Corneal Regeneration: Methods and Protocols [1st ed.] 9781071605981, 9781071605998

This volume details different laboratory techniques used to develop regenerative therapies that help treat corneal blind

490 116 8MB

English Pages XII, 250 [253] Year 2020

Report DMCA / Copyright


Polecaj historie

Corneal Regeneration: Methods and Protocols [1st ed.]
 9781071605981, 9781071605998

Table of contents :
Front Matter ....Pages i-xii
Isolation and Culture of Corneal Stromal Stem Cells (Richard M. Nagymihaly, Morten C. Moe, Goran Petrovski)....Pages 1-15
In Vitro Expansion of Corneal Endothelial Cells for Transplantation (Kim Santerre, Isabelle Xu, Mathieu Thériault, Stéphanie Proulx)....Pages 17-27
Primary Culture of Cornea-Limbal Epithelial Cells In Vitro (Finbarr O’Sullivan)....Pages 29-37
Optimization of Human Limbal Stem Cell Culture by Replating a Single Limbal Explant (Marina López-Paniagua, Teresa Nieto-Miguel, Sara Galindo, Laura García-Posadas, Ana de la Mata, Rosa M. Corrales et al.)....Pages 39-49
A Guide to the Development of Human CorneaOrganoids from Induced Pluripotent Stem Cells in Culture (James W. Foster, Karl J. Wahlin, Shukti Chakravarti)....Pages 51-58
Gene Editing for Corneal Stromal Regeneration (Tara Moore, Connie Chao-Shern, Larry DeDionisio, Kathleen A. Christie, M. Andrew Nesbit)....Pages 59-75
Preparation and Administration of Adeno-associated Virus Vectors for Corneal Gene Delivery (Liujiang Song, Jacquelyn J. Bower, Matthew L. Hirsch)....Pages 77-102
The Self-assembly Approach as a Tool for the Tissue Engineering of a Bi-lamellar Human Cornea (Gaëtan Le-Bel, Pascale Desjardins, Camille Couture, Lucie Germain, Sylvain L. Guérin)....Pages 103-118
Formation of Corneal Stromal-Like Assemblies Using Human Corneal Fibroblasts and Macromolecular Crowding (Mehmet Gürdal, Gülinnaz Ercan, Dimitrios I. Zeugolis)....Pages 119-141
Preparation of Dried Amniotic Membrane for Corneal Repair (Andrew Hopkinson, Emily R. Britchford, Laura E. Sidney)....Pages 143-157
Fabrication of Corneal Extracellular Matrix-Derived Hydrogels (Mark Ahearne, Julia Fernández-Pérez)....Pages 159-168
Synthesis and Application of Collagens for Assembling a Corneal Implant (Elle Edin, Fiona Simpson, May Griffith)....Pages 169-183
Development and Validation of a 3D In Vitro Model to Study the Chemotactic Behavior of Corneal Stromal Fibroblasts (Evrim Ceren Kabak, Julia Fernández-Pérez, Mark Ahearne)....Pages 185-196
Femtosecond Laser-Assisted Surgery for Implantation of Bioengineered Corneal Stroma to Promote Corneal Regeneration (Neil Lagali, Mehrdad Rafat)....Pages 197-214
The Use of Animal Models to Assess Engineered Corneal Tissue (Robert Thomas Brady, Peter W. Madden)....Pages 215-230
X-Ray Diffraction Imaging of Corneal Ultrastructure (Keith M. Meek, Andrew J. Quantock, Sally Hayes, James Bell)....Pages 231-247
Back Matter ....Pages 249-250

Citation preview

Methods in Molecular Biology 2145

Mark Ahearne Editor

Corneal Regeneration Methods and Protocols




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

For further volumes:

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

Corneal Regeneration Methods and Protocols

Edited by

Mark Ahearne Trinity Centre for Biomedical Engineering, Trinity College Dublin, Dublin, Ireland

Editor Mark Ahearne Trinity Centre for Biomedical Engineering Trinity College Dublin Dublin, Ireland

ISSN 1064-3745 ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-0716-0598-1 ISBN 978-1-0716-0599-8 (eBook) © Springer Science+Business Media, LLC, part of Springer Nature 2020 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Humana imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer Nature. The registered company address is: 1 New York Plaza, New York, NY 10004, U.S.A.

Preface Corneal blindness is one of the most common causes of blindness worldwide. To treat this condition, there has been much interest in trying to develop new therapies to regenerate or repair damaged and diseased corneal tissue. This book will provide a detailed overview of several laboratory techniques that are used to develop regenerative therapies to help treat corneal blindness. These include how to optimize cell culture conditions, how to apply gene-editing techniques, how to prepare different types of scaffold for corneal regeneration, and how to evaluate the success of these therapies using in vitro and in vivo models and cell and material characterization techniques. This book will be of interest to new and experienced laboratory researchers working on different aspects of corneal regeneration as well as ophthalmologists and patients interested in learning more about the latest techniques and technology. Dublin, Ireland

Mark Ahearne


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

v ix

1 Isolation and Culture of Corneal Stromal Stem Cells . . . . . . . . . . . . . . . . . . . . . . . . Richard M. Nagymihaly, Morten C. Moe, and Goran Petrovski 2 In Vitro Expansion of Corneal Endothelial Cells for Transplantation . . . . . . . . . . Kim Santerre, Isabelle Xu, Mathieu The´riault, and Ste´phanie Proulx 3 Primary Culture of Cornea-Limbal Epithelial Cells In Vitro . . . . . . . . . . . . . . . . . . Finbarr O’Sullivan 4 Optimization of Human Limbal Stem Cell Culture by Replating a Single Limbal Explant. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Marina Lo pez-Paniagua, Teresa Nieto-Miguel, Sara Galindo, Laura Garcı´a-Posadas, Ana de la Mata, Rosa M. Corrales, Margarita Calonge, and Yolanda Diebold 5 A Guide to the Development of Human CorneaOrganoids from Induced Pluripotent Stem Cells in Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . James W. Foster, Karl J. Wahlin, and Shukti Chakravarti 6 Gene Editing for Corneal Stromal Regeneration . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tara Moore, Connie Chao-Shern, Larry DeDionisio, Kathleen A. Christie, and M. Andrew Nesbit 7 Preparation and Administration of Adeno-associated Virus Vectors for Corneal Gene Delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Liujiang Song, Jacquelyn J. Bower, and Matthew L. Hirsch 8 The Self-assembly Approach as a Tool for the Tissue Engineering of a Bi-lamellar Human Cornea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gae¨tan Le-Bel, Pascale Desjardins, Camille Couture, Lucie Germain, and Sylvain L. Gue´rin 9 Formation of Corneal Stromal-Like Assemblies Using Human Corneal Fibroblasts and Macromolecular Crowding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ¨ rdal, Gu ¨ linnaz Ercan, and Dimitrios I. Zeugolis Mehmet Gu 10 Preparation of Dried Amniotic Membrane for Corneal Repair . . . . . . . . . . . . . . . . Andrew Hopkinson, Emily R. Britchford, and Laura E. Sidney 11 Fabrication of Corneal Extracellular Matrix-Derived Hydrogels . . . . . . . . . . . . . . Mark Ahearne and Julia Ferna´ndez-Pe´rez 12 Synthesis and Application of Collagens for Assembling a Corneal Implant . . . . . Elle Edin, Fiona Simpson, and May Griffith 13 Development and Validation of a 3D In Vitro Model to Study the Chemotactic Behavior of Corneal Stromal Fibroblasts . . . . . . . . . . . . . . . . . . . Evrim Ceren Kabak, Julia Ferna´ndez-Pe´rez, and Mark Ahearne



17 29


51 59



119 143 159 169




15 16


Femtosecond Laser-Assisted Surgery for Implantation of Bioengineered Corneal Stroma to Promote Corneal Regeneration. . . . . . . . . . 197 Neil Lagali and Mehrdad Rafat The Use of Animal Models to Assess Engineered Corneal Tissue . . . . . . . . . . . . . 215 Robert Thomas Brady and Peter W. Madden X-Ray Diffraction Imaging of Corneal Ultrastructure . . . . . . . . . . . . . . . . . . . . . . . 231 Keith M. Meek, Andrew J. Quantock, Sally Hayes, and James Bell

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


Contributors MARK AHEARNE • Department of Mechanical and Manufacturing Engineering, School of Engineering, Trinity College Dublin, The University of Dublin, Dublin, Ireland; Trinity Centre for Biomedical Engineering, Trinity Biomedical Sciences Institute, Trinity College Dublin, The University of Dublin, Dublin, Ireland JAMES BELL • Structural Biophysics Group, School of Optometry and Vision Sciences, Cardiff University, Cardiff, UK JACQUELYN J. BOWER • Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA ROBERT THOMAS BRADY • Department of Ophthalmology, Mater Misericordiae University Hospital, Dublin, Ireland EMILY R. BRITCHFORD • Academic Ophthalmology, Division of Clinical Neuroscience, School of Medicine, University of Nottingham, Nottingham, UK; NuVision Biotherapies Ltd, MediCity, Nottingham, UK MARGARITA CALONGE • Grupo de Superficie Ocular, Instituto Universitario de Oftalmobiologı´a Aplicada (IOBA), Universidad de Valladolid, Valladolid, Spain; Centro de Investigacion Biome´dica en Red de Bioingenierı´a, Biomateriales y Nanomedicina (CIBER-BBN). Instituto de Salud Carlos III, Madrid, Spain SHUKTI CHAKRAVARTI • Department of Ophthalmology and Pathology, NYU Langone Health, Alexandria Life Sciences Center, New York, NY, USA CONNIE CHAO-SHERN • Biomedical Sciences Research Institute, Ulster University, Coleraine, Northern Ireland, UK; Avellino Lab USA, Inc., Menlo Park, CA, USA KATHLEEN A. CHRISTIE • Biomedical Sciences Research Institute, Ulster University, Coleraine, Northern Ireland, UK ROSA M. CORRALES • Grupo de Superficie Ocular, Instituto Universitario de Oftalmobiologı´a Aplicada (IOBA), Universidad de Valladolid, Valladolid, Spain; Centro de Investigacion Biome´dica en Red de Bioingenierı´a, Biomateriales y Nanomedicina (CIBER-BBN). Instituto de Salud Carlos III, Madrid, Spain CAMILLE COUTURE • Centre universitaire d’ophtalmologie - recherche (CUO-Recherche) et, Universite´ Laval, Que´bec, QC, Canada; Centre de recherche en organoge´ne`se expe´rimentale de l’Universite´ Laval/LOEX, Centre de recherche du CHU de Que´bec-Universite´ Laval, Universite´ Laval, Que´bec, QC, Canada; De´partement de chirurgie, Faculte´ de me´decine, Universite´ Laval, Que´bec, QC, Canada; De´partement d’ophtalmologie, Faculte´ de Me´ decine, Universite´ Laval, Que´bec, QC, Canada LARRY DEDIONISIO • Avellino Lab USA, Inc., Menlo Park, CA, USA PASCALE DESJARDINS • Centre universitaire d’ophtalmologie - recherche (CUO-Recherche) et, Universite´ Laval, Que´bec, QC, Canada; Centre de recherche en organoge´ne`se expe´rimentale de l’Universite´ Laval/LOEX, Centre de recherche du CHU de Que´bec-Universite´ Laval, Universite´ Laval, QCQue´bec, Canada; De´partement d’ophtalmologie, Faculte´ de Me´decine, Universite´ Laval, Que´bec, QC, Canada YOLANDA DIEBOLD • Grupo de Superficie Ocular, Instituto Universitario de Oftalmobiologı´a Aplicada (IOBA), Universidad de Valladolid, Valladolid, Spain; Centro de Investigacion Biome´dica en Red de Bioingenierı´a, Biomateriales y Nanomedicina (CIBER-BBN). Instituto de Salud Carlos III, Madrid, Spain




ELLE EDIN • Maisonneuve-Rosemont Hospital Research Centre, Montre´al, QC, Canada; Department of Ophthalmology and Institute of Biomedical Engineering, Universite´ de Montre´al, Montre´al, QC, Canada GU¨LINNAZ ERCAN • Faculty of Medicine, Department of Medical Biochemistry, Ege University, Izmir, Turkey; Department of Stem Cell, Institute of Health Sciences, Ege University, Izmir, Turkey JULIA FERNA´NDEZ-PE´REZ • Department of Mechanical and Manufacturing Engineering, School of Engineering, Trinity College Dublin, The University of Dublin, Dublin, Ireland; Trinity Centre for Biomedical Engineering, Trinity Biomedical Sciences Institute, Trinity College Dublin, The University of Dublin, Dublin, Ireland JAMES W. FOSTER • The Wilmer Eye Institute, Johns Hopkins University, Baltimore, MD, USA SARA GALINDO • Grupo de Superficie Ocular, Instituto Universitario de Oftalmobiologı´a Aplicada (IOBA), Universidad de Valladolid, Valladolid, Spain; Centro de Investigacion Biome´dica en Red de Bioingenierı´a, Biomateriales y Nanomedicina (CIBER-BBN). Instituto de Salud Carlos III, Madrid, Spain LAURA GARCI´A-POSADAS • Grupo de Superficie Ocular, Instituto Universitario de Oftalmobiologı´a Aplicada (IOBA), Universidad de Valladolid, Valladolid, Spain LUCIE GERMAIN • Centre universitaire d’ophtalmologie - recherche (CUO-Recherche) et, Universite´ Laval, Que´bec, QC, Canada; Centre de recherche en organoge´ne`se expe´rimentale de l’Universite´ Laval/LOEX, Centre de recherche du CHU de Que´bec-Universite´ Laval, Universite´ Laval, QCQue´bec, Canada; De´partement de chirurgie, Faculte´ de me´decine, Universite´ Laval, Que´bec, QC, Canada; De´partement d’ophtalmologie, Faculte´ de Me´ decine, Universite´ Laval, Que´bec, QC, Canada MAY GRIFFITH • Maisonneuve-Rosemont Hospital Research Centre, Montre´al, QC, Canada; Department of Ophthalmology and Institute of Biomedical Engineering, Universite´ de Montre´al, Montre´al, QC, Canada SYLVAIN L. GUE´RIN • Centre universitaire d’ophtalmologie - recherche (CUO-Recherche) et, Universite´ Laval, Que´bec, QC, Canada; Centre de recherche en organoge´ne`se expe´rimentale de l’Universite´ Laval/LOEX, Centre de recherche du CHU de Que´bec-Universite´ Laval, Universite´ Laval, Que´bec, QC, Canada; De´partement d’ophtalmologie, Faculte´ de Me´ decine, Universite´ Laval, Que´bec, QC, Canada MEHMET GU¨RDAL • Regenerative, Modular & Developmental Engineering Laboratory (REMODEL), National University of Ireland Galway (NUI Galway), Galway, Ireland; ´ RAM), Science Foundation Ireland (SFI) Centre for Research in Medical Devices (CU National University of Ireland Galway (NUI Galway), Galway, Ireland; Faculty of Medicine, Department of Medical Biochemistry, Ege University, Izmir, Turkey SALLY HAYES • Structural Biophysics Group, School of Optometry and Vision Sciences, Cardiff University, Cardiff, UK MATTHEW L. HIRSCH • Gene Therapy Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; Department of Ophthalmology, University of North Carolina, Chapel Hill, NC, USA ANDREW HOPKINSON • Academic Ophthalmology, Division of Clinical Neuroscience, School of Medicine, University of Nottingham, Nottingham, UK; NuVision Biotherapies Ltd, MediCity, Nottingham, UK EVRIM CEREN KABAK • Trinity Centre for Biomedical Engineering, Trinity Biomedical Sciences Institute, Trinity College Dublin, The University of Dublin, Dublin, Ireland; Biotechnology Unit, Nobel Pharmaceuticals A.S, Gebze, Kocaeli, Turkey



NEIL LAGALI • Division of Ophthalmology, Institute for Biomedical and Clinical Sciences, Linko¨ping University, Linko¨ping, Sweden GAE¨TAN LE-BEL • Centre universitaire d’ophtalmologie - recherche (CUO-Recherche) et, Universite´ Laval, Que´bec, QC, Canada; Centre de recherche en organoge´ne`se expe´rimentale de l’Universite´ Laval/LOEX, Centre de recherche du CHU de Que´bec-Universite´ Laval, Universite´ Laval, Que´bec, QC, Canada; De´partement de chirurgie, Faculte´ de me´decine, Universite´ Laval, Que´bec, QC, Canada; De´partement d’ophtalmologie, Faculte´ de Me´ decine, Universite´ Laval, Que´bec, QC, Canada MARINA LO´PEZ-PANIAGUA • Grupo de Superficie Ocular, Instituto Universitario de Oftalmobiologı´a Aplicada (IOBA), Universidad de Valladolid, Valladolid, Spain; Centro de Investigacion Biome´dica en Red de Bioingenierı´a, Biomateriales y Nanomedicina (CIBER-BBN). Instituto de Salud Carlos III, Madrid, Spain PETER W. MADDEN • Department of Mechanical and Manufacturing Engineering, School of Engineering, Trinity College Dublin, University of Dublin, Dublin, Ireland; Trinity Centre for Biomedical Engineering, Trinity Biomedical Science Institute, Trinity College Dublin, University of Dublin, Dublin, Ireland ANA DE LA MATA • Grupo de Superficie Ocular, Instituto Universitario de Oftalmobiologı´a Aplicada (IOBA), Universidad de Valladolid, Valladolid, Spain; Centro de Investigacion Biome´dica en Red de Bioingenierı´a, Biomateriales y Nanomedicina (CIBER-BBN). Instituto de Salud Carlos III, Madrid, Spain KEITH M. MEEK • Structural Biophysics Group, School of Optometry and Vision Sciences, Cardiff University, Cardiff, UK MORTEN C. MOE • Department of Ophthalmology, Center for Eye Research, Oslo University Hospital, Oslo, Norway; Faculty of Medicine, Institute of Clinical Medicine, University of Oslo, Oslo, Norway TARA MOORE • Biomedical Sciences Research Institute, Ulster University, Coleraine, Northern Ireland, UK; Avellino Lab USA, Inc., Menlo Park, CA, USA RICHARD M. NAGYMIHALY • Department of Ophthalmology, Center for Eye Research, Oslo University Hospital, Oslo, Norway M. ANDREW NESBIT • Biomedical Sciences Research Institute, Ulster University, Coleraine, Northern Ireland, UK TERESA NIETO-MIGUEL • Grupo de Superficie Ocular, Instituto Universitario de Oftalmobiologı´a Aplicada (IOBA), Universidad de Valladolid, Valladolid, Spain; Centro de Investigacion Biome´dica en Red de Bioingenierı´a, Biomateriales y Nanomedicina (CIBER-BBN). Instituto de Salud Carlos III, Madrid, Spain FINBARR O’SULLIVAN • National Institute for Cellular Biotechnology and SSPC-SFI, Centre for Pharmaceuticals, Dublin City University, Dublin, Ireland GORAN PETROVSKI • Department of Ophthalmology, Center for Eye Research, Oslo University Hospital, Oslo, Norway; Faculty of Medicine, Institute of Clinical Medicine, University of Oslo, Oslo, Norway STE´PHANIE PROULX • Centre de recherche du Centre hospitalier universitaire (CHU) de Que´ bec—Universite´ Laval, axe me´decine re´ge´ne´ratrice, Hoˆpital du Saint-Sacrement, Que´bec, QC, Canada; De´partement d’Ophtalmologie et d’oto-rhino-laryngologie-chirurgie cervicofaciale, Faculte´ de me´decine, Universite´ Laval, Que´bec, QC, Canada; Centre de recherche en organoge´ne`se expe´rimentale de l’Universite´ Laval/LOEX, Que´bec, QC, Canada ANDREW J. QUANTOCK • Structural Biophysics Group, School of Optometry and Vision Sciences, Cardiff University, Cardiff, UK



MEHRDAD RAFAT • Department of Biomedical Engineering, Linko¨ping University, Linko¨ping, Sweden; LinkoCare Life Sciences AB, Linko¨ping, Sweden KIM SANTERRE • Centre de recherche du Centre hospitalier universitaire (CHU) de Que´bec— Universite´ Laval, axe me´decine re´ge´ne´ratrice, Hoˆpital du Saint-Sacrement, Que´bec, QC, Canada; De´partement d’Ophtalmologie et d’oto-rhino-laryngologie-chirurgie cervicofaciale, Faculte´ de me´decine, Universite´ Laval, Que´bec, QC, Canada; Centre de recherche en organoge´ne`se expe´rimentale de l’Universite´ Laval/LOEX, Que´bec, QC, Canada LAURA E. SIDNEY • Academic Ophthalmology, Division of Clinical Neuroscience, School of Medicine, University of Nottingham, Nottingham, UK FIONA SIMPSON • Maisonneuve-Rosemont Hospital Research Centre, Montre´al, QC, Canada; Department of Ophthalmology and Institute of Biomedical Engineering, Universite´ de Montre´al, Montre´al, QC, Canada LIUJIANG SONG • Gene Therapy Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA; Department of Ophthalmology, University of North Carolina, Chapel Hill, NC, USA MATHIEU THE´RIAULT • Centre de recherche du Centre hospitalier universitaire (CHU) de Que´bec—Universite´ Laval, axe me´decine re´ge´ne´ratrice, Hoˆpital du Saint-Sacrement, Que´ bec, QC, Canada; De´partement d’Ophtalmologie et d’oto-rhino-laryngologie-chirurgie cervico-faciale, Faculte´ de me´decine, Universite´ Laval, Que´bec, QC, Canada; Centre de recherche en organoge´ne`se expe´rimentale de l’Universite´ Laval/LOEX, Que´bec, QC, Canada KARL J. WAHLIN • Shiley Eye Institute, UC San Diego, La Jolla, CA, USA ISABELLE XU • Centre de recherche du Centre hospitalier universitaire (CHU) de Que´bec— Universite´ Laval, axe me´decine re´ge´ne´ratrice, Hoˆpital du Saint-Sacrement, Que´bec, QC, Canada; De´partement d’Ophtalmologie et d’oto-rhino-laryngologie-chirurgie cervicofaciale, Faculte´ de me´decine, Universite´ Laval, Que´bec, QC, Canada; Centre de recherche en organoge´ne`se expe´rimentale de l’Universite´ Laval/LOEX, Que´bec, QC, Canada DIMITRIOS I. ZEUGOLIS • Regenerative, Modular & Developmental Engineering Laboratory (REMODEL), National University of Ireland Galway (NUI Galway), Galway, Ireland; ´ RAM), Science Foundation Ireland (SFI) Centre for Research in Medical Devices (CU National University of Ireland Galway (NUI Galway), Galway, Ireland

Chapter 1 Isolation and Culture of Corneal Stromal Stem Cells Richard M. Nagymihaly, Morten C. Moe, and Goran Petrovski Abstract An increasing body of evidence authenticates the benefit of corneal stroma-derived stem cells (CSSCs) in tissue engineering and regeneration oriented research, and potentially in the development of clinically relevant cellular therapies. Postmortem corneal tissue obtained from otherwise discarded material after keratoplasties is oftentimes the source of the cells for ex vivo research. Relatively easy to isolate and cultivate as well as inexpensive to culture, CSSCs now represent a well-described cell type with attributes of mesenchymal stem cells (MSCs). These include differentiation- and immunosuppressive potential, as well as a favorable capacity to expand in vitro. Here, we in detail describe two straightforward methods to isolate and establish CSSC cultures ex vivo. Key words Cornea, Corneal stroma, Keratocytes, Corneal fibroblasts, Corneal stromal stem cells, Mesenchymal stem cells, Isolation, Cell culturing


Introduction The corneal stroma takes 90% of the volume and mass of the cornea. It is characterized by a unique dome-shape and transparency, a result of its components, mostly collagen I and bridging molecules—proteoglycans. The complex tissue layer is 0.5–0.7 mm thick and consists of approximately 200 layers of collagen lamellae, orthogonally stacked on each other. Embedded between these layers lie the keratocytes, the most abundant cells of the stroma. The keratocytes are scarcely distributed across the stroma, concentrated in the anterior part (20,000–25,000 cells/mm2) (Fig. 1), decreasing in the middle and increasing in the posterior part again [1]. These cells remain quiescent in healthy conditions, displaying a dendritic-like morphology; however, upon activation by either disease or injury, the keratocytes may differentiate into fibroblasts or myofibroblasts [2]. The function of the latter two is to assist stromal wound healing through the secretion of matrix metalloproteinases (MMPs) and deposition of de novo extracellular matrix (ECM) components, in order to remodel the tissue [3, 4].

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Richard M. Nagymihaly et al.

Fig. 1 Corneal cross-sectional structure. Arrows mark the nuclei of keratocytes embedded between the layers of collagen in the corneal stroma (hematoxylin and eosin staining)

Cells isolated from the corneal stroma are known to develop a fibroblastic/myofibroblastic phenotype in vitro, when cultured in the presence of serum as a growth supplement [5–8]. Moreover, these cultured corneal stroma-derived cells have been attributed mesenchymal-stem cell like properties, such as trilineage differentiation potential into bone, cartilage, and adipose tissue, as well as immunomodulatory/immunosuppressive potential [9]. Early communications by research groups identified precursors of the corneal epithelium, the limbal stem cells as the multipotent stem cells of the stroma; however later, as more evidence and results were generated, it became apparent that in culture, the two cell types represent distinct cell groups, with distinct functions and potential [9, 10]. The relative ease of access to tissue and having such unique characteristics helped the corneal stroma-derived stem cells (CSSCs) attain a reputation in tissue engineering, regenerative medicine, and cellular therapy and become a popular target for corneal research [10–13]. Research groups worldwide use the explant or enzymatic method to obtain the cells from postmortem whole corneas graded unsuitable for corneal transplantation or discarded material after keratoplasties. The CSSCs are now a well-studied phenomenon, and in this chapter, we are going to discuss the means of extraction and expansion to establish primary cell cultures from human corneal postmortem tissue.

Corneal Stroma-Derived Stem Cells



Materials The equipment and solutions listed here are required to carry out the work detailed in this chapter:

2.1 Lab Equipment and Instruments

1. Cell culture Petri dish (vented, with lid) 60  15 mm, sterile. 2. Cell culture Petri dish (vented, with lid) 110  17 mm, sterile. 3. Cell strainer, 70 μm, individually packed, sterile. 4. Conjunctival scissors (curved). 5. Corneal Trephine Blade, 8 mm. 6. Corneal Trephine Blade, 18 mm. 7. CO2 incubator. 8. Cryogenic tube with cap, 1.8 mL, self-standing, sterile. 9. Electronic pipette controller and serological pipette tips (5, 10, 25 mL). 10. Filter unit, 0.2 μm, individually packed, sterile. 11. Hettich Universal 30F centrifuge (4 150  g). 12. Laminar flow cabinet. 13. Liquid nitrogen tank. 14. Nalgene 5100-0001 PC/HDPE Mr. Frosty Cryogenic Freezing Container. 15. Precisa Balance XT 120 A scale. 16. Rocking shaker, 2D rocking. 17. Sterile, individually packed disposable scalpels. 18. Sterile, individually packed 12- or 24-well non-treated cell culturing plates. 19. Sterile, individually packed 60 mL enteral syringes. 20. Sterile water, bottled 1 L. 21. Sterilized, metallic Iris forceps (preferably non-toothed). 22. Ultralow temperature (ULT) freezer (~ 80  C). 23. Upright brightfield microscope. 24. Water bath, capable of warming to 37  C. 25. 1 mL micropipette and suitable pipette tips (100–1000 μL). 26. 15 mL Centrifuge tubes, conical bottom, with screw cap, sterile. 27. 25 cm2 Cell culture flasks, cell culture treated, sterile. 28. 50 mL Centrifuge tubes, screw cap, sterile.



Richard M. Nagymihaly et al.


1. 5% Betadine povidone-iodine ophthalmic solution. 2. Dulbecco’s Modified Eagle Medium (DMEM) Low glucose with Glutamax™: 1 g/L D-glucose, 110 mg/L sodium pyruvate, phenol red, sterile. 3. Antibiotic Antimycotic Solution (100): 10,000 units/mL penicillin, 10 mg/mL streptomycin, 25 μg/mL amphotericin B. Sterile-filtered using 0.22 μm filter unit and syringe. 4. Dulbecco’s Phosphate Buffered Saline (DPBS) 1: Calcium chloride and magnesium chloride free, sterile. 5. Culture medium: DMEM, 10% (v/v%) fetal bovine serum (FBS) cell culture tested and heat inactivated, 1% (v/v%) antibiotic antimycotic solution (100) (see Note 1). Use a 0.22 μm filter unit and a 60 mL syringe to sterile-filter the culture medium aliquots in clean, sterile 50 mL tubes. Store in 50 mL aliquots at 4  C for up to a maximum of 3 weeks. 6. Collagenase solution: 0.5 mg/mL collagenase from clostridium histolyticum (type I, 0.25–1.0 FALGPA unit/mg solid, 125 CDU/mg solid) in serum-free medium. 7. Freezing medium: 10% (v/v%) dimethyl sulfoxide (DMSO), 90% (v/v%) culture medium. Always prepare fresh before using. 8. Trypan Blue, sterile-filtered. 9. FNC Coating Mix®: 10 μg/mL fibronectin, 35 μg/mL collagen type 1, 1 mg/mL bovine serum albumin (BSA), 200 μg/ mL potassium chloride, 1 μg/mL phenol red, 1.7 mg/mL D-glucose, 4.8 mg/mL HEPES buffer, 7 mg/mL sodium chloride, 1.7 mg/mL sodium phosphate.


Methods All procedures detailed here are in accordance with the directives of the Helsinki Declaration and all tissue harvesting was approved by the Regional Committee for Medical and Health Research Ethics in Norway.



Corneal tissue can be obtained from postmortem whole corneas graded unsuitable for corneal transplantation or discarded material after keratoplasties. Prior to enucleation, eyes are disinfected with 5% Betadine at the local Department of Pathology, then transferred to the ophthalmology department and processed for eye banking. Briefly, eyes are rinsed in sterile water and an 18 mm trephine is centered around the cornea and a groove is cut into the sclera and incised in a laminar flow cabinet. Conjunctival scissors are used to cut around the corneal disc. Adhering tissue from the iris is

Corneal Stroma-Derived Stem Cells


Fig. 2 Corneal button with the scleral collar, obtained after a DMEK surgery in a Petri dish. Note the suture placed in at the 12 o’clock position

removed using forceps, and then the corneal button is placed in sterile saline. A suture is usually placed in the sclera to mark the 12 o’ clock position. Following assessment in the microscope and scoring the corneal endothelium, the tissue is placed in organ culture medium until transplantation in a 31  C incubator. The leftover corneal tissue is most commonly obtained after Descemet’s membrane endothelial keratoplasty (DMEK) and are received endothelium-free by the research laboratory. When the intact corneal button is collected, the endothelium is removed from the posterior stroma using an 8 mm trephine and peeled away by forceps under a stereomicroscope. In the following section, two methods to isolate CSSCs are described: (1) an explanted tissue technique and (2) a technique using enzymatic digestion. 3.2 Isolation of CSSCs (Explant Technique)

All steps are performed in a laminar flow hood (see Note 2) 1. Remove the corneal button from the storage or transportation medium and place it in a sterile 110  17 mm Petri dish with 2–3 mL serum-free culturing medium (see Note 3) (Fig. 2). Keep both sides of the tissue hydrated. 2. It is crucial to remove the corneal epithelium before the rest of the steps are performed. A clean, sterile scalpel is the best option to use for this (see Note 5). Grab the tissue using an Iris forceps and scrape away the epithelium. Repeat the scraping from other angles. Rinse off the floating epithelium pieces with more medium, and then move the corneal button to a clean Petri dish.


Richard M. Nagymihaly et al.

Fig. 3 Images showing the removal of the sclera (a) and the leftover corneal stroma, devoid of the epithelium and endothelium (b)

Fig. 4 Pieces of stroma in a 12-well plate with culturing medium

3. Cut around and remove the scleral collar using surgical scissors (Fig. 3a). At this point the “naked” stroma is ready for cell isolation (Fig. 3b). 4. Hold the tissue on one side with the forceps and use the blade to chop the stroma into pieces of approximately 2  2 mm. Add 1–2 mL serum-free medium to the Petri dish (see Note 3) to prevent the tissue from desiccation (see Note 4). 5. Pick up the tissue pieces using forceps and place 1–2 pieces per well in a 12-well plate (see Note 6), as depicted in Fig. 4. Make sure to prevent the tissue from drying out by adding a few drops of culture medium in the wells (see Note 4). 6. When all the pieces have been placed in the wells, add 200–250 μL culture medium to the plate to barely cover the tissue explants (see Note 7), and then place the 12-well plate in the incubator. Do not disturb the plate for the next 24 h. 7. After the first day, carefully add pre-warmed culture medium up to 1–1.5 mL to the wells with the tissue pieces (see Note 4). Exercise caution not to touch or lift the attached tissue. Culture cells by refreshing the medium every 3–4 days. Observe

Corneal Stroma-Derived Stem Cells


Fig. 5 Corneal stroma explant (∗) in the top right corner of the image is shown (a), with a dense cell layer around it (b), at 4 and 10 magnification, respectively

growth of adherent elongated or spindle-shaped cells from attached tissue bits around days 6–7 under the microscope (Fig. 5) (see Note 8). 3.3 Isolation of CSSCs (Collagenase Digestion Technique)

Steps 1, 2, 3, and 6 are performed in a laminar flow hood. 1. Remove the corneal button from the organ culture bottle or other storage container and clear away the corneal endothelium and epithelium, followed by the scleral collar as already described. 2. Take a sterile 60  15 mm Petri dish and add 1 mL of the 0.5 mg/mL collagenase solution. Add the endothelium- and epithelium-free corneal stromal tissue to the dish and cut it into four quadrants using the surgical blade and the forceps. Use the scalpel to chop up the four quadrants into a mince, similarly as seen in Fig. 6. When done, pipette 4 mL of collagenase solution into the Petri dish, then cover and label accordingly. 3. Decontaminate a 2D rocker/shaker device thoroughly using 70% ethanol solution. Place the device into the incubator along with the Petri dish containing the corneal tissue in collagenase solution, as shown in Fig. 7. Turn on the rocker device and set it to gentle rocking. Close the door of the incubator and incubate overnight at 37  C (see Note 9). 4. Take out the Petri dish from the incubator. Observe the suspension/solution looking for leftover tissue pieces. If necessary, the undigested parts can be exposed to a higher collagenase concentration (see Note 10). Transfer the digest to a clean 15 mL tube, through a strainer with at least 70 μm pore size (see Note 11), as shown in Fig. 8. Include additional culture medium to neutralize enzymatic activity (see Note 4). Centrifuge the tube at 100  g for 10 min.


Richard M. Nagymihaly et al.

Fig. 6 Image showing minced corneal tissue in a plastic Petri dish

Fig. 7 2D Rocker/shaker in the 37  C incubator with the Petri dish containing the minced corneal tissue in a collagenase solution

Fig. 8 Image shows straining of the digest of stromal tissue

Corneal Stroma-Derived Stem Cells


Fig. 9 Colonies of cells obtained by digestion of the corneal stroma at 4 and 10 magnifications, respectively

5. Observe the cell pellet and aspirate the supernatant carefully, using a micropipette. Gently triturate pellet in 1 mL pre-warmed culture medium. The cells are now ready for seeding. Fill a 25 cm2 flask with 4 mL pre-warmed culture medium (see Note 6). Use a 1 mL micropipette to seed cells in the 25 cm2 culture dish and then place it in the incubator. Do not disturb the plate for the next 24 h. 6. Refresh culture medium (pre-warmed) between 24 and 48 h after seeding. Observe colonies of adherent cells under the microscope (Fig. 9). 3.4 Sub-culturing CSSCs

Steps 2, 4, and 6 are performed in a laminar flow hood. 1. Observe and estimate the confluency of cells under an upright microscope. When the cells reach 60–80% confluency (see Note 12) in the culture flask, they may be transferred to a larger container. Place aliquots of culture medium and DPBS in a 37  C water bath 20 min before commencing the transfer process (see Note 4). 2. Aspirate and discard culture medium using an electronic pipette controller (motorized pipette) and a sterile serological pipette. Add 0.2 mL pre-warmed DPBS per cm2 culture surface area (see Note 3) to the cells and wash for 10–15 s. Discard the buffer and add enough TrypLE (see Note 13) to cover the surface of the cell culture dish, usually 0.02 mL/cm2, and place the flask into the incubator at 37  C. 3. Incubate the flask for 2–3 min and observe cell detachment under the upright microscope. Tap the container gently to assist cell detachment. Add culture medium, after cells separated from the plastic surface, to neutralize the enzymatic effect.


Richard M. Nagymihaly et al.

Fig. 10 Counting CSSCs in a Bu¨rker chamber

4. Collect cell suspension using the pipette controller into a sterile, 15 mL centrifuged tube through a 70 μm cell strainer to filter out protein clots or extracellular matrix elements. Centrifuge at 100  g for 10 min at room temperature. 5. Observe the cell pellet and decant or aspirate the supernatant. Resuspend and carefully triturate cells in 1–2 mL culture medium. Determine cell number and viability in a Bu¨rker chamber and staining by trypan blue, by mixing 50 μL cell suspension with 50 μL stain (see Note 14) (Fig. 10). 6. Fill a clean cell culture dish of the desired size with the recommended volume of pre-warmed culture medium (see Note 15). Use culture medium to adjust cell density of the suspension to achieve at least 2000–5000 cells/cm2 coverage of the culturing surface area. Seed cells in the new container, close the lid, and incubate at 37  C. CSSCs take between 4–6 h to attach. Refresh medium every 3–4 days (Fig. 11). 3.5 Freezing CSSC Aliquots

All steps are performed in a laminar flow hood. 1. Observe cell density in the desired culture dish under the microscope. Upon reaching 60–80% confluency, cells may be stored frozen. 2. Repeat steps 2 and 3 described in Subheading 3.4 (Sub-culturing CSSCs) to detach the cells from the culture surface. 3. Adjust cell density to approximately 1,000,000 cells per mL medium suspension. Divide the cell suspension into aliquots that contain 500,000 to 1,000,000 cells in 15 mL centrifuge tubes (see Note 16) and centrifuge at 150  g for 5 min at room temperature. 4. Decant the supernatant and resuspend cells in 0.5 mL to 1 mL freezing medium. Make sure to triturate cell pellet thoroughly but gently. Transfer an aliquot of cells in freezing medium to a

Corneal Stroma-Derived Stem Cells


Fig. 11 Refreshing the culture medium

Fig. 12 Freezing CSSC aliquots

freezing tube (Fig. 12). Repeat the process with all aliquots of cells (see Note 17). Place all tubes in a freezing container. Transfer the freezing container to an ultra-freezer ( 86  C), for overnight cooling, as soon as possible. 5. The next day, remove the freezing tubes from the container and transfer them to a liquid nitrogen tank. Frozen cells can be safely stored in liquid nitrogen for extended time [14].


Richard M. Nagymihaly et al.

3.6 Thawing CSSC Aliquots

Steps 3 and 4 are to be performed in a laminar flow hood. 1. Pre-warm medium in a water bath for at least 20 min before thawing cells. Fill the desired cell culture flask(s) with the pre-warmed medium. 2. Remove cell aliquot(s) from liquid nitrogen and thaw by handholding tubes for 1–2 min in the water bath at 37  C. 3. Open the lid of the freezing tube and carefully triturate the cell suspension using a 1 mL micropipette. Adjust cell number to the desired density with culture medium before seeding in the flasks with the pre-warmed medium to achieve coverage of approximately 2000–5000 cells/cm2 culturing surface (see Note 18). Placed cell culture flask(s) in the incubator. 4. The next day, observe cell attachment under the microscope and refresh culture medium to remove any remaining DMSO from the culture. Cells may now be used for any type of investigation.


Notes 1. DMEM with low glucose content is the most commonly used culture medium for CSSCs; although other types of media have been used [6, 15]. Generally, 10% FBS is added to the medium; however, different research laboratories use various concentrations ranging from 2% to 20% FBS. Notably, a few studies have used animal-free medium to successfully expand the cells; however, without serum in the culture medium the cells fail to exhibit mesenchymal stem cell (MSC) like characteristics and adopt a more keratocyte-like resting state [7, 16–18]. Cells require a source of glutamine and this essential amino acid should be added to the culture medium at a 20 mM final concentration, if it has not already been incorporated in the basal medium. 2. For all steps performed in a laminar flow hood, disinfect surfaces using a 70% ethanol solution, or equivalent surface disinfectant solution. 3. Other sterile, physiological buffers and solutions will suffice, such as HBSS. 4. Make sure to warm the culture medium aliquots and other washing buffers in a 37  C water bath before applying it to the cells. 5. Alternatively, or complementary to the scraping, the corneal epithelium can be removed using digestion in dispase. 6. Coating the surface of the cell culture dish is an optional step. The authors did not observe decreased cell attachment, when

Corneal Stroma-Derived Stem Cells


Fig. 13 Epithelial contamination in a culture with explanted stromal tissue

coating was omitted. Alternatively, a coating solution, containing fibronectin and collagen I, may be used. For details, refer to the Materials section. 7. It is not necessary to cover the surface of the wells with any type of attachment agent; however, it is encouraged to add enough culturing medium to barely cover the tissue bits. More liquid will float the tissue and prevent it from attachment. This may reduce the success of establishing cultures. Supplementary medium may be added following the first 24 h. 8. When establishing explant cultures using corneal tissue from rims, occasionally cells displaying epithelial morphology may emerge in the culture wells at passage 0 (Fig. 13). Possibly, these contaminating cells originate from limbal epithelial cell niches, located peripherally in the basal epithelial layer. Wells or cultures contaminated by epithelial cells should not be used to enrich CSSC populations. 9. Alternatively, a 50 mL tube may be used, if a proper centrifuge and a necessary setup are available. 10. It is possible to use a stronger collagenase solution, up to 1.5–2 mg/mL, for a shorter incubation time (4–6 h). The rocking platform speeds up the digestion process; although when not available, overnight digestion is encouraged. Success of the digestion process may be assessed macroscopically. When leftover, undigested tissue pieces are present in the solution/ suspension, a stronger concentration of 2 mg/mL collagenase may be used for 1–2 h after the overnight incubation to completely digest the tissue and maximize cell yield. 11. The authors’ personal experience is that when isolating cells from corneal rims or from corneas stored in organ culture for extended times (>3 weeks), the cell yield is lower, while


Richard M. Nagymihaly et al.

straining the digest may cause additional cell loss and should be avoided. 12. Mesenchymal stem cell cultures should be prevented from reaching high confluency, as cells in overgrown cultures tend to transdifferentiate and lose the MSC phenotype. 13. Alternatively, trypsin can be used; however, the enzyme is known for its harshness on the cell membrane. Notably, the authors discovered no adverse effects on cell morphology, when exposure to trypsin was used to detach the cells for a maximum of 5 minutes. 14. When determining the cell count, remember to double the number, to correct for the dilution with trypan blue. Blue cells in the chamber are dead and should be excluded from the calculations. 15. Refer to the manufacturer of the culture dish. Generally, 0.2 mL medium per cm2 of culturing surface is sufficient. 16. Alternatively, the whole container (well or flask) of cells can be frozen as one aliquot; however, freezing cells in extreme high numbers is advised against. 17. Make sure that everything is prepared before suspending the cells in freezing medium. DMSO is necessary to ensure that ice crystal formation in the cells is blocked; however, the substance is toxic. In order to minimize the toxic effect, work effectively and quickly. Do not expose the cells to freezing medium at room temperature for extended periods of time (>10 min). 18. Seeding cell density may vary depending on the application or intended use.

Acknowledgements The work presented was supported by the South-Eastern Norway Regional Health Authority and by the Oslo University Hospital, Norway. References 1. Forrester J, Dick AD, McMenamin PG, Roberts F, Pearlman E (2016) Chapter 1anatomy of the eye and orbit. In: Forrester J, Dick AD, McMenamin PG, Roberts F, Pearlman E (eds) The eye, 4th edn. Elsevier, Amsterdam, Pages 1–102.e2. https://www. B9780702055546000010 2. Ljubimov AV, Saghizadeh M (2015) Progress in corneal wound healing. Prog Retin Eye Res

49:17–45. teyeres.2015.07.002 3. Wilson SE, Chaurasia SS, Medeiros FW (2007) Apoptosis in the initiation, modulation and termination of the corneal wound healing response. Exp Eye Res 85(3):305–311. 4. Kureshi AK, Funderburgh JL, Daniels JT (2014) Human corneal stromal stem cells exhibit survival capacity following isolation

Corneal Stroma-Derived Stem Cells from stored organ-culture corneas. Invest Ophthalmol Vis Sci 55(11):7583–7588. 5. Funderburgh JL, Mann MM, Funderburgh ML (2003) Keratocyte phenotype mediates proteoglycan structure: a role for fibroblasts in corneal fibrosis. J Biol Chem 278 (46):45629–45637. 1074/jbc.M303292200 6. Sidney LE, Branch MJ, Dua HS, Hopkinson A (2015) Effect of culture medium on propagation and phenotype of corneal stroma-derived stem cells. Cytotherapy 17(12):1706–1722. 7. Lynch AP, O’Sullivan F, Ahearne M (2016) The effect of growth factor supplementation on corneal stromal cell phenotype in vitro using a serum-free media. Exp Eye Res 151:26–37. 2016.07.015 8. Nagymihaly R, Vereb Z, Facsko A, Moe MC, Petrovski G (2017) Effect of isolation technique and location on the phenotype of human corneal stroma-derived cells. Stem Cells Int 2017:9275248. 1155/2017/9275248 9. Vereb Z, Poliska S, Albert R, Olstad OK, Boratko A, Csortos C, Moe MC, Facsko A, Petrovski G (2016) Role of human corneal stroma-derived mesenchymal-like stem cells in corneal immunity and wound healing. Sci Rep 6:26227. srep26227 10. Szabo DJ, Noer A, Nagymihaly R, Josifovska N, Andjelic S, Vereb Z, Facsko A, Moe MC, Petrovski G (2015) Long-term cultures of human cornea limbal explants form 3d structures ex vivo - implications for tissue engineering and clinical applications. PLoS One 10 (11):e0143053. journal.pone.0143053 11. Ahearne M, Wilson SL, Liu KK, Rauz S, El Haj AJ, Yang Y (2010) Influence of cell and collagen concentration on the cell-matrix mechanical relationship in a corneal stroma wound healing model. Exp Eye Res 91(5):584–591.


