Hair Cell Regeneration 3031206606, 9783031206603

Table of contents :
The Acoustical Society of America
Series Preface
Preface 1992
Volume Preface
Contents
Contributors
Chapter 1: Sensory Regeneration in the Inner Ear: History, Strategies, and Prospects
1.1 Introduction
1.2 Historical Overview of Otic Regeneration
1.2.1 Postnatal Generation of Sensory Receptors in Vertebrates
1.2.2 Postnatal Addition of Hair Cells in Cold-Blooded Vertebrates
1.2.3 Sensory Regeneration in the Avian Inner Ear
1.3 Overview of Contents
1.4 Conclusions
References
Chapter 2: Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological and Functional Recovery
2.1 Introduction
2.2 Nonmammalian Hair Cell Regeneration: An Overview
2.3 Supporting Cell Populations and Their Functions During Regeneration
2.3.1 Identities and Locations of Hair Cell Progenitors in Fish
2.3.2 The Role of Peripheral Supporting Cells in Fish
2.3.3 Supporting Cell Diversity in Birds
2.4 Approaches to Define New Molecular Regulators Using Nonmammals
2.4.1 Transcriptional Profiling
2.4.2 Genetic and Molecular Screening
2.5 Molecular Regulation of Supporting Cells
2.5.1 Transcription Factors Regulate Hair Cell Regeneration in Nonmammals
2.5.2 Cell-Cell Signaling Molecules That Regulate Hair Cell Regeneration in Birds and Fish
2.5.2.1 Notch Signaling
2.5.2.2 Wnt Signaling
2.5.2.3 Other Signaling Pathways
2.5.3 Epigenetic Mechanisms Controlling Nonmammalian Hair Cell Regeneration
2.6 Conclusion
References
Chapter 3: Cell Junctions and the Mechanics of Hair Cell Regeneration
3.1 Introduction
3.2 Shape Change Controls Proliferation of Supporting Cells
3.3 Actomyosin Contractility at Apical Junctions Accelerates Wound Closure in the Lesioned Vestibular Epithelium
3.4 Maturational Reinforcement of Adherens Junctions Coincides with Age-Related Declines in the Plasticity of Mammalian Supporting Cells
3.4.1 The Unique Circumferential F-actin Bands in Mammalian Supporting Cells
3.4.2 Potential Mechanical Influence of the Thick F-actin Bands that Develop in Mammalian Supporting Cells
3.4.3 Structure and Regulation of the Circumferential F-actin Bands in Supporting Cells
3.4.3.1 Sarcomeric Actomyosin Network at Cochlear Apical Junctions
3.4.3.2 Regulation of the Circumferential F-actin Bands by Rho GTPases
3.5 E-cadherin Accumulates at Supporting Cell Junctions in the Mammalian Vestibular Epithelium
3.5.1 Hypothesized Role for N-cadherin in Limiting Supporting Cell Proliferation
3.5.2 A Special Case: Apical Junctions in the Anolis Lizard
3.6 Regulation and Perturbation of Apical Junctions in Mammalian Supporting Cells
3.6.1 Potential Interaction of Notch Signaling and E-cadherin Adhesion
3.7 Intracellular Signaling Downstream of Mechanical Signals
3.7.1 YAP/TAZ and the Hippo Pathway
3.7.1.1 YAP-TEAD Regulate Cell Cycle Arrest and Size Control in Hair Cell Epithelia
3.7.1.2 YAP-TEAD Signaling in Repair and Regeneration
3.7.2 Canonical Wnt Signaling
3.8 Supporting Cell-Extracellular Matrix Interactions
3.9 Summary
3.9.1 A Hypothetical Model for Mechanical Control of Hair Cell Replacement
3.9.2 Outstanding Questions and Opportunities
3.10 Conclusions
References
Chapter 4: Mammalian Hair Cell Regeneration
4.1 Introduction
4.2 Structural and Developmental Considerations
4.2.1 Structure of the Mammalian Vestibular Sensory Epithelia
4.2.2 Development of the Vestibular Sensory Epithelia
4.2.3 Structure of the Mammalian Auditory Epithelium, the Organ of Corti
4.2.3.1 Hair Cell Types
4.2.3.2 Supporting Cells
4.2.3.3 Features of Organ of Corti Development
4.3 Hair Cell Generation and Regeneration in the Immature Inner Ear
4.3.1 Generation of Supernumerary Hair Cells
4.3.2 Hair Cell Regeneration in the Immature Inner Ear
4.3.3 Stem Cells in the Sensory Epithelia
4.4 Hair Cell Generation and Regeneration in the Mature Vestibular Sensory Epithelia
4.4.1 Characteristics of Spontaneous Regeneration in Adult Mammalian Utricles
4.4.2 Origin of Regenerated Hair Cells
4.4.3 Functionality of Regenerated Vestibular Hair Cells
4.5 Enhancing Hair Cell Regeneration by Phenotypic Conversion in Mature Animals
4.5.1 Vestibular Sensory Epithelia
4.5.1.1 Notch Pathway Inhibition
4.5.1.2 Overexpression of Atoh1
4.5.2 Inducing Hair Cell Regeneration in the Mature Organ of Corti in Vivo
4.5.2.1 Overexpression of Atoh1
4.5.2.2 Notch Pathway Inhibition
4.5.3 Additional Factors to Promote Differentiation of Regenerated Hair Cells
4.5.4 Clinical Trials
4.6 Regeneration of Vestibular Hair Cells: Summary
4.7 Cochlear Cellular Pathology and Challenges to Hair Cell Regeneration and Recovery of Auditory Function
References
Chapter 5: Specification and Plasticity of Mammalian Cochlear Hair Cell Progenitors
5.1 Introduction
5.2 Induction of the Inner Ear and the Development of Prosensory Patches
5.3 Regulation and Function of the Atoh1 Transcription Factor During HC Development
5.4 Promotion of Supporting Cell Fate Through Notch-Mediated Lateral Inhibition from Hair Cells
5.5 Regulation of Supporting Cell Fate Decisions: Extracellular Signals and Intracellular Transcription Factors
5.6 Toward Hair Cell Regeneration: Lessons from Non-mammalian Models and Neonatal Mice
5.7 Enhancing Mammalian Hair Cell Regeneration: Reprogramming of Supporting Cells into Hair Cells
5.8 Epigenetic Regulation of Gene Expression in the Cochlea
5.9 Summary
References
Chapter 6: Inner Ear Cells from Stem Cells: A Path Towards Inner Ear Cell Regeneration
6.1 Introduction
6.1.1 Pluripotent Stem Cells
6.2 Two-Dimensional Culture Systems
6.2.1 Inner Ear Replacement Parts and Hair Cells from Scratch
6.2.2 Hair Cell-Like Cells from Pluripotent Stem Cells
6.2.3 Limitations of Two-Dimensional Culture Systems
6.3 Three-Dimensional Culture and Organoids
6.3.1 Derivation of Inner Ear Organoids
6.3.2 Limitations of Three-Dimensional Culture Systems
6.4 Stem Cells in the Adult Inner Ear
6.4.1 Stem Cells Are the Source of the Strong Regenerative Capacity of the Avian Inner Ear
6.4.2 Stem Cells in the Inner Ear of Adult Rodents Are Restricted to the Vestibular System
6.5 Stem Cells in the Neonatal Rodent Cochlea
6.5.1 Supporting Cells of the Neonatal Organ of Corti Display Stem Cell-Like Capacity
6.5.2 Proliferation of Dissociated Neonatal Organ of Corti Supporting Cells Can be Manipulated with Growth Factors and Small Molecules
6.6 Emerging New Methods Towards Hair Cell Regeneration
6.6.1 CRISPR Genome Editing to Enhance Stem Cell Applications?
6.6.2 Supporting Cell Reprogramming
6.7 Conclusion
References
Chapter 7: Spiral Ganglion Neuron Regeneration in the Cochlea: Regeneration of Synapses, Axons, and Cells
7.1 Introduction to Spiral Ganglion Neurons
7.1.1 Peripheral Connections of Spiral Ganglion Neurons
7.1.2 Glia and Myelination
7.1.3 Central Connections of SGNs
7.1.4 Neurotrophic Factors
7.2 Synaptopathy and Synapse Regeneration
7.2.1 Primary Degeneration
7.2.2 Secondary SGN Degeneration After Hair Cell Loss
7.3 Regeneration of SGN Axons and Guidance of Their Growth
7.3.1 Stimulation of Growth by Neurotrophic Factors
7.3.2 Intracellular Signaling for Neurite Growth/Retraction
7.3.3 Directional Guidance by Neurotrophic Factors
7.3.4 Directional Guidance by Substrate Pattern/Materials
7.4 Cochlear Implants Present a Model for SGN Regeneration
7.4.1 Cochlear Implants Depend on SGN Health
7.4.2 Cochlear Trauma and Inflammation Likely Reduce SGN Health and CI Performance
7.5 Regeneration of SGNs from Stem Cells
7.5.1 Assessments of Success
References
Chapter 8: Genetic and Epigenetic Strategies for Promoting Hair Cell Regeneration in the Mature Mammalian Inner Ear
8.1 Introduction
8.2 Mouse Models for Altering Gene Expression
8.2.1 Cre/lox Technology
8.2.1.1 Cell Type-Specific and Temporal Control of Recombination
8.2.1.2 Generation of Cre and CreERT Mouse Lines
8.2.1.3 Characterizing Cre/CreERT Expression
8.2.1.4 Interpreting Cre/CreERT Reporter Expression
8.2.1.5 Cre/CreERT Efficiency
8.2.1.6 Caveats for Cre/CreERT Studies
8.2.2 CreERT Alleles Expressed in Supporting Cells
8.2.2.1 CreERT Alleles for Adult Auditory Supporting Cells
8.2.2.2 CreERT Alleles for Adult Vestibular Supporting Cells
8.2.3 Tetracycline-Responsive Gene Regulation
8.2.4 CRISPR/Cas Approaches
8.2.4.1 Generation of Knockout Mice
8.2.4.2 Advantages Over Conventional Mouse Mutagenesis
8.2.4.3 Genetic Knock-In Using CRISPR/Cas
8.2.4.4 Cas9-Expressing Mouse Lines
8.2.4.5 Reducing Off-Target Effects
8.2.4.6 DNA Base Editing with Dead Cas9
8.3 Targeting Epigenetic Processes for Hair Cell Regeneration
8.4 Virus-Mediated Gene Transfer in Adult Supporting Cells
8.4.1 Introduction to Adeno-Associated Virus
8.4.1.1 AAV Infection and Tropism
8.4.1.2 Recombinant Adeno-Associated Viral Vectors
Capsid Discovery to Achieve Novel Transduction
8.4.1.3 Packaging Capacity
8.4.2 Gain of Function in Supporting Cells
8.4.2.1 Supporting Cell Transduction Efficiency
8.5 Loss of Function in Supporting Cells
8.5.1 MicroRNAs and Small Interfering Ribonucleic Acids
8.5.2 Antisense Oligonucleotides
8.6 Summary: Refocusing Hair Cell Regeneration Model Systems
References

Citation preview

Springer Handbook of Auditory Research

Mark E. Warchol · Jennifer S. Stone Allison B. Coffin · Richard R. Fay  Arthur N. Popper   Editors

Hair Cell Regeneration With 33 Illustrations

Springer Handbook of Auditory Research Volume 75 Series Editors  Arthur N. Popper, Ph.D., University of Maryland Richard R. Fay, Ph.D., Loyola University Chicago

Editorial Board Karen Avraham, Ph.D., Tel Aviv University, Israel Andrew Bass, Ph.D., Cornell University Lisa Cunningham, Ph.D., National Institutes of Health Bernd Fritzsch, Ph.D., University of Iowa Andrew Groves, Ph.D., Baylor University Ronna Hertzano, M.D., Ph.D., School of Medicine, University of Maryland Colleen Le Prell, Ph.D., University of Texas, Dallas Ruth Litovsky, Ph.D., University of Wisconsin Paul Manis, Ph.D., University of North Carolina Geoffrey Manley, Ph.D., University of Oldenburg, Germany Brian Moore, Ph.D., Cambridge University, UK Andrea Simmons, Ph.D., Brown University William Yost, Ph.D., Arizona State University

The ASA Press ASA Press, which represents a collaboration between the Acoustical Society of America and Springer Nature, is dedicated to encouraging the publication of important new books as well as the distribution of classic titles in acoustics. These titles, published under a dual ASA Press/Springer imprint, are intended to reflect the full range of research in acoustics. ASA Press titles can include all types of books that Springer publishes, and may appear in any appropriate Springer book series. Editorial Board Mark F. Hamilton, University of Texas at Austin Timothy F. Duda, Woods Hole Oceanographic Institution Gary Elko, mh Acoustics Robin Glosemeyer Petrone, Threshold Acoustics William M. Hartmann (Ex Officio), Michigan State University Darlene R. Ketten, Boston University James F. Lynch (Ex Officio), Woods Hole Oceanographic Institution Philip L. Marston, Washington State University Andrew Norris, Rutgers University Arthur N. Popper (Ex Officio), University of Maryland Christine H. Shadle, Haskins Laboratories G. Christopher Stecker, Boys Town National Research Hospital Ning Xiang (Chair), Rensselaer Polytechnic Institute

Mark E. Warchol  •  Jennifer S. Stone Allison B. Coffin  •  Arthur N. Popper Richard R. Fay Editors

Hair Cell Regeneration

Editors Mark E. Warchol Department of Otolaryngology Washington University School of Medicine St. Louis, MO, USA Allison B. Coffin Department of Integrative Physiology and Neuroscience Washington State University Vancouver, WA, USA Richard R. Fay Department of Psychology Loyola University Chicago Chicago, IL, USA

Jennifer S. Stone Department of Otolaryngology/Head and Neck Surgery Virginia Merrill Bloedel Hearing Research Center University of Washington School of Medicine Seattle, WA, USA Arthur N. Popper Department of Biology University of Maryland College Park, MD, USA

ISSN 0947-2657     ISSN 2197-1897 (electronic) Springer Handbook of Auditory Research ISBN 978-3-031-20660-3    ISBN 978-3-031-20661-0 (eBook) https://doi.org/10.1007/978-3-031-20661-0 © Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

The Acoustical Society of America On 27 December 1928, a group of scientists and engineers met at Bell Telephone Laboratories in New York City to discuss organizing a society dedicated to the field of acoustics. Plans developed rapidly, and the Acoustical Society of America (ASA) held its first meeting on 10–11 May 1929 with a charter membership of about 450. Today, ASA has a worldwide membership of about 7000. The scope of this new society incorporated a broad range of technical areas that continues to be reflected in ASA’s present-day endeavors. Today, ASA serves the interests of its members and the acoustics community in all branches of acoustics, both theoretical and applied. To achieve this goal, ASA has established Technical Committees charged with keeping abreast of the developments and needs of membership in specialized fields, as well as identifying new ones as they develop. The Technical Committees include acoustical oceanography, animal bioacoustics, architectural acoustics, biomedical acoustics, engineering acoustics, musical acoustics, noise, physical acoustics, psychological and physiological acoustics, signal processing in acoustics, speech communication, structural acoustics and vibration, and underwater acoustics. This diversity is one of the society’s unique and strongest assets since it so strongly fosters and encourages cross-disciplinary learning, collaboration, and interactions. ASA publications and meetings incorporate the diversity of these Technical Committees. In particular, publications play a major role in the society. The Journal of the Acoustical Society of America (JASA) includes contributed papers and patent reviews. JASA Express Letters (JASA-EL) and Proceedings of Meetings on Acoustics (POMA) are online, open-access publications, offering rapid publication. Acoustics Today, published quarterly, is a popular open-access magazine. Other key features of ASA’s publishing program include books, reprints of classic acoustics texts, and videos. ASA’s biannual meetings offer opportunities for attendees to share information, with strong support throughout the career continuum, from students to retirees. Meetings incorporate many opportunities for professional and social interactions, and attendees find the personal contacts a rewarding experience. These experiences result in building a robust network of fellow scientists and engineers, many of whom become lifelong friends and colleagues. From the society’s inception, members recognized the importance of developing acoustical standards with a focus on terminology, measurement procedures, and criteria for determining the effects of noise and vibration. The ASA Standards Program serves as the Secretariat for four American National Standards Institute Committees and provides administrative support for several international standards committees. Throughout its history to present day, ASA’s strength resides in attracting the interest and commitment of scholars devoted to promoting the knowledge and practical applications of acoustics. The unselfish activity of these individuals in the development of the society is largely responsible for ASA’s growth and present stature. v

Series Preface

Springer Handbook of Auditory Research The following preface is the one that we published in volume 1 of the Springer Handbook of Auditory Research back in 1992. As anyone reading the original preface, or the many users of the series, will note, we have far exceeded our original expectation of eight volumes. Indeed, with books published to date and those in the pipeline, we are now set for 7 volumes in SHAR. Once volume 77 is completed, we are turning the series over to new Series Editors who will carry on with additional volumes. We are very proud that there seems to be consensus, at least among our friends and colleagues, that SHAR has become an important and influential part of the auditory literature. While we have worked hard to develop and maintain the quality and value of SHAR, the real value of the books is very much because of the numerous authors who have given their time to write outstanding chapters and to our many co-editors who have provided the intellectual leadership to the individual volumes. We have worked with a remarkable and wonderful group of people, many of whom have become great personal friends of both of us. We also continue to work with a spectacular group of editors at Springer. Indeed, several of our past editors have moved on in the publishing world to become senior executives. But the truth is that the series would and could not be possible without the support of our families, and we want to take this opportunity to dedicate all of our SHAR books to them. Our wives, Catherine Fay and Helen Popper, and our children, Michelle Popper Levit, Melissa Popper Levinsohn, Christian Fay, and Amanda Fay Sierra, have been immensely patient as we developed and worked on this series. We thank them and state, without doubt, that this series could not have happened without them. We also dedicate the future of SHAR to our next generation of (potential) auditory researchers – our grandchildren – Ethan and Sophie Levinsohn, Emma Levit, Nathaniel, Evan, and Stella Fay, and Sebastian Sierra. vii

Preface 1992

The Springer Handbook of Auditory Research presents a series of comprehensive and synthetic reviews of the fundamental topics in modern auditory research. The volumes are aimed at all individuals with interests in hearing research including advanced graduate students, post-doctoral researchers, and clinical investigators. The volumes are intended to introduce new investigators to important aspects of hearing science and to help established investigators to better understand the fundamental theories and data in fields of hearing that they may not normally follow closely. Each volume presents a particular topic comprehensively, and each serves as a synthetic overview and guide to the literature. As such, the chapters present neither exhaustive data reviews nor original research that has not yet appeared in peer-­ reviewed journals. The volumes focus on topics that have developed a solid data and conceptual foundation rather than on those for which a literature is only beginning to develop. New research areas will be covered on a timely basis in the series as they begin to mature. Each volume in the series consists of a few substantial chapters on a particular topic. In some cases, the topics will be ones of traditional interest for which there is a substantial body of data and theory, such as auditory neuroanatomy (Vol. 1) and neurophysiology (Vol. 2). Other volumes in the series deal with topics that have begun to mature more recently, such as development, plasticity, and computational models of neural processing. In many cases, the series editors are joined by a co-­ editor having special expertise in the topic of the volume. SHAR logo by Mark B. Weinberg, Potomac, Maryland, used with permission Richard R. Fay*, Chicago, IL, USA Arthur N. Popper, College Park, MD, USA

Deceased

*

ix

Volume Preface

The chapters in this volume review current knowledge of the mechanisms that underly the generation of hair cells in the mature ear and discuss how such new hair cells might reestablish functional connections with the brain. Although older findings are discussed, our emphasis is on work conducted since about 2005 because much exciting work has taken place since chapters in earlier volumes of the Springer Handbook of Auditory Research. Chapter 2, by Madeleine Hewitt, David W. Raible, and Jennifer S. Stone, gives an overview of hair cell regeneration in non-mammalian vertebrates (primarily fishes and birds). They review the evidence that first led scientists to conclude that hair cells can be replaced after damage, and they detail the cellular processes by which supporting cells can give rise to new hair cells. This is followed by Chap. 3, where Mark A. Rudolf and Jeffrey T. Corwin focus on the basic biophysical and mechanical properties of supporting cells during regeneration. Together, Chaps. 2 and 3 build a foundation of understanding for the remarkable capacity for hair cell regeneration shared by nonmammalian vertebrates and establish the context for the next three chapters, which focus on the mammalian ear. In Chap. 4, Andrew Forge and Ruth Taylor give an overview of hair cell regeneration in the mammalian inner ear. The authors provide a brief review of the development and structure of the mammalian vestibular and auditory sensory epithelia as well as review the phenomenology of natural regeneration in the mature mammalian vestibular organs, how it occurs through phenotypic conversion of supporting cells, and its limitations. One frequently proposed strategy for inducing regeneration in the mature cochlea is to reactivate the signaling pathways that produce hair cells during embryonic development. In Chap. 5, Melissa M. McGovern and Andrew K. Groves describe the molecular mechanisms that govern organ of Corti development, resulting in the establishment of correct cell phenotypes, numbers, and patterning. The authors then explain how insights gleaned from development have shaped studies of auditory hair cell regeneration. Another approach to inducing repair in the mammalian inner ear involves the reactivation or surgical introduction of stem cells. Such efforts are reviewed by xi

xii

Volume Preface

Amanda Janesick, Stefan Heller, and Eri Hashino in Chap. 6. The chapter describes current strategies for generating inner ear cell types from pluripotent stem cells in vitro as well as somatic stem and progenitor cells that are present in the ears of non-mammalian vertebrates, in the adult mammalian vestibular organs, and in the cochleae of neonatal rodents. In Chap. 7, Steven H.  Green, Sepand Bafti, Benjamin M.  Gansemer, A.  Eliot Shearer, Muhammad Taifur Rahman, Mark E.  Warchol, and Marlan R.  Hansen summarize the causes of spiral ganglion neuron death and review several approaches to their possible regeneration. Finally, in Chap. 8, Brandon C. Cox, John V. Brigande, and Bradley J. Walters focus on some newer approaches to the biological restoration of hearing. They provide a detailed overview of such methodology and potential applications to otic regeneration. This volume conveys many of the critical advances in our understanding of hair cell regeneration that have occurred since the 2005 SHAR volumes. We have framed these advances in the context of our shared goal of defining ways to stimulate regeneration, and recovery of hearing and balance function, in humans. In addition, this volume discusses many of the cutting-edge approaches that are being applied, or soon will be applied, to hasten progress toward this goal. Mark E. Warchol, St. Louis, MO, USA   Jennifer S. Stone, Seattle, WA, USA   Allison B. Coffin, Vancouver, WA, USA   Richard R. Fay, Chicago, IL, USA   Arthur N. Popper, College Park, MD, USA

Contents

1

Sensory Regeneration in the Inner Ear: History, Strategies, and Prospects����������������������������������������������������������    1 Mark E. Warchol and Jennifer S. Stone

2

Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological and Functional Recovery������������������������������������������������������������������������   11 Madeleine N. Hewitt, David W. Raible, and Jennifer S. Stone

3

 Cell Junctions and the Mechanics of Hair Cell Regeneration ������������   41 Mark A. Rudolf and Jeffrey T. Corwin

4

 Mammalian Hair Cell Regeneration������������������������������������������������������   73 Ruth Taylor and Andrew Forge

5

Specification and Plasticity of Mammalian Cochlear Hair Cell Progenitors������������������������������������������������������������������������������  105 Melissa M. McGovern and Andrew K. Groves

6

Inner Ear Cells from Stem Cells: A Path Towards Inner Ear Cell Regeneration������������������������������������������������������������������  135 Amanda Janesick, Eri Hashino, and Stefan Heller

7

Spiral Ganglion Neuron Regeneration in the Cochlea: Regeneration of Synapses, Axons, and Cells ����������������������������������������  163 Steven H. Green, Sepand Bafti, Benjamin M. Gansemer, A. Eliot Shearer, Muhammad Taifur Rahman, Mark E. Warchol, and Marlan R. Hansen

8

Genetic and Epigenetic Strategies for Promoting Hair Cell Regeneration in the Mature Mammalian Inner Ear ����������  195 Brandon C. Cox, John V. Brigande, and Bradley J. Walters

xiii

Contributors

Sepand Bafti  Nortis Inc., Woodinville, WA, USA John  V.  Brigande  Oregon Hearing Research Center, Oregon Health & Science University, Portland, OR, USA Jeffrey T. Corwin  Department of Neuroscience, University of Virginia School of Medicine, Charlottesville, VA, USA Department of Cell Biology, University of Virginia School of Medicine, Charlottesville, VA, USA Brandon  C.  Cox  Southern Springfield, IL, USA

Illinois

University

School

of

Medicine,

Andrew Forge  UCL Ear Institute, London, UK Benjamin  M.  Gansemer  Department of Biology, University of Iowa, Iowa City, IA, USA Steven H. Green  Department of Biology, University of Iowa, Iowa City, IA, USA Andrew  K.  Groves  Department of Neuroscience, Baylor College of Medicine, Houston, TX, USA Department of Human and Molecular Genetics, Baylor College of Medicine, Houston, TX, USA Marlan  R.  Hansen  Department of Otolaryngology  – Head & Neck Surgery, University of Iowa, Iowa City, IA, USA Eri  Hashino  Department of Otolaryngology – Head & Neck Surgery, Indiana University School of Medicine, Indianapolis, IN, USA Stefan Heller  Department of Otolaryngology – Head & Neck Surgery, Stanford University School of Medicine, Stanford, CA, USA Institute for Stem Cell Biology and Regenerative Medicine, Stanford University School of Medicine, Stanford, CA, USA xv

xvi

Contributors

Madeleine  N.  Hewitt  Molecular and Cellular Biology Graduate Program, University of Washington School of Medicine, Seattle, WA, USA Department of Biological Structure, University of Washington School of Medicine, Seattle, WA, USA Department of Otolaryngology/Head and Neck Surgery, University of Washington School of Medicine, Seattle, WA, USA Amanda  Janesick  Department of Otolaryngology–Head & Neck Surgery, Stanford University School of Medicine, Stanford, CA, USA Institute for Stem Cell Biology and Regenerative Medicine, Stanford University School of Medicine, Stanford, CA, USA Melissa M. McGovern  Department of Neuroscience, Baylor College of Medicine, Houston, TX, USA Muhammad  Taifur  Rahman  Department of Biology, University of Iowa, Iowa City, IA, USA David W. Raible  Molecular and Cellular Biology Graduate Program, University of Washington School of Medicine, Seattle, WA, USA Department of Biological Structure, University of Washington School of Medicine, Seattle, WA, USA Department of Otolaryngology/Head and Neck Surgery, University of Washington School of Medicine, Seattle, WA, USA Mark  A.  Rudolf  Department of Neuroscience, University of Virginia School of Medicine, Charlottesville, VA, USA Department of Cell Biology, University of Virginia School of Medicine, Charlottesville, VA, USA A.  Eliot  Shearer  Department of Otolaryngology, Boston Children’s Hospital, Harvard Medical School, Boston, MA, USA Jennifer  S.  Stone  Department of Otolaryngology/Head and Neck Surgery, Virginia Merrill Bloedel Hearing Research Center, University of Washington School of Medicine, Seattle, WA, USA Ruth Taylor  UCL Ear Institute, London, UK Bradley J. Walters  University of Mississippi Medical Center, Jackson, MS, USA Mark E. Warchol  Department of Otolaryngology, Washington University School of Medicine, St. Louis, MO, USA

Chapter 1

Sensory Regeneration in the Inner Ear: History, Strategies, and Prospects Mark E. Warchol and Jennifer S. Stone

Abstract  Hair cells are the sensory receptors of the vertebrate inner ear that detect sound vibrations (to initiate hearing) and head movements (to facilitate balance). In humans, injury or loss of these cells can cause permanent hearing impairment, disequilibrium, and/or improper visual reflexes. For this reason, the development of methods to promote the regeneration of lost hair cells is a topic of great translational interest. The demonstration that the ears of nonmammalian vertebrates can regenerate hair cells after injury has led to considerable research focused on understanding the biological basis of this regenerative process. Other work has attempted to induce repair in the mammalian inner ear via gene therapy or introduction of stem cells. Such research is still ongoing, and it is not clear which approaches will ultimately yield clinical strategies for restoring hearing and motion perception. This chapter provides a brief history of this discipline and summarizes the contents of this volume. Keywords  Hair cell · Cochlea · Vestibular · Basilar papilla · Acoustic trauma Ototoxicity · Regeneration

1.1 Introduction The inner ear is a complex region of the body that houses specialized end organs for the auditory and vestibular senses. Injury or pathology can affect numerous cell types in the inner ear, ultimately leading to hearing loss or disorders of balance and equilib­rium. Hair cells are one of the most critical elements of the auditory and M. E. Warchol (*) Department of Otolaryngology, Washington University School of Medicine, St. Louis, MO, USA e-mail: [email protected] J. S. Stone Department of Otolaryngology/Head and Neck Surgery, Virginia Merrill Bloedel Hearing Research Center, University of Washington School of Medicine, Seattle, WA, USA e-mail: [email protected] © Springer Nature Switzerland AG 2023 M. E. Warchol et al. (eds.), Hair Cell Regeneration, Springer Handbook of Auditory Research 75, https://doi.org/10.1007/978-3-031-20661-0_1

1

2

M. E. Warchol and J. S. Stone

vestibular sensory systems. They detect mechanical motion generated by either sound vibrations or head movements and provide synaptic input to the afferent neurons that convey information to the brain. In humans, the loss of hair cells is the leading cause of sensorineural deafness. The death of hair cells can also be an underlying cause of vestibular dysfunction. The focus of the present volume is on biological pro­cesses that regulate the regeneration of hair cells after injury. Although the mammalian ear has a very limited ability for regeneration, the ears of nonmammalian vertebrates are capable of spontaneously regenerating hair cells and such regenera­tion is accompanied by functional recovery (reviewed by Warchol 2011; Burns and Corwin 2013). Hair cell-containing organs of some vertebrates also show evidence for cell turnover, in which hair cells die under apparently normal conditions (i.e., without damaging stimuli or agents), and new hair cells are then formed to replace them. Such turnover has been demonstrated to occur in lateral line neuromasts of fishes (Williams and Holder 2000; Cruz et al. 2015) and in the vestibular organs of birds and mammals (Jørgensen and Mathiesen 1988; Kirkegaard and Jørgensen 2000; Bucks et al. 2017) and may yield insights into the possibility for regen­eration after injury. This book reviews current research on sensory regeneration in the inner ear, including some of the key questions under investigation, the scientific approaches that are being applied, and the insights that have been gleaned. It has been roughly 15 years since hair cell regeneration was covered by this book series and, during that time, the field has evolved considerably. We have invited some of the leading investigators in the field to describe the progress that has been made across a range of research topics. This book should serve to update seasoned investigators and to help early-stage scientists establish a foundation of knowledge on hair cell regeneration and the factors controlling hair cell development, hair cell innervation by spiral ganglion neurons, and other topics closely connected to hair cell regeneration. The authors provide insights into how research progress has shaped the questions we are now asking, and how technological advances have enabled more sophisticated and efficient approaches to address these questions. Research into the possibility of promoting regeneration in the inner ear was first stimulated by the observations that the ears of fishes and amphibians continue to produce new hair cells throughout mature life and that birds can spontaneously regenerate hair cells after injury (see Chap. 2). Many efforts are aimed at inducing similar forms of biological repair in the human ear. Such work is still in its early stages, and there is no consensus as to which particular approach will ultimately lead to restoration of inner ear function. For this reason, the present volume contains reviews of several differing conceptual and methodological strategies for promoting regeneration in the cochlea and vestibular organs. One straightforward approach to the development of regenerative therapies is to identify the genetic and molecular signals responsible for such regeneration in nonmammals and to then use such information to guide methods for inducing similar forms of repair in the ears of mammals. A related approach is inspired by the observation that, during embryonic development, the mammalian ear is capable of some degree of self-repair but this ability is lost during maturation (reviewed in Chap. 4).

