprinted from Chemistry & Biology, 21 (7), N. Kim, J. M. Kim, M. Lee, C. Y. Kim, K.-Y. Chang and W. D. Heo, Spatiotemporal control of fibroblast growth factor receptor signals by blue light, 903–912, Copyright 2014, with permission from Elsevier. Parts (d) and (e) reprinted with permission from J. M. Kim, M. Lee, N. Kim and W. D. Heo, Proc. Natl. Acad. Sci. U. S. A., 2016, 113 (21), 5952–5957. Parts (f) and (g) reprinted with permission from Macmillan Publishers Ltd: Nature Communications, 5, K.-Y. Chang, D. Woo, H. Jung, S. Lee, S. Kim, J. Won, T. Kyung, H. Park, N. Kim, H. W. Yang, J.-Y. Park, E. M. Hwang, D. Kim and W. D. Heo, Light-inducible receptor tyrosine kinases that regulate neurotrophin signaling, 4057, Copyright 2014.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

190

Nury Kim, Hyerim Park, Doyeon Woo and Won Do Heo

Figure 10.4.  Downstream activation and cellular behavior under different light-stimulation conditions. (a, b) Relative efficiencies of optoFGFR1 activation by lasers with different wavelengths. OptoFGFR1- and R-GECO1-coexpressing HeLa cells were illuminated with light twice: first with the indicated wavelength (405–647 nm) and power (0.5–405.9 µW) and second at 488 nm and 3 µW. Images were obtained every 15 s. The initial stimulation was at 0 s (red arrow) and the second stimulation was at 180 s (blue arrow). Quantification of intensities of R-GECO1 fluorescence from time-lapse images (a). Relative activation levels, calculated by dividing the maximum intensity induced by the first stimulation by that induced by the second stimulation (b). (c, d) Relative efficiencies of optoFGFR1 activation by lasers with different wavelengths. The indicated loops of the 488 nm laser at 0.1 µW were applied to the whole field of view. Change in relative R-GECO1 intensity in responding cells (c) and relative number of cells that responded to the light within the field of view (d) were quantified. (e) Relationship between light intensity and migration speed. The migration speed of HUVECs stimulated with distinct laser intensities (5, 25 and 50 µW) was analyzed in a phototaxis model. (f, h) Quantification of the effect of illumination intensity on the presence of neurites (f), total neurite length (g) and length of the longest neurite (h). Neurite-presenting cells are defined as those with at least one neurite more than one cell-body diameter in length. Error bars: ± SEM. Parts (a) and (b) reprinted from Optogenetics, Optogenetic control of fibroblast growth factor receptor signaling, 1408,

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

Building Light-inducible Receptor Tyrosine Kinases

191

in this process. Using optoFGFR1, we were able to induce a highly localized influx of Ca2+, detected optically in the form of “sparklets.” Local activation of optoFGFR1 by illumination on one side of a cell resulted in an overall increase in Ca2+ in the activated area and very small, pinpoint increases in Ca2+ (sparklets) in the peripheral region (Figure 10.3d).19 This activation pattern is very different from the calcium “puffs,” which are much larger (about half of the whole cell area), but still localized, elementary events associated with Ca2+ release through IP3 receptors. Sparklets are involved in the generation and maintenance of intracellular Ca2+ gradients and are mediated by the nimodipine-sensitive, L-type, voltage-dependent Ca2+ channel (Figure 10.3e).19 Repeated local influx of Ca2+ in the form of sparklets can increase global Ca2+ and thereby contribute to the rearrangement of intracellular calcium gradients, and thereby controls the localization of compartments involved in cell polarity. Another typical cellular phenotype induced by activation of an RTK is cellular differentiation. Neurotrophins are well-known inducers of neurite outgrowth that are associated with sustained activation of ERK.20 Under optimal activation conditions (5 s of illumination with light at 5 min intervals, 5.5 µW of 470 nm LED), light can induce neurite outgrowth in optoTrkB-expressing PC12 cells comparable to that of BDNF stimulation (Figure 10.3f and g).8 These results show that light activation of optoRTKs recapitulates natural RTK signaling, but more conveniently than other approaches and with potential benefits for cell biology and, possibly, in vivo experiments. 10.3.4. Responses to Different Light Conditions Because of its versatile applications, optoFGFR1 has been used here as an example of the light-responsive properties of optoRTKs. Although light responsiveness will depend mainly on the properties of the PHR domain, the characteristics of the cytoplasmic region of the FGFR1 and attached FPs may affected the dimerization efficiency upon exposure to light.9 To test the effect of wavelength, we applied lasers with different wavelengths to activate optoFGFR1. A 488 nm laser was optimal for activation, as predicted, and the 514 nm laser yielded comparable results (Figure 10.4a and b).13 A 405 nm laser was also able to activate optoFGFR1, albeit less efficiently than 488 and 2016, 345–362, N. Kim, J. M. Kim and W. D. Heo, © Springer Science + Business Media, New York, 2016, with permission of Springer. Part (e) reprinted from Chemistry & Biology, 21 (7), N. Kim, J. M. Kim, M. Lee, C. Y. Kim, K.-Y. Chang and W. D. Heo, Spatiotemporal control of fibroblast growth factor receptor signals by blue light, 903–912, Copyright 2014, with permission from Elsevier. Parts (f), (g) and (h) reprinted with permission from Macmillan Publishers Ltd: Nature Communications, 5, K.-Y. Chang, D. Woo, H. Jung, S. Lee, S. Kim, J. Won, T. Kyung, H. Park, N. Kim, H. W. Yang, J.-Y. Park, E. M. Hwang, D. Kim and W. D. Heo, Light-inducible receptor tyrosine kinases that regulate neurotrophin signaling, 4057, Copyright 2014.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

192

Nury Kim, Hyerim Park, Doyeon Woo and Won Do Heo

514 nm lasers, but lasers with a wavelength equal to or longer than 561 nm were unable to induce any activation. In terms of the power of light stimulation, we measured the activation strength and responsiveness, which indicate the amount of activation measured in each cell and in a group of cells, respectively. No changes in intracellular Ca2+ level were observed following the application of a single, low-power stimulus (0.1 µW, the lowest available power setting) to cells expressing optoFGFR1. However, cells started to respond to low-power light stimulation if multiple rounds of exposure were applied (Figure 10.4c), a trend that was particularly evident in a population of cells (Figure 10.4d). The stimulation intensity had only modest effects on functional outcome. Whereas increasing laser power had no effect on cell migration speed (Figure 10.4e), increased light intensity resulted in a small, but statistically significant, enhancement in neurite outgrowth efficiency and length of neurites (Figure 10.4f–h).8 With optimization of light wavelength and power, the amount of activation can be modulated according to the purpose of the experiment. 10.3.5. Application of Diverse Actuators Optogenetic actuators are the critical functional determinants of optoRTKs. Different optogenetic modules yield different efficiencies and, by improving optogenetic actuators, it is possible to improve optoRTK signaling efficiency. To demonstrate this, we compared optoFGFR1 with an optogenetic module of murine Fgfr1 (opto-mFgfr1), which incorporates the LOV domain of aureochrome photoreceptors from Vaucheria frigida.21 With optoFGFR1, ERK was activated more rapidly and its nuclear translocation was relatively highly maintained (Figure 10.5a). It is likely that these differences reflect differences between the LOV domain and PHR, which induce dimerization and oligomerization, respectively, upon light stimulation. This interpretation is further supported by results obtained with a PHR variant with enhanced oligomerization efficiency (Figure 10.5c). However, we cannot rule out the possibility that these differences result from species-specific differences in the FGFR1 gene. During studies on PHR, a very short (nine-residue) peptide that substantially enhances light-induced CRY2 clustering when appended at the C-terminus was serendipitously discovered. This enhanced PHR variant was named CRY2clust,9 succeeding the name CRY2olig,22 a previously described, enhanced PHR harboring a single mutation. To assess the properties of CRY2clust as actuator of homo- and hetero-interaction, we investigated the efficiency of CRY2clust clustering with CIBN, the N-terminal region of the CRY2-binding protein CIB1. Cells expressing CRY2clust formed clusters upon stimulation with light (Figure 10.5b); however, when coexpressed with CIB1, the efficiency of PHR self-clustering decreased. Clustering was observed only in cells with a low level of CIB1 expression. This suggests that the region involved in PHR self-clustering is related to the region involved in interacting with CIB1, implying that CRY2clust can be used only in modules that

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

Building Light-inducible Receptor Tyrosine Kinases

193

Figure 10.5.  Effects of different actuators in optoRTK modules. (a) Quantification of activation of HeLa cells expressing optoFGFR1 constructs containing a PHR or LOV module following continuous light stimulation (488 nm, 5 µW). (b) Cells expressing a CRY2olig or CRY2clust construct, with or without CIBN, exposed to continuous light stimulation (488 nm, 5 µW). The numbers of clusters were counted and are presented as a time course. (c, d) A comparison of the activation efficiency of HeLa cells expressing optoRTK constructs containing different PHR variants. Activation profiles of ERK in cells expressing OptoFGFR1 (c) or optoTrkB (d) are shown. Error bars: ± SEM.

require homo-multimerization. To test the effect of enhanced oligomerization in optoRTKs, we changed the PHR domain in optoFGFR1 and optoTrkB to CRY2clust. The resulting optoFGFR1 variant was indistinguishable from the original, but CRY2clust-containing optoTrkB exhibited increased activation kinetics and a change in activated status (Figure 10.5c and d). CRY2clust also exhibited faster deactivation kinetics, suggesting that it could increase the temporal resolution of stimulation. These results illustrate that enhancing optogenetic actuators is a powerful approach for improving optogenetic tools. Although enhanced oligomerization of PHR enhanced the signaling efficiency, it should be noted that the differences in properties between various optogenetic actuators do not necessarily indicate superiority; in particular, the PHR domain and CRY2clust were unable to induce optogenetic activation

View Online

194

Nury Kim, Hyerim Park, Doyeon Woo and Won Do Heo

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

of EGFR, whereas the LOV domain did activate signals.21 In our view, the palette of opto-RTKs can be expanded only with more trial-and-error studies using various combinations of different modules and different RTKs.