12. Szabo DJ, Nagymihaly R, Vereb Z, Josifovska N, Noer A, Liskova P, Facsko A, Moe MC, Petrovski G (2018) Ex vivo 3d human corneal stroma model for schnyder corneal dystrophy - role of autophagy in its pathogenesis and resolution. Histol Histopathol 33 (5):455–462. HH-11-928 13. Shojaati G, Khandaker I, Sylakowski K, Funderburgh ML, Du Y, Funderburgh JL (2018) Compressed collagen enhances stem cell therapy for corneal scarring. Stem Cells Transl Med 7(6):487–494. sctm.17-0258 14. Kumar A, Xu Y, Yang E, Du Y (2018) Stemness and regenerative potential of corneal stromal stem cells and their secretome after long-term storage: Implications for ocular regeneration. Invest Ophthalmol Vis Sci 59(8):3728–3738. 15. Matthyssen S, Ni Dhubhghaill S, Van Gerwen V, Zakaria N (2017) Xeno-free cultivation of mesenchymal stem cells from the corneal stroma. Invest Ophthalmol Vis Sci 58 (5):2659–2665. iovs.17-21676 16. Wilson SL, Yang Y, El Haj AJ (2014) Corneal stromal cell plasticity: In vitro regulation of cell phenotype through cell-cell interactions in a three-dimensional model. Tissue Eng Part A 20(1–2):225–238. ten.TEA.2013.0167 17. Sidney LE, Hopkinson A (2018) Corneal keratocyte transition to mesenchymal stem cell phenotype and reversal using serum-free medium supplemented with fibroblast growth factor-2, transforming growth factor-beta3 and retinoic acid. J Tissue Eng Regen Med 12(1): e203–e215. 2316 18. Foster JW, Gouveia RM, Connon CJ (2015) Low-glucose enhances keratocytecharacteristic phenotype from corneal stromal cells in serum-free conditions. Sci Rep 5:10839. srep10839

Chapter 2 In Vitro Expansion of Corneal Endothelial Cells for Transplantation Kim Santerre, Isabelle Xu, Mathieu The´riault, and Ste´phanie Proulx Abstract The corneal endothelium forms a leaky barrier between the corneal stroma and the aqueous humor of the anterior chamber. This cell monolayer maintains the corneal stroma in a state of relative dehydration, a process called deturgescence, which is required in order to obtain corneal stromal transparency. Endothelial dysfunctions lead to visual impairment that ultimately can only be treated surgically via the corneal transplantation of a functional endothelium. Shortages of corneas suitable for transplantation has motivated research toward new alternatives involving in vitro corneal endothelial cell (CEC) expansion. This chapter describes current methods that allow isolate and culture CECs. In brief, Descemet membrane is peeled out of the cornea and digested in order to obtain CECs. Cells are then seeded and cultured. Key words Cornea, Corneal endothelium, Cell culture, Cell expansion


Introduction The corneal endothelium consists of a monolayer of corneal endothelial cells (CECs), which lie on a thick basal membrane named the Descemet’s membrane. It is located at the posterior side of the cornea and faces the aqueous humor of the anterior chamber. CECs play a key role in maintaining the corneal stroma partially dehydrated, which is required for corneal transparency. This is achieved through a “pump/leak balance,” where the leaky endothelial barrier prevents the cornea from swelling by slowing the intake of liquid into the stroma and pumping it back to the anterior chamber [1]. CECs do not proliferate in vivo. This is thought to be caused by a strong contact inhibition mechanism and the presence of transforming growth factor (TGF)-β in the aqueous humor, which maintain a strong expression of the cyclin-dependent kinase inhibitor p27 within the cells [2, 3]. Upon damage to the corneal endothelium, the remaining cells will migrate and enlarge in order

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Kim Santerre et al.

to cover the damaged zone and reform the endothelial barrier [4]. Endothelial cell density in healthy adult ranges from 2500 to 3000 cells/mm2 [5]. The endothelium remains however functional down to a breaking point of 400–700 cells/mm2 [5]. A dysfunctional endothelium will result in corneal edema, serious vision impairments, and even blindness [6]. Endothelium keratoplasty (EK) is currently the only available treatment for endothelial dysfunctions or trauma. Both Descemet stripping automated endothelial keratoplasty (DSEAK) and Descemet membrane endothelial keratoplasty (DMEK) can be performed with cadaveric tissues suitable for transplantation [7]. In North America, approximately 25,000 EKs are performed each year [8]. With a growing and aging population, this number will inevitably increase the demand for the limited available corneas eligible for transplantation. New alternatives were therefore needed to address these limitations. A possible solution would be to increase the treating potential of each corneal tissue by isolating and expanding the number of cells. At present, investigated strategies include the creation of tissue-engineered corneal endothelium and the intracameral injection of a cell suspension [9, 10]. These promising alternatives to conventional corneal transplantation both require cell expansion. This chapter details the methodology and the materials used to successfully isolate CECs with the peel-and-digest method and to expand CECs in vitro, including media changes, cell passages, cell count, cryopreservation, and cell thawing.

2 2.1

Materials Labware

1. Filtration unit: The filtration unit is mounted with a 47 mm diameter and 0.22 μm filter set. 2. 15 mL and 50 mL tubes. 3. Low-binding 0.22-μm disposable filter. 4. 1.5 mL tube. Sterilize by autoclaving. 5. 60  15 mm Tissue culture dish. 6. Sterile gauze. 7. 22-scalpel blade. 8. Dissecting curved scissor. 9. Parafilm. 10. Dissecting curved forceps. 11. Dissecting straight forceps.

In Vitro Expansion of Corneal Endothelial Cells


12. Plastic culture dishes: 24-well culture plate with 1.9 cm2 surface, 12-well culture plate with 3.8 cm2 surface, 6-well culture plate with 9.5 cm2 surface, 25 cm2 culture flask. 13. Flint glass disposable Pasteur pipets. Sterilize by autoclaving. 14. Hemocytometer. 15. Glass coverslips. 16. Hand tally counter. 17. Nalgene freezing container. 18. Cryogenic vials. 19. Sterile plastic pipette: 1 mL pipette, 2 mL pipette, 5 mL pipette, 10 mL pipette, and 25 mL pipette. 2.2

Culture Media

2.2.1 Solutions

1. Phosphate buffered saline (PBS): 127 mM NaCl, 2.7 mM KCl, 6.5 mM Na2HPO4, 1.5 mM KH2PO4. Yields a 10 stock solution. Store at room temperature. Working solution is done by mixing 10 PBS with pyrogenic ultrapure water and sterilized by filtration through a filtration unit. 2. Collagenase A: Collagenase A (0.15 units/mg) is dissolved in expansion medium or sterile 1 PBS at a final concentration of 1 mg/mL (0.15 units/mL). Solution is sterilized by filtration through a 0.22 μm low-binding disposable filter. Collagenase A must be used within a day. 3. Trypsin/EDTA: Trypsin/EDTA 0.05%-0.53 mM is aliquoted in sterile 15 mL tubes and stored at 20  C (long-term) or 4  C (short-term) (see Note 1). 4. Dimethyl sulfoxide (DMSO). 5. FNC-coating mix. 6. Trypan blue solution 0.4%: Aliquot in 1.5 mL tubes. 7. HCl 0.01 N. 8. Anhydrous ethyl alcohol (ethanol) 99%.

2.2.2 Base Media

1. DMEM: Dulbecco’s Modified Eagle Minimal Essential Medium. Store at 4  C. 2. Opti-MEM: Complete with 0.20 g/L calcium chloride dehydrate and 2.4/L sodium bicarbonate. Adjust pH to 7.1. Sterilize through a filtration unit. Store at 4  C.

2.2.3 Additives

1. Epidermal growth factor (EGF): Dissolve one vial (100 μg) of EGF in 2 mL of HCl 0.01 N. Add the mix to 38 mL of expansion medium. Sterilize by filtration through a 0.22 μm low-binding disposable filter. Aliquot in 1.5 sterile tubes and store at 20  C (see Note 1). Thaw at room temperature.


Kim Santerre et al.

2. Penicillin/Streptomycin: Aliquot penicillin/streptomycin 100 in sterile 15 mL tubes and store at 20  C (long-term preservation) or 4  C (short-term preservation) (see Note 1). Aliquots are thawed at room temperature or in 37  C water bath prior to their use. 3. Chondroitin sulfate: Dissolve chondroitin sulfate A sodium salt from bovine trachea in expansion medium to yield a 1.6% solution. Sterilize through a filtration unit. Aliquot in 15 mL tubes. Store at 20  C (long-term preservation) or 4  C (short-term preservation) (see Note 1). Thaw at room temperature or in a 37  C water bath. 4. Ascorbic acid: Dissolve 2 mg/mL of L-ascorbic acid in expansion medium. Sterilize by filtration through a 0.22 μm low-binding disposable filter. Store at 4  C and use within a day. 5. Fetal bovine serum (FBS): Thaw in cold water (1–2 days). Inactivate in hot water (56  C) for 30 min. Distribute in single use aliquots (50 mL tubes) and store at 20  C (see Note 1). 2.2.4 Complete Media

1. Extraction medium: DMEM supplemented with 10% FBS and penicillin/streptomycin. Store at 4  C. 2. Expansion medium: Opti-MEM supplemented with 8% FBS, 5 ng/mL EGF, penicillin/streptomycin (working concentration: 1). Store at 4  C. 3. Complete expansion medium: Expansion medium completed with 20 μg/mL ascorbic acid and 0.08% chondroitin sulfate. Complete expansion medium must be used within a day. 4. Freezing medium: Freezing medium is composed of 10% DMSO in FBS. Freezing medium is kept on ice or stored at 4  C. Freezing medium must be used within a day. 5. Optisol-GS.


Donor Tissues

1. Human globes or corneas unsuitable for transplantation (see Note 2) are normally provided by an Eye Bank. In our work, tissues are provided by the Centre Universitaire d’Ophtalmologie (CUO) Eye Bank, Hoˆpital Saint-Sacrement, Que´bec, Canada. Globes are kept at 4  C until dissection. Corneas are kept in Optisol-GS or Extraction medium at 4  C until Descemet membrane peeling.

In Vitro Expansion of Corneal Endothelial Cells




3.1 Corneal Endothelial Cells Isolation 3.1.1 Globe Dissection

1. Surround the eye specimen with a folded sterile gauze to avoid touching it (Fig. 1a) and place it in a 100 mm sterile Petri dish. 2. Make a first opening in the sclera with a 22-scalpel blade (Fig. 1b). Use a curved scissor to carve out the cornea (Fig. 1c). 3. Remove the iris and the lens (Fig. 1d). 4. Store the cornea in extraction medium endothelial side facingup in a 60 mm Petri dish (Fig. 1e) and seal it with parafilm until Descemet peeling. Corneas must be used within a few days and kept at 4  C.

3.1.2 Descemet Membrane Peeling

1. Place the cornea with the endothelial side facing up in a 60 mm Petri dish, with enough extraction medium to cover the cornea (Fig. 1e). Under a binocular microscope (see Note 3), use dissecting curved forceps to maintain the cornea in place (Fig. 1f). 2. Make a scratch at the periphery of Descemet membrane and gently unroll strips from the stroma using dissecting straight forceps (Fig. 1g, h). 3. Add the strips to 11 mL of expansion medium and incubate at 37  C overnight.

Fig. 1 Steps from eye globe dissection to Descemet membrane peeling. (a–d) Macroscopic images showing the successive steps of corneal removal from a human eye globe; (e, f) Macroscopic images of a human cornea with the endothelium facing up; (g, h) Stereomicroscope images showing the peeling of Descemet’s membrane


Kim Santerre et al.

3.1.3 Collagenase A Digestion

CECs isolated with Collagenase A can form aggregates. To generate single cells, additional steps can be performed (see Note 4). Alternatively, isolation can be done using EDTA (see Note 5). 1. Centrifuge Descemet membrane strips at 300  g for 10 min and remove supernatant. 2. Incubate Descemet membrane strips in 1 mL of collagenase A 1 mg/mL for 2–4 h at 37  C. 3. Centrifuge digested Descemet membrane strips at 300  g 10 min and remove supernatant. 4. Dissolve cell pellet in complete expansion medium (see Note 4). 5. Seed CECs on a FNC-coated plastic culture dish (see Notes 6 and 7).


Cell Expansion

3.2.1 Medium Change

Cell morphology (fibroblast-like or endothelial morphology; Fig. 2) and confluence are assessed under a phase contrast microscope (see Note 8). Medium is changed every 2–3 days (see Note 9). 1. Warm complete expansion medium in a 37  C water bath. 2. Remove cell medium with a sterile glass pipette branched to a vacuum. 3. Add fresh warmed complete expansion medium to cells with a sterile plastic pipette. Volume added will vary depending on culture dish used.

3.2.2 CECs Passages (Subculture)

Prior to cell passages, trypsin/EDTA must be warmed in a 37  C water bath.

Fig. 2 Phase contrast images of cultured corneal endothelial cells. CECs with an endothelial, a fibroblastic, or a mixed morphology are shown. Scale bar ¼ 200 μm

In Vitro Expansion of Corneal Endothelial Cells


1. Remove medium from confluent or nearly confluent CECs (see Note 10) with a sterile glass pipette. 2. Rinse CECs with warm trypsin/EDTA or PBS. 3. Add trypsin/EDTA to cells. Monitor cell detachment every 5 min under a phase contrast microscope. 4. Add expansion medium or FBS when CECs are detached to neutralized trypsin/EDTA. Do up-and-downs to remove cells that are still present on the plastic dish. 5. Collect cells with a sterile plastic pipette and put them in a 15 mL tube. Rinse plastic dish with expansion medium and add it to the 15 mL tube. 6. Centrifuge the tube at 300  g for 10 min and remove supernatant. 7. Resuspend cells in 1 mL of warm complete expansion medium. Take an aliquot (50 μL) for cell counting (see Subheading 3.3). 8. Seed cells on plastic culture dishes at a density of 20,000 cells/ cm2. 3.3

Cell Count

This step can be performed under non-sterile conditions. Prior to its use, clean the hemocytometer and glass coverslip with ethanol 99%, rinse with water and dry (see Note 11). 1. Mix 1:1 cells with trypan blue colorant (see Note 12). 2. Fill both chambers of the hemocytometer underneath the coverslip, allowing cell suspension to be drawn out by capillary action. 3. Place the hemocytomer under a phase contrast microscope. Focus on the grid lines with a 10 objective. 4. Count cells manually with a hand tally counter from five squares for each chamber. Viable cells are unstained while dead cells are blue. 5. Using the following formula, calculate cell concentration (cells/mL). Here the dilution factor is 2 and the surface equals to 104: Cell concentration ¼

Total cell counted  Dilution factor Number of squares  Surface

6. Viability can be assessed with the formula: %viability ¼

Live cells  100 Live cells þ Dead cells



Kim Santerre et al.


Nalgene freezing container must be filled with 99% ethanol and store at 20  C prior its use. 1. Follow steps 1–7 from Subheading 3.2.2. 2. Centrifuge tubes and discard supernatant. 3. Resuspend cells at the desired concentration in freezing medium and put the tube on ice (see Note 13). 4. Aliquot in cryogenic vials on ice and place the vials in the Nalgene freezing container. 5. Store the Nalgene freezing container at 80  C overnight. 6. Transfer the cryogenic vials in liquid nitrogen for long-term preservation.


Cells Thawing

1. Place cryogenic tubes in a 37  C water bath. Do not let the cell suspension completely thaw; a small ice pellet has to remain. 2. Add the cells to 9 mL of cold expansion medium. 3. Centrifuge tubes and remove supernatant. 4. Dissolve cell pellet in 1 mL of complete expansion medium. Take an aliquot (50 μl) for cell counting and assessing viability (see Subheading 3.3). 5. Seed cells on plastic culture dish at a density of 20,000 cells/ cm2.


Notes 1. Avoid freeze-thaw cycle that could damage serum or additives. 2. Protocols need to be approved by the institution’s committee for the protection of human subjects. Young donors and short postmortem delays usually yield better CECs cultures 3. Descemet peeling must be done under sterile conditions next to a Bunsen burner to avoid bacterial contamination. Forceps are sterilized after each dissection. 4. Additional steps for CEC isolation can be performed to separate aggregates into single cells. Before seeding CECs, 1 mL of trypsin/EDTA or TrypLE reagent is added for 5–10 min [11]. Trypsin/EDTA or TrypLE are inactivated with the same amount of expansion medium. Cells are centrifuged, supernatant is removed, and cells are resuspended in complete expansion medium before being seeded on plastic culture dish. 5. CECs isolation can be performed with 0.02% EDTA solution instead of collagenase A [12]. Descemet strips are incubated in 1 mL of EDTA for 45 min followed by up-and-down pipetting using a glass pipette. Cells are centrifuged and resuspended in

In Vitro Expansion of Corneal Endothelial Cells


complete expansion medium to seed them on FNC-coated plastic culture dishes. Afterward, Descemet strips without cells attached can be removed from culture under a binocular microscope with sterile straight forceps. Alternatively, they can be removed at the next media change through aspiration with a sterile glass pipette branched to a vacuum. 6. FNC mix is added to plastic culture for 1 min at room temperature before rinsing. This allows to obtain FNC-coated culture plastic dishes. FNC-coated plastic culture has been shown to improve endothelial morphology [13]. 7. A pair of corneas (diameter of ~11 mm) allows to isolate enough cells for the seeding of one well of a 6-well plate. 8. For clinical purpose, cell quality controls should be performed on CECs at the final passage. The quality of the CECs can be established in many ways, such as using morphometric analysis and cell surface markers expression [10, 14–21]. Cell surface markers associated with an endothelial phenotype include CD56 [15], CD166 [19], CD200 [17], CD248 [15], coxsackie adenovirus receptor (CAR) [15], collagen type VIII [14, 16], N-cadherin [16], TGF-β2 [16], SLC4A11 [14], and glypican 4 [17], while endothelial with a fibroblastic-like phenotype have been shown to express CD24 [18], CD26 [18], CD44 [19], CD105 [18], CD109 [15], and others [16] 9. When CECs reach confluency, the complete expansion medium can be switched for a maturation medium for 7–28 days [22, 23]. The dual media approach has been shown to improve endothelial phenotype throughout passages. Basic maturation medium composition: Opti-MEM, 8% FBS and 1 penicillin/streptomycin. 10. Confluency is assessed by the percentage of CECs covering plastic culture. 11. Cell count can also be performed with an automated cell counter. 12. Trypan blue coloration is cytotoxic. Cell counting should be done in less than 5 min. 13. DMSO is a toxic oxidative agent at temperatures above 10  C. Working with cells in contact with a solution containing DMSO must be done quickly and on ice.


Kim Santerre et al.

Acknowledgements Procurement of eyes for research was possible thanks to a partnership with He´ma-Que´bec, the CUO Eye Bank, and a “Fonds de recherche du Que´bec – Sante´ (FRQ-S)” Vision Health Research Network (VHRN) Infrastructure Program (S.P.). I.X. was a recipient of Master Training Awards from Universite´ Laval (WilbrodBhe´rer), the Fondation du CHU de Que´bec, the LOEX Center, and the VHRN. References 1. Bonanno JA (2003) Identity and regulation of ion transport mechanisms in the corneal endothelium. Prog Retin Eye Res 22(1):69–94 2. Joyce NC, Harris DL, Mello DM (2002) Mechanisms of mitotic inhibition in corneal endothelium: Contact inhibition and tgf-beta2. Invest Ophthalmol Vis Sci 43 (7):2152–2159 3. Yoshida K, Kase S, Nakayama K, Nagahama H, Harada T, Ikeda H, Harada C, Imaki J, Ohgami K, Shiratori K, Ilieva IB, Ohno S, Nishi S, Nakayama KI (2004) Involvement of p27kip1 in the proliferation of the developing corneal endothelium. Invest Ophthalmol Vis Sci 45(7):2163–2167 4. Matsuda M, Sawa M, Edelhauser HF, Bartels SP, Neufeld AH, Kenyon KR (1985) Cellular migration and morphology in corneal endothelial wound repair. Invest Ophthalmol Vis Sci 26 (4):443–449 5. Edelhauser HF (2006) The balance between corneal transparency and edema: The proctor lecture. Invest Ophthalmol Vis Sci 47 (5):1754–1767. iovs.05-1139 6. American Academy of Opthalmology A (1997) Corneal endothelial photography - three~year revision. Ophthalmology 104(8):1360–1365 7. Talajic JC, Straiko MD, Terry MA (2013) Descemet’s stripping automated endothelial keratoplasty: then and now. Int Ophthalmol Clin 53(2):1–20. 0b013e31827eb6ba 8. EBAA (2016) 2015 eye banking statistical report. Washington, DC 9. Proulx S, Audet C, Uwamaliya J, Deschambeault A, Carrier P, Giasson CJ, Brunette I, Germain L (2009) Tissue engineering of feline corneal endothelium using a devitalized human cornea as carrier. Tissue Eng Part A 15(7):1709–1718. 10.1089/ten.tea.2008.0208

10. Kinoshita S, Koizumi N, Ueno M, Okumura N, Imai K, Tanaka H, Yamamoto Y, Nakamura T, Inatomi T, Bush J, Toda M, Hagiya M, Yokota I, Teramukai S, Sotozono C, Hamuro J (2018) Injection of cultured cells with a rock inhibitor for bullous keratopathy. N Engl J Med 378 (11):995–1003. NEJMoa1712770 11. Li W, Sabater AL, Chen YT, Hayashida Y, Chen SY, He H, Tseng SC (2007) A novel method of isolation, preservation, and expansion of human corneal endothelial cells. Invest Ophthalmol Vis Sci 48(2):614–620. https:// 12. Senoo T, Obara Y, Joyce NC (2000) Edta: a promoter of proliferation in human corneal endothelium. Invest Ophthalmol Vis Sci 41 (10):2930–2935 13. Engler C, Kelliher C, Speck CL, Jun AS (2009) Assessment of attachment factors for primary cultured human corneal endothelial cells. Cornea 28(9):1050–1054. 1097/ICO.0b013e3181a165a3 14. Chng Z, Peh GS, Herath WB, Cheng TY, Ang HP, Toh KP, Robson P, Mehta JS, Colman A (2013) High throughput gene expression analysis identifies reliable expression markers of human corneal endothelial cells. PLoS One 8 (7):e67546. nal.pone.0067546 15. Bartakova A, Alvarez-Delfin K, Weisman AD, Salero E, Raffa GA, Merkhofer RM Jr, Kunzevitzky NJ, Goldberg JL (2016) Novel identity and functional markers for human corneal endothelial cells. Invest Ophthalmol Vis Sci 57(6):2749–2762. iovs.15-18826 16. Ueno M, Asada K, Toda M, SchlotzerSchrehardt U, Nagata K, Montoya M, Sotozono C, Kinoshita S, Hamuro J (2016) Gene signature-based development of elisa assays for reproducible qualification of cultured

In Vitro Expansion of Corneal Endothelial Cells human corneal endothelial cells. Invest Ophthalmol Vis Sci 57(10):4295–4305. 17. Cheong YK, Ngoh ZX, Peh GS, Ang HP, Seah XY, Chng Z, Colman A, Mehta JS, Sun W (2013) Identification of cell surface markers glypican-4 and cd200 that differentiate human corneal endothelium from stromal fibroblasts. Invest Ophthalmol Vis Sci 54 (7):4538–4547. iovs.13-11754 18. Hamuro J, Toda M, Asada K, Hiraga A, Schlotzer-Schrehardt U, Montoya M, Sotozono C, Ueno M, Kinoshita S (2016) Cell homogeneity indispensable for regenerative medicine by cultured human corneal endothelial cells. Invest Ophthalmol Vis Sci 57 (11):4749–4761. iovs.16-19770 19. Okumura N, Hirano H, Numata R, Nakahara M, Ueno M, Hamuro J, Kinoshita S, Koizumi N (2014) Cell surface markers of functional phenotypic corneal endothelial cells. Invest Ophthalmol Vis Sci 55(11):7610–7618. 1167/iovs.14-14980


20. Okumura N, Ishida N, Kakutani K, Hongo A, Hiwa S, Hiroyasu T, Koizumi N (2017) Development of cell analysis software for cultivated corneal endothelial cells. Cornea 36 (11):1387–1394. ICO.0000000000001317 21. Toda M, Ueno M, Hiraga A, Asada K, Montoya M, Sotozono C, Kinoshita S, Hamuro J (2017) Production of homogeneous cultured human corneal endothelial cells indispensable for innovative cell therapy. Invest Ophthalmol Vis Sci 58(4):2011–2020. 22. Peh GS, Chng Z, Ang HP, Cheng TY, Adnan K, Seah XY, George BL, Toh KP, Tan DT, Yam GH, Colman A, Mehta JS (2015) Propagation of human corneal endothelial cells: a novel dual media approach. Cell Transplant 24(2):287–304. 3727/096368913X675719 23. Beaulieu Leclerc V, Roy O, Santerre K, Proulx S (2018) Tgf-beta1 promotes cell barrier function upon maturation of corneal endothelial cells. Sci Rep 8(1):4438. 1038/s41598-018-22821-9

Chapter 3 Primary Culture of Cornea-Limbal Epithelial Cells In Vitro Finbarr O’Sullivan Abstract The cultivation of corneal-limbal cells in vitro represents an excellent means to generate models to study cornea function and disease processes. These in vitro expanded cornea-limbal epithelial cell cultures are rich in stem cells for cornea, and hence can be used as a cell therapy for cornea-limbal deficiency. This chapter details the primary culture of these cornea-limbal cells, which can be used as model for further studies of the cornea surface. Key words Cornea, Limbal, Primary culture, Feeder cells, Human amniotic membrane, Explants


Introduction The cornea epithelium accounts for approximately 10% of the total cornea thickness. It rests on a basement membrane or basal lamina, in close contact with Bowman’s membrane, separating it from the mesenchymal stroma. The cornea epithelium is composed of three groups of cells, a superficial squamous cells with a flat polygonal morphology of about 2–3 cells in thickness, an intermediate cell layer of 2–3 cells, called wing cells, and a final basal layer of epithelium with a cuboidal/columnar morphology. At the border between the transparent cornea and the opaque sclera is the limbus. This region was first proposed in the early 1970s as having a role in cornea renewal. The epithelium of the limbus consists of multilayered cells rich in minute vessels with the presence of Langerhans and melanocyte cells (which produce and secrete pigments) and a gradient of epithelial cells that includes progenitors of the corneal epithelium. The “X, Y, Z” hypothesis of corneal epithelial maintenance, developed by Thoft and Friend, proposes that corneal epithelial stem cells are located in the limbus [1]. It is these cells that are responsible for producing transient amplifying cells that proliferate and migrate toward the center of the cornea in a centripetal manner, replacing epithelial cells lost from the surface of the cornea. As this region has a rich stem cells population that differentiate

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Finbarr O’Sullivan

into cornea epithelia cells and has a greater proliferative potential when compared to central cornea, it is this area of the cornea that is selected preferentially for the generation of in vitro cultures. Limbal-corneal cultures can be generated either by the outgrowth from tissue explants or the dissociation of the tissue by enzymatic means. For both methods contradictory reports exist as to which method is superior [2, 3]. In truth both methods have their advantages and disadvantages and some researchers have found only minimal differences in culture outcomes. Explant cultures involve plating a piece of tissue approximately 1 mm by 1 mm in size on a tissue culture surface and allowing cell outgrowth to occur. As no harsh enzymes are being used, cell damage should be minimal, improving the chances of viable cell outgrowth. In addition to avoiding cell loss during tissue processing the explant technique maintains the stem cell niche of limbal tissue. This niche effect is reflected in the culture obtained itself as studies have shown a decrease in stem cell markers in the cells the greater the distance from the explant [4]. The alternative method to explant cultures is to dissociate the cells from tissue by enzymatic means to generate a suspension culture. One of the main advantages this method has is that it allows the purification of a homogeneous cell population. The common enzymes used in cell dissociation methods for limbal corneal epithelial cultures are dispase and trypsin [2, 5, 6]. Dispase is considered the more gentle of the two enzymatic methods and this is reflected in the long incubation times, that range from 1 h to 18 h [5, 7]. However, prolonged incubation with dispase can still result in cellular damage [8]. In addition to these two methods for initiating a primary cell culture of limbal-cornea epithelial cells, a myriad of cultivation methods exists. These methods range of different matrices, feeder cell systems to advanced carrier systems and serum-free media that seek to replicate the stem cell niche. Of the various matrices used for limbal-cornea epithelial cell cultures, human amniotic membrane (HAM) has probably proved to be the most popular. The reasons for this popularity are due in part to its long history in ophthalmic surgery, having low or no immunogenicity and it also is a convenient carrier to use clinically for expanded cell therapy [9]. In addition to low immunogenicity, a number of growth factors associated with epithelial growth and proliferation have been found on HAM [10]. Perhaps even more importantly HAM has been shown as being a good match for the basement membrane where limbal-corneal epithelium resides, with a similar constitution of collagens type IV and VII, fibronectin, and laminins such as α3, β1, and β2 [11, 12]. Furthermore, there is some evidence the ratio of components and the bio-mechanical properties of HAM are a close match to that found in the limbal niche [13]. However, HAM does require the use of donor tissue, so supply can be scarce and

Primary Culture of Cornea-Limbal Epithelial Cells


even when tissue is sourced from accredited tissue banks there is always a small concern regarding viral disease transmission. The in vitro culture of some cell types, particularly those more fastidious cells such as stem cells, often requires the co-culture of reproductively inactive cells known as feeder cells. The first reported use of feeder cells in cell culture dates back to 1956 by Puck and colleagues [14]. Feeder cells help by simulating the cell microenvironment by secreting extracellular metabolites that maintain the stem cells. This microenvironment is a combination of different mechanisms that have not been fully defined. Theses mechanisms are believed to include cell-to-cell and cell-to-extracellular matrix (ECM) interaction [15] in addition to the production of soluble factors promoting growth and inhibit apoptosis [16, 17]. The positive impact of fibroblast feeder cells, such as the NIH 3T3 cell line or limbal-cornea fibroblast cells, on the longterm survival and propagation of limbal-cornea epithelial cells has been demonstrated in numerous studies [17–19]. One of the most common media for culturing the cells used is based on the original work of Rheinwald and Green in the 1970s for keratinocyte cell culture. This consists of a medium of minimal essential medium (MEM) and Ham’s F-12 (3: 1) supplement with hydrocortisone, insulin, triiodothyronine, adenine, cholera toxin, and EGF [20]. Each of these supplements has a positive role to play in the promotion of cellular growth and the prevention of differentiation [20–22]. As this media contains the undefined component of serum and in order to improve the defined nature and performance of the media developed by Rheinwald and Green, various serum-free medias have been developed. Over the last 30 years various studies have been conducted into the development of a defined serum-free media to successfully culture limbal-cornea epithelial cells in vitro [17]. These have included the addition of such additives as bovine pituitary extract to B-27 serum-free supplement growth-factor all with the aim to eliminate the requirement of serum [23–25]. Commercial serum-free media that have been refined for the cultivation of keratinocytes such as EpiLife are now available [26].



2.1 Preparation of Media Supplements

1. Recombinant epidermal growth factor (rhEGF) stock solution: 100 μg/mL in 0.1% (w/v) bovine serum albumin (BSA), 10 mM acetic acid solution. Prepare a 10 mM acetic acid solution from a stock of glacial acetic acid by slowly adding 57 μL to 25 mL of deionized water and adjust the final volume to 100 mL with deionized water using a graduated cylinder. Add 5 mg of BSA slowly to 5 mL of 10 mM acetic acid. In a laminar flow cabinet, observing aseptic techniques use a syringe


Finbarr O’Sullivan

fitted with a 0.22 μm filter to sterilize the solution. Add 200 μg of rhEGF powder to 2 mL of this sterile solution to yield a stock solution with 100 μg/mL rhEGF. Dispense the solution as 10 μL aliquots into sterile micro-centrifuge tubes and store at 20  C, avoiding repeated freeze thaw cycles. 2. Hydrocortisone Stock Solution: 0.83 μM in basal media. In a laminar flow cabinet, observing aseptic technique, use a syringe fitted with a 0.22 μm filter to sterilize 5 mL of 95% (v/v) ethanol. Add 250 μL of the sterilized ethanol to 1 mg of hydrocortisone (sterile). When dissolved, add 250 μL of hydrocortisone solution to 2.25 mL of basal media. Dispense the 8.3 μM stock of hydrocortisone as 100 μL aliquots into sterile micro-centrifuge tubes and store at 20  C; avoid repeated freeze thaw cycles. 3. Triiodothyronine Stock Solution: 2 μM in PBS-A. In a laminar flow cabinet, observing aseptic technique dissolve 1 mg of triiodothyronine in 1 mL of sterile PBS-A to generate a 1.5 mM solution. Add 26.6 μL of 1.5 mM triiodothyronine solution to 19.973 mL of sterile PBS-A to yield a stock solution of 2 μM triiodothyronine. Using good aseptic technique, use a syringe fitted with a 0.22 μm filter to sterilize the triiodothyronine stock solution and dispense as 100 μL aliquots into sterile micro-centrifuge tubes and store at 20  C; avoid repeated freeze thaw cycles. 4. Cholera Toxin A Subunit Stock Solution: 1 mg/mL in sterile ultrapure water (UPW). Using good aseptic technique, in a laminar flow add 1 mL of sterile UHP water to 1 mg of cholera toxin A subunit (sterile). Dispense as 100 μL aliquots in sterile micro-centrifuge tubes and store at +4  C. Do not freeze solution. 2.2

Culture Media

1. Corneal Epithelial Growth Media: 67.5% (v/v) 1 Dulbecco’s modified Eagle media (DMEM), 22.5% (v/v) 1 Ham’s F-12 nutrient media, 10% (v/v) fetal calf serum (FCS), 10 ng/mL rhEGF, 0.4 μg/mL hydrocortisone solution, 2  109 M triiodothyronine solution, 100 ng/mL cholera toxin A subunit solution. Store corneal epithelial growth media at 4  C for up to 2 weeks. 2. NIH 3T3 Growth Media: 95% (v/v) DMEM, 5% (v/v) fetal calf serum. Store NIH 3T3 growth media at 4  C for up to 3 weeks.


Other solutions

1. PHEM Buffer: 60 mM 1,4-piperazinediethanesulfonic acid (PIPES), 25 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 10 mM egtazic acid (EGTA), 2 mM MgCl2. To a glass beaker add 9.072 g of PIPES, 13.254 g of HEPES,

Primary Culture of Cornea-Limbal Epithelial Cells


11.902 g of EGTA, 0.203 g of MgCl2. Add 400 mL of UHP water to dissolve. Adjust the pH to 6.9 with 10 M KOH (approximately 7 mL). Transfer to a graduated cylinder and make to the 500 mL. This can be kept at room temperature for 3 weeks. 2. HANKS Buffered Saline Solution (HBSS). 3. Trypsin-EDTA solution. 4. Phosphate-buffered saline A (PBS-A). 2.4 Immunofluorescence Reagents

1. Methanol. 2. Normal goat serum. 3. Blocking buffer: 5% (v/v) normal goat serum in PBS-A. 4. Tween-20: 0.25% (v/v) in PHEM. 5. Mouse anti-cytokeratin 3. 6. Rabbit anti-cytokeratin 12. 7. Mouse anti-ABC-G2. 8. Rabbit anti-ΔNP63. 9. Goat anti-mouse IgG Alexa Fluor 488. 10. Goat anti-rabbit IgG Alexa Fluor 555. 11. DAPI (40 , 6- diamidino-2-phenylindole).



3.1 Option A: Irradiation of NIH 3T3 Cells

1. Culture NIH 3T3 fibroblast cells in DMEM supplemented with 5% (v/v) fetal calf serum, until 60–70% confluence is reached. 2. Trypsinize the cells and wash in HBSS. Resuspend the NIH 3T3 cells in HBSS and place vial in gamma irradiator (see Note 1). Set the gamma irradiator for 60 Gray (see Note 2). 3. When irradiation is complete, centrifuge the cells at 100  g for 5 min. 4. Resuspend the cells in growth media, perform a cell count, plate the irradiated NIH-3T3 cells at a density of 2.4  104 cells/cm2, and allow 24 h for attachment to occur (see Note 3).

3.2 Option B: Preparation of Denuded (HAM)

1. Thaw overnight at 4  C cyro-preserved human amniotic membrane (HAM). Carefully remove the HAM from its freezing medium and carefully peel from the nitrocellulose backing paper with sterile forceps on a sterile 90 mm dish. 2. Stretch the HAM out on the dish, make sure to place the epithelial side face up, and add 20 mL of 0.25% trypsin-


Finbarr O’Sullivan

EDTA solution. Cover the dish and place in a 37  C incubator for 15–20 min. 3. Following the 15–20 min incubation gently scrubbed the HAM surface with a cell scraper to remove the amniotic epithelium (see Note 4). 4. Remove the trypsin solution with the detached cells. Wash the HAM twice with PBS-A to remove scraped cells. 5. Carefully turn the membrane over to the endothelial side and scrape the HAM with a cell scraper to remove mucus. Wash twice with PBS-A. The resulting membrane is now termed as denuded human amniotic membrane (dHAM). 6. Depending on the size available, trim the dHAM to pieces of 2–2.5 cm2 in area. Wrap the dHAM pieces around a 1.5 cm2 glass slide that has a thickness of approximately 2 mm. Ensure that the HAM is tucked in under the glass slide (see Note 5). Place in a well of a 6-well plate or a 35 mm2 dish and slowly add growth media to just cover the dHAM. The dish is now ready to receive limbal explants. 3.3 Generation of Limbal-Cornea Explants

1. In a sterile 90 mm dish place the cornea-scleral tissue and rinse with growth media to remove any loose material. 2. Using a scalpel cut the cornea-sclera into four quadrants and remove any excess tissue such as sclera, iris, cornea to isolate the cornea-scleral rim. 3. Locate the limbal ring by identifying a pale pink-brown ring in the endothelial side of the cornea-scleral rim. With the scale blade gently scrape off the endothelial layer. 4. Place a sterile glass microscope slide into a 90 mm dish; this will act as a hard surface for cutting the ring out on. Cut away all excess sclera and cornea to isolate the limbal ring. 5. Using a firm chopping motion cut the limbal ring into 1  1 mm2 explants. Place the 1  1 mm2 explants epithelial side down in the center of well. The well will either be option A pre-seeded irradiated NIH-3T3 feeder cells or option B dHAM. 6. Place approximately 6–7 explants in a ring centripetally in the center to the irradiated NIH-3T3 feeder cells or dHAM. Media level should be low so that the growth surface is moist, but the explants should not be floating. 7. Cover the cell culture dish and carefully place in a 37  C CO2 incubator. Allow the explants to settle for 1–2 h before removing from the incubator and very slowly and carefully adding 1 mL of growth media.

Primary Culture of Cornea-Limbal Epithelial Cells


8. The following day refeed explant cultures with growth media. Examine the explant culture for adhesion and cell outgrowth. Outgrowth should become visible by day 2–5 as a sheet of cobble-like cells growing from the explant. 9. Refeed the culture on alternative days until day 15–20; at this stage the culture should be confluent. 3.4 Confirmation of Limbal-Cornea Stem Cells

To confirm the identity of the cells that have grown from the explant conduct immunofluorescence for specific markers of limbal-cornea cells. 1. Wash the cell culture sample gently for 5 min, three times, with PBS-A to remove traces of growth media and any debris. 2. Fix the cell culture sample using ice-cold methanol for 10 min. 3. Block the cell culture sample with the blocking buffer for 1 h at room temperature. 4. Remove the blocking buffer and wash the cell culture sample once with PHEM buffer for 5 min and once with 0.25% (v/v) Tween-20 in PHEM buffer for 5 min. 5. Remove the buffer and add primary antibody diluted appropriately in 1% (v/v) normal goat serum in PHEM buffer. Primary antibodies considered suitable are mouse anti-cytokeratin 3, rabbit anti-cytokeratin 12, mouse anti-ABC-G2, and rabbit anti-ΔNP63. Incubate the cell culture sample and primary antibody overnight at 4  C. 6. Remove the primary antibody the following morning and wash the cell culture sample for 5 min, three times, with 0.25% (v/v) Tween-20 in PHEM buffer. 7. Add the appropriate secondary antibody, e.g., Goat anti-mouse IgG Alexa Fluor 488 or Goat anti-rabbit IgG Alexa Fluor 555, diluted in 1% (v/v) normal goat serum in PHEM buffer. Incubate at room temperature for 1 h in the dark (see Note 6). 8. Remove the secondary antibody solutions and wash the cell culture sample for 5 min, three times, with 0.25% (v/v) Tween20 in PHEM buffer. 9. Counterstain the nuclei with DAPI for 2 min, then wash twice PHEM Buffer and mount the cell culture sample with an antifade mountant and coverslip the sample. 10. Limbal-cornea epithelial stem cells should show bright nuclear staining for ΔNP63, ABC-G2, while differentiating limbalcornea epithelial cells will show weak staining for these markers and will be strongly positive for cytokeratin 3 and 12 (see Note 7).



Finbarr O’Sullivan

Notes 1. Avoid using media with phenol red during the irradiation process as free radical can be generated that will lead to cell death. 2. Batches of irradiated NIH-3 T3 cells can be prepared and stored in liquid nitrogen to avoid having to irradiate fresh cells each time. Cells can be cryopreserved using a 10% DMSO—90% FCS Solution. DMSO should be added slowly to cells to avoid shock and cells cooled at 1  C/min. 3. Mitomycin C is an alternative means to generate feeder cells using a concentration of 10 μg/mL for 2–4 h. Care must be taken to ensure no Mitomycin C is present after preparation of the feeder cells, as it can negatively impact on cell growth. 4. When scrapping the surface of the HAM care must be taken not to press too hard as this might tear the underlying basement membrane. This method removes of 90–100% of epithelium from HAM. 5. It is important to use a glass slide with sufficient weight under which the amniotic membrane is tucked. Use of a light coverslip will not hold the amniotic membrane in place during the duration of the cell culture. 6. The fluorescent secondary antibodies are light sensitive and should be prepared out of direct light in dim conditions. Once the secondary has been added, all incubation steps should be in the dark to avoid “quenching” of the fluorescent signal. 7. Although a variety of putative limbal-cornea stem cell markers have been proposed, the expression of no maker is truly definitive. The general consensus is that the co-expression of a number of makers represents the best means of identifying cells as be of being limbal-corneal epithelium.