1  Sensory Regeneration in the Inner Ear: History, Strategies, and Prospects

3

Identifying the changes in gene expression and cell signaling that accompany this transition may reveal which signaling pathways are permissive or inhibitory for regeneration. There is also much interest in the use of stem cell technologies to induce repair in the ear, either by activating endogenous stem cells that might remain in the mature ear or by transplantation of stem cells that have been guided toward becoming replacement sensory receptors (e.g., Chap. 6). Finally, new methods for gene therapy offer the potential to reprogram cells within the damaged ear to undergo self-repair or to change phenotype into new sensory receptors (see Chap. 8). A major objective in assembling the present volume was to provide a thorough review of the scientific issues relevant for this field of study and background information on the various approaches currently being employed.

1.2 Historical Overview of Otic Regeneration 1.2.1 Postnatal Generation of Sensory Receptors in Vertebrates A pioneering study of the developing mouse inner ear, conducted in the 1960s by Robert Ruben, had demonstrated that the proliferation of hair cell precursors terminates at around the time of birth (Ruben 1967). This finding suggested that all mammalian hair cells were produced during embryogenesis and implied that the ear may lack the ability to generate new hair cells later in life. In contrast, studies of other sensory modalities indicated that, in some circumstances, other types of receptor cells could be produced throughout life. For example, adult mammals can regenerate neurons of the olfactory epithelium and olfactory bulb (Altman 1969; Graziadei 1979; Graziadei and Monti Graziadei 1985). Gustatory receptors also exhibit a normal turnover throughout life and can be replaced after injury or loss (Farbman 1980). When considering these observations, it is notable that the sensory receptors for both taste and smell are directly exposed to the external environment, making them particularly vulnerable to damage, and it is possible that their capacity for renewal and regeneration is a consequence of evolutionary pressure to maintain proper sensory function. However, not all generation of sensory cells in mature animals appears to be linked to injury and repair. The retinae of fishes and amphibians, for example, continue to add photoreceptors throughout life. These animals also continue to grow throughout mature life (albeit at increasingly slower rates with age), and such cell addition is common in many organ systems as they grow larger. In the case of the fish retina, addition of new photoreceptors parallels the overall growth of the animal (Johns 1982) and results in slightly enhanced visual acuity with age (Fernald 1990). It is further notable that the growing retina also adds new ganglion cells (Johns and Easter 1977), which extend axons to the optic tectum. Age-related growth and addition of neurons also occurs in the tectum so that the increase in retinal input is matched by a corresponding increase in the numbers of target cells in the CNS (Meyer 1978; Raymond and Easter 1983).

4

M. E. Warchol and J. S. Stone

1.2.2 Postnatal Addition of Hair Cells in Cold-Blooded Vertebrates Several classic studies demonstrated that hair cells in the lateral line neuromasts of aquatic vertebrates could be produced or replaced throughout life. For example, certain salamanders show a remarkable ability to regenerate lost body parts, such as limbs or tails (e.g., Brockes and Kumar 2008). Studies of axolotls, conducted in the 1930s, demonstrated that amputation of distal tail tips that contained a lateral line neuromast resulted in a regenerated tail segment that also contained a newly formed neuromast (Stone 1937). Such observations might have been interpreted as evidence for the possibility of hair cell regeneration, but the generation of new neuromasts in salamanders occurs as part of a more global regenerative process, and it was not clear how – or whether – these findings could be applied to the sensory receptors of inner ear. The potential for postnatal hair cell production in the vertebrate inner ear was first demonstrated by Jeffrey Corwin, who showed a dramatic age-related increase in hair cell numbers in elasmobranchs. Quantitative studies demonstrated that the maculae of young charcharinid sharks contained about 40,000 hair cells, while the number of hair cells in the maculae of mature animals could exceed 200,000 (Corwin 1977, 1981). Later, morphological observations by Li and Lewis (1979) suggested that the inner ears of bullfrogs (Rana catesbeiana) might also add new hair cells during maturity, and postembryonic addition of hair cells was subsequently demonstrated in other cold-blooded vertebrates, including rays (Raja clavata; Corwin 1983), teleost fish (Astronotus ocellatus; Popper and Hoxter 1984), and toads (Bufo marinus; Corwin 1985).

1.2.3 Sensory Regeneration in the Avian Inner Ear Together, these earlier studies demonstrated that cold-blooded vertebrates continue to produce new sensory hair cells well into maturity, but they did not demonstrate regeneration per se. Injury-induced production of new hair cells was first observed in the hearing organ of birds, known as the basilar papilla. This sensory organ shares many similarities with the mammalian cochlea in that it contains hair cells that are arranged in a tonotopically organized array, resting upon a basilar membrane that vibrates in response to sound input (reviewed by Fettiplace 2020). Like many scientific advances, the demonstration of hair cell regeneration in the avian ear occurred through an indirect route. Frequency tuning in the basilar papilla is accomplished (in part) by a traveling wave that propagates along the basilar membrane, with a locus of peak displacement that varies in a tonotopic fashion (von Békésy 1960). It had further been suggested that the frequency map of the basilar papillae underwent a systematic shift during development, such that the basal/proximal regions initially detect low-frequency sounds and then become best sensitive to

1  Sensory Regeneration in the Inner Ear: History, Strategies, and Prospects

5

higher frequencies as the ears matured (e.g., Rubel 1978). Evidence for such remapping was first provided by Rubel and Ryals (1983), who showed that the position of maximal damage caused by exposure to a high-intensity pure tone shifted along the basilar membrane, from proximal to distal, during maturation. In addition, the neural frequency map within the auditory brainstem (which is determined by the pattern of afferent input from the cochlea) also undergoes a similar shift during this same developmental period (Lippe and Rubel 1983). Age-related changes in susceptibility to noise injury were further studied by Cotanche and colleagues, who used scanning electron microscopy to image the patterns of hair cell loss across the surface of the basilar papilla after exposure to intense pure tones (Cotanche et al. 1988). In a subsequent study, Cotanche quantified the loci of such lesions both immediately following the noise exposure, and also after several days of recovery. As expected, papillae that were examined shortly after the sound exposure contained large regions that were devoid of hair cells. But after several days of recovery, these lesioned regions were found to contain numerous hair cells that possessed morphological features characteristic of immature hair cells (Cotanche 1987). This finding constituted strong and dramatic evidence that the injured hair cells had been replaced. At nearly the same time, Cruz, Lambert, and Rubel were studying patterns of aminoglycoside ototoxicity in the chick inner ear. In mammals, systemic treatment with gentamicin permanently destroys auditory hair cells. However, at several weeks after such treatments, in chicks, Cruz et al. (1987) found that hair cells had apparently recovered in the basilar papilla. The demonstration that the avian inner ear could regenerate hair cells triggered great interest among auditory researchers but also raised many questions. Were new hair cells being produced or did damaged hair cells undergo some form of repair? Was cell division occurring in the injured papillae or were other cells converting into a hair cell phenotype? These issues were partly resolved by parallel studies that showed evidence for the injury-evoked proliferation of supporting cells and the subsequent differentiation of some of the progeny as replacement hair cells (Corwin and Cotanche 1988; Ryals and Rubel 1988). A similar form of hair cell regeneration was also demonstrated to occur in the vestibular organs of birds, both after spontaneous hair cell death (Jørgensen and Mathiesen 1988) and after injury by ototoxic antibiotics (Weisleder and Rubel 1993). In addition, subsequent studies found that renewed proliferation was not the only mechanism that can generate new hair cells after injury. In some case, supporting cells can undergo a direct change of phenotype, thereby becoming hair cells without first dividing (e.g., Adler and Raphael 1996; Baird et al. 2000). Finally, evidence emerged for a limited – but demonstrable – recovery of hair cells in the vestibular organs of mammals (Forge et al. 1993), which was accompanied by a low level of supporting cell proliferation (Warchol et al. 1993). These findings motivated numerous efforts to determine whether the mammalian cochlea also possessed some degree of regenerative ability. The outcomes of such studies were uniformly negative (e.g., Roberson and Rubel 1994), and it is currently believed that the mature mammalian cochlea lacks any ability to replace sensory cells after injury.

6

M. E. Warchol and J. S. Stone

1.3 Overview of Contents The chapters in this volume review current knowledge of the mechanisms that underlie the generation of hair cells in the mature ear and discuss how these new hair cells might reestablish functional connections with the brain. Although older findings are discussed, our emphasis is on work conducted since 2008. The second chapter, Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological and Functional Recovery, by Madeleine Hewitt, David W. Raible, and Jennifer Stone, gives an overview of hair cell regeneration in nonmammalian vertebrates (primarily fishes and birds). They review the evidence that first led scientists to conclude that hair cells can be replaced after damage, and they detail the cellular processes by which supporting cells can give rise to new hair cells. This chapter also describes progress in understanding how hair cell regeneration is controlled at the molecular level, such as which cellular signaling systems, transcription factors, and epigenetic regulators trigger supporting cells to form new hair cells after damage and promote precursor cells to acquire the hair cell-specific features required to restore function. The third chapter, Cell Junctions and the Mechanics of Hair Cell Regeneration, by Mark A.  Rudolf and Jeffrey T.  Corwin, focuses on the basic biophysical and mechanical properties of supporting cells. The morphology of supporting cells is significantly altered by the loss of hair cells, and such mechanical cues may be a key factor that either triggers or limits the induction of regeneration. Supporting cells in the ears of nonmammals (i.e., fishes, amphibians, and birds) undergo shape changes in response to loss of hair cells, and the extension of these cells may enable them to leave the quiescent state and divide. In contrast, supporting cells of the mammalian ear possess large cytoskeletal elements that have been hypothesized to limit their proliferative ability. In other cell types, such mechanical factors are known to activate the Hippo and Wnt pathways, which regulate cell division. As such, it is likely that cellular mechanics is one factor that may help explain the differences in the regenerative ability of the ears of nonmammals vs. mammals. Together, Chaps. 2 and 3 build a foundation of understanding for the remarkable capacity for hair cell regeneration shared by nonmammalian vertebrates and establish the context for the next three chapters, which focus on the mammalian ear. In the fourth chapter, Mammalian Hair Cell Regeneration, Andrew Forge and Ruth Taylor give an overview of hair cell regeneration in the mammalian inner ear. As noted above, the mammalian cochlea does not regenerate hair cells, and only a limited amount of regeneration occurs in the vestibular organs. Current research in this area is directed toward uncovering a possible latent potential for hair cell regeneration, either by negating putative signals that might prevent spontaneous regeneration or by inducing pro-regenerative responses. The authors give a brief review of the development and structure of the mammalian vestibular and auditory sensory epithelia. Studies have shown that the immature inner ears of mammals possess some ability to regenerate but that this capacity is lost with maturation. Much current research is aimed at identifying factors that are permissive for a regenerative

1  Sensory Regeneration in the Inner Ear: History, Strategies, and Prospects

7

response in immature ear but limit regeneration in mature ears. The authors also review the phenomenology of natural regeneration in the mature mammalian vestibular organs, how it occurs through phenotypic conversion of supporting cells, and its limitations. They further describe current approaches to enhance or stimulate hair cell regeneration in adult mammals and the outcomes of such attempts. Finally, the authors review the structural reorganization that occurs in the mature organ of Corti after hair cell damage and how it may impact possible regenerative strategies. One frequently proposed strategy for inducing regeneration in the mature cochlea is to reactivate the signaling pathways that produce hair cells during embryonic development. In Chap. 5, Specification and Plasticity of Mammalian Cochlear Hair Cell Progenitors, Melissa M. McGovern and Andrew K. Groves describe the molecular mechanisms that govern organ of Corti development, resulting in the establishment of correct cell phenotypes, numbers, and patterning. The authors then explain how insights gleaned from development have shaped studies of auditory hair cell regeneration. Of particular interest is the identification of signals that might suppress pro-regenerative behaviors by supporting cells in the mature cochlea after damage. They further discuss evidence that age-dependent modifications of chromatin might block the activation of genetic programs that have the potential to drive hair cell regeneration. This chapter highlights how an understanding of the mechanisms that regulate the phenotype and proliferation of supporting cells is a critical step toward the development of regenerative therapies. Another approach to inducing repair in the mammalian inner ear involves either the reactivation or surgical introduction of stem cells. Such efforts are reviewed by Amanda Janesick, Stefan Heller, and Eri Hashino in Chap. 6, Inner Ear Cells from Stem Cells – a Path Toward Inner Ear Cell Regeneration. The investigation of otic regeneration has been limited by the difficulty in accessing the sensory organs of the inner ear and by the relatively small numbers of hair cells and supporting cells that are present in the mammalian ear. Large numbers of cells are necessary for employment of techniques such as high-throughput in vitro drug screens to identify pro-­ regenerative drugs and gene therapies with the potential to promote regeneration. This chapter describes current strategies for generating inner ear cell types from pluripotent stem cells in vitro. Also discussed are somatic stem and progenitor cells that are present in the ears of nonmammalian vertebrates, in the adult mammalian vestibular organs, and in the cochleae of neonatal rodents. Extracting knowledge from these naturally proliferative cell populations may lead us closer to the possibility of reawakening regenerative potential in the ears of adult mammals. Hair cells are not the only type of sensory cell that can be damaged or lost from the mature ear. The neurons of the spiral ganglion convey information from cochlear hair cells to the nuclei of the auditory brainstem. These neurons can be injured or lost by many of the same insults that kill cochlear hair cells. Moreover, the loss of spiral ganglion neurons (SGNs) can also occur after exposure to noise levels that do not kill hair cells (Kujawa and Liberman 2006, 2009). SGNs are essential for both normal hearing and for the restoration of hearing provided by cochlear implants. In Chap. 7, Spiral Ganglion Neuron Regeneration in the Cochlea: Regeneration of Synapses, Axons, and Cells, Steven H. Green, Sepand

8

M. E. Warchol and J. S. Stone

Bafti, Benjamin M. Gansemer, A. Eliot Shearer, Muhammad Taifur Rahman, Mark E. Warchol, and Marlan R. Hansen summarize the causes of SGN death and also review several approaches to their possible regeneration. The degeneration of SGNs often occurs slowly, and it is possible that restoring their synaptic connections with hair cells may prevent their loss. Current research emphasizes the use of certain neurotrophic factors to promote such regrowth. However, it is also possible that SGNs can be replaced by transplantation of new neurons derived from stem cells. The chapter discusses current efforts to generate SGNs from stem cells and the challenges facing these attempts. The final chapter, Genetic and Epigenetic Strategies for Promoting Hair Cell Regeneration in the Mature Mammalian Inner Ear, by Brandon C.  Cox, John V. Brigande, and Bradley J. Walters, focuses on some newer approaches to the biological restoration of hearing. The development of successful strategies for inducing regeneration will likely involve the genetic manipulation of experimental animals, in order to either delete or overexpression of certain genes or to introduce novel genetic constructs into cells of the inner ear. Chapter 8 provides a detailed overview of such methodology and potential applications to otic regeneration. Inducing repair in the mature inner ear will be a complex process, requiring the manipulation of multiple genes and/or alterations of epigenetic modifiers. The chapter reviews techniques such as CRISPR/Cas and CreER/lox methodology. Complementary strategies using virus-mediated gene transfer, small interfering RNA, antisense oligonucleotides, and microRNAs are also highlighted.

1.4 Conclusions As noted earlier, the investigation of hair cell regeneration is currently proceeding along several distinct and parallel paths. We hope that readers will find this book to be a useful resource for initiating or continuing research into sensory regeneration and that it will enhance their understanding of the rationale, potential, and limitations of the current approaches and methodologies. We further hope that this information will stimulate researchers to uncover new facts and develop more effective techniques that will guide this field toward its ultimate objective: the biological repair of the human inner ear.

References Adler HJ, Raphael Y (1996) New hair cells arise from supporting cell conversion in the acoustically damaged chick inner ear. Neurosci Lett 205:17–20. https://doi. org/10.1016/0304-­3940(96)12367-­3 Altman J (1969) Autoradiographic and histological studies of postnatal neurogenesis. IV.  Cell proliferation and migration in the anterior forebrain, with special reference to persisting

1  Sensory Regeneration in the Inner Ear: History, Strategies, and Prospects

9

neurogenesis in the olfactory bulb. J Comp Neurol 137:433–457. https://doi.org/10.1002/ cne.901370404. PMID: 5361244 Baird RA, Burton MD, Lysakowski A, Fashena DS, Naeger RA (2000) Hair cell recovery in mitotically blocked cultures of the bullfrog saccule. Proc Natl Acad Sci U S A. 97(22):11722–11729. https://doi.org/10.1073/pnas.97.22.11722 Brockes JP, Kumar A (2008) Comparative aspects of animal regeneration. Annu Rev Cell Dev Biol 24:525–549. https://doi.org/10.1146/annurev.cellbio.24.110707.175336. PMID: 18598212 Bucks SA, Cox BC, Vlosich BA, Manning JP, Nguyen TB, Stone JS (2017) Supporting cells remove and replace sensory receptor hair cells in a balance organ of adult mice. 6:e18128. https://doi.org/10.7554/eLife.18128. PMID: 28263708; PMCID: PMC5338920 Burns JC, Corwin JT (2013) A historical to present-day account of efforts to answer the question: “what puts the brakes on mammalian hair cell regeneration?”. Hear Res. 297:52–67. https:// doi.org/10.1016/j.heares.2013.01.005. Epub 2013 Jan 17. Corwin JT (1977) Morphology of the macula neglecta in sharks of the genus Carcharhinus. J Morphol 152:341–362. https://doi.org/10.1002/jmor.1051520306. PMID: 875041 Corwin JT (1981) Postembryonic production and aging in inner ear hair cells in sharks. J Comp Neurol 201:541–553. https://doi.org/10.1002/cne.902010406. PMID: 7287934 Corwin JT (1983) Postembryonic growth of the macula neglecta auditory detector in the ray, Raja clavata: continual increases in hair cell number, neural convergence, and physiological sensitivity. J Comp Neurol 217:345–356. https://doi.org/10.1002/cne.902170309. PMID: 6886057 Corwin JT (1985) Perpetual production of hair cells and maturational changes in hair cell ultrastructure accompany postembryonic growth in an amphibian ear. Proc Natl Acad Sci U S A 82:3911–3915. https://doi.org/10.1073/pnas.82.11.3911. PMID: 3923484; PMCID: PMC397899 Corwin JT, Cotanche DA (1988) Regeneration of sensory hair cells after acoustic trauma. Science 240:1772–1774. https://doi.org/10.1126/science.3381100. PMID: 3381100 Cotanche DA (1987) Regeneration of hair cell stereociliary bundles in the chick cochlea following severe acoustic trauma. Hear Res 30:181–195. https://doi.org/10.1016/0378-­5955(87)90135-­3. PMID: 3680064 Cotanche DA, Saunders JC, Tilney LG (1988) Hair cell damage produced by acoustic trauma in the chick cochlea. Hear Res 25:267–286. https://doi.org/10.1016/0378-­5955(87)90098-­0. PMID: 3558135 Cruz RM, Lambert PR, Rubel EW (1987) Light microscopic evidence of hair cell regeneration after gentamicin toxicity in chick cochlea. Arch Otolaryngol Head Neck Surg 113:1058–1062. https://doi.org/10.1001/archotol.1987.01860100036017. PMID: 3620125 Cruz IA, Kappedal R, Mackenzie SM, Hailey DW, Hoffman TL, Schilling TF, Raible DW (2015) Robust regeneration of adult zebrafish lateral line hair cells reflects continued precursor pool maintenance. Dev Biol. 402(2):229–238. https://doi.org/10.1016/j.ydbio.2015.03.019. Epub 2015 Apr 11. PMID: 25869855; PMCID: PMC4450121 Farbman AI (1980) Renewal of taste bud cells in rat circumvallate papillae. Cell Tissue Kinet 13:349–357. https://doi.org/10.1111/j.1365-­2184.1980.tb00474.x. PMID: 7428010 Fernald RD (1990) Teleost vision: Seeing while growing. J Exp Zool Suppl 5:167–180. https://doi. org/10.1002/jez.1402560521. PMID: 1982493 Fettiplace R (2020) Diverse mechanisms of sound frequency discrimination in the vertebrate cochlea. Trends Neurosci 43:88–102. https://doi.org/10.1016/j.tins.2019.12.003. Epub 2020 Jan 15. PMID: 31954526; PMCID: PMC7015066 Forge A, Li L, Corwin JT, Nevill G (1993) Ultrastructural evidence for hair cell regeneration in the mammalian inner ear. Science 259:1616–1619. https://doi.org/10.1126/science.8456284. PMID: 8456284 Graziadei GA, Graziadei PP (1979) Neurogenesis and neuron regeneration in the olfactory system of mammals. II. Degeneration and reconstitution of the olfactory sensory neurons after axotomy. J Neurocytol 8:197–213. https://doi.org/10.1007/BF01175561. PMID: 469573 Graziadei PP, Monti Graziadei GA (1985) Neurogenesis and plasticity of the olfactory sensory neurons. Ann N Y Acad Sci 457:127–142. https://doi.org/10.1111/j.1749-­6632.1985. tb20802.x. PMID: 3913359

10

M. E. Warchol and J. S. Stone

Johns PR (1982) Formation of photoreceptors in larval and adult goldfish. J Neurosci 2:178–198. https://doi.org/10.1523/JNEUROSCI.02-­02-­00178.1982. PMID: 7062105; PMCID: PMC6564304 Johns PR, Easter SS (1977) Growth of the adult goldfish eye. II. Increase in retinal cell number. J Comp Neurol 176:331–341. https://doi.org/10.1002/cne.901760303. PMID: 915041 Jørgensen JM, Mathiesen C (1988) The avian inner ear. Continuous production of hair cells in vestibular sensory organs, but not in the auditory papilla. Naturwissenschaften 75:319–320. https://doi.org/10.1007/BF00367330. PMID: 3205314 Kirkegaard M, Jørgensen JM (2000) Continuous hair cell turnover in the inner ear vestibular organs of a mammal, the Daubenton’s bat (Myotis daubentonii). Naturwissenschaften. 87(2):83–86. https://doi.org/10.1007/s001140050015. PMID: 10663140 Kujawa SG, Liberman MC (2006) Acceleration of age-related hearing loss by early noise exposure: evidence of a misspent youth. J Neurosci 26(7):2115–2123. https://doi.org/10.1523/ JNEUROSCI.4985-05.2006 Kujawa SG, Liberman MC (2009) Adding insult to injury: cochlear nerve degeneration after “temporary” noise-induced hearing loss. J Neurosci 29(45):14077–14085. https://doi.org/10.1523/ JNEUROSCI.2845-09.2009 Li CW, Lewis ER (1979) Structure and development of vestibular hair cells in the larval bullfrog. Ann Otol Rhinol Laryngol 88:427–437. https://doi.org/10.1177/000348947908800323. PMID: 313736 Lippe W, Rubel EW (1983) Development of the place principle: tonotopic organization. Science 219:514–516. https://doi.org/10.1126/science.6823550. PMID: 6823550 Meyer RL (1978) Evidence from thymidine labeling for continuing growth of retina and tectum in juvenile goldfish. Exp Neurol 59:99–111. https://doi.org/10.1016/0014-­4886(78)90204-­2 Popper AN, Hoxter B (1984) Growth of a fish ear: 1. Quantitative analysis of hair cell and ganglion cell proliferation. Hear Res 15(2):133–142. https://doi.org/10.1016/0378-­5955(84)90044-­3 Raymond PA, Easter SS (1983) Postembryonic growth of the optic tectum in goldfish. I. Location of germinal cells and numbers of neurons produced. J Neurosci 3:1077–1091. https://doi. org/10.1523/JNEUROSCI.03-­05-­01077.1983. PMID: 6842282; PMCID: PMC6564515 Roberson DW, Rubel EW (1994) Cell division in the gerbil cochlea after acoustic trauma. Am J Otol 15(1):28–34. Rubel EW (1978) Ontogeny of Structure and Function in the Vertebrate Auditory System. In: Jacobson M (ed) Development of Sensory Systems. Handbook of Sensory Physiology, vol 9. Springer, Berlin, Heidelberg Rubel EW, Ryals BM (1983) Development of the place principle: acoustic trauma. Science 219:512–514. https://doi.org/10.1126/science.6823549 Ruben RJ (1967) Development of the inner ear of the mouse: a radioautographic study of terminal mitoses. Acta Otolaryngol: Suppl 220:1–44 Ryals BM, Rubel EW (1988) Hair cell regeneration after acoustic trauma in adult Coturnix quail. Science 240:1774–1776. https://doi.org/10.1126/science.3381101 Stone LS (1937) Further experimental studies of the development of lateral-line sense organs in the amphibians observed in living preparations. J Comp Neurol 68:83–115 Von Békésy G (1960) Experiments in hearing. McGraw Hill, New York Warchol ME (2011) Sensory regeneration in the vertebrate inner ear: differences at the levels of cells and species. Hear Res 273(1–2):72–79. https://doi.org/10.1016/j.heares.2010.05.00 Warchol ME, Lambert PR, Goldstein BJ, Forge A, Corwin JT (1993) Regenerative proliferation in inner ear sensory epithelia from adult guinea pigs and humans. Science 259:1619–1622. https://doi.org/10.1126/science.8456285 Weisleder P, Rubel EW (1993) Hair cell regeneration after streptomycin toxicity in the avian vestibular epithelium. J Comp Neurol 331:97–110. https://doi.org/10.1002/cne.903310106 Williams JA, Holder N (2000) Cell turnover in neuromasts of zebrafish larvae. Hear Res 143(1–2):171–181. https://doi.org/10.1016/s0378-5955(00)00039-3

Chapter 2

Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological and Functional Recovery Madeleine N. Hewitt, David W. Raible, and Jennifer S. Stone

Abstract  This chapter provides an overview of hair cell regeneration in nonmammalian vertebrates. First, we review the early foundational research on hair cell replacement. Then, we discuss research in the past 15 years that underlies our current understanding of the mechanistic basis of hair cell regeneration, including the identity, properties, and epithelial locations of supporting cells, the progenitors to new hair cells. We also describe some of the new approaches used to study hair cell regeneration, and we discuss molecules that have been found to either block or drive supporting cells to generate new hair cells after damage. This chapter focuses on chickens and zebrafish, which are the most commonly used animal models for this type of research. Keywords  Hair cell · Supporting cell · Regeneration · Molecular regulation Zebrafish · Chicken

M. N. Hewitt · D. W. Raible Molecular and Cellular Biology Graduate Program, University of Washington School of Medicine, Seattle, WA, USA Department of Biological Structure, University of Washington School of Medicine, Seattle, WA, USA Department of Otolaryngology/Head and Neck Surgery, University of Washington School of Medicine, Seattle, WA, USA e-mail: [email protected]; [email protected] J. S. Stone (*) Department of Otolaryngology/Head and Neck Surgery, Virginia Merrill Bloedel Hearing Research Center, University of Washington School of Medicine, Seattle, WA, USA e-mail: [email protected] © Springer Nature Switzerland AG 2023 M. E. Warchol et al. (eds.), Hair Cell Regeneration, Springer Handbook of Auditory Research 75, https://doi.org/10.1007/978-3-031-20661-0_2

11

12

M. N. Hewitt et al.

2.1 Introduction Studies conducted in the 1980s revealed that nonmammalian vertebrates have a remarkable ability to regenerate the sensory hair cells of the inner ear. Subsequent work has unveiled some of the cellular and molecular mechanisms that underlie this process and lead to restoration of hearing and balance function after hair cell injury. These studies advanced our fundamental understanding of how regeneration can occur in a highly specialized tissue. They also provided important clues as to why hair cell regeneration fails to occur in the auditory organ of mammals and why it is highly limited in the mammalian vestibular organs. This chapter reviews advances in our understanding of hair cell regeneration in nonmammalian vertebrates. The focus is on studies published since 2008, when the last version of the Springer Handbook of Auditory Research (SHAR) books on regeneration was released (Oesterle and Stone 2008). Because most of these studies in the past 14 years have examined chickens (Gallus gallus) and zebrafish (Danio rerio), our chapter focuses on those organisms. However, much of this work builds upon studies of postembryonic hair cell production in other nonmammals, such as sharks and goldfish (Corwin 1981; Presson et al. 1996). To learn about studies on nonmammalian hair cell regeneration prior to 2008, we recommend the following reviews: BerminghamMcDonogh and Rubel 2003; Stone and Cotanche 2007; Breuskin et  al. 2008; Brignull et  al. 2009; Ma and Raible 2009; Cotanche and Kaiser 2010; Warchol 2011; Rubel et al. 2013; Ryals et al. 2013; and Kelley and Stone 2016. Mammalian regeneration is discussed in Forge and Taylor, Chap. 4, and McGovern and Groves, Chap. 5.