10.4. Conclusions The idea of applying light-responsive modules as rapid and efficient actuators of signaling dynamics in vitro and in vivo has opened up a new era of synthetic biology – optogenetics. Although the term optogenetics has been used to refer to the use of various channelrhodopsins and halorhodopsins to modulate membrane potential, it has now expanded to include the regulation of cellular signaling pathways. A good example of these latter applications is optoRTKs. Starting from optoTrkB, we were able to identify other RTKs, including optoFGFR1, that could be made light responsive through the introduction of various optogenetic modules. We hope that the studies described here, which provide an overview of signal activation and subsequent changes in cell morphology and migration in response to stimulation by light of varying wavelengths and intensities, will inform future uses of optoRTKs. With advances in optogenetic modules, including the serendipitously discovered CRY2clust, it will be possible to improve optoRTKs and apply them more efficiently, especially in vivo. It has also been demonstrated that additional functionalities can be achieved through combination with other optogenetic methods such as LARIAT.23 In this example, the emerging new tool IM-LARIAT can be used to inhibit RTK signaling, providing a technique for optogenetically investigating the regulation of vesicle trafficking.24 We expect that optoRTKs, empowered by advances in other technologies such as wavefront shaping,25 are and will continue to be versatile tools for biological studies, allowing users to modulate cell behavior precisely for their specific experimental purposes.

Acknowledgements This work was supported by the Institute for Basic Science (No. IBS-R001-G1), Republic of Korea.

References 1. M. A. Lemmon and J. Schlessinger, Cell signaling by receptor tyrosine kinases, Cell, 2010, 1117–1134. 2. D. Tischer and O. D. Weiner, Illuminating cell signalling with optogenetic tools, Nat. Rev. Mol. Cell Biol., 2014, 15, 551–558. 3. J. E. Toettcher, C. A. Voigt, O. D. Weiner and W. A. Lim, The promise of optogenetics in cell biology: interrogating molecular circuits in space and time, Nat. Methods, 2011, 8, 35–38.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

Building Light-inducible Receptor Tyrosine Kinases

195

4. M. J. Kennedy, R. M. Hughes, L. A. Peteya, J. W. Schwartz, M. D. Ehlers and C. L. Tucker, Rapid blue-light-mediated induction of protein interactions in living cells, Nat. Methods, 2010, 7, 973–975. 5. L. J. Bugaj, A. T. Choksi, C. K. Mesuda, R. S. Kane and D. V. Schaffer, Optogenetic protein clustering and signaling activation in mammalian cells, Nat. Methods, 2013, 10, 249–252. 6. L. Duan, J. Hope, Q. Ong, H.-Y. Lou, N. Kim, C. McCarthy, V. Acero, M. Z. Lin and B. Cui, Understanding CRY2 interactions for optical control of intracellular signaling, Nat. Commun., 2017, 8, 547. 7. N. Kim, J. M. Kim, M. Lee, C. Y. Kim, K. Y. Chang and W. D. Heo, Spatiotemporal control of fibroblast growth factor receptor signals by blue light, Chem. Biol., 2014, 21, 903–912. 8. K.-Y. Chang, D. Woo, H. Jung, S. Lee, S. Kim, J. Won, T. Kyung, H. Park, N. Kim, H. W. Yang, J.-Y. Park, E. M. Hwang, D. Kim and W. D. Heo, Light-inducible receptor tyrosine kinases that regulate neurotrophin signalling, Nat. Commun., 2014, 5, 4057. 9. H. Park, N. Y. Kim, S. Lee, N. Kim, J. Kim and W. D. Heo, Optogenetic protein clustering through fluorescent protein tagging and extension of CRY2, Nat. Commun., 2017, 8, 30. 10. Y. Zhao, S. Araki, J. Wu, T. Teramoto, Y.-F. Chang, M. Nakano, A. S. Abdelfattah, M. Fujiwara, T. Ishihara, T. Nagai and R. E. Campbell, An expanded palette of genetically encoded Ca2+ indicators, Science, 2011, 333, 1888–1891. 11. H. W. Yang, M.-G. Shin, S. Lee, J.-R. Kim, W. S. Park, K.-H. Cho, T. Meyer and W. D. Heo, Cooperative activation of PI3K by Ras and Rho family small GTPases, Mol. Cell, 2012, 47, 281–290. 12. D. M. Shcherbakova and V. V. Verkhusha, Near-infrared fluorescent proteins for multicolor in vivo imaging, Nat. Methods, 2013, 10, 751–754. 13. N. Kim, J. M. Kim and W. D. Heo, Optogenetic control of fibroblast growth factor receptor signaling, in Optogenetics. Methods in Molecular Biology, Humana Press, New York, NY, 2016, vol. 1408. 14. H. Park, S. Lee and W. D. Heo, Protein inactivation by optogenetic trapping in living cells, in Optogenetics. Methods in Molecular Biology, Humana Press, New York, NY, 2016, vol. 1408. 15. I. Maruyama, Mechanisms of activation of receptor tyrosine kinases: monomers or dimers, Cells, 2014, 3, 304–330. 16. J. E. Toettcher, O. D. Weiner and W. A. Lim, Using optogenetics to interrogate the dynamic control of signal transmission by the Ras/Erk module, Cell, 2013, 155, 1422–1434. 17. O. D. Weiner, Regulation of cell polarity during eukaryotic chemotaxis: the chemotactic compass, Curr. Opin. Cell Biol., 2002, 14, 196–202. 18. A. J. Ridley, M. A. Schwartz, K. Burridge, R. A. Firtel, M. H. Ginsberg, G. Borisy, J. T. Parsons and A. R. Horwitz, Cell migration: integrating signals from front to back, Science, 2003, 302, 1704–1709. 19. J. M. Kim, M. Lee, N. Kim and W. D. Heo, Optogenetic toolkit reveals the role of Ca2+ sparklets in coordinated cell migration, Proc. Natl. Acad. Sci. U. S. A., 2016, 113, 5952–5957. 20. S. Traverse, N. Gomez, H. Paterson, C. Marshall and P. Cohen, Sustained activation of the mitogen-activated protein (MAP) kinase cascade may be required for differentiation of PC12 cells. Comparison of the effects of nerve growth factor and epidermal growth factor, Biochem. J., 1992, 288(Pt 2), 351–355.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00181

196

Nury Kim, Hyerim Park, Doyeon Woo and Won Do Heo

21. M. Grusch, K. Schelch, R. Riedler, E. Reichhart, C. Differ, W. Berger, A. Ingles-Prieto and H. Janovjak, Spatio-temporally precise activation of engineered receptor tyrosine kinases by light, EMBO J., 2014, 33, 1713–1726. 22. A. Taslimi, J. D. Vrana, D. Chen, S. Borinskaya, B. J. Mayer, M. J. Kennedy and C. L. Tucker, An optimized optogenetic clustering tool for probing protein interaction and function, Nat. Commun., 2014, 5, 4925. 23. S. Lee, H. Park, T. Kyung, N. Y. Kim, S. Kim, J. Kim and W. D. Heo, Reversible protein inactivation by optogenetic trapping in cells, Nat. Methods, 2014, 1–6. 24. M. K. Nguyen, C. Y. Kim, J. M. Kim, B. O. Park, S. Lee, H. Park and W. D. Heo, Optogenetic oligomerization of Rab GTPases regulates intracellular membrane trafficking, Nat. Chem. Biol., 2016, 12, 431–436. 25. J. Yoon, M. Lee, K. Lee, N. Kim, J. M. Kim, J. Park, H. Yu, C. Choi, W. D. Heo and Y. Park, Optogenetic control of cell signaling pathway through scattering skull  using wavefront shaping, Sci. Rep., 2015, 5, 13289.

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

Chapter 11

Mechanotransduction and Optogenetics Adèle Kerjouan and Olivier Destaing* DYSAD, Institute for Advanced Biosciences (IAB), Centre de Recherche UGA/ Inserm, U 1209/CNRS UMR 5309, 38042, Grenoble, France *E-mail: [email protected]

Table of Contents 11.1.  I ntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 11.2.  Optogenetic Regulation of Mechanotransduction   at the Tissue Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 11.2.1.  Small Rho GTPases Are Key Regulators of   Force Modulation in Tissues. . . . . . . . . . . . . . . . . . . . . . .    202 11.2.2.  How Can a Predefined Spatial Constraint   Have an Impact on Tissue Dynamics?. . . . . . . . . . . . . . .    204 11.3.  Photocontrol of Mechanotransduction Processes   at the Cellular Level. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207 11.3.1.  Light-sensitive Increases in Contractility   Can Mimic the Initiation of Cytokinesis . . . . . . . . . . . . .    207 11.3.2.  Optogenetics Regulation of Cell Migration. . . . . . . . . . . 208 11.4.  Controlling the Individual Regulators of   Mechanotransduction by Optogenetics. . . . . . . . . . . . . . . . . . . . . . 212 11.4.1.  Optogenetic Control of ECM Receptors,   the Integrins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .    213 11.4.2.  Optogenetic Control of Signalling Elements   Downstream of Integrins. . . . . . . . . . . . . . . . . . . . . . . . . .    214 11.4.3.  Optogenetic Control of Actin Remodelling. . . . . . . . . . . 215

  Optogenetics: Light-driven Actuators and Light-emitting Sensors in Cell Biology Edited by Sophie Vriz and Takeaki Ozawa © European Society for Photobiology 2019 Published by the Royal Society of Chemistry, www.rsc.org

197

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

198

Adèle Kerjouan and Olivier Destaing

11.5.  O  ptogenetics Sheds Light on the Spatiotemporal   Regulation of Signalling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 11.5.1.  Temporal Regulation of a Signal Transduction. . . . . . . . 216 11.5.2.  Spatial Regulation of a Signal Transduction. . . . . . . . . . 217 11.5.3.  Diffusion of a Signal as a Feature of Spatial   Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .    217 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218