Acknowledgements This work is supported by Science Foundation Ireland cofunded by ERDF, grant no 12/RC/2275_P2. References 1. Thoft RA, Friend J (1983) The X, Y, Z hypothesis of corneal epithelial maintenance. Invest Ophthalmol Vis Sci 24:1442–1443 2. Koizumi N, Cooper LJ, Fullwood NJ et al (2002) An evaluation of cultivated corneal limbal epithelial cells, using cell-suspension culture. Investig Ophthalmol Vis Sci 43:2114–2121

3. Ghoubay-Benallaoua D, Basli E, Goldschmidt P et al (2011) Human epithelial cell cultures from superficial limbal explants. Mol Vis 17:341–354 4. Kolli S, Lako M, Figueiredo F et al (2008) Loss of corneal epithelial stem cell properties in outgrowths from human limbal explants cultured on intact amniotic membrane. Regen Med

Primary Culture of Cornea-Limbal Epithelial Cells 3:329–342. 17460751.3.3.329 5. Liu S, Li J, Wang C et al (2006) Human limbal progenitor cell characteristics are maintained in tissue culture. Ann Acad Med Singap 35:80–86 6. Pellegrini G, Traverso CE, Franzi AT et al (1997) Long-term restoration of damaged corneal surfaces with autologous cultivated corneal epithelium. Lancet 349:990–993. 11188-0 7. Espana EM, Romano AC, Kawakita T et al (2003) Novel enzymatic isolation of an entire viable human limbal epithelial sheet. Investig Ophthalmol Vis Sci 44:4275–4281. https:// 8. Spurr SJ, Gipson IK (1985) Isolation of corneal epithelium with Dispase II or EDTA. Effects on the basement membrane zone. Invest Ophthalmol Vis Sci 26:818–827 9. Kubo M, Sonoda Y, Muramatsu R, Usui M (2001) Immunogenicity of human amniotic membrane in experimental xenotransplantation. Invest Ophthalmol Vis Sci 42:1539–1546 10. Koizumi N, Inatomi T, Sotozono C, et al (2000) Growth factors in amniotic membrane Growth factor mRNA and protein in preserved human amniotic membrane. Curr Eye Res. 20(3) : 173–177 11. Moriyama T, Asahina I, Ishii M et al (1994) Basement membrane assembly and differentiation of cultured corneal cells: importance of culture environment and endothelial cell interaction. Exp Cell Res 4:415–427 12. Fukuda K, Chikama T, Nakamura M, Nishida T (1999) Differential distribution of subchains of the basement membrane components type IV collagen and laminin among the amniotic membrane, cornea, and conjunctiva. Cornea 18:73–79. 00003226-199901000-00013 13. Chen B, Jones RR, Mi S et al (2012) The mechanical properties of amniotic membrane influence its effect as a biomaterial for ocular surface repair. Soft Matter 8:8379. https://doi. org/10.1039/c2sm26175h 14. Puck TT, Marcus PI, CieciuraI SJ (1956) Clonal growth of mammalian cells in vitro; growth characteristics of colonies from single HeLa cells with and without a feeder layer. J Exp Med 103:273–283. 1084/jem.103.2.273 15. Ehmann UK, Stevenson MA, Calderwood SK, DeVries JT (1998) Physical connections between feeder cells and recipient normal mammary epithelial cells. Exp Cell Res 243:76–86. 1998.4054


16. Burroughs J, Gupta P, Blazar BR, Verfaillie CM (1994) Diffusible factors from the murine cell line M2-10B4 support human in vitro hematopoiesis. Exp Hematol 22:1095–1101 17. Tseng SC, Kruse FE, Merritt J, Li DQ (1996) Comparison between serum-free and fibroblast-cocultured single-cell clonal culture systems: evidence showing that epithelial antiapoptotic activity is present in 3T3 fibroblastconditioned media. Curr Eye Res 15:973–984. 02713689609017643 18. Kim MK, Lee JL, Oh JY et al (2008) Efficient cultivation conditions for human limbal epithelial cells. J Korean Med Sci 23:864. https:// 19. Pellegrini G, Golisano O, Paterna P et al (1999) Location and clonal analysis of stem cells and their differentiated progeny in the human ocular surface. J Cell Biol 145:769–782. 145.4.769 20. Rheinwald JG, Green H (1975) Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell 6:331–343. https://doi. org/10.1016/s0092-8674(75)80001-8 21. Sun TT, Green H (1977) Cultured epithelial cells of cornea, conjunctiva and skin: absence of marked intrinsic divergence of their differentiated states. Nature 269:489–493. https:// 22. Yaeger PC, Stiles CD, Rollins BJ (1991) Human keratinocyte growth-promoting activity on the surface of fibroblasts. J Cell Physiol 149:110–116. 1041490114 23. Yokoo S, Yamagami S, Yanagi Y et al (2005) Human corneal endothelial cell precursors isolated by sphere-forming assay. Invest Ophthalmol Vis Sci 46:1626–1631. https:// 24. Brejchova K, Trosan P, Studeny P et al (2018) Characterization and comparison of human limbal explant cultures grown under defined and xeno-free conditions. Exp Eye Res 176:20–28. 2018.06.019 25. Zamudio A, Wang Z, Chung S-H, Wolosin JM (2016) Inhibition of TGFβ cell signaling for limbal explant culture in serumless, defined xeno-free conditions. Exp Eye Res 145:48–57. 2015.10.021 26. Lekhanont K, Choubtum L, Chuck RS et al (2009) A serum- and feeder-free technique of culturing human corneal epithelial stem cells on amniotic membrane. Mol Vis 15:1294–1302

Chapter 4 Optimization of Human Limbal Stem Cell Culture by Replating a Single Limbal Explant Marina Lo´pez-Paniagua, Teresa Nieto-Miguel, Sara Galindo, Laura Garcı´a-Posadas, Ana de la Mata, Rosa M. Corrales, Margarita Calonge, and Yolanda Diebold Abstract Cultured limbal epithelial stem cell transplantation is a clinical procedure used to regenerate the corneal epithelium in patients with limbal stem cell deficiency. The protocols used to expand limbal epithelial cells in vitro need to be optimized, since the scarcity of human ocular tissue donors is limiting the potential use of this procedure. Here, we describe a method to consecutively expand a single human limbal explant. With this method it is possible to obtain up to three limbal epithelial primary cultures from the same explant, thus increasing the efficiency of the in vitro cell culture. Key words Cornea, Corneal epithelium, Corneoscleral samples, Limbus, Limbal explant, Limbal stem cells, Limbal primary cultures, In vitro expansion


Introduction Limbal epithelial stem cells (LESCs) are responsible for corneal epithelium renewal [1, 2]. Corneal epithelium integrity is paramount for maintaining the corneal transparency required for optimal vision. However, several diseases, as well as chemical or physical insults, can compromise the integrity of the corneal epithelium, causing wounds and/or opacities. In many of these cases, corneal transplantation is a solution that may be used to restore sight. However, in the case of the destruction or dysfunction of LESCs or their niche, the outcome of corneal transplantation is usually poor [3]. In these cases, it is necessary to replace the missing stem cells to regenerate the corneal epithelium. To accomplish this, clinicians carry out cultured limbal epithelial transplantation (CLET). This therapy was first developed by Pellegrini et al. in 1997 [4] and has been reproduced in various clinical trials [5– 11]. Since a limiting factor of this method is obtaining of enough

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Marina Lo´pez-Paniagua et al.

LESCs, several protocols have been proposed to optimize this procedure in an attempt to increase its performance [12–14]. In this protocol we propose a method based on the consecutive expansion of a single human limbal explant obtained from cadaveric donors. Using this protocol, it is possible to expand the same explant up to three times, allowing clinicians to obtain larger numbers of limbal epithelial cells that retain stem cell properties [15, 16]. These cultured LESCs can be used for either research or clinical purposes.


Materials All the solution preparations should be performed in sterile conditions in a laminar flow hood irradiated with ultraviolet light and cleaned with ethanol 70% (v/v) in water. Surgical instruments should be sterilized by steam sterilization (autoclaved) prior to use. Complete culture media should be kept at 4  C and pre-warmed to 37  C in a thermostatic bath prior to use.

2.1 Medium and Solutions

1. Shipping and culture medium (see Notes 1 and 2): Dulbecco’s modified Eagle medium/Ham’s F12 solution (DMEM/F12) (1:1), 2.5 ng/mL epidermal growth factor (EGF), 10 μg/mL insulin, 5.5 μg/mL transferrin, 5 ng/mL sodium-selenite, 0.01 μg/mL hydrocortisone, 0.5% dimethyl sulfoxide (DMSO), 132.5 ng/mL cholera toxin, 50 μg/mL gentamicin, 2.5 μg/mL amphotericin B, and 5% fetal bovine serum (FBS). 2. Wash solution: Hanks Balanced Salt Solution (HBSS), 50 μg/ mL gentamicin, and 2.5 μg/mL amphotericin B. 3. Biosafe culture medium (see Note 3): DMEM/F12 (1:1), 2.5 ng/mL EGF, 10 μg/mL insulin, 5.5 μg/mL transferrin, 5 ng/mL sodium-selenite, 0.5 μg/mL hydrocortisone, 1 μM isoproterenol, 0.18 mM adenine, 2 nM triiodotironina, 50 μg/ mL gentamicin, 2.5 μg/mL amphotericin B, and 10% human serum (HS). 4. Fetal bovine serum (FBS).


Plastic and Glass

1. Plastic or glass sterile container for the shipment of corneoscleral tissues. 2. 35 mm Petri dishes. 3. 5 mL disposable serological pipettes. 4. Micropipette tips (100–1000 μL and 20–200 μL). 5. Tissue culture 12-well polystyrene plates (3.8 cm2/well).

Consecutive Culture of a Human Limbal Explant

2.3 Surgical Instruments


1. Fine-tipped tweezers. 2. Blunt-tipped tweezers. 3. Scissors. 4. Crescent knife. 5. 7.5 mm trephine. 6. Scalpel. 7. Surgical compass.



1. Laminar flow hood. 2. Autoclave sterilizer. 3. Thermostatic bath. 4. 100–1000 μL micropipette. 5. 20–200 μL micropipette. 6. Cell culture CO2 incubator. 7. Inverted phase contrast microscope.


Methods All the steps described in this section must be carried out in a laminar-flow hood previously exposed to ultraviolet radiation and cleaned with ethanol 70% (v/v) in water to maintain sterile conditions.

3.1 Preservation and Shipment of Cadaveric Human Corneoscleral Tissues

Human corneoscleral samples can be collected in a different geographic location from which they are processed. In such cases, samples should be stored and shipped under specific conditions until their processing: 1. Fill a sterile container with shipping medium (see Notes 4 and 5). 2. Place the cadaveric human corneoscleral tissue (see Note 6) into the container filled with shipping medium and send it to the laboratory where tissues are going to be processed. The shipment temperature should be maintained at 4  C (see Note 7).

3.2 Limbal Explant Preparation

1. Remove the corneoscleral tissue from the container used for the shipment. Remove the sample from the scleral area using blunt-tipped tweezers in order to avoid damaging the limbal region. 2. Place the corneoscleral tissue (Fig. 1a) in a 35 mm Petri dish and add 4 mL of wash solution (see Note 8 and Fig. 1b).


Marina Lo´pez-Paniagua et al.

Fig. 1 Limbal explant isolation. Corneoscleral tissues (a) are rinsed with wash solution (b). Excess of conjunctiva, iris, and corneal endothelium are removed (c and d). Subsequently, central cornea is removed from the corneoscleral button using a trephine (e–i). Finally, a corneoscleral ring is cut into 1–3 mm2 limbal explants (j and k), which are plated into 12-well polystyrene plates (l)

3. Remove the wash solution and add other 4 mL of the same solution to perform a second wash. Repeat this step twice for a total of three washes. 4. Excise the conjunctival tissue from the corneoscleral sample using fine-tipped tweezers and scissors. To do this, grab a small portion of the conjunctival tissue with the tweezers and cut it using scissors. Repeat this procedure until all the conjunctival tissue is removed (Fig. 1c). 5. Remove both the wash solution and the remnants of conjunctiva from the Petri dish using a 5 mL pipette (see Note 9). 6. Add 4 mL of wash solution to the Petri dish and turn over the corneoscleral sample so that the corneal endothelium and the iris face upward. Take a sample from the scleral area using blunt-tipped tweezers to avoid damaging the limbal region. 7. Remove the iris and the corneal endothelium. Hold the corneoscleral sample using the fine-tipped tweezers and scratch the superior surface of the sample with a crescent knife (see Note 10 and Fig. 1d). 8. Repeat step 5 to remove remnants of the iris and corneal endothelium from the Petri dish.

Consecutive Culture of a Human Limbal Explant


9. Add 4 mL of wash solution to the Petri dish, turn over the corneoscleral sample, and excise the scleral tissue until only 1–2 mm of sclera remain around the limbal ring. To perform this procedure, take different small pieces of sclera with finetipped tweezers and remove them using scissors. 10. Repeat step 5 to remove scleral debris and add 4 mL of wash solution to the Petri dish. 11. Excise the central corneal button from the corneoscleral sample using a trephine that is 7.5 mm in diameter (see Note 11). Place the corneoscleral sample with the corneal stroma facing up and the corneal epithelium facing down over the concave surface of the trephine and press the corneoscleral sample with the trephine knife (see Notes 12 and 13, and Fig. 1e–h). This will remove the central cornea allowing you to obtain the corneoscleral ring, which includes scleral, limbal, and corneal tissues (Fig. 1g). 12. Place the corneoscleral ring with the epithelium facing down on the lid of the Petri dish (Fig. 1i). 13. Prepare limbal explants. Hold the corneoscleral ring with finetipped tweezers and cut limbal portions of 1–3 mm2 using a scalpel (see Notes 14 and 15, Figs. 1j, k and 2). 3.3 Initial Limbal Primary Culture (LPC)

1. Place the limbal explants with the limbal epithelium facing up in a 12-well plate (a single explant per well of 3.8 cm2) (Fig. 1l).

Fig. 2 Optimal size of limbal explants. Diagram showing optimal limbal explant characteristics (see Note 15)


Marina Lo´pez-Paniagua et al.

Fig. 3 Cell culture growth from a single limbal explant. (a–e) Representative phase contrast microphotographs of cell outgrowths from a single limbal explant replated for generating consecutive limbal primary cultures (LPC). (f–j) Representative phase contrast microphotographs of confluent LPC samples. LPC0-3 shows homogeneous, cuboidal, and epithelial-like morphology, while LPC4 shows elongated and fibroblast-like morphology. Scale bar: 100 μm

Keep the plate uncovered and inside the laminar-flow hood for 30 min (see Note 16). 2. Add 50 μL of FBS to each well and incubate overnight at 37  C, 5% CO2, and 95% relative humidity in a cell culture incubator (see Notes 3, 17, and 18). 3. Add 500 μl of culture medium to each well and incubate the plate at 37  C, 5% CO2, and 95% relative humidity. 4. Monitor each limbal explant daily under an inverted phase contrast microscope and carefully change the culture medium every 2–3 days (see Notes 19 and 20). 5. Remove the limbal explant from its well when cells migrating from the edges of the limbal explant form a ring of cells around it. In order to perform this step, take the explant from the scleral area with fine-tipped tweezers and use other similar tweezers to break the connection between the cell outgrowth and the limbal tissue by making a quick upward movement (see Note 21). 6. Once an explant is removed from a well, maintain the cell outgrowth under the same culture conditions until the primary culture reaches confluence (see Note 22 and Fig. 3), at which time designate it “LPC0.” 3.4

Consecutive LPC

1. Place the removed limbal explant into a new 3.8 cm2 polystyrene well, as in step 1 from Subheading 3.3. 2. Follow steps 2–6 from Subheading 3.3 in order to obtain consecutive LPCs (see Notes 23–25, and Figs. 3 and 4).

Consecutive Culture of a Human Limbal Explant


Fig. 4 Experimental timeline for generating consecutive human limbal epithelial stem cell cultures by replating a single limbal explant. In order to establish each limbal primary culture (LPC) the explant is maintained in culture until a cell outgrowth has surrounded it. Then, the explant is removed and replated in order to obtain the next LPC. After removing the explant, cells are maintained in culture until they reach confluence, at which time the culture is considered an established LPC. The total time needed from limbal plating to reach a confluent LPC0 is about 26 days, while the time required to obtain confluent LPC1 and LCP2 cultures is around 30 and 24 days, respectively. In addition, time elapsed between when the first limbal explant is plated into a well of a 12-well plate until a confluent LPC2 is obtained is about 42 days (see Note 25)


Notes 1. The shipping and culture media can be stored at 4  C for 1 month. If they have not been used during this period of time, they should be discarded. 2. To carry out this protocol, we recommend using this composition for the shipping medium. Nevertheless, another composition could also be valid for sending corneoscleral samples intended to perform this protocol. 3. This protocol could also be carried out using a biosafe culture medium. 4. Fill the bottle/container with shipping medium completely to ensure that the ocular sample never gets dry and damaged. We strongly recommend not using low volumes of shipping medium in order to prevent, as much as possible, the potential consequences of evaporation or freezing due to changes in environmental temperature. 5. Each bottle/container should be used to store and ship only one corneoscleral sample. 6. We recommend collecting corneoscleral tissues from the eyeball leaving at least 3 mm of scleral tissue surrounding the corneal area. We suggest marking the superior area in order to orient the tissue sample once is out from the eyeball, since


Marina Lo´pez-Paniagua et al.

limbal epithelial stem cells are mainly located in the superior and the inferior limbal ring areas [17]. The marking can be done by leaving a small projection of scleral tissue in the superior area. 7. Do not keep ocular tissues under storage conditions for more than 5 days. 8. At the end of this step, the corneoscleral tissue must be placed with corneal epithelium facing up. 9. We recommend avoiding the use of a smaller pipette, since remnants of conjunctival tissue could obstruct the pipette and prevent the proper collection and removal of tissue debris. 10. In order to avoid damaging the limbus, it is important to take the corneoscleral sample from the scleral region without touching the limbal or the corneal areas. 11. First divide the trephine into its two parts by separating the part containing the blade from the part with the concave surface where the corneoscleral sample should be placed (Fig. 1e–h). 12. To cut a regular limbal ring, make sure that the corneoscleral tissue is perfectly centered on the concave surface of the trephine (Fig. 1e). 13. To cut the central corneal button press the corneoscleral sample with the blade only once (Fig. 1f). If you press the tissue with the blade two or more times, it is possible that the tissue could move over the concave surface of the trephine and, consequently, produce an irregular limbal ring. 14. First, we suggest dividing the corneoscleral ring into two equal portions, using a scalpel. Keep one of them in a Petri dish with 4 mL of wash solution to prevent tissue drying and damaging. Second, we recommend cutting the other half portion into two equal pieces, again submerging one of them in the same wash solution. Once you have ¼ of the corneoscleral ring on the lid of the Petri dish, divide this tissue portion into 6 equal sections, in order to obtain limbal explants of 1–3 mm2 each. Divide each ¼ of the corneoscleral portion into 6 equal limbal explants this way (Fig. 1j, k). 15. The size of each limbal explant can be verified using a surgical compass. 16. The 12-well plate with the limbal explants should be placed without its cover at the back of the laminar-flow hood. In this way, the air flow will favor the evaporation of the remaining wash solution and, consequently, the successful attachment of the explant to the polystyrene.

Consecutive Culture of a Human Limbal Explant


17. If using the biosafe culture medium to carry out this protocol, 50 μL of HS instead of FBS must be added in the step. 18. The FBS (or HS in cases where biosafe culture medium is used, see Note 17) must be added to the top of the limbal explants using a micropipette (20–200 μL) that is maintained perpendicular to the limbal explant as serum is added. 19. A micropipette (100–1000 μL) should be used to routinely change the culture medium. The tip of the micropipette should be positioned next to the wall of the well during culture medium changes to prevent touching, and possibly hoisting, limbal tissue. In addition, fresh culture medium should be slowly added to the well in order to ensure that limbal explants remain in the same position. 20. We suggest making a mark in the cover of the plate indicating which wells contain migrating cells growing from limbal explants. This will allow you to add 1 mL of culture medium only to marked wells when changing the culture medium. The outgrowth of the different limbal explants may vary, meaning that within the same plate you may have to add 0.5 mL of culture medium/well (if no outgrowth is observed) or 1 mL of culture medium/well (when outgrowth is observed). 21. It is possible that some cells may detach from the substratum when the limbal explant is removed from the well. 22. We consider an LPC successful when it reaches more than 80% confluence. 23. The same limbal explant can be consecutively cultivated up to seven times, obtaining confluent LPCs from LPC0 to LPC6. However, only the first three LPCs (LPC0-LPC2) will maintain an epithelial cell morphology and a limbal epithelial cell phenotype [15]. If using the biosafe culture medium to carry out the protocol (see Note 3), only two consecutive LPCs (LPC0 and LPC1) will maintain a limbal epithelial cell phenotype [16]. 24. The percentage of confluent LPCs obtained from a corneoscleral sample can vary depending on the tissue donor age, since there is a decreasing trend in the growth potential of limbal tissues with increasing donor age [15, 18, 19]. 25. In case of using the biosafe culture medium for performing this protocol, note that both time elapsed between limbal explant plating to limbal explant removal and time elapsed between


Marina Lo´pez-Paniagua et al.

limbal explant removal to reach a confluent LPC might be slightly different from times mentioned here [16].

Acknowledgments Financial Support: Ministerio de Economı´a y Competitividad and Fondo Europeo de Desarrollo Regional, Spain (SAF2015-63594-R MINECO/FEDER, EU). Centro de Investigacio´n Biome´dica en Red de Bioingenierı´a, Biomateriales y Nanomedicina (CIBERBBN). Instituto de Salud Carlos III, Spain. Centro en Red de Medicina Regenerativa y Terapia Celular de Castilla y Leo´n, Spain. References 1. Cotsarelis G, Cheng SZ, Dong G, Sun TT, Lavker RM (1989) Existence of slow-cycling limbal epithelial basal cells that can be preferentially stimulated to proliferate: implications on epithelial stem cells. Cell 57:201–209 2. Notara M, Alatza A, Gilfillan J, Harris AR, Levis HJ, Schrader S, Vernon A, Daniels JT (2010) In sickness and in health: corneal epithelial stem cell biology, pathology and therapy. Exp Eye Res 90:188–195 3. Garg P, Krishna PV, Stratis AK, Gopinathan U (2005) The value of corneal transplantation in reducing blindness. Eye 19:1106–1114. 4. Pellegrini G, Traverso CE, Franzi AT, Zingirian M, Cancedda R, De Luca M (1997) Long-term restoration of damaged corneal surfaces with autologous cultivated corneal epithelium. Lancet 349:990–993. https://doi. org/10.1016/S0140-6736(96)11188-0 5. Shortt AJ, Secker GA, Rajan MS, Meligonis G, Dart JK, Tuft SJ, Daniels JT (2008) Ex vivo expansion and transplantation of limbal epithelial stem cells. Ophthalmology 115:1989–1997. OPHTHA.2008.04.039 6. Rama P, Matuska S, Paganoni G, Spinelli A, De Luca M, Pellegrini G (2010) Limbal stem-cell therapy and long-term corneal regeneration. N Engl J Med 363:147–155. 10.1056/NEJMoa0905955 7. Baylis O, Figueiredo F, Henein C, Lako M, Ahmad S (2011) 13 years of cultured limbal epithelial cell therapy: a review of the outcomes. J Cell Biochem 112(4):993–1002 8. Zakaria N, Possemiers T, Dhubhghaill SN, Leysen I, Rozema J, Koppen C, Timmermans JP, Berneman Z, Tassignon MJ (2014) Results of a phase I/II clinical trial: standardized,

non-xenogenic, cultivated limbal stem cell transplantation. J Transl Med 12:58. https:// 9. Zhao Y, Ma L (2015) Systematic review and meta-analysis on transplantation of ex vivo cultivated limbal epithelial stem cell on amniotic membrane in limbal stem cell deficiency. Cornea 34:592–600. 1097/ICO.0000000000000398 10. Ramı´rez BE, Sa´nchez A, Herreras JM, Ferna´ndez I, Garcı´a-Sancho J, Nieto-Miguel T, Calonge M (2015) Stem cell therapy for corneal epithelium regeneration following good manufacturing and clinical procedures. Biomed Res Int 2015:408495. 1155/2015/408495 11. Calonge M, Pe´rez I, Galindo S, Nieto-MiguelT, Lo´pez-Paniagua M, Ferna´ndez I, Alberca M, Garcı´a-Sancho J, Sa´nchez A, Herreras JM (2019) A proof-of-concept clinical trial using mesenchymal stem cells for the treatment of corneal epithelial stem cell deficiency. Transl Res 206:18–40. 1016/J.TRSL.2018.11.003 12. Lindberg K, Brown ME, Chaves HV, Kenyon KR, Rheinwald JG (1993) In vitro propagation of human ocular surface epithelial cells for transplantation. Investig Ophthalmol Vis Sci 34:2672–2679 13. Shortt AJ, Secker GA, Notara MD, Limb GA, Khaw PT, Tuft SJ, Daniels JT (2007) Transplantation of ex vivo cultured limbal epithelial stem cells: a review of techniques and clinical results. Surv Ophthalmol 52:483–502. SURVOPHTHAL.2007.06.013 14. Mariappan I, Maddileti S, Savy S, Tiwari S, Gaddipati S, Fatima A, Sangwan VS, Balasubramanian D, Vemuganti GK (2010) In

Consecutive Culture of a Human Limbal Explant vitro culture and expansion of human limbal epithelial cells. Nat Protoc 5:1470–1479. 15. Lo´pez-Paniagua M, Nieto-Miguel T, de la Mata A, Galindo S, Herreras JM, Corrales RM, Calonge M (2013) Consecutive expansion of limbal epithelial stem cells from a single limbal biopsy. Curr Eye Res 38:537–549. 767350 16. Lo´pez-Paniagua M, Nieto-Miguel T, de la Mata A, Galindo S, Herreras JM, Corrales RM, Calonge M (2017) Successful consecutive expansion of limbal explants using a biosafe culture medium under feeder layer-free conditions. Curr Eye Res 42:685–695. https://doi. org/10.1080/02713683.2016.1250278 17. Utheim TP, Raeder S, Olstad OK, Utheim ØA, de la Paz M, Cheng R, Huynh TT, Messelt E,


Roald B, Lyberg T (2009) Comparison of the histology, gene expression profile, and phenotype of cultured human limbal epithelial cells from different limbal regions. Investig Ophthalmol Vis Sci 50:5165–5172. https:// 18. Notara M, Shortt AJ, O’Callaghan AR, Daniels JT (2013) The impact of age on the physical and cellular properties of the human limbal stem cell niche. Age (Omaha) 35:289–300. 19. James SE, Rowe A, Ilari L, Daya S, Martin R (2001) The potential for eye bank limbal rings to generate cultured corneal epithelial allografts. Cornea 20:488–494. 10.1097/00003226-200107000-00010

Chapter 5 A Guide to the Development of Human CorneaOrganoids from Induced Pluripotent Stem Cells in Culture James W. Foster, Karl J. Wahlin, and Shukti Chakravarti Abstract The cornea is the outermost transparent and refractive barrier surface of the eye necessary for vision. Development of the cornea involves the coordinated production of extracellular matrix, epithelial differentiation, and endothelial cell expansion to produce a highly transparent tissue. Here we describe the production of multilayered three-dimensional organoids from human-induced pluripotent stem cells. These organoids have the potential for multiple downstream applications which are currently unattainable using traditional in vitro techniques. Key words Cornea, Organoids, iPSC, Corneal epithelium, Corneal stroma


Introduction The cornea is the outermost barrier tissue of the eye and may be viewed as a specialized skin to the eye. From a tissue engineering point of view, made up of three major resident cell types, it is a relatively simple tissue. The outer most layer of the cornea consists of a stratified epithelium, a central stroma of orthogonally stacked collagen fibril lamellae interspersed with flattened cells, the keratocytes, and an innermost single cell layered endothelium. The selfrenewal potential is highest in the basal epithelial cells of the stratified epithelium, decreasing dramatically in the stromal and endothelial cells [1–3]. In development, the cornea arises from the cranial ectoderm and neural crest cells which subsequently differentiate into the Pax6 positive ocular surface ectoderm (giving rise to the corneal epithelium) and the neuroectoderm (keratocytes and endothelial cells) [4, 5]. These cell types must then coordinate to produce the highly aligned collagenous extracellular matrix necessary for the unobstructed passage of incident light [6]. The generation of induced pluripotent stem cells (iPSC) has revolutionized the field of regenerative medicine and has allowed

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



James W. Foster et al.

for investigations into development and disease and opened new avenues for therapeutic intervention [7]. In the eye, the focus has often been on the development of the retina and in this the development of optic cups has been a major advance [8–12]. Since these early investigations the field has progressed to produce ever more specific cell types [10, 13–15]. In the cornea field, there has been progress in the generation of iPSC-derived corneal epithelial cells [16], keratocytes [17], endothelial cells [18], and all cell types in 2D culture [19]. Thus, we sought to leverage the development of iPSC-derived corneal progenitors in a self-assembly model of corneal development to produce stem cell-derived corneal organoids [20]. Here we describe these techniques in detail.



2.1 IMR 90.4 iPSC Cell Line Culture

2.2 Small Molecules and Supplements

IMR-90.4 iPSC cells are maintained by clonal propagation on growth factor-reduced Matrigel in mTeSR1 medium under hypoxia (10% CO2, 5% O2) in a copper-lined incubator. 1. 10 mM all-trans-retinoic acid (ATRA): 50 mg ATRA, 16 mL of dimethyl sulfoxide (DMSO). Store at 80  C for up to 4 months. 2. 10 mM ( )-Blebbistatin: 1 mg blebbistatin, 340 μL DMSO. Make 10 aliquots and store at 80  C. 3. E6 supplement: 100 mL water, 7.5 g NaHCO3, 97 mg insulin, 53.5 mg holo-transferrin, 320 mg L-ascorbic acid, 70 mg sodium selenite. Mix and make 10 mL aliquots into 15 mL tubes. Store in designated box at 80  C. 4. 3 mM IWR-1-endo: 10 mg IWR, 8.1 mL DMSO. Make 50 μL aliquots and store at 80  C. 5. 3.7 M sodium chloride (NaCl): 10.95 g NaCl, 50 mL cell culture grade water. NaCl stock solution cannot be easily filter sterilized. 6. 100 μM smoothened agonist (SAG): 1 mg SAG, 16 mL DMSO. Make 100 μL aliquots and store at 80  C. 7. 400 mM Taurine: 500 mg taurine, 10 mL sterile cell culture grade water. Make 1.25 mL aliquots and store at 80  C. 8. Matrigel-Growth Factor Reduced (GFR). 9. Accutase. 10. Hanks buffered saline solution (HBSS).

2.3 Medium Solutions

1. mTeSR1 medium. 2. Neural induction medium (NIM): Dulbecco’s modified Eagle medium (DMEM), 1% B27 minus vitamin A, 1 mM pyruvate,

iPSC-Derived Corneal Organoids


1 nonessential amino acids (NEAA), 1 Glutamax, 2 E6 supplement, 0.88 g/L NaCl. NaCl should raise the osmolarity by +30 mOsm to ~330–340 mOsm. Sterilize using a 0.22 μm filter. 3. Optic vesicle induction medium: NIM, 100 nM of the smoothened agonist (SAG). 4. Long-term retina maturation medium (LTR): 3:1 mix of DMEM:F12, 1% B27 supplement, 10% heat-inactivated fetal bovine serum (FBS), 1 mM pyruvate, 1 NEAA, 1 Glutamax, 1 mM taurine. LTR is sterilized with a 0.22 μm sterile filter. (For inclusion of 500 nM all-trans-retinoic acid, see Note 1). 2.4 Tungsten Needles for Neural Vesicle Excision


Tungsten needles are made by embedding a 0.64 mm diameter tungsten wire into a hollow glass rod cut to size with a glass cutter and mounted using epoxy resin. The tungsten wire is connected to the cathode of a gel electrophoresis DC power supply; a carbon rod is connected to the anode. Tips are immersed in a 0.5 M NaOH solution and low voltage (30 V) applied. Bubbles forming at the tip of the tungsten wire indicate active electrolytic sharpening. The tip of the tungsten wire should be repeatedly immersed in an up/down motion until the tip becomes very sharp. The first-time electrodes are sharpened it can be helpful to bevel the edges with a sharpening stone. Sharpening is recommended prior to each cutting session.

Methods This method is permissive for generating retinal organoids, corneal organoids, and hybrids. It is at the discretion of the investigator to select for those organoids which meet their needs. A visual representation of this method is provided in Fig. 1.


Stem Cell Culture

1. IMR-90.4 iPSCs should be maintained in mTeSR1 medium on dishes coated with 1% (v/v) Matrigel-GFR™ at 37  C under hypoxic conditions prior to reaggregation. 2. Replace medium every day as per standard cell culture conditions. 3. Once colonies reach 70% of a field of view under 10 magnification, treat cells with Accutase for 8–10 min. 4. Dissociate into a single cell suspension and quench Accutase activity with mTeSR1 plus 5 μM blebbistatin (B) to improve single cell survival. 5. Spin cells at 80  g for 5 min. 6. Resuspended in mTeSR1 + B and plated at 5000 cells per 35 mm well.


James W. Foster et al.

Fig. 1 Visual representation of corneal organoid methodology. IPS cells are taken through sequential stages of differentiation from pluripotency, neural phenotype, optic induction, and finally corneal selection over 30 days

iPSC-Derived Corneal Organoids


7. After 48 h, feed cells with mTeSR1 alone. 8. The entire protocol is carried out without antibiotics to minimize cell stress. 3.2 Forced Aggregation

1. On day 1, passage cells as described in Subheading 3.1. 2. Resuspend cells in mTeSR1 containing 5 μM blebbistatin and count with a hemocytometer. 3. For a full 96-well plate, 3.0  105 cells are diluted in 5.0 mL of media (60,000 cells/mL) and mixed by inversion 10 times. 4. Transfer 50 μL of cell suspension (corresponding to 3000 cells) to a ChannelMate reservoir and add to each well of an ultralow attachment U-bottom 96-well plate using a multichannel pipette. This approach quickly dispenses cells and minimizes variability. 5. After plating, place cells back in hypoxia overnight at 37  C to facilitate cell survival and aggregate formation (see Note 2).


Neural Induction

1. On day 2 induce forced aggregates to a neuronal phenotype by supplementing with an equal volume (50 μL) of NIM supplemented with 2% Matrigel and 6 μM IWR1-e. 2. Transfer organoids to normoxia on day 2 (5% CO2, 20% O2) and maintain for the duration of the experiment. 3. On days 3 and 4, feed aggregates daily by addition of 50 μL of NIM supplemented with 1% Matrigel and 3 μM IWR1-e. 4. On days 5–7, feed aggregates daily with a 50% media change (100 μL) with NIM supplemented with 1% Matrigel and 3 μM IWR1-e. 5. On day 9, feed aggregates with a 50% media change (100 μL) with NIM supplemented without matrigel or 3 μM IWR1-e. Media is changed every other day at this point. 6. On day 11, pool aggregates into 15 mL tubes and are allowed to sink (see Notes 3 and 4). 7. Rinse the aggregates three times with Hanks buffered saline solution (HBSS) to remove Matrigel, cell debris, and other components of NIM. 8. Resuspend in 20 mL of optic vesicle medium in ultralow attachment T-75 flasks or 15 mL of medium in untreated deep dish 10 cm polystyrene petri dishes and keep in a standard tissue culture incubator. 9. Feed vesicles every other day with >75% media exchange (see Note 5).


James W. Foster et al.

3.4 Neural Vesicle Excision

1. On days 11 and 13, feed organoids with NIM media + 100nM SAG. On days 11–15, visually identify and manually excise vesicles from the rest of the organoids using a microscope and a scissoring motion with fine-tipped tungsten needles (see Note 6). 2. Once isolated from the central mass, place the presumptive neural vesicles back in optic vesicle medium containing 100 nM SAG in ultralow binding T-75 flasks or untreated 10 cm polystyrene petri dishes in a standard normoxic incubator at 37  C (see Note 7).

3.5 Organoid Maturation and Selection



1. On days 15–19, feed vesicles with LTR media + SAG every 2 days. 2. On day 21, feed vesicles with LTR + ATRA every 2–3 days for the duration of the experiment (see Note 8). By day 31, corneal organoids should be apparent by their translucent cystic appearance. These structures differ from retinal organoids or RPE organoids that have distinctive cup-shaped or darkly pigmented appearances, respectively (see Note 9). Corneal organoids can be differentiated from retinal progenitors by their lack of pigmented cells, transparent appearance, and larger cystic morphology. For molecular validation we recommend assaying by immunofluorescence where good antibodies are available or a transcript level assessment by RT-PCR: 1. PAX6—Ocular phenotype. 2. P63α—Limbal epithelium. 3. KERA—Stromal keratocyte. 4. KRT14—Basal epithelium. 5. COL8A1—Bowman’s membrane. 6. Six6—Retinal phenotype (should be absent as tested by RT-PCR). 7. OCT4 and Nanog—Should be absent. Additional information on assays and antibodies for these markers are provided in our previous work [15, 20].


Notes 1. ATRA is not added to LTR between day 15 and 20; it is added fresh for each feeding after day 20. 2. It is strongly recommended to tap the sides of the 96-well dish intermittently for up to 1 h after plating to ensure that solitary organoids form.

iPSC-Derived Corneal Organoids


3. To minimize damage to the aggregates we recommend using large diameter 1000 μL tips for collection or 200 μL tips which have had the lower ¼ removed to expand their smallest diameter. 4. Failure to form well-organized vesicles from between days 6 and 12 indicates a failure of differentiation which can lead to unreliable corneas. These aggregates should be discarded. 5. Organoids are visualized every day until day 10, to verify that neural vesicles budding from the central mass have formed. 6. Detailed images of the budding neuronal vesicles can be found in “Wahlin, K. J. et al. Photoreceptor Outer Segment-like Structures in Long-Term 3D Retinas from Human Pluripotent Stem Cells. Sci. Rep. 7, 766 (2017)”. 7. Too many organoids can adversely affect cell survival; thus, it may be necessary to titrate different numbers of organoids for long-term survival if other cell lines are used. A range from 20 to 50 vesicles per 10 cm dish is recommended. 8. Corneal organoids can be allowed to mature at the discretion of the researcher. It is possible to mature these organoids >120 days if so desired. 9. It is common for organoids to contain areas of dense pigmented cells and nonpigmented cells which correspond to presumptive RPE and corneal regions, respectively.

Acknowledgements This work was supported by the NIH EY026104 (SC) and R00EY024648 (KW). References 1. Cotsarelis G, Cheng SZ, Dong G et al (1989) Existence of slow-cycling limbal epithelial basal cells that can be preferentially stimulated to proliferate: implications on epithelial stem cells. Cell 57:201–209 2. Lavker RM, Tseng SCG, Sun T-T (2004) Corneal epithelial stem cells at the limbus: looking at some old problems from a new angle. Exp Eye Res 78:433–446 3. Scott S-G, Jun AS, Chakravarti S (2011) Sphere formation from corneal keratocytes and phenotype specific markers. Exp Eye Res 93:898–905. exer.2011.10.004 4. Lwigale PY (2015) Corneal development: different cells from a common progenitor. Prog

Mol Biol Transl Sci 134:43–59. https://doi. org/10.1016/bs.pmbts.2015.04.003 5. Graw J (2010) Eye development. Curr Top Dev Biol 90:343–386. 1016/S0070-2153(10)90010-0 6. Hassell JR, Birk DE (2010) The molecular basis of corneal transparency. Exp Eye Res 91:326–335. exer.2010.06.021 7. Carey BW, Markoulaki S, Hanna J et al (2009) Reprogramming of murine and human somatic cells using a single polycistronic vector. Proc Natl Acad Sci 106:157–162. 10.1073/pnas.0811426106 8. Eiraku M, Takata N, Ishibashi H et al (2011) Self-organizing optic-cup morphogenesis in


James W. Foster et al.

three-dimensional culture. Nature 472:51–56. 9. Nakano T, Ando S, Takata N et al (2012) Selfformation of optic cups and storable stratified neural retina from human ESCs. Cell Stem Cell 10:771–785. stem.2012.05.009 10. Zhong X, Gutierrez C, Xue T et al (2014) Generation of three-dimensional retinal tissue with functional photoreceptors from human iPSCs. Nat Commun 5:4047. https://doi. org/10.1038/ncomms5047 11. Vergara MN, Flores-Bellver M, AparicioDomingo S et al (2017) Three-dimensional automated reporter quantification (3D-ARQ) technology enables quantitative screening in retinal organoids. Development 144:3698–3705. dev.146290 12. Eldred MK, Charlton-Perkins M, Muresan L, Harris WA (2017) Self-organising aggregates of zebrafish retinal cells for investigating mechanisms of neural lamination. Development 144:1097–1106. 1242/dev.142760 13. Maruotti J, Wahlin K, Gorrell D et al (2013) A simple and scalable process for the differentiation of retinal pigment epithelium from human pluripotent stem cells. Stem Cells Transl Med 2:341–354. 2012-0106 14. Wahlin KJ, Maruotti J, Zack DJ (2014) Modeling retinal dystrophies using patient-derived induced pluripotent stem cells. Adv Exp Med

Biol 801:157–164. 978-1-4614-3209-8_20 15. Wahlin KJ, Maruotti JA, Sripathi SR et al (2017) Photoreceptor outer segment-like structures in long-term 3D retinas from human pluripotent stem cells. Sci Rep 7:766. 16. de la Torre RA MG, Nieto-Nicolau N, Morales-Pastor A, Casaroli-Marano RP (2017) Determination of the culture time point to induce corneal epithelial differentiation in induced pluripotent stem cells. Transplant Proc 49:2292–2295. 10.1016/j.transproceed.2017.09.047 17. Joseph R, Srivastava OP, Pfister RR (2014) Generation of induced pluripotent stem cells from normal and keratoconus corneal fibroblasts using viral- and non-viral methods. Invest Ophthalmol Vis Sci 55:523–523 18. Zhao JJ, Afshari NA (2016) Generation of human corneal endothelial cells via in vitro ocular lineage restriction of pluripotent stem cells. Invest Ophthalmol Vis Sci 57:6878–6884. iovs.16-20024 19. Hayashi R, Ishikawa Y, Sasamoto Y et al (2016) Co-ordinated ocular development from human iPS cells and recovery of corneal function. Nature 531(7594):376–380. https://doi. org/10.1038/nature17000 20. Foster JW, Wahlin K, Adams SM et al (2017) Cornea organoids from human induced pluripotent stem cells. Sci Rep 7:41286. https://doi. org/10.1038/srep41286

Chapter 6 Gene Editing for Corneal Stromal Regeneration Tara Moore, Connie Chao-Shern, Larry DeDionisio, Kathleen A. Christie, and M. Andrew Nesbit Abstract CRISPR/Cas9 gene editing holds the promise of sequence-specific alteration of the genome to achieve therapeutic benefit in the treated tissue. Cas9 is an RNA-guided nuclease in which the sequence of the RNA can be altered to match the desired target. However, care must be taken in target choice and RNA guide design to ensure both maximum on-target and minimum off-target activity. The cornea is an ideal tissue for gene therapy due to its small surface area, accessibility, immune privilege, avascularity, and ease of visualization. Herein, we describe the design, testing, and delivery of Cas9 and guide RNAs to target genes expressed in the cornea. Key words Gene editing, sgRNA expression construct, Dual luciferase assay, In vitro digest, Lymphoblastoid cell line, Nucleofection, Polymerase chain reaction (PCR), Corneal endothelial and epithelial cell culture, In vivo imaging, Intracameral injection, Adeno-associated virus (AAV)


Introduction The discovery, development, and application of Clustered Regularly Interspaced Palindromic Repeats (CRISPR)/CRISPR associated protein (Cas) systems has brought the promise of gene therapy, tailored specifically for the patient, closer to realization. CRISPR/Cas9 is a bacterial RNA-guided endonuclease that has been modified for use in mammalian cells, consisting of a Cas9 nuclease that forms a complex with a simplified single guide RNA (sgRNA), formed by the fusion of crRNA and tracrRNAs, which guides the nuclease to its DNA target [1–3]. In addition to a match between the RNA and DNA sequences, the Cas9 nuclease itself recognizes and binds to a protospacer adjacent motif (PAM) directly upstream of the guide RNA sequence (Fig. 1). Following target recognition, the Cas9 nuclease makes a double strand break (DSB) which triggers the DNA repair machinery of the cell, leading to either error-prone nonhomologous end joining (NHEJ) or precise homology directed repair (HDR) [4]. It is by harnessing

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Tara Moore et al.

Fig. 1 S. pyogenes Cas9 (purple outline) can be directed to cut any sequence in the genome (DNA target in grey), provided it is directly upstream of a protospacer adjacent motif known as PAM (pink box). This can be achieved by altering the 20 nucleotide guide sequence, which is associated with a 82 nucleotide scaffold [8]

these different cellular responses that different forms of gene editing can be achieved. It is important to understand, however, that DNA cleavage by Cas9 nuclease can occur in some cases where there are mismatches between the guide sequence and the target [5, 6]. This can lead to off-targeting elsewhere in the genome or, as we have shown [7], failure to discriminate between mutant and wild-type alleles. Selectivity between wild-type and mutant alleles can be improved if the mutation is located within the PAM [7–9]. Successful application of targeted CRISPR/Cas9 gene editing requires a stepwise approach by (1) careful selection of the target, (2) testing the efficiency with which the Cas9/sgRNA nuclease cleaves the selected target (on-target), (3) the specificity for the on-target compared to off-target, and (4) consideration of the method of delivery to the chosen tissue.



2.1 SpCas9 sgRNA Design

1. Sequence of target gene.

2.2 sgRNA Construct Preparation

1. S. pyogenes Cas9 vector plasmid or S. aureus Cas9 vector plasmid.

2. Online design program (Benchling, CasOFFinder, etc.).

2. BbsI restriction endonuclease. 3. Guide oligos containing the following template: 4. Top guide oligo—50 -CACCG - insert 20 nt of sgRNA – 30 . 5. Bottom guide oligo—50 -C - insert complement of 20 nt CAAA – 30 .

Gene Editing for Corneal Stromal Regeneration


6. T4 DNA ligase. 7. LB Broth. 8. LB Agar. 9. Antibiotic—check plasmid map to see which is needed. 10. dH5α E. coli competent cells. 11. Plasmid DNA isolation kit. 2.3 Human Embryonic Kidney Cells (HEK) AD293 Maintenance

1. HEK AD293 cells. 2. DMEM low glucose. 3. Fetal bovine serum. 4. Trypsin/EDTA (0.25%). 5. Phosphate-buffered saline.

2.4 Dual Luciferase Assay

1. DMEM low glucose. 2. Fetal bovine serum. 3. S. pyogenes Cas9 vector plasmid containing sgRNA of interest pSpCas9(BB)-2A-Puro (PX459) V2.0. 4. psiTEST-LUC-target vector containing target sequence of sgRNA. 5. Renilla luciferase reporter plasmid. 6. Lipofectamine 2000. 7. Optimem. 8. Dual-Luciferase® Reporter Assay System. 9. BMG Labtech, LUMIstar Optima plate reader (or equivalent).