2.2 Nonmammalian Hair Cell Regeneration: An Overview In the late 1980s, several key studies sparked the exploration of hair cell regeneration as a way to restore function following hair cell injury and death. Initial papers presented morphological evidence that hair cells are replaced after damage to the chicken basilar papilla (BP; the avian hearing organ and analogue of the mammalian cochlea). These studies were conducted with early post-hatch chicks, in which the morphology and function of the auditory organ is largely mature (e.g., Tilney et al. 1992; Jones et al. 2006a). Cruz et al. (1987) used the aminoglycoside ototoxin gentamicin to induce hair cell death in the BP of hatchling chicks. Although a hair cell lesion was evident several days after gentamicin treatment, the authors noted an unexpected recovery of hair cell numbers after 3–4 weeks. Another study used noise injury to induce hair cell lesions in the chick BP (Cotanche 1987). Imaging with scanning electron microscopy demonstrated the emergence over time of tiny bundles of stereocilia – the hearing organelle in hair cells – that resembled those of embryonic hair cells. This finding further suggested that some hair cells can be replaced after damage (Cotanche 1987).

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

13

Next, two studies (Corwin and Cotanche 1988; Ryals and Rubel 1988) published definitive evidence that these hair cells were regenerated and were the result of renewed cell proliferation. These studies focused on regeneration after noise trauma in both young chicks and senescent quail (Coturnix coturnix japonica). Radioactive (tritiated) thymidine, which becomes incorporated into newly replicated DNA, was administered to birds for several days after noise damage. Detection of this marker in the nuclei of replacement hair cells revealed that they had been regenerated by cell division and did not simply recover from damage or migrate from nearby uninjured areas. Importantly, the study of Ryals and Rubel also demonstrated that even old birds could replace hair cells after damage. Subsequent studies sought to identify hair cell progenitors, which are the cell type that divides and generates new hair cells after hair cell damage in birds (reviewed in Stone and Cotanche 2007). Markers such as bromodeoxyuridine [BrdU], tritiated thymidine, and DNA-binding agents were delivered into birds after hair cell damage, and organs were examined shortly thereafter to catch cells as they underwent DNA synthesis and mitosis. These studies demonstrated that, within 24 hours after damage, supporting cells reentered the cell cycle and divided, and daughter cells became hair cells, demonstrating that supporting cells are the hair cell progenitors. These studies showed that new supporting cells were also formed after hair cell loss, indicating that hair cell progenitors are replenished during regeneration. Following the discovery of hair cell regeneration in birds, researchers turned to other animals to examine the capacities for hair cell replacement in maturity. They determined that, following hair cell damage induced by laser-mediated ablation or ototoxic drugs, hair cells are also regenerated in amphibians (Balak et  al. 1990; Baird et  al. 1993), fishes (Song et  al. 1995), and reptiles (Avallone et  al. 2003). Studies with thymidine analogs or visual monitoring of live sensory epithelia showed that, like birds, supporting cells in these animals divide and form new hair cells shortly after hair cells degenerate and/or are lost from the sensory epithelia (e.g., Jones and Corwin 1993; Baird et al. 1996; Lanford et al. 1996). Later studies broadened our understanding of nonmammalian hair cell regeneration by showing that supporting cells can form new hair cells using mechanisms other than mitotic division. This conclusion was based on the findings that (1) hair cells were replaced after damage when supporting cell division was blocked by drugs (AraC or aphidicolin) and (2) some regenerated hair cells lacked a cell division marker even when it was provided for long periods of time after damage. These studies were conducted in frog saccules (Baird et al. 2000) and chick BPs (Adler and Raphael 1996; Roberson et al. 2004; Shang et al. 2010). Thus, while supporting cells can generate new hair cells by undergoing cell division, they can also undergo direct transdifferentiation, in which a highly specialized cell phenotypically converts into another cell without an intervening mitosis (Fig. 2.1). This process had been documented in invertebrates but was considered to be quite unusual in vertebrates (reviewed in Morest and Cotanche 2004). Some of the earliest work on postembyronic hair cell production in fish was conducted in sharks and goldfish (e.g., Corwin 1981; Presson et al. 1996). Later,

14

M. N. Hewitt et al.

Fig. 2.1  Known mechanisms of hair cell regeneration in nonmammals. (a) In the absence of damage, hair cells and supporting cells are regularly arranged within the sensory epithelium. Hair cells have regular, staircase-shaped stereocilia, and supporting cells interdigitate between hair cells. (b) When hair cells are damaged, the stereocilia can become splayed and disorganized, the supporting cells may facilitate extrusion of the apical region of the hair cell (including the stereocilia), and the damaged hair cells may be apically extruded or undergo intra-epithelial degeneration. Following hair cell loss, some supporting cells generate new hair cells, either through proliferation or direct transdifferentiation. (c) During proliferative regeneration, which occurs in lateral line neuromasts and chick inner ear organs, one supporting cell divides symmetrically to form two new hair cells or supporting cells, or asymmetrically to form a hair cell and a supporting cell; two new hair cells are shown in this example. (d) During direct transdifferentiation, which occurs in the avian vestibular system, a single supporting cell directly converts to a hair cell fate. (Figure © 2022 Madeleine Hewitt, all rights reserved)

many researchers adopted the zebrafish larval model, in large part for its experimental advantages over other species, such as small body size and widespread use for genetic studies. Research on hair cell regeneration in the zebrafish lateral line began with the demonstration that hair cells that are eliminated by exposure to ototoxic compounds are completely replaced after several days of recovery (Williams and Holder 2000; Harris et al. 2003). Regenerating hair cells are also properly reinnervated by directionally sensitive neurons (Nagiel et al. 2008; Faucherre et al. 2009). The lateral line system grows in complexity as the animal matures, adding new neuromasts from budding of existing sensory organs and from de novo generation from latent precursors (Nuñez et al. 2009; Iwasaki et al. 2020). Like in amphibians, lateral line neuromasts regenerate after tail amputation in zebrafish (Dufourcq et al. 2006). Postembryonic hair cell addition has also been observed in the inner ear of amphibians (Lewis and Li 1973) and aquatic vertebrates (Corwin 1981; Popper and Hoxter 1984). It is therefore perhaps not surprising that zebrafish neuromasts maintain the ability to produce new hair cells throughout life (Cruz et al. 2015; PintoTeixeira et al. 2015).

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

15

In contrast to birds and amphibians, direct transdifferentiation does not appear to be a major mechanism of hair cell regeneration in the fish lateral line. One study (Hernández et  al. 2007) reported BrdU-negative hair cells during regeneration, which would suggest that some hair cells were not produced via cell division. However, other studies found that most hair cells produced during regeneration harbored a marker of DNA synthesis and therefore arose from cell division (Ma et al. 2008; Wibowo et  al. 2011). In addition, treatment with the cell cycle inhibitors aphidicolin, flubendazole, or nocodazole blocked hair cell regeneration in the zebrafish lateral line (Wibowo et al. 2011; Mackenzie and Raible 2012). Time-lapse studies of regenerating neuromasts have also indicated that new hair cells arise from symmetrically dividing precursors (Wibowo et  al. 2011; Romero-Carvajal et  al. 2015). While it is possible that direct transdifferentiation sometimes occurs in lateral line neuromasts of fish, supporting cell mitosis appears to be the primary mode of hair cell regeneration. Compared to the lateral line, fewer studies have examined hair cell regeneration in the zebrafish inner ear. Zebrafish inner ear hair cells can be damaged and killed by exposure to noise trauma (Schuck and Smith 2009; Breitzler et  al. 2020). Following noise-induced hair cell loss, hair cell numbers visibly recover within 7 days, and functional recovery occurs within 2 weeks (Breitzler et al. 2020). Hair cell regeneration in the fish inner ear is therefore rapid and robust. However, it remains unclear whether hair cells in the fish inner ear are regenerated through direct transdifferentiation, proliferation, or both. Mature supporting cells in the inner ears of the goldfish Carrassius auratus maintain the ability to enter the cell cycle postembroynically, which suggests that they may be able to act as mitotic precursors for hair cells during regeneration (Presson et al. 1996). One study suggested that direct transdifferentiation may contribute to hair cell regeneration in the larval zebrafish inner ear (Millimaki et al. 2010). However, hair cell regeneration was assayed at a stage when developmental hair cell addition was still occurring. When hair cells are damaged by noise stimulation in adult zebrafish, regeneration is accompanied by cell proliferation (Schuck and Smith 2009). Supporting cell division was also observed when hair cells were laser ablated in the zebrafish lateral crista (Rubbini et al. 2015). Whether this increased proliferation functions to produce regenerated hair cells or replenish the supporting cell population is not clear. The full process of hair cell regeneration in birds, defined as the creation and maturation of new hair cells, reinnervation, and functional recovery, takes about 1 month in the auditory organ and more than 6 months in vestibular organs (reviewed in Bermingham-McDonogh and Rubel 2003; Oesterle and Stone 2008; Rubel et al. 2013). A major limitation on the speed of vestibular regeneration is the maturation of afferent innervation and formation of calyx afferents that envelope type I hair cells, which can require about 6 months (Haque et al. 2009). In zebrafish neuromasts, hair cell regeneration is more rapid and is completed within days rather than months. Within 72 h of nearly complete hair cell death, larval zebrafish neuromasts can regenerate a full complement of hair cells with mature features (Harris et al. 2003; Ma et al. 2008). Ribbon synapse maturation and reinnervation likewise occurs within 72 h post-injury (Faucherre et al. 2009; Suli et al. 2016) In older larvae, the

16

M. N. Hewitt et al.

onset of regeneration is likewise rapid, but complete functional recovery requires at least 5 days (Hardy et al. 2021). Studies on songbirds in the 1990s demonstrated that hair cell regeneration restores sensory function, bringing hearing sensitivity back to near-normal levels (e.g., Marean et al. 1993; Dooling et al. 1997), enabling perception of specific bird song (Woolley and Rubel 2002; Woolley et al. 2001; Ryals et al. 2013), and restoring the ability to sense head motions (Carey et al. 1996; Goode et al. 1999; Matsui et al. 2003; Dickman and Lim 2004). Similarly, studies demonstrate that regeneration of hair cells in fish lateral line also restores behaviors related to its functions, including escape responses, rheotaxis, and schooling (McHenry et  al. 2009; Suli et al. 2012; Mekdara et al. 2018). In birds, amphibia, and fishes, hair cells in vestibular, auditory, and lateral line sensory organs undergo turnover; that is, hair cells have a finite lifespan, at which point they die and are replaced at a relatively low rate (reviewed in Brignull et al. 2009; see Coffin et al. 2012; Cruz et al. 2015). This process is distinct but related to hair cell regeneration, which occurs after hair cells are damaged by an external stimulus (ototoxins, noise, or laser ablation). It is also distinct from the peripherally contained addition of hair cells in fish and amphibian vestibular organs, which grow in size throughout life (e.g., Corwin 1981; Corwin 1985). Importantly, no new hair cells appear to be added to auditory organs in mature birds (Jørgensen and Mathiesen 1988; Oesterle and Rubel 1993).

2.3 Supporting Cell Populations and Their Functions During Regeneration As discussed in Sect. 2.2, supporting cells serve as progenitors for replacement hair cells. Supporting cells reside alongside hair cells. They serve multiple functions, including maintaining the three-dimensional structure and the ionic balance of the sensory epithelium, generating overlying extracellular matrix materials (tectorial membrane, otoconial membrane, and cupula), and wrapping afferent and efferent neurites of the sensory nerves (reviewed in Wan et al. 2013). The finding that supporting cells are the source of new hair cells after damage in nonmammalian vertebrates introduced a new function for them and raised several questions: Where are the hair cell progenitors located, how are they regulated, and how are they maintained? Are all supporting cells in a given sensory epithelium functionally identical, or do distinct populations exist? In the past 10 years, a considerable amount of work addressing these questions has been conducted in the lateral line of zebrafish and in the auditory and vestibular organs of the chick inner ear. At first glance, supporting cells in nonmammalian lateral line and inner ear organs do not seem to be as specialized or diverse as their counterparts in the mammalian cochlea. For instance, although supporting cells in chick inner ear organs vary in shape depending on their locations in the epithelium, differences are subtle

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

17

compared to the mammalian cochlea. Likewise, early studies of neuromasts only identified the most peripheral cells (also known as mantle cells) as representing a distinct population of supporting cells; the rest were simply grouped together as “internal” supporting cells (Jones and Corwin 1993). New technologies have enabled more detailed interrogation of the diversity of cell types and lineage relationships in sensory epithelia, and through this work, it has become increasingly apparent that supporting cells in nonmammals are more diverse in function and behavior than previously recognized. This part of our chapter will review insights on supporting cell identities, locations, and functions in zebrafish neuromasts (Sects. 2.3.1 and 2.3.2) and chick inner ear epithelia (Sect. 2.3.3).

2.3.1 Identities and Locations of Hair Cell Progenitors in Fish As discussed in Sect. 2.2, hair cell regeneration in lateral line neuromasts occurs primarily via proliferating precursors. A study using markers of DNA synthesis and mitosis following neomycin-induced hair cell death found that proliferation of internal supporting cells peaked around 15–18 h posttreatment (Ma et al. 2008). A later study used transgenic lines that labeled hair cells and supporting cells and identified hair cell progenitors located near the dorsal-ventral (DV) compartments of anteroposteriorly polarized neuromasts (Wibowo et al. 2011; Fig. 2.2). These hair cell progenitors expressed high levels of the hair cell fate specification gene atoh1a (described in Sect. 2.5.1) and always divided just once, giving rise to two hair cells. These findings implied that DV-localized supporting cells are distinct from other internal supporting cells and have specialized functions during regeneration. The finding that new hair cells arise from symmetrically dividing progenitor cells rather than an asymmetrically dividing stem cell population raised the question of whether hair cell progenitors would eventually be depleted. However, mature neuromasts retain the capacity to regenerate hair cells after repeated rounds of hair cell death and regeneration (Cruz et al. 2015; Pinto-Teixeira et al. 2015). Time-lapse imaging of regenerating neuromasts has shown that supporting cells can also divide symmetrically to produce more supporting cells (Romero-Carvajal et  al. 2015). These supporting cell-producing divisions also occurred in DV compartments of the neuromast but were less centrally located than hair cell-producing divisions. Moreover, supporting cells produced by proliferation during a first round of hair cell regeneration can divide to form new hair cells after a second round of damage (Thomas and Raible 2019). Thus, after depletion by symmetric division and differentiation of progeny into new hair cells, precursors are replaced. These studies suggest that progenitor behavior within the neuromast depends on spatial location. Thomas and Raible (2019) used CRISPR-Cas9 technology (Sect. 2.4.2) to knock in fluorescent reporters into genes that were previously shown to have distinct and polarized expression patterns within neuromasts. Three genes generated distinct spatial expression patterns: sfrp1a drove expression in peripheral supporting cells, sost in DV supporting cells, and tnfsf10l3 in a ring of supporting

18

M. N. Hewitt et al.

Fig. 2.2  Anatomical locations and axes for the three sensory epithelia described in this chapter: the neuromast, utricle, and basilar papilla (not to scale). (a) The neuromast is divided into the central region (c), dorsal and ventral (DV) compartments, and anterior and posterior (AP) compartments. Peripheral supporting cells forms a ring around the neuromast (light gray). (b) The utricle is divided into the striola (dark gray, center) and the extrastriolar regions, with the line of polarity reversal (LPR, dashed line) dividing the center. (c) The basilar papilla is defined by the proximaldistal (sometimes also called basal-apical) and the neural-abneural axes. The neural-­abneural axis is so named because only hair cells on the neural side receive afferent innervation. (Figure © 2022 Madeleine Hewitt, all rights reserved)

cells near the periphery (but primarily anteroposterior [AP] supporting cells). Using lineage tracing, the authors tested the ability of each supporting cell population to give rise to hair cells after neomycin-induced hair cell death. DV supporting cells gave rise to the majority of regenerated hair cells (~60%), with lesser contributions from AP supporting cells (~20%) and peripheral supporting cells (~4%). It was further shown that specifically ablating DV cells reduced hair cell regeneration, suggesting that other supporting cells did not compensate for their loss. In addition, AP cells, peripheral cells, and more centrally localized DV cells could all regenerate DV cells. Single cell RNA sequencing (scRNA-Seq) has further supported the existence of multiple supporting cell subtypes (Lush et al. 2019). Sequencing at the single cell level allows for identification of rare cell types and developmental transitions that might be missed by population RNA sequencing (Sect. 2.4.1). Using this technique, the authors isolated over 1500 neuromast cells from 5-day-old zebrafish and identified seven distinct cell types: mantle cells, central supporting cells, AP-pole supporting cells, DV-pole amplifying supporting cells, differentiating hair cell progenitors, and young hair cells, and mature hair cells. Such data should provide a useful starting point for identifying factors that regulate supporting cell identity and

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

19

hair cell differentiation. Although this study examined neuromasts under homeostatic conditions, future studies may perform similar RNA-Seq experiments in regenerating neuromasts to see how gene expression profiles and developmental trajectories shift during hair cell regeneration.

2.3.2 The Role of Peripheral Supporting Cells in Fish Studies of amphibians, such as the axolotl salamanders Ambystoma punctatum and Ambystoma mexicanum, have shown that existing lateral line neuromasts can give rise to entirely new neuromasts during the process of tail-tip regeneration (Stone 1937; Jones and Corwin 1993). Similar phenomena occur in zebrafish (Dufourcq et al. 2006). The source of new neuromasts are the peripheral supporting cells of an existing neuromast close to the amputation site, which divide and form a placode that migrates along the regenerating tail. Similarly, during the growth of the adult lateral line system in zebrafish, new neuromasts bud from the periphery of existing neuromasts (Wada et al. 2010; Iwasaki et al. 2020). In both of these cases, peripheral supporting cells divide and give rise to migrating clusters of cells resembling the lateral line primordia that deposit neuromasts during development. Based on these results, peripheral supporting cells are thought to be capable of producing all cell lineages within the neuromast. Understanding conditions that provoke peripheral supporting cells to divide and produce other cell types within neuromasts has proven difficult. In the absence of damage, peripheral supporting cells divide rarely, and when they do, they are usually observed only to produce other peripheral supporting cells (Romero-Carvajal et al. 2015). When hair cells are killed using neomycin, peripheral supporting cells maintain relatively low levels of proliferation and do not usually give rise to new hair cells (Romero-Carvajal et al. 2015; Thomas and Raible 2019). In another fish species, medaka (Oryzias latipes), long-term lineage tracing showed that peripheral supporting cells can give rise to clones containing peripheral supporting cells, internal supporting cells, and hair cells (Seleit et al. 2017). Thus, peripheral supporting cells might continuously produce internal supporting cells at a very low rate during homeostasis, perhaps explaining why this phenomenon has not been seen in studies that only tracked cell divisions over several days. Given these data, some have proposed that peripheral supporting cells are a latent stem cell population that only responds to severe injuries. As described previously in this section, tail-tip amputation (which ablates entire neuromasts) induces peripheral supporting cell division, but hair cell ablation alone does not. Several studies have attempted to address whether larger scale injuries within the neuromast induce peripheral supporting cell division. In one study, the authors laser ablated 40% to 95% of neuromast cells in medaka, leaving behind a few peripheral supporting cells (Seleit et al. 2017). As few as four peripheral supporting cells could regenerate the neuromast and all cell types within it. In a similar study using laser ablation of neuromast cells in zebrafish, a mixture of two to three peripheral supporting cells and

20

M. N. Hewitt et al.

two to seven internal supporting cells was enough to regenerate the neuromast to 70% of its normal size at 7 days post-injury (Viader-Llargués et al. 2018). Time-­ lapse imaging of regenerating neuromasts revealed that internal supporting cells also produced all three major cell types of the neuromast, but peripheral supporting cells only appeared to produce other peripheral supporting cells. What could explain the differences in peripheral supporting cell behavior observed in these studies? One possibility is that peripheral supporting cells behave differently in medaka compared to zebrafish. Another possibility is that peripheral supporting cells only contribute substantially to regenerating other cell types when damage includes internal supporting cells as well as hair cells. Supporting this idea, peripheral supporting cells gave rise to DV-localized supporting cells when DV cells were ablated (Thomas and Raible 2019). One caveat is that different markers for peripheral supporting cells were used in all three of these studies, so it is possible that these studies are not investigating the exact same cell subtypes. More research is needed to resolve these discrepancies and conclusively define their role within the neuromast. It also remains unknown what signals regulate peripheral supporting cell identity. One possibility is that neuromast border cells (nBCs) provide signals to peripheral supporting cells (Seleit et  al. 2017). nBCs are specialized epithelial cells surrounding the neuromast that are induced to form as neuromasts develop. They constitute an independent epithelial lineage and can regenerate themselves after damage. The existence of nBCs seems to be conserved across fish species. As cells that surround the neuromast and directly contact peripheral supporting cells, nBCs are a good candidate for cells that might form a niche to maintain peripheral supporting cell identity.

2.3.3 Supporting Cell Diversity in Birds Relative to zebrafish, fewer studies have explored supporting cell heterogeneity in birds. In the chick BP, there are differences in supporting cell shape across the radial width of the epithelium: While nuclei of supporting cells in the neural half of the epithelium reside close to the basal lamina, those in the abneural half reside almost halfway between the basal lamina and the lumen (e.g., Scheibinger et  al. 2018). Another sign of supporting cell heterogeneity in birds is their regionalized expression of certain genes under normal conditions and after damage. Janesick et  al. (2021) used scRNA-seq to determine that, while supporting cells throughout the uninjured BP express many common genes, some genes are transcribed at very different levels in supporting cells located in different zones in the BP (neural versus abneural). Similarly, in the undamaged chick utricle (a vestibular organ), transcripts for some Notch pathway genes are abundant in supporting cells in the striola (a specialized region near the center of the epithelium) but are scarce outside of this region (Warchol et al. 2017). Following gentamicin treatment, the chick BP upregulated expression of Notch pathway genes Ser1 and Hes5, and a Notch pathway

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

21

modulator Lfng (Daudet et al. 2009). The increased expression of these genes was observed only in supporting cells in the neural region, despite global hair cell loss. It is unclear if these restricted patterns of gene expression reflect functionally significant differences in supporting cell populations. The behaviors of supporting cells also vary across the zones of the damaged BP. The majority of supporting cell divisions following gentamicin damage occurred in the neural (superior) portion of the sensory epithelium (Cafaro et  al. 2007). Consistent with this, most regenerated hair cells in the neural half of the BP were derived via supporting cell mitosis, while most cells in the abneural half appeared to have been formed by direct transdifferentiation (Cafaro et al. 2007). It is not clear why these two paths toward a new hair exist, whether supporting cells are predetermined for one path or the other, of if a given supporting cell is simply influenced by its microenvironment to divide or directly transdifferentiate. These are interesting areas for further study, as is the question of the molecular mechanisms that guide a supporting cell toward the hair cell fate using either path. There is evidence that stemlike cells reside among supporting cells in avian inner ear epithelia (reviewed in Stone and Cotanche 2007; Kelley and Stone 2016). For instance, cell divisions in the chick BP and utricle can yield cells with asymmetric fate outcomes (one hair cell and one supporting cell; Stone and Rubel 2000; Roberson et al. 1992; Scheibinger et al. 2018). Furthermore, the quail BP regenerates hair cells after repeated rounds of hair cell damage (Niemiec et al. 1994). These findings strongly suggest that bipotential cells (i.e., cells that can form at least two cell types) exist, and these cells can replenish hair cell progenitors. For a more detailed consideration of inner ear stem cells, see Janesick, Heller, and Hashino, Chap. 6.

2.4 Approaches to Define New Molecular Regulators Using Nonmammals When supporting cells regenerate hair cells after injury, they leave their normal “resting” state and either divide or change phenotype. The specific molecular events that initiate these processes are not fully known. They could be associated with disruption of signals that healthy hair cells normally send to supporting cells via direct cell-cell contacts or diffusible substances. Alternatively, supporting cell transdifferentiation may be triggered by shifts in other aspects of the epithelial microenvironment during or after damage. Afferent and efferent nerves, which course through the epithelium, appear dispensable for avian hair cell regeneration because new hair cell-like cells still form in sensory organs that were dissected away from other cells in the nervous system and grown in vitro (e.g., Matsui et al. 2000) and in cultures of dissociated supporting cells (e.g., Stone et al. 1996). It is also unlikely that immune cells such as macrophages are necessary for hair cell regeneration in avian auditory organs because depletion of macrophages in the BP does not affect

22

M. N. Hewitt et al.

numbers of hair cells that get replaced (Warchol et al. 2012). Similarly, although macrophages respond to hair cell damage in the zebrafish lateral line, they are not required for regeneration (Warchol et al. 2021). Although innervation is not necessary for the initial development of zebrafish lateral line hair cells (Grant et al. 2005; Suli et  al. 2016), changes in innervation result in reduced hair cell regeneration (Hardy et al. 2021). It is of great interest to identify the molecules and molecular pathways that direct hair cell regeneration in nonmammals because such knowledge may lead to methods for promoting a similar form of regeneration in the ears of mammals. Scientists have taken advantage of new genomic methods to identify candidate regulators, which include transcriptional profiling, mutagenesis studies, and small-molecule testing. Each of these approaches is unique and described separately below.

2.4.1 Transcriptional Profiling Identification of proteins that are key regulators of cellular processes such as cell death, division, differentiation/maturation, and innervation during hair cell regeneration is an important focus of research. Because cells tend to “upregulate” or “downregulate” synthesis of proteins as they become needed or dispensable, one approach toward identifying such potential regulators is to determine which proteins are generated during important phases of regeneration (e.g., as hair cells are dying, supporting cells are dividing, or hair cells are differentiating and synapses are forming). Direct identification of proteins and their patterns of expression can be accomplished with proteomics and immunolabeling. However, proteomics requires large amounts of biological material, which is difficult to obtain with inner ear epithelia, and most immunolabeling approaches require knowledge a priori of proteins of interest. Instead, determination of the types and numbers of mRNA transcripts that are present at key times after damage is often used as a proxy. Although every transcript will not necessarily yield a functional protein, transcriptional profiling is very effective for identifying potential protein candidates for further study. The first studies to apply transcriptional profiling to study nonmammalian hair cell regeneration were conducted in utricles and cochlea of chickens (Hawkins et al. 2003; Hawkins et al. 2007; Frucht et al. 2010). A similar approach by Frucht et al. (2010) used microarrays to identify changes in gene expression associated with activation of adenylyl cyclase (AC), which promotes mitotic hair cell regeneration in the chick BP.  These groups identified numerous transcripts whose numbers changed significantly in sensory epithelial cells after drug damage or AC activation, and they validated changes in transcripts using reverse quantitative polymerase chain reaction or RNA in situ hybridization. Furthermore, Frucht et al. (2010, 2011) predicted through analysis in silico that microRNA 181a controls the levels of transcripts whose expression changed in association with AC activation, implicating microRNA 181a as a regulator of supporting cell proliferation, and they obtained evidence to support that hypothesis. Additional microarray analysis of cells from

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

23

the sensory epithelium of the chick utricle indicates that an RNA-binding protein called MSI1 may play a role in avian hair cell regeneration (Wakasaki et al. 2017). Microarrays have also been used to profile gene expression in the regenerating inner ear (Schuck et al. 2011) and lateral line (Behra et al. 2012; Steiner et al. 2014) of zebrafish. Other studies performed transcriptional profiling using RNA sequencing (RNASeq), a technique that significantly expands the number of transcripts that can be examined. Using this approach, Ku et  al. (2014) identified FGF20 as a negative regulator of supporting cell proliferation in the chick utricle. A subsequent microarray analysis showed that FGFs and their receptors are also dynamically regulated at the transcriptional level in the BP following hair cell damage (Lee et al. 2016) and that the transcription factor CMYC is another potent regulator of supporting cell division during avian hair cell regeneration. Further studies showed dynamic expression of genes in the Notch, FGF, Wnt, and BMP pathways after hair cell damage (Jiang et al. 2018; Matsunaga et al. 2020). Finally, RNA-Seq approaches have also been used to compare supporting cell types in adult zebrafish and mouse ears (Giffen et al. 2019) and to study regeneration in the adult zebrafish ear (Liang et al. 2012) and the larval lateral line (Jiang et al. 2014), revealing potential roles for Stat, Wnt, and Notch signaling. A more refined transcriptional profiling method, single cell RNA sequencing (scRNA-Seq), has subsequently emerged as a method to analyze gene expression of individual cells. The method yields expression data that would otherwise be lost with whole tissue extraction of mRNA. This method has been used to profile zebrafish lateral line cells (Lush et al. 2019), which has identified distinct cell phenotypes and also suggested roles for Fgf signaling in hair cell regeneration. Another study used scRNA-Seq to explore properties of quiescent supporting cells in the undamaged chick BP (Janesick et al. 2021). We expect that this powerful method will play an increasingly important role in regeneration research.