View Online

Mechanotransduction and Optogenetics

199

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

11.1. Introduction Cell homeostasis is sustained by sensing activities that influence cell reactions to the complexity of the microenvironment. The challenge of cellular sensing activities is to integrate both biochemical and physical parameters in order for cells to adapt to continuous changes in the external world. The biological interpretation of physical parameters (rigidity, stiffness, topology, curvature, pressure, neighbourliness, temperature, pH, osmolarity, etc.) supposes that cells transform external information into manageable biochemical signals. Mechanotransduction is a concept that integrates mechanical constraints, forces and physical properties into biochemical information.1 Reciprocally, cells also generate different types of forces in order to shape and adapt their microenvironment; these forces influence multiple processes such as cell division, differentiation and migration. Mechanotransduction processes occur at multiple scales of space and time, from molecular stretching to the shaping of tissues that occurs during development. Mechanotransduction is molecularly integrated into multiple adhesive structures that mediate cell–cell contact or interactions between cells with the extracellular matrix (ECM). The ability of these structures to support mechanotransduction is maintained by the accumulation of transmembrane receptors such as cadherins, which mediate cell–cell contacts, or integrins, which mediate cell–ECM contacts. Cadherins and integrins are mechanically active receptors that function through their dynamic biochemical links with the different elements of the cytoskeleton, primarily the mostly acto-myosin cytoskeleton, microtubules and intermediate filaments.2,3 Adhesive structures are characterized by their dynamics and by their ability to be activated both mechanically and biochemically (Figure 11.1). For example, the adaptor protein p130Cas undergoes mechanically dependent conformational switching that is necessary to reveal phosphorylation sites that support signal transduction.4 Biological systems have integrated complex and entangled signalling networks in order to couple the sampling of the mechanical environment with the generation of coordinated cellular responses. Intracellular signalling is sensitive to a high number of cellular inputs passing through specific receptors at the plasma membrane and is capable of integrating them to induce appropriated cell responses. Proteomic approaches have listed numerous Tyr–Ser/Thr kinases, GTPases, integrins and actin cytoskeleton regulators that are organized in a complex network to regulate mechanotransduction.5,6 In addition to the diversity of regulators identified biochemically, their spatial and temporal organization need to be considered in order to understand the cooperativity of different cellular responses that supports mechanotransduction. On the one hand, proper signal transduction requires a high level of local regulation to support the asymmetric distributions necessary for a mechanoresponse. For example, the development of FRET-based biosensors revealed the dynamics of numerous regulators of mechanotransduction such

Published on 11 September 2018 on https://pubs.rsc.org |

200 Adèle Kerjouan and Olivier Destaing

Figure 11.1.  Mechanical response to ECM is sustained by a functional chain of mechanotransduction. Force sensing and production are under the activation of a chain of mechanosensitive modules comprising clustered integrins, an associated mechanosignalling hub and the acto-myosin network. Integrins, RTKs and GPCRs receive physical and/or biochemical information from the ECM. Their signalling activity will affect the activity of GEFs through accumulations of specific phospholipids and, then, control the activity of Rho family small GTPases. These specific signalling elements will regulate concomitantly actin polymerization, adhesion site dynamics and myosin activity in order to modulate cell contractility. The motor pulling activity of myosins applied to the mechanical chain composed of actin cytoskeleton-adhesion sites–ECM will generate mechanical constraints in the cell microenvironment.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

Mechanotransduction and Optogenetics

201

as GTPases and Tyr–Ser/Thr kinases, for submicron processes occurring on the seconds scale.7,8 On the other hand, mechanical modulations can be achieved in less than 0.1 s and transmitted through a few microns.9 Defining the causal links between both the dynamic and spatial organization of biochemical signalling and mechanical properties is one of the great challenges to understanding mechanotransduction. Genetic and pharmacological approaches highlighted the importance of many molecular determinants in the formation of mechanical cellular responses. However, these cellular processes present different spatiotemporal characteristics that eventually made the traditional tools difficult to use for probing and controlling mechanotransduction and migration processes so that they could be studied on the multiple spatiotemporal scales on which they occur. Optogenetics is a new technological opportunity to modulate specific biological functions with high spatial precision and high temporal resolution that are closer to the native subcellular conditions of many dynamic cellular responses implicated in mechanotransduction. This synthetic approach allows the exploration of new aspects of molecular physiology that are poorly accessible with genetic and pharmacological approaches. Optogenetics has been used to reorganize directly the acto-myosin network, to trigger force generation and ECM remodelling and to regulate the inside-out signalling that occurs in adhesion sites. Optogenetics suitably complements the new tools designed to measure and induce mechanical constraints.10 In this chapter, we highlight some of the major optogenetic approaches and the light-sensitive tools that have been applied recently to investigate the spatiotemporal aspects of mechanotransduction on multiple scales. We first focus on the force transmission between cells at the tissue level to define the power and the range of the forces implicated in tissue organization. Then we describe how optogenetics allows us to decipher the spatial regionalization of force transmitters at the subcellular level. Indeed, the polarization of acto-myosin organization and force production is essential to achieve specific cellular processes such as division and migration. Next, we describe the multiple tools designed to control the molecular aspects of cell signalling that occurs in adhesive sites. Finally, we explore how optogenetics is not only a new technique for increasing the resolution at which such activities can be studied but also supports the development of new paradigms about the notion of spatiotemporal signalling activity in biology and the processing of information by biological systems.

11.2. Optogenetic Regulation of Mechanotransduction at the Tissue Level Tissue remodelling is a process regulated at the cell level through division and apoptosis, cell differentiation, cell migration and changes in cell shapes. By being mostly responsible for the mechanical forces, the acto-myosin

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

202

Adèle Kerjouan and Olivier Destaing

cytoskeleton network modulates contraction–relaxation cycles that directly regulate cell size and shape. In the case of highly cohesive tissues such as the epithelium or endothelium, remodelling requires supracellular coordination of the acto-myosin network to maintain tissue homeostasis. The acto-myosin network allows long-range transmission of forces and can produce the large-scale changes necessary for morphogenesis and the modification of morphology or polarity/directionality. However, how mechanical cues are integrated and propagated at the tissue level is a debated question for which optogenetics could open up new directions of research. As a complementary method to classical genetic mosaic or pharmacological approaches, optogenetics allows the study of the impact of regionalized mechanical constraints or changing frequency of contraction–relaxation cycles on tissue remodelling. Indeed, light allows the correlation of the spatial patterning of cell contractility and general tissue behaviour to be followed. Thus, the spatiotemporal coordination between cells at the tissue level and the determination of the characteristic distance of action of mechanical constraints can be explored. Cellular contractility is generated by a suite of different actors that directly affect the organization of the acto-myosin cytoskeleton and the efficiency of force transmission. Force production is dependent on a functional chain linking the transmembrane receptors clustered in adhesive sites with the regions of polymerized actin structured as architectural elements (crosslinkers and regulators of actin polymerization) that support the actions of cytoskeletal motors (composed of a large family of myosins). Although highly specific, the challenge of choosing an optogenetics strategy to control contractility globally is to act cooperatively on all these modules. For over 20 years, members of the Rho family small GTPases have been described as general switches mediating adhesive sites and acto-myosin cytoskeleton organization.11 Therefore, optogenetic control of Rho GTPases and their regulators has been used to modulate cell contraction in tissue. 11.2.1. Small Rho GTPases Are Key Regulators of Force Modulation in Tissues The transient and local control of small GTPases is essential for many developmental processes. Such spatiotemporal control is especially necessary for the small GTPase Rac1 during the collective migration of border cells in the Drosophila ovary. Border cells are formed by a group of 6–8 epithelial cells that migrate collectively from the surrounding epithelium, through the nurse cells into the egg chamber. Classical genetic and mutant expression approaches showed that Rac1 had an essential role in promoting the migration of the border cells. However, the mechanism by which the modulation of the GTPase in this group of cells could regulate such collective behaviour was unclear. Wu et al. took advantage of the photoactivatable Rac1 (PA-Rac1) probe to control directly spatiotemporal Rac1 activation with light. PA-Rac1 is a fusion of Rac1 and the photoreactive light-, oxygen- and voltage-sensing  (LOV) domain of phototropin that can selectively block the interactions

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

Mechanotransduction and Optogenetics

203

of Rac1 with its effectors.12 In response to blue light, the conformational change of the α-helix between the LOV domain and Rac1 allows the GTPase to bind multiple effectors and activate its classical downstream signalling pathways leading to the regulation of the acto-myosin cytoskeleton. The expression of PA-Rac1 in the Drosophila ovary allowed the induction of the collective migration of epithelial cells and especially borders cells. Regionalization of PA-Rac1 activation was used to induce the collective migration in all directions. A striking feature revealed by this approach is that activation of PA-Rac1 in only one cell of the group is sufficient to induce a collective migration of these borders cells, suggesting the important mechanical and signalling coherence between them to sustain this migration behaviour. In order to achieve both an increase and a decrease in cell contractility in a monolayer of epithelial cells, light control of the GTPase RhoA was developed.13 To activate RhoA, Valon et al. designed an optogenetic probe to control the association of the specific RhoA activator ARHGEF11 with either the plasma membrane or the mitochondrial membrane. Guanine nucleotide exchange factors (GEFs) display characteristic selectivity for different members of the Rho GTPases family. Each GEF is characterized by the presence of both a Dbl domain (DH) and a pleckstrin homology domain (PH).13 Thus, the recruitment of GEFs at the plasma membrane allows the binding of the PH domain to lipids, leading to its activation and the subsequent activation of small GTPase(s). The photosensitive molecule CRY2 was fused with ARHGEF11 to target specifically RhoA. The light-sensitive heterodimerization of CRY2 with its partner CIBN was used to control the spatial localization in discrete subcellular compartments by illumination.14 The expression of CIBN fused to either a plasma membrane-anchoring domain (CIBN–GFP–CAAX) or a mitochondria-anchoring domain (mito–CIBN–GFP) allowed the recruitment of ARHGEF11–CRY2 between two different compartments. Light-dependent relocalization of ARHGEF11 to the plasma membrane increases its proximity with the activating membrane and endogenous RhoA, leading to the activation of the RhoA GTPase and the stimulation of the contractile machinery. In contrast, the light-dependent titration of ARHGEF11 at mitochondria kept it away from the membrane and reduced its ability to activate endogenous RhoA. Based on traction force microscopy, the cellular contraction at the level of a monolayer was determined by integrating the difference between the contractile force applied to the ECM (measured by the mean traction on the ECM, in Pa) and intercellular tension (in mN m−1). The photosensitive relocalization of ARHGEF11 to the plasma membrane increased the global cellular tension while the titration of ARHGEF11 at mitochondria decreased the cellular contractility; both effects appeared on a time scale of minutes. The modulation of cell contraction was also correlated with minutes-scale actin polymerization events. The differential control of ARHGEF11 localization yielded an important dynamic range in terms of mechanical constraints (from a 1.5-fold increase to a fourfold decrease in cell contraction). Hence it was possible to determine how the