2.5 In Vitro DNA Cleavage Assay

1. S. pyogenes EnGen Cas9 S.pyogenes and reaction buffer. 2. Modified synthetic sgRNA. 3. Nuclease-free H2O. 4. Proteinase K. 5. DNA cleavage template containing 20 bp target site and adjacent PAM (The cleavage template can either be circular or linearized plasmid, PCR products, or synthesized oligonucleotides).

2.6 Generate Lymphocyte Cell Line (LCL)

1. RPMI media. 2. Fetal bovine serum. 3. Ficoll-Paque PLUS. 4. Epstein-Barr virus (Human gammaherpesvirus 4 (HHV-4), ATCC® VR-1492). 5. Phytohemagglutinin.


Tara Moore et al.

2.7 Nucleofection of LCLs with Ribonucleic Proteins (RNPs)

1. S. pyogenes EnGen Cas9 NLS. 2. Modified synthetic sgRNA. 3. SF Cell line 4D-Nucleofector X kit. 4. Lonza 4D nucleofector. 5. QIAamp DNA Mini Kit.

2.8 Polymerase Chain Reaction (PCR)

1. Dream Taq. 2. Primers flanking region of interest. 3. DNA to be amplified. 4. Thermocycler.

2.9 Intrastromal Injection of CRISPR/ Cas9

1. Capillary Glass, 1.0 mm outer diameter, 0.58 mm inner diameter, 10.16 cm (4 in.). 2. DMZ universal puller (or equivalent). 3. Hamilton 701 RN 10 μL syringe without needle. 4. Hamilton RN Compression Fitting 1 mm. 5. Ketamine. 6. Xylazine. 7. dH2O for injections, 8. Tropicamide. 9. Phenylephrine. 10. Tooth (palate) bar only from model 923-B Mouse Gas Anesthesia Head Holder. 11. Gas mask. 12. Surgical microscope. 13. Fusidic gel.

2.10 IVIS In Vivo Imaging of Fluorescence

1. Isoflurane. 2. Isoflurane chamber. 3. Xenogen IVIS Lumina in vivo imager. 4. Custom mouse holder tube (Fig. 2). 5. 1.5–2% isoflurane.

3 3.1

Methods sgRNA Design

sgRNAs are designed using an online design program (Benchling, CasOFFinder, etc.) and chosen on the basis of good on- and off-target scores. Overhangs for the BbsI restriction site (shown in bold) are added to the 20 bp sgRNA sequence of the target site.

Gene Editing for Corneal Stromal Regeneration


Fig. 2 Custom-made mouse holder for use in the IVIS that allows eyes to be placed at optimal angle to the camera

Top and bottom sgRNA oligos are ordered using the following template: Top guide oligo—50 -CACCG insert 20 nt of sgRNA - 30 Bottom guide oligo—50 -C insert complement of 20 nt CAAA - 30 3.2 CRISPR/Cas9 Construct Preparation

1. Digest pSpCas9(BB)-2A-Puro (PX459) V2.0 with BbsI and Buffer 2.1 for 2 hours at 37  C using the following reaction mixture: Volume H2O

Up to 50 μL

Buffer 2.1 (10)

5 μL

Plasmid DNA

1 μg

BbsI restriction enzyme

1 μL

2. Digestion mix is then electrophoresed on a 1% agarose gel alongside a DNA ladder; linearized pSpCas9(BB)-2A-Puro (PX459) V2.0 plasmid will run at 9 kb. Purify gel fragment containing your cloning backbone. 3. Resuspend the top and bottom strands of oligos for each guide oligo previously designed to a final concentration of 10 μM. Prepare the following mixture for annealing the sgRNA oligos (top and bottom strands):


Tara Moore et al.

Volume Top Oligo (100 μM)

1 μL

Bottom Oligo (100 μM)

1 μL


7 μL

5 DNA Ligase Buffer

2 μL

4. Anneal the oligos in a thermocycler by using the following parameters: 37  C for 30 min; 95  C for 5 min; ramp down to 25  C at 5  C/min. 5. Dilute the oligos (2 μL annealed oligo: 98 μL water). 6. Set up a ligation reaction for each sgRNA, as described below. Include a no-insert, pSpCas9(BB)-only negative control for ligation. Volume Digested SpCas9 Plasmid

100 ng

Diluted annealed Oligos

2 μL

5 DNA Ligation Buffer

4 μL


X μL

DNA Ligase Buffer

1 μL

Total Reaction Volume

20 μL

7. Incubate the ligation mixture for 1 hour at room temperature. 8. Add 4 μL ligation mixture to 45 μL of competent E. coli strain e.g. DH5α cells in a prechilled tube. (a) Leave on ice for 30 min. (b) Heat shock for 45 s in a preheated water bath at 42  C. (c) Leave on ice for 2 min. (d) Add 150 μL of prewarmed LB broth to centrifuge tube. (e) Incubate in shaking incubator at 37  C for 1 hour . (f) Plate all 200 μL transformation mix onto an LB agar plate containing 100 μg/mL ampicillin and place in 37  C incubator overnight. 9. Inspect the plates for colony growth. Pick colonies with a sterile tip and place into 20 μL LB broth in a 0.5 mL tube, select ~4 colonies per plate. Shake tip in LB to dislodge bacteria, remove, and discard tip. 10. Vortex each tube briefly to disperse bacterial colony.

Gene Editing for Corneal Stromal Regeneration


11. Perform colony PCR to determine successful ligation using the reaction below: Reagents

1 Reaction

5 Dream Taq Buffer

4 μL

Forward primer 10 μM (Cas9 BB seq Fwd–5-gggaaacgcctggtatcttt-30 )

1 μL

Reverse primer 10 μM (Bottom guide oligo—custom for each sgRNA)

1 μL

Colony broth

2 μL


11.92 μL

Dream Taq

0.08 μL

Total reaction volume

20 μL

12. Run PCR program: Pre-PCR holding stage

95  C for 3 min

Cycling stage (35 cycles)

95  C for 15 s 60  C for 15 s 72  C for 30 s

Post-PCR holding stage

72  C for 5 min

13. After PCR, run the material on 1% agarose 1 TBE gel. 14. If the sgRNA has been inserted, there should be a product size of 210 bp. The reverse primer used in the PCR reaction is the bottom guide oligo; therefore, this customized for each reaction. The forward primer primes off the pSpCas9(BB)-2APuro (PX459) V2.0 backbone. Consequently, there should only be a product in cases where the 20 bp sgRNA has been inserted into the digested plasmid. 15. Inoculate 3 mL LB broth with 100 μg/mL ampicillin with broth containing dispersed insert-positive colony in LB broth. 16. Incubate culture in a shaking incubator at 37  C overnight. 17. Isolate plasmid DNA from the cultures using a plasmid DNA purification kit, following the manufacturer’s protocol. 3.3 HEK AD293 Cell LINE MAINTENANCE

1. AD293 cells are cultured in 1 DMEM (containing 1 g/L glucose, 4.0 mM L-Glutamine, 1.0 mM Sodium Pyruvate) and 10% heat-inactivated fetal bovine serum (FBS) in an incubator at 37  C with 5% CO2.


Tara Moore et al.

2. Subculture conditions: Split sub-confluent cultures (70–80%) from 1:2 to 1:3, seeding at 1  106 cells/cm2 in a 75 cm2 flask, using 1 trypsin. 3. Remove the growth medium by aspiration. 4. Wash cells once with 2 mL phosphate-buffered saline. 5. Trypsinize cells for 1–3 min in 1 trypsin/EDTA at 37  C. 6. Dilute the cells with growth medium (volume at least equal to the volume of trypsin/EDTA solution added) to inactivate the trypsin. 7. Transfer cell suspension to a conical tube and centrifuge at 1000  g for 5 min at room temp. 8. Aspirate the supernatant and resuspend cells in 10 mL growth medium. 9. Seed cells at appropriate cell density. Place the cells in a 37  C incubator at 5% CO2. Monitor cell density daily. 3.4 Preparation of psiTEST-LUC-target Vector

1. Co-digest the psiTEST-LUC-target vector with NheI and KpnI in Buffer 1.1 for 2 hours at 37  C. This plasmid will be used for both the dual-luciferase assay and the in vitro digest reaction. Volume H2O

Up to 50 μL

Buffer 2.1 (10)

5 μL

Plasmid DNA

1 μg

KpnI restriction enzyme

1 μL

NheI restriction enzyme

1 μL

Digestion mix is then electrophoresed on a 1% agarose gel alongside a DNA ladder; desired band will run at 5223 bp and excised band will run at 12 bp. Purify gel fragment containing your cloning backbone. 2. Design a 50 bp template containing your 20 bp target sequence and adjacent PAM including overhangs complementary to the NheI and KpnI restriction sites. 3. Following steps 3–8 in Subheading 3.2 ligate the 50 bp dsODN into the digested psiTEST-LUC-target vector. 4. Verify the 50 bp dsODN has been inserted into the digested backbone using a colony PCR as previously described (Subheading 3.2, steps 9–13). Correct insertion will result in a PCR product of 641 bp; an empty backbone will result in a PCR product of 591 bp.

Gene Editing for Corneal Stromal Regeneration


5. This construct can now be used directly in the dual-luciferase assay. 6. For the in vitro digest use the psiTEST-LUC-target vector containing the 50 bp region encompassing your target site and adjacent PAM for PCR. Prepare PCR product containing the target sequence using the following primers: (a) FWD: 50 – ACCCCAACATCTTCGACGCGGGC-30 (b) REV: 50 – TGCTGTCCTGCCCCACCCCA – 30 7. Purify PCR product and use in the in vitro cleavage reaction at the specified concentration of 3 nM. The PCR product generated using the primers described in step 6 generates a template of 587 bp which has a M.W of 356701.7 g/mol (see Note 6). 3.5 Dual Luciferase Assay

1. Dual Luciferase assay workflow (Fig. 3). 2. The day prior to transfection seed AD293 cells in a volume of 100 μL in a 96-well plate at a density of 6.5  103 cells per well. 3. Cells are transfected using Lipofectamine 2000 according to the manufacturers’ instructions. 4. For each well the reaction mix is as follows: Reagents

1 reaction

Firefly luciferase reporter plasmid

20 ng

CRISPR/Cas9 expression construct

80 ng

Renilla luciferase plasmid

1 ng

Lipofectamine 2000

0.2 μL


25 μL

DMEM with 10% FBS

25 μL

5. Prepare a mastermix for your experimental design. 6. Combine Lipofectamine 2000 and Optimem and incubate at room temperature for 5 min. 7. Add Renilla plasmid. 8. Add Reporter construct plasmid. 9. Add sgRNA construct plasmid and incubate at room temperature for 5 min. 10. Add DMEM with 10% FBS. 11. Remove 50 μL culture medium from each well. 12. Add 50 μL transfection mixture to each well. 13. Incubate for 72 h FBS in an incubator at 37  C with 5% CO2. 14. Prepare the lysates using the Dual-Luciferase® Reporter Assay System kit, following the manufacturer’s instructions (see Note 1).

Fig. 3 Explanation of dual-luciferase assay. Diagram showing the different constructs that were transfected for the dual luciferase assay. The Renilla luciferase plasmid (Red) is used to normalize the firefly luciferase read, as it is not affected by the addition of sgGeneWT or sgGeneMut(ant). Firefly luciferase is shown in orange, while the sgRNAs for GeneWT and GeneMut(ant) are shown in green and blue, respectively [8]

Gene Editing for Corneal Stromal Regeneration

3.6 Post-transfection Luciferase Readout


1. Perform post-transfection luciferase readout on the LUMIstar Optima microplate reader. 2. Aliquot the Dual-Luciferase® Reporter Assay System reagents prior the experiment into 15 mL centrifuge tubes and store at 80  C. 3. Defrost the LAR and Stop&Glo reagents immediately before reading the plate (see Note 2). 4. To prepare the Stop&Glo reagent, dilute 50 Stop and Glo reagent (supplied with kit and kept at 20  C) with the S&G buffer previously aliquoted to make a 1 solution. 5. In the LUMIstar Reagent Box, insert reagent injector 1 into the LAR solution and reagent injector 2 into the S&G solution. Ensure each tube is inserted properly (see Note 3). 6. Open the LUMIstar software. 7. Prime pumps (see Note 4). 8. Click test protocols and select the pre-made program (see Note 5). 9. Edit the layout, based on the experiment. 10. Press OK and click the measurement icon (Traffic lights icon). 11. When the plate run is complete, open the LUMIstar analysis software and select the plate run from the drop-down menu. 12. Save data as Excel spreadsheet for analyses. 13. After using the machine, wrap the remaining solutions in foil and store for future use. 14. Remove injector needles from holes in the bottom of the machine and place in a waste tube. 15. Place tubing 1 and 2 in a tube containing dH2O, click the priming icon, and perform the first wash of machine using dH2O. 16. Place tubing 1 and 2 in a tube containing isopropanol, click the priming icon, and perform the second wash of machine using isopropanol. 17. Finally, remove the tubing from isopropanol and prime pumps using air, ensuring that the needles are still in waste tube.

3.7 In Vitro DNA Cleavage Assay

1. Prepare a cleavage template containing the target sgRNA sequence and an adjacent PAM. The cleavage template can either be circular or linearized plasmid, PCR products, or synthesized oligonucleotides (Subheading 3.4). 2. Prepare the reaction at room temperature in the following order:


Tara Moore et al.




20 μL

10 Cas9 Nuclease Reaction Buffer

3 μL

300 nM sgRNA

3 μL (30 nM)

1 μM Cas9 S. pyogenes

1 μL (30 nM)

Reaction volume

27 μL

3. Pre-incubate for 10 min at 25  C. 4. Add 3 μL 30 nM substrate DNA to the reaction (3 nM final). 5. Incubate at 37  C for 15 min. 6. Add 1 μL of Proteinase K to each sample, mix thoroughly, and pulse-spin in a microfuge. 7. Incubate at room temperature for 10 min. 8. Run digested products on a 1% agarose 1 TBEgel. 9. The fraction of the PCR product (587 bp) that is digested indicates the activity of the sgRNA. 10. An example of an in vitro cleavage reaction output is shown below in Fig. 4. 3.8 Generate Patient-Derived LCL

1. Take 5 mL of freshly collected whole blood and place in a sterile 50 mL conical tube. 2. Add an equal volume of RPMI medium containing 20% fetal calf serum and mix by gentle inversion. 3. Place 6.25 mL of Ficoll-Paque PLUS in a separate sterile 50 mL conical tube. 4. Very carefully add the 10 mL of blood/media mix to the FicollPaque. Hold the tube with the Ficoll-Paque at an angle of 45 and then using a sterile 3 mL aspirating pipette (Pastette) gently run the blood down the side of the tube so that it forms a separate layer above the Ficoll-Paque (see Note 7). 5. Spin the tube at 400  g for 20 min at room temperature using slow acceleration and slow deceleration (brake off). 6. Remove the tube. The red blood cells will collect at the bottom of the tube above which will be the Ficoll-Paque layer. The lymphocytes should form a layer on top of the Ficoll-Paque layer, while the top layer will be the medium. 7. Insert a clean sterile Pastette to just above the Ficoll-Paque layer and draw off the lymphocytes, which are placed in a sterile 15 mL conical tube. 8. Centrifuge the lymphocytes at 60–100  g for 4 min. 9. Carefully aspirate and discard the media.

Gene Editing for Corneal Stromal Regeneration


Fig. 4 Confirmation of the specificity achieved using a guide-specific system targeted to prevalent TGFBI mutations. In vitro digestion of either wild-type or respective mutant TGFBI sequence via Cas9 protein complexed with an sgRNA [8]


Tara Moore et al.

10. Wash the pellet of lymphocytes with RPMI media (20% FCS). 11. Centrifuge the lymphocytes at 60–100  g for 3 min. 12. Aspirate and discard the majority of the media. 13. Resuspend the pelleted lymphocytes in the residual media. 14. Rapidly thaw an aliquot of EBV at 37  C. 15. Add the thawed EBV suspension to the resuspended lymphocytes and mix gently. 16. Incubate for 1 hour at 37  C (infection period). 17. Add RPMI, 20% FCS media to EBV-treated lymphocytes to give a total volume of 3 mL. 18. Add 40 μL of 1 mg/mL phytohemagglutinin and mix gently. 19. Aliquot 1.5 mL of the lymphocyte mixture to 2 of the middle wells of a 24-well plate. 20. Add PBS to the surrounding wells of the plate to maintain humidity. 21. Incubate for 24 hours at 37  C with 5% CO2 in tissue culture incubator. 22. After 24 hours, the lymphoblastoid cells should be observed to be aggregating. 23. When the media begins to turn yellow, replace with fresh media and expand cells as appropriate into 6-well plates and small flasks. 24. Once in flasks, the serum content of the media can be reduced to 10%. 3.9 Nucleofection of LCLs with RNPs

1. RNPs are formed directly in the Lonza Nucleofector SF solution (SF Cell line 4D-Nucleofector X kit), and incubated for 10 min at room temperature. 2. Prepare the reagents according to the following order: Reagent


Synthego modified sgRNA (30 pmol)

2.66 μL

NEB EnGen SpCas9 (20 pmol)

2 μL

Lonza Nucleofector SF solution

20.34 μL

Total Volume

25 μL

3. Using a hemocytometer calculate the number of cells required per transfection and total number of cells required for the experiment. 4. Collect the total number of cells by centrifugation (300  g x 5 min) and resuspend in Nucleofector solution by gently pipetting (5 μL/per reaction).

Gene Editing for Corneal Stromal Regeneration


5. 5 μL of each cell solution was added to 25 μL of corresponding preformed RNPs, mixed and transferred to the nucleofector 16-well strip. 6. The cells were electroporated using the 4D Nucleofector program DN-100, 70 μL of pre-warmed media was added to each well and allowed to recover at room temperature for 5 min. 7. The transfected cells were then transferred to a 24-well plate containing 200 μL medium per well. 8. Incubate cells at 37  C for 48 hours. 9. 48 hours post nucleofection, extract gDNA from cells using the QIAamp DNA Mini Kit following the manufacturer’s instructions. 3.10 PCR for Tracking of Indels by Decomposition (TIDE)

1. Design primers flanking the target site. Primers should generate a ~700 bp product, the target site should be ~200 bp from the sequencing start site. 2. Perform a gradient PCR to determine the annealing temperature. 3. Set up a PCR reaction as follows, include a non-edited control: Reagents

1 Reaction

5 Dream Taq Buffer

4 μL

Forward primer 10 μM

1 μL

Reverse primer 10 μM

1 μL

DNA template

Plasmid DNA ¼ 1 pg to 1 ng, Human DNA ¼ 50–100 ng


Make up 20 μL

Dream Taq

0.08 μL

4. Electrophorese PCR products on a 1% agarose gel alongside a DNA ladder to confirm the PCR reaction has produced a specific band at the expected size. 5. Purify the PCR product and send test and control PCR products for Sanger sequencing. 6. Upload the Sanger traces to the TIDE online program to quantify indels in your test sample. 3.11 Intrastromal Injection of CRISPR/ Cas9 Constructs

1. Glass needles are prepared by pulling glass capillaries using program 17 on a DMZ Universal Puller. 2. Glass needles are fitted onto a Hamilton 10 μL syringe using a compression fitting.


Tara Moore et al.

3. Anesthetize animals by an intra-peritoneal injection of ketamine and xylazine. 4. Position mouse on custom mouse tooth holder under the surgical microscope. Rotate the head of the mouse to have a bird’s-eye view of the cornea. 5. Dilate pupils by topically applying one drop of tropicamide and phenylephrine on each eye. 6. 2 μg of CRISPR/Cas9 expression construct is back-filled into a glass needle (1.0 mm) attached to a Hamilton 10 μL syringe. 7. The glass needle is used to make a track in the stroma just above the periphery of the cornea, when the needle feels secure in the stroma inject the CRISPR/Cas9 expression construct. 8. Fusidic gel was applied topically following injection as an antibiotic agent. 3.12 IVIS In Vivo Imaging of Fluorescence in Mouse Eyes

1. Anesthetize mice using 1.5–2% isoflurane in ~1.5 L/min flow of oxygen (setting 2.5 on isoflurane control). 2. Mice are placed in a custom mouse holder tube (Fig. 2) and placed inside a Xenogen IVIS Lumina in vivo imager. 3. A custom program to detect fluorescence or luminescence is selected and the signal is quantified.


Notes 1. Remove media from wells and wash with 100 μL PBS. Then add 20 μL 1 Passive Lysis buffer to each well, shake gently for 15 min on a plate shaker. Finally insert into the LUMIstar Optima plate reader. 2. Cover the 15 mL tubes completely in tin foil as reagents are light sensitive. Put the reagents in the LUMIstar compartment and close lid. 3. Care needs to be taken to prevent tube mix up. 4. Pump priming icon is on the top row of buttons in the LUMIstar software. 5. Use Greiner 96 F-bottom black wall microplate with 36 intervals, reading direction #. Volume for each injection is 36 μL. 6. Template ¼ 587 bp – M.W of 356701.7 g/mol, require 3 nM. 7. Avoid excessive mixing of the blood into the Ficoll-Paque layer. 8. Flasks must be coated with 10 μg/mL laminin and 10 mg/mL chondroitin sulfate (acts like the Descemet’s membrane).

Gene Editing for Corneal Stromal Regeneration


References 1. Jinek M, Chylinski K, Fonfara I, Hauer M, Doudna JA, Charpentier E (2012) A programmable dual-rna-guided DNA endonuclease in adaptive bacterial immunity. Science (New York, NY) 337(6096):816–821. https:// 2. Mali P, Yang L, Esvelt KM, Aach J, Guell M, DiCarlo JE, Norville JE, Church GM (2013) Rna-guided human genome engineering via cas9. Science (New York, NY) 339 (6121):823–826. ence.1232033 3. Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, Hsu PD, Wu X, Jiang W, Marraffini LA, Zhang F (2013) Multiplex genome engineering using crispr/cas systems. Science (New York, NY) 339(6121):819–823. https:// 4. Ran FA, Hsu PD, Wright J, Agarwala V, Scott DA, Zhang F (2013) Genome engineering using the crispr-cas9 system. Nat Protoc 8 (11):2281–2308. nprot.2013.143 5. Hsu PD, Scott DA, Weinstein JA, Ran FA, Konermann S, Agarwala V, Li Y, Fine EJ, Wu X, Shalem O, Cradick TJ, Marraffini LA, Bao G, Zhang F (2013) DNA targeting specificity of rna-guided cas9 nucleases. Nat Biotechnol

31(9):827–832. 2647 6. Fu Y, Foden JA, Khayter C, Maeder ML, Reyon D, Joung JK, Sander JD (2013) Highfrequency off-target mutagenesis induced by crispr-cas nucleases in human cells. Nat Biotechnol 31(9):822–826. nbt.2623 7. Moore CBT, Christie KA, Marshall J, Nesbit MA (2018) Personalised genome editing - the future for corneal dystrophies. Prog Retin Eye Res 65:147–165. preteyeres.2018.01.004 8. Christie KA, Courtney DG, DeDionisio LA, Shern CC, De Majumdar S, Mairs LC, Nesbit MA, Moore CBT (2017) Towards personalised allele-specific crispr gene editing to treat autosomal dominant disorders. Sci Rep 7(1):16174. 9. Courtney DG, Moore JE, Atkinson SD, Maurizi E, Allen EH, Pedrioli DM, McLean WH, Nesbit MA, Moore CB (2016) Crispr/ cas9 DNA cleavage at snp-derived pam enables both in vitro and in vivo krt12 mutation-specific targeting. Gene Ther 23(1):108–112. https://

Chapter 7 Preparation and Administration of Adeno-associated Virus Vectors for Corneal Gene Delivery Liujiang Song, Jacquelyn J. Bower, and Matthew L. Hirsch Abstract Gene delivery approaches using adeno-associated virus (AAV) vectors are currently the preferred method for human gene therapy applications and have demonstrated success in clinical trials for a diverse set of diseases including retinal blindness. To date, no clinical trials using AAV gene therapy in the anterior eye have been initiated; however, corneal gene delivery appears to be an attractive approach for treating both corneal and ocular surface diseases. Multiple preclinical studies by our lab and others have demonstrated efficient AAV vector-mediated gene delivery to the cornea for immunomodulation, anti-vascularization, and enzyme supplementation. Interestingly, the route of AAV vector administration and nuances such as administered volume influence vector tropism and transduction efficiency. In this chapter, a detailed protocol for AAV vector production and specific approaches for AAV-mediated gene transfer to the cornea via subconjunctival and intrastromal injections are described. Key words Adeno-associated virus (AAV), Cornea, Gene delivery, Subconjunctival injection, Intrastromal injection, Purification, Titering, Purity


Introduction Adeno-associated virus (AAV) is a small single-stranded DNA virus initially discovered in 1965 as a contaminant of an adenovirus preparation [1, 2]. In the 1980s, it was reported that all AAV viral coding sequences can be substituted with transgenic DNA and packaged in the AAV capsid, thus establishing the field of AAV gene therapy [3, 4]. To date, at least 12 naturally occurring serotypes and hundreds of variants have been isolated via tissue mining experiments in various species [5]. In general, AAV serotypes demonstrate altered tropisms which utilize various cell surface entry receptors or co-receptors [5–17]. Although the exact mechanism of cellular entry, nuclear entry/trafficking, and virion uncoating remains unknown, reports have demonstrated that desirable physiological properties can be generated by random or rational capsid

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Liujiang Song et al.

mutagenesis, including restricted tissue transduction or enhanced whole body transduction [5, 18–22]. Over the past few decades, several AAV production systems have been developed that yield high viral titers of transduction competent AAV vectors for both preclinical and clinical applications. Producer cell lines include the mammalian HEK293 cell system [4, 23] (adherent and serum free suspension), Vero cells (Herpes Simplex Virus-based platform), and insect cell lines such as Sf9 cells (Baculovirus-based system) [24–27]. Each packaging system features scale-up options ranging from several dishes/flasks to large-scale bioreactors (such as the iCellIS 500 m2 bioreactor or the Abec 4000-L bioreactor) [28]. Ion exchange/affinity chromatography is the most commonly used purification technique for highpurity clinical-grade AAV manufacturing [25, 29], whereas gradient centrifugation (such as cesium chloride or iodixanol gradient) is still the major method for general laboratory-scale AAV vector purification [30, 31]. AAV titering methods include genome characterization and quantification via quantitative PCR (qPCR), droplet digital PCR (ddPCR), Southern dot blot, and alkaline gel electrophoresis. AAV capsids are characterized by Western dot blots, ELISA, electron microscopy, and silver staining conveying the purity of the preparation [32]. Among the different production methods, a 30-year-old triple transfection protocol using adherent HEK293 cells remains the most common and reproducible method to produce small-scale AAV preparations for preclinical studies [33]. A combination of iodixanol gradient centrifugation and ion exchange chromatography utilizing prepacked columns is the most widely used AAV purification strategy to ensure sufficient purity. When used in combination, these methods consistently and efficiently produce high yields of AAV vectors for preclinical use in corneal gene delivery applications [34, 35]. The eye is a particularly attractive gene therapy target, due to its unique anatomic accessibility and presumed immunological privilege. In fact, AAV has taken center stage as the gene therapy of choice for many ocular diseases; the most notable of these is Luxturna (voretigene neparvovec, STN:125610, Spark Therapeutics, Inc.), which was the first commercially available gene therapy drug approved by the FDA in the United States to treat a rare disease of the posterior eye [36, 37]. Its success has encouraged the gene therapy community to expand AAV applications to multiple ocular diseases including those of the cornea, which is the anterior transparent avascular tissue that acts as the major refractive surface of the eye. It is naturally organized into 3 general layers: (1) the anterior epithelia, (2) the relatively large central stroma region composed of precisely aligned collagen lamellae interspaced primarily by quiescent keratocytes, and (3) a posterior single layer of generally nondividing endothelial cells. For corneal diseases, allogeneic cornea transplantation is a commonly employed strategy for vision

Preparation and Administration of AAV for Corneal Gene Delivery


restoration. However, a lack of suitable transplantation-grade donor corneas and immunologic rejection remain important general obstacles for this treatment [38, 39]. Alternatively, several in vivo animal studies have demonstrated efficient AAV gene delivery to the cornea for the correction of genetic or acquired corneal diseases. For instance, AAV encoding an endostatin, decorin, or angiostatin gene were reported to decrease or reverse corneal fibrosis after injury or infection [35, 40–42] and prevent corneal vascularization [35, 43–45]. Furthermore, AAV-mediated immunomodulatory gene expression (such as HLA-G) may re-establish tolerance in ocular surface immune-mediated diseases [35], and delay/prevent rejection of allogeneic, and perhaps even xenogeneic, corneal transplants. Moreover, gene addition strategies may prevent corneal opacity in children with hereditary lysosomal storage diseases and have also been utilized for the treatment of Fuchs’ dystrophy [46, 47]. AAV serotype specificity and the route of administration play key roles in AAV vector biodistribution and transgene expression levels; thus, identification of the optimal serotype and route for AAV delivery is a critical step for successful corneal gene therapy. Currently, topical [40, 41, 45, 48–53], intrastromal [46, 54], subconjunctival [34, 43, 44], and intracameral [55–59] delivery routes have been utilized for AAV gene delivery to the cornea (Fig. 1). While all of these administration routes are approved for clinical use, some are not necessarily commonplace, and each delivery route exhibits its own inherent potential complications. For example, topical administration appears to be the most attractive route of administration due to its minimal invasiveness. However, without corneal epithelial cell removal, AAV transduction via topical eye drops is minimal [48, 49, 60]. Corneal intrastromal and intracameral injections of AAV vectors have been well characterized and demonstrate impressive serotype-dependent transduction of

Fig. 1 Administration routes for corneal drug delivery. 1. Topical; 2. intrastromal; 3. intracameral; 4. subconjunctival


Liujiang Song et al.

different anterior eye compartments [46, 54–59]. In addition, our lab has recently demonstrated corneal transduction following a simple and safe subconjunctival injection [34]. Although serotype dependence has not been completely characterized for each route of administration, the choice of the best serotype will be diseasedependent, as different cells can be targeted even within an optimal administration route. This chapter provides a detailed description of essential methods for AAV production in adherent HEK293 cells by triple plasmid transfection using polyethylenimine (PEI), purification by iodixanol density gradient ultracentrifugation and ion exchange chromatography, titration by qPCR, and particle characterization including genome integrity and vector purity for corneal gene delivery (Fig. 2). Furthermore, protocols for AAV administration via subconjunctival and intrastromal injections are also described.

Fig. 2 Diagram of the major steps in AAV preparation

Preparation and Administration of AAV for Corneal Gene Delivery



Materials (See Note 1)

2.1 Vector Production

1. HEK293 cells (ATCC: CRL-1573). 2. AAV production plasmids: pXX680, pXR2, pAAV-transgene (These plasmids are commercially available at the University of North Carolina at Chapel Hill Vector Core Facility or other companies). 3. T-150 tissue culture dishes. 4. DMEM medium (with L-glutamine, high glucose, and sodium pyruvate) supplemented with 10% heat-inactivated fetal bovine serum (FBS), and 100 U/mL penicillin/100 μg/mL streptomycin (for cell culture). 5. DMEM medium transfection).






6. Phosphate-buffered saline (PBS). 7. PEI (linear, MW 25 kD). Dissolve in PBS at 1 mg/ml, adjust the pH between 4~5 and filter sterilize. 8. 0.05% trypsin/EDTA. 9. 50 mL polystyrene conical tubes. 10. 250 mL centrifugation bottles. 11. Cell scrapers. 12. Serological pipettes. 13. 37  C water bath. 14. Dry ice/ethanol bath. 15. Vortex mixer. 16. Benzonase nuclease. 17. 4.8 M MgCl2. 2.2

AAV Purification

1. 60% Iodixanol: OptiPrep Density Gradient Medium. 2. 4.8 M MgCl2. 3. 2.5 M KCl. 4. 0.22 μm filtered Milli Q H2O. 5. 5 M NaCl. 6. 5 mL syringes with 21 G needles. 7. OptiSeal centrifuge tubes. 8. Tube rack. 9. Ultracentrifuge. 10. Syringe pump. 11. Prepacked strong ion-exchange columns.


Liujiang Song et al.

12. Buffer A: 20 mM Tris, pH 9.0. 13. Buffer B: 20 mM Tris, pH 9.0, 2 M NaCl. 14. Buffer C: 20 mM Tris, pH 9.0, 200 mM NaCl. 15. Dialysis buffer: PBS with 50 g/L D-sorbitol, 210 mM NaCl. 16. Slide-A-Lyzer dialysis cassette. 1. 0.5 M EDTA.

2.3 AAV Titration by qPCR

2. DNase digestion solution: 10 mM Tris (pH 7.5), 10 mM MgCl2, 200 mM CaCl2, 100 μg/mL DNase I. 3. Proteinase solution: 100 μg/mL Proteinase K, 5 mM NaCl, and 1% Sarkosyl. 4. Molecular-grade H2O. 5. 10 mM Tris, pH 8.0 6. Standard plasmid, such as a maxi preparation of pAAVtransgene plasmid used to generate the AAV vector, or an AAV reference standard stock from ATCC. 7. LightCycler® 480 SYBR Green I Master Mix. 8. Primers designed for the specific AAV vector, such as the following primer sets are designed for an AAV2 vector with a GFP transgene, a CMV promoter, and an SV40 polyA tail.


Primer ID

Sequence (50 !30 )










SV40 polyA-F


SV40 polyA-R


CMV promoter

SV40 polyA

9. Gilson Pipetman. 10. Pipette tips. 11. qPCR 96-well plates and plate sealer sheets. 12. LightCycler 480 qPCR machine.

Preparation and Administration of AAV for Corneal Gene Delivery

2.4 AAV Titering by Dot/Slot Blot


1. 100 mM NaOH. 2. Benzonase nuclease. 3. ddH2O. 4. Church buffer: 1% (w/v) bovine serum albumin, 1 mM EDTA (pH 8.0), 0.5 M phosphate buffer (0.5 M phosphate buffer is 134 g of Na2HPO4 · 7H2O, 4 mL of 85% H3PO4 (concentrated phosphoric acid), H2O to 1 L), 7% (w/v) SDS. 5. Bio-Rad thin absorbent filter papers or Whatman 3MM filter papers. 6. Positively charged nylon transfer membrane. 7. Maxi preparation of pAAV-transgene plasmid used to generate the AAV vector standard plasmid or AAV reference standard stock from ATCC. 8. 96-Well dot-blot apparatus. 9. Stratagene Stratalinker 1800 UV Crosslinker. 10. Probe template (such as a restriction enzyme digested fragment of the targeted vector). 11. Random Primer DNA Labeling Kit. 12. 20 SSC (saline-sodium citrate): 3 M NaCl, 0.3 M sodium citrate, pH 7.0. 13. Low salt buffer: 0.1 SSC, 0.1% SDS. 14. High salt buffer: 2 SSC, 0.1% SDS. 15. Hybridization bottle. 16. X-ray film. 17. 8  10 in. autoradiography cassette.

2.5 scAAV Genome Integrity by Alkaline Gel Electrophoresis

1. 1 M NaOH. 2. Benzonase nuclease. 3. ddH2O. 4. 20 mM EDTA. 5. 1 kb DNA marker. 6. 10 alkaline gel buffer: Add 50 mL of 10 N NaOH and 20 mL of 0.5 M EDTA (pH 8.0) to 800 mL of H2O and then adjust the final volume to 1 L. 7. 1 TAE buffer: 40 mM Tris-acetate, 1 mM EDTA, pH 8.3. 8. 6 Alkaline Gel-loading Buffer: 300 mM NaOH, 6 mM EDTA, 18% (w/v) Ficoll, 0.15% (w/v) bromocresol green, 0.25% (w/v) xylene cyanol. 9. General electrophoresis agarose powder. 10. 37  C water bath.


Liujiang Song et al.

11. 55  C water bath. 12. SYBR™ Gold Nucleic Acid Gel Stain. 2.6 Determination of AAV Purity

1. NuPAGE 4–12% Bis-Tris gels. 2. 2-Mercaptoethanol. 3. NuPAGE™ Sample Reducing Agent (10). 4. NuPAGE™ LDS 4 Loading buffer. 5. Silver staining kit. 6. Protein marker with broad range allowing for the determination of the AAV capsid protein sizes (87, 72, and 62 kDa). 7. 10 SDS-PAGE Running Buffer. 8. SDS-PAGE gel electrophoresis apparatus.

2.7 Corneal Injections for Mouse Models

1. Mouse model of choice. 2. 1% Fluorescein solution, filter sterilized 3. 0.5% Proparacaine hydrochloride eye drops 4. GenTeal ocular lubricant gel. 5. 1% Tropicamide 6. Oxygen (tank). 7. Isoflurane vaporizer and induction chamber set. 8. Mouse nose cone for anesthesia. 9. Standard Infuse/Withdraw Programmable Syringe pump. 10. Polyethylene tubing, I.D. 0.38 mm, O.D. 1.09 mm. 11. 10 μL Hamilton gastight removable needle syringe 12. Beveled 36 G needles. 13. Syringes with 27 G needle (O.D. about 0.41 mm). 14. Stereoscopic microscope. 15. Finely pointed precision forceps. 16. Heating pad. 17. Parafilm.



3.1 Vector Production

1. Culture HEK293 cells to a cell density of 80–90% confluence. (see Note 2). 2. Calculate the amount of each plasmid needed for a 1:1:1 molar ratio with a total mass of 62.4 μg/plate. For example, the following volumes are calculated for ten T-150 plates for packaging of AAV2-CMV-EGFP vector:

Preparation and Administration of AAV for Corneal Gene Delivery


(a) The length of the pXX680, pXR2, and pAAV-CMVEGFP is 18,322 bp, 7462 bp, and 5848 bp, respectively. Thus, the number of total base pairs will be: 18,322 + 7462 + 5848 ¼ 31,632 bp (b) Calculate the μg/bp: 0.0019769 μg/bp.





(c) Mass needed for each plasmid for 10 plates: l



pXX680: 0.0019769 μg/bp  18,322 bp  10 ¼ 361.435508 μg pXR2: 0.0019769 μg/bp  7462 bp  10 ¼ 147.516278 μg pAAV-CMV-EGFP: 0.0019769 μg/bp  5848 bp  10 ¼ 115.609112 μg

(d) Then according to the plasmid concentration, determine the volume needed for each plasmid. Assuming each plasmid is approximately 1 μg/μL, then a total volume of 624 μL containing all three plasmids will be used for the transfection of 10 plates. (e) If transfecting more/less plates, simply scale the reagents up/down for the total number of desired plates (see Notes 3 and 4). 3. Add the calculated amounts of DMEM (without FBS or antibiotics) and plasmid DNA to a tube and mix well. Then add the PEI to the DNA/DMEM solution, and vortex briefly to mix (see Notes 5–8). Regent

Volume (mL) for 10 plates


1.25 mL


0.624 mL


8.126 mL


10 mL

4. Incubate the DMEM/DNA-PEI solution for 10 min at room temperature to allow complex formation. 5. Carefully add 1 mL (per plate) of the DMEM/DNA-PEI complexes dropwise to the cells. 6. Return the cells to the 37  C/5% CO2 incubator, incubate 4–8 h (or overnight). Cells will be more viable and give better yields if the media is changed 4–8 h post transfection. 7. Scrape the dishes to harvest transfected cells at 72 h post transfection (see Note 9).


Liujiang Song et al.

8. Spin at 500  g and 4  C for 10 min. Discard the supernatant (see Note 10). 9. Resuspend the cell pellet in ~5 mL of autoclaved MilliQ H2O. 10. Freeze/thaw by placing the resuspended pellet in a dry ice/ethanol bath and subsequently thawing at 37  C for 10 min. Vortex the sample aggressively. Repeat the freeze/ thaw/vortex steps two additional times (the crude lysate may be stored at 80  C for an extended period of time at this point in the protocol). 11. Sonicate the sample on ice for 2 min. 12. Add 1 μL of 4.8 M MgCl2 (a final concentration of 1 mM), and 20 U of Benzonase (about 2 U of Benzonase for each T-150 plate harvested). Vortex and incubate at 37  C for 1 h. 13. Spin down at 2000  g at 4  C for 20 min and collect the supernatant. 3.2

AAV Purification

1. Prepare iodixanol gradient as shown in the table below (see Note 11): Components





Iodixanol 60% stock

13.5 mL

21 mL

33.3 mL

50 mL

1 M Tris–HCl, pH 9.0

1.25 mL

1.25 mL

1.25 mL


Autoclaved MilliQ H2O

25 mL

27.5 mL

15.3 mL


5 M NaCl 1 M MgCl2 2.5 M KCl

10 mL 0.1 mL 0.1 mL

0.1 mL 0.1 mL

0.1 mL 0.1 mL

0.1 mL 0.1 mL

2. In a biosafety cabinet hood, load the iodixanol gradient stepwise with a glass Pasteur pipette and a pipette aid into a QuickSeal tube (29.9 mL capacity) in the following order, mark the 40% and 60% interface with a “+” (see Note 12): (a) 6 mL of 17% iodixanol (b) 6 mL of 25% iodixanol (c) 7 mL of 40% iodixanol (d) 6 mL of 60% iodixanol (e) Crude lysate (at the top) 3. Spin in an ultracentrifuge with an appropriate rotor at 504,000  g (70,000 rpm in a Beckman Coulter Ti70 rotor, for example) for 1 h at 18  C. 4. Collect the fraction at the 40–60% iodixanol interface by inserting a 5 mL syringe with an 18 G needle (bevel up) between the 40% and 60% gradient fractions. Take ~ 4–5 mL of the fraction (see Note 13).

Preparation and Administration of AAV for Corneal Gene Delivery


5. Transfer each viral particle fraction obtained from an iodixanol gradient/Quickseal tube into an individual 50 mL conical tube and bring the final volume of all tubes to 12 mL with Buffer A. 6. Assemble the ion exchange apparatus by placing a syringe on the pump and connecting the syringe to the prepacked column using the appropriate tubes. 7. Load the following buffers to the syringe in succession and collect the fractions eluted with Buffer C: (a) 5 mL of 0.5 M NaOH (set speed at 1 mL/min) (b) 5 mL of Buffer B (set speed at 1 mL/min) (c) 10 mL of Buffer A (set speed at 1 mL/min) (d) 12 mL of sample (diluted in buffer A) (set speed at 0.5 mL/min) (e) 10 mL of Buffer A (f) 6 mL of Buffer C 8. Collect the fraction eluted with Buffer C into multiple tubes, the first tube 0.5 mL, the second tube 1 mL, the third tube 1 mL, then 0.5 mL for each tube until finish the collection (see Notes 14). 3.3

Titration by qPCR

1. Add 10 μL of virus to 90 μL of the DNase I digestion solution. 2. Incubate for 1 h at 37  C to digest any DNA remaining outside of the viral particles. 3. Add 6 μL of 0.5 M EDTA and vortex to inactivate the DNase I. 4. Add 120 μL of Proteinase K (ProK) solution to the DNasetreated AAV samples to digest the capsid and mix. 5. Incubate at 55  C for 2 h to digest the viral particles and release the AAV DNA. 6. Incubate the sample for 10 min at 95  C to inactivate the ProK. 7. Dilute the sample 100-fold in molecular-grade water or 10 mM Tris (pH 8.0) for qPCR. 8. Prepare the standard plasmid by making serial 1:10 dilutions with 10 mM Tris (pH 8.0) at concentrations ranging from 50 pg/μL to 0.05 fg/μL (see Notes 15 and 16). 9. Plan the sample layout for the 96-well qPCR plate. Negative controls and plasmid standards should be assayed at least in duplicate. To minimize the risk of cross-contamination, assign the plasmid standard to the end of the plate only after adding all of the samples (Fig. 3). 10. Determine the number of PCR reactions to be performed including plasmid control dilutions, samples, negative controls, and add additional 10% to the sample number in order


Liujiang Song et al.

Fig. 3 qPCR plate layout template

to calculate the total number of reactions (N) required for the master mix solution. 11. Dilute the primer to a concentration of 20 μM (see Note 17). 12. Set up the qPCR SYBR master mix reactions (N ¼ total number of reactions): Reagent

Volume (μL)

2 SYBR mix

5 μL  N

Forward Primer

0.25 μL  N

Reverse Primer

0.25 μL  N


2.5 μL  N

13. Load 8 μL of the master mix described above into each well (see Note 18). 14. Pipette 2 μL of DNA (sample, standard, H2O, or vehicle control) into each well. 15. Seal the plate with the adhesive plate seal.

Preparation and Administration of AAV for Corneal Gene Delivery


16. Spin down the plate briefly. 17. Cycle in the qPCR thermocycler. Set up the cycling conditions according to the primer sets used and the manufacturer’s instructions. For example, the GFP primer set listed in the materials uses the following cycling conditions: Step 1

95  C for 10 min (1 cycle)

Step 2

95  C for 10 s 58  C for 10 s (40 cycles) 72  C for 10 s

Step 3

72  C for 10 s (1 cycle)

Step 4

Melting curve analysis

18. Calculate the viral titer (see Note 19). (a) Perform the data analysis using the instrument’s software: (1) input each sample name and the plasmid standard mass (fg) in the “Sample Editor,” (2) choose the “Ab Quant/2nd derivative max” and click “calculate,” (3) the software will automatically generate the standard curve and corresponding mass for each sample from their respective Ct values. (b) Determine the copy number in each well: Copy number ¼

fg  6:022  1023 650  size of plasmidðbpÞ  1015

Average weight of a DNA base pair ¼ 650 Da. Avogadro’s number ¼ 6.022  1023 molecules/mol. (c) Determine the sample volume (vol) dilution factor (DF) in steps 2–7: DF ¼

starting volume þ DNAse mix þ EDTA þ Prok mix starting volume  dilution in step 7

(d) Determine the virus titer: For self-complementary AAV: Titer ðvg=mLÞ ¼

Copy number  DF  103 The final sample vol added into each well in step 14 For single-stranded AAV:

Titer ðvg=mLÞ ¼

Copy number  DF  2  103 The final sample vol added into each well in step 14


Liujiang Song et al.