2.4.2 Genetic and Molecular Screening Another powerful approach toward identifying regulators of cellular processes underlying hair cell regeneration is genetic screening. During a forward genetic screen, researchers induce random mutations to identify genes that affect hair cell regeneration. Forward screens are common in organisms that breed rapidly, such as the fruit fly Drosophila melanogaster. Although fruit flies have hearing organs (called Johnston’s organs) and investigators have begun to use them as a model to study hair cell genetics (e.g., Li et al. 2016, 2018), we do not know if they regenerate auditory receptors after damage. Given their speed of reproduction, zebrafish have been used in forward genetic screens to study hair cell development. For instance, a novel gene, phoenix, was identified by Behra and colleagues as part of an insertional mutagenesis screen in zebrafish (Behra et al. 2009). Mutants developed normal neuromasts but, in contrast to wild-type fish, hair cells failed to regenerate

24

M. N. Hewitt et al.

after damage induced by copper and neomycin because supporting cells did not divide. The mutation is lethal, indicating that phoenix has other functions besides regeneration. So far, there have been few reports of screens designed to identify genes that are specific regulators of hair cell regeneration. Targeting individual genes for inactivation is often called reverse genetics, as the gene identity is known before the loss of function phenotype. In this approach, genes of interest are often identified by transcriptional profiling and inactivated using antisense oligonucleotides or RNA interference (RNAi). Transcription factor pathways including PAX and AP1 were identified by an RNAi screen of genes involved in regeneration of avian sensory epithelia (Alvarado et al. 2011). Rapid advances in genomic engineering using CRISPR and other approaches have made the reverse genetic approach feasible for assessing large numbers of targeted mutations by phenotypic screening. This approach can avoid some of the off-target effects of targeted interference with antisense oligonucleotides or RNAi. Pei and colleagues have used this approach to screen a set of mutations in genes that are differentially expressed in the adult zebrafish inner ear by transcriptional profiling (Pei et al. 2016, 2018). Systemic screening of 254 genes identified seven genes that, when mutated, affected hair cell regeneration. Disruption of these genes also altered other types of regeneration including fin and liver, suggesting that a common set of genes may play general roles in these regenerative processes. As an alternative to genetic screens, small-molecule screening provides a way to assess different potential regulatory pathways. Witte, Montcouquiol, and colleagues screened a panel of 13 signaling pathway inhibitors to identify those that altered proliferation in chick (Witte et  al. 2001) and mouse (Montcouquiol and Corwin 2001) utricle cultures. In parallel, these studies found that blocking PI3K and mTOR pathways reduced proliferation of supporting cells. Similar studies have been performed using zebrafish to identify drugs that alter hair cell regeneration in the lateral line (Namdaran et  al. 2012). Screening a 1680 compound library of drugs with known biological activity revealed that glucocorticoids enhanced hair cell addition, while drugs that blocked mitosis inhibited regeneration. Although drug screens can be effective in identifying potential signaling pathways involved in biological processes, drug action can be the result of off-target effects rather than effects on the pathway the drug was developed for. If the target is unknown, it can be very challenging to determine the mechanistic cause of drug action. In conclusion, the application of the genetic approaches has, and will continue to, expand and accelerate our understanding of how nonmammalian hair cell regeneration is regulated. These approaches will also provide important clues as to how hair cell replacement may be stimulated in the hearing organ, and augmented in the vestibular organs, of mammals. Although considerable progress has been made over the past 10 years using these genetic tools, considerably more work is needed to identify the set of genes that are necessary to drive full morphological and functional recovery.

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

25

2.5 Molecular Regulation of Supporting Cells In this section, we review current knowledge of the signals that control hair cell regeneration in nonmammalian vertebrates. These insights were obtained using the tools described in Sect. 2.4, as well as “best guess analysis” based on published data from other systems or developing hair cell epithelia. For additional reviews, we point readers to Stone and Cotanche (2007), Oesterle and Stone (2008), Warchol (2011), and Kelley and Stone (2016).

2.5.1 Transcription Factors Regulate Hair Cell Regeneration in Nonmammals Transcription factors are proteins that regulate gene expression and act in context-­ dependent manners (i.e., in specific times and specific cell types) to maintain a cell’s status or to effect a change in status. Much is known about how transcription factors control the formation of hair cells during embryogenesis (discussed in McGovern and Groves, Chap. 5). Because hair cell regeneration may, in many ways, recapitulate development, a number of studies have explored how developmentally relevant transcription factors could influence hair cell replacement later in life. Two examples of these factors are ATOH1 and SOX2. ATOH1 is a transcription factor in the basic helix-loop-helix family that appears necessary for acquisition of a hair cell phenotype in all vertebrates (e.g., Bermingham et  al. 1999; Millimaki et  al. 2007). Moreover, misexpression of Atoh1 (forced expression at an abnormal time, place, or level) induces cells in the developing mammalian inner ear to acquire the hair cell fate (e.g., Zheng et al. 2000; Liu et al. 2012). In mature sensory organs lacking hair cell damage, there is little expression of ATOH1/atoh1a/atoh1b in hair cells or supporting cells (e.g., Cafaro et al. 2007; Ma et al. 2008; Fig. 2.3), although small levels of expression are always present in organs that undergo hair cell turnover (e.g., chick utricle and zebrafish neuromasts). Following damage, however, ATOH1/atoh1a/atoh1b expression at mRNA and protein levels becomes highly upregulated in supporting cells (Cafaro et al. 2007; Ma et al. 2008). As described in these studies and others, increased expression of Atoh1 is one of the earliest pro-regenerative responses in nonmammalian supporting cells. Atoh1 becomes elevated in at least some dividing supporting cells, as well as in supporting cells that undergo direct transdifferentiation and in differentiating hair cells. Its expression returns to baseline levels once regenerated hair cells have matured. In the chick BP, misexpression of ATOH1 causes overproduction of hair cells and stimulates supporting cell division (Lewis et al. 2012). Because of Atoh1’s powerful roles in hair cell development and nonmammalian hair cell regeneration, several studies have examined whether Atoh1 misexpression also induces hair cell regeneration in the ears of mammals. Results of those studies are discussed in Forge and Taylor, Chap. 4, and McGovern and Groves, Chap. 5.

26

M. N. Hewitt et al.

Fig. 2.3  Different stages of hair cell regeneration in chickens and description of the molecules known to either inhibit (red text) or activate (green text) each stage. (Figure © 2022 Jennifer Stone, all rights reserved)

In contrast to ATOH1, some transcription factors are broadly expressed in supporting cells under normal conditions (i.e., in mature undamaged epithelia). Two examples are the SRY box transcription factor SOX2 (Neves et al. 2007; Oesterle et al. 2008) and the basic-loop-helix transcription factor ID2/3 (Lewis et al. 2012), both of which are required for hair cell development (Kiernan et al. 2005; Jones et al. 2006b; Kamaid et al. 2010). Following damage to the chick BP, expression of both SOX2 and ID2/3 is downregulated in supporting cells in spatial and temporal concert with the increased expression of ATOH1, as supporting cells directly transdifferentiate into hair cells (Cafaro et  al. 2007; Lewis et  al. 2018). This reduced expression of supporting cell-specific transcription factors presumably occurs as supporting cells downregulate expression of genes that maintain the supporting cell fate and upregulate genes that promote either hair cell fate or cell division. Further study will be necessary in order to understand the specific roles of ID and SOX transcription factors during avian hair cell regeneration. Studies of zebrafish have revealed that mutations in sox2 and the closely related family member sox3 affect the initial formation of the otic placode as well as subsequent development of hair cells and innervating neurons (Gou et al. 2018a, 2018b). Both transcription factors cooperate during the establishment of the placode but appear to play distinct roles in the differentiation of sensorineural structures: Sox2 promotes hair cell differentiation, while Sox3 promotes neuronal differentiation. Sox2 is also expressed in the migrating lateral line primordium and in supporting cells in mature neuromasts (Hernández et  al. 2007). Following hair cell death,

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

27

sox2-expressing supporting cells divide and downregulate sox2 as they differentiate into hair cells (Hernández et al. 2007). Because some later divisions also produce more supporting cells, the sox2-expressing population is usually not depleted during hair cell regeneration. More severe injuries in zebrafish can deplete the sox2-expressing population. For example, when fish are first treated with Notch inhibitor to force hair cell overproduction (see Sect. 2.5.2.1) and then treated with neomycin to ablate hair cells, the number of sox2-positive supporting cells is reduced (Romero-Carvajal et al. 2015; Ye et al. 2020). However, sox2-expressing supporting cells were shown to regenerate from surviving atoh1a + supporting cells through yap1-mediated upregulation of the RNA-binding protein lin28a (Ye et al. 2020). Lin28a inhibits the microRNA let7, which activates Wnt signaling and promotes progenitor renewal. This suggests that some atoh1a + supporting cells within neuromasts can respond to severe injury by upregulating sox2 and dividing to renew the sox2-positive population.

2.5.2 Cell-Cell Signaling Molecules That Regulate Hair Cell Regeneration in Birds and Fish Cellular behaviors are influenced by signaling molecules that enable cells to communicate with each other. The vast majority of molecules known to control hair cell regeneration are proteins, but signaling agents can also be ions, lipids, or other molecules. Signaling molecules can act at close range upon adjacent cells, or they can diffuse or be transported to cells at a distance. In this section, we discuss signaling pathways known to regulate nonmammalian hair cell regeneration, as summarized in Fig. 2.3. 2.5.2.1 Notch Signaling Probably the best studied signals that control nonmammalian hair cell regeneration are those that act through the Notch receptor. Notch receptors are bound by Delta/ Serrate/Jagged ligands that are usually found on the membranes of adjacent cells. When such ligands bind Notch, they cause its intracellular domain (NICD) to be proteolytically cleaved by gamma secretase. NICD then translocates to the nucleus, where it acts as a transcriptional activator. NICD promotes expression of Hes/Hey/ Her transcriptional repressors that reduce expression of Atoh1 and Notch ligands (Zheng et  al. 2000), thereby preventing the cell from adopting the hair cell fate. Because Notch-ligand interactions require cell-cell contacts, cells not adjacent to the ligand-expressing cell are permitted to become hair cells. As a result, Notch signaling establishes a pattern of alternating hair cells and supporting cells and ensures the formation of the normal mosaic structure of the sensory epithelium (reviewed in Kelley 2003; Kelley and Stone 2016; Chitnis et al. 2012). Disruption

28

M. N. Hewitt et al.

of Notch signaling during embryogenesis leads to production of excess hair cells, inappropriate organ size, and improper cellular patterning. These changes occur because Notch inhibition increases proliferation of supporting cells and causes inappropriate cell fate determination: cells are diverted from becoming supporting cells and instead acquire the hair cell fate (e.g., Haddon et  al. 1998; Itoh and Chitnis 2001). In mature sensory epithelia, Notch pathway components continue to be expressed under homeostatic conditions. NOTCH1 is expressed by supporting cells in the chick BP (Stone and Rubel 1999), while several Notch receptors are expressed by supporting cells in zebrafish neuromasts (notch1a/b, notch3) (Romero-Carvajal et al. 2015). Jagged family ligands (SERRATE1 in chick, jagged2b in zebrafish) are expressed in supporting cells (Stone and Rubel 1999; Romero-Carvajal et al. 2015). Although the function of Jagged ligands in mature sensory epithelia has not been well explored, it is required in mice for maintenance of Hensen’s cells, which are laterally position supporting cells in the organ of Corti (Chrysostomou et al. 2020). Delta ligands are expressed in hair cell precursors and in regenerated hair cells that have yet to fully differentiate (Stone and Rubel 1999; Ma et al. 2008; Jiang et al. 2014). Pharmaceutical inhibition of Notch after hair cell damage caused hair cell overproduction in lateral line neuromasts (Ma et al. 2008; Romero-Carvajal et al. 2015) as well as in chick BPs (Daudet et al. 2009; Jiang et al. 2018) and utricles (Warchol et al. 2017) grown in culture. The increase in hair cell numbers occurs at the expense of supporting cells. Thus, regenerated hair cells express Delta, which signals to Notch in surrounding supporting cells and prevents them from adopting a hair cell fate. This mechanism ensures that the proper ratio of hair cells to supporting cells is reestablished after damage. Given the developmental role of Notch in inhibiting hair cell production and the finding that Notch is expressed postnatally, it was first postulated that Notch signaling might, under homeostatic conditions, maintain supporting cell quiescence and prevent them from transdifferentiating into hair cells. However, while Notch inhibition led to increased hair cell production after neomycin-induced hair cell loss, it had no effect on hair cell number in undamaged neuromasts (Ma et  al. 2008; Wibowo et al. 2011). Similarly, Notch inhibition in the regenerating chick BP led to an overproduction of replacement hair cells (Daudet et  al. 2009). Notably, such overproduction only occurred in regions where significant hair cell damage occurred but not in undamaged regions of the BP. Notch inhibition therefore appears to be insufficient to trigger hair cell production in both the BP and lateral line neuromasts under homeostatic conditions. In the regenerating chick BP, supernumerary hair cells that were produced in response to blocking Notch activity were generated by both cell division and direct transdifferentiation, indicating that either supporting cells or postmitotic precursor cells could be coaxed to form hair cells in the absence of Notch (Daudet et al. 2009). However, Notch inhibition did not result in increased supporting cell division, so a subset of supporting cells transdifferentiated into hair cells and were not replenished. A radically different response to Notch inhibition was observed in the chick utricle. Inhibiting of Notch in undamaged utricles led to a dramatic increase in

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

29

supporting cell division without significant hair cell loss (Warchol et  al. 2017). Thus, the homeostatic chick utricle seems to have fewer intrinsic brakes on supporting cell division than the BP. This finding is not surprising because there is significant turnover of hair cells throughout chick utricles under normal conditions (Kil et al. 1997), and the process of turnover would be hampered if cell division were suppressed under homeostatic conditions. The mechanisms that block cell proliferation and new hair cell production in regions with normal hair cell densities have not been identified, but they are most likely derived from the hair cells themselves. Furthermore, the reasons why zebrafish neuromasts and chick utricles respond differently to Notch inhibition under apparently homeostatic conditions are not known. The distinct responses that each hair cell epithelium exhibits in response to Notch inhibition demonstrate clearly that molecular regulation of supporting cell division is unique in each context. Furthermore, these findings reveal that other inhibitory mechanisms beside Notch signaling prevent supporting cells from dividing and/or transdifferentiating into new hair cells. This finding has important implications for the development of therapies for promoting hair cell regeneration in mammals. Indeed, the potent inhibitory role of Notch that occurs during development in vertebrates and during regeneration in nonmammalian vertebrates had prompted the exploration of whether Notch signaling might block hair cell regeneration in mature mammals. A number of studies have found that inhibition of Notch activity promotes regeneration in the damaged mouse utricle but not the cochlea. These studies are discussed in detail in Forge and Taylor, Chap. 4, and in McGovern and Groves, Chap. 5. In addition to its role in regulating supporting cell proliferation and hair cell differentiation, Notch signaling also appears to play an important role in regulating hair cell maturation, controlling the cell polarity that determines their directional sensitivity to mechanical stimulation. The transcription factor Emx2 plays a central role in this process (Jiang et al. 2017; Ji et al. 2018). Emx2 is expressed in about half of all hair cells in the developing mouse utricle. Those Emx2-positive hair cells all display the same polarity, while Emx2-negative hair cells have the opposite polarity. Emx2 was shown to be both necessary and sufficient for determining hair cell polarity across multiple species and sensory epithelia. Studies suggest Notch signaling acts to determine which hair cells express Emx2 (Jacobo et al. 2019; Kozak et al. 2020). To what extent these different functions of Notch signaling – regulating supporting cell proliferation, cell fate choice, and differentiation of hair cell polarity – are controlled by distinct Notch receptors and ligands is unknown. In addition, understanding how Notch signals result in different outcomes in different contexts will require further research. 2.5.2.2 Wnt Signaling Wnt has emerged as an important regulator of hair cell regeneration in birds and fish. Wnt ligands regulate cell division and differentiation in a variety of cell types in the body. Expression of WNTs, their receptors (FZDs), their effectors (e.g.,

30

M. N. Hewitt et al.

ß-catenin), and their modulators (e.g., DKK’s, SFRP’s) is altered in supporting cells following hair cell damage in the chick utricle (Alvarado et al. 2011; Ku et al. 2014). Knockdown of WNT4 in cell cultures of avian utricular supporting cells inhibits their proliferation (Alvarado et  al. 2011), while activation of the WNT effector ß-catenin in the chick BP promotes supporting cell division (Jiang et  al. 2018). Similar findings have been reported in studies of zebrafish lateral line neuromasts, in which activation of Wnt signaling increases supporting cell proliferation during development and regeneration, leading to increased hair cell numbers (Head et al. 2013; Jacques et al. 2014). Unlike Notch, Wnt does not appear to affect fate decisions of dividing supporting cells because both hair cell and supporting cell numbers increase in response to Wnt activation (Jacques et al. 2014; Romero-Carvajal et al. 2015). Wnt pathway genes are not expressed during homeostasis in mature neuromasts, and during regeneration, proliferation begins prior to significant upregulation of Wnt (Jiang et  al. 2014). This suggests that signals other than Wnt are responsible for the initial phase of regeneration in the lateral line. Studies of the zebrafish lateral line have further elucidated the interplay between Notch and Wnt signaling during homeostasis and regeneration. Components of both Notch and Wnt signaling have polarized expression patterns within neuromasts (Jiang et al. 2014; Romero-Carvajal et al. 2015). Notch normally inhibits Wnt signaling during homeostasis by promoting expression of the Wnt inhibitor dkk2 (Romero-Carvajal et al. 2015). Following hair cell death, Notch is transiently downregulated, and subsequently, Wnt is upregulated (Jiang et al. 2014; Romero-Carvajal et  al. 2015). Treatment with a gamma secretase inhibitor disrupts the polarity of supporting cell divisions during homeostasis and regeneration, causing divisions to be found throughout the neuromast instead of localized to the DV compartment (Romero-Carvajal et al. 2015). It is possible that Notch signaling more strongly suppresses proliferation outside the DV region, perhaps by inhibiting Wnt, which may explain why DV cells contribute more to hair cell regeneration (Wibowo et al. 2011; Thomas and Raible 2019). 2.5.2.3 Other Signaling Pathways Work in the past 10  years revealed at least three tyrosine kinase receptors to be important regulators of hair cell regeneration in chick epithelia. Both epidermal growth factor (EGF) and vascular endothelial growth factor (VEGF) are required for supporting cell division in the aminoglycoside-treated chick BP. EGF and VEGF ligands bind to their corresponding receptors, known as the EGF receptor (EGFR) and the VEGF receptor (VEGFR), to initiate downstream signaling. Inhibition of EGFR in cultured chick cochlear ducts significantly reduced supporting cell division (White et al. 2012) as did VEGFR antagonists (Wan et al. 2020). As anticipated, treatment of cultures with recombinant VEGF caused higher numbers of hair cells to be regenerated (Wan et al. 2020). VEGFA is expressed highly in normal hair cells, while VEGFRs are expressed in supporting cells. Wan et al. (2020) suggest that release of VEGFA from damaged hair cells may promote supporting cell

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

31

division. This model is consistent with classical observations that supporting cell division is initiated following visible injury to auditory hair cells (reviewed in Stone and Cotanche 2007). Fibroblast growth factor (FGF) signaling appears to antagonize hair cell regeneration in the chick ear (Ku et al. 2014; Jiang et al. 2018). In chick utricles, transcripts encoding FGF20 and its cognate receptor FGFR3 decreased after hair cell death. Addition of FGF20 to cultured chick utricles curtailed the supporting cell division that follows hair cell loss (Ku et al. 2014), while inhibition of FGFR activity in damaged chick BPs increased numbers of both dividing supporting cells and regenerated hair cells (Jiang et al. 2018). The function of Fgf signaling in regeneration of hair cells in zebrafish neuromasts is less clear. Fgf3 is expressed in the central region of the neuromast, while fgf receptors are expressed in supporting cells outside the central region (Lee et al. 2016). Fgf pathway components, including fgf3, are transiently downregulated following hair cell death in neuromasts (Jiang et al. 2014; Lee et al. 2016; Lush et al. 2019). Lee et al. (2016) found that Fgf inhibition reduced hair cell regeneration and interpreted the loss of fgf3 transcripts following hair cell death to mean it was expressed in hair cells. However, a later study in which a transgenic knock-­in line was created to label fgf3-expressing cells found that fgf3 is expressed in central supporting cells rather than hair cells (Lush et al. 2019). Furthermore, loss of Fgf activity caused increased supporting cell proliferation and hair cell production during homeostasis and regeneration (Lush et al. 2019). More work is needed to conclusively define the role of Fgf signaling in zebrafish neuromasts. Retinoic acid (RA) signaling plays many roles during development (reviewed in Rhinn and Dollé 2012). RA is a small-molecule synthesized from retinol (also known as vitamin A) by dehydrogenase enzymes. RA binds to retinoic acid receptors (RARs) that bind directly to genomic RA response elements (RAREs). RA binding changes the activity of RARs from transcriptional corepressors to coactivators. In zebrafish, RA pathway components were transiently upregulated in both the neuromasts and the cristae of the inner ear following hair cell death (Rubbini et al. 2015). Using a dominant negative RAR (dnRAR), the authors found that inhibiting RA signaling reduced supporting cell proliferation and delayed hair cell regeneration. Expression of dnRAR suppressed downregulation of the cyclin-dependent kinase inhibitor p27kip, which inhibits reentry into the cell cycle. RA signaling was also required for normal downregulation of sox2 expression following hair cell loss. These data support a model whereby hair cell loss leads to RA-mediated inhibition of p27kip and sox2 in supporting cells, thereby allowing them to reenter the cell cycle. It is not known which cells normally act as a source of RA in neuromasts. The gene aldh1a2, which encodes a dehydrogenase involved in RA biosynthesis, is expressed in peripheral supporting cells of zebrafish neuromasts (Pittlik and Begemann 2012). A similar expression pattern has been observed in axolotls (Monaghan and Maden 2012). Thus, it is possible that peripheral supporting cells act as a source of RA during hair cell regeneration, although this has yet to be explicitly tested.

32

M. N. Hewitt et al.

Bone morphogenetic proteins (BMPs) bind to receptors with serine-threonine kinase activity and regulate a variety of developmental processes (reviewed in Wang et al. 2014). In the chick BP, BMP4 is highly expressed in hair cells, and its receptors are expressed in supporting cells (Lewis et al. 2018; Jiang et al. 2018). During hair cell loss, BMP4 transcripts decrease substantially, and expression of a BMP4-­ regulated gene – the transcriptional cofactor, ID2/3 – is also reduced. When exogenous BMP4 was added to cultured BPs after streptomycin-induced hair cell death, there was significantly less supporting cell division and very few hair cells were regenerated. Furthermore, there was little upregulation of ATOH1, which is required for hair cell differentiation. In contrast, inhibition of BMP4 signaling after hair cell loss increased numbers of regenerated hair cells. These findings indicate that BMP4 is a potent inhibitor of supporting cell regenerative responses, suggesting that steady-state BMP4 signaling by hair cells might prevent supporting cells from forming new hair cells in the absence of damage. Several other signaling pathways have been implicated in regulation of nonmammalian hair cell regeneration including growth hormone (Sun et al. 2011) and the Hippo pathway (Rudolf et al. 2020; see Rudolf and Corwin, Chap. 3). Growth hormone also regulates hair cell addition in fish inner ear organs (Sun et  al. 2011). Finally, it is important to acknowledge that some signaling pathways work together to exert additive effects upon hair cell regeneration (e.g., Bai et al. 2020).

2.5.3 Epigenetic Mechanisms Controlling Nonmammalian Hair Cell Regeneration As discussed above, the process of hair cell regeneration involves changes in gene expression. Gene expression is regulated by the interaction of transcription factors with the promoter and enhancer regions associated with particular genes, and this process is critically dependent on the local structure of chromatin. Chromatin can exist in either “open” or “condensed” configurations, and its structure is determined by enzymatic modification of both histones and DNA.  Chromatin remodeling occurs when histones are enzymatically modified (e.g., acetylated or methylated) at specific amino acids, altering chromatin accessibility and the likelihood that a gene or set of genes will be transcribed into mRNA. DNA methylation is another mechanism that alters gene transcription. Together, these types of modifications are called epigenetic, as they alter gene expression without altering the gene sequence encoded in DNA. Such epigenetic regulation is likely to play an influential role in how, when, and to what extent hair cells are regenerated in all vertebrates. At this point, there has been little work done on epigenetic regulation of hair cell regeneration, except with respect to chromatin remodeling (see below). In contrast, there is a rapidly growing body of work on epigenetic control of hair cell development (see Yizhar-Barnea et  al. 2018; reviewed Doetzlhofer and Avraham 2017). One key result is that epigenetic modification of Atoh1 regulatory elements is correlated temporally with the loss of regeneration potential in the mammalian cochlea

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

33

during the early postnatal period (Stojanova et al. 2015). This work and other epigenetic studies pertaining to mammalian hair cell development and regeneration are discussed in Forge and Taylor, Chap. 4, as well as McGovern and Groves, Chap. 5. Changes in chromatin structure can occur via either acetylation or methylation of histones or DNA.  Several studies have examined how histone acetylation affects hair cell regeneration in chicks and fish. Slattery et al. (2009) showed that treatment with several inhibitors of histone deacetylases (HDACs) resulted in decreased proliferation of supporting cells but did not affect hair cell replacement. A similar study examined the effects of the HDAC inhibitors on regeneration of lateral line hair cells in larval zebrafish and observed reduced supporting cell division and fewer regenerated hair cells (He et al. 2014). Since histone acetylation is typically associated with “open” chromatin, these results suggest that the expression of certain genes needs to be repressed (via deacetylation of histones) in order for regeneration to occur. He et  al. (2016) further found that an inhibitor of histone methylation, lysine-specific demethylase 1 (Lsd1), may also be a significant regulator of hair cell regeneration in zebrafish neuromasts. Blockade of Lsd1 using a pharmaceutical agent reduced supporting cell division after neomycin treatment (He et al. 2016) and also reduced cell division in developing neuromasts (He et al. 2013). Together, the results of these three studies imply that epigenetic changes may be necessary for the induction of regeneration in nonmammals. However, all studies employed pharmacological inhibitors that may have “off-target” effects. Additional studies are needed to identify regeneration-relevant genes that are impacted by epigenetic modifications.

2.6 Conclusion Nonmammalian vertebrates have a natural capacity to regenerate hair cells in their lateral line, hearing, and vestibular sensory organs. Chickens and zebrafish have been the most common nonmammalian animal subjects for scientists to build an understanding of the cellular and molecular mechanisms that stimulate supporting cells to form new hair cells after damage. It is anticipated that studies in these species will continue to point us in the direction of therapies that can promote hair cell regeneration in mammals and result in improved hearing and balance function following hair cell loss. Acknowledgments  The authors thank Mark Warchol, Allison Coffin, and Art Popper for comments on this manuscript. Compliance with Ethics Requirements  Madeleine N. Hewitt declares that she has no conflict of interest. David W. Raible declares that he has no conflict of interest. Jennifer S. Stone declares that she has no conflict of interest.