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

204

Adèle Kerjouan and Olivier Destaing

local modulation of mechanical constraints affects tissue properties. Particle image velocimetry (PIV) was used to correlate the velocity of the cell shape changes, cell movements and mechanical constraints in response to photosensitive contraction–relaxation cycles. The activation of ARGEF11 in a subregion of only 50 cells in an MDCK monolayer induced their centripetal movements inside the region of photostimulation, leading to tissue deformation and a significant increase in cellular density. In the absence of photostimulation, tissue relaxation supports the rapid mechanical adaptation of the cellular monolayer. In contrast, a light-sensitive decrease in cell contractility induced centrifugal cell movements from the centre of the illuminated square to its periphery (Figure 11.2). In contrast with light-dependent increased contractility, cell relaxation of the illuminated cells affected the cells' behaviour at a distance from the stimulated area, suggesting that the imposed mechanical constraint had long-range action. The relationship between the induced mechanical constraints and the frequency or duration of ARHGEF11 modulation unravelled how temporal control of mechanics affects tissue morphogenesis. In fact, the same method was used to investigate other mechanosensitive signalling pathways and, in particular, the relocalization of the transcription factor YAP. It was shown that actin polymerization reaches a maximum on the minutes scale whereas YAP nuclear relocalization reaches a maximum only after tens of minutes. The ability to compare mechanical cues with their associated biochemical signalling events opens up a new dimension in understanding the functional difference between pulsed and long-term mechanical constraints on both cell movement and cell differentiation. 11.2.2. How Can a Predefined Spatial Constraint Have an Impact on Tissue Dynamics? A decrease in cellular contractility in tissue has been achieved in vivo by using the CRY2–CIBN system to modulate the level of PI(4,5)P2 at the plasma membrane of blastodermic epithelial cells in Drosophila embryos.15,16 Controlling the membrane localization of the lipid 5-phosphatase OCRL allows a decrease in the content of PI(4,5)P2 lipids, which regulate many cellular processes, including clathrin-mediated endocytosis, protein recruitment, GEF activation and, notably, actin polymerization. Light-dependent depletion of PI(4,5)P2 in the aforementioned blastoderm epithelial cells affects actin polymerization, observed as a decrease in the recruitment of the actin-binding protein moesin from the plasma membrane on a time scale of minutes.16 The Drosophila embryo is a reliable model for studying the modulation of cell contractility by tracking and analysing its effects on mechanodependent developmental processes. A good example of such a process is the invagination of the 3D mesoderm following the formation of the ventral furrow. This process is executed by a small group of approximately 1000 cells organized rectangularly along the antero-posterior (a-p) axis of the embryo. The coordinated constriction of the cell apical surface generates a specific tissue-level

View Online

205

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

Mechanotransduction and Optogenetics

Figure 11.2.  Contractility control by light-sensitive RhoA optogenetic probes. (A) Modulation of RhoA activity by the photosensitive CRY2–CIBN heterodimerization system. DH–PH domains of ARHGEF11 were fused to CRY2 and formed a cytosolic photosensitive element that can be either recruited/activated at the membrane

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

206

Adèle Kerjouan and Olivier Destaing

force that affects the furrow and then facilitates invagination. Finally, the constriction of this rectangular group of cells also influences the furrow geometry and induces an eccentric tissue morphology (named a-p anisotropy) due to a preferential reduction in cell size in one direction, perpendicularly to the a-p axis. In addition to perturbing actin polymerization, the membrane recruitment of OCRL blocks the contraction process leading to furrow formation in Drosophila embryos. Because of the transient nature of the contraction events and the small number of blastoderm cells involved, optogenetics emerges as a key approach to perturb the spatiotemporal contractility that occurs during this developmental invagination process. The main challenge in implementing the optogenetic approaches in embryos is the selective activation of cells located deep within a 3D environment. Hence a two-photon illumination strategy was used to activate CRY2 and locally photoactivate the tissue within the 3D structure. The stimulation of whole embryos globally decreases cell contractility and prevents tissue invagination after furrow constriction is induced. Coupling the local optogenetics-dependent relaxation with furrow constriction helped to determine how cell behaviour can influence global tissue remodelling. Indeed, the laser power delivered by two-photon excitation allowed the inhibition of contractility in only a subset group of cells (approximately 75 cells in 1000) in the furrow. The relaxation of this subgroup of cells was sufficient to block completely the whole ventral furrow constriction, indicating that long-range mechanical coordination between cells is necessary for the developmental process. Based on the geometric changes that occur during a-p anisotropy, the regionalization of the contractility decrease was used to explore the importance of the rectangular geometry of a group of 1000 cells during furrow formation. Modulation of the distance between regions in which OCRL was activated revealed that a change from rectangular to square geometry at the tissue level decreased the a-p anisotropy associated with furrow invagination. Light-dependent local inhibition of cell contractility induces a loss of coordinated contractile behaviour and allowed the limiting factors (cell numbers and geometry) that interfere with long-range force transmission along the supracellular acto-myosin meshwork of the ventral furrow to be described. In conclusion, optogenetics opens up new routes to understanding the interplay between tissue geometry, force transmission and biological processing.

if expressed with CIBN–GFP–CAAX or sequestered/inhibited at the mitochondria if expressed with mito–CIBN–GFP. The expression of each couple expressed in MDCK monolayers allows a local increase or decrease in cell contractility to be induced and, then, different cell dynamics. (B) Activation of RhoA by the photosensitive TULIPs heterodimerization system. The LOV/ePDZ system was used to obtain a photosensitive membrane receptor (Stargazine) able to heterodimerize with the DH domain of LARG, which can activate endogenous RhoA with high precision in response to light. Highly local RhoA activation was used to initiate cleavage furrow formation in different stages of the cell cycle.

View Online

Mechanotransduction and Optogenetics

207

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

11.3. Photocontrol of Mechanotransduction Processes at the Cellular Level The polarization process is defined by an anisotropic distribution (molecules, second messengers, organelles, cytoskeleton, etc.) that is sustained over time. Numerous cell polarization processes are characterized by a heterogeneous organization of the acto-myosin cytoskeleton. Anisotropy of contractility distribution is thus a general process in many changes of cell shape. For example, polarization in migrating cells is defined by the induction of local protrusions at the front of migration and contraction events in adhesive sites and cytoskeleton at the rear of the cell. Moreover, efficient cell migration supposes an important coordination of these multiple mechanical elements between the protrusive edge and the rear edge in order to produce an efficient migration of the cell.17–19 Regionalization and coordination of mechanotransduction processes at the cellular level are still an open question. In this section, we illustrate how optogenetics can be used to investigate the properties of the spatiotemporal regulation of mechanical signals in cell division and migration. 11.3.1. Light-sensitive Increases in Contractility Can Mimic the Initiation of Cytokinesis Cytokinesis is the last step of cell division and consists of cell separation leading to the formation of two daughter cells. The process of cytokinesis is dependent on the formation of a furrow composed of a contractile actin ring that bends membranes and forces their fusion. RhoA regulates the formation of the contractile ring by acting on myosins and F-actin assemblies, through the activation of formins. In general, the contractile ring is formed in the middle of the mother cell. By controlling the spatial and temporal activation of RhoA with the optogenetics system TULIPs,20 Wagner and Glotzer explored how patterns of local excitation of this contractile ring could affect the efficiency of cytokinesis. For that purpose, the PH domain of LARG (another GEF specific for RhoA) was fused to an ePDZ domain tag that can bind a peptide that is inaccessible while fused to a LOV domain (LOVpep). In response to blue light, the PH domain of LARG heterodimerizes with LOVpep, which is localized at the plasma membrane through its fusion with the transmembrane receptor Stargazine and then locally activates endogenous RhoA.21 By patterning the light in a very fine equatorial region, precise RhoA activation generated a contractile furrow during a non-contractile anaphase. However, light-induced furrows were insufficiently mature to ingress fully and to give rise to cell separation. Moreover, light-induced furrows regressed temporarily if there was no continuous photoactivation. This result confirms the intense polarized RhoA dynamic activation necessary to sustain a contractile structure and to form the furrow. Indeed, no furrow was generated if the photostimulation patterns covered too large an area or targeted only one side of the cells. This optogenetic framework is characterized by the presence of the

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

208

Adèle Kerjouan and Olivier Destaing

light-sensitive element at the plasma membrane that allows the maintenance of a sharp recruitment of GEF activity to locally activate endogenous RhoA. Diffusion modelling revealed that if the photosensitive element is cytosolic (such as in the CRY2–CIBN system), the minimal spatial resolution of optogenetic relocalization is larger, namely between 5 and 10 µm.22 Tight spatial control of RhoA activation supports furrow formation at both the cell poles and equator and at different steps of cell division such as metaphase and anaphase. The induction of a contractile ring even during the interphase confirmed that RhoA-dependent furrow formation is poorly regulated by the cell cycle. However, light-induced furrows were formed during the interphase only in the absence of adhesive structures between the cell and the ECM. Indeed, inducing the rounding of cells in the interphase with trypsin allowed furrow formation at more than 70% of ingression. Induction of synthetic furrow formation confirmed the importance of the relationship between the local increase in contractility of the actin rings and the increase in isotropic cortical tension during mitotic entry.23 To explore this relationship, Wagner and Glotzer combined pharmacological treatment for globally decreasing cortical tension with optogenetics in order to control the rate of furrow ingression (Figure 11.2). While the localized activation of RhoA is not sufficient to induce the full process of cytokinesis, the regulation of contractility appears also to be an insufficient signal for total cell division. Again, the optogenetics approach brings new perspectives to the equilibrium of contractility that is needed to generate efficient furrows during the process of cell division. 11.3.2. Optogenetics Regulation of Cell Migration Directed migration is subsequent to spatial compartmentalization of different mechanical processes over time. The migration process requires the spatiotemporal coordination of adhesion sites turnover and acto-myosin cytoskeleton remodelling to produce membrane protrusions at the front edge and contractile retraction at the rear edge (Figure 11.3). During migration, multiple molecular actors orchestrate this global coordination. Among them, the small Rho GTPases and the GPCR transmembrane receptors are key coordinators of cell migration because of their ability to regulate spatiotemporally second messengers such as calcium or PI(3,4,5)P3.8 However, understanding the role of each player is particularly complicated because of the high coordination of all signalling pathways in charge of cell migration. This section describes the different optogenetic approaches used to control the initiation and persistence of cell migration. In this case, the use of optogenetics is particularly adapted owing to the high spatial and temporal resolution of this technique. For example, using light patterning, the activity of regulators of migration is asymmetrically induced in order to address the general question of how biochemical signalling permits the coordination of the mechanical machinery that supports cell movement. The importance of cell polarization (with front–rear axis) to the promotion of efficient migration will be illustrated here.