3.4 Titration by Dot Blot

1. Add 40 μL of ddH2O to 10 μL of virus. 2. Aliquot 100 ng plasmid and bring the volume to 50 μL. 3. Add 0.2 μL of Benzonase to both the virus and plasmid samples to generate the functional Benzonase control sample. Incubate at 37  C for 1 h. 4. Aliquot another 100 ng control plasmid and bring the volume to 50 μL to generate the control plasmid sample. 5. Add 50 μL of 100 mM NaOH to each sample. 6. Incubate at 65  C for 30 min, then chill on ice for 5 min. 7. Cut the filter paper and nylon membrane to the size of the slot blot apparatus. 8. Mark a corner to orient the blot using a pencil. 9. Rinse the membrane in ddH2O for 30 min at RT. 10. Soak the membrane in 10 SSC (diluted from 20 SSC) for 10 min at RT. 11. Rinse three sheets of filter paper in 10 SSC. 12. Set up the slot blot apparatus. 13. Place three sheets of filter paper on the bottom and place the membrane on top of the filter paper. Connect to a vacuum flask. 14. Add 100 μL of 10 SSC to the slot blot apparatus to check the vacuum leak. 15. Add 100 μL of 20 SSC to each of the samples + 1 μL of 6 DNA loading dye to visualize sample loading. 16. Transfer all samples and plasmids into a 96-well plate. Add 100 μL of each sample onto the blot membrane using a multichannel pipette. 17. Add 100 μL of 10 SSC to each sample, mix. 18. Take 100 μL of the diluted sample to the next well on the membrane. 19. Repeat steps 17 and 18 until you run out of wells at the end of the slot blot apparatus. 20. Remove the membrane and insert it into the UV Crosslinker. Use the following settings: “Energy,” “Optimal Crosslink,” and “Start.” The UV crosslinker will automatically stop when finished. 21. Place the membrane in a hybridization bottle with the side containing the crosslinked DNA facing toward the center of the bottle. 22. Add 15 mL of Church Buffer, tightly seal the bottle, and place it in a rotating hybridization incubator at 68  C for at least 1 h.

Preparation and Administration of AAV for Corneal Gene Delivery


23. Make a radioactively labeled probe following the manufacturer’s instructions included in the Random Primer DNA Labeling Kit (see Note 20). 24. Add the probe to the Church buffer in the hybridization bottle. 25. Hybridize at 68  C for overnight. 26. Wash the membrane with high salt buffer for 10 min at 68  C. Repeat this step with a fresh buffer change. 27. Wash with low salt buffer for 30 min at 68  C. 28. Wrap the blot with plastic wrap, ensuring that no air bubbles are trapped between the blot membrane and the wrap. Place it into the film cassette and tape the membrane to keep it in position. Scan with the radioactivity meter to roughly get a sense about the signal strength of radioactivity. 29. Add a film over the membrane in the dark room. Fold and then unfold a corner of the film to mark the orientation of the dot blot membrane. 30. Leave the cassette tightly closed in a drawer to protect from light and decide approximately how long to wait to develop the film according to the signal strength obtained at step 28. The wait times can range from several minutes to days. One film exposed for about 30 min may be used to obtain a rough idea of the signal strength and then a decision can be made as to whether a longer or shorter wait time is necessary for a second film exposure. Several exposures may be needed to obtain the optimal exposure quality. 31. Scan the exposed film to assess each sample’s corresponding mass compared to the plasmid standards. 32. Calculate the titer using the same formula/method in the qPCR titering section. 3.5 Vector Genome Integrity by Alkaline Gel Analysis

1. Aliquot 10 μL of purified virus, and add 10 μL of ddH2O. 2. Add 0.2 μL of Benzonase and incubate at 37  C for 1 h. 3. Add 1 μL of 1 M NaOH. 4. Incubate at 65  C for 30 min, then chill on ice for 5 min. 5. Add 1.3 μL of 20 mM EDTA, 2.5 μL ddH2O, and 5 μL of 6 alkaline gel-loading buffer to the virus sample. 6. Add 1.5 μL of 1 M NaOH and 1.5 μL of 20 mM EDTA to 5 μL of the DNA Marker and 22 μL of ddH2O. 7. Prepare the agarose solution by adding 2 g of agarose powder to 180 mL of ddH2O, and boil it in the microwave until agarose has dissolved and the solution is clear (approximately 2–3 min).


Liujiang Song et al.

8. Cool the agarose solution to 55  C by placing the flask in 55  C water bath. 9. Add 20 mL of 10 alkaline gel buffer. 10. Perform the electrophoresis using 1 Alkaline gel buffer (Dilute the 10 alkaline buffer with ddH2O to generate a 1 working solution immediately before use) at 35 V overnight. 11. Stain the gel using SYBR™ Gold Nucleic Acid Gel Stain in 1 TAE buffer (see Note 21). 3.6

Silver Staining

1. Aliquot 10 μL of virus. 2. Add 5%, (v/v) 2-Mercaptoethanol to the 4 loading buffer (make sure that a fresh solution of 2-Mercaptoethanol is used). 3. Add 4 loading buffer mix and 1/10 volume of NuPAGE™ Sample Reducing Agent to the sample. 4. Boil the samples for 10 min. 5. Load the samples onto the SDS-PAGE gel. 6. Start the denaturing gel electrophoresis at 120 V for 1 h using 1 SDS-PAGE Running Gel Buffer. 7. Perform the silver staining following the manufacturer’s instructions included in the kit.



1. Set up the injection stage as shown in Fig. 4a in a biosafety cabinet. A simple injection stage consists of a stereo microscope, a syringe pump, and the anesthesia system. 2. Cut the polyethylene tubing at a length as needed (usually around 50 cm is convenient for operation). 3. Insert the 36 G needle shank to the polyethylene tubing. 4. Fill a syringe with sterile water, manually flush the tubing/ needle to ensure no leaks, clogs, or damage throughout the tubing. 5. Flush 70% EtOH through the tubing to disinfect. 6. Use a syringe to evacuate with air and rinse with water. 7. Fill the tubing with sterile water using a syringe and pull the syringe out slowly to create an excess drop of water on a parafilm at the tubing opening; leave the opening in the water drop (see Note 22). 8. Fill a Hamilton syringe and eject some water out to ensure there is no air in the syringe. Place the filled Hamilton syringe needle tip in the drop of water on the parafilm and insert it into the water filled tubing. 9. Once the tubing is connected with the Hamilton syringe (Fig. 4b), gently place the Hamilton syringe on the microinjection pump.

Preparation and Administration of AAV for Corneal Gene Delivery


Fig. 4 Injection stage setup. A simple injection stage consists of a stereo microscope, a syringe pump, and the anesthesia system (a). A Hamilton syringe and a 36G needle connected with polyethylene tubing (b). An anesthesia nose mask designed for a tight fit to the mouse’s face and allowing access to the eyes (c). Introducing a small air bubble in the tubing to create a gap between the sterile water and the virus (d)

10. Eject the water out in the Hamilton syringe (the tubing is still filled with water), pull back the Hamilton syringe, and introduce a small air bubble in the tubing/needle.


Liujiang Song et al.

Fig. 5 Assessment of the subconjunctival injection. Microscopic view of AAV vector administration to the murine subconjunctival space. One percent fluorescein is added directly to the 70 μL AAV vector preparation to allow visualization of bleb formation during the procedure using an operating microscope

11. Place the needle tip into AAV vector and withdraw virus. There should be a visible air bubble remaining between the virus and the water in the tubing (Fig. 4d, see Note 23). 12. Allow the tubing/needle to sit at room temperature for 10 min with viral vector solution to allow the saturation of potential binding of virus on the wall of the needle and/or tubing; discard the virus. 13. Place needle tip in the viral prep and withdraw desired amount of vector. 14. Set the syringe pump at a rate of 10 μL/min (for the subconjunctival injection) or 2 μL/min (for intrastromal injection) (see Note 24). 15. In an anesthesia chamber, anesthetize the mouse with 2.5% isoflurane and an oxygen flow rate of 1 L per min until the mouse reaches a state of unconsciousness. 16. Transfer the mouse from the anesthesia chamber to the surgical pad, and place a nose cone (Fig. 4c) over the mouse’s nose to maintain anesthesia. 17. Apply 0.5% proparacaine hydrochloride eye drops topically to the mouse’s eyes. 18. Use the forceps to expose the eyeball and immobilize the conjunctiva. 19. For a subconjunctival injection, use the dominant hand to hold the needle with the bevel facing upward. Insert the needle into the conjunctiva (Fig. 5). For an intrastromal injection, hold the needle horizontally with the bevel facing upward. Penetrate the cornea at a distance of approximately one-third of the corneal radius as measured

Preparation and Administration of AAV for Corneal Gene Delivery


from the temporal limbus. Allow the needle tip to remain in the stroma layer (see Notes 25–29). 20. Start the injection by using the footpad to control the pump (see Note 30). 21. Hold the needle in place for at least 10 s before removing the needle from the conjunctiva. 22. Put a drop of the GenTeal lubricant gel on the mouse’s eyes and then place the mouse on a heating pad to recover.


Notes 1. Most of the kits, regents, or apparatuses listed in the materials are commercially available from several different suppliers. Choose cost-effective and consistent ones according to each individual’s preferences. 2. Always keep good track of the passage number of the HEK293 cells and use low passage cells to avoid potential cell line passage effects. 3. The molar ratio of the three plasmids used for the triple transfection protocol may vary among labs [30, 32, 61]. 4. The total mass of plasmid used for transfection of one T-150 plate ranges from 28 μg to 120 μg, all of which are reported to produce good AAV yields [32]. 5. The plasmid:PEI ratio is approximately 1:2. If less or more total mass of the plasmid/plate is used, the amount of PEI should also be adjusted accordingly. 6. PEI with MW of 25,000 and PEI-Max with MW of 40,000 both have been reported to produce good yields of AAV [30]. 7. It is recommended to do a triple transfection with a fluorescent reporter gene in parallel as a transfection positive control when packaging a non-reporter gene especially for a beginner. 8. AAV package capacity, remains unknown, however appears to be limited to ~5 kb (for single-stranded AAV) [62, 63] and ~2.3 kb [64, 65] (for self-complementary AAV) in length, respectively. Vector constructs that attempt to exceed this length are most commonly packaged as sub-genomic fragments [66]; however, successful capsid packaging of larger fragments were demonstrated by several groups [67, 68]. 9. After transfection, the media can be changed after 4–8 h to increase cell viability and virus yield. The total incubation time can vary anywhere from 2 to 6 days. 10. AAV will be present in the cells and in the medium in a serotype-dependent fashion. For example, the majority of


Liujiang Song et al.

AAV2 particles is reported to be inside the cells, whereas the majority of AAV1 and AAV8 particles are present in the supernatant [69]. The supernatant could also be collected by PEG precipitation and then purified using the same protocol [70]. 11. When preparing the 25% and 60% gradient fractions, it is advisable to add a few drops of phenol red to help with visualization of the fractions at later steps. 12. Always prepare the gradient very carefully and avoid disturbing the gradient. When handling multiple samples, a variable speed peristaltic pump may be used to speed up loading of the iodixanol gradient, and to minimize the loading differences between tubes to ensure tubes are balanced and gradient layers are undisturbed. 13. One critical step for the iodixanol purification is to aspirate the correct layer from the 40%/60% gradient boundary after ultracentrifugation. For this purpose, gently and slowly collect the 40%/60% boundary, avoiding the use of too much force. When the top part of the 40% fraction is reached during collection with the syringe, turn the bevel down to minimize the risk of collecting contaminants such as the empty capsid. Do not collect the top of the 40% fraction which contains cellular protein and empty particles. 14. The second and third tube is usually where the majority of the viral particles are located. Dialysis can be performed at 4  C for overnight to exchange the buffer. Do not leave the virus in buffer C long term; its high pH value may lead to AAV vector instability. 15. For the standard plasmid serial dilutions, it is highly recommended that the plasmid DNA obtained from a Maxipreparation be diluted to 10 ng/μL initially in 10 mM Tris (pH 8.0), aliquoted, and stored at 20  C, as this will ensure consistency in the plasmid stock used to generate the standard curve. Use one aliquot for each viral titration experiment and avoid repeat usage of the DNA once it has been thawed. 16. AAV reference stock standard materials from ATCC (Cat#: VR-1616 for AAV2 and VR-1816 for AAV8) [71] or previously titered AAV may also serve as qPCR standards. However, to ensure consistency among batches, use the same reference materials for each viral titration experiment and record the standard Ct values each time. 17. Different primer sequences and position of the target amplicon within the vector genome may influence titers. It is recommended to use 3 sets of primers for genome quantification (50 , central, and 30 ) to reduce concerns of incomplete genome packaging and use the same primer set for different batches of the same vector to minimize concerns regarding primer biases.

Preparation and Administration of AAV for Corneal Gene Delivery


18. To minimize pipetting errors, an electronic repeating pipettor can be used to dispense the master mix into the wells. 19. Always check that the Ct value obtained for the control virus or plasmid is within the previously obtained range. Always check the PCR efficiency, melting curve, and the quality of the standard curve to ensure the quality of the qPCR assay. 20. Nonradioactive labeling kit is also available such as digoxigenin DNA labeling if radioactive probe is not preferred. 21. SYBR Gold is sensitive enough to detect 1010 viral genomes. If the viral genome titer is less than 109, skip the SYBR Gold staining step and proceed to southern hybridization steps following standard protocol. 22. It is critical to minimize the amount of air present in the injecting syringe/tube/needle to ensure accuracy of the injection volume. 23. As an optional step, mixing the virus with sodium fluorescein (final conc. 0.01%) is helpful for visualizing the virus/air bubble interface during injection and monitoring AAV distribution and/or any leakage after injection (Figs. 5 and 6). 24. When setting up the syringe pump parameters, it is very important to select the right syringe diameter and then set the speed of injection. 25. For intrastromal injection (Fig. 5), 1~2 μL for mouse cornea is well tolerated. One of the major difficulties of intrastromal injections of a mouse cornea for beginners is endothelial

Fig. 6 Assessment of the intrastromal injection. Corneal optical coherence tomography (OCT) (left panel) and fluorescence imaging (right panel) using a Micron IV (Phoenix Research Labs, Pleasanton, CA, USA) following a 1 μL of intrastromal injection of AAV vectors in a 0.1% fluorescein/PBS solution


Liujiang Song et al.

perforation due to the small thickness of the cornea and the size/design of commercially available needles. 26. A 1~2 μL intrastromal injection results in a high distribution area in mice (Fig. 6) [72], which is visible as a temporary corneal plaque. This is usually completely resolved within 24 h post-injection. 27. Subconjunctival injection has less of a restriction on the administered volume (1~100 μL for a mouse). However, volume differences may play a role in AAV biodistribution and transduction. In addition, larger volume of injection was reported to be used to create the conjunctival scarring model [68]. 28. Upon subconjunctival injection, a bleb may appear (Fig. 5). It is usually completely resolved a few hours post injection. 29. A Micro IV imaging system (Phoenix Research Labs) could be used to monitor the resolution of the injected solution. 30. The movement of the air should be synchronized with the movement of the Hamilton plunger. A delay indicates excess air in the injecting system or loose connection between the tubing/needle/syringe. References 1. Atchison RW, Casto BC, Hammon WM (1965) Adenovirus-associated defective virus particles. Science 149(3685):754–756 2. Atchison RW, Casto BC, Hammon WM (1966) Electron microscopy of adenovirusassociated virus (aav) in cell cultures. Virology 29(2):353–357 3. Hermonat PL, Muzyczka N (1984) Use of adeno-associated virus as a mammalian DNA cloning vector: transduction of neomycin resistance into mammalian tissue culture cells. Proc Natl Acad Sci U S A 81(20):6466–6470. 4. Senapathy P, Carter BJ (1984) Molecular cloning of adeno-associated virus variant genomes and generation of infectious virus by recombination in mammalian cells. J Biol Chem 259 (7):4661–4666 5. Asokan A, Schaffer DV, Samulski RJ (2012) The aav vector toolkit: poised at the clinical crossroads. Mol Ther 20(4):699–708. 6. Qing K, Mah C, Hansen J, Zhou S, Dwarki V, Srivastava A (1999) Human fibroblast growth factor receptor 1 is a co-receptor for infection by adeno-associated virus 2. Nat Med 5 (1):71–77. 7. Kashiwakura Y, Tamayose K, Iwabuchi K, Hirai Y, Shimada T, Matsumoto K,

Nakamura T, Watanabe M, Oshimi K, Daida H (2005) Hepatocyte growth factor receptor is a coreceptor for adeno-associated virus type 2 infection. J Virol 79(1):609–614. https:// 8. Summerford C, Bartlett JS, Samulski RJ (1999) Alphavbeta5 integrin: a co-receptor for adeno-associated virus type 2 infection. Nat Med 5(1):78–82. 1038/4768 9. Akache B, Grimm D, Pandey K, Yant SR, Xu H, Kay MA (2006) The 37/67-kilodalton laminin receptor is a receptor for adenoassociated virus serotypes 8, 2, 3, and 9. J Virol 80(19):9831–9836. 10.1128/JVI.00878-06 10. Di Pasquale G, Davidson BL, Stein CS, Martins I, Scudiero D, Monks A, Chiorini JA (2003) Identification of pdgfr as a receptor for aav-5 transduction. Nat Med 9 (10):1306–1312. nm929 11. Weller ML, Amornphimoltham P, Schmidt M, Wilson PA, Gutkind JS, Chiorini JA (2010) Epidermal growth factor receptor is a co-receptor for adeno-associated virus serotype 6. Nat Med 16(6):662–664. 10.1038/nm.2145

Preparation and Administration of AAV for Corneal Gene Delivery 12. Bell CL, Vandenberghe LH, Bell P, Limberis MP, Gao GP, Van Vliet K, AgbandjeMcKenna M, Wilson JM (2011) The aav9 receptor and its modification to improve in vivo lung gene transfer in mice. J Clin Invest 121(6):2427–2435. 1172/JCI57367 13. Shen S, Bryant KD, Brown SM, Randell SH, Asokan A (2011) Terminal n-linked galactose is the primary receptor for adeno-associated virus 9. J Biol Chem 286(15):13532–13540. 14. Wu Z, Miller E, Agbandje-McKenna M, Samulski RJ (2006) Alpha2,3 and alpha2,6 n-linked sialic acids facilitate efficient binding and transduction by adeno-associated virus types 1 and 6. J Virol 80(18):9093–9103. 15. Pillay S, Meyer NL, Puschnik AS, Davulcu O, Diep J, Ishikawa Y, Jae LT, Wosen JE, Nagamine CM, Chapman MS, Carette JE (2016) An essential receptor for adeno-associated virus infection. Nature 530(7588):108–112. 16. Summerford C, Johnson JS, Samulski RJ (2016) Aavr: a multi-serotype receptor for aav. Mol Ther 24(4):663–666. https://doi. org/10.1038/mt.2016.49 17. Pillay S, Zou W, Cheng F, Puschnik AS, Meyer NL, Ganaie SS, Deng X, Wosen JE, Davulcu O, Yan Z, Engelhardt JF, Brown KE, Chapman MS, Qiu J, Carette JE (2017) Aav serotypes have distinctive interactions with domains of the cellular receptor aavr. J Virol. https://doi. org/10.1128/JVI.00391-17 18. Sen D (2014) Improving clinical efficacy of adeno associated vectors by rational capsid bioengineering. J Biomed Sci 21:103. https://doi. org/10.1186/s12929-014-0103-1 19. Shi W, Arnold GS, Bartlett JS (2001) Insertional mutagenesis of the adeno-associated virus type 2 (aav2) capsid gene and generation of aav2 vectors targeted to alternative cellsurface receptors. Hum Gene Ther 12 (14):1697–1711. 104303401750476212 20. Bowles DE, McPhee SW, Li C, Gray SJ, Samulski JJ, Camp AS, Li J, Wang B, Monahan PE, Rabinowitz JE, Grieger JC, Govindasamy L, Agbandje-McKenna M, Xiao X, Samulski RJ (2012) Phase 1 gene therapy for duchenne muscular dystrophy using a translational optimized aav vector. Mol Ther 20(2):443–455. mt.2011.237 21. Li W, Asokan A, Wu Z, Van Dyke T, DiPrimio N, Johnson JS, Govindaswamy L, Agbandje-McKenna M, Leichtle S, Redmond


DE, McCown TJ, Petermann KB, Sharpless NE, Samulski RJ (2008) Engineering and selection of shuffled aav genomes: a new strategy for producing targeted biological nanoparticles. Mol Ther 16(7):1252–1260. https:// 22. Zhong L, Li B, Mah CS, Govindasamy L, Agbandje-McKenna M, Cooper M, Herzog RW, Zolotukhin I, Warrington KH, WeigelVan Aken KA, Hobbs JA, Zolotukhin S, Muzyczka N, Srivastava A (2008) Next generation of adeno-associated virus 2 vectors: point mutations in tyrosines lead to high-efficiency transduction at lower doses. Proc Natl Acad Sci U S A 105(22):7827–7832. 10.1073/pnas.0802866105 23. Xiao X, Li J, Samulski RJ (1998) Production of high-titer recombinant adeno-associated virus vectors in the absence of helper adenovirus. J Virol 72(3):2224–2232 24. Urabe M, Ding C, Kotin RM (2002) Insect cells as a factory to produce adeno-associated virus type 2 vectors. Hum Gene Ther 13 (16):1935–1943. 10430340260355347 25. Chahal PS, Aucoin MG, Kamen A (2007) Primary recovery and chromatographic purification of adeno-associated virus type 2 produced by baculovirus/insect cell system. J Virol Methods 139(1):61–70. https://doi. org/10.1016/j.jviromet.2006.09.011 26. Mietzsch M, Grasse S, Zurawski C, Weger S, Bennett A, Agbandje-McKenna M, Muzyczka N, Zolotukhin S, Heilbronn R (2014) Onebac: platform for scalable and high-titer production of adeno-associated virus serotype 1-12 vectors for gene therapy. Hum Gene Ther 25(3):212–222. https://doi. org/10.1089/hum.2013.184 27. Joshi PRH, Cervera L, Ahmed I, Kondratov O, Zolotukhin S, Schrag J, Chahal PS, Kamen AA (2019) Achieving high-yield production of functional aav5 gene delivery vectors via fedbatch in an insect cell-one baculovirus system. Mol Ther Methods Clin Dev 13:279–289. 003 28. Powers AD, Piras BA, Clark RK, Lockey TD, Meagher MM (2016) Development and optimization of aav hfix particles by transient transfection in an icellis(®) fixed-bed bioreactor. Hum Gene Ther Methods 27(3):112–121. 29. Gao G, Qu G, Burnham MS, Huang J, Chirmule N, Joshi B, Yu QC, Marsh JA, Conceicao CM, Wilson JM (2000) Purification of recombinant adeno-associated virus vectors by column chromatography and its performance


Liujiang Song et al.

in vivo. Hum Gene Ther 11(15):2079–2091. 104303400750001390 30. Crosson SM, Dib P, Smith JK, Zolotukhin S (2018) Helper-free production of laboratory grade aav and purification by iodixanol density gradient centrifugation. Mol Ther Methods Clin Dev 10:1–7. omtm.2018.05.001 31. Strobel B, Miller FD, Rist W, Lamla T (2015) Comparative analysis of cesium chloride- and iodixanol-based purification of recombinant adeno-associated viral vectors for preclinical applications. Hum Gene Ther Methods 26 (4):147–157. hgtb.2015.051 32. Grieger JC, Choi VW, Samulski RJ (2006) Production and characterization of adenoassociated viral vectors. Nat Protoc 1 (3):1412–1428. nprot.2006.207 33. Samulski RJ, Chang LS, Shenk T (1989) Helper-free stocks of recombinant adenoassociated viruses: normal integration does not require viral gene expression. J Virol 63 (9):3822–3828 34. Song L, Llanga T, Conatser LM, Zaric V, Gilger BC, Hirsch ML (2018) Serotype survey of aav gene delivery via subconjunctival injection in mice. Gene Ther 25(6):402–414. https:// 35. Hirsch ML, Conatser LM, Smith SM, Salmon JH, Wu J, Buglak NE, Davis R, Gilger BC (2017) Aav vector-meditated expression of hla-g reduces injury-induced corneal vascularization, immune cell infiltration, and fibrosis. Sci Rep 7(1):17840. 1038/s41598-017-18002-9 36. Bennett J, Wellman J, Marshall KA, McCague S, Ashtari M, DiStefano-Pappas J, Elci OU, Chung DC, Sun J, Wright JF, Cross DR, Aravand P, Cyckowski LL, Bennicelli JL, Mingozzi F, Auricchio A, Pierce EA, Ruggiero J, Leroy BP, Simonelli F, High KA, Maguire AM (2016) Safety and durability of effect of contralateral-eye administration of aav2 gene therapy in patients with childhoodonset blindness caused by rpe65 mutations: a follow-on phase 1 trial. Lancet 388 (10045):661–672. S0140-6736(16)30371-3 37. Russell S, Bennett J, Wellman JA, Chung DC, Yu ZF, Tillman A, Wittes J, Pappas J, Elci O, McCague S, Cross D, Marshall KA, Walshire J, Kehoe TL, Reichert H, Davis M, Raffini L, George LA, Hudson FP, Dingfield L, Zhu X, Haller JA, Sohn EH, Mahajan VB, Pfeifer W, Weckmann M, Johnson C, Gewaily D,

Drack A, Stone E, Wachtel K, Simonelli F, Leroy BP, Wright JF, High KA, Maguire AM (2017) Efficacy and safety of voretigene neparvovec (aav2-hrpe65v2) in patients with rpe65mediated inherited retinal dystrophy: a randomised, controlled, open-label, phase 3 trial. Lancet 390(10097):849–860. https://doi. org/10.1016/S0140-6736(17)31868-8 38. Williams AM, Muir KW (2018) Awareness and attitudes toward corneal donation: challenges and opportunities. Clin Ophthalmol 12:1049–1059. OPTH.S142702 39. Gain P, Jullienne R, He Z, Aldossary M, Acquart S, Cognasse F, Thuret G (2016) Global survey of corneal transplantation and eye banking. JAMA Ophthalmol 134 (2):167–173. jamaophthalmol.2015.4776 40. Mohan RR, Tandon A, Sharma A, Cowden JW, Tovey JC (2011) Significant inhibition of corneal scarring in vivo with tissue-selective, targeted aav5 decorin gene therapy. Invest Ophthalmol Vis Sci 52(7):4833–4841. 41. Chaudhary K, Moore H, Tandon A, Gupta S, Khanna R, Mohan RR (2014) Nanotechnology and adeno-associated virus-based decorin gene therapy ameliorates peritoneal fibrosis. Am J Physiol Renal Physiol 307(7): F777–F782. ajprenal.00653.2013 42. Sharma A, Ghosh A, Hansen ET, Newman JM, Mohan RR (2010) Transduction efficiency of aav 2/6, 2/8 and 2/9 vectors for delivering genes in human corneal fibroblasts. Brain Res Bull 81(2–3):273–278. 1016/j.brainresbull.2009.07.005 43. Lai LJ, Xiao X, Wu JH (2007) Inhibition of corneal neovascularization with endostatin delivered by adeno-associated viral (aav) vector in a mouse corneal injury model. J Biomed Sci 14(3):313–322. s11373-007-9153-7 44. Cheng HC, Yeh SI, Tsao YP, Kuo PC (2007) Subconjunctival injection of recombinant aav-angiostatin ameliorates alkali burn induced corneal angiogenesis. Mol Vis 13:2344–2352 45. Mohan RR, Tovey JC, Sharma A, Schultz GS, Cowden JW, Tandon A (2011) Targeted decorin gene therapy delivered with adenoassociated virus effectively retards corneal neovascularization in vivo. PLoS One 6(10): e26432. pone.0026432 46. Vance M, Llanga T, Bennett W, Woodard K, Murlidharan G, Chungfat N, Asokan A, Gilger B, Kurtzberg J, Samulski RJ, Hirsch

Preparation and Administration of AAV for Corneal Gene Delivery ML (2016) Aav gene therapy for mps1associated corneal blindness. Sci Rep 6:22131. 47. Luca T, Givogri MI, Perani L, Galbiati F, Follenzi A, Naldini L, Bongarzone ER (2005) Axons mediate the distribution of arylsulfatase a within the mouse hippocampus upon gene delivery. Mol Ther 12(4):669–679. https:// 48. Igarashi T, Miyake K, Suzuki N, Kato K, Takahashi H, Ohara K, Shimada T (2002) New strategy for in vivo transgene expression in corneal epithelial progenitor cells. Curr Eye Res 24(1):46–50 49. Mohan RR, Schultz GS, Hong JW, Wilson SE (2003) Gene transfer into rabbit keratocytes using aav and lipid-mediated plasmid DNA vectors with a lamellar flap for stromal access. Exp Eye Res 76(3):373–383 50. Mohan RR, Sharma A, Netto MV, Sinha S, Wilson SE (2005) Gene therapy in the cornea. Prog Retin Eye Res 24(5):537–559. https:// 51. Sharma A, Tovey JC, Ghosh A, Mohan RR (2010) Aav serotype influences gene transfer in corneal stroma in vivo. Exp Eye Res 91 (3):440–448. exer.2010.06.020 52. Mohan RR, Sharma A, Cebulko TC, Tandon A (2010) Vector delivery technique affects gene transfer in the cornea in vivo. Mol Vis 16:2494–2501 53. Mohan RR, Sinha S, Tandon A, Gupta R, Tovey JC, Sharma A (2011) Efficacious and safe tissue-selective controlled gene therapy approaches for the cornea. PLoS One 6(4): e18771. pone.0018771 54. Hippert C, Ibanes S, Serratrice N, Court F, Malecaze F, Kremer EJ, Kalatzis V (2012) Corneal transduction by intra-stromal injection of aav vectors in vivo in the mouse and ex vivo in human explants. PLoS One 7(4):e35318. 0035318 55. Tsai ML, Chen SL, Chou PI, Wen LY, Tsai RJ, Tsao YP (2002) Inducible adeno-associated virus vector-delivered transgene expression in corneal endothelium. Invest Ophthalmol Vis Sci 43(3):751–757 56. Bogner B, Boye SL, Min SH, Peterson JJ, Ruan Q, Zhang Z, Reitsamer HA, Hauswirth WW, Boye SE (2015) Capsid mutated adenoassociated virus delivered to the anterior chamber results in efficient transduction of trabecular meshwork in mouse and rat. PLoS One 10


(6):e0128759. nal.pone.0128759 57. Buie LK, Rasmussen CA, Porterfield EC, Ramgolam VS, Choi VW, Markovic-Plese S, Samulski RJ, Kaufman PL, Borra´s T (2010) Self-complementary aav virus (scaav) safe and long-term gene transfer in the trabecular meshwork of living rats and monkeys. Invest Ophthalmol Vis Sci 51(1):236–248. https:// 58. O’Callaghan J, Crosbie DE, Cassidy PS, Sherwood JM, Flu¨gel-Koch C, Lu¨tjen-Drecoll E, Humphries MM, Reina-Torres E, Wallace D, Kiang AS, Campbell M, Stamer WD, Overby DR, O’Brien C, Tam LCS, Humphries P (2017) Therapeutic potential of aav-mediated mmp-3 secretion from corneal endothelium in treating glaucoma. Hum Mol Genet 26 (7):1230–1246. hmg/ddx028 59. Wang L, Xiao R, Andres-Mateos E, Vandenberghe LH (2017) Single stranded adenoassociated virus achieves efficient gene transfer to anterior segment in the mouse eye. PLoS One 12(8):e0182473. 1371/journal.pone.0182473 60. Rocha EM, Di Pasquale G, Riveros PP, Quinn K, Handelman B, Chiorini JA (2011) Transduction, tropism, and biodistribution of aav vectors in the lacrimal gland. Invest Ophthalmol Vis Sci 52(13):9567–9572. 61. Ling C, Lu Y, Cheng B, McGoogan KE, Gee SW, Ma W, Li B, Aslanidi GV, Srivastava A (2011) High-efficiency transduction of liver cancer cells by recombinant adeno-associated virus serotype 3 vectors. J Vis Exp (49). 62. Dong JY, Fan PD, Frizzell RA (1996) Quantitative analysis of the packaging capacity of recombinant adeno-associated virus. Hum Gene Ther 7(17):2101–2112. https://doi. org/10.1089/hum.1996.7.17-2101 63. Hermonat PL, Quirk JG, Bishop BM, Han L (1997) The packaging capacity of adenoassociated virus (aav) and the potential for wild-type-plus aav gene therapy vectors. FEBS Lett 407(1):78–84 64. McCarty DM, Fu H, Monahan PE, Toulson CE, Naik P, Samulski RJ (2003) Adenoassociated virus terminal repeat (tr) mutant generates self-complementary vectors to overcome the rate-limiting step to transduction in vivo. Gene Ther 10(26):2112–2118. 65. Wang Z, Ma HI, Li J, Sun L, Zhang J, Xiao X (2003) Rapid and highly efficient transduction by double-stranded adeno-associated virus


Liujiang Song et al.

vectors in vitro and in vivo. Gene Ther 10 (26):2105–2111. 66. Hirsch ML, Agbandje-McKenna M, Samulski RJ (2010) Little vector, big gene transduction: fragmented genome reassembly of adenoassociated virus. Mol Ther 18(1):6–8. 67. Grieger JC, Samulski RJ (2005) Packaging capacity of adeno-associated virus serotypes: impact of larger genomes on infectivity and postentry steps. J Virol 79(15):9933–9944. 68. Dong B, Nakai H, Xiao W (2010) Characterization of genome integrity for oversized recombinant aav vector. Mol Ther 18 (1):87–92. 2009.258 69. Okada T, Nonaka-Sarukawa M, Uchibori R, Kinoshita K, Hayashita-Kinoh H, NitaharaKasahara Y, Takeda S, Ozawa K (2009) Scalable purification of adeno-associated virus

serotype 1 (aav1) and aav8 vectors, using dual ion-exchange adsorptive membranes. Hum Gene Ther 20(9):1013–1021. https://doi. org/10.1089/hum.2009.006 70. Ayuso E, Mingozzi F, Montane J, Leon X, Anguela XM, Haurigot V, Edmonson SA, Africa L, Zhou S, High KA, Bosch F, Wright JF (2010) High aav vector purity results in serotype- and tissue-independent enhancement of transduction efficiency. Gene Ther 17 (4):503–510. 2009.157 71. Moullier P, Snyder RO (2008) International efforts for recombinant adeno-associated viral vector reference standards. Mol Ther 16 (7):1185–1188. mt.2008.125 72. Matthaei M, Meng H, Bhutto I, Xu Q, Boelke E, Hanes J, Jun AS (2012) Systematic assessment of microneedle injection into the mouse cornea. Eur J Med Res 17:19. https://

Chapter 8 The Self-assembly Approach as a Tool for the Tissue Engineering of a Bi-lamellar Human Cornea Gae¨tan Le-Bel, Pascale Desjardins, Camille Couture, Lucie Germain, and Sylvain L. Gue´rin Abstract Tissue engineering is a flourishing field of regenerative medicine that allows the reconstruction of various tissues of our body, including the cornea. In addition to addressing the growing need for organ transplants, such tissue-engineered substitutes may also serve as good in vitro models for fundamental and preclinical studies. Recent progress in the field of corneal tissue engineering has led to the development of new technologies allowing the reconstruction of a human bi-lamellar cornea. One unique feature of this model is the complete absence of exogenous material. Indeed, these human corneal equivalents are exclusively composed of untransformed human corneal fibroblasts (hCFs) entangled in their own extracellular matrix, as well as untransformed human corneal epithelial cells (hCECs), both of which isolated from donor corneas. The reconstructed human bi-lamellar cornea thereby exhibits a well-organized stroma as well as a well-differentiated epithelium. This chapter describes the methods used for the isolation and culture of hCFs, the production and assembly of hCFs stromal sheets, the seeding of hCECs, and the maturation of the tissue-engineered cornea. Key words Cornea, Human corneal fibroblasts, Tissue engineering, Stroma, Epithelium


Introduction The cornea, the transparent anterior section of the eye, is a particularly attractive tissue for tissue engineering. Its composition is quite simple: a stratified epithelium composed of 5–7 layers, a stroma essentially made up of collagen, and an endothelium, a cell monolayer that plays a very critical function in corneal deturgescence [1, 2]. To reconstruct this tissue is, however, much more complex than it seems. Tissue-engineered corneas have been widely improved over time in order to better mimic the characteristics of the native cornea, such as a well-differentiated epithelium and a fully organized stroma, which are both essential to ensure proper

Gae¨tan Le-Bel and Pascale Desjardins contributed equally with all other contributors. Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Gae¨tan Le-Bel et al.

corneal transparency [3]. Many laboratories first used collagen gels mixed with cells that were cultured from animals such as the pig [4], the rabbit [5], and the bovine [6]. At that time, the use of animal cells was more convenient due to their availability. However, these reconstructed corneas were not suitable for grafting. Since then, great efforts have been made to replace these animal cells by human cell lines. To maintain long-term culture of reconstructed corneas, immortalized cells were used at first [7, 8]. However, both the fact that they were virally transformed and unable to faithfully mimic the behavior of the native corneal cells restricted their use for corneal reconstruction. Today, due to refinements in tissue engineering, we created a reconstructed cornea that is entirely made of untransformed human cells and devoid of exogenous material such as collagen gels. In our model, stromal and epithelial cells are cultured directly from postmortem corneas and preserved in appropriate culture conditions over long time periods. Our model is also self-assembled, as supplementation of the culture medium with ascorbic acid triggers the synthesis of collagen by the stromal fibroblasts [9] and increases its stability [10, 11]. This method allows the formation of a stroma whose composition and organization are very similar to those seen with the native cornea [12–14]. This chapter details the methods for producing a bi-lamellar cornea composed of a stratified epithelium laying on a complex stroma using the self-assembly approach. Particular attention is given to the culture of human corneal cells that come directly from postmortem corneas, the use of ascorbic acid, which is required for the production of the self-assembled stromal matrix devoid of any exogenous material, and the use of the air-liquid interface to trigger the appropriate differentiation of the corneal epithelium.

2 2.1

Materials Cell Culture

2.1.1 Base Media

1. Dulbecco’s modified Eagle’s medium (DMEM): Dilute DMEM powder in apyrogenic ultrapure water. Add 3.07 g/L of NaHCO3 (final concentration: 36.5 mM). Adjust pH to 7.1. Sterilize by filtration through a 0.22 μm low-binding disposable filter and store in the dark at 4  C. 2. DME-Ham: Combine 3 parts DMEM with 1 part Ham’s medium in apyrogenic ultrapure water. Add 3.07 g/L NaHCO3 (final concentration: 36.5 mM), 24.3 mg/L of 0.18 mM adenine solubilized in 312.5 μL/L of 2 N HCl. Adjust pH to 7.1. Sterilize by filtration through a 0.22 μm low-binding disposable filter and store in the dark at 4  C.

Bi-lamellar Cornea Produced by Tissue-Engineering 2.1.2 Sera


1. Fetal calf serum. 2. Fetal clone II serum. Thaw sera at 4  C or in cold water (see Note 1). Gently swirl it to resuspend its components. Inactivate in a 56  C water bath for 30 min. To avoid repeated thawing and freezing cycles, distribute in single-use aliquots. Store at 20  C or at 80  C for long-term storage.

2.1.3 Additives

1. Insulin: 5 mg/mL in 5 mM HCl. Yields a 1000 stock solution. Can be stored at 4  C. 2. Epidermal growth factor (EGF): 200 μg/mL in 10 mM HCl. Dilute (1:20) the solution with DME-Ham containing 10% v/v fetal clone II serum. Yields a 1000 stock solution. 3. Hydrocortisone: 5 mg/mL in 95% ethanol. Dilute (1:25) the solution with DME-Ham. Yields a 500 stock solution. 4. Penicillin G/Gentamicin: 50000 IU/mL Penicillin G and 12.5 mg/mL gentamicin in apyrogenic ultrapure water. Yields a 500 stock solution. 5. Isoproterenol hydrochloride: Store at 4  C. Vials are at 0.2 mg/mL and are single use. Leftover isoproterenol should not be stored. 1000 stock solution. 6. Fungizone: 0.25 mg/mL amphotericin B in apyrogenic ultrapure water. Yields a 500 stock solution. For all additives, except isoproterenol, sterilize by filtration through a 0.22 μm low-binding disposable filter. To avoid repeated thawing and freezing cycles, distribute in single-use aliquots. Store at 80  C.

2.1.4 Complete Media

1. Tissue transport medium (tDMEM): High (4.5 g/L) glucose DMEM (containing sodium pyruvate and L-glutamine), 10% (v/v) fetal calf serum, 0.2% (v/v) penicillin G/gentamicin, 0.2% (v/v) fungizone. Store in the dark at 20  C to 80  C for up to 6 months or at 4  C for up to 10 days. 2. Complete corneal fibroblast culture medium (cfDMEM): DMEM, 10% (v/v) fetal calf serum, 0.2% (v/v) penicillin G/gentamicin. Store in the dark at 4  C for up to 10 days. 3. Complete corneal epithelial cell culture medium (ccDMEHam): DME-Ham, 5% (v/v) fetal clone II serum, 0.1% (v/v) insulin (see Note 2), 1.06 mL/L isoproterenol, 0.1% (v/v) epidermal growth factor, 0.2% hydrocortisone, and 0.2% (v/v) penicillin G/gentamicin. Store in the dark at 4  C for up to 10 days.


Gae¨tan Le-Bel et al.

4. Cryopreservation medium for human corneal fibroblast: 10% (v/v) dimethyl sulfoxide (DMSO), 90% (v/v) fetal calf serum. Keep on ice or store at 4  C and use within 24 h (see Note 3). 5. Air-liquid corneal epithelial cell culture medium (alcDMEHam): DME-Ham, 5% (v/v) fetal clone II serum, 0.1% (v/v) insulin (see Note 2), 1.06 mL/L isoproterenol, 0.1% (v/v) epidermal growth factor, 0.2% hydrocortisone, and 0.2% (v/v) penicillin G/gentamicin. Store in the dark at 4  C for up to 10 days (see Note 4). All frozen components can be thawed at 4  C (see Note 1). 2.1.5 Solutions

1. Phosphate-buffered saline (PBS): 127 mM NaCl, 2.7 mM KCl, 6.5 mM Na2HPO4, 1.5 mM KH2PO4 in apyrogenic ultrapure water. Verify pH is between 7.35 and 7.45. Yields a 10 stock solution. Store at room temperature. 2. PBS—Penicillin G/Gentamicin/Fungizone (PBS-P/G/F): Dilute 10 PBS to 1 with apyrogenic ultrapure water. Sterilize by filtration through a 0.22 μm low-binding disposable filter. Add Penicillin G/Gentamicin and Fungizone (dilute to 1). Store at 4  C. 3. 10 HEPES/KCl/NaCl: 0.1 M 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 67 mM KCL, and 1.42 M NaCl in apyrogenic ultrapure water. Adjust pH to 7.3. Yields a 10 stock solution. Store in the dark at 4  C (see Note 5). 4. HEPES/KCl/NaCl - CaCl2: Dilute 10 HEPES/KCl/NaCl to 1 with apyrogenic ultrapure water. Add CaCl2 to 1mM final. Adjust pH to 7.45. Store in the dark at 4  C. 5. Collagenase H: 0.125 U/mL collagenase H in cfDMEM. Sterilize by filtration through a 0.22 μm low-binding disposable filter. Preheat at 37  C and use immediately . 6. Dispase II: 2.5 mg/mL dispase II in HEPES/KCl/NaCl – CaCl2. Verify that pH is at 7.4 with pH paper. Sterilize by filtration through a 0.22 μm low-binding disposable filter. Store at 4  C and use within the hour. 7. Trypsin/EDTA: 0.05% (w/v) trypsin, 0.01% (w/v) EDTA, 2.8 mM D-glucose in 1 PBS. Add 100,000 IU/L penicillin G, 25 mg/L active gentamicin, 0.00075% (v/v) pre-sterile filtered 0.1% phenol red-water solution. Adjust pH to 7.45. Sterilize by filtration through a 0.22 μm low-binding disposable filter. To avoid repeated thawing and freezing cycles, distribute in singleuse aliquots. Store at 20  C to 80  C. 8. L-Ascorbic acid for corneal fibroblasts: 10 mg/mL L-ascorbic acid in cfDMEM. Sterilize by filtration through a 0.22 μm low-binding disposable filter. Store at 4  C and use within the day.