34

M. N. Hewitt et al.

References Adler HJ, Raphael Y (1996) New hair cells arise from supporting cell conversion in the acoustically damaged chick inner ear. Neurosci Lett 205:17–20 Alvarado DM, Hawkins RD, Bashiardes S et al (2011) An RNA interference-based screen of transcription factor genes identifies pathways necessary for sensory regeneration in the avian inner ear. J Neurosci 31:4535–4543 Avallone B, Porritiello M, Esposito D et al (2003) Evidence for hair cell regeneration in the crista ampullaris of the lizard Podarcis sicula. Hearing Res 178:79–88 Bai H, Yang S, Xi C et al (2020) Signaling pathways (Notch, Wnt, Bmp and Fgf) have additive effects on hair cell regeneration in the chick basilar papilla after streptomycin injury in vitro: Additive effects of signaling pathways on hair cell regeneration. Hear Res 401:108161. https:// doi.org/10.1016/j.heares.2020.108161 Baird RA, Torres MA, Schuff NR (1993) Hair cell regeneration in the bullfrog vestibular otolith organs following aminoglycoside toxicity. Hearing Res 65:164–174 Baird RA, Steyger PS, Schuff NR (1996) Mitotic and nonmitotic hair cell regeneration in the bullfrog vestibular otolith organs. Ann N Y Acad Sci. 781:59–70 Baird RA, Burton MD, Fashena DS, Naeger RA (2000) Hair cell recovery in mitotically blocked cultures of the bullfrog saccule. Proc Natl Acad Sci U S A 97:11722–11729 Balak KJ, Corwin JT, Jones JE (1990) Regenerated hair cells can originate from supporting cell progeny: evidence from phototoxicity and laser ablation experiments in the lateral line system. J Neurosci 10:2502–2512 Behra M, Bradsher J, Sougrat R et al (2009) Phoenix is required for mechanosensory hair cell regeneration in the zebrafish lateral line. PLoS Genet 5(4):e1000455. https://doi.org/10.1371/ journal.pgen.1000455 Behra M, Gallardo VE, Bradsher J et al (2012) Transcriptional signature of accessory cells in the lateral line, using the Tnk1bp1:EGFP transgenic zebrafish line. BMC Dev Biol 12:6. https:// doi.org/10.1186/1471-­213X-­12-­6 Bermingham NA, Hassan BA, Price SD et al (1999) Math1: an essential gene for the generation of inner ear hair cells. Science. 284:1837–1841 Bermingham-McDonogh O, Rubel EW (2003) Hair cell regeneration: winging our way towards a sound future. Curr Opin Neurobiol 13:119–126 Breitzler L, Lau IH, Fonseca PJ et al (2020) Noise-induced hearing loss in zebrafish: investigating structural and functional inner ear damage and recovery. Hear Res 391:107952 Breuskin I, Bodson M, Thelen N et al (2008) Strategies to regenerate hair cells: identification of progenitors and critical genes. Hearing Res 236:1–10 Brignull HR, Raible DW, Stone JS (2009) Feathers and fins: Non-mammalian models for hair cell regeneration. Brain Res 1277:12–23 Cafaro J, Lee GS, Stone JS (2007) Atoh1 expression defines activated progenitors and differentiating hair cells during avian hair cell regeneration. Dev Dynam 236:156–170 Carey JP, Fuchs AF, Rubel EW (1996) Hair cell regeneration and recovery of the vestibuloocular reflex in the avian vestibular system. J Neurophysiol 76:3301–3312 Chitnis AB, Dalle Nogare D, Matsuda M (2012) Building the posterior lateral line system in zebrafish. Dev Neurobiol 72:234–255 Chrysostomou E, Zhou L, Darcy YL et al (2020) The Notch ligand Jagged1 is required for the formation, maintenance, and survival of Hensen’s Cells in the mouse cochlea. J Neurosci. 40(49):9401–9413 Coffin AB, Mohr RA, Sisneros JA (2012) Saccular-specific hair cell addition correlates with reproductive state-dependent changes in the auditory saccular sensitivity of a vocal fish. J Neurosci 32:1366–1376 Corwin JT (1981) Postembryonic production and aging of inner ear hair cells in sharks. J Comp Neurol 201:541–553

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

35

Corwin JT (1985) Perpetual production of hair cells and maturational changes in hair cell ultrastructure accompany postembryonic growth in an amphibian ear. Proc Natl Acad Sci U S A 82:3911–3915 Corwin JT, Cotanche DA (1988) Regeneration of sensory hair cells after acoustic trauma. Science. 240:1772–1774 Cotanche DA (1987) Regeneration of hair cell stereociliary bundles in the chick cochlea following severe acoustic trauma. Hearing Res 30:181–195 Cotanche DA, Kaiser CL (2010) Hair cell fate decisions in cochlear development and regeneration. Hearing Res 266:18–25 Cruz RM, Lambert PR, Rubel EW (1987) Light microscopic evidence of hair cell regeneration after gentamicin toxicity in chick cochlea. Arch Otolaryngol Head Neck Surg 113:1058–1062 Cruz IA, Kappedal R, Mackenzie SM et al (2015) Robust regeneration of adult zebrafish lateral line hair cells reflects continued precursor pool maintenance. Dev Biol 402:229–238 Daudet N, Gibson R, Shang J et al (2009) Notch regulation of progenitor cell behavior in quiescent and regenerating auditory epithelium of mature birds. Dev Biol 326:86–100 Dickman JD, Lim I (2004) Posture, head stability, and orientation recovery during vestibular regeneration in pigeons. J Assoc Res Otolaryngol 5:323–336 Doetzlhofer A, Avraham KB (2017) Insights into inner ear-specific gene regulation: Epigenetics and non-coding RNAs in inner ear development and regeneration. Semin Cell Dev Biol 65:69–79 Dooling RJ, Ryals BM, Manabe K (1997) Recovery of hearing and vocal behavior after hair-cell regeneration. Proc Natl Acad Sci U S A 94:14206–14210 Dufourcq P, Roussigné M, Blader P et al (2006) Mechano-sensory organ regeneration in adults: The zebrafish lateral line as a model. Mol Cell Neurosci 33:180–187 Faucherre A, Pujol-Martí J, Kawakami K et al (2009) Afferent neurons of the zebrafish lateral line are strict selectors of hair-cell orientation. PLoS One 4:e4477. https://doi.org/10.1371/journal. pone.0004477 Frucht CS, Uduman M, Duke JL et al (2010) Gene expression analysis of forskolin treated basilar papillae identifies microRNA181a as a mediator of proliferation. PLoS One 5(7):e11502. https://doi.org/10.1371/journal.pone.0011502 Frucht CS, Santos-Sacchi J, Navaratnam DS (2011) MicroRNA181a plays a key role in hair cell regeneration in the avian auditory epithelium. Neurosci Lett 493(1–2):44–48 Giffen KP, Liu H, Kramer KL, He DZ (2019) Expression of protein-coding gene orthologs in zebrafish and mouse inner ear non-sensory supporting cells. Front Neurosci 13:1117. https:// doi.org/10.3389/fnins.2019.01117 Goode CT, Carey JP, Fuchs AF et al (1999) Recovery of the vestibulocolic reflex after aminoglycoside ototoxicity in domestic chickens. J Neurophysiol 81:1025–1035 Gou Y, Guo J, Maulding K et al (2018a) sox2 and sox3 cooperate to regulate otic/epibranchial placode induction in zebrafish. Dev Biol 435:84–95 Gou Y, Vemaraju S, Sweet EM et al (2018b) sox2 and sox3 play unique roles in development of hair cells and neurons in the zebrafish inner ear. Dev Biol 435:73–83 Grant KA, Raible DW, Piotrowski T (2005) Regulation of latent sensory hair cell precursors by glia in the zebrafish lateral line. Neuron 45:69–80 Haddon C, Jiang YJ, Smithers L et al (1998) Delta-Notch signalling and the patterning of sensory cell differentiation in the zebrafish ear: evidence from the mind bomb mutant. Development 125:4637–4644 Haque A, Zakir M, Dickman JD (2009) Regeneration of vestibular horizontal semicircular canal afferents in pigeons. J Neurophysiol 102:1274–1286 Hardy K, Amariutei AE, De Faveri F et al (2021) Functional development and regeneration of hair cells in the zebrafish lateral line. J Physiol. https://doi.org/10.1113/JP281522 Harris JA, Cheng AG, Cunningham LL et al (2003) Neomycin-induced hair cell death and rapid regeneration in the lateral line of zebrafish (Danio rerio). J Assoc Res Otolaryngol 4:219–234

36

M. N. Hewitt et al.

Hawkins RD, Bashiardes S, Helms CA et  al (2003) Gene expression differences in quiescent versus regenerating hair cells of avian sensory epithelia: implications for human hearing and balance disorders. Hum Mol Genet 12:1261–1272 Hawkins RD, Bashiardes S, Powder KE et  al (2007) Large scale gene expression profiles of regenerating inner ear sensory epithelia. PLoS One 2(6):e525. https://doi.org/10.1371/journal. pone.0000525 He Y, Yu H, Sun S et al (2013) Trans-2-phenylcyclopropylamine regulates zebrafish lateral line neuromast development mediated by depression of LSD1 activity. Int J Dev Biol 57:365–373 He Y, Cai C, Tang D et al (2014) Effect of histone deacetylase inhibitors trichostatin A and valproic acid on hair cell regeneration in zebrafish lateral line neuromasts. Front Cell Neurosci 8:382. https://doi.org/10.3389/fncel.2014.00382 He Y, Tang D, Cai C et al (2016) LSD1 is required for hair cell regeneration in zebrafish. Mol Neurobiol 53:2421–2434 Head JR, Gacioch L, Pennisi M et  al (2013) Activation of canonical Wnt/β-catenin signaling stimulates proliferation in neuromasts in the zebrafish posterior lateral line. Dev Dynam 242:832–846 Hernández PP, Olivari FA, Sarrazin AF et al (2007) Regeneration in zebrafish lateral line neuromasts: Expression of the neural progenitor cell marker sox2 and proliferation-dependent and -independent mechanisms of hair cell renewal. Dev Neurobiol 67:637–654 Itoh M, Chitnis AB (2001) Expression of proneural and neurogenic genes in the zebrafish lateral line primordium correlates with selection of hair cell fate in neuromasts. Mech Develop 102:263–266 Iwasaki M, Yokoi H, Suzuki T et al (2020) Development of the anterior lateral line system through local tissue-tissue interactions in the zebrafish head. Dev Dynam 249:1440–1454 Jacobo A, Dasgupta A, Erzberger A et  al (2019) Notch-mediated determination of hair-bundle polarity in mechanosensory hair cells of the zebrafish lateral line. Curr Biol 29:3579–3587.e7 Jacques BE, Montgomery WH, Uribe PM et  al (2014) The role of Wnt/β-catenin signaling in proliferation and regeneration of the developing basilar papilla and lateral line. Dev Neurobiol 74:438–456 Janesick A, Scheibinger M, Benkafadar N et al (2021) Cell-type identity of the avian cochlea. Cell Rep 34(12):108900. https://doi.org/10.1016/j.celrep.2021.108902 Ji YR, Warrier S, Jiang T et al (2018) Directional selectivity of afferent neurons in zebrafish neuromasts is regulated by Emx2 in presynaptic hair cells. eLife 7:e35796. https://doi.org/10.7554/ eLife.35796 Jiang L, Romero-Carvajal A, Haug JS et al (2014) Gene-expression analysis of hair cell regeneration in the zebrafish lateral line. Proc Natl Acad Sci U S A 111:E1383–E1392 Jiang T, Kindt K, Wu DK (2017) Transcription factor Emx2 controls stereociliary bundle orientation of sensory hair cells. eLife 6:e23661. https://doi.org/10.7554/eLife.23661 Jiang L, Xu J, Jin R et  al (2018) Transcriptomic analysis of chicken cochleae after gentamicin damage and the involvement of four signaling pathways (Notch, FGF, Wnt and BMP) in hair cell regeneration. Hearing Res 361:66–79 Jones JE, Corwin JT (1993) Replacement of lateral line sensory organs during tail regeneration in salamanders: identification of progenitor cells and analysis of leukocyte activity. J Neurosci 13:1022–1034 Jones TA, Jones SM, Paggett KC (2006a) Emergence of hearing in the chicken embryo. J Neurophysiol 96(1):128–141 Jones JM, Montcouquiol M, Dabdoub A et al (2006b) Inhibitors of differentiation and DNA binding (Ids) regulate Math1 and hair cell formation during the development of the Organ of Corti. J Neurosci 26:550–558 Jørgensen JM, Mathiesen C (1988) The avian inner ear. Continuous production of hair cells in vestibular sensory organs, but not in the auditory papilla. Die Naturwissenschaften 75:319–320 Kamaid A, Neves J, Giráldez F (2010) Id gene regulation and function in the prosensory domains of the chicken inner ear: a link between Bmp signaling and Atoh1. J Neurosci 30:11426–11434

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

37

Kelley MW (2003) Cell adhesion molecules during inner ear and hair cell development, including notch and its ligands. Curr Top Dev Biol. 57:321–356 Kelley M, Stone JS (2016) Hair cell development and regeneration. In: Salvi R, Fay R, Popper A (eds) Auditory Hair Cell Protection and Regeneration. Springer-Verlag, New York Kiernan AE, Cordes R, Kopan R et al (2005) The Notch ligands DLL1 and JAG2 act synergistically to regulate hair cell development in the mammalian inner ear. Development 132:4353–4362 Kil J, Warchol ME, Corwin JT (1997) Cell death, cell proliferation, and estimates of hair cell life spans in the vestibular organs of chicks. Hearing Res 114:117–126 Kozak EL, Palit S, Miranda-Rodríguez JR et al (2020) Epithelial planar bipolarity emerges from Notch-mediated asymmetric inhibition of Emx2. Curr Biol 30:1142–1151 Ku Y-C, Renaud NA, Veile RA et al (2014) The transcriptome of utricle hair cell regeneration in the avian inner ear. J Neurosci 34:3523–3535 Lanford PJ, Presson JC, Popper AN (1996) Cell proliferation and hair cell addition in the ear of the goldfish, Carassius auratus. Hearing Res 100:1–9 Lee SG, Huang M, Obholzer ND et al (2016) Myc and Fgf are required for zebrafish neuromast hair cell regeneration. PLoS One 11:e0157768. https://doi.org/10.1371/journal.pone.0157768 Lewis ER, Li CW (1973) Evidence concerning the morphogenesis of saccular receptors in the bullfrog (Rana catesbeiana). J Morphol 139:351–361 Lewis RM, Hume CR, Stone JS (2012) Atoh1 expression and function during auditory hair cell regeneration in post-hatch chickens. Hear Res 289:74–85 Lewis RM, Keller JJ, Wan L et  al (2018) Bone morphogenetic protein 4 antagonizes hair cell regeneration in the avian auditory epithelium. Hearing Res 364:1–11 Li T, Giagtzoglou N, Eberl DF et al (2016) The E3 ligase Ubr3 regulates Usher syndrome and MYH9 disorder proteins in the auditory organs of Drosophila and mammals. eLife 5:e15258. https://doi.org/10.7554/eLife.15258 Li T, Bellen HJ, Groves AK (2018) Using Drosophila to study mechanisms of hereditary hearing loss. Dis Model Mech 11:dmm031492. https://doi.org/10.1242/dmm.031492 Liang J, Wang D, Renaud G et al (2012) The stat3/socs3a pathway is a key regulator of hair cell regeneration in zebrafish. J Neurosci 32:10662–10673 Liu Z, Dearman JA, Cox BC et al (2012) Age-dependent in vivo conversion of mouse cochlear pillar and Deiters’ cells to immature hair cells by Atoh1 ectopic expression. J Neurosci 32:6600–6610 Lush ME, Diaz DC, Koenecke N et al (2019) scRNA-Seq reveals distinct stem cell populations that drive hair cell regeneration after loss of Fgf and Notch signaling. eLife 8:e44431. https:// doi.org/10.7554/eLife.44431 Ma EY, Raible DW (2009) Signaling pathways regulating zebrafish lateral line development. Curr Biol 19:R381–R386 Ma EY, Rubel EW, Raible DW (2008) Notch signaling regulates the extent of hair cell regeneration in the zebrafish lateral line. J Neurosci 28:2261–2273 Mackenzie SM, Raible DW (2012) Proliferative regeneration of zebrafish lateral line hair cells after different ototoxic insults. PLoS One 7:e47257. https://doi.org/10.1371/journal.pone.0047257 Marean GC, Burt JM, Beecher MD et al (1993) Hair cell regeneration in the European starling (Sturnus vulgaris): Recovery of pure-tone detection thresholds. Hearing Res 71:125–136 Matsui JI, Oesterle EC, Stone JS et al (2000) Characterization of damage and regeneration in cultured avian utricles. J Assoc Res Otolaryngol 1:46–63 Matsui JI, Haque A, Huss D et al (2003) Caspase inhibitors promote vestibular hair cell survival and function after aminoglycoside treatment in vivo. J Neurosci 23:6111–6122 Matsunaga M, Kita T, Yamamoto R et al (2020) Initiation of supporting cell activation for hair cell regeneration in the avian auditory epithelium: An explant culture model. Front Cell Neurosci 14:583994. https://doi.org/10.3389/fncel.2020.583994 McHenry MJ, Feitl KE, Strother JA et al (2009) Larval zebrafish rapidly sense the water flow of a predator’s strike. Biol Lett 5:477–479

38

M. N. Hewitt et al.

Mekdara PJ, Schwalbe MAB, Coughlin LL et  al (2018) The effects of lateral line ablation and regeneration in schooling giant danios. J Exp Biol 22:175166. https://doi.org/10.1242/ jeb.175166 Millimaki BB, Sweet EM, Dhason MS et al (2007) Zebrafish atoh1 genes: classic proneural activity in the inner ear and regulation by Fgf and Notch. Development 134:295–305 Millimaki BB, Sweet EM, Riley BB (2010) Sox2 is required for maintenance and regeneration, but not initial development, of hair cells in the zebrafish inner ear. Dev Biol 338:262–269 Monaghan JR, Maden M (2012) Visualization of retinoic acid signaling in transgenic axolotls during limb development and regeneration. Dev Biol 368:63–75 Montcouquiol M, Corwin JT (2001) Brief treatments with forskolin enhance S-phase entry in balance epithelia from the ears of rats. J Neurosci 21:974–982 Morest DK, Cotanche DA (2004) Regeneration of the inner ear as a model of neural plasticity. J Neurosci Res 78:455–460 Nagiel A, Andor-Ardó D, Hudspeth AJ (2008) Specificity of afferent synapses onto plane-­polarized hair cells in the posterior lateral line of the zebrafish. J Neurosci 28:8442–8453 Namdaran P, Reinhart KE, Owens KN et al (2012) Identification of modulators of hair cell regeneration in the zebrafish lateral line. J Neurosci 32:3516–3528 Neves J, Kamaid A, Alsina B et al (2007) Differential expression of Sox2 and Sox3 in neuronal and sensory progenitors of the developing inner ear of the chick. J Comp Neurol 503:487–500 Niemiec AJ, Raphael Y, Moody DB (1994) Return of auditory function following structural regeneration after acoustic trauma: behavioral measures from quail. Hear Res 75(1–2):209–224. https://doi.org/10.1016/0378-­5955(94)90072-­8 Nuñez VA, Sarrazin AF, Cubedo N et al (2009) Postembryonic development of the posterior lateral line in the zebrafish. Evol Dev 11:391–404 Oesterle EC, Rubel EW (1993) Postnatal production of supporting cells in the chick cochlea. Hear Res 66:213–224 Oesterle EC, Stone JS (2008) Hair cell regeneration: Mechanisms guiding cellular proliferation and differentiation. In: Salvi RJ, Popper AN, Fay RR (eds) Hair Cell Regeneration, Repair, and Protection. Springer, New York, pp 141–197 Oesterle EC, Campbell S, Taylor RR et  al (2008) Sox2 and Jagged1 expression in normal and drug-damaged adult mouse inner ear. J Assoc Res Otolaryngol 9:65–89 Pei W, Huang SC, Xu L et al (2016) Loss of Mgat5a-mediated N-glycosylation stimulates regeneration in zebrafish. Cell Regen. https://doi.org/10.1186/s13619-­016-­0031-­5 Pei W, Xu L, Huang SC et  al (2018) Guided genetic screen to identify genes essential in the regeneration of hair cells and other tissues. NPJ Regen Med. https://doi.org/10.1038/ s41536-­018-­0050-­7 Pinto-Teixeira F, Viader-Llargués O, Torres-Mejía E et al (2015) Inexhaustible hair-cell regeneration in young and aged zebrafish. Biol Open 4:903–909 Pittlik S, Begemann G (2012) New sources of retinoic acid synthesis revealed by live imaging of an Aldh1a2-GFP reporter fusion protein throughout zebrafish development. Dev Dyn 241:1205–1216 Popper AN, Hoxter B (1984) Growth of a fish ear: 1. Quantitative analysis of hair cell and ganglion cell proliferation. Hear Res 15:133–142 Presson JC, Lanford PJ, Popper AN (1996) Hair cell precursors are ultrastructurally indistinguishable from mature support cells in the ear of a postembryonic fish. Hear Res 100:10–20 Rhinn M, Dollé P (2012) Retinoic acid signalling during development. Development 139:843–858 Roberson DF, Weisleder P, Bohrer PS, Rubel EW (1992) Ongoing production of sensory cells in the vestibular epithelium of the chick. Hearing Res 57:166–174 Roberson DW, Alosi JA, Cotanche DA (2004) Direct transdifferentiation gives rise to the earliest new hair cells in regenerating avian auditory epithelium. J Neurosci Res 78:461–471 Romero-Carvajal A, Navajas Acedo J, Jiang L et  al (2015) Regeneration of sensory hair cells requires localized interactions between the Notch and Wnt pathways. Dev Cell 34:267–282

2  Nonmammalian Hair Cell Regeneration: Cellular Mechanisms of Morphological…

39

Rubbini D, Robert-Moreno À, Hoijman E et al (2015) Retinoic acid signaling mediates hair cell regeneration by repressing p27kip and sox2 in supporting cells. J Neurosci 35:15752–15766 Rubel EW, Furrer SA, Stone JS (2013) A brief history of hair cell regeneration research and speculations on the future. Hear Res 297:42–51 Rudolf MA, Andreeva A, Kozlowski MM et  al (2020) YAP mediates hair cell regeneration in balance organs of chickens, but LATS kinases suppress its activity in mice. J Neurosci 40:3915–3932 Ryals BM, Rubel EW (1988) Hair cell regeneration after acoustic trauma in adult Coturnix quail. Science 240:1774–1776 Ryals BM, Dent ML, Dooling RJ (2013) Return of function after hair cell regeneration. Hear Res 297:113–120 Scheibinger M, Ellwanger DC, Corrales CE et al (2018) Aminoglycoside damage and hair cell regeneration in the chicken utricle. J Assoc Res Otolaryngol 19:17–29 Schuck JB, Smith ME (2009) Cell proliferation follows acoustically-induced hair cell bundle loss in the zebrafish saccule. Hearing Res 253:67–76 Schuck JB, Sun H, Penberthy WT et al (2011) Transcriptomic analysis of the zebrafish inner ear points to growth hormone mediated regeneration following acoustic trauma. BMC Neurosci 12:88. https://doi.org/10.1186/1471-­2202-­12-­88 Seleit A, Krämer I, Riebesehl BF et al (2017) Neural stem cells induce the formation of their physical niche during organogenesis. eLife 6:e29173. https://doi.org/10.7554/eLife.29173 Shang J, Cafaro J, Nehmer R et al (2010) Supporting cell division is not required for regeneration of auditory hair cells after ototoxic injury in vitro. J Assoc Res Otolaryngol 11:203–222 Slattery EL, Speck JD, Warchol ME (2009) Epigenetic influences on sensory regeneration: Histone deacetylases regulate supporting cell proliferation in the avian utricle. J Assoc Res Otolaryngol 10:341–353 Song J, Yan HY, Popper AN (1995) Damage and recovery of hair cells in fish canal (but not superficial) neuromasts after gentamicin exposure. Hearing Res 91:63–71 Steiner AB, Kim T, Cabot V et al (2014) Dynamic gene expression by putative hair-cell progenitors during regeneration in the zebrafish lateral line. Proc Natl Acad Sci U S A 111:E1393–E1401 Stojanova ZP, Kwan T, Segil N (2015) Epigenetic regulation of Atoh1 guides hair cell development in the mammalian cochlea. Development 142:3529–3536 Stone LS (1937) Further experimental studies of the development of lateral-line sense organs in amphibians observed in living preparations. J Comp Neurol 68:83–115 Stone JS, Cotanche DA (2007) Hair cell regeneration in the avian auditory epithelium. Int J Dev Biol 51:633–647 Stone JS, Rubel EW (1999) Delta1 expression during avian hair cell regeneration. Development 126:961–973 Stone JS, Rubel EW (2000) Temporal, spatial, and morphologic features of hair cell regeneration in the avian basilar papilla. J Comp Neurol 417:1–16 Stone JS, Leaño SG, Baker LP et al (1996) Hair cell differentiation in chick cochlear epithelium after aminoglycoside toxicity: In vivo and in vitro observations. J Neurosci 16:6157–6174 Suli A, Watson GM, Rubel EW et al (2012) Rheotaxis in larval zebrafish is mediated by lateral line mechanosensory hair cells. PLoS One 7:e29727. https://doi.org/10.1371/journal.pone.0029727 Suli A, Pujol R, Cunningham DE et al (2016) Innervation regulates synaptic ribbons in lateral line mechanosensory hair cells. J Cell Sci 129:2250–2260 Sun H, Lin C-H, Smith ME (2011) Growth hormone promotes hair cell regeneration in the zebrafish (Danio rerio) inner ear following acoustic trauma. PLoS One 6:e28372. https://doi. org/10.1371/journal.pone.0028372 Thomas ED, Raible DW (2019) Distinct progenitor populations mediate regeneration in the zebrafish lateral line. eLife 8:e43736. https://doi.org/10.7554/eLife.43736 Tilney LG, Cotanche DA, Tilney MS (1992) Actin filaments, stereocilia and hair cells of the bird cochlea. VI.  How the number and arrangement of stereocilia are determined. Development 116(1):213–226

40

M. N. Hewitt et al.

Viader-Llargués O, Lupperger V, Pola-Morell L et  al (2018) Live cell-lineage tracing and machine learning reveal patterns of organ regeneration. eLife Sciences 7:e30823. https://doi. org/10.7554/eLife.30823 Wada H, Ghysen A, Satou C et al (2010) Dermal morphogenesis controls lateral line patterning during postembryonic development of teleost fish. Dev Biol 340:583–594 Wakasaki T, Niiro H, Jabbarzadeh-Tabrizi S et al (2017) Musashi-1 is the candidate of the regulator of hair cell progenitors during inner ear regeneration. BMC Neurosci 18:64. https://doi. org/10.1186/s12868-­017-­0382-­z Wan G, Corfas G, Stone JS (2013) Inner ear supporting cells: Rethinking the silent majority. Semin Cell Dev Biol 24:448–459 Wan L, Lovett M, Warchol ME et  al (2020) Vascular endothelial growth factor is required for regeneration of auditory hair cells in the avian inner ear. Hearing Res 385:107839. https://doi. org/10.1016/j.heares.2019.107839 Wang RN, Green J, Wang Z et al (2014) Bone Morphogenetic Protein (BMP) signaling in development and human diseases. Genes Dis 1:87–105 Warchol ME (2011) Sensory regeneration in the vertebrate inner ear: Differences at the levels of cells and species. Hearing Res 273:72–79 Warchol ME, Schwendener RA, Hirose K (2012) Depletion of resident macrophages does not alter sensory regeneration in the avian cochlea. PLoS One 7:e51574. https://doi.org/10.1371/ journal.pone.0051574 Warchol ME, Stone J, Barton M et al (2017) ADAM10 and γ-secretase regulate sensory regeneration in the avian vestibular organs. Dev Biol 428:39–51 Warchol ME, Schrader A, Sheets L (2021) Macrophages respond rapidly to ototoxic injury of lateral line hair cells but are not required for hair cell regeneration. Front Cell Neurosci 14:613246. https://doi.org/10.3389/fncel.2020.613246 White PM, Stone JS, Groves AK et al (2012) EGFR signaling is required for regenerative proliferation in the cochlea: Conservation in birds and mammals. Dev Biol 363:191–200 Wibowo I, Pinto-Teixeira F, Satou C et  al (2011) Compartmentalized Notch signaling sustains epithelial mirror symmetry. Development 138:1143–1152 Williams JA, Holder N (2000) Cell turnover in neuromasts of zebrafish larvae. Hearing Res 143:171–181 Witte MC, Montcouquiol M, Corwin JT (2001) Regeneration in avian hair cell epithelia: identification of intracellular signals required for S-phase entry. Eur J Neurosci 14:829–838 Woolley SMN, Rubel EW (2002) Vocal memory and learning in adult bengalese finches with regenerated hair cells. J Neurosci 22:7774–7787 Woolley SMN, Wissman AM, Rubel EW (2001) Hair cell regeneration and recovery of auditory thresholds following aminoglycoside ototoxicity in Bengalese finches. Hearing Res 153:181–195 Ye Z, Su Z, Xie S et al (2020) Yap-lin28a axis targets let7-Wnt pathway to restore progenitors for initiating regeneration. eLife 9:e55771. https://doi.org/10.7554/eLife.55771 Yizhar-Barnea O, Valensisi C, Jayavelu ND et  al (2018) DNA methylation dynamics during embryonic development and postnatal maturation of the mouse auditory sensory epithelium. Sci Rep 8:17348. https://doi.org/10.1038/s41598-­018-­35587-­x Zheng JL, Shou J, Guillemot F et al (2000) Hes1 is a negative regulator of inner ear hair cell differentiation. Development 127:4551–4560

Chapter 3

Cell Junctions and the Mechanics of Hair Cell Regeneration Mark A. Rudolf and Jeffrey T. Corwin

Abstract  Hundreds of millions of people worldwide have disabling hearing loss or imbalance. A major cause of these impairments is sensory hair cell loss from loud sounds, ototoxic drugs, and aging. Birds, fish, and amphibians recover hearing and vestibular function after hair cell loss, as neighboring supporting cells retain a lifelong capacity to divide and differentiate into replacement hair cells. In mammals, supporting cell plasticity declines sharply during maturation. Thus, major research goals are to define the mechanisms of hair cell replacement in nonmammals and identify unique maturational changes in mammalian supporting cells that restrict these regenerative mechanisms. After the first report of self-repair in the chicken basilar papilla described pronounced shape changes of supporting cells within acoustic lesions, it was hypothesized that mechanical forces associated with that shape change trigger regenerative proliferation. Subsequent investigations found that age-related changes in the intercellular junctions of mammalian supporting cells strongly correlate with reductions in cellular shape change and supporting cell proliferation, while supporting cells in chickens do not change with age and retain proliferative capacity. Mammalian supporting cell junctions are bracketed by exceptionally thick circumferential F-actin bands and express high levels of E-cadherin, which is barely detectable in most nonmammalian supporting cells. This chapter outlines the hypothesis that these specialized junctions of mammalian supporting cells impede changes in tension that occur in nonmammalian supporting cells, reviews findings implicating the mechano-responsive Hippo and Wnt pathways in regenerative proliferation, and highlights unanswered questions regarding the role of epithelial mechanics in hair cell regeneration. Keywords  Inner ear · Vestibular · Auditory · Utricle · Cochlea · Adherens junction · E-cadherin · F-actin · YAP · Hippo · β-catenin · Wnt M. A. Rudolf (*) · J. T. Corwin Department of Neuroscience, University of Virginia School of Medicine, Charlottesville, VA, USA Department of Cell Biology, University of Virginia School of Medicine, Charlottesville, VA, USA e-mail: [email protected]; [email protected] © Springer Nature Switzerland AG 2023 M. E. Warchol et al. (eds.), Hair Cell Regeneration, Springer Handbook of Auditory Research 75, https://doi.org/10.1007/978-3-031-20661-0_3

41

42

M. A. Rudolf and J. T. Corwin

3.1 Introduction To contextualize this chapter and the open questions and research opportunities it reviews, a brief history of how the topic of cell junctions and mechanical forces gained attention may be helpful. After discovering that the ears of sharks produce and accumulate hundreds of thousands of hair cells (HCs) throughout their life span (Corwin 1981), the senior author presented autoradiographic evidence that supporting cell (SC) divisions were the source of those newly produced HCs. Based on those findings, it was proposed that the ears of nonmammalian vertebrates might be capable of regenerating HCs and recovering from damage and that the ears of mammals might add new HCs at levels so low as to have been unrecognized. A few years later, Douglas Cotanche presented striking scanning electron microscopic evidence that rapid self-repair occurred at the surface of the basilar papilla sensory epithelium in the cochleae of young chickens just days after the ear had been damaged by loud sound (Cotanche 1987). At that point, it was unclear whether the self-repair occurred via regrowth of damaged HC bundles or through the replacement of lost HCs. Administering tritiated thymidine to chickens during the week following the acoustic damage showed that the repair resulted from proliferation, since radioactively labeled DNA was present in the nuclei of SCs and newly produced HCs at the sites of acoustic lesions (Corwin and Cotanche 1988; Ryals and Rubel 1988). Although other cells were proposed as the source of the HCs produced during regeneration, recovery after laser microbeam ablation of HCs in lateral line neuromasts was followed in continuous time-lapse imaging and established that SCs are the source of the regenerated HCs (Balak et al. 1990; Jones and Corwin 1996). Those discoveries gave rise to the central question: “What differences between mammalian and nonmammalian SCs, or the signals they encounter, cause them to have different capacities for producing replacement HCs after damage?” The specialized structures and differentiation of SCs in the mammalian organ of Corti clearly differ from those attributes of SCs in the hearing organs in nonmammals. Unique to therian (placental and marsupial) mammals, the organ of Corti appears to have arisen in evolution approximately 220 million years ago; more than 100 million years after the origins of ancestral fish, amphibians, and reptiles; and about 55–70 million years before the first birds are believed to have evolved from the therapod dinosaurs. Seven named SC types (inner border cells, inner and outer pillar cells, three rows of Deiters’ cells, Hensen’s cells, Boettcher’s cells, and Claudius cells) are distinguishable from each other due to their specialized structures, their precise interfaces with cellular neighbors, and their stereotyped positions along the organ of Corti’s highly ordered longitudinal and radial axes. In contrast to the organ of Corti, vestibular epithelia in humans and other mammals appear quite similar, although not identical to their counterparts in fish, amphibians, reptiles, and birds. Knowing this, the vestibular epithelia from humans and other mammals were investigated, which were soon found to express the basic machinery for HC replacement (Forge et al. 1993; Warchol et al. 1993). However, those and subsequent investigations have shown that the self-repair processes in the

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

43

vestibular organs of mature mammals operate at considerably lower rates and differ in other ways from what had been observed in birds, fish, and amphibians. After those discoveries, more laboratories turned their attention to the possibility of discovering therapeutic means for enhancing the regeneration of HCs in the ears of humans and other mammals. Initial efforts assessed SC responsiveness in tests of various mitogenic growth factors that were known to promote mitosis in some other cells and tissues. The neuregulins GGF2 and heregulin, EGF, TGF-α, FGF2, IGF-1, insulin, and the undefined mix of growth factors present in serum were all found to increase the incidence of proliferation in vestibular SCs that had been harvested from the ears of young rodents (Montcouquiol and Corwin 2001). It was soon discovered, however, that the proliferative responsiveness of the SCs within sheets of vestibular epithelium sharply declined when the epithelia were harvested from rodents at progressively older ages during the first month of life (Gu et al. 2007). The degree of decline was consistent with increasing maturational age and independent of whether the epithelia were treated with GGF2, insulin, or serum, which suggested that it did not depend on the downregulation of specific growth factor receptors or other components within the different growth factor pathways. Rather, the findings suggested that maturation of the SCs was inhibiting proliferative responsiveness at a level that could act downstream of the growth factor pathways (Burns and Corwin 2013). The “Parking Brake Hypothesis” was proposed on the basis of those results and posited that maturing SCs in mammalian ears develop unique characteristics that limit their capacity to respond to the mitogenic signals that effectively stimulate SC proliferation in young mammals and throughout the life span in nonmammals. Growth factor stimulation increases proliferation in neonatal mammals like force on a car’s accelerator pedal increases speed, and the unidentified maturation-dependent mechanism that appeared to inhibit such proliferative responses was envisioned as a downstream limit analogous to the secondary limit a car’s parking brake provides downstream of the accelerator pedal. The hunt for the biological basis of that parking brake eventually led to discoveries about the intercellular junctions and cytoskeletons of SCs in mammals and nonmammals as well as to discoveries about epithelial mechanics, mechanically controlled signaling pathways within those cells, and many yet-to-be-answered questions, which this chapter will cover.