View Online

209

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

Mechanotransduction and Optogenetics

Figure 11.3.  Induction of cell migration by optogenetic control of front–rear polarization. (A) Chemoattractant or rigidity gradients are sensed by integrins or RTKs/ GPCRs and then lead to a front–rear polarization. This process is supported by an increase in actin polymerization and formation of adhesion sites in the front edge (leading edge or lamellipodia). In parallel, the rear edge is characterized by an increase in intracellular tensions due to the activation of the RhoA–stress fibres–myosinII pathway. (B) Local photostimulation of optogenetic probes (GTPases, bOpsin, FGFR, etc.) can induce different steps in cell migration. Local PA-Rac1 photoactivation permits front formation without rear modulation. However, local photoactivation of Cdc42, FGFR or bOpsin is sufficient to induce a full front–rear polarization through the formation of a PI(3,4,5)P3 gradient, RhoA activation and Ca2+ sparklets at the rear of the migrating cell.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

210

Adèle Kerjouan and Olivier Destaing

11.3.2.1.  Cell Polarization During Migration Induced by Light-sensitive Small Rho GTPases.  Several studies have used FRET-based biosensors to show the importance of the localized activity of Rho family GTPases in inducing and sustaining cell migration.24–26 Thus, photocontrol of GTPases emerges as a tractable method of controlling cell migration. The previously introduced PA-Rac1 fusion was used to initiate cell migration in prostate cancer (PC-3) cells through the polarized activation of Rac1.27 Limited light stimulation at one border of the PA-Rac1-expressing PC-3 cells induced the minute-scale formation of lamellipodia in the stimulated area only after the first activation. The coupling of optogenetics and specific pharmacological inhibitors revealed that the activation of PI3 kinase is a downstream effect of PA-Rac1 signalling necessary for lamellipodia extension, but is not necessary for other Rac1-dependent phenotypes such as 3D membrane ruffling. However, the appearance of a migration front defined by membrane protrusion was not sufficient to induce the full migration. When the effects of PA-Rac1 and other optogenetic probes on cell migration were compared, it seemed that PA-Rac1 is not fully capable of inducing cell migration. The transmembrane receptor tyrosine kinase (RTK) FGFR1 phosphorylates substrates in response to its clustering. The photoactivation of FGFR1 fused to the light-sensitive CRY2 protein induced its homo-oligomerization and led to the phosphorylation of multiple substrates. Local activation of an optoFGFR1–CRY2 fusion in HUVECs was sufficient to induce both lamellipodia formation and rear-edge contraction, both of which are required for cell movement.28 In contrast, the same local activation of PA-Rac1 sustains only lamellipodia formation and not the necessary front–rear coordination. This pattern seems not to be the case for other small Rho GTPases. The local photoactivation of Cdc42 in the RAW 264.7 macrophage cell line induces the full migration process. The expression of an improved light-induced dimer (iLID)-based system29,30 allowed the membrane recruitment of the PH–DH domains of the GEF intersectin to be controlled to activate specifically Cdc42.31 Continuous photoactivation of Cdc42 at the border of RAW 264.7 cells induced their polarization and led to a directed migration process. By tracking GTP-bound domains of small GTPases, it appears that light-stimulated Cdc42 activated the Rac1 proteins at the front edge and activated RhoA–ROCK–myosin II pathways at the rear edge. However, light-dependent Cdc42 activation was not sufficient to induce the classical polarized pattern of PI(3,4,5)P3 seen in migrating cells. Targeting Cdc42, then, appears to be sufficient to induce the global front–rear polarization necessary for the support of cell migration. The precise mechanism by which a local action can induce a long-range organization is an open matter. Further work towards the precise quantification of the local activity of a small Rho GTPase and its range of action (proportional to the cell size) on multiple downstream signalling pathways or cellular responses should illuminate this point. 11.3.2.2.  Cell Polarization During Migration Induced by Light-sensitive G-protein Coupled Receptors (GPCRs).  GPCRs are inducers of cell migration that are sensitive to external chemical cues. To understand better the full

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

Mechanotransduction and Optogenetics

211

process behind the initiation of migration, the non-rhodopsin opsin was used to activate spatially G-protein signalling to guide migration.32 The photoreceptor blue opsin (bOpsin) was used to activate endogenous Gi on one side of RAW 264.7 macrophage cells. In contrast to Cdc42 activation, bOpsin-dependent Gi activation was able to induce the formation of both the full front–rear edge polarization and the classical accumulation of PI(3,4,5)P3 specifically at the front of migrating RAW 264.7 cells.33 Single-cell analysis allowed the sensitivity of RAW 264.7 cells to bOpsin stimulation to be tested. RAW 264.7 cells having the same level of bOpsin expression responded unequally to blue light and two different populations appeared. Early migrants needed less light stimulation to initiate their migration or PI(3,4,5)P3 polarization than did late migrants. To explain this differential sensitivity, the dynamics of PI(3,4,5)P3 were modelled by considering the diffusion of both a membrane activator and a cytosolic inhibitor of PI(3,4,5)P3. This model proposes that a spatially constrained threshold level of PI(3,4,5)P3 influences cell reactivity to light stimulation. This sensitive switch-like signalling process appears to influence considerably changes in decision making at the cellular level in response to GPCR regulation. In addition to creating a PI(3,4,5)P3 gradient from the front to the rear of migrating cells, local photostimulation can also induce a gradient of G-protein activity. Activation of bOpsin dissociates the G-protein from the GPCR, leading to the membrane anchoring of Gαi and the cytosolic release of Gβγ into the cytosol, where it can diffuse throughout the cell. Siripurapu et al. investigated the different signalling pathways downstream of local bOpsin activation that leads to RAW 264.7 cell migration. Coupling the local activation of bOpsin with the Gβγ inhibitor (GRK2Ct-CRY2)34 showed that Gαi poorly supports the induction of light-dependent polarization and migration of RAW 264.7 cells. Thus, after bOpsin activation, Gβγ can diffuse throughout the cell and activate the RhoA–ROCK–myosin II pathway at the rear edge.35 The regulation of local GPCR activation appears to be a simple mechanism for polarizing migrating cells at long distances through a diffusive transport mechanism. 11.3.2.3.  Optogenetic Probing Reveals a Different Ca2+ Pool Implicated in Cell Migration.  Besides PI(3,4,5)P3 metabolism and small GTPase regulation, it was tempting to hypothesize that second messengers could also support front–rear polarization in response to local stimulation of migration regulators. Interestingly, bOpsin activation cannot induce migration in RAW 642.7 cells when Ca2+ release is blocked either by using a Ca2+-free medium or by chelating intracellular Ca2+.35 The control of Ca2+ signalling by Gβγ could support long-distance coordination based on the ability of this signalling pathway to form an intracellular Ca2+ gradient. To decipher the spatial patterning of Ca2+ release and investigate its role in the coordination between processes at the front and rear edges of a migrating cell, Kim and co-workers controlled migration by using two previously described optogenetic probes, optoFGFR1 and PA-Rac1.28 As already discussed, the local activation of optoFGFR1 induces a rapid (within

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

212

Adèle Kerjouan and Olivier Destaing

20 min) front–rear polarization based on the local reorganization of F-  actin from stress fibres to branched actin at the front of the cell. The activation of PA-Rac1 can also coordinate front–rear polarization, but at a slower rate (within 90 min). Based on these different optogenetic strategies for inducing migratory phenotypes, the authors correlated this light-induced morphological transition with Ca2+ release measured through the Lyn-RGECO1 sensor. The photoactivation of optoFGFR1 induced not only a large Ca2+ burst around the illuminated area but also small Ca2+ sparklets outside the stimulated area, primarily at the rear edge of the cell. These Ca2+ release events are specific to the photostimulation of optoFGFR1 and are not observed in response to the photostimulation of PA-Rac1. Comparison of the distribution, amplitude and duration of the Ca2+ sparklets and the Ca2+ bursts made it clear that optoFGFR1 induces a front–rear Ca2+ gradient. This result raised a question about the implication of the Ca2+ gradient in the mechanical coordination of front–rear polarization. Targeting specific Ca2+ channels (nimodipine sensitive) perturbs this gradient and leads to the inhibition of the front–rear coordination and the cell migration in response to optoFGFR1 activation. The use of biosensors to correlate optogenetic stimulation with the modulation of downstream signalling activities or cellular outputs is the main limitation of this type of approach. The use of optogenetic probes limits multiplexed imaging, especially the use of FRET-based probes, because blue light cannot be used strictly for observation. However, the precise choice of compatible fluorescent probes and biosensors has allowed the quantification of the evolution of downstream signalling elements (such as phosphoinositide formation, Ca2+ release and GTP-binding domain recruitment) in response to a precise optogenetic activation.36 This quantification is essential for the correlation of signal transfer efficiency with the dynamics of cellular responses.

11.4. Controlling the Individual Regulators of Mechanotransduction by Optogenetics The optogenetic targeting of small Rho GTPases, GPCRs and RTKs induced a global reorganization of cells illustrated by front–rear polarization and cell migration. This result shows that the sustained stimulation of the targeted inducers was sufficient to activate downstream effectors implicated in the different aspects of mechanotransduction (Figure 11.1). This section describes methods for the direct phototargeting of specific elements that are essential in the chain of events involved in mechanotransduction. The optogenetics toolkit is beginning to be broad enough for the design of multiple strategies to control signalling or effector elements of mechanotransduction. The general strategy adopted in the design of light-sensitive probes is to generate a conformation that is tightly folded and inactive in the absence of photostimulation but becomes active upon photostimulation.