Bi-lamellar Cornea Produced by Tissue-Engineering


9. L-Ascorbic acid for corneal epithelial cell: 10 mg/mL L-ascorbic acid in ccDME-Ham. Sterilize by filtration through a 0.22 μm low-binding disposable filter. Store at 4  C and use within the day. 2.1.6 Tissues and Cells

1. Human corneal fibroblasts (hCFs) are isolated, cultured, and cryopreserved from surgically removed corneal stroma. Corneas from human postmortem donors unsuitable for transplantation (Banque nationale d’yeux du CHU de Que´bec, Que´bec, Canada). 2. Human corneal epithelial cells (hCECs) are isolated, cultured, and cryopreserved from corneal epithelium. Corneas from human postmortem donors unsuitable for transplantation (Banque nationale d’yeux du CHU de Que´bec, Que´bec, Canada).

2.1.7 Labware

1. For volumes inferior to 100 mL: 0.22 μm low-binding disposable filter. For volumes superior to 100 mL: filtration unit mounted with a 47 mm diameter and 0.22 μm filter set. 2. Sterile containers. 3. 15 and 50 mL centrifuge tubes. 4. 35  40 mm and 100  15 mm cell culture Petri dishes. 5. Dissecting curved forceps. 6. Size 4 and 22 scalpel blades. 7. Trypsination unit, Celstir® 50 mL suspension culture flask. 8. Parafilm® M. 9. 25 or 75 cm2 tissue culture flasks. 10. Sterile cryogenic vials. 11. Freezing container. 12. Sterile 7 cm  7 cm gauze. 13. Dissecting curved scissor. 14. 8 mm diameter trephine. 15. Dissecting stereomicroscope 16. 6-Well tissue-culture plates. 17. Anchoring papers, made from Whatman® grade 3 qualitative filter paper (see Note 6). 18. Plastic rings (see Note 7). 19. Sterile air-liquid stands (see Note 8). 20. Small ligating clips LIGACLIP® EXTRA. 21. Surgical single-clip applier LIGACLIP®. 22. 100  25 mm cell culture Petri dishes.


3 3.1

Gae¨tan Le-Bel et al.

Methods Cell culture

3.1.1 Tissue Sampling and Transport

Ethical approval and informed consent must be obtained for each human tissue. 1. Shortly after donor’s death, eyes or only the corneas are extracted and transferred in a sterile container filled with cold (4  C) tDMEM by qualified staff. 2. Samples must be kept on ice and transported to a cell culture facility without delay. Manipulations must be performed under a sterile laminar flow hood cabinet.

3.1.2 Isolation of Human Corneal Fibroblasts (hCFs)

1. hCFs are obtained from postmortem human donors corneas that are unsuitable for transplantation (see Note 9). 2. Wash the eye specimen in a 50 mL centrifuge tube containing 30 mL PBS-P/G/F. Agitate gently for 1–2 min. With sterile curved forceps, transfer the eye specimen into another tube filled with PBS-P/G/F. Repeat this step three times. 3. Place the eye specimen into a 100 mm Petri dish. 4. Surround the eye specimen with a folded sterile gauze. This helps in holding the eye without having to touch it. 5. With the size 22 scalpel blade, make a small opening of 2–3 mm in the sclera. 6. With curved scissors, cut out the cornea to obtain only the limbus and the central cornea. Avoid cutting the sclera to prevent contamination with conjonctival epithelial cells. 7. With two curved forceps, peel off the iris. Do this step while holding the cornea in the air to avoid any damage to the epithelium during the procedure. 8. Place the central cornea into a 35 mm tissue culture Petri dish. The central cornea is obtained by separating the limbus from the central cornea with an 8 mm diameter trephine (see Note 10). Hold the epithelium upward and add 5 mL of cold (4  C) dispase II. Seal the Petri dish with parafilm. 9. Incubate overnight at 4  C. 10. With two curved forceps, mechanically detach the epithelium from the stroma under a dissecting microscope. 11. With curved forceps, transfer the corneal stroma in a new 35 mm tissue culture Petri dish. With the size 22 scalpel blade, cut the corneal stroma into small pieces. Add 2 mL of collagenase H and keep cutting the corneal stroma until you get small pieces of 1–2 mm2.

Bi-lamellar Cornea Produced by Tissue-Engineering


12. Collect the 2 mL of collagenase H containing the small pieces of stroma and transfer it in a Celstir® suspension culture flask containing 18 mL warm (37  C) collagenase H. 13. Incubate under agitation for 2–3 h at 37  C, until the small pieces of corneal stroma have been digested by the enzyme (see Note 11). 14. Stir well and collect the supernatant, put it in a 50 mL centrifuge tube, and add 20 mL of cfDMEM. 15. Centrifuge (300  g) the hCFs suspension for 10 min at room temperature. 16. Remove the supernatant, resuspend with 20 mL of cfDMEM, and centrifuge (300  g) the hCFs suspension for 10 min at room temperature (see Note 12). 17. Remove the supernatant and resuspend hCFs in 5 mL of warm (37  C) cfDMEM. 18. Seed hCFs into a culture flask and add cfDMEM. Total medium volume should not exceed 7 mL/25 cm2. 19. When the hCFs reach 90% confluence, subculture or freeze cells. 3.1.3 Culture

1. Place hCFs seeded culture flasks in an 8% CO2 and 100% humidity atmosphere incubator at 37  C. 2. Change the culture medium three times a week, every 2–3 days. Remove the medium from the culture flask. Replace it with warm (37  C) cfDMEM. 3. Monitor cell confluence (see Note 13) daily under a microscope. 4. When cells reach 75–95% confluence, subculture (Fig. 1) or cryopreserve them. Do not let cells reach 100% confluence.

3.1.4 Subculture

1. Remove the culture medium. 2. Depending on culture flask size, swiftly rinse cells with either 1 or 2 mL (for either 25 or 75 cm2 culture flasks, respectively) trypsin/EDTA. Remove it. 3. Depending on culture flask size, add either 2 or 3 mL (for either 25 or 75 cm2 culture flasks, respectively) of trypsin/ EDTA into the culture flask. 4. Incubate at 37  C until all cells are completely detached from the flask (verify cell detachment under a microscope). Time for complete detachment should be around 10 min. Do not incubate for more than 10 min (see Note 14). 5. Depending on culture flask size, neutralize trypsin activity by adding either 2 or 3 mL of cfDMEM.


Gae¨tan Le-Bel et al.

Fig. 1 Growth and morphological characteristics of cultured human corneal fibroblasts. Human corneal fibroblasts (hCFs) isolated from the central corneal

Bi-lamellar Cornea Produced by Tissue-Engineering


6. Vigorously pipette the cell suspension up and down at least ten times to ensure suspension homogeneity. 7. Transfer the cell suspension into a 50 mL centrifuge tube. 8. Depending on culture flask size, thoroughly rinse the culture flask with 2 or 4 mL cfDMEM. Transfer the suspension into the 50 mL tube. 9. Use an automated cell counter or a hemocytometer to count cells. 10. Use trypan blue staining and a hemocytometer to estimate cell viability. Cell viability is expected to be greater than 95%. 11. Centrifuge (300  g) the cell suspension for 10 min at room temperature. 12. Remove the supernatant and resuspend cells at the desired concentration in cfDMEM. 13. Seed cells at no less than 7000 cells/cm2 (see Note 15) into a culture flask. Seed directly into the culture medium cfDMEM. 1. Fill a freezing container with 100% isopropyl alcohol. Store at 4  C until cool.

3.1.5 Cryopreservation

2. Follow steps 1 through 11 from the previous section (Subheading 3.1.4). 3. Remove the supernatant and resuspend cells at the desired concentration (max. 1  107/mL) in hCF cryopreservation medium. Put the tube on ice. 4. Aliquot in cryogenic vials on ice. 5. Put the cryogenic vials in the freezing container. 6. Store the container overnight at 80  C. Under these conditions, cell temperature should drop 1  C/min. 7. Store cryogenic vials in liquid nitrogen for long-term storage. 1. Put the cryogenic vial in a 37  C water bath. Do not let the cell suspension thaw completely. A small ice pellet should remain.

3.1.6 Thawing

2. Add 0.5–1 mL of cold (4  C) cfDMEM into the cryogenic vial. 3. As soon as the remaining ice has melted, transfer the content of the cryogenic vial into a 50 mL centrifuge tube containing 8–10 mL of cold (4  C) cfDMEM.

ä Fig. 1 (continued) stroma of a postmortem human eye were isolated and seeded in 75 cm2 tissue culture flasks. hCFs reached 25% confluence after 4 days in culture (a), 40% confluence after 7 days in culture (b) and 95% confluence after 9 days in culture (c). Scale bar: 25 μm


Gae¨tan Le-Bel et al.

4. Centrifuge (300  g) the cell suspension for 10 min at room temperature. 5. Remove the supernatant and resuspend cells in 10 mL of warm (37  C) cfDMEM. 6. Use an automated cell counter or a hemocytometer to count the cells. 7. Use trypan blue staining and a hemocytometer to estimate cell viability. Cell viability is expected to be greater than 80%. 8. Centrifuge (300  g) the cell suspension for 10 min at room temperature. 9. Remove the supernatant and resuspend cells at the desired concentration in cfDMEM (see Note 16). 10. Seed cells at no less than 7000 cells/cm2 (see Note 15) into a culture flask. Seed directly into the culture medium cfDMEM. 3.2 Tissue-Engineered Human Cornea 3.2.1 Production of hCF Sheets

All further manipulations must be performed under a sterile laminar flow hood cabinet.

1. In a 6-well plate, dispose an anchoring paper in each well and a plastic ring on top of it. 2. Add ascorbic acid solution in cfDMEM to obtain a final concentration of 50 μg/mL in the medium. 3. Seed 105 hCFs per well (9.6 cm2) in the prepared 6-well culture plate in 5 mL of cfDMEM containing 50 μg/mL ascorbic acid. Seed directly into the culture medium (see Note 17). 4. Incubate in 8% CO2, 100% humidity atmosphere at 37  C for 40 days. Change culture medium three times a week.

3.2.2 Assembly of hCF Sheets for Stromal Reconstruction

1. Completely remove culture medium from two wells at a time and remove temporarily the plastic ring. 2. With a curved forceps, gently scrub the inside of the well all around the anchoring paper and carefully detach the stromal sheet from the bottom of the well. 3. Transfer one stromal sheet on top of the other. 4. With a sterile surgical single-clip applier, take one ligating clip at a time. 5. Using a sterile curved forceps in one hand and the surgical single-clip applier in the other, attach together the two stromal sheets by placing the ligating clip around the two anchoring papers.

Bi-lamellar Cornea Produced by Tissue-Engineering


6. Repeat steps 4 and 5 three times while disposing the ligating clips evenly all around the anchoring papers. 7. Replace the plastic ring on top of the reconstructed stroma and add 5 mL of cfDMEM containing 50 μg/mL ascorbic acid. 8. Incubate in 8% CO2, 100% humidity atmosphere at 37  C for 1 week to merge. Change culture medium every 2 or 3 days with cfDMEM containing 50 μg/mL ascorbic acid. 3.2.3 Seeding of Human Corneal Epithelial Cells (hCECs) and Maturation at the Air-Liquid Interface

1. One week after the assembly of fibroblast sheets, hCECs can be seeded on the reconstructed stroma. hCECs between their second and fourth passages at the time of seeding are favored. 2. Resuspend hCECs at 2  105 or 2.5  105 cells/mL in ccDME-Ham. 3. Add 10 mg/mL ascorbic acid solution to ccDME-Ham to obtain a final concentration of 50 μg/mL. 4. Remove culture medium of reconstructed stromas and replace it with 5 mL of warm ccDME-Ham containing 50 μg/mL ascorbic acid. 5. Seed 1  106 hCECs per reconstructed stroma drop by drop everywhere within the plastic ring. Replace the 6-well plate in the incubator carefully to limit the dispersion of hCECs within the well (Fig. 2). 6. Incubate in 8% CO2, 100% humidity atmosphere at 37  C. Change culture medium three times a week with ccDMEHam containing 50 μg/mL ascorbic acid. 7. After 1 week, remove culture medium and plastic ring in each well. 8. Place an air-liquid stand in a 100  25 mm cell culture Petri dish. 9. Using a curved forceps, carefully detach the reconstructed cornea from the bottom of the well. 10. Place the reconstructed tissue on the air-liquid stand with the hCECs on top (Fig. 2). 11. Add the right volume of 10 mg/mL ascorbic acid solution in alcDME-Ham to obtain a final concentration of 50 μg/mL. 12. Depending on the air-liquid stand, add the right volume (here, 18 mL) of alcDME-Ham containing 50 μg/mL ascorbic acid. Make sure the bottom of the reconstructed cornea is in contact with culture medium while the top is kept dry (see Note 18). 13. Incubate in 8% CO2, 100% humidity atmosphere at 37  C and keep at the air-liquid interface for 1 week (Fig. 3). Change culture medium every 2 or 3 days with alcDME-Ham containing 50 μg/mL ascorbic acid.


Gae¨tan Le-Bel et al.

Fig. 2 Detail of the procedure used for the production of a human tissue-engineered bi-lamellar cornea. (a) After assembly and maturation of hCF sheets for stromal reconstruction, hCECs were seeded on the reconstructed human corneal stromas. (b and c) An air-liquid stand produced by 3D printing in PLA and covered with a nylon membrane. (d and e) The reconstructed tissue is deposited on the air-liquid stand with hCECs on top and left in culture for 1 week in order to induce vertical stratification of the epithelium. (f) Mature, human bi-lamellar cornea


Notes 1. Sera and additives can be thawed more rapidly at room temperature or in a 37  C water bath. However, do not refreeze. Rather, we recommend using immediately or dilute in culture medium at working dilution for further utilization. 2. Serum must be added first, followed by insulin. Insulin must be added with a new sterile plastic pipette.

Bi-lamellar Cornea Produced by Tissue-Engineering


Fig. 3 Characteristics of the human tissue-engineered bi-lamellar cornea (a and b) Macroscopic views showing the transparency of the human bi-lamellar cornea. (c) Electron microscopic analysis of the tissueengineered corneal stroma. Organization of collagen fibers is very similar to that observed in the native corneal stroma with collagen fibers perpendicular to each other. BM basement membrane; hCEC human corneal epithelial cell; C collagen fibers; hCFs human corneal fibroblasts. (d) Histological view of the human bi-lamellar cornea that shows the stratified epithelium made up of five to seven cell layers attached to the corneal stroma containing many hCFs (sections were stained with Masson trichrome; cells are purple and collagen is bluish). Scale bar: 20 μm

3. DMSO is a toxic oxidative agent at temperatures above 10  C. Working with cells in contact with a solution containing DMSO must be done quickly and on ice. 4. The composition of the alcDME-Ham is substantially the same as for the ccDME-Ham, except that EGF is absent, allowing hCECs differentiation rather than proliferation. 5. HEPES may undergo degradation when exposed to light and might become toxic. 6. To make the anchoring paper, cut a ring with a 31.8 mm external diameter and a 22.2 mm internal diameter in the Whatman filter paper grade 3. Anchoring papers are sterilized by autoclave.


Gae¨tan Le-Bel et al.

7. Plastic rings are made of 50 mL collection tube. Caps are removed and tubes are cut at approximately 1.5 cm from the aperture of the tube. The 1.5 cm high plastic ring is notched at its base to allow the culture medium to move freely within the well. Plastic rings are used to prevent the anchoring paper (with cells attach to it) to float at the surface of the culture medium. Ingots or anchoring rings in stainless steel may be used as well. Sterilization of the plastic rings is performed by gas sterilization with ethylene oxide (Fig. 2). 8. Air-liquid stands are made by 3D printing. Dimensions are 56 mm external diameter, 32 mm internal diameter, and 6 mm height. Three legs of 4 mm high support the air-liquid stands. Stands are made of polylactic acid (PLA) and are covered with a nylon membrane fixed to the plastic stands with chloroform (Fig. 2). Since it deforms the plastic, gas sterilization is not suitable for those air-liquid stands. Instead, we sterilize them by immersing the stands in 100% isopropyl alcohol, followed by ultraviolet irradiation under a sterile laminar flow hood cabinet. Each side of the air-liquid stand is exposed to ultraviolet irradiation for at least 30 min. 9. Protocols must be approved by the institution’s committee for the protection of human subjects. It is preferable to use corneas with as short as possible postmortem time to improve cultivation of hCFs and hCEC. 10. Corneal epithelial stem cells are found in the limbal basal epithelium, the transition area separating the cornea and underlying sclera. Discarding the limbal ring prevents the formation of hCECs colonies caused by corneal epithelial stem cells contaminating the corneal fibroblast cultures as hCECs can efficiently proliferate in cfDMEM. 11. Collagenase H breaks the peptide bonds in collagen, destroying extracellular matrix structures. Collagenase treatment should be stopped as soon as there are no more visible pieces in suspension. Indeed, collagenase turns out to be toxic for cells when incubated for a long time. It is therefore necessary to check the suspension every 30 min. 12. Contrary to trypsin, collagenase H activity cannot be neutralized with fetal calf serum contained in cfDMEM. For longterm use, collagenase H is toxic for the cells. A double wash is then required to completely remove remaining collagenase H from the suspension. 13. Here, we define confluence as the approximate percentage of the culture flask surface covered by hCFs. It is estimated under a microscope by assessing how much of a given field of vision is occupied by hCFs. Do not let more confluent areas differentiate for the sake of obtaining a higher mean confluence.

Bi-lamellar Cornea Produced by Tissue-Engineering


14. For optimal efficiency, do not stack culture flasks on top of each other in the incubator. Temperature is typically higher on the flask surface directly in contact with the incubator shelf. 15. The seeding density is determined by the operator. hCF densities are expected to double each day. Proliferation can vary from one population to another and decreases as cells are passaged. It is recommended to seed at no less than 7000 cells/cm2 with an uncharacterized population. 95% confluence should be reached within 5–7 days for the first few passages. 16. A double wash is preferable to completely remove DMSO from the suspension. 17. hCFs between their fifth and seventh passages are favored for generating sheets. 18. Try to avoid air bubbles beneath the reconstructed cornea to maintain the air-liquid interface. If air bubble should form, remove it with the vacuum.

Acknowledgements The authors would like to thank current and former members of the LOEX and CUO-Recherche laboratories who contributed to develop and improve the foregoing protocols. This work was supported by the Canadian Institutes for Health Research (CIHR) grant MOP-12087 and FDN-143213 (L.G.), the Fondation des Pompiers du Que´bec pour les Grands Bruˆle´s (FPQGB), the Fonds de Recherche du Que´bec-Sante´ (FRQS), and the Re´seau de the´rapie cellulaire, tissulaire et ge´nique du Que´bec -The´Cell (a thematic network supported by the FRQS). The Banque d’yeux Nationale is partly supported by the Re´seau de Recherche en Sante´ de la Vision from the FRQS. P.D. and C.C. were supported by studentships from the FRQS. L.G. is the recipient of a Tier 1 Canadian Research Chair on Stem Cells and Tissue Engineering and a Research Chair on Tissue-Engineered Organs and Translational Medicine of the Fondation de l’Universite´ Laval. References 1. Bonanno JA (2012) Molecular mechanisms underlying the corneal endothelial pump. Exp Eye Res 95(1):2–7. 1016/j.exer.2011.06.004 2. Eghrari AO, Riazuddin SA, Gottsch JD (2015) Overview of the cornea: Structure, function, and development. Prog Mol Biol Transl Sci 134:7–23. pmbts.2015.04.001

3. DelMonte DW, Kim T (2011) Anatomy and physiology of the cornea. J Cataract Refract Surg 37(3):588–598. 1016/j.jcrs.2010.12.037 4. Schneider AI, Maier-Reif K, Graeve T (1999) Constructing an in vitro cornea from cultures of the three specific corneal cell types. In Vitro Cell Dev Biol Anim 35(9):515–526. https://


Gae¨tan Le-Bel et al.

5. Zieske JD, Mason VS, Wasson ME, Meunier SF, Nolte CJ, Fukai N, Olsen BR, Parenteau NL (1994) Basement membrane assembly and differentiation of cultured corneal cells: Importance of culture environment and endothelial cell interaction. Exp Cell Res 214(2):621–633. 6. Parnigotto PP, Bassani V, Montesi F, Conconi MT (1998) Bovine corneal stroma and epithelium reconstructed in vitro: Characterisation and response to surfactants. Eye (Lond) 12 (Pt 2):304–310. eye.1998.70 7. Griffith M (1999) Functional human corneal equivalents constructed from cell lines. Science 286(5447):2169–2172. 1126/science.286.5447.2169 8. Griffith M (2002) Artificial human cornea. Cornea 21(2):54–61. 1097/01.ico.0000263120.68768.f8 9. Geesin JC, Darr D, Kaufman R, Murad S, Pinnell SR (1988) Ascorbic acid specifically increases type i and type iii procollagen messenger rna levels in human skin fibroblasts. J Invest Dermatol 90(4):420–424. 10.1111/1523-1747.ep12460849 10. Chan D, Lamande SR, Cole WG, Bateman JF (1990) Regulation of procollagen synthesis and processing during ascorbate-induced

extracellular matrix accumulation in vitro. Biochem J 269:175–181 11. Michel M, L’Heureux N, Pouliot R, Xu W, Auger FA, Germain L (1999) Characterization of a new tissue-engineered human skin equivalent with hair. In Vitro Cell Dev Biol Anim 35 (6):318–326. s11626-999-0081-x 12. Couture C, Zaniolo K, Carrier P, Lake J, Patenaude J, Germain L, Gue´rin SL (2016) The tissue-engineered human cornea as a model to study expression of matrix metalloproteinases during corneal wound healing. Biomaterials 78:86–101. 1016/j.biomaterials.2015.11.006 13. Carrier P, Deschambeault A, Talbot MV, Giasson CJ, Auger FA, Gue´rin SL, Germain L (2008) Characterization of wound reepithelialization using a new human tissue-engineered corneal wound healing model. Invest Ophthalmol Vis Sci 49(4):1376–1385. https://doi. org/10.1167/iovs.07-0904 14. Proulx S, d’Arc Uwamaliya J, Carrier P, Deschambeault A, Audet C, Giasson CJ, Gue´rin SL, Auger FA, Germain L (2010) Reconstruction of a human cornea by the selfassembly approach of tissue engineering using the three native cell types. Mol Vis 16:2192–2201

Chapter 9 Formation of Corneal Stromal-Like Assemblies Using Human Corneal Fibroblasts and Macromolecular Crowding Mehmet Gu¨rdal, Gu¨linnaz Ercan, and Dimitrios I. Zeugolis Abstract Tissue engineering by self-assembly allows for the formation of living tissue substitutes, using the cells’ innate capability to produce and deposit tissue-specific extracellular matrix. However, in order to develop extracellular matrix-rich implantable devices, prolonged culture time is required in traditionally utilized dilute ex vivo microenvironments. Macromolecular crowding, by imitating the in vivo tissue density, dramatically accelerates biological processes, resulting in enhanced and accelerated extracellular matrix deposition. Herein, we describe the ex vivo formation of corneal stromal-like assemblies using human corneal fibroblasts and macromolecular crowding. Key words Corneal stroma, Tissue engineering by self-assembly, Macromolecular crowding, Extracellular matrix


Introduction The cornea is a highly specialized transparent tissue located at the anterior most surface of the eye. The cornea is composed of three layers: an outer epithelial layer, a middle stromal layer, and an inner endothelial layer. The corneal stromal layer is synthesized by corneal keratocytes, makes up 90% of the corneal thickness, and comprises a heterodimeric complex of type I and type V collagen fibers, which are arranged in bundles referred to as lamellae [1–5]. Tissue engineering by self-assembly approaches utilize the natural sophistication of the cells to create tissue-specific or native supramolecular assemblies [6, 7]. Ex vivo, the formation of native supramolecular assemblies is dependent on the rate of extracellular matrix (ECM) synthesis and deposition, which, under the dilute traditional cell culture conditions, is extremely slow. In vivo, cells reside in the dense network of the ECM, where the enzymatic processing of procollagen to collagen type I is rapid. In vitro, in the dilute culture media, the conversion of water-soluble

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Mehmet Gu¨rdal et al.

procollagen to water-insoluble collagen type I is very slow; thus, prolonged culture times are required to produce an implantable device [8, 9]. Macromolecular crowding (MMC) is a biophysical phenomenon that accelerates biological processes (such as the enzymatic conversion of procollagen to collagen), resulting in amplified and accelerated ECM deposition in vitro in permanently differentiated and stem cell cultures [10–14]. Among the various macromolecular crowders that have been used to-date, carrageenan, a highly sulfated polysaccharide, has been shown to induce the highest and fastest ECM deposition in various cell types due to its polydispersity and negative charge [15, 16]. Herein, we describe the ex vivo formation of corneal stromallike assemblies using human corneal fibroblasts and ascending concentrations of carrageenan in order to identify the optimal concentration for maximum ECM deposition. The influence of carrageenan in corneal fibroblasts cultures was assessed with cell morphology, metabolic activity, viability and proliferation and sodium dodecyl sulfate poly-acrylamide gel electrophoresis (SDS-PAGE) and immunocytochemistry.


Materials All cell culture materials should be handled following aseptic technique. Human primary cells should be handled using Biosafety Level 2 practices and containment. Diligently follow all the waste disposal regulations of your country/institution when disposing any kind of waste material.


Cell Culture

1. Human corneal fibroblasts (HCFs), cryopreserved. 2. Fibroblast Medium Kit (see Note 1). 3. Poly(L-lysine) solution. 4. Double distilled water (ddH2O), sterile. 5. Hank’s Balanced Salt Solution (HBSS), sterile. 6. Penicillin-streptomycin solution (PS) 100: 10,000 units/ml penicillin, 10 mg streptomycin/mL. 7. Maintaining growth medium: Fibroblast basal medium, 2% fetal bovine serum (FBS), 1% fibroblast growth supplement (FGS), 1% PS. 8. Dimethyl sulfoxide (DMSO), suitable for cell culture. 9. Dulbecco’s modified eagle medium: nutrient mixture F12 (DMEM/F12). 10. Newborn calf serum (NBCS), heat inactivated. 11. Human basic fibroblast growth factor (FGF-2).

Macromolecular Crowding in Human Corneal Fibroblast Culture


12. Growth medium: DMEM/F12, 10% NBCS, 5 ng/mL FGF-2, 1% PS. 13. 0.25% Trypsin-EDTA solution. 14. Carrageenan powder, suitable for gel preparation. 15. 100 mM L-ascorbic acid 2-phosphate solution: 100 mM Lascorbic acid 2-phosphate in ddH2O. When dissolved filter with a syringe and a 0.2 μm syringe filter. Store it in frozen aliquots at 20  C and protect from the light. 16. 0.2 μm surfactant-free cellulose acetate sterile syringe filters. 17. 75 cm2 tissue culture flask. 18. 175 cm2 tissue culture flask. 19. Poly(L-lysine)-coated 75 cm2 tissue culture treated flask: 0.1 mg/mL poly(L-lysine) solution is prepared and 1.5 mL of solution is transferred to 75 cm2 flask (see Note 2). The flask is gently rocked to evenly coat the surface. After 5 min, the excess solution is removed and the surface is thoroughly rinsed with sterile ddH2O and allowed to dry for several hours. 20. 24- and 48-well tissue culture treated plates. 21. MMC growth medium: Weigh the appropriated amount of carrageenan powder in a 1.5 mL microcentrifuge tube in order to prepare 1 mL of medium per well of 24-well tissue culture plate or 0.5 mL of medium per well of 48-well tissue culture plate to be treated, considering the concentration of carrageenan. Always prepare 2 mL extra in order to compensate for pipetting errors. In order to disinfect the carrageenan, irradiate it with UV-C light for 15 min (see Note 3). Supplement growth medium with 1 μL the 100 mM ascorbic acid solution per mL of medium in order to reach a concentration of 100 μM. This will be used as the control medium and will also be used to prepare crowding MMC growth medium by adding carrageenan. For the preparation of MMC growth medium, recover the disinfected carrageenan from the 1.5 mL microcentrifuge tube by suspending it in 1 mL of growth medium supplemented with ascorbic acid and transfer the volume to a tube with the remaining medium. Repeat this step at least twice in order to ensure the complete recovery of the carrageenan from the tube. For non-crowded control wells, keep aside the appropriated volume of growth medium supplemented with L-ascorbic acid 2-phosphate for this purpose. For the solubilization of the carrageenan in the growth medium supplemented with ascorbic acid, incubate the tube in a thermostatic bath at 37  C for at least 30 min (see Note 4).


Mehmet Gu¨rdal et al.

2.2 Collagen Deposition Analysis

1. HBSS. 2. 0.5 M Acetic acid: 26.622 mL glacial acetic acid in 973.378 mL ddH2O. 3. 100 μg/mL pepsin from porcine gastric mucosa: 1 mg pepsin in 0.5 M acetic acid and dilute 1/10 (v/v) in HBSS. Use pepsin solution in 30 min. 4. Phenol red solution: 10 mg of phenol red in 50 mL of ddH2O. 5. 1 N sodium hydroxide (NaOH): 4 g of NaOH in ddH2O, final volume 100 mL (see Note 5). 6. 37% hydrochloric acid (HCl). 7. 1.875 M Tris–HCl, pH 8.8: 22.70 g Tris-base, 80 mL ddH2O, 2 mL 37% HCl. Leave overnight to equilibrate, adjust pH to 8.8 with a few drops concentrated HCl and make it up to 100 mL with ddH2O. Keep it at 4  C. 8. 1.25 M Tris–HCl, pH 6.8: 15.14 g Tris-base, 70 mL ddH2O, 7 mL 37% HCl. Leave overnight to equilibrate, adjust pH to 6.8 with a few drops concentrated HCl and make it up to 100 mL with ddH2O. Keep it at 4  C. 9. Sodium dodecyl sulfate (SDS) (see Note 6). 10. 5 sample buffer: 0.25 g SDS, 0.625 mL 1.25 M Tris–HCl, pH 6.8, 2 mL ultrapure water. Leave overnight for the foam to settle. Top up with glycerol to 5 mL (approximately 2.3 mL). Add 2.5 mg bromophenol blue per 10 mL buffer. Also prepare 1 sample buffer diluted from 5 sample buffer with ddH2O. 11. 5 running buffer: 15.1 g Tris-base, 72 g glycine, 5 g SDS, 1 L ddH2O. Store at 4  C. 1 running buffer is made to run the gel from 5 running buffer by diluting in ddH2O. 12. 30% Acrylamide/Bis (37.5:1). 13. 10% (w/v) SDS: 10 g SDS in 90 ml ddH2O and adjust the volume to 100 mL. This solution can be stored at room temperature for up to 6 months. 14. 100 mg/mL Ammonium Persulfate (APS): 500 mg APS, 5 mL ddH2O. Aliquot in microcentrifuge tubes and keep them at 20  C. The solution is active for a few months (see Note 7). 15. N,N,N0 ,N0 -Tetramethylethylenediamine (TEMED) (see Note 8). 16. Absolute ethanol. 17. 10% (v/v) Ethanol: Dilute 10 mL of absolute ethanol in 90 mL of ddH2O. Mix thoroughly. Store at room temperature for up to 1 year. 18. 70% (v/v) Ethanol: Add 30 mL of ddH2O to 70 mL of absolute ethanol. Mix thoroughly. Store at room temperature for up to 1 year.

Macromolecular Crowding in Human Corneal Fibroblast Culture


19. 1 mg/mL collagen type I standard: 1 mg collagen type I, 1 mL 0.5 M acetic acid. Make further dilution in 0.5 M acetic acid, if needed. 20. Hamilton syringe. 21. SDS PAGE gel making and electrophoresis apparatus. 22. SilverQuest™ Silver Staining Kit (see Note 9). 23. Fixing solution: Mix 40 mL absolute ethanol, 10 mL glacial acetic acid and adjust volume to 100 mL ddH2O. Mix thoroughly. Prepare the fixing solution immediately before using. 24. Sensitizing solution: Mix 30 mL absolute ethanol, 10 mL sensitizer and adjust volume to 100 mL with ddH2O. Mix thoroughly. Prepare the sensitizing solution immediately before using. 25. Staining solution: Mix 1 mL stainer and 99 mL ddH2O. Mix thoroughly. Prepare the staining solution immediately before using. 26. Developing solution: Mix 10 mL developer, 1 drop developer enhancer and adjust volume to 100 mL with ddH2O. Mix thoroughly. Prepare the developing solution immediately before using. 27. 30% (v/v) Ethanol: Dilute 30 mL of absolute ethanol in 70 mL of ddH2O and mix thoroughly. Store at room temperature for up to 1 year. 2.3 Metabolic Activity Assessment

1. HBSS. 2. AlamarBlue® solution. 3. 96-Well plate. 4. Microplate reader.

2.4 Cell Viability Assessment

1. DMSO. 2. HBSS. 3. 4 mM Calcein AM in DMSO. 4. 2 mM Ethidium homodimer-1 in DMSO. 5. Inverted fluorescence microscope.

2.5 Cell Proliferation Assessment

1. HBSS. 2. DNase-free water. 3. Quant-iT™ PicoGreen® dsDNA assay kit. 4. 96-Well plate. 5. Microplate reader.


Mehmet Gu¨rdal et al.

2.6 Immunofluorescent Characterization of ECM

1. Phosphate-buffered saline (1 PBS): 137 mM sodium chloride (NaCl), 2.7 mM potassium chloride (KCl), 10 mM phosphate buffer. Add 8 g of NaCl, 0.2 g of KCl, 1.44 g of Na2HPO4, and 0.24 g of KH2PO4 to 800 mL of ddH2O. Adjust the pH to 7.4 with HCl before adding ddH2O to 1 L. 2. Fixing solution: 4% (w/v) paraformaldehyde (PFA) in 1 PBS. Weigh 0.4 g of PFA 95% powder and dissolve it by heating and stirring in a glass beaker containing 10 mL of 1 PBS (see Note 10). Allow dissolution for 1 h (see Note 11). Once the PFA has been completely dissolved, transfer the resulting fixation solution to a 50 mL centrifuge tube. Store at 4  C. Filter the solution through a 0.2 μm sterile syringe filter before use. 3. Blocking solution: 3% bovine serum albumin (BSA) in 1 PBS. Weigh 0.3 g of BSA and dissolve it in 10 mL of 1 PBS in a 15 mL centrifuge tube by vortexing. Keep it at 4  C for shortterm or 20  C for long-term storage. 4. Primary antibodies: rabbit polyclonal anti-collagen type I (Boster, UK; PA2140-2), rabbit polyclonal anti-collagen type IV (Abcam, UK; ab6586), rabbit polyclonal anti-collagen type V (Abcam, UK; ab7046), rabbit polyclonal anti-collagen type VI (Abcam, UK; ab6588), rabbit polyclonal anti-fibronectin (Abcam, UK; ab2413) (see Note 12). Prepare the dilutions of the primary antibodies in 1 PBS as indicated: rabbit polyclonal anti-collagen type I (1:200), rabbit polyclonal anticollagen type IV (1:200), rabbit polyclonal anti-collagen type V (1:200), rabbit polyclonal anti-collagen type VI (1:200), rabbit polyclonal anti-fibronectin (1:200). 5. Secondary antibody: Goat anti-rabbit conjugated to AlexaFluor®488 (Invitrogen, Thermo Fisher Scientific, Ireland; A-32731) (see Note 13). Prepare the dilution of the secondary antibody in 1 PBS as indicated: goat anti-rabbit AlexaFluor®488 (1:400) for collagen types I, IV, V, VI and fibronectin immunofluorescent staining. Protect it from exposure to light (see Note 14). 6. 20 mM Hoechst 33342 Solution. 7. Olympus IX-81 inverted fluorescence microscope, equipped with 10 objective, filters suitable for Hoechst 33342 (Ex/Em: 358/461 nm), AlexaFluor®488 (Ex/Em: 495/519 nm), and a connected camera.


Methods For the preparation of corneal stromal-like assemblies rich in ECM, routine cell culture practices should be applied.

Macromolecular Crowding in Human Corneal Fibroblast Culture

3.1 Preparation of Corneal Stromal-Like Assemblies Rich in ECM


1. Fill the appropriated number of 75 cm2 poly(L-lysine)-coated tissue culture flasks (T75) with 9 mL of maintaining growth medium. Equilibrate the flasks for 30 min in a cell culture incubator humidified at 37  C at 5% CO2 level. 2. Thaw a vial of cryopreserved HCFs, low passage preferred (see Note 15) in a thermostatic bath at 37  C and seed the pre-equilibrated tissue culture flasks at 5000 cells/cm2. Incubate the cells overnight at 37  C and 5% CO2 under humidified atmosphere and replace the medium with 9 mL of fresh maintaining growth medium per T75 flask. Replace the maintaining growth medium every 2–3 days. 3. Once the cell cultures have reached approximately 85% confluency, discard the medium and wash the cell layers twice with 5 mL of HBSS per T75 flask. Discard the HBSS and add 1 mL of trypsin-EDTA solution to each flask and incubate for 5 min at 37  C. Assess cell detachment by microscopical examination, and once all cells are detached from the flasks, add 2 mL of maintaining growth medium, collect the cell suspension, and centrifuge it for 5 min at 360  g. Discard the supernatant and resuspend the cell pellet in 1 mL of growth medium per each cultured T75. Determine the cell concentration with the aid of a hematocytometer. 4. To plate the desired number of wells in 24 or 48-well tissue culture treated plates, adjust the cell concentration in order to seed 25,000 cells/cm2 and ensure that each well has 1 mL of growth medium for 24-well plates or 0.5 mL of growth medium for 48-well plates. Incubate the plates at 37  C and 5% CO2. In order to appreciate the effects of MMC on ECM deposition, at least three wells are used per analysis or per immunofluorescence marker to be analyzed, and three wells without macromolecular crowder are used as control group. Plate the necessary number of wells per immunofluorescence assay in duplicate in order to have non-primary antibody controls in triplicate per condition. 5. After 24 h of culture, examine the cells by microscopy and ensure that they are healthy and properly attached to the cell culture substrate before replacing the medium. 6. Replace the medium of the wells to be treated with MMC medium with 1 mL (for 24-well plate) or 0.5 mL (for 48-well plate) of the growth medium containing ascorbic acid and carrageenan or only ascorbic acid for control wells. Incubate the plates at 37  C and 5% CO2. 7. According to determined time points (in this study day 4, 7, and 10), the plates can be examined by microscopy (Fig. 1) and processed by SDS-PAGE, metabolic activity assessment, viability assessment, proliferation assessment, and immunofluorescent staining.


Mehmet Gu¨rdal et al.

Fig. 1 Morphology of HCFs in the absence and presence of increasing concentrations of macromolecular crowder carrageenan (CR). Phase contrast microscopy revealed that HCFs maintained their spindle-shaped morphology at each time point (day 4, 7, and 10) independently of the presence or absence of CR. However, higher than 100 μg/mL CR concentrations caused cell losses 3.2


3.2.1 Pepsin Treatment and Neutralization of Samples

1. At the end of each time point, aspirate the medium from the wells of 24-well plate. 2. Wash the cell layer portion with HBSS. 3. Add 100 μg/mL pepsin solution to the cell layer (150 μL/well of 24-well plate). 4. Incubate the plate at 37  C for 2 h with continuous shaking at 200 rpm using a Thermo Scientific, MaxQ 4000 Benchtop Orbital Shaker, or equivalent. 5. After 2 h of incubation, scrape off the cell layer portion using 1 mL pipette tips and transfer to pre-labeled tubes. Cell layer sample from 3 wells can be pooled in a 1.5 mL microcentrifuge tube. 6. Add 22.5 μL of phenol red solution to 450 μL of cell layer sample. The samples will turn into yellow color. Add 22.5 μL of 1 N NaOH to 450 μL of cell layer sample. Repeat this step until the samples will turn into pink color (see Note 16). 7. Vortex briefly. 8. Store at 4  C for short-term storage or at 20  C for long-term storage.

3.2.2 Preparation and Running of SDS-PAGE

1. Prepare 250 μg/mL concentration collagen type I in 0.5 M acetic acid. Prepare the standard by adding 42 μL of ddH2O, 12 μL of 5 sample buffer, 2 μL of 1 N NaOH, and 4 μL of standard (250 μg/mL) into a 1.5 mL microcentrifuge tube to get 15:1 dilution. 2. Prepare cell layer samples by adding 24 μL of ddH2O, 12 μL of 5 sample buffers, and 24 μL of each cell layer sample into

Macromolecular Crowding in Human Corneal Fibroblast Culture


Table 1 5% Separation Gel (1 mm thickness) for collagen for mini gel Component

1 Gel (μL)

30% Acrylamide/Bis (37.5:1) 1.875 M Tris–HCl pH 8.8 10% SDS ddH2O APS (100 mg/mL) TEMED Total

2 Gels (μL)















Table 2 3% Stacking Gel (1 mm thickness) for collagen for mini gel Component

1 Gel (μL)

2 Gels (μL)

30% Acrylamide/Bis (37.5:1)



1.875 M Tris–HCl pH 8.8













10% SDS ddH2O APS (100 mg/mL) TEMED Total

1.5 mL microcentrifuge tubes to get 2.5:1 dilution (see Note 17). 3. Vortex the samples and centrifuge them briefly. Store them at 4  C. 4. Prior to running SDS-PAGE, denature the samples and standard by heating at 95  C for 5 min. 5. Vortex and then centrifuge the samples briefly. 6. Take out of 20  C an aliquot of APS to thaw (put it back at 20  C at the end). 7. Clean glass plates with 70% ethanol and wipe dry with microscope tissue papers. 8. Set the gel making apparatus ensuring that the glass plates fit snugly to the platform (mini gel: 1 mm space using appropriate spacers) (see Note 18). 9. Add the gel ingredients to make the 5% resolving gel according to the Table 1. This can be done in a 15 mL centrifuge tube.


Mehmet Gu¨rdal et al.

10. Make sure to add the APS and TEMED last, right before the gels are to be poured. 11. Using a pipette, pour the prepared mixture carefully into the space between the two glass plates to reach about 1 cm (mini gel) from the bottom of the wells etched out by the comb (keep the excess solution to check how quickly the gels will be polymerized). 12. Overlay the gel with 10% ethanol to cut off oxygen in contact with the gels. 13. Leave it aside for approximately 30 min (check with the excess solution remained). 14. During the setting period, prepare the 3% stacking gel according to Table 2 (Do not add the APS and TEMED until the gel is ready for pouring). 15. A line at the ethanol-gel interface that initially had disappeared will reappear after the 30 min period indicating that polymerization is complete. 16. Carefully aspirate the ethanol out of the glass plates using a syringe and imbibe any traces using filter paper. 17. Now add the APS and TEMED to the stacking gel and carefully pour it on top of the polymerized resolving gel. Immediately insert the comb taking care to avoid trapping any air bubbles (keep the excess solution to check how quickly the gels will be polymerized). 18. Allow it to set for 10–15 min and, in the meantime, denature samples and standard at 95  C as described above. 19. After the gels have been set (10–15 min, check it with the excess solution), remove slowly the combs. 20. Assemble the electrophoresis apparatus; for small gel apparatus, fit the gel plates on the electrode bar and fit the set into the inner chamber and clamp them. 21. Fill the upper/inner chamber with 1 running buffer. 22. Wash the wells by squirting buffer into the wells with a hypodermal needle syringe to remove all air bubbles. 23. Load the standard and samples using a 50 μL Hamilton syringe. Wash the syringe in between using the running buffer in the chamber (at least 5-times). Load 10 μL per well of 15-well. Load 15 μL of 1 sample buffer per empty well of 15-well. 24. Put the upper chamber on the main chamber, close the lid, and run the gel(s). 25. For the mini gel ! run at constant voltage: 50 V until the front reaches the end of the stacking gel (30–40 min), then 120 V until the front reaches the end of the separating gel (1 h).

Macromolecular Crowding in Human Corneal Fibroblast Culture


26. Remove the glass using a Hoefer wonder wedge, cut the lower right-hand corner, and release the gel slowly into ddH2O. 27. Proceed Silver staining. 28. Add 50 mL fixing solution onto the gel. Incubate the gel for 20 min at room temperature with slight shaking. After incubation, remove the fixing solution. 29. Wash the gel with 50 mL of 30% (v/v) ethanol for 10 min at room temperature with slight shaking. After washing, remove the solution. 30. Add 50 mL sensitizer solution onto the gel. Incubate the gel for 10 min at room temperature with slight shaking. After incubation, remove the sensitizer solution. 31. Wash the gel with 50 mL of 30% (v/v) ethanol for 10 min at room temperature with slight shaking. After washing, remove the solution. 32. Wash the gel with 50 mL of ddH2O for 10 min at room temperature with slight shaking. After washing, remove the solution. 33. Add 50 mL staining solution onto the gel. Incubate the gel for 15 min at room temperature with slight shaking. After incubation, remove the staining solution. 34. Wash the gel with 50 mL of ddH2O for 1 min at room temperature with slight shaking. After washing, remove the solution. 35. Add 50 mL developing solution onto the gel. Incubate the gel for 4–8 min (see Note 19) at room temperature with slight shaking. 36. Add 5 mL stopper solution directly to developing solution. Incubate the gel for 10 min at room temperature with slight shaking. After incubation, remove the developing solution. 37. Wash the gel with 50 mL of ddH2O for 10 min at room temperature with slight shaking. 3.2.3 Protein Band Quantification Using ImageJ (Densitometric Analysis)

1. Scan the SDS-PAGE gels on Scanner with Active Transparency Adapter (Fig. 2a). 2. Measure the band density of α1(I) and α2(I) bands with ImageJ software. 3. Open ImageJ. 4. Go to File ! Open ! (your image). 5. If the image looks too dark or too light go to Image ! Adjust ! Brightness/contrast. 6. Save the image with an updated name. 7. Go to Analyze ! Set measurements ! Tick the following boxes.