3.2 Shape Change Controls Proliferation of Supporting Cells What could be responsible for “parking brake” inhibition that restricts proliferative responses of SCs in mature mammals, but not those in embryonic mammals or the ears of young and mature birds, fish, and amphibians? Earlier research provided clues suggesting that mechanical signals arising from changes in cell shape and modulation of interepithelial tension might influence the regenerative replacement of HCs.

44

M. A. Rudolf and J. T. Corwin

Fig. 3.1  Regeneration of HCs occurs in the chicken basilar papilla after prolonged exposure to loud sound. Arrows show expansion of SC surfaces. Arrowhead shows damaged hair bundle. (Adapted with permission and source from Cotanche 1987)

As mentioned in Sect. 3.1, HC loss results in expansion of the SC apical surfaces (Fig.  3.1) (Forge 1985; Cotanche 1987), and SCs throughout the chicken basilar papilla remained completely quiescent until HC loss occurred nearby (Corwin and Cotanche 1988). Furthermore, regenerative proliferation was restricted to SCs within and very near to sites of damage, a finding later quantified through laser microbeam experiments in the basilar papilla where the region of HC ablation could be precisely controlled (Warchol and Corwin 1996). Thus, HC loss triggers proliferation in SCs that are nearby, which are the cells whose shapes change the most after damage (Corwin et al. 1991). In many epithelia, cell loss and proliferation are precisely matched through mechanisms that appear to be of regulated on the basis of mechanical signals. Ongoing cell growth and proliferation lead to compression of neighboring cells and their apoptosis or extrusion (Matamoro-Vidal and Levayer 2019). In addition, cell loss in epithelia appears to trigger proliferation by affecting the shape and tension of remaining cells through what has been called a “demand-driven” system (Mesa et al. 2018). The latter paradigm appears consistent with findings from HC epithelia. For instance, in the chicken utricle, HCs spontaneously degenerate and are replaced by new HCs produced when SCs divide with such precise matching of death and replacement that the spatial density of HCs remains constant throughout life (Jørgensen and Mathiesen 1988; Kil et al. 1997; Goodyear et al. 1999). Furthermore, when treatments with caspase inhibitors were used to block the spontaneous HC loss and turnover of the HC population that resulted in the suppression of the normally high incidence of SC proliferation that occur in avian vestibular epithelia (Fig. 3.2) (Matsui et al. 2002). Additional evidence that changes in shape correlate with and may influence the probability that SCs will proliferate came from experiments on cultures of isolated pieces of sensory epithelia that were enzymatically delaminated from the underlying stroma in the utricles of chickens and mice. When such epithelia adhere to rigid substrates in tissue culture, SCs at the cut edges of the epithelia change from their normal columnar shape to flat, squamous shapes, and their area expands as they spread out. Soon after changing shape, large numbers of SCs in the epithelia from chickens and young mice entered mitosis and proliferated (Warchol 2002;

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

45

Fig. 3.2  The caspase inhibitor BAF prevents neomycin-mediated HC loss in the chicken utricle. It also prevents HC loss and SC proliferation that results from ongoing turnover. (Adapted with permission and source from Matsui et al. 2002, copyright 2002 Society for Neuroscience)

Davies et al. 2007). However, in identically prepared explants from utricles of mice that were 6 days old, shape changes were slower to occur, and SCs near the cut edge of the explant often changed shape and proliferated most (Fig. 3.3) (Davies et al. 2007). In explants from mice that were more than 2 weeks old, few SCs changed shape or proliferated, suggesting a link between the capacity for SCs to change shape and their propensity for proliferation. Contrasting with those results, when pieces of sensory epithelium were cultured after delamination from the utricles of young and old chickens, there was no measurable decline in SC spreading or their high incidence of proliferation (Burns et al. 2008). The results showed that maturational changes in mammalian HC epithelia led to decreases in the capacity for vestibular SCs to change shape and proliferate, but such changes do not occur in epithelia from young or old chickens. The search was on to find the cause, and the results above strongly suggested that the limited regeneration of HCs observed in

46

M. A. Rudolf and J. T. Corwin

Fig. 3.3  In isolated sheets of utricular epithelia from mice, SC shape changes precede and correlate with proliferation. The capacity for shape change and proliferation each decline with age. Insets depict the initial size of the explant at the start of culture, the images depict the explants after 72  h in culture, and proliferation (BrdU) images depict isolated sheets of SCs in utricles from embryonic mice readily change shape and proliferate. E18, embryonic day 18. P6, postnatal day 6. BF, brightfield. BrdU is a marker of S-phase entry. (Adapted with permission and source from Davies et al. 2007)

mammals was not due to the absence or reduction of a gene product expressed in chickens and other nonmammals where SCs readily change shape, proliferate, and replace lost HCs. Instead, it appeared that mammalian SCs gained something as those cells matured and became more differentiated. Potential mechanisms had not been identified at that point, but the evidence suggested that mammals had evolved some age-related changes not observed in chickens, which were limiting the capacity both for the SC shape change and for the high rates of SC proliferation that vestibular epithelia from neonatal mammals clearly exhibited but quickly lost. What caused the progressive, age-dependent decline in the capacity of mammalian SCs to change shape, which the SCs from the ears of young and fully adult birds never showed? Could the basis for that be identified by comparing the properties of SCs older mammals to those of younger mammals and nonmammals?

3.3 Actomyosin Contractility at Apical Junctions Accelerates Wound Closure in the Lesioned Vestibular Epithelium HCs have convex, flask-like shapes and are more turgid than the SCs that conform to them. (One exception is the mammalian organ of Corti, in which SCs surround HCs but adopt highly specialized morphologies themselves.) SCs span from the basement membrane to the apical surface of the sensory epithelium, which in the organ of Corti was historically called the reticular lamina. HC loss requires remodeling of apical junctions to maintain the epithelium’s ionic barrier function; as mature HCs do not contact the basement membrane, HC loss does not immediately perturb SC contacts with the extracellular matrix (ECM).

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

47

Early studies that investigated how SCs in the organ of Corti remodel their apical junctions following HC loss found that tight junctions between the degenerating HC and neighboring SCs remain intact temporarily as SCs extend beneath the degenerating HC’s cuticular plate and form a new tight junction, often referred to as a “scar” (Leonova and Raphael 1997). Several SCs surrounding the dying HC participate in the new junction. Since the age-related declines in the capacity of mammalian SCs to change shape and proliferate had only been investigated using delaminated sheets of HC epithelium, there was the possibility that they were artifacts or significantly influenced by the nature of that experimental preparation. It was later found that similar age-­ related changes occurred when wounds were created in the utricular sensory epithelium that remained in situ (i.e., on the stroma), with the roof and otoconia removed for visualization and experimental access (Meyers and Corwin 2007). The HC epithelium was wounded using a micropunch and a sharpened tungsten needle to excise a region of the sensory epithelium from the center of the mouse utricle’s macula while leaving the underlying stroma intact. Time-lapse microscopic recordings and phalloidin labeling of F-actin showed that excision wounds in utricles from E18.5 mouse embryos closed within 24 h. SCs in a wide region around the wound changed shape to participate in the closure process, and the apical surfaces of SCs in the resealed area were only slightly larger than those outside of the lesioned area. The time-lapse recordings, when combined with pharmacological treatments and controls, revealed an initial “circularization” process that could be blocked by treatment with cytochalasin D, an actin depolymerizing agent. In circularization, the irregular boundaries of the lesion became smooth as a thick F-actin cable formed inside the portion of each of the SCs bordering the wound edge. Once that multicellular F-actin cable formed, the wound front closed in a “purse-string” process, which was dependent on myosin-mediated contraction, since it was reversibly blocked by inhibiting the Rho kinase (ROCK). Most significantly, the rapid closure of those wounds resulted in SCs in and around the wound site changing shape and proliferating. Mice that were at least 2 weeks old were investigated using the same procedures as in those from E18.5 mice. The wounds in the older utricles all remained open for at least 48  h, but eventually most closed by 72–96  h (Meyers and Corwin 2007; Collado et al. 2011a). Treatment with lysophosphatidic acid (LPA), an agent known to promote Rho/ROCK activity, resulted in the excision wounds closing within 48 h. Unlike what had occurred in neonatal mice, only a small number of SCs bordering the wound edge changed shape, and they greatly expanded in area in order to reseal the epithelium. The results showed that the age-related declines observed in delaminated epithelia persist when HC epithelia are in situ within the utricle. The findings also were consistent with the notion that actomyosin-generated tension, increases in the planar area, or a combination of those potential signals raises the probability that a given SC will proliferate. Results of a subsequent study comparing the responses of delaminated HC epithelia from young and old mice with those from young and old chickens showed that the differences in SC shape change and proliferation that had

48

M. A. Rudolf and J. T. Corwin

been found using delaminated epithelial sheets were replicated using this in situ preparation (Collado et al. 2011a). While it still was not clear whether mechanical tension influenced proliferation directly, or indirectly through its effect on cellular shape change, both have been shown to influence proliferation in other epithelia (Uroz et al. 2018).

3.4 Maturational Reinforcement of Adherens Junctions Coincides with Age-Related Declines in the Plasticity of Mammalian Supporting Cells 3.4.1 The Unique Circumferential F-actin Bands in Mammalian Supporting Cells The close phylogenetic and developmental associations between cellular shape change and proliferation in HC epithelia showed the need to understand how the capacity to change shape became restricted as mammalian SCs matured. The crucial clue appeared when fluorescently labeled phalloidin was used to label and compare F-actin at the apical junctions of SCs in utricles from neonatal and mature mice, which showed a great increase in intercellular junction-associated F-actin in the utricles from more mature mice (Meyers and Corwin 2007). Detailed measurements found the thickness of the circumferential F-actin bands that bracket SC-SC junctions increased 13-fold during the postnatal maturation of the mouse utricle (Burns et al. 2008). In fact, the bands of F-actin that form thin belts in epithelial cells in other tissues grow so large in the vestibular SCs of mature mice that they extend across 89% of the average SC at the level of the adherens junction (Fig. 3.4). In addition, the growth of the F-actin bands was found to be exceptionally correlated to the age-­ related declines in spreading (R  =  −0.989) and SC proliferation (R = −0.975) that had been measured in cultured sheets of delaminated utricular epithelia. Similarly thick circumferential F-actin bands were found in the human utricle, and maturational thickening also occurred in Deiters’ cells within the auditory epithelia of mice (Fig. 3.4) (Burns et al. 2008, 2013). In comparison, the circumferential F-actin bands in SCs of the chicken utricle were found to remain thin throughout life, consistent with the observation that their ability to spread and readily proliferate does not appear to change between hatching and older adulthood (Fig. 3.4) (Burns et al. 2008). An additional phylogenetic analysis showed that the circumferential F-actin belts in SCs within the vestibular organs of sharks, bony fish, frogs, and turtles also remained thin from birth or hatching into adulthood (Burns et al. 2013).

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

49

Fig. 3.4  F-actin bands in SCs of the mammalian utricle and cochlea thicken postnatally. F-actin bands in chicken SCs remain thin throughout life. Scale bars denote 10  μm for mouse utricle images and 5 μm for images of the mouse cochlea and chicken utricle. (Adapted with permission and source from Burns et al. 2008 and Burns et al. 2013)

3.4.2 Potential Mechanical Influence of the Thick F-actin Bands that Develop in Mammalian Supporting Cells In other epithelia, adherens junctions couple cells into a mechanical syncytium and serve as an interface through which they sense the mechanical forces arising from cell loss (Harris et  al. 2012). Such collective behavior was evident in utricles of chickens and neonatal mice that were subject to large excision lesions as discussed in Sect. 3.3. SCs many cell diameters away from the initial wound edge changed shape and participated in closure of the wound (Collado et al. 2011a). The collective behavior was more restricted in adult mice. There, wounds closed more slowly, and the SCs that were within three cell diameters of the lesion edge were the only ones that changed shape substantially, participated in wound closure, and often began to proliferate. Similar phenomena were found to occur during the wound closure events after aminoglycoside-mediated poisoning of HCs. In utricles of chickens and newborn mice, SCs formed a supracellular F-actin cable that constricted like a purse string beneath the dying HC’s cuticular plate, with SCs that were several cell diameters away also noticeably changing in shape and position (Bird et al. 2010; Burns and Corwin 2014). In other epithelia, wounds with negative curvature, like the negative curvature where SCs border a dying HC, promote the formation of actomyosin cables that close via purse-string contraction, contrasting with the lamellipodial protrusions that tend to form at wound edges with positive (convex) curvature (Ravasio et al. 2015). After mild aminoglycoside poisoning that led to individual HC losses in the mature mouse utricle, the thick F-actin bands in surrounding SCs remained in place, and F-actin purse strings did not form. Rather, F-actin in the SCs adjacent to the dying HC formed lamellipodial structures that appeared to seal the potential breech in the epithelium (Burns and Corwin 2014). In addition to this

50

M. A. Rudolf and J. T. Corwin

qualitative age and species difference in the organization of the F-actin response, the speed of wound closure differed. In the chicken utricle, the new, damage-evoked intercellular junctions formed in only 3–6 min (Bird et al. 2010), while damage-­ evoked junction formation occurred within 1 h of HC death in utricles from newborn mice and required up to 4 h in utricles from mice that were 30 days old (Burns and Corwin 2014). This slowing may be consequential, because epithelial proliferation is not only affected by the degree of cellular shape change but also the rate of change and the time course of the resulting strain (Irvine and Shraiman 2017). These studies strongly suggest that the maturational reinforcement of F-actin bands in mammalian SCs makes them more resistant to deformation, a known trigger for proliferation. Measurements of the surface mechanical properties of the sensory epithelium have provided direct evidence supporting this hypothesis. One such study used atomic force microscopy to examine how stiffness of the murine organ of Corti’s reticular lamina changes between E16 and P5. This revealed that stiffness of the reticular lamina doubles from ~3.7 to ~7.4 kPa in the region containing outer HCs and Deiters’ cells as F-actin and acetylated tubulin accumulate during early postnatal life (Szarama et al. 2012). Micropipette aspiration studies showed that the apical surface of the mouse utricle becomes more resistant to deformation with age, and depolymerization of the F-actin bands reduces its stiffness. Furthermore, the apical surface of the adult mouse utricle is significantly stiffer than the apical surface of the chicken utricle (Rudolf et al. 2022). Further studies may investigate whether dismantling the circumferential F-actin bands of mature mammalian SCs also restores their capacity for shape change.

3.4.3 Structure and Regulation of the Circumferential F-actin Bands in Supporting Cells 3.4.3.1 Sarcomeric Actomyosin Network at Cochlear Apical Junctions The circumferential actomyosin network in the organ of Corti of neonatal rodents is stereotypical of other simple epithelia. The circumferential F-actin bands associate with regularly spaced (~400–600  nm), bipolar filaments of nonmuscle myosin II that interlace with the actin crosslinker α-actinin1, reminiscent of a sarcomeric network (Fig.  3.5) (Ebrahim et  al. 2013). Actomyosin contractility exerts tension at these apical junctions, as treatments with blebbistatin, an inhibitor of myosin II contractility, caused a reversible expansion of the apical surfaces of HCs and SCs in the neonatal organ of Corti (Ebrahim et  al. 2013; Cohen et  al. 2020). It seems unlikely that such an expansion would occur in the mature organ of Corti, which has a stiffer reticular lamina, thicker circumferential F-actin bands in SCs, and deforms little after HC loss (Anttonen et al. 2017). Although the dimensions of the F-actin bands in mammalian SCs have been extensively measured at the light microscopic level, their ultrastructural organization has received only a cursory examination. Whether the F-actin is present as

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

51

Fig. 3.5  Sarcomeric network at HC-SC apical junctions in the P2 rat organ of Corti. Scale bar represents 3 μm. The middle panel depicts an enlarged view of the bracketed region along with relative fluorescence intensity traces of nonmuscle myosin IIB (green) and α-actinin 1 (red). (Adapted with permission and source from Ebrahim et al. 2013)

bundles or branched filaments can influence the contractility of F-actin-­nonmusclemyosin II assemblies and their efficient generation of tension (Ennomani et  al. 2016). In typical epithelial cells, the apical domain size is regulated by continuous actomyosin contractility along the zonula adherens. The exceptionally thick and highly stable F-actin bands that bracket the apical junctions of mature mammalian SCs may reduce the energetic cost for mature mammalian SCs to maintain stable shape, by reducing ATP expended in myosin II contraction. Robust and stable circumferential bands of F-actin that reduce actomyosin contractility also could limit the activation of intracellular signaling pathways that respond to changes in epithelial tension, such as YAP-TEAD (Pan et al. 2016). If the mechanical stability of mature mammalian HC epithelia are maintained independent of actomyosin contractility, SCs may expend less energy maintaining their shape than the SCs of young mammals or nonmammals, which require continuous activity of nonmuscle myosin II (an ATP-hydrolyzing enzyme) to do so (Ebrahim et al. 2013, Rudolf et al. 2022). Perhaps the reinforced junctions provide mechanical stability at low metabolic cost, at the expense of plasticity, to improve frequency sensitivity in the organ of Corti or responsiveness of the vestibular epithelia. 3.4.3.2 Regulation of the Circumferential F-actin Bands by Rho GTPases Rho GTPases regulate cytoskeletal dynamics at cell junctions and assume a variety of functions depending on the local availability of scaffold proteins, effector molecules, guanine nucleotide exchange factors, GTPase activating proteins, and guanine nucleotide dissociation inhibitors (Arnold et al. 2017). RhoA and Cdc42 are two well-studied members of this family of 20 proteins whose function in the inner ear has been investigated. Deletion of RhoA in cochlear SCs did not affect junctional maturation in SCs (Anttonen et al. 2017), but RhoA has been implicated in wound closure in the mouse utricle. There, closure of excision lesions was accelerated by treatments with

52

M. A. Rudolf and J. T. Corwin

lysophosphatidic acid, which activates RhoA (Meyers and Corwin 2007). RhoA is known to interact with formin and promote F-actin bunding and the activation of ROCK, which phosphorylates myosin regulatory light chain to promote actomyosin contractility (Arnold et al. 2017). Consistent with that, pharmacological inhibition of ROCK did not prevent F-actin cable formation but did prevent wound closure (Meyers and Corwin 2007). Deletion of Cdc42 in Deiters’ and pillar cells of the neonatal mouse cochlea did affect development of adherens junctions (Anttonen et al. 2012), resulting in shortened, fragmented F-actin bands and reduced mediolateral dimensions of the organ of Corti. These structural defects appeared to disrupt the mechanical stability of the reticular lamina, eventually leading to collapse of the basal organ of Corti. Deiters’ cells also require Cdc42 to remodel their junctions during wound closure after loss of outer HCs (Anttonen et al. 2014).

3.5 E-cadherin Accumulates at Supporting Cell Junctions in the Mammalian Vestibular Epithelium In addition to dictating tissue-level mechanical properties, adherens junctions are a focal point of intracellular signaling. Their cadherin-based adhesions link to the F-actin cytoskeleton via catenins, which have both structural and signaling functions (Fig. 3.6). E-cadherin is a tumor suppressor that restricts proliferation through a variety of mechanisms, including activation of p27kip1 and inhibition of EGFR, canonical Wnt signaling, and YAP-TEAD transcriptional activity (Mendonsa et al. 2018). E-cadherin was found to play an important role in the long-established phenomenon of contact inhibition of proliferation; homophilic ligation of E-cadherin adhesions activates LATS1/2 kinases, thereby sequestering YAP in the cytoplasm and limiting proliferation (Kim et al. 2011). The expression of E-cadherin increases sixfold between birth and adulthood in the mouse utricle (Fig. 3.7), where it is enriched at SC-SC junctions (Collado et al. 2011b). In the organ of Corti, E-cadherin increases 56% between P0 and P7, where it is enriched in the lateral domain including pillar cells and Deiters’ cells (Luo et al. 2018). SCs in birds, fish, and amphibians express little E-cadherin (Fig.  3.7) (Warchol 2007; Burns et al. 2013). Thus, E-cadherin expression, like the thickening of circumferential F-actin bands, correlates with age-dependent declines in the plasticity of mammalian SCs and with the striking differences in the capacity for regeneration between mammals and other vertebrates. Rather surprisingly, the expression of β-catenin at cell membranes in the utricle is reported not to increase with age despite the maturational increase in E-cadherin expression in mammalian SCs (Collado et al. 2011b). However, β-catenin is notoriously difficult to localize and measure with immunohistochemistry. In contrast, the expression p120-catenin, which stabilizes cadherin/catenin complexes, has been shown to increase during maturation of the organ of Corti, colocalizing with E-cadherin (Fig. 3.7) (Luo et al. 2018).

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

53

Ca2+

Ca2+ Ca2+

Ca2+

Ca2+

Ca2+

Ca2+

Ca2+

Ca2+

Ca2+

E-cadherin

p120

p120

 -cat

 -cat

 -c

at

 -c

at

Vinculin

Fmn F-actin

NMII

Fig. 3.6  Schematic of cadherin-catenin complex at the adherens junction

3.5.1 Hypothesized Role for N-cadherin in Limiting Supporting Cell Proliferation N-cadherin is expressed in the intercellular junctions of HCs and SCs throughout the utricular macula in birds and mammals as well as in the inner HCs and phalangeal cells of the organ of Corti. An early and intriguing finding was the discovery that microbeads coated with an antibody that blocks N-cadherin stopped

Fig. 3.7  Levels of p120-catenin and E-cadherin increase during postnatal maturation in the mammalian inner ear. In chicken utricles, N-cadherin is expressed within the sensory epithelium (SE) and E-cadherin is expressed in the surrounding nonsensory epithelium (NSE). In mouse utricles, E-cadherin is expressed both in the SE and NSE. The scale bars denote 20 μm in the cochlear images and 10 μm in the images of utricles. (Cochlear images adapted with permission and source from Luo et al. 2018, and mouse utricle images were adapted with permission and source from Collado et al. 2011b, each under the Creative Commons Attribution 4.0 International License (CC-­ BY). Chicken utricle images adapted with permission and source from Warchol 2007)

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

55

proliferation when those beads bound to SCs at the outer edges of epithelial sheets from chicken utricles (Warchol 2002). One hypothesis is that the decoupling of N-cadherin adhesions at HC-SC junctions could function as an injury signal that triggers SC proliferation. Alternatively, the binding of microbeads to N-cadherin expressed in SC junctions might impose mechanical constraints that could limit Hippo pathway signaling. In any case, the surprising nature of this finding suggests that follow-up investigations could be rewarded by discoveries about the fundamental nature of the controls that limit the proliferation of SCs.

3.5.2 A Special Case: Apical Junctions in the Anolis Lizard The perijunctional F-actin of vestibular SCs from the lizard genus Anolis is unusual, developing into a reticular, weblike structure by adulthood (Burns et al. 2013). Their SCs also appear to express E-cadherin at levels more comparable to those of mammals. One experiment presented in that study was supportive of the potential for SCs in the Anolis to undergo regenerative proliferation, but the data were preliminary and limited to a crude preparation with incidental mechanical damage. The features of SC junctions in Anolis do not appear to be universal among reptiles; for instance, SCs in adult turtles have relatively thin F-actin bands and hardly express E-cadherin, reminiscent of SCs in birds, fish, and amphibians (Burns et al. 2013). Thus, a more thorough investigation of the potential for HC regeneration in lizards would be informative. Comparative investigations of the SCs in the ears of different species of reptiles, such as crocodilians, other lizards besides Anolis, snakes, and rhynchocephalians may shed light on how F-actin reinforcement and stabilization of E-cadherin could contribute to any physiological differences in the regenerative capacity.

3.6 Regulation and Perturbation of Apical Junctions in Mammalian Supporting Cells As shown in Sects. 3.4 and 3.5, the maturational decline in the plasticity of mammalian SCs mirrors their development of E-cadherin-rich junctions bracketed by exceptionally thick F-actin bands. Yet, correlation does not equal causation: The question remains as to whether these junctional components causally limit repair responses in the mammalian ear, or whether they are merely hallmarks of SC differentiation, downstream of some other crucial underlying process of SC maturation. Testing this will require experiments that directly manipulate these junctional components. It is presently unclear how the circumferential F-actin bands could be precisely manipulated, although it may be possible to identify and delete actin crosslinkers or GTPases that may be important for maintenance of the F-actin

56

M. A. Rudolf and J. T. Corwin

bands. Targeting E-cadherin appears to be more straightforward; however, this too can be associated with pitfalls. For instance, conditional knockout of E-cadherin in Plp1+ SCs of the mouse utricle led to compensatory upregulation of N-cadherin in those cells (M.S. Collado and J.T. Corwin, unpublished observation). It was shown in an SC-derived cell line that transfection of E-cadherin increased the number of intercellular junctions, reduced proliferation, and inhibited expression of certain HC-specific genes including Myo7a (Hackett et al. 2002). These promising results warrant further studies that directly manipulate levels of E-cadherin or other components of adherens junction in SCs. Whatever the case, understanding the regulation of E-cadherin has revealed clues as to how SCs mature and may be reverted to a more progenitor-like state. Inhibiting GSK3β or expressing stabilized and constitutively active β-catenin depleted E-cadherin and increased SC proliferation in the developing utricle as well as in the cochlea (Lu and Corwin 2008; Shi et al. 2014), indicating that canonical Wnt signaling suppresses E-cadherin levels in SCs. GSK3β activity is also known to mediate acetylation of the E-cadherin promoter by repressing Snail and Slug. Consistent with that possible mode of action, treatment with a histone deacetylase inhibitor increased E-cadherin levels and reduced proliferation in the neonatal mouse utricle (Lu and Corwin 2008). Work by McLean and colleagues has shown that a cocktail of seven factors (EGF, IGF-1, FGF2, vitamin C, and pharmacologic inhibitors to GSK3β, HDAC, and TGFRβI) stimulated significant proliferation in cultures of SCs that were dissociated from the neonatal mouse cochlea (McLean et  al. 2017). This breakthrough enabled robust amplification of Lgr5+ progenitors in three-dimensional culture and development of HC-producing organoids at unprecedented scale (Janesick, Hashino, and Heller, Chap. 6). Inspired by those results, a subsequent study investigated the effects that this cocktail would have on semi-intact utricles cultured from young and mature mice. Although the composition of that cocktail was designed to simultaneously activate multiple growth factor pathways known to stimulate proliferation of SCs while blocking activity in pathways associated with inhibition of SC proliferation, it was found that when utricles from neonatal, juvenile, and more mature mice were cultured with that cocktail, the incidence of SC proliferation followed the same age-­ related declines observed in prior studies (Kozlowski et  al. 2020). Other results showed that two of the seven factors from that cocktail, EGF and a GSK3β inhibitor (CHIR), caused the robust circumferential F-actin bands at the junctions of SCs of more mature mice to become thinner throughout those utricles (Fig. 3.8) (Kozlowski et al. 2020). Then, as the treatment with EGF and CHIR continued, the levels of E-cadherin, β-catenin, and YAP decreased significantly, but only within those SCs that were located within the central epithelial region called the striola (Fig.  3.8). Subsequent to those changes, the striolar SCs in those more mature utricles exited their long period of quiescence and exhibited robust proliferation, which occurred even in utricles from adult mice. Importantly, the junctional remodeling induced by the combination of EGF and CHIR preceded their ability to stimulate proliferation

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

57

Fig. 3.8  EGF and the GSK3 inhibitor CHIR-99021 thin the F-actin bands of SCs in cultured mouse utricles. Scale bar denotes 5 μm. Culture with these agents depletes E-cadherin, but not N-cadherin, from striolar SC junctions of adult mouse utricles. Levels of the transcriptional coactivators β-catenin and YAP are reduced in the apical domains of those striolar SCs. (Adapted with permission and source from Kozlowski et al. 2020 under the Creative Commons Attribution 4.0 International License (CC-BY))

in mature SCs. These findings raised several questions: Does thinning the F-actin bands restore the capacity for cellular shape change? By what mechanisms do EGF and the GSK3β inhibitor cause the depletion of E-cadherin in striolar SCs? What differences in gene expression may be linked to the spatial restriction of E-cadherin depletion and renewed proliferation to SCs within the striola while extrastriolar SCs remain quiescent? Finally, would the progeny produced from the SCs that divide in the striola of adult mice be capable of differentiating into HCs?