View Online

Mechanotransduction and Optogenetics

213

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

11.4.1. Optogenetic Control of ECM Receptors, the Integrins Integrins are transmembrane receptors that form a mechanical link between the extracellular matrix and the intracellular cytoskeleton. Integrin clustering also supports the formation of signalling platforms that transduce information from the environment into the cell, thus adapting cellular behaviour and vice versa (outside-in and inside-out signalling). Protein-based photosensitizers have been used to improve chromophore-assisted light irradiation (CALI) approaches and inactivate rapidly a specific integrin. KillerRed is one of the photosensitizers that produces reactive oxygen species (ROS) through the formation of a water channel close to its chromophore site.37,38 The fusion of KillerRed with β1-integrins was used to determine the immediate action of β1-integrin inactivation by ROS on invadosomes. Invadosome structures support the invasion of cells into the ECM by coupling both acto-adhesive functions with local ECM degradation.39 The transient nature of invadosomes, which have a life-span of less than 20 min, makes it very difficult to assess their spatiotemporal characteristics and their regulator functions with classical genetic experiments (gene silencing or conditional knockout models). In β1-integrinlox/lox cells, the loss of β1-integrin proteins after the induction of the conditional knockout takes between 2 and 4 days and completely blocks invadosome formation. Although invadosome inhibition could be an indirect consequence of the loss of β1-integrins, the minute-scale depletion of β1-integrin allows its direct influence on invadosome dynamics to be deciphered. Thus, β1-integrins fused to KillerRed (β1–KR) was expressed in β1-integrin−/− cells to restore invadosome formation and to allow the rapid functional inactivation of β1–KR in response to red light. In this example, ROS production induced by the photostimulation of β1–KR functionally inactivated the protein fused to KR, rapidly disorganized the invadosomes and induced the formation of focal adhesions and stress fibres.40 Comparison of the inactivation of dynamin (a regulator of actin formation in invadosome) and β1-integrins revealed that this integrin functions as a positional marker of invadosome formation rather than as an essential structural element necessary for actin polymerization.41 The stimulation of β1–KR by red light induces its specific functional inactivation and on-target ROS production, whereas the activation of β1–KR in the presence of either the endogenous proteins or β1–GFP does not lead to invadosome destruction. The appeal of the KillerRed approach is the ability to mimic rapidly classical genetic loss-of-functions experiments and its limitation consists primarily of the inability to localize repeatedly and spatially the functional inactivation of the target protein. The use of photosensitive elements to modulate the affinity between proteins is a complementary strategy for achieving cyclic stimulations of integrins in a subcellular region. Integrins are an interesting example of the use of optogenetics to modulate subtle biochemical equilibria in live cells. Indeed, the short cytoplasmic domain of integrins can bind numerous partners, albeit not concurrently and with medium affinity values. Integrins

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

214

Adèle Kerjouan and Olivier Destaing

can switch easily between different partners to maintain the high dynamic response and plasticity of adhesion sites. This observation motivates interest in modulating the affinity between integrins and their activators such as talin and kindlin. As a proof of principle, the fusion of CRY2 to the F3 domain of the integrin activator talin allowed the possibility of increasing the local talin concentration by light-sensitive homo-oligomerization, leading to the formation of β3-integrin clusters.42 The second example of integrin photomodulation is illustrated by the light-sensitive recruitment of kindlin-2 to β3-integrins.43 For that purpose, the authors used the TULIPs system to fuse LOVpep to a mutant of β3-integrin (β3 integrins ΔRGT) that cannot bind kindlins, whereas kindlin-2 was fused with ePDZ1. Indeed, one key for a suitable optogenetic protein–protein interaction-based approach is to have no interaction between the two targeted proteins in the absence of photostimulation (dark). Expression of this system in HUVEC cells showed that inducing interaction between kindlin-2 and β3-integrins ΔRGT activates different mechanotransduction responses. The interaction between β3-integrins and kindlin-2 promotes the migration of HUVECs through fibrinogen-coated Transwell membranes (ECM bound by β3-integrins), HUVECs sprouting and invadosome formation. In addition to defining the functional relevance of this interaction, mapping the different domains of kindlin-2 implicated in the induction of these cellular responses is possible. Combining relevant mutations in kindlin-2 and in β3-integrins with this optogenetic method allows the investigation of the precise molecular mechanisms implicated in HUVEC adhesion, migration and sprouting and invadosome formation. 11.4.2. Optogenetic Control of Signalling Elements Downstream of Integrins Integrin clustering forms signalling platforms and recruits numerous signalling elements such as the non-receptor tyrosine kinases such as c-Src and FAK. Focal adhesion kinase (FAK) is a signal transducer activated by outside-in signalling and is essential for the phosphorylation of numerous substrates/effectors directly implicated in mechanical responses. The design of a light-sensitive FAK is based on the fusion of FAK and the photosensitive CRY2 protein.44 To be precise, the FAK-W266A mutant presents a reduced ability to self-dimerize and is poorly localized in focal adhesions in the absence of photostimulation. In response to stimulation with blue light, optoFAK (FAK-W266A–CRY2) oligomerization induces the autophosphorylation of a key marker of FAK activation (pY397). The light-sensitive activation of FAK induces both relocalization to focal adhesions and the phosphorylation of many classical FAK partners such as FAK itself, paxillin, Src and p130Cas. OptoFAK is an interesting tool for manipulating mechanotransduction independently of integrin inputs and also for specifically studying inside-out adhesion signalling.

View Online

Mechanotransduction and Optogenetics

215

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

11.4.3. Optogenetic Control of Actin Remodelling Optogenetic activation of different small Rho GTPases has allowed the activation of actin polymerization and general cytoskeletal organization. The precise manipulation of individual molecular functions that are directly implicated in the actin cytoskeleton dynamics is also of interest. Diaphanous-related formins (DRFs) are components of the actin polymerization machineries that promote elongation of the barbed end of actin filaments. The direct control of actin polymerization by light has been accomplished using a LOV domain strategy to control the activation of the formin mDIA1. The transition from a closed to an open conformation, which defines mDIA1 activation, is supported by an intramolecular bond between the diaphanous inhibitory domain (DID) and the diaphanous autoregulatory domain (DAD). The LOV2-Jα domain of Avena sativa was fused to the DAD in order to inhibit its binding ability in the absence of light. Thus, photostimulation of the optogenetic probe released a functional DAD that can compete with the inhibitory intramolecular bond of endogenous mDIA1 and induces its open/active conformation.45 The photostimulation of LOV2-Jα–DAD at the cellular level induced actin polymerization in stress fibres without affecting recruitment of myosins and focal adhesion morphology. The cellular responses were directly due to the photoactivation of LOV2-Jα–DAD and were inhibited by preincubation with a formins inhibitor. Actin dynamics are characterized by an essential coupling between polymerization and disassembly processes, and it is very interesting to control these two types of events with light. ADF/cofilin family members are actin-binding proteins that regulate the disassembly of actin filaments through stochastic disassembly, debranching or severing. Cofilin binds actin filaments as a complex molecule that displays a stronger affinity for ADP-bound actin subunits. Cofilin binding dynamically affects actin ATPase activity and its binding activity induces severing and depolymerization activities.46 Based on substantial biochemical data, a light-sensitive cofilin was designed for the direct spatiotemporal control of actin depolymerization. To modulate the binding equilibrium between actin filaments and cofilin, the optogenetic heterodimerizing system CRY2–CIBN was used to for the optical control of cofilin concentration on actin filaments bound by the Lifeact peptide. To obtain a probe that had a poor binding affinity for actin filaments and a reduction in the important severing activity, the cofilin-S3A-S120A–CRY2 mutant was designed and was used to remodel actin locally in cells expressing Lifeact–CIBN. Light stimulation was used to induce the local translocation of cofilin-S3A-S120A–CRY2 from the cytosol to the polymerized actin cytoskeleton, which led to the formation of filopodia or protrusions. By controlling the actin filament dynamics, it is possible to change the actin architecture and to direct cell motility with local light stimulation.

View Online

216

Adèle Kerjouan and Olivier Destaing

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

11.5. Optogenetics Sheds Light on the Spatiotemporal Regulation of Signalling As described above, the regulation of mechanotransduction is highly complex, since it involves numerous molecular players located in various cell compartments (from membrane to cytoskeleton), together with a whole hierarchy of spatial and temporal scales. Following relative concentration of an activated effector downstream, a signalling element or a structural element can be easily quantifiable by live microscopy using fluorescent probes and FRET biosensors. Understanding the spatiotemporal regulation of a signalling element is complex but necessary for the precise definition and quantification of signalling transfer in space and time. Classically, information is defined as the content of news exchange between partners. A signalling element can transfer information by directly interacting with effectors leading to their activation (as for small GTPases) or by inducing post-translational modifications that affect their repertoire of binding partners (phosphorylation, ubiquitination, SUMOylation, etc.). 11.5.1. Temporal Regulation of a Signal Transduction The temporal character of a biological signal can be defined semiquantitatively by specifying whether it is transient or stable or appears before or after a reference event. The temporal profiles of a signalling node induce different signal information for specific pathways.47 Multiple studies showed that temporal modulation of a signal encodes cellular outputs. For example, the modulation of the quantity of p53 over time is sufficient to induce distinct cellular phenotypes.48 Signalling outputs also depend on the activation frequency of a signalling node. Indeed, the small GTPase Ras induces a specific phenotype mostly in response to a characteristic activation frequency.49 The next challenge is to define the consequences of the temporal modulations of signalling over time. To our knowledge, it is not yet possible to follow quantitatively how information propagates from signalling molecules (GTPases, ubiquitin ligase, kinases, etc.) towards their direct substrates/effectors. For example, considering a signalling kinase as a typical signalling element, a FRET-based biosensor could measure its phosphorylation activity on a synthetic substrate but not the global signalling transfer over time (the quantity of all phosphorylated substrates downstream of this kinase). The high temporal resolution of optogenetics could one day allow us to modulate certain dynamic properties on a signalling node and follow the information transfer in response to mechanotransduction. For example, the modelling and characterization of the dynamics of the optogenetic heterodimerizer system CRY2–CIBN already showed that it was possible to tune the local quantity of a signalling element by playing mostly on the frequency of light stimulation.23