Mehmet Gu¨rdal et al.

Fig. 2 Collagen I deposition at the cell layers in the absence and presence of increasing concentrations of macromolecular crowder carrageenan (CR). SDS-PAGE (a) and complementary densitometric analysis (b) revealed that in the presence of 50 μg/mL and 100 μg/mL CR, the highest amount of collagen I was deposited. In addition, 50 μg/mL CR deposited the highest amount of collagen I at day 10. Data shown are mean  3. ∗P < 0.05, ∗∗P < 0.0001

8. Area, Mean Gray Value, Standard Deviation. 9. Select the rectangle tool, and draw a box around the lane. 10. Go to Analyze ! Measure. 11. A new box with all the results will appear. 12. Repeat steps 8 and 9 for all the bands of your interest. 13. Be consistent with the area of rectangle box. 14. Copy all results and paste into Microsoft Excel sheet. 15. Add the mean band intensity of α1(I) with α2(I) band.

Macromolecular Crowding in Human Corneal Fibroblast Culture


16. Normalize the mean value of collagen type I standard. (For example, if the mean band density is 500 for 250 μg/mL of collagen type I standard, multiply all the mean value of sample with 2). 17. Plot the bands intensity as μg/mL of sample and standard bands (Fig. 2b). 3.3 Cellular Metabolic Activity

1. Prepare a 10% alamarBlue® solution in HBSS. 2. Remove culture medium from the cells that are in 24-well plate and wash with HBSS. 3. Add 500 μL of the diluted alamarBlue® solution to the cells and a negative control of alamarBlue® at 10% alone. 4. To obtain the background absorbance, add HBSS to empty wells. 5. Incubate for 3 h at 37  C, 5% CO2. 6. Transfer 100 μL of the alamarBlue® solution and of the negative control and background to a clear 96-well plate. 7. Measure the absorbance at 550 nm and at 595 nm using a microplate reader. 8. Subtract the values of HBSS to the values of alamarBlue® alone from both absorbances to obtain the absorbance of alamarBlue®. For 550 nm this value is called absorbance of the oxidized form at lower wavelength (AOLW) and for 595 nm it is called absorbance of the oxidized form at higher wavelength (AOHW). 9. Calculate the correlation factor: Ro ¼ AOLW/AOHW. 10. To calculate the percentage of alamarBlue® reduced (AR) by the cells, use the following: AR ¼ ALW  (AHW  Ro)  100 (Fig. 3a).


Cellular Viability

1. Prepare staining solution by diluting calcein AM to 4 μM and ethidium homodimer-1 to 2 μM in HBSS. 2. To prepare a negative control, sample can be immersed in DMSO to kill all cells before staining. 3. Remove culture medium from the cells and wash cells with HBSS. 4. Add staining solution to cells (enough volume to cover completely the sample). 5. Incubate at 37  C, 5% CO2 for 30 min. 6. Image under inverted fluorescence microscope: For Calcein AM use FITC filter, for Ethidium homodimer-1: use Texas Red filter (Fig. 3b).


Mehmet Gu¨rdal et al.

Fig. 3 Cell metabolic activity (a), viability (b), and proliferation (c) of HCFs in the absence and presence of 50 μg/mL CR (concentration that induced the highest collagen I deposition at day 10, as per SDS-PAGE and complementary densitometric analysis). Macromolecular crowding (50 μg/mL CR) did not affect cellular metabolic activity (a), viability (b), and DNA content (c) at a given time point as revealed by alamarBlue assay, Live/Dead assay, and Quant-iT PicoGreen dsDNA assay, respectively. Data shown are mean  SD. The N values for each group are 3 3.5

Cell Proliferation

1. Remove the media from the cells that are in 24-well plate and gently rinse the cells with HBSS. 2. Add 250 μL of DNase-free water. 3. Freeze-thaw cells three times (freeze at 80  C for 15 min minimum and thaw at room temperature until it is completely defrosted). 4. Prepare a standard curve in a 96-well plate in accordance with Table 3 using 1 TE buffer (initial solution at 20, dilute in DNase-free water). 5. Make up 2 μg/mL DNA stock solution from 100 μg/mL DNA standard and 50 ng/mL DNA stock solution from 2 μg/mL stock solution. 6. Transfer 100 μL of each sample in the 96-well plate. 7. Make up diluted PicoGreen® solution by adding 5.376 mL 1 TE and 27 μL concentrated PicoGreen® (enough for the standard curve in triplicate and 24 samples). 8. Add 100 μL of diluted PicoGreen® to each well. 9. Incubate at room temperature for 2–5 min in the dark. 10. Read the plate for fluorescence (excitation: 480 nm, emission: 520 nm). 11. Plot a graph concentration vs. the fluorescence values. Determine the concentration of DNA as a function of the standard curve (Fig. 3c).

Macromolecular Crowding in Human Corneal Fibroblast Culture


Table 3 Standard curve for PicoGreen Final DNA concentration (ng/mL)

1 TE Volume (μL)

Volume of 2 μg/mL stock DNA solution (μL)

Volume of 50 ng/mL stock DNA solution (μL)

























3.6 Immunofluorescent Characterization of the ECM

All the following steps must be performed at room temperature (unless otherwise stated). 1. At the end of each time point, aspirate the cell culture media of the corresponding 48-well plate and wash each well three times during 5 min with 500 μL of 1 sterile HBSS. Fix the samples with 150 μL/well of the filtered fixation solution (4% PFA) pre-cooled at 4  C for 15 min (see Note 20). 2. Remove the fixation solution and wash the wells three times during 5 min with 250 μL of 1 PBS. Samples can be kept at 4  C in 1 PBS if immunofluorescent staining is not to be performed right away. 3. Add 150 μL per well of BSA blocking solution for 30 min to block unspecific binding sites for the primary antibody. 4. After the aforementioned 30 min, drain the blocking solution and incubate the samples with 80 μL per well of the corresponding primary antibody solution for overnight at 4  C. For negative control wells, add 80 μL of 1 PBS per well (see Note 21). 5. Remove the primary antibody solution and wash the wells three times during 5 min with 250 μL of 1 PBS (see Note 22). Incubate with 80 μL per well of the secondary antibody solution for 1 h. In this case, all the samples including the negative controls must be incubated with the secondary antibody solution (see Note 23). Protect the samples from light from here on.


Mehmet Gu¨rdal et al.

6. Remove the secondary antibody solution and wash three times during 5 min with 250 μL of 1 PBS. 7. Remove the PBS and incubate the samples for 5 min with 80 μL per well of diluted Hoechst 33342 solution in order to stain nuclei. 8. Remove the Hoechst 33342 solution and wash three times during 5 min with 250 μL of 1 PBS. 9. Add 80 μL per well of 1 PBS in each well. 10. Analyze the samples by inverted fluorescence microscope equipped with 10 objective, filters suitable for Hoechst 33342 (Ex/Em: 358/461 nm), AlexaFluor®488 (Ex/Em: 495/519 nm), and a connected camera. 3.6.1 Imaging of Immunofluorescent Staining

Perform the steps relative to the image acquisition process in dark, in order to prevent the loss of fluorescence intensity from the fluorophores as a consequence of light excitation. 1. Once the software has been launched, position the plate on the sample stage of the microscope and visualize a live preview from the camera input; be sure that the light path selector of the microscope is set on “camera” to view the image on screen. 2. It is pivotal to keep the same exposure time for all the samples examined for the same antigen (see Note 24). Adjust the exposure time for Hoechst 33342, and carefully adjust the focus on the cell layer. Select the correct filter to excite the fluorophore bound to the secondary antibody of interest, regulate the exposure time in order to avoid saturation, and take note of the longest exposure time void of saturation. Repeat these actions for all the samples, excluding the negative controls, then set the lowest obtained exposure time value to examine all of the samples and controls. 3. Position the plate on the sample stage exposing the first sample to the light path, select the violet filter, and acquire a 10 magnified picture of the nuclei stained with Hoechst 33342. Without moving the sample, turn the filter block turret to the appropriated filter for the fluorophore bound to the secondary antibody of interest, regulate the focus, and acquire the image. Repeat this step five times in randomly selected fields of the well for each replicate of each condition. 4. Using the emission filter for the secondary antibody of interest, acquire five pictures from five randomly selected fields of the non-primary antibody negative controls (see Note 25). Save all the acquired pictures in TIFF format before proceeding with the analysis of fluorescence intensity, which can be performed with the ImageJ software.

Macromolecular Crowding in Human Corneal Fibroblast Culture


5. To create a representative picture of the assay, open the desired image in ImageJ software together with the corresponding Hoechst 33342 field. For both of the pictures, open the “Image” menu, select “Type” and click on “32-bit.” Once all the images that need to be merged in a single picture have been turned to 32-bit format, open the “Image” menu, select “Colour,” then “Merge channels”: match each picture with the desired channel, tick the “create composite” box and click “Ok” to obtain the composite picture (see Note 26). Once the picture has been saved in the preferred format, it is possible to add a scale bar using the dedicated ImageJ tool (see Note 27) (Fig. 4).

Fig. 4 Immunofluorescent labeling of different extracellular matrix proteins in HCFs for each time point in the absence and presence of macromolecular crowding (50 μg/mL CR; concentration that induced the highest collagen I deposition at day 10, as per SDS-PAGE and complementary densitometric analysis). Macromolecular crowding enhanced collagenous proteins (I, IV, V, VI), which are the main components of corneal stroma, in the cell layer of HCFs at each time point



Mehmet Gu¨rdal et al.

Notes 1. Fibroblast medium kit is a kit purchased from Innoprot (Spain) that includes 500 mL fibroblast basal medium, 10 mL fetal bovine serum (FBS), 5 mL fibroblast growth supplement (FGS), and 5 mL penicillin/streptomycin solution. 2. Poly(L-lysine)-coated 175 cm2 flask: 3 mL 0.1 mg/mL poly(Llysine) solution is transferred to 175 cm2 flask. The flask is gently rocked to evenly coat the surface. After 5 min, the excess solution is removed and the surface is thoroughly rinsed with sterile ddH2O and allowed to dry for several hours. 3. The efficacy of UV-C disinfection depends on many parameters, including the intensity and wavelength of the UV radiation, the time of exposure, the distance from the source of irradiation, the presence of particles that can protect the microorganisms from UV, and a microorganism’s ability to withstand UV during its exposure. We currently disinfect the carrageenan with success inside of an open 1.5 mL microcentrifuge tube in vertical position with the UV lamps typically included in the biological safety cabinets used for cell culture, but more dedicated equipment can also be used with the same efficacy if similar conditions are met. 4. Once the carrageenan is suspended into the medium, if the tube is vortexed and observed through a light source, particles in suspension can be appreciated. Once the carrageenan is completely dissolved into the medium, no particles can be appreciated following the same observation procedure. It is very important to ensure the completely dissolution of the carrageenan into the medium as non-dissolved particles can deposit on the bottom of the plates and negatively affect the cell viability. 5. Dissolving sodium hydroxide in water creates an exothermic reaction, which produces large amounts of heat. Concentrated sodium hydroxide may result in boiling; thus, the flask should not be touched, and personal protective equipment (PPE) is required. 6. SDS is a respiratory, skin, and eye irritant. It affects fertility. It is harmful if swallowed, inhaled, or absorbed through the skin. The use of PPE is recommended. Weigh SDS under a fume hood to avoid eye and skin contact and inhalation. 7. APS is highly toxic. It causes respiratory, skin, and eye irritation. It is harmful if swallowed. The use of PPE is recommended. Weigh APS under a fume hood to avoid eye and skin contact and inhalation.

Macromolecular Crowding in Human Corneal Fibroblast Culture


8. TEMED is highly flammable and it is harmful if swallowed or inhaled. The use of PPE is recommended. Work with TEMED under a fume hood. 9. SilverQuest™ Silver Staining Kit includes Sensitizer, Stainer, Developer, Developer Enhancer, and Stopper solution. 10. Avoid temperatures over 60  C (optimal are between 55–57  C), which result in methanol formation and the reduction of the effective concentration of PFA. This can damage the cytoskeleton of the cells and increase the auto-fluorescence of the samples. 11. For safety reasons, make the fixation solution under a chemical hood and cover the glass beaker, with aluminum foil for example, to prevent the release of PFA toxic vapors and the evaporation of the PFA. Likewise, all the steps that involve the use of the fixation solution must be executed under a chemical hood due to the PFA toxicity. 12. To perform the localization of several molecules in the same sample by indirect immunofluorescent staining, it is required for the primary antibodies against the antigens of interest to be raised in different species in order to be able to be recognized by different specific secondary antibodies (e.g., the primary antibody anti-collagen type I has been raised in mouse and can be combined with any of the other primary antibodies raised in rabbit for their use in multicolor immunostaining). 13. For the selection of the secondary antibodies, several guidelines must be followed. First, the secondary antibody must be specific for the target species corresponding to the primary antibody (e.g., goat anti-mouse secondary antibodies for primary antibodies derived from mice and goat anti-rabbit secondary antibodies for primary antibodies derived from rabbit). Moreover, to avoid cross-reactivity with the non-intended primary antibody targets in multicolor immunostaining and increase specificity, secondary antibodies cross-adsorbed against IgGs from the nontarget species and sera from different species can be used (e.g., for the present study, we used goat anti-mouse secondary antibody cross-adsorbed against human serum and rabbit and goat IgGs, as well as goat anti-rabbit secondary antibody cross-adsorbed against mouse and goat IgGs). The selection of the fluorochromes conjugated with the secondary antibodies and the appropriated filter set it is crucial to have the minimal signal overlapping and maximal specificity in case of detecting several molecules at the same time in multicolor immunofluorescent stainings. For this, it is very important to excite each fluorochrome with wavelengths as close as possible to its maximal excitation, avoiding at the same time to excite the other fluorochromes present in the


Mehmet Gu¨rdal et al.

sample. Also, for collecting the fluorescence of each fluorochrome specifically, the emission filter should match as close as possible the maximal emission wavelength of the corresponding fluorochrome, avoiding at the same time the emission of the other fluorochromes present in the sample. 14. Protect the samples from the light to avoid the loss of fluorescent signal due to photobleaching. The photobleaching effect (also termed as fading) causes a permanent photochemical alteration, by cleaving covalent bonds in the fluorochromes. These irreversible modifications make the fluorochrome unable to emit fluorescence and thereby decrease the signal in the samples. 15. The HCFs that can be acquired from Innoprot (Sapin) are generally in early passages (p0 - p1). To thaw the cells, wear protective gloves and face shield and take out the tube containing the frozen cells from liquid nitrogen cylinder. Thaw the contents of cryotube by rubbing in palm or water bath at 37  C. Transfer the contents of the tube in a 15 mL conical tube containing 10 mL of maintaining growth medium inside the laminar air flow hood. Centrifuge the tube at 360  g for 5 min. Remove the supernatant and discard it. Add 1 mL of pre-warmed (37  C) maintaining growth medium into the same falcon tube and mix properly with gentle aspiration to distribute the cells homogeneously. Count the cells using hemocytometer. Transfer appropriate amount of the cell suspension (5000 cells/cm2) to a poly(L-lysine)-coated T75 flask containing 9 mL (20 mL for T175) of pre-warmed maintaining growth medium. Label the flasks with name, date, cell type, and passage no. The subculture routine included washing with 5 mL (10 mL for T175) of 1 HBSS, incubating with 1 mL (2 mL for T175) of trypsin-EDTA solution for 5 min at 37  C, cell recovery with 3 mL (6 mL for T175) of maintaining growth medium per each T75 flask, centrifugation for 5 min at 360  g, resuspension, counting, and plating at 5000 cells/ cm2 to new poly(L-lysine)-coated flask. Normally, the cells reach confluency after a week of culture, and medium is changed thrice per week, using 9 mL (20 mL for T175) per T75. For cell cryopreservation, a cell density of 500,000 cells/ mL of 10% DMSO in growth medium is used. In this study, passage 4 cells were used for MMC experiments. 16. The amount of NaOH solution used for neutralization is sample-dependent (e.g., dependent upon the acidity of the solution) and therefore may require optimization. 17. To make densitometric analyze, dilution factor of collagen type I standard will be 15 and dilution factor of each sample will be 2.5.

Macromolecular Crowding in Human Corneal Fibroblast Culture


18. Check for any leaks by pouring ddH2O into the gel-making apparatus and then removing the ddH2O with filter paper before making the gel. 19. After adding the developing solution onto the gel, closely observe the bands as they form within 0 to 8 min. Once all bands become completely visible, immediately add the stopper solution to prevent a dark background. 20. Be extremely careful during fixation and washing steps to prevent the detachment of the cell layer. 21. Alternatively, incubation with the primary antibody solutions can be performed 90 min at room temperature. 22. When working with a high number of experimental groups or samples, always proceed with the washes in a fractional way in order to avoid drying the samples. Drying may cause a background increase due to unspecific binding of the primary antibody to the sample. 23. In order to demonstrate the nonspecific binding of the secondary antibodies to the sample, the non-primary antibody control wells should be incubated with the corresponding secondary antibody solution. For multicolor immunostaining, these wells should be incubated with a solution containing the mixed secondary antibodies as applied to the sample wells. 24. The selection of an optimal exposure time suitable for all the examined samples, defined as the maximal exposure time at which the sample with the highest fluorescence intensity does not show signs of saturation, will allow for the comparison of the collected data between the different samples. The acquisition of an excessively saturated image would result in an underestimation of fluorescence intensity. 25. Background fluorescence can result from several factors, such as nonspecific binding of the secondary antibodies, or autofluorescence from the culture plate, the cultured cells, and from collagen itself. The quantification of non-primary antibody controls fluorescence is therefore performed in order to estimate the amount of this nonspecific fluorescence, which will eventually be subtracted from the mean gray value measurements of the corresponding samples. 26. It is possible to merge more than two pictures, or channels, in this step, in order to appreciate the relative localization of the examined antigens. To reduce the background interference that can derive from nonspecific binding of the antibodies as well as from autofluorescence of several biological molecules, or even from the plastic substrate, the “Subtract background” ImageJ tool can be useful. To use it, right after the images have been turned to 32-bit format, open the “Process” menu, click


Mehmet Gu¨rdal et al.

on “Subtract background,” select the same rolling ball radius for every picture in the assay, and confirm with “OK,” then proceed with the merging of the channels. Choose a larger rolling ball radius to maximize the preservation of positive pixels intensity, or a smaller one to maximize background subtraction. 27. There are several options to add a scale bar with ImageJ. If the pixel/length ratio is known, it is possible to open the “Analyze” menu, select “Set scale,” type “1” in the “Pixels” box and type the corresponding distance in the “Known distance” box, then specify the unit of length and confirm with “OK.” If the pixel/length ratio is not known, it is possible to use the “Straight” tool from ImageJ toolbar in order to trace a segmented line covering a known distance in the picture, then open the “Set scale” window and fill the “Known distance” and “unit of length” boxes, then confirm with “OK.” A third option provides that a second picture, with the same pixel/ length ratio of the former and already containing its scale bar is open and selected in ImageJ: in such case, it is only needed to tick the “Global” box in the “Set scale” window before clicking “OK” to apply the same scale to all the open pictures. To finally add the scale bar to the desired picture, all the listed options provide that the “Analyze” menu is open, in order to select “Tools > Scale bar,” personalize the scale bar and confirm with “OK.”

Acknowledgements This work has been supported from: Science Foundation Ireland, Career Development Award Programme (grant agreement number: 15/CDA/3629) and Science Foundation Ireland and the European Regional Development Fund (grant agreement number: 13/RC/2073). Mehmet Gu¨rdal was supported by The Scientific € ˙ TAK), Sciand Technological Research Council of Turkey (TUBI ence Fellowships and Grant Programmes Department (BI˙DEB), Programme of 2214-A Ph.D. Research Scholarship for Abroad. The authors have no competing interests. References 1. Fini ME, Stramer BM (2005) How the cornea heals: cornea-specific repair mechanisms affecting surgical outcomes. Cornea 24(8 Suppl): S2–S11 2. Radner W, Zehetmayer M, Aufreiter R, Mallinger R (1998) Interlacing and cross-angle

distribution of collagen lamellae in the human cornea. Cornea 17(5):537–543 3. Fini ME (1999) Keratocyte and fibroblast phenotypes in the repairing cornea. Prog Retin Eye Res 18(4):529–551 4. Hay ED (1980) Development of the vertebrate cornea. Int Rev Cytol 63:263–322

Macromolecular Crowding in Human Corneal Fibroblast Culture 5. Kumar P, Pandit A, Zeugolis D (2016) Progress in corneal stromal repair: from tissue grafts and biomaterials to modular supramolecular tissue-like assemblies. Adv Mater 28 (27):5381–5399. adma.201503986 6. Guillame-Gentil O, Semenov O, Roca AS, Groth T, Zahn R, Voros J, Zenobi-Wong M (2010) Engineering the extracellular environment: strategies for building 2d and 3d cellular structures. Adv Mater 22(48):5443–5462. 7. Peck M, Dusserre N, McAllister TN, L’Heureux N (2011) Tissue engineering by selfassembly. Mater Today 14(5):218–224. 70117-1 8. Canty EG, Kadler KE (2005) Procollagen trafficking, processing and fibrillogenesis. J Cell Sci 118(Pt 7):1341–1353. 1242/jcs.01731 9. Sorushanova A, Delgado L, Wu Z, Shologu N, Kshirsagar A, Raghunath R, Mullen A, Bayon Y, Pandit A, Raghunath M, Zeugolis D (2019) The collagen suprafamily: from biosynthesis to advanced biomaterial development. Adv Mater 31(1):e1801651. 10.1002/adma.201801651 10. Kumar P, Satyam A, Fan X, Collin E, Rochev Y, Rodriguez BJ, Gorelov A, Dillon S, Joshi L, Raghunath M, Pandit A, Zeugolis DI (2015) Macromolecularly crowded in vitro microenvironments accelerate the production of extracellular matrix-rich supramolecular assemblies. Sci Rep 5:8729. srep08729 11. Kumar P, Satyam A, Fan X, Rochev Y, Rodriguez BJ, Gorelov A, Joshi L, Raghunath M, Pandit A, Zeugolis DI (2015) Accelerated


development of supramolecular corneal stromal-like assemblies from corneal fibroblasts in the presence of macromolecular crowders. Tissue Eng Part C Methods 21(7):660–670. 0387 12. Graceffa V, Zeugolis D (2019) Carrageenan enhances chondrogenesis and osteogenesis in human bone marrow stem cell culture. Eur Cell Mater 37:310–332. 22203/eCM.v037a19 13. Graceffa V, Zeugolis D (2019) Macromolecular crowding as a means to assess the effectiveness of chondrogenic media. J Tissue Eng Regen Med 13(2):217–231. 10.1002/term.2783 14. Cigognini D, Gaspar D, Kumar P, Satyam A, Alagesan S, Sanz-Nogues C, Griffin M, O’Brien T, Pandit A, Zeugolis DI (2016) Macromolecular crowding meets oxygen tension in human mesenchymal stem cell culture - a step closer to physiologically relevant in vitro organogenesis. Sci Rep 6:30746. 10.1038/srep30746 15. Satyam A, Kumar P, Fan X, Gorelov A, Rochev Y, Joshi L, Peinado H, Lyden D, Thomas B, Rodriguez B, Raghunath M, Pandit A, Zeugolis D (2014) Macromolecular crowding meets tissue engineering by selfassembly: a paradigm shift in regenerative medicine. Adv Mater 26(19):3024–3034. https:// 16. Gaspar D, Fuller K, Zeugolis D (2019) Polydispersity and negative charge are key modulators of extracellular matrix deposition under macromolecular crowding conditions. Acta Biomater 88:197–210. 1016/j.actbio.2019.02.050

Chapter 10 Preparation of Dried Amniotic Membrane for Corneal Repair Andrew Hopkinson, Emily R. Britchford, and Laura E. Sidney Abstract Amniotic membrane transplantation is an established therapeutic and biological adjunct for several clinical situations, including treatment of diabetic foot ulcers and ocular surface disease. However, poorly standardized and validated clinical preparation and storage procedures can render the final product highly variable and an unpredictable biomaterial. We have therefore developed a novel, standardized method for processing and dry-preserving amniotic membrane, minimizing biochemical, compositional, and structure damage to produce a potentially superior membrane suitable for clinical use. The intellectual property associated with this methodology was patented by the University of Nottingham and licensed to NuVision® Biotherapies which formed the basis of the Tereo® manufacturing process which is used to manufacture Omnigen®. Key words Amniotic membrane, Standardization, Preparation, Transplantation, Cornea, Repair


Introduction Amniotic membrane (AM) is the innermost layer of the placental membrane that surrounds the fetus during pregnancy. At birth, the placenta is normally discarded as waste; however, following elective caesarean sections, the AM can be harvested and processed into an allograft transplant material for the repair of many soft-tissue damage situations, including the ocular surface. The clinical objectives of AM transplantation (AMT), particularly in corneal regeneration, are to: (1) support and replace damaged tissue, (2) protect the defect from further degeneration from external factors, and (3) promote effective recellularization [1]. This is achieved through a combination of beneficial biological properties, including immune-privilege (minimizing the risk of an immune response) [2], ability to modulate and inhibit inflammation [3, 4], angiogenesis [3, 5, 6], fibrosis [7], tissue adhesions [7, 8], and cancer [9–13], while preserving and nurturing the host’s stem cell population [14, 15]. Combined, these biological properties facilitate recellularization to accelerate wound healing [28]. The physical

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Andrew Hopkinson et al.

nature of AM, as a biological bandage, also significantly improves pain [16–19], and delivers antimicrobial benefits [20–22]. Since its first clinical use as a skin transplant in 1910 [23], AM has been adopted in numerous soft-tissue transplantation modalities, including restoration of hearing [24], grafting of the abdominal cavity [25, 26], and replacement of the vaginal [27, 28], urethral [29], oral, and pharyngeal mucous membranes [30, 31]. Moreover, AMT has been widely adopted in the treatment of ocular surface disease, as well as more recently for the treatment of chronic wounds [32]. Application of AM in the treatment of corneal disease has been reported to accelerate wound healing [33], especially in infected corneas by reducing infection-causing microorganisms [34]. AMT has been shown to be superior to antibiotic treatment alone [35, 36]. Although the mechanism of action of AM is still not fully understood, it is broadly attributed to bioactive substances such as , inflammatory-mediators, proteoglycans, such as hyaluronic acid, and other classifications of beneficial proteins that enhance the function of structural constituents, working synergistically as a complex matrix to support local cell survival, adhesion, proliferation, and migration [37]. An effective method for processing and preserving AM for clinical application must therefore aim to preserve the integrity of the tissue without compromising the structural and biochemical composition, and therefore therapeutic potential of the AM. Freshly collected AM is rarely used directly for clinical application due to an increased risk of disease transmission, posttransplant complications [38], and a very short shelf life. The use of fresh AM is also not permitted by many regulatory authorities. Therefore, commercial clinical AM products are subject to various processing and preservation steps [37]. The conventional methods for preserving AM are: (1) cryopreservation (extra-cold freeze-preservation at around 80  C) [39], (2) freeze-drying (lyophilization of 80  C frozen tissue via sublimation) [40], and (3) heat (up to 50  C) [41] or air-dehydration. Despite the use of cryoprotectants such as glycerol or dimethyl sulfoxide (DMSO), cryopreservation (wet-preservation) procedures have been shown to damage the AM [42], which results in a significant decrease in the levels of many bioactive substances, including important wound healing factors, such as epidermal growth factors (EGF) and transforming growth factors (TGF) [6, 42–46]. Cryopreservation also creates major cold-chain storage and shipping, shelf life, and quality assurance challenges. Furthermore, cryopreserved AM must be thawed and washed to remove cryoprotectants before clinical use, which further depletes beneficial proteins from the damaged tissue. Though offering practicality, storage, and logistic advantages, heat-dried AM followed by radiation-sterilization has been shown to also reduce growth factor content [40], increased fragility, and compromised efficacy. More recently, the use of freeze-dried AM

Dried Amniotic Membrane Preparation


has been proposed in combination with sterilization, and in the presence of novel complex saccharide lyoprotectants [47]. Unfortunately, this involves aforementioned deleterious freezing of the tissue before drying. Further challenges of dry-preserving AM is friability of tissues in the dry state, and subsequently effective rehydration of the material to restore normal physical, biochemical, and functional characteristics. This chapter describes an alternative process for AM preservation that overcomes the issues of conventional wet- and dry-preservation methods, to produce a standardized and highquality AM substrate, which is robust and malleable in the dehydrated state, and rapidly and effectively rehydrates when applied to the wound in vivo. AM is delicately dehydrated using a unique freeze- and heat-free low temperature vacuum evaporation (Patent WO2014195699A3, licensed to NuVision Biotherapies Ltd) process which preserves the integrity and biochemistry of AM, comparable to fresh AM, as a devitalized stable stromal matrix. During pregnancy and whilst in vivo, AM is considered sterile; however, potential microbial “bioburden” contamination can be introduced from skin and vaginal flora during delivery of the placenta [48]. Though elective caesarean section reduces the risk of bioburden, the AM manufacturing process must involve measures to effectively decontaminate potential bioburden and allow for long-term preservation of AM. Due to potential degradation by terminal (radiation) sterilization techniques [49], AM can be prepared aseptically, involving antibiotics, in a Good Manufacturing Practice (GMP) environment in order to ensure natural bioburden is eliminated and no further contamination is introduced during processing. In this protocol, decontamination is achieved through a combination of: (1) an aseptic procedure, (2) a sequential washing process involving a potent antibiotic cocktail decontamination step, and (3) a drying procedure. Recent research at the University of Nottingham has demonstrated that this combined process can be considered a sterilization process [50, 51]. This chapter describes the research preparation of dried AM using a standardized laboratory manufacturing process. The final product is produced by isolating the amniotic layer from the placenta, chorion, and spongy layer (SL) prior to washing and drying. The salient features of this technique are: l

Human Tissue Authority (HTA) approved protocol for clinical transplantation of AM on human ocular surface (see Note 1).


Minimized inter- and intra-donor variation between AM preparations.


Unique spongy layer removal process (Patent GB1441939, licensed to NuVision Biotherapies Ltd).


A xeno-free preparation system suitable for clinical, in vivo, and in vitro research.


2 2.1

Andrew Hopkinson et al. l

Unique dehydration process.


A dried AM matrix as an ideal tissue engineering substrate for direct ex vivo co-culturing of various cell types.

Materials Reagents

1. 0.9% Sodium chloride (NaCl): 9 g NaCl in 1 L H2O. 2. Phosphate buffered saline (PBS): 72 mg potassium phosphate monobasic (KH2PO4), 397.5 mg sodium phosphate dibasic (Na2HPO4), and 4500 mg NaCl in 0.5 L H2O, pH 7.4. 3. Antibiotic/antifungal concentrate 5 mL: gentamicin sulfate, imipenem, nystatin dihydrate, polymyxin B sulfate, vancomycin hydrochloride (broad-spectrum).

2.2 Disposable Instruments

1. 175 cm2 (T175) straight neck vent flask. 2. Scissors Supersnip B/B sterile. 3. Adson forceps (Bayonet) non-toothed 18 cm. 4. Bio assay dish standard. 5. Syringes 50 mL. 6. Serological pipette 25 mL. 7. Pipette aid. 8. Sartorius syringe filters 0.22 μm. 9. Blades No. 22 sterile. 10. Disposable scalpels No. 22 sterile. 11. Surgical spongy spears. 12. Bold line surgical pen. 13. Parafilm® M roll.



1. Class II laminar air flow hood (biological safety cabinet). 2. Rocker. 3. Incubator. 4. Freeze Dryer.

2.4 Additional Materials

1. Gloves. 2. Sharps bin. 3. Marker pen. 4. Cleaning tissues. 5. Clinical waste bags. 6. Black general waste bags. 7. Cleaning reagents (70% IMS).

Dried Amniotic Membrane Preparation



Methods The preparation of all materials and methods should be performed in a class II laminar air flow hood (safety cabinet) following Good Laboratory Practice (GLP), unless otherwise stated for clinical use (see Note 2). Remove all packaging inside the safety cabinet to maintain sterility and cleanliness.

3.1 Tissue Collection Flask

Prepare the tissue collection flask (TCF) prior to tissue collection. 1. Transfer the following materials to the safety cabinet: (a) 1  T175 flask (b) 1  1 L NaCl (c) Marker pen 2. Clearly label the TCF using the marker pen. 3. Mark the 250 mL line and decant 250 mL NaCl into the flask. 4. Pre-weight the Raffinose powder (D-(+)-Raffinose pentahydrate 99 + %) in a T175 flask (14.86 g). Label “RAFFINOSE” using the marker pen (see Note 3).

3.2 Donor Consenting and Amniotic Membrane Tissue Procurement

1. Identify and consent potential donors undergoing elective caesarean sections (donor informed consent is mandatory before processing) (see Notes 4 and 5). 2. Prepare a collection kit by placing the following items into a collection box: (a) 1  TCF (prepared in Subheading 3.1) (b) 1  Scissors (c) 1  18 cm forceps 3. Collect the AM tissue immediately after birth from the delivery suite sluice room. 4. Use the forceps and scissors to search along the edges of the isolated sac for areas where the AM and chorion have separated. Separate the AM using general force (by hand) to peel/pull the chorion away from the AM. Carefully cut the AM around the base of the cord to release the tissue from the placenta (see Note 6). 5. Use the forceps to place the isolated AM tissue in the TCF (containing sterile saline) ready for transportation. 6. Dispose of the placental body. 7. Discard the scissors and forceps in the sharps bin (within sluice room). 8. Transport the TCF to the laboratory within 2 h of birth.


Andrew Hopkinson et al.

3.3 Preparation for Manufacturing

1. Ensure appropriate Personal Protective Equipment (PPE) is worn before entering the laboratory (see Note 7). 2. Collect 1 antibiotics from the 80  C freezer before entering the laboratory. Thaw the vial at room temperature. 3. Keep the TCF outside the safety cabinet until Subheading 3.5, step 5. 4. Switch on equipment—rocker, incubator, and safety cabinet. 5. Clean the working area in the safety cabinet with IMS (see Note 8). 6. Set aside IMS, tissues, sharps bin, and marker pen for later use.

3.4 Preparation of Washing Solutions

1. Transfer the following materials to the safety cabinet: (a) 3  T175 flasks (b) 1  1 L NaCl 2. Label the T175 flasks “WASH 1,” “WASH 2,” and “WASH 3” using the marker pen. 3. Mark the 250 mL line on each flask and decant 250 mL NaCl into each flask.

3.5 Amniotic Membrane Washing and Incubation

1. Transfer the following materials to the safety cabinet: (a) 1  Square dish pack (containing 4  square tray pairs) (b) 1  18 cm forceps 2. Open the square dish pack and take one square tray base to use as an “instruments” tray. Place forceps on the instruments tray. 3. Take another tray base to use as a “working” tray. 4. Set aside the remaining square dishes for later use. 5. Transfer the TCF to the safety cabinet. 6. Use the forceps to remove the AM from the TCF and spread it over the working tray for inspection (see Note 9). 7. Use the forceps to place the AM in the WASH 1 flask (rest the cap upward on the cabinet surface). 8. Remove the WASH 1 flask from the safety cabinet and place on the rocker for 20 min, speed ¼ 50 RPM. 9. Transfer the following materials to the safety cabinet: (a) 1  0.5 L PBS (b) 1  RAFFINOSE flask 10. Mark the 250 mL line on the flask and decant 250 mL PBS into the flask. Retain the PBS bottle for later use. 11. Remove the RAFFINOSE flask from the safety cabinet and place on the rocker for at least 25 min. 12. When the rocking time for the WASH 1 flask has finished, transfer it back to the safety cabinet.

Dried Amniotic Membrane Preparation


13. Keep the RAFFINOSE flask on the rocker. 14. Use forceps to transfer the AM from the WASH 1 to the WASH 2 flask (rest the flask caps upward on the cabinet surface). 15. Remove the WASH 2 flask from the safety cabinet and place on the rocker for 20 min. 16. Discard the WASH 1 flask in the clinical waste bag. 17. Transfer the following materials to the safety cabinet: (a) 2  T175 flask (b) 2  50 mL syringe (c) 2  0.22 μm filters (d) 1  25 mL Serological pipette (e) 1  Pipette aid 18. Place all materials on the instruments tray. 19. Label the flasks with “INCUBATION” and “INCUBATION WASH” using the marker pen. 20. Mark the 250 mL line on INCUBATION WASH flask. 21. When the rocking time for the WASH 2 flask has finished, transfer it back to the safety cabinet. 22. Transfer the RAFFINOSE flask to the safety cabinet (make sure the raffinose is fully dissolved). 23. Use the forceps to transfer the AM from the WASH 2 to the WASH 3 flask (rest the caps upward on the cabinet surface). 24. Remove the WASH 3 flask from the safety cabinet and place on the rocker for 20 min. 25. Discard the WASH 2 flask in the clinical waste bag. 26. Use the 50 mL syringes and 0.22 μm filters to filter the raffinose solution from the RAFFINOSE flask into the INCUBATION flask. 27. Transfer the following materials to the safety cabinet: (a) 1  Antibiotics 28. Add the antibiotics to the INCUBATION flask containing PBS. Swirl the flask to mix. 29. Use a serological pipette to withdraw 25 mL solution containing antibiotics from the INCUBATION flask and add to the INCUBATION WASH flask. 30. Add PBS to the INCUBATION WASH flask up to 250 mL (retain the PBS bottle for later use). 31. Remove the INCUBATION flask from the safety cabinet and place in the incubator (set to 37  C). 32. Set aside the INCUBATION WASH flask for later use. 33. Discard the RAFFINOSE flask, filters, syringes, and pipettes in the clinical waste bag.


Andrew Hopkinson et al.

3.6 Spongy Layer Removal

1. Transfer the following materials to the safety cabinet: (a) 1  No. 22 blade (b) 2  Pack of spongy spears (c) 1  Bold surgical pen 2. When the rocking time for the WASH 3 flask has finished, transfer it back to the safety cabinet. 3. Take a new square tray base to use as working tray. 4. Use the forceps to remove the AM from the WASH 3 flask and place it on the working tray (rest the flask cap upward on the cabinet surface). 5. Use fresh, dry, and clean spongy spears to detect the SL. 6. Carefully dab a spongy spear onto the AM and lift it. The spear will stick to the SL but not the epithelial side. If it does not stick, it is most likely facing the epithelial side. Flip the AM and repeat test with a new spongy spear to ensure your acceptation is correct (see Note 10). 7. When the SL side has been determined, spread the AM with its epithelial side (reverse to SL side) up. Confirm you have the correct side with a new spongy spear. 8. Use the surgical pen to write: “UP” on the epithelial side. Take a picture of the whole AM, showing the marking. 9. Flip the AM to the SL side and spread out as much as possible. 10. Add sufficient washing solution from the WASH 3 flask to maintain AM and SL hydration. 11. Use the reverse side of the scalpel to firmly but carefully peel, but not scrape, back the SL off the AM as a continuous layer (from the direction of the front of the cabinet to the back). Take care not to cut the underlying AM stroma. Remove substantially all the SL from the AM (see Note 11). 12. Use new spongy spears to check substantially all the SL has been removed from across the AM (the spear should not detect any SL). 13. Transfer the following materials to the safety cabinet: (a) INCUBATION flask 14. Use the forceps to place the AM in the INCUBATION flask (rest the cap upward on the cabinet surface). 15. Transfer out the INCUBATION flask and place in the incubator (set to 37  C) for 2 h (wipe off any contaminating liquids before placing on the rocker with IMS). 16. Transfer any removed SL to the WASH 3 flask. 17. Discard the WASH 3 flask and working tray in the clinical waste bag. Discard sharps in the sharps bin.

Dried Amniotic Membrane Preparation


18. Switch on dryer (20 min before the incubation ends). 19. Transfer the following materials to the safety cabinet: (a) INCUBATION flask (b) 1  18 cm forceps 20. Place the forceps on the instruments tray. 21. Use the forceps to transfer the AM from the INCUBATION to the INCUBATION WASH flask. 22. Remove the INCUBATION WASH flask from the safety cabinet and place on the rocker for 15 min, speed ¼ 50 RPM. 23. Discard the INCUBATION flask in the clinical waste bag. 3.7 Amniotic Membrane Drying (Patent PCT/GB2014/ 051722)

1. Transfer the following materials to the safety cabinet: (a) 1  Scalpel (b) Parafilm 2. Take 2–3 (depending on AM size) square lids. Place one pre-cut Parafilm in each tray lid. 3. Transfer the following materials to the safety cabinet: (a) INCUBATION WASH flask 4. Prepare the dryer ready for drying (see Note 12). 5. Take a new square tray base to use as a working tray. 6. Use the forceps to transfer the AM from the INCUBATION WASH flask to the tray. Observe the AM inside the INCUBATION WASH flask, if it has coiled onto itself, uncoil it while in the liquid before taking it to the tray. 7. Spread the AM out, epithelial side upward (use the UP mark to identify). 8. If the AM is bigger than the tray, use a scalpel to cut it into smaller pieces, and place each AM piece on a tray, epithelial side up. 9. Transfer trays out of the safety cabinet. 10. Place AM trays in dryer. Run recipe (see Note 13). 11. Transfer the following materials to the safety cabinet: (a) TCF 12. Discard the remaining PBS to the INCUBATION flask. 13. Take the TCF flask and transfer the collection liquid to the PBS bottle for bioburden analysis. 14. Transfer out all remaining materials and close the sharps bin. 15. Clean inside the safety cabinet thoroughly with IMS. 16. Discard the TCF in the clinical waste bag. 17. Turn off equipment—rocker, incubator, and safety cabinet.


Andrew Hopkinson et al.

18. After drying cycle has finished, remove the dried AM (see Note 14). 19. Transfer the dried AM to the safety cabinet. Cut or store AM as necessary (see Note 15). 20. Turn off dryer.


Notes 1. For preparation of dried AM for clinical use, the laboratory must conform to strict regulatory bodies, e.g., the HTA in accordance with the Human Tissue Act 2004, the Human Tissue (Quality and Safety for Human Applications) Regulations 2007, the Quality and Safety of Organs Intended for Transplantation Regulations 2012, or other EU or international regulatory bodies if processing outside the UK. Hepatitis B immunization is required for individuals working with human tissue and blood. 2. For clinical use, the preparation of all materials and methods should be performed in a Grade A positive pressure environment following GMP. Settle plates and continuous particle counting (using a particle analyzer) need to be used during processing to ensure that the isolator conforms with EU GMP Annex 1 Microbial Contamination limits for Grade A environments. The environment needs to be monitored for relative humidity and temperature. 3. Raffinose is a natural preservative sugar protectant used to prepare the tissue for drying. As an alternative, the AM can be processed with and without raffinose or in the presence of other novel complex saccharide lyoprotectants such as trehalose. 4. Ensure appropriate training and permission has been granted from approved hospitals prior to consent and procurement. Research ethics or regulatory approved consenting policies (for clinical use) must be in place. Consent should be taken by trained personnel on the morning of the patient undergoing elective caesarean surgery. 5. For clinical use, donor blood samples should be taken during the surgical process for communicable disease testing including serological testing for HIV/HIV 2 antibodies (HIV antigen/ antibody), hepatitis B HBsAg, hepatitis B Anti HBc, HCV antibodies, syphilis (Treponema Pallidum), HTLV antibodies, and molecular nucleic acid amplification test testing for HIV RNA, HBV DNA, HCV RNA, and CMV DNA. Tests must be sent to an accredited pathology laboratory for the screening of the mandatory infectious diseases in accordance with the requirements of Annex II Directive 2006/17/EC. Blood

Dried Amniotic Membrane Preparation


tubes for testing (purple top EDTA and yellow top) should be placed in the collection box prior to use. 6. The fetal sac (placental membrane) is a thin tissue composed of two layers, the AM and the chorion. The chorion forms the outermost layer. It is thicker in appearance, opaque and vascular, with a rough texture. The AM forms the innermost layer and is smoother in appearance, transparent, and avascular. 7. PPE must be worn, including lab coats and gloves, at all times. Ensure gloves are category II PPE registered. For clinical use, ensure PPE aligns with appropriate regulatory bodies for processing human tissue. 8. Ensure that the working area is frequently cleaned during processing. For clinical use, GMP grade cleaning reagents must be used. 9. AM is a biological variable tissue, which can present with significant physical, color, and visual abnormalities, depending on the mother health, well-being, and lifestyle. The position in the womb can also influence the AM physical appearance. For clinical use, all raw tissue needs to be assessed by completion of an in-process raw tissue assessment form to assess whether the tissue is acceptable for processing. Examine the fetal sac for any discoloration, debris, or signs of damage. 10. The “sticky” physical property of the SL is exploited in the theatre by ophthalmology surgeons to identify the SL side of the cryopreserved AM, though not always visible on all clinical grade cryopreserved AM (Fig. 1). Tseng et al. in 2007 [52] advocates the use of Weckcel cellulose surgical spears to detect the SL, and orientate the stromal surface of AmnioGraft®. Similarly, the proprietary CryoTek® method is also employed to manufacture PROKERA®, meaning both these products contain variable amounts of SL with ultimately variable biochemistry. Having such close contact with the ocular surface, the SL may therefore be integral for the reported “properties” of the AM. Consequently, this may be crucial in explaining the variable clinical efficacy of different membranes observed in clinical practice [53]. The spongy spear test has therefore been developed as a nondestructive validation test to ensure SL removal (Fig. 1). Spongy spears are used to dab the surface of the processed AM. Due to the sticky nature of the SL, even single fibers adhere to the spongy spears as it lifts from the AM. The absence of any stickiness will indicate substantially all the SL has been removed. 11. Removal of the SL is only effective after prolonged washing steps (Fig. 2). The excessive swelling enables the easy removal of substantially all the SL, almost intact. The reverse edge of the scalpel blade is used to break the SL, which is then


Andrew Hopkinson et al.