58

M. A. Rudolf and J. T. Corwin

3.6.1 Potential Interaction of Notch Signaling and E-cadherin Adhesion Juxtacrine signals in the Notch pathway regulate pattern development in many tissues. During embryonic development of the ear, cells throughout the forming sensory epithelium express Notch receptors. As cells differentiate toward an HC fate, they express the ligands Jag1 and Dll1, which bind Notch receptors on neighboring cells. Ligation of Notch receptors inhibits HC differentiation and drives cells toward an SC differentiation program that restricts their expression of Jag1 and Dll1 (Campbell et al. 2016) and thus develops the alternating mosaic pattern of HCs and SCs in the otic sensory epithelia. Pharmacological blockade of the Notch pathway can be achieved by treating epithelia with an inhibitor of gamma secretase and induces phenotypic conversion of SCs into HCs in the developing vestibular and auditory epithelia (Lin et al. 2011). In utricles from young mice, gamma secretase inhibition causes SCs to pinocytotically internalize membrane E-cadherin and convert to an HC phenotype (Collado et al. 2011b; Luo et al. 2018). In those experiments, the depletion of E-cadherin was not associated with changes in E-cadherin transcription but did require protein synthesis (Collado et  al. 2011b). It remains unclear whether the accumulation of E-cadherin limits phenotypic conversion. Also, it remains to be determined how age-related changes limit the effects of that pharmacological blockade of Notch signaling, including the internalization of E-cadherin and the conversion of cell phenotype.

3.7 Intracellular Signaling Downstream of Mechanical Signals Cell losses cause changes in mechanical tension that can propagate widely through most epithelia (Karsch et al. 2017). These forces are transduced into mechanosensitive biochemical signaling cascades, such as the Hippo and canonical Wnt signaling pathways, that can facilitate regenerative responses.

3.7.1 YAP/TAZ and the Hippo Pathway YAP and its paralog TAZ are transcriptional coactivators and Hippo pathway effectors that integrate mechanical and biochemical signals to control cell behavior (Fig.  3.9) (Totaro et  al. 2018). In the evolutionarily conserved pathway, MST1/2 kinases (Hippo in Drosophila) phosphorylate LATS1/2 (Warts in Drosophila), which in turn phosphorylate YAP/TAZ to induce their cytoplasmic sequestration and degradation. Absent this inhibitory phosphorylation, YAP/TAZ accumulate in

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

59

nuclei and partner with transcription factors such as TEADs to mediate changes in gene expression that can control proliferation and stemness. SC proliferation is associated with changes in intraepithelial tension, cell density, substrate stiffness, and cadherin adhesion. These stimuli control YAP/TAZ activity in other epithelia, and the role of YAP in SC proliferation has been directly explored. 3.7.1.1 YAP-TEAD Regulate Cell Cycle Arrest and Size Control in Hair Cell Epithelia Just as increases in SC area can trigger proliferation, decreases in SC area can contribute to cell cycle arrest during development. A model proposed by Gnedeva and colleagues specified that elastic forces from the surrounding nonsensory tissue resist growth of the utricular macula (Fig.  3.9). Simulations showed that this

Fig. 3.9  Overview of YAP/TAZ regulation. There is a hypothesized relationship between utricular growth, cell density, YAP-TEAD transcriptional activity, and proliferation during development. The graph depicts the inverse relationship between SC density and terminal mitoses in the developing mouse utricle

60

M. A. Rudolf and J. T. Corwin

mechanism explains the observed increases in cell density that occur throughout development, which coincide with inactivation of YAP-TEAD transcriptional activity and cell cycle exit (Gnedeva et al. 2017). Using “bubble cultures” in which the external mechanical constraints could be experimentally reduced, utricles grew to supraphysiological sizes, leading to a reduction in cell density and YAP-TEADdependent SC proliferation. A similar paradigm of mechanical crowding could theoretically govern the cessation of regenerative proliferation, but this remains to be tested. A subsequent study investigated the role of YAP-TEAD signaling in the developing organ of Corti, finding that TEAD transcription factors bind to putative regulatory elements of numerous genes important for stemness and proliferation in cycling E12 progenitors (Gnedeva et al. 2020). The accessibility of these chromatin regions declined by E13.5 when progenitors exited the cell cycle, coinciding with activation of Hippo pathway components and inhibitory phosphorylation of YAP. Taken together, the results suggested that mechanical constraints contribute to cell cycle exit in progenitors via LATS1/2-mediated phosphorylation and inactivation of YAP-TEAD signaling. Further underscoring the importance of YAP during developmental growth, conditional deletion of YAP in the developing inner ear significantly reduced proliferation and led to a smaller cochlea and vestibular organs. It seems likely that similar mechanisms control growth and cell cycle arrest in nonmammalian ears. This remains to be tested, though it has been shown that manipulation of Hippo pathway components can control the size of the lateral line primordium in the developing zebrafish (Loh et al. 2014; Agarwala et al. 2015). 3.7.1.2 YAP-TEAD Signaling in Repair and Regeneration In addition to its importance in developmental growth, studies have uncovered a role of YAP-TEAD signaling in governing repair responses in SCs. Mechanical lesions in neonatal mouse utricles caused YAP to accumulate in SC nuclei and mediate TEAD-dependent proliferation (Gnedeva et al. 2017). The degree of nuclear accumulation correlated to the extent of cellular shape change, indicating that mechanical strain contributes to the activation of YAP in mammalian SCs (Rudolf et al. 2020). However, HC loss was not sufficient to induce detectable nuclear accumulation of YAP in SCs of the murine utricle. This stands in contrast to the chicken utricle, where HC loss induced nuclear translocation of YAP in SCs, and pharmacologic inhibition of YAP-TEAD signaling attenuated regenerative proliferation (Rudolf et al. 2020; Borse et al. 2021). There is also evidence that YAP-TEAD target genes increase following HC loss in the chicken basilar papilla (Matsunaga et al. 2020). Thus, the damage-evoked activation of YAP occurs more readily in nonmammalian SCs than those in mammals and may contribute to species differences in regenerative proliferation. It will be important to test the effects of genetic

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

61

inactivation of YAP in nonmammalian models to confirm its role in regenerative proliferation. What limits the activation of YAP in mammalian SCs compared to those of nonmammals? Many upstream regulators in the Hippo signaling network converge on the inhibitory kinases LATS1/2. This led to the hypothesis that bypassing LATS1/2 could spur proliferation in mammalian SCs. Indeed, YAP variants that evade inhibitory phosphorylation entered SC nuclei and spurred TEAD-dependent proliferation, but ectopic wild-type YAP was phosphorylated and sequestered in the cytoplasm (Rudolf et al. 2020). Conditional deletion of LATS1/2 reduced inhibitory phosphorylation of YAP at Ser127 and evoked proliferation in SCs of the mouse utricle, even in adults (Rudolf et  al. 2020). Furthermore, a novel pharmacologic inhibitor of LATS1/2 kinase activity induced YAP-dependent proliferation in the murine utricle and made the transcriptomes of mature SCs resemble those of embryonic SCs (Kastan et al. 2021). These studies revealed that constitutive activity of LATS1/2 kinases is required to inhibit YAP-TEAD signaling and maintain quiescence in mature vestibular SCs. LATS1/2 inhibition did not induce proliferation of cochlear SCs, consistent with the notion that the cochlea contains additional layers of inhibitory regulation. However, viral-mediated expression of constitutively active YAP-5SA did induce a modest level of S-phase entry within the postnatal organ of Corti following HC loss (Gnedeva et al. 2020). It is likely that the E-cadherin-rich junctions in mammalian SCs contribute to LATS1/2-mediated inhibition of YAP. A clue is that E-cadherin has been shown to stimulate LATS1/2 activity independent of MST1/2 in an epithelial cell line (Kim et al. 2011). Also, treatments with an MST1/2 inhibitor induced nuclear translocation in SCs of chicken utricles, but not those of mice (Rudolf et al. 2020; Kastan et al. 2021). Secondly, pharmacological treatments that led to depletion of E-cadherin in striolar SCs of the mouse utricle also decreased YAP in the apical domains of those cells (Fig. 3.8) (Kozlowski et al. 2020). Factors downstream of YAP/TAZ inhibitory phosphorylation also may contribute to the age-dependent decline in plasticity of mammalian SCs. TEAD2 expression declines during postnatal maturation of the utricle, paralleling the downregulation of SoxC transcription factors for which it is a direct target (Gnedeva and Hudspeth 2015). This would explain observations that the rate of proliferation declines as SCs mature despite conditional deletion of LATS1/2 or expression of non-phosphorylatable variants of YAP (Rudolf et al. 2020). YAP/TAZ also regulate differentiation in epithelial cells (Totaro et al. 2017), but their role in HC differentiation remains relatively unexplored. In epidermis, YAP/ TAZ activity is high in cells that contact the basal lamina and low in cells that lose contact with the basal lamina and differentiate (Elbediwy et  al. 2016). HCs lose contact with the basal lamina as they progress in differentiation, but it remains untested whether downregulation of YAP/TAZ is required for SCs to assume an HC fate. Direct transdifferentiation of SCs into HCs is the major mode of regeneration in adult mouse utricles (Fig. 3.10) (Golub et al. 2012), so it will be important to identify what role YAP/TAZ or other SC-ECM signaling mechanisms play in this process.

62

M. A. Rudolf and J. T. Corwin

Fig. 3.10  Non-mitotic HC replacement in adult mice following diphtheria toxin (DT)-mediated damage in the Pou4f3+/DTR model system. Most new HCs are found in the lateral extrastriolar region and arise via direct transdifferentiation of SCs rather than via proliferative regeneration. S denotes striola, mES denotes medial extrastriola, and lES denotes lateral extrastriola. (Adapted with permission and source from Golub et  al. 2012 under the Creative Commons Attribution 4.0 International License (CC-BY))

3.7.2 Canonical Wnt Signaling Wnts are a family of secreted ligands that bind Frizzled and LRP5/6 receptors. Signaling through the canonical pathway results in nuclear accumulation of β-catenin, which serves as a transcriptional coactivator by modulating gene expression with TCF/Lef transcription factors (Nelson and Nusse 2004). In the absence of canonical Wnt signaling, the cytoplasmic pool of β-catenin is degraded by a mechanism that involves casein kinase I phosphorylation of Ser45 and then GSK3β phosphorylation of multiple serine and threonine residues. Analyses of TCF/Lef and Lgr5 reporter mice have shown that canonical Wnt signaling is active in the embryonic cochlear duct (Chai et al. 2011; Jacques et al. 2012) and in striolar SCs of the developing utricle (Wang et al. 2015). β-catenin regulates SC proliferation as well as HC differentiation through its role in Wnt signaling (Jacques et al. 2012; Shi et al. 2014). It also has a structural role in adherens junctions; comparison of mice with conditional knockout of β-catenin to those that express a transcriptionally inactive β-catenin variant revealed that the protein’s structural function is required for establishment of the mediolateral boundary of the organ of Corti, independent of its function in Wnt signaling (Jansson et al. 2019). Expression of a gain-of-function variant of β-catenin potentiated regenerative proliferation and mitotic HC replacement in Lgr5+ SCs of the neonatal mouse utricle and cochlea after diphtheria toxin mediated HC ablation (Wang et  al. 2015; Atkinson et al. 2018). In those studies, β-catenin activation had no mitogenic effect

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

63

in the absence of damage (Wang et al. 2015; Atkinson et al. 2018). This corresponds to functional evidence that HC loss activates canonical Wnt signaling, leading to the upregulation of the canonical Wnt target gene Lgr5 in striolar SCs of the neonatal mouse utricle (Wang et al. 2015), and Wnt-induced proliferation in the zebrafish lateral line (Romero-Carvajal et  al. 2015). However, the evidence that β-catenin accumulates in nuclei of SCs after damage is rather limited. Some nuclear β-catenin was observed in forskolin-treated sheets of utricular epithelium from the rat (Kim et al. 2004) and in SCs of the delaminated chicken utricle (Warchol 2002). The proliferative response to β-catenin stabilization declines with age in the mouse cochlear and utricular epithelia (Lu and Corwin 2008; Samarajeewa et al. 2018), despite evidence that TCF/Lef reporters can still be activated and Wnt pathway components continue to be expressed (Geng et al. 2016). This has been attributed to age-dependent changes in chromatin accessibility or crosstalk with other signaling pathways. Another possibility is that β-catenin’s dual functions in adhesion and canonical Wnt signaling may be antagonistic, competing for β-catenin in the cell. A simple, but untested, hypothesis accounting for these findings is that increased levels of E-cadherin in mature mammalian SCs sequester β-catenin and limit canonical Wnt signaling. Supporting that notion, GSK3β inhibition and expression of activated β-catenin each reduced E-cadherin levels and enhanced proliferation in HC epithelia (Lu and Corwin 2008; Shi et al. 2014). Extended culture of adult mouse utricles in EGF and a GSK3 inhibitor led to depletion of E-cadherin in striolar SCs, which was accompanied by a reduction in apical β-catenin expression and proliferation of those SCs (Fig. 3.8) (Kozlowski et al. 2020). Mechanical tension also induces β-catenin transcriptional activity. Strain applied to the E-cadherin-based adhesions of MDCK monolayers causes β-catenin to accumulate in nuclei and mediate S-phase entry with TCF/Lef (Benham-Pyle et  al. 2015). Should the thick circumferential F-actin bands of mature mammalian SCs limit these shape changes as hypothesized, that would lead to blunting of such tension-­mediated β-catenin regulation.

3.8 Supporting Cell-Extracellular Matrix Interactions Although the apical junctions of SCs have garnered considerable attention, SC contacts with the underlying stroma are relatively unexplored, despite their likely profound impact on SC behavior. A general rule in epithelia is that increased footprint on the basement membrane triggers proliferation and stem cell renewal, while decreased footprint promotes differentiation (Totaro et  al. 2017). While this basic principle has never been rigorously shown in HC epithelia, it is consistent with the cytoarchitecture; progenitors and SCs remain in constant contact with the basement membrane, and HCs lose their association with the basement membrane as they differentiate. What then does this imply? For one, adhesions to the basement membrane may promote an SC phenotype and actively inhibit HC differentiation. Secondly, when an SC’s contact area

64

M. A. Rudolf and J. T. Corwin

with the basement membrane increases (e.g., when a neighboring progenitor/SC differentiates into an HC), it would be expected to activate mitogenic signaling cascades, such as the integrin-FAK-Src-YAP axis (Kim and Gumbiner 2015). Concordant with that possibility, SCs within delaminated epithelial sheets from the chicken utricle exhibit positive immunoreactivity for FAK and phosphotyrosine as they spread and proliferate (Warchol 2002). The same principle could also underlie the delayed regenerative proliferation that occurs in the chicken utricle, which allows for the SC density to remain constant after an initial wave of direct transdifferentiation (Scheibinger et al. 2018). In addition to contact area, other parameters affecting SC-basement membrane interactions include the expression of integrins, as well as the composition and mechanical properties of the substrate. Little is known about these specifics in native HC epithelia or how integrin-based contacts form focal adhesions or hemidesmosomes in SCs. Transmission electron micrographs of the SC-ECM interface of the mouse utricle revealed an age-dependent maturation of electron-dense regions suggestive of maturing hemidesmosomal plaques (Fig. 3.11) (Davies et al. 2007). It is easy to envision robust hemidesmosomes serving as literal anchors that impede HC differentiation. Treatment with phorbol myristate acetate, a pharmacologic activator of protein kinase C, induced redistribution of integrin β4 immunostaining and induced both spreading and proliferation in explants from P6 mice (Davies et al. 2007). It will be necessary to investigate the role of integrin signaling in vivo or in organotypic culture, where SCs remain in contact with their native ECM. As the notion that physical characteristics of the ECM could direct stem cell lineage was emerging among bioengineers (Engler et al. 2006), the effect of substrate stiffness on SC spreading and proliferation was being assessed. Sheets of utricular epithelia from young mice were cultured on glass-bottom dishes coated either with thick deposits of Matrigel (producing a soft substrate) or on a thin layer of Matrigel (producing a biochemically identical stiff substrate). SCs spread and proliferated readily in the stiff condition, but not when plated atop the soft substrate in which case almost all of the SCs remained columnar and did not spread or proliferate (Collado et al. 2011a). The findings established that SCs spreading and proliferation are modulated by substrate stiffness, composition, and integrin signaling, but outside of a few studies (Davies et al. 2007; Ishiyama et al. 2009), very little is known about these parameters in vivo. Do composition and stiffness of the underlying stroma change as mammalian vestibular epithelia mature? Do the expression and the types of integrin at SC-ECM contacts change during maturation and differ between mammals and nonmammals? Does HC loss activate signaling cascades from focal adhesions? If so, are these cascades required for regenerative proliferation in nonmammals and immature mammals?

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

65

Fig. 3.11  Transmission electron micrographs of the SC-basal lamina interface from utricles of mice at different ages. Arrows depict progressive maturation of hemidesmosome-like plaques. SC, supporting cell. CT, connective tissue. Scale bar denotes 250 nm. (Adapted with permission and source from Davies et al. 2007)

3.9 Summary 3.9.1 A Hypothetical Model for Mechanical Control of Hair Cell Replacement Current evidence suggests that YAP-TEAD transcriptional activity is high during embryonic development of HC epithelia, facilitating robust progenitor cell proliferation that leads to organ growth (Gnedeva et al. 2017). Eventually, mechanical constraints imposed by the surrounding tissue or the bony labyrinth restrict growth, so further proliferation leads to an increase in cell density (Gnedeva et al. 2017). In

66

M. A. Rudolf and J. T. Corwin

other epithelia, such crowding reduces cell-matrix adhesion and cortical tension and increases cell-cell adhesion (Miroshnikova et al. 2018), all of which serve to reduce YAP-TEAD transcriptional activity. These factors may collaborate with changes to gene expression to arrest proliferation in progenitor cells (Warchol 2002; Gnedeva et  al. 2017). Such developmental regulation is likely to occur in HC epithelia of mammals and nonmammals alike. In nonmammals and juvenile mammals, HC elimination involves rapid shape changes in SCs, which coordinate to form an F-actin purse string that undergoes ROCK-dependent constriction and reseals the epithelial barrier (Meyers and Corwin 2007; Bird et al. 2010). Waves of Ca2+ signaling may facilitate the endocytosis of E-cadherin and formation of a supracellular F-actin purse string (Hunter et  al. 2015). These changes in cell shape and contractility increase the likelihood of SC proliferation (Meyers and Corwin 2007; Collado et al. 2011a), possibly via increased mechanical tension that would lead to nuclear accumulation of YAP and β-catenin, which stimulate TEAD- and TCF/Lef-dependent transcription, respectively (Benham-Pyle et al. 2015). SCs that undergo either direct transdifferentiation or symmetric differentiating division leave an SC-sized vacancy on the basal lamina, that in theory would increase the cell-ECM contact area of remaining SCs – a known stimulus of YAP-­ TEAD transcriptional activity in other cell types (Dupont et al. 2011). The signaling evoked by this increased footprint may promote the delayed onset of amplifying divisions that occur in nonmammals and replenish the SC population (Scheibinger et al. 2018). Such damage-evoked responses appear to be impeded in SCs of mature mammals, whose junctions are reinforced by exceptionally thick circumferential F-actin bands (Burns et al. 2008; Burns and Corwin 2014) and high levels of E-cadherin (Collado et al. 2011b; Luo et al. 2018). These reinforced junctions could form a mechanical block by limiting the rate and restricting the propagation of mechanical deformation and SC shape changes after HC loss (Burns and Corwin 2014; Anttonen et al. 2017). In addition, those junctional attributes could contribute to a biochemical block by restricting signaling from growth factor receptors, activating inhibitory LATS1/2 kinases, and sequestering β-catenin (Nelson and Nusse 2004; Perrais et al. 2007; Kim et al. 2011).

3.9.2 Outstanding Questions and Opportunities Some important open questions related to the role of cell junctions and epithelial mechanics in the regenerative replacement of HCs are as follows: 1. What processes lead to the development of E-cadherin-rich junctions and thick F-actin bands in mammalian SCs, but not those of fish, amphibians, or birds? Can these maturational processes be recapitulated in organoids or other in vitro systems to enable high-throughput studies?

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

67

2. Does experimental deletion of E-cadherin activate YAP-TEAD signaling and canonical Wnt signaling and evoke proliferation of mature mammalian SCs? Do the thick F-actin bands stabilize E-cadherin at SC-SC junctions? 3. What upstream signaling processes lead to activation of LATS1/2 kinases in mammalian SCs? After YAP and TEAD mediate cell cycle reentry, are progeny able to differentiate as HCs? Does HC loss evoke YAP-TEAD transcriptional activity in SCs of nonmammals? 4. How significant is integrin signaling at the basal lamina to the regenerative proliferation of SCs? Does the molecular composition or mechanical properties of the basal lamina change with age? Does integrin signaling differ systematically between the HC epithelia of mammals and nonmammals? Does loss of contact with the basement membrane guide HC differentiation? 5. How are the regenerative responses that occur in nonmammalian HC epithelia terminated? What is the role of mechanics in this process? 6. What role does the speed (i.e., dynamic time course) of HC loss play in triggering SC responses?

3.10 Conclusions There appears to be no evidence that mammals are missing a critical component found in nonmammals that have relatively rapid HC regeneration that leads to hearing recovery after deafening and the restoration of balance after HC losses in vestibular epithelia. Instead, the evidence summarized in this chapter indicates that properties unique to the intercellular junctions of mature SCs in mammals appear to have arisen as that class of vertebrates diverged from reptilian ancestors, and that those properties inhibit the regenerative proliferation of mammalian SCs that has been shown to occur in neonates (Burns et al. 2012; Cox et al. 2014). Comparative approaches comparing mammals with nonmammals in regard to cellular structures, tissue organization, responses to damage, and gene expression patterns have proven to be powerful and are likely to continue to be fruitful. Comparative investigations like genetic screens derive power from the discounting of similarities (in this case between mammals and nonmammals) and focus research on those properties that differ and may be responsible for limiting mammalian HC regeneration. In a way, nonmammals serve as an important conceptual and experimental control. The development of reinforced junctions and cytoskeletal features in mammalian SCs appear to be unique process and appears to represent a promising area for further discovery. Direct experimental manipulations of the junctional and cytoskeletal components are needed to further test whether the strong correlations between junctional reinforcement, stiffening of SCs, and decreased SC plasticity are indicative of a causal relationship. Acknowledgments  The authors thank Jennifer Stone, Mark Warchol, and Art Popper for helpful comments.

68

M. A. Rudolf and J. T. Corwin

Compliance with Ethics Requirements  Mark A. Rudolf declares that he has no conflict of interest. Jeffrey T. Corwin declares that he has no conflict of interest.

References Agarwala S, Duquesne S, Liu K et al (2015) Amotl2a interacts with the Hippo effector Yap1 and the Wnt/β-catenin effector Lef1 to control tissue size in zebrafish. eLife 4:e08201. https://doi. org/10.7554/eLife.08201 Anttonen T, Kirjavainen A, Belevich I et al (2012) Cdc42-dependent structural development of auditory supporting cells is required for wound healing at adulthood. Sci Rep 2. https://doi. org/10.1038/srep00978 Anttonen T, Belevich I, Kirjavainen A et al (2014) How to bury the dead: Elimination of apoptotic hair cells from the hearing organ of the mouse. JARO J Assoc Res Otolaryngol 15:975–992. https://doi.org/10.1007/s10162-­014-­0480-­x Anttonen T, Belevich I, Laos M et al (2017) Cytoskeletal Stability in the Auditory Organ In Vivo: RhoA Is Dispensable for Wound Healing but Essential for Hair Cell Development. eNeuro 4. https://doi.org/10.1523/ENEURO.0149-­17.2017 Arnold TR, Stephenson RE, Miller AL (2017) Rho GTPases and Actomyosin: partners in regulating epithelial cell-cell junction structure and function. Exp Cell Res 358:20–30. https://doi. org/10.1016/j.yexcr.2017.03.053 Atkinson PJ, Dong Y, Gu S et  al (2018) Sox2 haploinsufficiency primes regeneration and Wnt responsiveness in the mouse cochlea. J Clin Invest 128:1641–1656. https://doi.org/10.1172/ JCI97248 Balak KJ, Corwin JT, Jones JE (1990) Regenerated hair cells can originate from supporting cell progeny: evidence from phototoxicity and laser ablation experiments in the lateral line system. J Neurosci 10:2502–2512. https://doi.org/10.1523/JNEUROSCI.10-­08-­02502.1990 Benham-Pyle BW, Pruitt BL, Nelson WJ (2015) Mechanical strain induces E-cadherin-dependent Yap1 and β-catenin activation to drive cell cycle entry. Science 348:1024–1027. https://doi. org/10.1126/science.aaa4559 Bird JE, Daudet N, Warchol ME, Gale JE (2010) Supporting cells eliminate dying sensory hair cells to maintain epithelial integrity in the avian inner ear. J Neurosci 30:12545–12556. https:// doi.org/10.1523/JNEUROSCI.3042-­10.2010 Borse V, Barton M, Arndt H et  al (2021) Dynamic patterns of YAP1 expression and cellular localization in the developing and injured utricle. Sci Rep 11:2140. https://doi.org/10.1038/ s41598-­020-­77775-­8 Burns JC, Corwin JT (2013) A historical to present-day account of efforts to answer the question: “What puts the brakes on mammalian hair cell regeneration?”. Hear Res 297:52–67. https:// doi.org/10.1016/j.heares.2013.01.005 Burns JC, Corwin JT (2014) Responses to cell loss become restricted as the supporting cells in mammalian vestibular organs grow thick junctional actin bands that develop high stability. J Neurosci 34:1998–2011. https://doi.org/10.1523/JNEUROSCI.4355-­13.2014 Burns J, Christophel JJ, Collado MS et al (2008) Reinforcement of cell junctions correlates with the absence of hair cell regeneration in mammals and its occurrence in birds. J Comp Neurol 511:396–414. https://doi.org/10.1002/cne.21849 Burns JC, Cox BC, Thiede BR et al (2012) In vivo proliferative regeneration of balance hair cells in newborn mice. J Neurosci 32:6570–6577. https://doi.org/10.1523/JNEUROSCI.6274-­11.2012 Burns JC, Collado MS, Oliver ER, Corwin JT (2013) Specializations of intercellular junctions are associated with the presence and absence of hair cell regeneration in ears from six vertebrate classes. J Comp Neurol 521:1430–1448. https://doi.org/10.1002/cne.23250

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

69

Campbell DP, Chrysostomou E, Doetzlhofer A (2016) Canonical Notch signaling plays an instructive role in auditory supporting cell development. Sci Rep 6. https://doi.org/10.1038/srep19484 Chai R, Xia A, Wang T et al (2011) Dynamic expression of Lgr5, a Wnt target gene, in the developing and mature mouse cochlea. J Assoc Res Otolaryngol 12:455–469. https://doi.org/10.1007/ s10162-­011-­0267-­2 Cohen R, Amir-Zilberstein L, Hersch M et  al (2020) Mechanical forces drive ordered patterning of hair cells in the mammalian inner ear. Nat Commun 11:5137. https://doi.org/10.1038/ s41467-­020-­18894-­8 Collado MS, Burns JC, Meyers JR, Corwin JT (2011a) Variations in shape-sensitive restriction points mirror differences in the regeneration capacities of avian and mammalian ears. PLoS ONE 6. https://doi.org/10.1371/journal.pone.0023861 Collado MS, Thiede BR, Baker W et al (2011b) The postnatal accumulation of junctional E-cadherin is inversely correlated with the capacity for supporting cells to convert directly into sensory hair cells in mammalian balance organs. J Neurosci Off J Soc Neurosci 31:11855–11866. https:// doi.org/10.1523/JNEUROSCI.2525-­11.2011 Corwin JT (1981) Postembryonic production and aging in inner ear hair cells in sharks. J Comp Neurol 201:541–553. https://doi.org/10.1002/cne.902010406 Corwin JT, Cotanche DA (1988) Regeneration of sensory hair cells after acoustic trauma. Science 240:1772–1774 Corwin JT, Jones JE, Katayama A et al (1991) Hair cell regeneration: the identities of progenitor cells, potential triggers and instructive cues. Ciba Found Symp. 160:103-120-130 Cotanche DA (1987) Regeneration of hair cell stereociliary bundles in the chick cochlea following severe acoustic trauma. Hear Res 30:181–195 Cox BC, Chai R, Lenoir A et al (2014) Spontaneous hair cell regeneration in the neonatal mouse cochlea in vivo. Development 141:816–829. https://doi.org/10.1242/dev.103036 Davies D, Magnus C, Corwin JT (2007) Developmental changes in cell–extracellular matrix interactions limit proliferation in the mammalian inner ear. Eur J Neurosci 25:985–998. https://doi. org/10.1111/j.1460-­9568.2007.05355.x Dupont S, Morsut L, Aragona M et al (2011) Role of YAP/TAZ in mechanotransduction. Nature 474:179–183. https://doi.org/10.1038/nature10137 Ebrahim S, Fujita T, Millis BA et al (2013) NMII forms a contractile transcellular sarcomeric network to regulate apical cell junctions and tissue geometry. Curr Biol CB 23:731–736. https:// doi.org/10.1016/j.cub.2013.03.039 Elbediwy A, Vincent-Mistiaen ZI, Spencer-Dene B et al (2016) Integrin signalling regulates YAP and TAZ to control skin homeostasis. Dev Camb Engl 143:1674–1687. https://doi.org/10.1242/ dev.133728 Engler AJ, Sen S, Sweeney HL, Discher DE (2006) Matrix elasticity directs stem cell lineage specification. Cell 126:677–689. https://doi.org/10.1016/j.cell.2006.06.044 Ennomani H, Letort G, Guérin C et al (2016) Architecture and connectivity govern actin network contractility. Curr Biol 26:616–626. https://doi.org/10.1016/j.cub.2015.12.069 Forge A (1985) Outer hair cell loss and supporting cell expansion following chronic gentamicin treatment. Hear Res 19:171–182 Forge A, Li L, Corwin JT, Nevill G (1993) Ultrastructural evidence for hair cell regeneration in the mammalian inner ear. Science 259:1616–1619 Geng R, Noda T, Mulvaney JF et  al (2016) Comprehensive expression of Wnt signaling pathway genes during development and maturation of the mouse cochlea. PloS One 11:e0148339. https://doi.org/10.1371/journal.pone.0148339 Gnedeva K, Hudspeth AJ (2015) SoxC transcription factors are essential for the development of the inner ear. Proc Natl Acad Sci USA 112:14066–14071. https://doi.org/10.1073/ pnas.1517371112 Gnedeva K, Jacobo A, Salvi JD et al (2017) Elastic force restricts growth of the murine utricle. eLife 6:e25681. https://doi.org/10.7554/eLife.25681