View Online

Mechanotransduction and Optogenetics

217

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

11.5.2. Spatial Regulation of a Signal Transduction In addition to being a new technology for controlling known signalling pathways, the use of optogenetics opens up new paradigms in signalling transfer. One of the greatest limitations of the classical activation of a signalling element is its spatial confinement. The localized activation of a specific node permits the downstream cellular responses to be measured at long range, supporting the concept of a characteristic distance of action of a signalling activation event. For example, the local activation of optoFGFR1 induces Ca2+ sparklets at a distance of a few microns. This signalling property seems extremely important for different mechanoresponses that are sustained by the cooperation of cellular structures at long distance, such as cell division, cell migration and tissue remodelling. Determining the molecular mechanism underlying the long-distance effect of local signal activation requires consideration of the diffusion of both the signalling element (transducer) and the molecule receiving the information (substrates/effectors). 11.5.3. Diffusion of a Signal as a Feature of Spatial Regulation Limiting spatially the diffusion of both signalling transducers and substrates/effectors is an important step occurring during signalling. Spatial reduction from a 3D space (cytosol) to a 2D surface (membrane or subcellular structure such an adhesion site) is an essential step to allow signalling transfer by increasing local concentration and limiting molecule diffusion.50,51 Integrating the difference in molecular mobility between 2D and 3D spaces is also essential for inducing the formation of a signalling gradient. Meyers et al. investigated how a change of cell shape could affect signalling. Their model considered a signalling transducer transferring post-translational modifications (kinases, ubiquitin ligase, etc.), which is controlled by both a cytosolic inhibitor and a membrane activator. During front spreading of a migrating cell, the local alterations of the cell shape increase considerably the membrane/cytoplasm ratio, which in turn modulates the ratio of the concentrated activator (membrane, 2D surface) to the diffusive inhibitor (cytosol, 3D space) and then directly controls activation of the signalling element.52 Defining the distance browsed by an activated signalling node diffusing on a surface over time is another element of space that is important for signalling transfer. Live super-resolution imaging showed that signalling elements such as c-Src kinase, the integrin mechanoreceptors or Rac1 GTPase switch between free diffusion, breaking and immobilization events over tens of nanometres.53–55 These diffusive behaviours define signalling clusters of different sizes that will transfer information (phosphorylation, conformational changes) on diffusive downstream substrates/effectors that should also be organized into clusters. Integration of the potential difference of size between signalling clusters and substrates/effectors will be essential for characterizing the spatial pattern of information transferred.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

218

Adèle Kerjouan and Olivier Destaing

The correlation between spatiotemporal patterns of signalling molecules and the properties of signalling transfer is not really clear. This raises the question of how information can be transferred through or even stored inside the dynamics of signalling molecules. The induction of dynamic signalling clusters is a key step in signalling transfer. Computational analysis proposed that the formation of signalling clusters discretizes a continuous input signal (e.g. the concentration of a chemoattractant or a fixed value of ECM rigidity), which is a continuous and smooth signal over time, into a digital signal, which is a binary step-like signal. For example, theory and stochastic simulations proposed that optimal nanometre-size clusters of Ras maximize its signalling fidelity and support the mapping of a maximal number of input states to distinct output states with minimal noise.56 By directly modulating biochemical equilibrium in space and time, optogenetic approaches reveal new aspects of the spatiotemporal regulation that occurs in signalling. Controlling the molecular mobility during a specific time period and over a characteristic distance is another approach to embed the specificity, constancy and fidelity of signal transmission in highly dynamic processes such as mechanotransduction.

Acknowledgements We thank Dr Faurobert and Professor Fourcade for critical reading, exciting discussions, valuable comments and their availability. This work was supported by the JCJC ANR Invadocontrol programme and by the Ligue Nationale contre le Cancer (LLNC) as Equipe Labellisée Ligue 2014. A. K. was funded by the LLNC and La Fondation pour la Recherche Médicale (FRM).

References 1. T. Iskratsch, H. Wolfenson and M. P. Sheetz, Nat. Rev. Mol. Cell Biol., 2014, 15, 825–833. 2. Z. Sun, S. S. Guo and R. Fässler, J. Cell Biol., 2016, 4, 445–456. 3. R. M. Mège and N. Ishiyama, Cold Spring Harb. Perspect. Biol., 2017, 9(5). 4. Y. Sawada, M. Tamada, B. J. Dubin-Thaler, O. Cherniavskaya, R. Sakai, S. Tanaka and M. P. Sheetz, Cell, 2006, 127, 1015–1026. 5. J. C. Kuo, J. Cell. Mol. Med., 2013, 17, 704–712. 6. H. B. Schiller, C. C. Friedel, C. Boulegue and R. Fäsignssler, EMBO Rep., 2011, 12, 259–266. 7. K. Honda, H. Yanai, H. Negishi, M. Asagiri, M. Sato, T. Mizutani, N. Shimada, Y. Ohba, A. Takaoka, N. Yoshida and T. Taniguchi, Nature, 2005, 434, 772–777. 8. R. D. Fritz and O. Pertz, F1000Research, 2016, 5, 749. 9. D. Mitrossilis, J. Fouchard, D. Pereira, F. Postic, A. Richert, M. Saint-Jean and A. Asnacios, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 16518–16523. 10. D. Seo, K. M. Southard, J. W. Kim, H. J. Lee, J. Farlow, J. U. Lee, D. B. Litt, T. Haas, A. P. Alivisatos, J. Cheon, Z. J. Gartner and Y. W. Jun, Cell, 2016, 165, 1507–1518.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

Mechanotransduction and Optogenetics

219

11. C. D. Nobes and A. Hall, Cell, 1995, 81, 53–62. 12. Y. I. Wu, D. Frey, O. I. Lungu, A. Jaehrig, I. Schlichting, B. Kuhlman and K. M. Hahn, Methods Enzymol., 2011, 497, 393–407. 13. L. Valon, A. Marín-Llauradó, T. Wyatt, G. Charras and X. Trepat, Nat. Commun., 2017, 8, 14396. 14. M. J. Kennedy, R. M. Hughes, L. A. Peteya, J. W. Schwartz, M. D. Ehlers and C. L. Tucker, Nat. Methods, 2010, 7, 973–975. 15. O. Idevall-Hagren, E. J. Dickson, B. Hille, D. K. Toomre and P. De Camilli, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, E2316–E2323. 16. G. Guglielmi, J. D. Barry, W. Huber and S. De Renzis, Dev. Cell, 2015, 35, 646–660. 17. G. Danuser, J. Allard and A. Mogilner, Annu. Rev. Cell Dev. Biol., 2013, 29, 501–528. 18. R. D. Fritz, M. Letzelter, A. Reimann, K. Martin, L. Fusco, L. Ritsma, B. Ponsioen, E. Fluri, S. Schulte-Merker, J. Van Rheenen and O. Pertz, Sci. Signaling, 2013, 6, 1–14. 19. O. Hoeller, J. E. Toettcher, H. Cai, Y. Sun, C. H. Huang, M. Freyre, M. Zhao, P. N. Devreotes and O. D. Weiner, PLoS Biol., 2016, 14, 1–35. 20. D. Strickland, Y. Lin, E. Wagner, C. M. Hope, J. Zayner, C. Antoniou, T. R. Sosnick, E. L. Weiss and M. Glotzer, Nat. Methods, 2012, 9, 379–384. 21. E. Wagner and M. Glotzer, J. Cell Biol., 2016, 213, 641–649. 22. L. Valon, F. Etoc, A. Remorino, F. di Pietro, X. Morin, M. Dahan and M. Coppey, Biophys. J., 2015, 109, 1785–1797. 23. M. P. Stewart, J. Helenius, Y. Toyoda, S. P. Ramanathan, D. J. Muller and A. A. Hyman, Nature, 2011, 469, 226–231. 24. C. M. Welch, H. Elliott, G. Danuser and K. M. Hahn, Nat. Rev. Mol. Cell Biol., 2011, 12, 749–756. 25. K. Kemp-O'Brien and M. Parsons, J. Microsc., 2013, 251, 270–278. 26. S. K. Donnelly, J. J. Bravo-Cordero and L. Hodgson, Cell Adhes. Migr., 2014, 8, 526–534. 27. T. Kato, K. Kawai, Y. Egami, Y. Kakehi and N. Araki, PLoS One, 2014, 9, 1–9. 28. N. Kim, J. M. Kim, M. Lee, C. Y. Kim, K. Y. Chang and W. D. Heo, Chem Biol., 2014, 21, 903–912. 29. G. Guntas, R. A. Hallett, S. P. Zimmerman, T. Williams, H. Yumerefendi, J. E. Bear and B. Kuhlman, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, 112–117. 30. S. P. Zimmerman, R. A. Hallett, A. M. Bourke, J. E. Bear, M. J. Kennedy and B. Kuhlman, Biochemistry, 2016, 55, 5264–5271. 31. P. R. ONeill, V. Kalyanaraman and N. Gautam, Mol. Biol. Cell, 2016, 27, 1442–1450. 32. W. K. A. Karunarathne, L. Giri, V. Kalyanaraman and N. Gautam, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, E1565–E1574. 33. W. K. A. Karunarathne, L. Giri, A. K. Patel, K. V. Venkatesh and N. Gautam, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, E1575–E1583. 34. L. M. Klayman and P. B. Wedegaertner, J. Biol. Chem., 2017, 292, 1773–1784. 35. P. Siripurapu, D. Kankanamge, K. Ratnayake, K. Senarath and A. Karunarathne, J. Biol. Chem., 2017, 292, 17482–17495. 36. J. E. Toettcher, D. Gong, W. A. Lim and O. D. Weiner, Nat. Methods, 2011, 8, 837–839. 37. M. E. Bulina, D. M. Chudakov, O. V. Britanova, Y. G. Yanushevich, D. B. Staroverov, T. V. Chepurnykh, E. M. Merzlyak, M. A. Shkrob, S. Lukyanov and K. A. Lukyanov, Nat. Biotechnol., 2006, 24, 95–99. 38. P. Carpentier, S. Violot, L. Blanchoin and D. Bourgeois, FEBS Lett., 2009, 583, 2839–2842.

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00197

220

Adèle Kerjouan and Olivier Destaing

39. O. Destaing, C. Petropoulos and C. Albiges-Rizo, Cell Adhes. Migr., 2014, 8, 256–262. 40. O. Destaing, E. Planus, D. Bouvard, C. Oddou, C. Badowski, V. Bossy, A. Raducanu, B. Fourcade, C. Albiges-Rizo, M. R. Block and I. A. Bonniot, Mol. Biol. Cell, 2010, 21, 4108–4119. 41. O. Destaing, S. M. Ferguson, A. Grichine, C. Oddou, P. De Camilli, C. Albiges-Rizo and R. Baron, PLoS One, 2013, 8, e77956. 42. L. J. Bugaj, D. P. Spelke, C. K. Mesuda, M. Varedi, R. S. Kane and D. V. Schaffer, Nat. Commun., 2015, 6, 1–10. 43. Z. Liao, A. Kasirer-Friede and S. J. Shattil, J. Cell Sci., 2017, 130, 3532–3541. 44. M. Hörner, C. Chatelle, W. W. D. Mühlhäuser, D. R. Stocker, M. Coats, W. Weber and G. Radziwill, Cell. Signalling, 2018, 42, 176–183. 45. M. V. Rao, P. H. Chu, K. M. Hahn and R. Zaidel-Bar, Cytoskeleton, 2013, 70, 394–407. 46. J. Mochida, T. Yamamoto, K. Fujimura-Kamada and K. Tanaka, Genetics, 2002, 160, 923–934. 47. H. K. Makadia, J. S. Schwaber and R. Vadigepalli, PLoS Comput. Biol., 2015, 11, e1004563. 48. J. E. Purvis and G. Lahav, Cell, 2013, 152, 945–956. 49. J. E. Toettcher, O. D. Weiner and W. A. Lim, Cell, 2013, 155, 1422–1434. 50. B. N. Kholodenko, Nat. Rev. Mol. Cell Biol., 2006, 7, 165–176. 51. M. Cebecauer, M. Spitaler, A. Sergé and A. I. Magee, J. Cell Sci., 2010, 123, 309–320. 52. J. Meyers, J. Craig and D. J. Odde, Curr. Biol., 2006, 16, 1685–1693. 53. O. Rossier, V. Octeau, J.-B. Sibarita, C. Leduc, B. Tessier, D. Nair, V. Gatterdam, O. Destaing, C. Albigès-Rizo, R. Tampé, L. Cognet, D. Choquet, B. Lounis and G. Giannone, Nat. Cell Biol., 2012, 14, 1057–1067. 54. H. Machiyama, T. Yamaguchi, Y. Sawada, T. M. Watanabe and H. Fujita, FEBS J., 2015, 282, 4034–4055. 55. A. Remorino, S. De Beco, F. Cayrac, F. Di Federico, G. Cornilleau, A. Gautreau, M. C. Parrini, J. B. Masson, M. Dahan and M. Coppey, Cell Rep., 2017, 21, 1922–1935. 56. E. Roob, N. Trendel, P. Rein Ten Wolde and A. Mugler, Biophys. J., 2016, 110, 1661–1669.

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00221

Subject Index accelerated sensor of action potentials (ASAP), 87 acto-myosin network, 202, 206 Akaike's information criterion (AIC) value, 165 Akt, 151–153 bimolecular fluorescence complementation (BiFC), 66 cadherins, 199 CAD–magnet system cell membrane dynamics induction, 112–113 kinetic study, 110–111 membrane morphology optical control, 112 cell ablation, 117–132 cell migration, optogenetics regulation, 208–212 cell polarization, G-protein coupled receptors (GPCRs), 210–211 cell polarization, small Rho GTPases, 210 optogenetic probing, Ca2+ pool, 211–212 cell polarization G-protein coupled receptors (GPCRs), 210–211 small Rho GTPases, 210 cellular contractility, 202, 204 cellular signaling processes spatiotemporal dynamics of, 101–114

channelrhodopsin, 103 chemically induced dimerization (CID), 139 chromophore-assisted light inactivation (CALI), 119 agents in cell biology, 122–131 chemical-based photosensi­ tizers, 122–125 future perspectives for, 131–132 genetically encoded photosensitizers, 125–131 molecular mechanism of, 119–122 photosensitization mechanism, 120–121 ROS effects, intracellular molecules, 121 specific, 122 confocal microscopy, 7 design strategies, optogenetic reporters, 66–72 bimolecular fluorescence complementation (BiFC), 66 dimerization-dependent FPs, 66–68 engineered allosteric effects, 69–71 FRET, 68–69 intrinsic FP sensitivities enhancement, 71–72 diaphanous inhibitory domain (DID), 215 221

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00221

222 diaphanous-related formins (DRFs), 215 digitally scanned light-sheet fluorescence microscope (DSLM), 10 dimerization-dependent FPs, 66–68 enhanced green fluorescent protein (EGFP), 125 excited-state proton transfer (ESPT), 70, 71 fast volumetric imaging, 5–22 flavin-based fluorescent proteins (FbFPs), 48 fluorescent proteins (FPs), 45, 65 fluorogen-based markers for advanced bioimaging, 43–58 bilirubin-binding green fluorescent proteins, 51–52 biliverdin-binding Far-red, 49–51 engineered from natural photoreceptors, 47–52 flavin-binding cyan–green fluorescent proteins, 48–49 glowing panoply of, 43–58 infrared fluorescent proteins, 49–51 semi-synthetic, 52–58 fluorogens, 46 focal adhesion kinase (FAK), 214 force modulation regulators, 202–204 Fӧrster resonance energy transfer (FRET), 49 FRET-based GECIs, 80–81 “Fucci” (fluorescent, ubiquitination-based cell cycle indicator), 72–74

Subject Index Gaussian optics, basics of, 11–12 genetically encoded Ca2+ indicators (GECIs), 80 FRET-based GECIs, 80–81 single FP-based GECIs, 82–83 genetically encoded photosensitizers, 125–131 for photodynamic therapy, 128–131 in vitro and in vivo applications of, 129–130 genetically encoded PhyB–PIF light-inducible dimerization, 139–146 genetically encoded voltage indicators (GEVIs), 86–88 green fluorescent protein (GFP), 45 guanine nucleotide exchange factors (GEFs), 203 integrins, 199 Jablonski diagram, 121 KillerRed, 126, 127, 132 kinase activity, quantitative control, 151–166 kinase translocation reporters (KTRs), 84, 85 light-driven actuators, cellular signaling processes, 101–114 light-induced dimerization (LID) systems, cell signaling control, 139–141 light-inducible receptor tyrosine kinases, 183–194 canonical signaling pathways activation, 186–189 diverse actuators, application, 192–194 downstream activation, functional validation, 189–191

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00221

Subject Index experimental, 183–185 image analysis, 184–185 light conditions, responses, 191–192 materials, 183–184 microscope setup, 184 sample preparation, 184 screening OptoRTKs with PHR, 185–186 light-inducible transcription systems, 171–173 neurogenesis and somitogenesis, Dll1 oscillations, 175–177 oscillatory expression, Dll1, 175–178 oscillatory information cell-to-cell transfer, Dll1 oscillations, 177–178 oscillatory versus sustained expression, Ascl1, 173–175 light-sheet fluorescence microscopy (LSFM), 5–22 building, 13–16 contrast enhancement, 16–17 high-speed volumetric imaging, 16 in next decade, 22 optical principles of, 10–13 optical sectioning, 5–10 origins of, 10–11 semi-transparent samples, imaging deeper, 17–18 sensorimotor integration, vertebrate brain, 20–21 single objective, 18–19 spatial resolution in, 11–16 whole-brain functional imaging, zebrafish larvae, 19–20 Lucifer Yellow, 119

223 Malachite Green, 120, 122–124 mechanotransduction process, 199–218 actin remodelling, optogenetic control, 215 cell migration, optogenetics regulation, 208–212 cytokinesis initiation, 207–208 individual regulators, controlling, 212–215 integrins, 213–214 optogenetic regulation, tissue level, 201–206 photocontrol, cellular level, 207–212 signal transduction, temporal regulation, 216 small Rho GTPases, force modulation regulators, 202–204 spatial constraint, tissue dynamics impact, 204–206 spatial regulation, signal diffusion, 217–218 spatial regulation, signal transduction, 217 spatiotemporal regulation of signalling, 216–218 membrane potential, 86–88 messenger molecules, 78–83 calcium ion, 80 neurotransmitters, 79–80 mini singlet oxygen generator (miniSOG), 49, 128, 132 monomeric IFP (mIFP), 50 near-infrared fluorescent protein (iRFP), 50 Nelder–Mead method, 164 optical sectioning, 5–10 optical transfer function (OTF), 33

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00221

224 optogenetic reporters applications of, 72–88 for cell biology and neuroscience, 65–89 cell cycle, 72–74 design strategies, 66–72 membrane potential, 86–88 messenger molecules, 78–83 pH sensing, 74–75 programmed cell death (PCD), 75–78 protein kinases, 83–85 optogenetic RTKs (optoRTKs), 183 optogenetic son of sevenless (optoSOS) system, 151 optogenetic tools, quantitative biology, 139–146 orthogonal-plane fluorescence optical sectioning (OPFOS), 11 PA-Akt system actin remodeling, spatial regulation, 159–161 for cellular signaling analysis, 157–165 construction design of, 154–155 experimental design of, 153–157 light illumination wavelength and strength, 155–157 light-induced Akt activation, Mathematical model, 161–165 optogenetic control of, 158–159 principle of, 153–154 signaling pathway, dissecting, 159 Per-Arnt-Sim (PAS) core, 108 photoinducible protein inactivation, 117–132 PhyB–PIF LID system, 143 cell signaling, quantitative manipulation, 143–145

Subject Index phycocyanobilin (PCB), 141, 142, 171 phytochrome chromophore synthesis, 141–143 point-spread function (PSF), 31, 32 programmed cell death (PCD), 75–78 apoptosis, 76–78 non-apoptotic PCD, 78 protein kinases, 83–85 Ras–Raf–MEK–Erk signaling pathway, 153 Rayleigh range, 12 reactive oxygen species, 117–132 receptor tyrosine kinases (RTKs), 183 light-inducible, 183–194 reversible saturable optical fluorescence transitions (RESOLFT), 34, 35 Rho GTPases, 202–204 semi-synthetic fluorogen-based protein markers, 52–57 fluorescence-activating and absorption-shifting tag, 56–57 fluorogen-activating proteins, 52–55 self-labeling tags, 55–56 semi-synthetic fluorogen-based RNA markers, 57–58 signal transduction, 139 single-molecule localization, in 3D, 31–32 single molecule localization microscopy (SMLM), 29, 30 spatial resolution, LSFM, 11–13 field of view and axial resolution, 13 Gaussian optics, basics of, 11–12 resolution, light sheet, 12–13 stochastic optical fluctuation imaging (SOFI) principle, 33

View Online

Published on 11 September 2018 on https://pubs.rsc.org | doi:10.1039/9781788013284-00221

Subject Index stochastic optical reconstruction microscopy (STORM), 30 super-resolution microscopy, 25–38 diffraction limit, 27 via fluorescence fluctuations correlation, 32–34 via localization microscopy, 27–31 via optical non-linear effects, 34–35 structured illumination, 36–38 transcription, light control, 171–178 tumor necrosis factor-α (TNFα), 78 two-photon microscopy, 7–8 vivid (VVD) protein, 103, 106 assembly method, magnet system performances, 109–110 bioluminescence assay, 104–105

225 CAD–magnet system, kinetic study, 110–111 cell culture, 104 cell membrane dynamics induction, CAD–magnet system, 112–113 confocal laser scanning microscopy imaging, 105–106 dish coating, 105 experimental procedures, 103–106 membrane morphology optical control, CAD– magnet system, 112 off kinetics, half-life evaluation, 106 plasmid construction, 103–104 switch-off kinetics and dimerization efficiencies, 108