70

M. A. Rudolf and J. T. Corwin

Gnedeva K, Wang X, McGovern MM et al (2020) Organ of Corti size is governed by Yap/Tead-­ mediated progenitor self-renewal. Proc Natl Acad Sci. https://doi.org/10.1073/pnas.2000175117 Golub JS, Tong L, Ngyuen TB et al (2012) Hair Cell replacement in adult mouse utricles after targeted ablation of hair cells with diphtheria toxin. J Neurosci 32:15093–15105. https://doi. org/10.1523/JNEUROSCI.1709-­12.2012 Goodyear RJ, Gates R, Lukashkin AN, Richardson GP (1999) Hair-cell numbers continue to increase in the utricular macula of the early posthatch chick. J Neurocytol 28:851–861 Gu R, Montcouquiol M, Marchionni M, Corwin JT (2007) Proliferative responses to growth factors decline rapidly during postnatal maturation of mammalian hair cell epithelia. Eur J Neurosci 25:1363–1372. https://doi.org/10.1111/j.1460-­9568.2007.05414.x Hackett L, Davies D, Helyer R et  al (2002) E-cadherin and the Differentiation of Mammalian Vestibular Hair Cells. Exp Cell Res 278:19–30. https://doi.org/10.1006/excr.2002.5574 Harris AR, Peter L, Bellis J et al (2012) Characterizing the mechanics of cultured cell monolayers. Proc Natl Acad Sci U S A 109:16449–16454. https://doi.org/10.1073/pnas.1213301109 Hunter MV, Lee DM, Harris TJC, Fernandez-Gonzalez R (2015) Polarized E-cadherin endocytosis directs actomyosin remodeling during embryonic wound repair. J Cell Biol 210:801–816. https://doi.org/10.1083/jcb.201501076 Irvine KD, Shraiman BI (2017) Mechanical control of growth: ideas, facts and challenges. Development 144:4238–4248. https://doi.org/10.1242/dev.151902 Ishiyama A, Mowry SE, Lopez IA, Ishiyama G (2009) Immunohistochemical distribution of basement membrane proteins in the human inner ear from older subjects. Hear Res 254:1–14. https://doi.org/10.1016/j.heares.2009.03.014 Jacques BE, Puligilla C, Weichert RM et al (2012) A dual function for canonical Wnt/β-catenin signaling in the developing mammalian cochlea. Development 139:4395–4404. https://doi. org/10.1242/dev.080358 Jansson L, Ebeid M, Shen JW et  al (2019) β-Catenin is required for radial cell patterning and identity in the developing mouse cochlea. Proc Natl Acad Sci USA 116:21054–21060. https:// doi.org/10.1073/pnas.1910223116 Jones JE, Corwin JT (1996) Regeneration of sensory cells after laser ablation in the lateral line system: hair cell lineage and macrophage behavior revealed by time-lapse video microscopy. J Neurosci 16:649–662 Jørgensen JM, Mathiesen C (1988) The avian inner ear. Continuous production of hair cells in vestibular sensory organs, but not in the auditory papilla. Naturwissenschaften 75:319–320. https://doi.org/10.1007/bf00367330 Karsch S, Kong D, Großhans J, Janshoff A (2017) Single-cell defects cause a long-range mechanical response in a confluent epithelial cell layer. Biophys J 113:2601–2608. https://doi. org/10.1016/j.bpj.2017.10.025 Kastan N, Gnedeva K, Alisch T et al (2021) Small-molecule inhibition of Lats kinases may promote Yap-dependent proliferation in postmitotic mammalian tissues. Nat Commun 12:3100. https://doi.org/10.1038/s41467-­021-­23395-­3 Kil J, Warchol ME, Corwin JT (1997) Cell death, cell proliferation, and estimates of hair cell life spans in the vestibular organs of chicks. Hear Res 114:117–126 Kim N-G, Gumbiner BM (2015) Adhesion to fibronectin regulates Hippo signaling via the FAK– Src–PI3K pathway. J Cell Biol 210:503–515. https://doi.org/10.1083/jcb.201501025 Kim T-S, Nakagawa T, Lee J-E et  al (2004) Induction of cell proliferation and beta-­ catenin expression in rat utricles in  vitro. Acta Oto-Laryngol Suppl 22–25. https://doi. org/10.1080/03655230310016672 Kim N-G, Koh E, Chen X, Gumbiner BM (2011) E-cadherin mediates contact inhibition of proliferation through Hippo signaling-pathway components. Proc Natl Acad Sci USA 108:11930–11935. https://doi.org/10.1073/pnas.1103345108 Kozlowski MM, Rudolf MA, Corwin JT (2020) EGF and a GSK3 Inhibitor Deplete Junctional E-cadherin and Stimulate Proliferation in the Mature Mammalian Ear. J Neurosci. https://doi. org/10.1523/JNEUROSCI.2630-­19.2020

3  Cell Junctions and the Mechanics of Hair Cell Regeneration

71

Leonova EV, Raphael Y (1997) Organization of cell junctions and cytoskeleton in the reticular lamina in normal and ototoxically damaged organ of Corti. Hear Res 113:14–28. https://doi. org/10.1016/s0378-­5955(97)00130-­5 Lin V, Golub JS, Nguyen TB et al (2011) Inhibition of notch activity promotes nonmitotic regeneration of hair cells in the adult mouse utricles. J Neurosci 31:15329–15339. https://doi. org/10.1523/JNEUROSCI.2057-­11.2011 Loh S-L, Teh C, Muller J et al (2014) Zebrafish yap1 plays a role in differentiation of hair cells in posterior lateral line. Sci Rep 4. https://doi.org/10.1038/srep04289 Lu Z, Corwin JT (2008) The influence of glycogen synthase kinase 3 in limiting cell addition in the mammalian ear. Dev Neurobiol 68:1059–1075. https://doi.org/10.1002/dneu.20635 Luo W, Wang X, Ma R et  al (2018) Junctional E-cadherin/p120-catenin Is correlated with the absence of supporting cells to hair cells conversion in postnatal mice cochleae. Front Mol Neurosci 11. https://doi.org/10.3389/fnmol.2018.00020 Matamoro-Vidal A, Levayer R (2019) Multiple influences of mechanical forces on cell competition. Curr Biol 29:R762–R774. https://doi.org/10.1016/j.cub.2019.06.030 Matsui JI, Ogilvie JM, Warchol ME (2002) Inhibition of caspases prevents ototoxic and ongoing hair cell death. J Neurosci 22:1218–1227 Matsunaga M, Kita T, Yamamoto R et al (2020) Initiation of supporting cell activation for hair cell regeneration in the avian auditory epithelium: an explant culture model. Front Cell Neurosci 14:583994. https://doi.org/10.3389/fncel.2020.583994 McLean WJ, Yin X, Lu L et al (2017) Clonal expansion of Lgr5-positive cells from mammalian cochlea and high-purity generation of sensory hair cells. Cell Rep 18:1917–1929. https://doi. org/10.1016/j.celrep.2017.01.066 Mendonsa AM, Na T-Y, Gumbiner BM (2018) E-cadherin in contact inhibition and cancer. Oncogene 37:4769–4780. https://doi.org/10.1038/s41388-­018-­0304-­2 Mesa KR, Kawaguchi K, Cockburn K et al (2018) Homeostatic epidermal stem cell self-renewal is driven by local differentiation. Cell Stem Cell 23:677–686.e4. https://doi.org/10.1016/j. stem.2018.09.005 Meyers JR, Corwin JT (2007) Shape change controls supporting cell proliferation in lesioned mammalian balance epithelium. J Neurosci 27:4313–4325. https://doi.org/10.1523/ JNEUROSCI.5023-­06.2007 Miroshnikova YA, Le HQ, Schneider D et al (2018) Adhesion forces and cortical tension couple cell proliferation and differentiation to drive epidermal stratification. Nat Cell Biol 20:69–80. https://doi.org/10.1038/s41556-­017-­0005-­z Montcouquiol M, Corwin JT (2001) Intracellular signals that control cell proliferation in mammalian balance epithelia: key roles for phosphatidylinositol-3 kinase, mammalian target of rapamycin, and s6 kinases in preference to calcium, protein kinase c, and mitogen-activated protein kinase. J Neurosci 21:570–580. https://doi.org/10.1523/JNEUROSCI.21-­02-­00570.2001 Nelson WJ, Nusse R (2004) Convergence of Wnt, beta-catenin, and cadherin pathways. Science 303:1483–1487. https://doi.org/10.1126/science.1094291 Pan Y, Heemskerk I, Ibar C et  al (2016) Differential growth triggers mechanical feedback that elevates Hippo signaling. Proc Natl Acad Sci USA 113:E6974–E6983. https://doi.org/10.1073/ pnas.1615012113 Perrais M, Chen X, Perez-Moreno M, Gumbiner BM (2007) E-cadherin homophilic ligation inhibits cell growth and epidermal growth factor receptor signaling independently of other cell interactions. Mol Biol Cell 18:2013–2025. https://doi.org/10.1091/mbc.e06-­04-­0348 Ravasio A, Cheddadi I, Chen T et al (2015) Gap geometry dictates epithelial closure efficiency. Nat Commun 6:1–13. https://doi.org/10.1038/ncomms8683 Romero-Carvajal A, Acedo JN, Jiang L et al (2015) Regeneration of sensory hair cells requires localized interactions between the Notch and Wnt pathways. Dev Cell 34:267–282. https://doi. org/10.1016/j.devcel.2015.05.025

72

M. A. Rudolf and J. T. Corwin

Rudolf MA, Andreeva A, Kozlowski MM et al (2020) YAP Mediates Hair Cell Regeneration in Balance Organs of Chickens, But LATS Kinases Suppress Its Activity in Mice. J Neurosci 40:3915–3932. https://doi.org/10.1523/JNEUROSCI.0306-­20.2020 Rudolf MA, Andreeva A, Kim CE et al (2022) Stiffening of circumferential F-Actin bands correlates with regenerative failure and may act as a Biomechanical Brake in the Mammalian Inner Ear. Front Cell Neurosci 16. https://doi.org/10.3389/fncel.2022.859882 Ryals BM, Rubel EW (1988) Hair cell regeneration after acoustic trauma in adult Coturnix quail. Science 240:1774–1776. https://doi.org/10.1126/science.3381101 Samarajeewa A, Lenz DR, Xie L et al (2018) Transcriptional response to Wnt activation regulates the regenerative capacity of the mammalian cochlea. Development 145:dev166579. https://doi. org/10.1242/dev.166579 Scheibinger M, Ellwanger DC, Corrales CE et al (2018) Aminoglycoside damage and hair cell regeneration in the chicken utricle. J Assoc Res Otolaryngol 19:17–29. https://doi.org/10.1007/ s10162-­017-­0646-­4 Shi F, Hu L, Jacques BE et  al (2014) β-Catenin Is required for hair-cell differentiation in the cochlea. J Neurosci 34:6470–6479. https://doi.org/10.1523/JNEUROSCI.4305-­13.2014 Szarama KB, Gavara N, Petralia RS et al (2012) Cytoskeletal changes in actin and microtubules underlie the developing surface mechanical properties of sensory and supporting cells in the mouse cochlea. Dev Camb Engl 139:2187–2197. https://doi.org/10.1242/dev.073734 Totaro A, Castellan M, Battilana G et al (2017) YAP/TAZ link cell mechanics to Notch signalling to control epidermal stem cell fate. Nat Commun 8:ncomms15206. https://doi.org/10.1038/ ncomms15206 Totaro A, Panciera T, Piccolo S (2018) YAP/TAZ upstream signals and downstream responses. Nat Cell Biol 20:888–899. https://doi.org/10.1038/s41556-­018-­0142-­z Uroz M, Wistorf S, Serra-Picamal X et al (2018) Regulation of cell cycle progression by cell–cell and cell–matrix forces. Nat Cell Biol 20:646–654. https://doi.org/10.1038/s41556-­018-­0107-­2 Wang T, Chai R, Kim GS et al (2015) Lgr5+ cells regenerate hair cells via proliferation and direct transdifferentiation in damaged neonatal mouse utricle. Nat Commun 6:6613. https://doi. org/10.1038/ncomms7613 Warchol ME (2002) Cell density and N-cadherin interactions regulate cell proliferation in the sensory epithelia of the inner ear. J Neurosci 22:2607–2616 Warchol ME (2007) Characterization of supporting cell phenotype in the avian inner ear: implications for sensory regeneration. Hear Res 227:11–18. https://doi.org/10.1016/j. heares.2006.08.014 Warchol ME, Corwin JT (1996) regenerative proliferation in organ cultures of the avian cochlea: identification of the initial progenitors and determination of the latency of the proliferative response. J Neurosci 16:5466–5477 Warchol ME, Lambert PR, Goldstein BJ et al (1993) Regenerative proliferation in inner ear sensory epithelia from adult guinea pigs and humans. Science 259:1619–1622

Chapter 4

Mammalian Hair Cell Regeneration Ruth Taylor and Andrew Forge

Abstract  In adult mammals, there is no hair cell regeneration in the cochlea and a limited capacity in the vestibular organs. Research in this area is directed toward uncovering a latent potential for regenerating hair cells either through removing blocks or by inducing regenerative responses. In this chapter, we consider the limitations and potential challenges to regeneration to restore functional hair cells in adult mammals. We review briefly features of development and of the structure of the mature mammalian vestibular and auditory sensory epithelia that are relevant to regenerating functional hair cells in adults. The immature inner ear can regenerate hair cells, but this capacity is lost with maturation. Factors that can induce regenerative responses in the immature tissue are discussed. The phenomenology of natural regeneration in the mature mammalian vestibular organs, how it occurs through phenotypic conversion of supporting cells, and its limitations are reviewed. Potential procedures to enhance or stimulate hair cell regeneration in adult mammals, and the outcomes of attempts to do so, including in vivo approaches as a prelude to clinical trials, are described. Finally, we consider how the structural reorganization of the mature organ of Corti after hair cell loss may impact upon possibilities for cochlear hair cell regeneration and suggest that a better understanding of the cellular and molecular environment of that reorganized epithelium is required to identify appropriate regenerative strategies. Keywords  Vestibular · Cochlea · Organ of Corti · Utricle · Atoh1 · Notch Wnt · Lgr5

R. Taylor · A. Forge (*) UCL Ear Institute, London, UK e-mail: [email protected]; [email protected] © Springer Nature Switzerland AG 2023 M. E. Warchol et al. (eds.), Hair Cell Regeneration, Springer Handbook of Auditory Research 75, https://doi.org/10.1007/978-3-031-20661-0_4

73

74

R. Taylor and A. Forge

4.1 Introduction Loss of the hair cells from the organ of Corti is a major cause of acquired hearing loss. Hair cell loss from the sensory epithelia of the vestibular system results in dizziness, vertigo, and balance disequilibrium that particularly affects older people and is a major underlying factor in falls in the elderly (Sayyid et al. 2018). As described by Hewitt, Raible, and Stone in Chap. 2, in nonmammalian vertebrates, natural regeneration of hair cells in both auditory and vestibular organs is spontaneously initiated very soon after loss of the original ones. It leads to near complete restoration of hair cell numbers, innervation of the regenerated hair cells, and recovery of apparently normal function. In contrast, in mammals, there is no regeneration of hair cells in the mature organ of Corti and only a limited capacity in the vestibular epithelia. Consequently, the functional deficits resulting from hair cell loss are permanent. It is perhaps remarkable that nonmammalian vertebrates exhibit rapid, complete hair cell regeneration following unnaturally extensive hair cell loss. There are few agents that are known to kill hair cells principally: loud noise, a small number of pharmaceutical and industrial chemicals, as well as aging. No animals, other than modern humans, in their natural environments will ever be exposed to ototoxins or damaging noise and most will probably die before age-related effects become manifest (e.g., Fabian and Flatt 2011). Thus, it is improbable that conditions that can result in hair cell death and acquired hearing loss and/or balance defects have provided selective pressure during evolution favoring hair cell regeneration. Hair cell regeneration in nonmammalian vertebrates is likely an epiphenomenon, possibly arising from the retention into maturity of developmental and tissue maintenance programs (Brockes and Kumar 2008). The lack of regenerative capacity in the mammalian cochlea could therefore be a consequence of the acquisition during late stages of maturation of the particular structural specializations of the organ of Corti that support the unique micromechanical activities underpinning auditory sensitivity in mammals. However, the structure of the vestibular sensory epithelia is highly conserved across vertebrate classes, and those of mammals are almost indistinguishable from those of birds. The very limited hair cell regenerative capacity of the mammalian ear may reflect more widespread limits on cell and tissue regeneration in mammals when compared to other vertebrates (e.g., Iismaa et al. 2018; Maden and Varholick 2020), or it might be the consequence of specific functional requirements in the mammalian inner ear, the development of which suppress regenerative responses (Rudolf and Corwin, Chap. 3). Discovering how to uncover a latent regenerative potential by overcoming any blocks or through cellular reprogramming is therefore of considerable biological interest and would provide insight into the possibilities for inducing or enhancing hair cell regeneration in the mammalian inner ear as a therapeutic route to ameliorate acquired hearing loss or balance dysfunction. In this chapter, we consider regeneration in the inner ear of mature mammals and the limitations and potential challenges to inducing restoration of hair cells and full

4  Mammalian Hair Cell Regeneration

75

functional recovery. We begin with a brief discussion of the development and structure of auditory and vestibular sensory epithelia of mammals because complete regeneration that restores function fully will require reestablishing the structural and architectural features of the undamaged, mature tissue.

4.2 Structural and Developmental Considerations As described in detail in other chapters in this volume and in excellent reviews (Basch et al. 2016; Driver and Kelley 2020), the two cell types that comprise the sensory epithelia, hair cells and non-sensory supporting cells, derive from common precursors. Those precursors express the transcription factor Atoh1 that determines differentiation as a hair cell. The first cells that begin to express Atoh1 inhibit those neighboring cells with which they are in contact from following the same fate through activation of the Notch signaling pathway (reviewed in Kiernan 2013). The inhibited cells differentiate as supporting cells. The consequent cellular arrangement in which each hair cell is surrounded by supporting cells and thereby separated from neighboring hair cells is of functional significance. Such “social distancing” allows each hair cell to act as an independent sensor of stimulating signals, physiologically insulated from the activities of its neighboring sensory cells, thereby enabling fine-grained sensory information to be conveyed to the brain. This mosaic pattern of hair cells and intervening supporting cells will need to be recreated when hair cells are regenerated if function is to be fully restored.

4.2.1 Structure of the Mammalian Vestibular Sensory Epithelia The sensory epithelia of the mammalian vestibular system – the maculae of the saccule and utricle and the cristae of the semicircular canals – contain two distinct hair cell types (reviewed in Eatock and Songer 2011). They are distinguishable by morphology, certain molecular markers (Desai et al. 2005; Oesterle et al. 2008; McInturff et al. 2018), electrophysiological signatures (Eatock 2018), and innervation patterns (Fernandez et al. 1990; Eatock and Songer 2011; Morley et al. 2017). Type I hair cells are flask-shaped with the entire basolateral surface enclosed within a single, large afferent nerve terminal called a calyx. Efferent nerves synapse with the afferent calyx. Type II hair cells are generally described as cylindrical but often possess basolaterally directed projections (Pujol et al. 2014) (Fig. 4.1a). They are innervated by several discrete bouton afferent nerve endings as well as directly by efferent endings. Type I hair cells provide rapid responses to stimuli and thus provide information on changes in head position more rapidly than type II (Eatock 2018). This would contribute to sensing an incipient fall and providing information necessary for posture and motion readjustments to prevent it.

76

R. Taylor and A. Forge

Fig. 4.1  Utricular macula. (a) Type I (TI) hair cells are flask shaped with cell body enclosed in an afferent nerve calyx, indicated by red arrows. Type II (TII) hair cells are cylindrical, some with basal projections (black arrowheads) and innervated by bouton endings of afferent nerves (blue arrows). Bodies of the supporting cells (SC) intervene between the base of the hair cell and the underlying basement membrane. Hair cells do not contact the basement membrane. (b) Luminal surface of the entire macula of a guinea pig. The approximate extent of the striola is outlined in red. (c) Region of reversal of polarity of hair bundles within the striola. Arrows indicate the direction of gradient of the increase in height of stereocilia toward the longer kinocilium that defines hair bundle polarity

In the middle of the epithelial sheet that comprises the macula of the utricle (the vestibular sensory patch most widely studied for regenerative events) and that of the saccule is a distinct band, the striola, that occupies about 15% of the macular area and is shaped as a C (or U) (Desai et al. 2005) (Fig. 4.1b). A number of molecular markers distinguish the striola from the extrastriolar region (e.g., Leonard and Kevetter 2002; Hoffman et al. 2018). Along the striola (and extending to the peripheries) is a line of “polarity reversal” where the hair bundles of all hair cells (type I and type II) on one side are oriented in the same direction and at precisely 180° to those on the opposite side (e.g., Denman-Johnson and Forge 1999; Li et al. 2008) (Fig. 4.1c). Type I hair cells predominate over type II in the striola; there are similar numbers of the two hair cell types in the extrastriolar regions, but the hair cells toward the periphery are almost exclusively type II (Fernandez et al. 1990; Desai et al. 2005). The physiological properties of the hair cells across the striola differ from those in extrastriolar regions (Eatock 2018). There are also differences in the expression of certain molecules between type 1 hair cells in the striola and those in extrastriolar regions (Leonard and Kevetter 2002; Simmons et al. 2010; McInturff et al. 2018). There is also some complexity in the pattern of vestibular hair cell innervation (reviewed in Eatock and Songer 2011; Burns and Stone 2017). A single afferent

4  Mammalian Hair Cell Regeneration

77

nerve may form calyceal endings with only one or more than one type I hair cell. Other afferent nerves, termed “dimorphic,” form calyces with type I and bouton endings with type II hair cells. A smaller proportion of afferent nerves form bouton endings exclusively, innervating one or more type II hair cells. Afferent nerves that form only calyceal endings are confined to the striola. Those that form only bouton endings are present in peripheral extrastriolar regions. The majority of afferent nerves are dimorphic (Fernandez et al. 1990; Eatock and Songer 2011). The precise functional significance of the various innervation patterns is not known, but if regenerated hair cells are to restore normal function, then recreation of something close to normal innervation patterns is likely to be required. Currently, it is not known how during development axons of the vestibular nerve target particular hair cell types. Thus, it is impossible to know how appropriate innervation patterns might be recreated for regenerated hair cells. Supporting cells in striolar and extrastriolar regions also have some distinct molecular characteristics (Yin et al. 2012; McInturff, et al. 2018). Transgenic mice that express specific fluorescent protein reporters have been developed to identify and enable fate mapping of supporting cells (Stone et al. 2018). Through the use of such technology, subpopulations of supporting cells have been recognized. Supporting cells in extrastriolar regions, but not those in the striola, express proteolipid protein 1 (PLP1) (Gomez-Casati et al. 2010; Burns et al. 2012), which is normally expressed by myelin-producing cells in the nervous system. In adult animals therefore, there appears to be at least two spatially distinct subpopulations of supporting cells, one confined to the striola and the other in the extrastriolar regions. There is no striolar region in the saddle-shaped cristae, and the hair bundles of all hair cells in an individual crista are oriented in the same direction. There are, however, differences in hair cell densities and in the relative proportions of type I and type II hair cells and their innervation patterns between the central crest and the more peripheral skirts of the cristae, with the crest resembling the striolae of the maculae (Goldberg et al. 1990; Eatock 2018). There are also regional similarities in sensitivity of maculae and cristae to agents that cause hair cell loss. With both aminoglycosides and aging, hair cell loss is first apparent at the crest of the cristae and across the macular striolae spreading progressively with time and/or dose down the skirts of the crista and into extrastriolar regions of the maculae (e.g., Lindeman 1969; Wright 1983).

4.2.2 Development of the Vestibular Sensory Epithelia Hair cells in the vestibular sensory epithelia are generated over a prolonged period: in mice from about embryonic day(E)12 (Denman-Johnson and Forge 1999) until around postnatal day (P)8. Over 50% of the mature adult, complement of hair cells are generated in the first postnatal week, and they are mostly added at the periphery of the sensory epithelium (Burns et al. 2012; Warchol, et al. 2019). From almost their earliest inception, type I and type II hair cells are differentiated

78

R. Taylor and A. Forge

morphologically, and separate physiological signatures are apparent (in mice) by E18 (Rusch et al. 1998; Geleoc et al. 2004). It is not clear how these two hair cell types are initially specified. Cell fate mapping has shown that almost all the hair cells generated during the postnatal period are type II (Wang et al. 2019; Warchol et al. 2019). They increase in number as the epithelium grows in size. However, the factors underlying the origin of type I hair cells and their distribution across the striola and into extrastriolar regions have not yet been elucidated. Although generated early, type I hair cells, identified by their characteristic molecular markers and morphology, only become innervated during later stages of maturation, in line with innervation of type II hair cells (Warchol et al. 2019). This temporal coordination may be one factor involved in setting the patterns of innervation seen in mature tissue.

4.2.3 Structure of the Mammalian Auditory Epithelium, the Organ of Corti 4.2.3.1 Hair Cell Types The architecture of the organ of Corti is precisely organized (Fig. 4.2) to support its particular functional specialization (Forge and Wright 2002). Hair cells with intervening supporting cells are regularly arranged in a single row of inner hair cells (IHC) and three rows of outer hair cells (OHC) (although there is somewhat greater irregularity in the organ of Corti of humans (Wright 1984)). IHCs are the primary receptor cells, each one innervated by ten or more individual afferent nerves that send auditory information to the brain. OHCs are primarily effector cells. They respond to sound by undergoing rapid length changes that act to locally enhance and finely sharpen the signal presented to the IHCs in their vicinity (reviewed in Ashmore et al. 2010). OHC activity is responsible for the ability to distinguish quieter sounds (85% of all supporting cell Walters et al. (2015) single injection subtypes; no hair cells P60, 75mg/kg Most supporting cells, inner and B. Walters, personal injected once outer sulcus cells, interdental cells, observation per day for 5 cells in the spiral ligament; no hair days cells, little expression in pillar cells

phalangeal, border, pillar, and Deiters’ cells, but there is significant variability across samples likely caused by X chromosome linkage (M.  McGovern and A. Groves, personal communication). While Lgr5CreERT2 mice have only been characterized after tamoxifen induction at neonatal ages, Lgr5 expression becomes confined to the third row of Deiters’ cells by postnatal day (P) 12 (Chai et al. 2011). Thus, this cell type could theoretically be targeted at adult ages using Lgr5CreERT2. Nestin-CreERT2 is expressed in a unique cell type, located medial to and below inner hair cells, that expresses the supporting cell marker Sox2 but not other markers of Schwann cells or supporting cells (Table 8.2) (Chow et al. 2015). Not much is known about these cells, but based on Nestin’s functions in other cell types, they may be a niche of latent stem cells that could be targeted to stimulate hair cell regeneration (Chow et al. 2016). Similarly, tamoxifen induction of Eya1CreERT2 at neonatal or juvenile ages targets a population of cells in the spiral ganglion that can clonally expand to produce hair cells and supporting cells in the undamaged cochlea, as well as produce regenerated hair cells after damage. While it is unknown whether these Eya1-positive cells persist in the adult cochlea, they may also be a stem cell niche worthy of targeting for hair cell regeneration studies (Xu et al. 2017). Given these limited options, new CreERT lines are critically needed to target individual supporting cell subtypes or to increase the number of supporting cells with CreERT activity while excluding other cell types. Recent single-cell RNA-seq studies using neonatal and adult cochleae can provide a resource to develop these needed tools (Cheng et al. 2019; Orvis et al. 2021).

206

B. C. Cox et al.

Table 8.3 CreERT alleles that target adult vestibular supporting cells

Allele name Id2CreERT2

GLAST-­ CreERT2 Lfng-CreERT2

Age and dose of tamoxifen induction 6 weeks; 9mg/40g 2 injections 6 weeks; 9mg/40g 2 injections P21; 9mg/40g 2 injections

Plp-CreERT2

6 weeks; 9mg/40g 2 injections

Sox2CreERT2

6 weeks; 9mg/40g 2 injections 6 weeks; 9mg/40g 2 injections

Sox9-CreERT2

CreERT expression pattern 29% supporting cells,