Fig. 1 Amnion SL removal. Image example of (a) spear testing of amnion before SL removal, and (b) following SL removal still detecting small amount of SL at the edge

Fig. 2 Diagram illustrating complete removal of SL. (a) Standard blood contaminated AM, (b) blood removal by continued washing without mechanical intervention (allowing SL layer to swell), (c) removal of swollen intact SL, and (d) image of SL removed

delicately peeled off the AM. This process is performed across the whole membrane, separating the spongy layer in its entirety. 12. Drying parameters have been optimized to ensure effective drying using an advantage Pro Laboratory benchtop freeze dryer. The recipe controls the temperature and vacuum at which the AM is exposed for a specific time in order to delicately dehydrate the tissue. Settings will need to be optimized and validated for different dryers. 13. Turn off the shelf and condenser from pre-freeze. Shelf temperatures should be set between 3.5  C and 6.0  C and condenser should be set between 60  C and 90  C. These can be altered by turning the shelf and condenser buttons on and off.

Dried Amniotic Membrane Preparation


14. Verify the dryer receipt is complete. Defrost shelves, condenser, and release vacuum. Wait until the system reaches atmospheric pressure (approximately 1070 mBar) and shelves reach 30  C before taking the trays out of the dryer. 15. Use new consumables for each AM cutting procedure. Identify holes and tears in the tissue and do not include them in the product. Use cutters or scalpel to cut AM into desired shapes. Use forceps (grip lightly to ensure no damage is caused) to carefully lift cut pieces from the tray. If any edges still appear to be connected, use a scalpel blade to gently cut the connections. Package product as required. References 1. Meller D, Pauklin M, Thomasen H, Westekemper H, Steuhl K-P (2011) Amniotic membrane transplantation in the human eye. Dtsch Arztebl Int 108(14):243–248. https:// 2. Dua HS, Gomes JA, King AJ, Maharajan VS (2004) The amniotic membrane in ophthalmology. Surv Ophthalmol 49(1):51–77 3. Hao Y, Ma DH, Hwang DG, Kim WS, Zhang F (2000) Identification of antiangiogenic and antiinflammatory proteins in human amniotic membrane. Cornea 19(3):348–352 4. Ogawa Y, He H, Mukai S, Imada T, Nakamura S, Su CW, Mahabole M, Tseng SC, Tsubota K (2017) Heavy chain-hyaluronan/ pentraxin 3 from amniotic membrane suppresses inflammation and scarring in murine lacrimal gland and conjunctiva of chronic graft-versus-host disease. Sci Rep 7:42195. 5. Bennett JP, Matthews R, Faulk WP (1980) Treatment of chronic ulceration of the legs with human amnion. Lancet 1 (8179):1153–1156 6. Hopkinson A, McIntosh RS, Shanmuganathan V, Tighe PJ, Dua HS (2006) Proteomic analysis of amniotic membrane prepared for human transplantation: characterization of proteins and clinical implications. J Proteome Res 5(9):2226–2235. 7. Tseng SC, Li DQ, Ma X (1999) Suppression of transforming growth factor-beta isoforms, tgf-beta receptor type ii, and myofibroblast differentiation in cultured human corneal and limbal fibroblasts by amniotic membrane matrix. J Cell Physiol 179(3):325–335. 199906)179:33.0.CO;2-

X. [pii] 10.1002/(SICI)1097-4652(199906) 179:33.0.CO;2-X 8. Meller D, Pires RT, Mack RJ, Figueiredo F, Heiligenhaus A, Park WC, Prabhasawat P, John T, McLeod SD, Steuhl KP, Tseng SC (2000) Amniotic membrane transplantation for acute chemical or thermal burns. Ophthalmology 107(5):980–989. discussion 990. S0161-6420(00)00024-5 [pii] 9. Kamiya K, Wang M, Uchida S, Amano S, Oshika T, Sakuragawa N, Hori J (2005) Topical application of culture supernatant from human amniotic epithelial cells suppresses inflammatory reactions in cornea. Exp Eye Res 80(5):671–679. 1016/j.exer.2004.11.018 10. Niknejad H, Yazdanpanah G (2014) Anticancer effects of human amniotic membrane and its epithelial cells. Med Hypotheses 82 (4):488–489. mehy.2014.01.034 11. Niknejad H, Yazdanpanah G, Ahmadiani A (2016) Induction of apoptosis, stimulation of cell-cycle arrest and inhibition of angiogenesis make human amnion-derived cells promising sources for cell therapy of cancer. Cell Tissue Res 363(3):599–608. 1007/s00441-016-2364-3 12. Niknejad H, Yazdanpanah G, Mirmasoumi M, Abolghasemi H, Peirovi H, Ahmadiani A (2013) Inhibition of hsp90 could be possible mechanism for anti-cancer property of amniotic membrane. Med Hypotheses 81 (5):862–865. mehy.2013.08.018 13. Magatti M, Munari S, Vertua E, Parolini O (2012) Amniotic membrane-derived cells inhibit proliferation of cancer cell lines by inducing cell cycle arrest. J Cell Mol Med 16


Andrew Hopkinson et al.

(9):2208–2218. 1582-4934.2012.01531.x 14. Prabhasawat P, Tesavibul N, Prakairungthong N, Booranapong W (2007) Efficacy of amniotic membrane patching for acute chemical and thermal ocular burns. J Med Assoc Thail 90(2):319–326 15. Tejwani S, Kolari RS, Sangwan VS, Rao GN (2007) Role of amniotic membrane graft for ocular chemical and thermal injuries. Cornea 26(1):21–26. 0b013e31802b4201. 00003226-20070100000005 [pii] 16. Chawla B, Sharma N, Tandon R, Kalaivani M, Titiyal JS, Vajpayee RB (2010) Comparative evaluation of phototherapeutic keratectomy and amniotic membrane transplantation for management of symptomatic chronic bullous keratopathy. Cornea 29(9):976–979. https:// 17. Tamhane A, Vajpayee RB, Biswas NR, Pandey RM, Sharma N, Titiyal JS, Tandon R (2005) Evaluation of amniotic membrane transplantation as an adjunct to medical therapy as compared with medical therapy alone in acute ocular burns. Ophthalmology 112 (11):1963–1969. ophtha.2005.05.022 18. Sharma N, Singh D, Maharana PK, Kriplani A, Velpandian T, Pandey RM, Vajpayee RB (2016) Comparison of amniotic membrane transplantation and umbilical cord serum in acute ocular chemical burns: a randomized controlled trial. Am J Ophthalmol 168:157–163. ajo.2016.05.010 19. Kheirkhah A, Tabatabaei A, Zavareh MK, Khodabandeh A, Mohammadpour M, Raju VK (2012) A controlled study of amniotic membrane transplantation for acute pseudomonas keratitis. Can J Ophthalmol 47 (3):305–311. jcjo.2012.03.014 20. Jensen O, Gluud B (1985) Bacterial growth in the conjunctival sac and the local defense of the outer eye. Acta Ophthalmol 63(S173):80–82 21. Sotozono CMDPD, Inagaki KMD, Fujita AMD, Koizumi NMDPD, Sano YMDPD, Inatomi TMDPD, Kinoshita SMDPD (2002) Methicillin-resistant staphylococcus aureus and methicillin-resistant staphylococcus epidermidis infections in the cornea. Cornea 21 (2):S94–S101 22. Goodman DF, Gottsch JD (1988) Methicillinresistant staphylococcus epidermidis keratitis treated with vancomycin. Arch Ophthalmol 106(11):1570–1571. 1001/archopht.1988.01060140738046

23. Davis J (1910) Skin transplantation with a review of 550 cases at the johns hopkins hospital. Johns Hopkins Med J 15:307 24. Schrimpf WJ (1954) Repair of tympanic membrane perforations with human amniotic membrane; report of fifty-three cases. Ann Otol Rhinol Laryngol 63(1):101–115 25. Trelford JD, Trelford-Sauder M (1979) The amnion in surgery, past and present. Am J Obstet Gynecol 134(7):833–845. doi:00029378(79)90957-8 [pii] 26. Matthews RN (1981) Human tissue response to amnion allograft. Lancet 2(8260–61):1428 27. Ashworth MF, Morton KE, Dewhurst J, Lilford RJ, Bates RG (1986) Vaginoplasty using amnion. Obstet Gynecol 67(3):443–446 28. Cox LW (1987) Constructive and reconstructive gynaecology: uterine and vaginal grafts. Aust N Z J Surg 57(5):284–286 29. Brandt FT, Albuquerque CD, Lorenzato FR (2000) Female urethral reconstruction with amnion grafts. Int J Surg Investig 1 (5):409–414 30. Lai DR, Chen HR, Lin LM, Huang YL, Tsai CC (1995) Clinical evaluation of different treatment methods for oral submucous fibrosis. A 10-year experience with 150 cases. J Oral Pathol Med 24(9):402–406 31. Goto Y, Noguchi Y, Nomura A, Sakamoto T, Ishii Y, Bitoh S, Picton C, Fujita Y, Watanabe T, Hasegawa S, Uchida Y (1999) In vitro reconstitution of the tracheal epithelium. Am J Respir Cell Mol Biol 20(2):312–318 32. Haugh AM, Witt JG, Hauch A, Darden M, Parker G, Ellsworth WA, Buell JF (2017) Amnion membrane in diabetic foot wounds: a meta-analysis. Plast Reconstr Surg Glob Open 5(4):e1302–e1302. 1097/GOX.0000000000001302 33. Guo Q, Hao J, Yang Q, Guan L, Ouyang S, Wang J (2011) A comparison of the effectiveness between amniotic membrane homogenate and transplanted amniotic membrane in healing corneal damage in a rabbit model. Acta Ophthalmol 89(4):e315–e319. https://doi. org/10.1111/j.1755-3768.2010.02097.x 34. Dua HS (1999) Amniotic membrane transplantation. Br J Ophthalmol 83(6):748–752 35. Tabatabaei SA, Soleimani M, Behrouz MJ, Torkashvand A, Anvari P, Yaseri M (2017) A randomized clinical trial to evaluate the usefulness of amniotic membrane transplantation in bacterial keratitis healing. Ocul Surf 15 (2):218–226 36. Chen H-C, Tan H-Y, Hsiao C-H, Huang SC-M, Lin K-K, Ma DH-K (2006) Amniotic membrane transplantation for persistent

Dried Amniotic Membrane Preparation corneal ulcers and perforations in acute fungal keratitis. Cornea 25(5):564–572 37. Dua HS, Rahman I, Miri A (2010) Variations in amniotic membrane: relevance for clinical applications. Br J Ophthalmol 94(8):963–964 38. Khokhar SMD, Sharma NMD, Kumar HMD, Soni AMD (2001) Infection after use of nonpreserved human amniotic membrane for the reconstruction of the ocular surface. Cornea 20 (7):773–774 39. Cheng AM, Zhao D, Chen R, Yin HY, Tighe S, Sheha H, Casas V, Tseng SC (2016) Accelerated restoration of ocular surface health in dry eye disease by self-retained cryopreserved amniotic membrane. Ocul Surf 14(1):56–63. 40. Paolin A, Trojan D, Leonardi A, Mellone S, Volpe A, Orlandi A, Cogliati E (2016) Cytokine expression and ultrastructural alterations in fresh-frozen, freeze-dried and gammairradiated human amniotic membranes. Cell Tissue Bank 17(3):399–406. 10.1007/s10561-016-9553-x 41. Koob TJ, Rennert R, Zabek N, Massee M, Lim JJ, Temenoff JS, Li WW, Gurtner G (2013) Biological properties of dehydrated human amnion/chorion composite graft: implications for chronic wound healing. Int Wound J 10 (5):493–500. 12140 42. Hopkinson A, McIntosh RS, Tighe PJ, James DK, Dua HS (2006) Amniotic membrane for ocular surface reconstruction: donor variations and the effect of handling on tgf-beta content. Invest Ophthalmol Vis Sci 47(10):4316–4322. 43. Gicquel JJ, Dua HS, Brodie A, Mohammed I, Suleman H, Lazutina E, James DK, Hopkinson A (2009) Epidermal growth factor variations in amniotic membrane used for ex vivo tissue constructs. Tissue Eng Part A 15 (8):1919–1927. ten.tea.2008.0432 44. Hopkinson A, McIntosh RS, Layfield R, Keyte J, Dua HS, Tighe PJ (2005) Optimised two-dimensional electrophoresis procedures for the protein characterisation of structural tissues. Proteomics 5(7):1967–1979. https:// 45. Dua HS, Maharajan VS, Hopkinson A (2005) Controversies and limitations of amniotic


membrane in ophthalmic surgery. In: Essentials in ophthalmology cornea and external eye disease. Springer, New York, pp 29–41 46. Allen CL, Clare G, Stewart EA, Branch MJ, McIntosh OD, Dadhwal M, Dua HS, Hopkinson A (2013) Augmented dried versus cryopreserved amniotic membrane as an ocular surface dressing. PLoS One 8(10):e78441. https:// 47. Nakamura T, Sekiyama E, Takaoka M, Bentley AJ, Yokoi N, Fullwood NJ, Kinoshita S (2008) The use of trehalose-treated freeze-dried amniotic membrane for ocular surface reconstruction. Biomaterials 29(27):3729–3737. 2008.05.023. S0142-9612(08)00378-5 [pii] 48. Aghayan HR, Goodarzi P, Baradaran-Rafii A, Larijani B, Moradabadi L, Rahim F, Arjmand B (2013) Bacterial contamination of amniotic membrane in a tissue bank from iran. Cell Tissue Bank 14(3):401–406. 1007/s10561-012-9345-x 49. Marsit NM, Sidney LE, Branch MJ, Wilson SL, Hopkinson A (2017) Terminal sterilization: conventional methods versus emerging cold atmospheric pressure plasma technology for non-viable biological tissues. Plasma Process Polym 14(7):e1600134. 1002/ppap.201600134 50. Marsit N (2019) Characterisation of vacuum dried amniotic membrane and validation of the processing protocol. PhD, The University of Nottingham Recently passed 51. Marsit NM, Sidney LE, Britchford ER, McIntosh OD, Allen CL, Ashraf W, Bayston R, Hopkinson A (2019) Validation and assessment of an antibiotic decontamination manufacturing protocol for vacuum-dried human amniotic membrane. Sci Rep 9(1):12854. https://doi. org/10.1038/s41598-019-49314-7 52. Tseng SC, Elizondo A, Cases V (2007) Amniotic membrane suturing techniques. In: Macsai MS (ed) Ophthalmic microsurgical suturing techniques. Springer, Berlin 53. Maharajan VS, Shanmuganathan V, Currie A, Hopkinson A, Powell-Richards A, Dua HS (2007) Amniotic membrane transplantation for ocular surface reconstruction: indications and outcomes. Clin Exp Ophthalmol 35 (2):140–147. 1442-9071.2006.01408.x

Chapter 11 Fabrication of Corneal Extracellular Matrix-Derived Hydrogels Mark Ahearne and Julia Ferna´ndez-Pe´rez Abstract Hydrogels derived from corneal extracellular matrix (ECM) represent a promising biomaterial for corneal repair and regeneration. To fabricate these hydrogels, first corneas need to be decellularized using repeated freeze-thaw cycles and nucleases to remove all nuclear and cellular components. The remaining corneal ECM is lyophilized to remove all water and milled into a fine powder. The ECM powder is weighed and dissolved in pepsin solution at a concentration of 20 mg/mL. Hydrogels are formed by neutralizing the pH of the solution and maintaining it at 37  C until fibrillogenesis has occurred. Corneal stromal cells may be suspended throughout the hydrogel solution prior to gelation to generate a corneal stromal substitute. Key words Hydrogel, Scaffold, Cornea, Stroma, Collagen, Extracellular matrix, Keratocyte


Introduction The global shortage of donor corneal tissues suitable for transplantation [1] has led many researchers to investigate alternative treatment modalities. One solution is to use tissue engineering to manufacture corneal tissue in vitro by combining either cornealderived cells or stem cells with a suitable three-dimensional biomaterial scaffold. One of the main challenges in developing scaffolds that are suitable for corneal stromal regeneration or the replacement of damaged corneal tissue is how to accurately replicate the composition of the real tissue. While the stroma primarily consists of collagen type 1, the presence of other collagens, proteoglycans, and cytokines in the extracellular matrix (ECM) is vital to regulating the phenotype of corneal stromal keratocytes [2–5] and maintaining the corneas structure and transparency [6–8]. While decellularized corneal scaffolds are capable of retaining most of these ECM components of the cornea, the main difficulty associated with these types of scaffolds is how to successfully repopulate the matrix with new healthy cells after decellularization [9, 10].

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Mark Ahearne and Julia Ferna´ndez-Pe´rez

Fig. 1 Overview of steps involved in fabricating corneal ECM hydrogels: (a) removal of cornea from eye; (b) dissection of cornea; (c) suspended for freeze-thaw decellularization; (d) ECM after freeze-drying; (e) ECM after cryomilling; (f) suspended in pepsin; (g) final hydrogel solutions; (h) final hydrogel

The recent development of corneal ECM-derived hydrogels represents a novel approach for generating biocompatible scaffolds that can retain most of the native tissues ECM components [11, 12]. These hydrogels form when solubilized collagen from ECM undergoes polymerization under specific temperature and pH conditions to form a water swollen collagen network that still contains other ECM components [13]. Unlike decellularized corneal scaffolds, cells can be easily suspended throughout the hydrogel prior to gelation. This allows for the generation of a three-dimensional corneal stromal construct embedded with viable cells. In addition, corneal ECM hydrogel solutions have previously been used as a bioink for the 3D bioprinting of corneal constructs [14, 15]. This chapter will explore the process of fabricating corneal ECM-derived hydrogels. While there are several variations in how to fabricate these hydrogels, they all tend to follow the same basic steps (Fig. 1). First, the corneal tissue needs to be decellularized to remove any cells and cellular components that could potentially illicit a negative immune response if present (see Note 1). Next, the tissue needs to be dehydrated and broken down into fine particles to allow the ECM to be more easily digested. The ECM particles are then digested in an acidic solution. Finally, by neutralizing the pH of the ECM digest, the collagen undergoes fibrillogenesis resulting in a stable hydrogel being formed. Since gelation of the hydrogel takes 20–30 min, cells can be mixed into the neutralized ECM solution prior to gelation occurring. Further crosslinking may then be applied to improve the stability and mechanical properties of the hydrogels [16].

Corneal ECM Hydrogels



Materials Where possible, fresh porcine corneas should be obtained directly from a local slaughterhouse or abattoir (see Notes 2 and 3). Ultrapure water is water that has been purified by double deionizing and filtering to 18 MΩ-cm resistivity at room temperature.


Lab Equipment

1. Freeze dryer. 2. Cryomill—a “SPEX Freezer/Mill Cryogenic Grinder” was used in this study. 3. Cryomill polycarbonate tube, stainless steel end plugs, and stainless steel impactor. 4. Type II biological safety cabinet. 5. CO2, temperate, and humidity controlled incubator. 6. Oven. 7.

80  C freezer.

8. Tube rotator. 9. Electronic weighing scale. 10. 12-Well plates. 11. 50 mL syringes. 12. 10 mL syringes. 13. 0.2 μm sterile filters. 14. 50 mm diameter Petri dishes. 15. 5 mL sterile tubes. 16. 1.5 mL sterile centrifuge tube. 17. 8 mm diameter biopsy punches. 18. Gilson pipettes (P1000 and P200) and sterile pipette tips. 19. Scalpel. 20. Spatula. 21. Tape. 22. Sharp scissors. 23. Sterile fine tipped forceps. 2.2 Preparation of Solutions

1. Reconstituting buffer: 5 mM CaCl2 in ultrapure water. Sterile filter solution using a syringe and 0.2 μm sterile filter. 2. 10 mM MgCl2 buffer solution: 203.3 mg of magnesium chloride hexahydrate in 100 mL of ultrapure water. Adjust the pH to 7.5 using 5 M sodium hydroxide solution (NaOH) and sterile filter solution using a syringe and 0.2 μm sterile filter. 3. 400 U/mL deoxyribonuclease I (DNAse): Dissolve 10 mg of 400 U/mg DNAse from bovine pancreas in 10 mL of reconstituting buffer. Aliquot and store at 20  C for up to 1 year.


Mark Ahearne and Julia Ferna´ndez-Pe´rez

4. 900 U/mL Ribonuclease (RNAse): Dissolve 50 mg of 90 U/mg RNAse A from bovine pancreas in 5 mL sterileautoclaved ultrapure water. For different initial U/mg adjust the mass accordingly. Aliquot and store at 20  C for up to 6 months. 5. DNAse/RNAse solution: Add 75 μL of DNAse solution and 33 μL of RNAse solution to 3 mL MgCl2 buffer solution. 6. 1 mg/mL pepsin solution: Add 20 mg of pepsin to 20 mL 0.1 M HCl. 7. 2% iodine solution. 8. Phosphate-buffered saline (PBS). 2.3 Hydrogel Fabrication

1. 1 N NaOH. 2. 10 PBS. 3. Whatman Grade 1 filter paper. 4. PTFE inserts (see Note 4).

3 3.1

Methods Decellularization

While there are numerous methods that have been developed to decellularize cornea [17], the method outlined here has been shown to be the most suitable for removing cells while still allowing a hydrogel to be formed [18]. 1. Under sterile conditions in a biological safety cabinet, rinse enucleated porcine eyes with 2% iodine solution followed by washes with PBS. 2. Using a sterile scalpel and forceps, remove the cornea from the eye and place the cornea into a Petri dish. Remove any non-corneal tissue from the cornea such as iris or conjunctiva and chop the remaining cornea into several pieces (2–3 mm3 each). 3. Place cornea pieces into a 5 mL tube with 4 mL sterile deionized water. 4. Freeze tube at

80  C for at least 4 h.

5. Allow water to fully thaw to room temperature without using the water bath and then replace the water with fresh ultrapure water. 6. Repeat steps 4 and 5 so the corneas undergo 5 freeze-thaw cycles in total. 7. Place each cornea in 3 mL of DNAse/RNAse solution and place in an oven under gentle rotation at 37  C for 1 h.

Corneal ECM Hydrogels


8. Remove solution and wash each cornea using ultrapure water three times for 24 h under rotation to remove any residual DNAse or RNAse (see Note 5). 3.2


1. Place the corneas in a Petri dish and submerge with sterile ultrapure water. 2. Place the dish in a freeze dryer (see Note 6) and program to reduce temperature to 30  C at a rate of 1  C/min from room temperature and hold at this temperature for 1 h. 3. Increase temperature to 10  C at a rate of 1  C/min. When the temperature reaches 10  C, switch on the vacuum to generate a pressure of 200 mbar inside the freeze dryer chamber and leave for 18 h. 4. Switch off vacuum and remove dish from freeze dryer. 5. Check that the tissue has been completely dried (see Note 7).



1. Sterilize cryomills polycarbonate tube, stainless steel end plugs, and stainless steel impactor before use (see Note 8). 2. Place lyophilized corneas in a polycarbonate tube with the impactor, seal both ends with the end plugs, and place tube into the cryomill (see Note 9). 3. Carefully fill tank with liquid nitrogen (see Note 10). 4. Close the cryomill chamber and run three times a 1-min cycle at a rate of 5 CPS with a 1-min break between each cycle (see Note 11). 5. Open and remove contents into a pre-weighed 5 mL tube in a flow hood. Calculate the weight of corneal ECM powder by re-weighing the tube (see Note 12).

3.4 Hydrogel Formation

1. Add 1 mg/mL pepsin solution to ECM powder to give a final concentration of 20 mg/mL ECM in solution. 2. Place in a rotator and gently rotate for at least 24 h at room temperature. Check that ECM powder has fully dissolved before continuing. 3. To make filter paper rings first use a 12 mm diameter circular disk to draw several circles on a sheet of filter paper. Cut around the circles with a sharp scissors to make 12 mm diameter filter paper disks. Next, use an 8 mm biopsy punch to cut holes in the center of the filter paper disks. Filter paper rings can be autoclaved to sterilize. 4. Place a PTFE disk into each well of a 12-well plate and place a filter paper ring on each disk. 5. To make 15 mg/mL hydrogels, cool a 1.5 mL tube on ice and add 42.7 μL of sterile water, 50 μL of 10 PBS, and 32.3 μL of


Mark Ahearne and Julia Ferna´ndez-Pe´rez

1 N NaOH and mix by pipetting up and down. Add 375 μL to tube to give a total volume of 500 μL and gentle mix by rotating the tube (see Notes 13–15). 6. Pipette 150 μL of the hydrogel solution into each well so to cover the filter paper ring and hole (see Notes 16 and 17). 7. Place in an incubator at 37  C for 1 h and then submerge in pre-warmed culture media. 8. Separate the hydrogel and filter paper from the PTFE disk and remove the disk from the well (see Note 18).


Notes 1. It is important that all cells and most of the cellular DNA be removed from the ECM to reduce the potential for negative immune responses in vivo. To confirm the success of decellularization it is believed that the ECM should contain less than 50 ng of double-stranded DNA per mg of ECM dry weight and any DNA fragments be less than 200 base pairs [19]. Cells or cell nuclei should also not be visible after histological sectioning and staining. 2. The quality of the eyes that are being used should be inspected before proceeding. Corneas that show signs of swelling, cloudiness, or physical damage should not be used. Ideally, the eyes should be removed shortly after the animal has been sacrificed. Some slaughterhouses or abattoirs will burn or chemically treat the pigs after death to remove hair, and if the eyes are still in place when this is happening, this will compromise the quality of the organs. 3. This protocol has been designed for use with porcine eyes but it should be possible to apply this protocol to eyes from different species. Some modifications to the decellularization process may have to be made and there is no guarantee that the outcome will be the same. 4. PTFE is less adhesive to proteins than cell culture plastic and allows the hydrogels to detach more easily. PTFE disks can be easily manufactured by purchasing a 0.5 mm to 1 mm thick sheet of PTFE and cutting it into circular disks using sharp scalpel. Thinner sheets may also be used but these tend to bend and can lead to difficulties when casting the hydrogel. Thicker sheets are more difficult to cut. PTFE can be sterilized by autoclaving. 5. Here we used a freeze-thaw process combined with nucleases to decellularize the tissue. We have tested several alternative decellularization procedures and found that this approach offered good retention of ECM components while still

Corneal ECM Hydrogels


allowing the removal cell and cellular components when compared to some alternative decellularization treatments. Other decellularization techniques could potentially work but would need to be tested. 6. When using a Petri dish in the freeze dryer, it is important to tape the lid onto the dish; otherwise it is likely to be sucked off once a vacuum is applied. Other containers, such as tubes, may also be used as alternative to a Petri dish; however it is important that there is a hole or space to allow water vapor to escape from the container. Make sure that the water is completely frozen before applying the vacuum. 7. After freeze-drying is complete, there should be no water or ice left in the Petri dish. If there is still water present, a longer vacuum cycle may be required and less water used in future. Alternatively, if the lid on the Petri dish or container housing the corneas is closed too tightly, this will prevent the water from escaping. 8. The cryomill polycarbonate tube, stainless steel end plugs, and impactor need to be sterilized before use. Particular care needs to be taken with the tube as it is likely to warp if autoclaved and undergo cracking if subjected to ethanol for prolonged periods of time. UV exposure in a cell culture flow hood for 30 min allows the tube to become sterile. Ethylene oxide may also be used to sterile the tube without damaging it. 9. We place between 10 and 20 freeze-dried corneas in the tube for cryomilling at the same time. Fewer corneas than this can lead to a low volume of ECM powder being obtained since the powder sticks to the side of the tube and grinder due to static energy. Higher numbers of corneas than this and the tube can become too full, preventing the impactor from successfully grinding the tissue into a powder. 10. It is recommended that several precautions are taken when using liquid nitrogen for cryomilling. Personnel protective equipment, especially a facemask and nitrogen handling gloves, should be worn as the nitrogen will probably spit and bubble when it first encounters the room temperate cryomill chamber. Adequate space and ventilation is also necessary to prevent the accumulation of nitrogen gas in the room. Ideally, an oxygen sensor should be present to detect if the nitrogen has displaced too much oxygen within the room. The nitrogen chamber in the cryomill should only be closed when the nitrogen has stopped bubbling. Care should also be taken when opening the chamber and removing the tube after the cryomilling cycles have been complete as liquid nitrogen is still present. The remaining nitrogen should be allowed to evaporate rather than trying to pour into a separate container.


Mark Ahearne and Julia Ferna´ndez-Pe´rez

11. The cryomill used in this description is very loud, so it is recommended that the user wear ear protection and warns any users nearby when the machine is going to start. 12. Some of the ECM will stick to the sides and ends of the tube and the grinder after cryomilling making it more difficult to extract. A sterile spatula is the best way to collect this ECM powder and moving it to a tube; however the user can expect some powder to be lost. The powder can be centrifuged once transferred to the tube. 13. Cells may be mixed into the hydrogel prior to gelation. This is best done by reducing the amount of dH2O and suspending the cells in culture media of an equivalent volume to that reduced. The cell suspension should be added after the other reagents have been mixed together; otherwise the pH of the hydrogel solution may not be evenly neutralized and result in cell death where the pH is too high or low. The cell suspension can be mixed by gentle inverting or stirring using a pipette tip. Pipetting up and down into a tip is not recommended as this results in the formation of bubbles. However for hydrogels with high concentrations (over 15 mg/mL) pipetting may be required to mix. 14. 10x DMEM (Dulbecco’s modified eagles medium) can be used as an alternative to 10 PBS when forming the hydrogels. The advantage of 10x DMEM is that it allows the user to visually determine if the solution has been evenly mixed due to its color. The color can also be used to indicate if the pH is approximately between 7 and 7.5. The drawback with using 10 DMEM is that the hydrogel is now red in color rather than being fully transparent. However this color can be removed after gelation by washing with PBS. 15. Try to avoid generating any bubbles while pipetting the hydrogel solution into the filter paper rings. If bubbles are present, they can be removed by either bursting with a needle tip or by pushing to the edge of the hydrogel using the pipette tip. 16. The shape and volume of the hydrogel can be changed depending on the user requirements. The use of filter paper rings when casting the hydrogels does have a number of advantages including it makes it easier for the hydrogels to be picked up and it limits contraction of the hydrogel by cells in the horizontal plane [20]. The size of the rings can also be varied if necessary. If using larger rings (e.g., 20 mm diameter), it is better to use two rings placed on top of each other rather than one. 17. Whatman grade 1 filter paper is effective at allowing hydrogel adhesion without absorbing too much liquid. We have test other grades of filter paper and found this to be the most practical to use.

Corneal ECM Hydrogels


18. The hydrogel can be detached from the PTFE disks by gently pushing it from the edge while submerged in solution, e.g., culture media. If the hydrogel is still sticking to the disk, lift the disk and use a scalpel to separate.

Acknowledgements The research is supported by funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (grant agreement no. 637460) and Science Foundation Ireland and Marie-Curie Action COFUND (grant no. 11/SIRG/B2104). References 1. Gain P, Jullienne R, He Z, Aldossary M, Acquart S, Cognasse F, Thuret G (2016) Global survey of corneal transplantation and eye banking. JAMA Ophthalmol 134 (2):167–173. jamaophthalmol.2015.4776 2. Fernandez-Perez J, Ahearne M (2019) Influence of biochemical cues in human corneal stromal cell phenotype. Curr Eye Res 44 (2):135–146. 02713683.2018.1536216 3. Petroll WM, Miron-Mendoza M (2015) Mechanical interactions and crosstalk between corneal keratocytes and the extracellular matrix. Exp Eye Res 133:49–57. https://doi. org/10.1016/j.exer.2014.09.003 4. Jester JV, Ho-Chang J (2003) Modulation of cultured corneal keratocyte phenotype by growth factors/cytokines control in vitro contractility and extracellular matrix contraction. Exp Eye Res 77(5):581–592 5. Lynch AP, O’Sullivan F, Ahearne M (2016) The effect of growth factor supplementation on corneal stromal cell phenotype in vitro using a serum-free media. Exp Eye Res 151:26–37. 2016.07.015 6. Hassell JR, Birk DE (2010) The molecular basis of corneal transparency. Exp Eye Res 91 (3):326–335. exer.2010.06.021 7. Massoudi D, Malecaze F, Galiacy SD (2016) Collagens and proteoglycans of the cornea: importance in transparency and visual disorders. Cell Tissue Res 363(2):337–349. https://doi. org/10.1007/s00441-015-2233-5 8. Meek KM (2009) Corneal collagen-its role in maintaining corneal shape and transparency.

Biophys Rev 1(2):83–93. 1007/s12551-009-0011-x 9. Lynch AP, Wilson SL, Ahearne M (2016) Dextran preserves native corneal structure during decellularization. Tissue Eng Part C Methods 22(6):561–572. ten.TEC.2016.0017 10. Wilson SL, Sidney LE, Dunphy SE, Dua HS, Hopkinson A (2016) Corneal decellularization: a method of recycling unsuitable donor tissue for clinical translation? Curr Eye Res 41 (6):769–782. 02713683.2015.1062114 11. Ahearne M, Lynch AP (2015) Early observation of extracellular matrix-derived hydrogels for corneal stroma regeneration. Tissue Eng Part C Methods 21(10):1059–1069. https:// 12. Lu Y, Yao QK, Feng B, Yan CX, Zhu MY, Chen JZ, Fu W, Fu Y (2015) Characterization of a hydrogel derived from decellularized corneal extracellular matrix. J Biomater Tiss Eng 5 (12):951–960. 2015.1410 13. Saldin LT, Cramer MC, Velankar SS, White LJ, Badylak SF (2017) Extracellular matrix hydrogels from decellularized tissues: structure and function. Acta Biomater 49:1–15. https://doi. org/10.1016/j.actbio.2016.11.068 14. Kim H, Park MN, Kim J, Jang J, Kim HK, Cho DW (2019) Characterization of cornea-specific bioink: high transparency, improved in vivo safety. J Tissue Eng 10:2041731418823382. 2041731418823382 15. Kim H, Jang J, Park J, Lee KP, Lee S, Lee DM, Kim KH, Kim HK, Cho DW (2019) Shearinduced alignment of collagen fibrils using 3d


Mark Ahearne and Julia Ferna´ndez-Pe´rez

cell printing for corneal stroma tissue engineering. Biofabrication 11(3). 1088/1758-5090/ab1a8b 16. Ahearne M, Coyle A (2016) Application of uva-riboflavin crosslinking to enhance the mechanical properties of extracellular matrix derived hydrogels. J Mech Behav Biomed Mater 54:259–267. 1016/j.jmbbm.2015.09.035 17. Fernandez-Perez J, Ahearne M (2019) Decellularization and recellularization of cornea: progress towards a donor alternative. Methods 171:86–96. ymeth.2019.05.009 18. Fernandez-Perez J, Ahearne M (2019) The impact of decellularization methods on

extracellular matrix derived hydrogels. Sci Rep 9(1):14933. s41598-019-49575-2 19. Crapo PM, Gilbert TW, Badylak SF (2011) An overview of tissue and whole organ decellularization processes. Biomaterials 32 (12):3233–3243. biomaterials.2011.01.057 20. Ahearne M, Liu KK, El Haj AJ, Then KY, Rauz S, Yang Y (2010) Online monitoring of the mechanical behavior of collagen hydrogels: influence of corneal fibroblasts on elastic modulus. Tissue Eng Part C Methods 16 (2):319–327. TEC.2008.0650

Chapter 12 Synthesis and Application of Collagens for Assembling a Corneal Implant Elle Edin, Fiona Simpson, and May Griffith Abstract Recombinant or artificial designer collagens have developed to a point where they are viable candidates for replacing extracted animal collagens in regenerative medicine applications. Biomimetic corneas made have shown promise as replacements for human donor corneas, and have previously been fabricated from several different collagens or collagen-like peptides (CLPs). Prokaryotic expression systems allow for cheap, rapid, gram scale production of collagens/CLPs. Here, we describe a procedure for production of collagen-like peptides for the manufacture of a biomimetic cornea. Key words Biomimetic, Artificial, Cornea, Artificial collagen, Collagen, Corneal regeneration, Transgenic



1.1 Collagen and Collagen-Like Peptides

Collagen is the most abundant protein present in the extracellular matrix that surrounds the cells of various tissues and organs in the mammalian body, including the cornea [1]. The defining feature of collagen is its unique supercoiled triple-helix structure [2, 3]. Fibrillar collagens, in particular, are robust structural macromolecules that contain cell-interactive domains. Hence, they have excellent properties for creating regenerative, cell-free scaffolds for corneal repair as seen in early clinical evaluation (Fig. 1) [4]. Most commercially available collagen is extracted from animal sources and purified using different methods, resulting in heterogeneity of size and helicity [5]. Recombinantly produced human collagens and short collagen mimetic peptides (CMPs) or collagen-like peptides (CLPs) developed as alternatives to animal collagens have the benefit of low heterogeneity. Also, unlike xenogeneic collagens [6], there is little/no risk of allergy to xenogeneic protein or zoonotic disease transfer. Collagen was initially considered a protein that is unique to multicellular animals, as hydroxyproline residues within collagen

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Elle Edin et al.

Fig. 1 Before and after photos of patients who had been grafted with recombinant human collagen-based implants to treat ulcers and scarring due to infection or burns. These patients showed stable integration of the implants and regenerated neo-corneas after an average of 2 years post-surgery. Modified from Fig. 2, Islam et al. [4]

have been considered the main determinants for structural stability [7]. However, CLPs have since been identified in prokaryotes, such as bacteria [8, 9]. These proteins have been isolated from biofilm, and they have been shown to also have triple helical structures and similar thermal stability to mammalian collagens [10]. As such, researchers have been able to design new CLPs that are based on bacterial collagen sequences and analyzed the structure-mechanical property relationships between these fibrils [8, 11, 12]. Collagens are chains of G-X-Y amino acid motifs, where the G amino acid (Gaa) is glycine [12]. The amino acid at the X (Xaa) position is frequently proline, and in animal collagens, the Y amino acid (Yaa) is often hydroxyproline. Important features that are known to affect collagen assembly, stability, as well as the related melting temperature include the Grand Average of Hydropathicity (GRAVY) score, hydroxyproline spacing, and the frequency of the six amino acid sequence Xaa1Yaa1Gaa1Xaa2Yaa2Gaa2 where the Yaa1 position hosts a lysine and the Xaa2 is occupied by a negatively charged residue (glutamic, or aspartic acid) [12–15]. In longer collagen peptides (>50 amino acids), assembly regions are often necessary for collagen fibril formation [16, 17]. In small CLPs that have high inter-strand interactions, assembly regions are not needed. Therefore, when selecting or designing a CLP, the experimenter needs to consider the availability of functional groups that can be used to stabilize the collagen helix, as well as stabilizing inter-fibrillar interactions. 1.2

CLP Production

Solid state synthesis is the method of choice for shorter CLPs ( 0.5. 3. 40 mL of bacterial culture is added to 3 L of media in baffled culture flask. 4. Culture is maintained at 37  C at 250 RPM until an OD of 1 is reached (see Note 9). 5. IPTG is added to culture to a final concentration of 1 mM. 6. Culture is brought to 5–8  C by placing in ice bath on a stirplate at 250 RPM. Monitor temperature with an analog thermometer. 7. Culture is placed at 5–16  C at 250 RPM for 24 h. 8. Culture is divided into 1 L centrifugation flasks and pelleted at 4  C, 4000 rcf. 9. Supernatant is discarded. 10. Pellet is dissolved in lysis buffer: 5 the pellet volume of buffer is used.

Synthesis and Application of Collagens for Assembling a Corneal Implant


11. Bacteria is lysed with sonication: 60% amplitude; 1 s on, 200 ms off; 4 2 min with 10 min of cooling on ice between each cycle. 12. Solution is centrifuged at 15 kRCF at 4  C for 30 min. 13. Supernatant is decanted and combined in a sterile glass flask. 14. Pellets are resuspended in 5 mL of lysis buffer with fresh PMSF. 15. Suspension is sonicated using the same parameters as above. 16. Perform SDS-PAGE of the supernatant from step 13 and solution from step 14 according to external protocol [33]. 17. If either of solutions from step 13 or 14 is devoid of target protein, that fraction is discarded. 18. Pool fractions that contain significant amounts of exColA. 3.4 FPLC Purification of CLP

1. The HisPrep™ FF 16/10 column containing Ni-loaded sepharose is attached to FPLC system (see Note 10). 2. System cleaning is performed according to FPLC system manufacturer handbook. 3. Column is equilibrated with 5 column volumes (CVs) of running buffer (see Note 11). 4. Sample from step 3.3.18 is loaded by direct injection at 3 mL/ min flow (see Note 12). 5. Wash the sample flow path with 1 CV of washing buffer. 6. Column with bound sample is washed with 10 CVs of washing buffer (see Note 13). 7. A linear gradient from washing buffer to elusion buffer is performed over 10 CVs. Fraction collection is performed over the whole span of the gradient, and one additional CV. Fraction volume is set to 20 mL (see Note 14). 8. FPLC system and column is washed according to manufacturer instructions. 9. Chromatogram is used to identify which fractions contain eluted protein. The peak is expected in the range of 100–400 mM of imidazole. Peak base width is expected to be no wider than 30 mL (see Note 15). 10. Dialysis tubing should be cut to the correct length for the sample volume. The length can be calculated using the formula:

Tubing Length ¼ ðð4 þ ððð376:992  Sample VolumeÞ=Tubing Flat WidthÞ= Tubing Flat WidthÞ  100Þ þ 0:5Þ=100 It is recommended to add 5–10 cm of tubing to this measurement to have space to attach the dialysis tubing clamps.


Elle Edin et al.

11. The dialysis tubing should be pre-equilibrated in the buffer to remove the residual preservatives from the tubing before use. 12. Clamp the bottom end of the dialysis tube by folding over the end of the tubing and attaching the weighted clamp over the folded edge. 13. Carefully pipette the solution into the tubing. 14. Open the tubing fully at the top end to allow air into the tubing above the sample. Fold the top end of the tubing over and clamp at the top ensuring that the air bubble remains between the clamp and the sample. The air is critical for the buoyancy of the sample during dialysis. 15. Fill the buffer container 80% with buffer and an appropriately sized stir bar. Add the sample and fill to 100% of the volume to prevent spillover. 16. Dialysis is performed by stepwise dialysis against urea solution. The sample is dialyzed against a total of 50 dialysis volumes of urea solution for a total of 12H per buffer step. 4 M, 3 M, 2 M, and 1 M is used and finally the solution is dialyzed against 100 dialysis volumes of TEV reaction buffer (see Notes 16 and 17). 17. The protein content is quantified using a Bradford assay according to the manufacturer instructions. 18. The sample is digested with TEV protease. 0.25 mg of TEV protease is used per mg of protein. 19. FPLC column is equilibrated with 5 CV of running buffer. 20. The sample is loaded on the column. Due to the lack of His-tag after digestion, only His-rich bacterial proteins should bind to the column while the protein of interest should run straight through. The flow-through is run to outlet and collected in a sterile glass bottle. Do not run to waste (see Note 18). 21. Wash buffer is run for 1 CV and collected in the same vessel as sample load flow through. 22. Bound protein is eluted using 3 CV of elution buffer. Flowthrough goes to waste (see Note 19). 3.5 Preparation of Lyophilized CLP

The protein flow-through should be dialyzed again from FPLC buffer until it is in ultrapure water. 1. Dialyzed exColA is transferred into 50 mL liquid nitrogen-safe tubes. 2. Sample tubes are frozen in liquid nitrogen for 10 min (see Note 20). 3. Sample tubes are opened slightly to allow air flow and placed in lyophilizer flasks. 4. The lyophilizer system is closed, and cycle is started.

Synthesis and Application of Collagens for Assembling a Corneal Implant

3.6 Preparation of Corneal-Shaped Implants


1. Remove the plunger from a sterile 10 mL syringe and wipe the interior of the barrel with a particle-free wipe to remove the syringe’s coating as it can interfere with hydrogel formation. Cap the syringe using a rubber cap held in place with parafilm. 2. Weigh the empty syringe and record the weight. Carefully transfer the lyophilized exColA into the syringe and weigh the assembly to determine the exColA mass. 3. Add ddH2O for a final concentration of 20% w/w. Cap the top of the syringe with a rubber stopper and parafilm. Centrifuge for 1 min at 200 rcf to ensure protein and water are in contact with one another. Dissolving can be expedited by cycles of heating to 37  C for 30 min followed by cooling on ice. Store at 4  C (see Note 21). 4. Centrifuge dissolved exColA at 1000 RCF at 4  C for 1 h; repeat until solution is free of visible bubbles. 5. Transfer 0.7 g of exColA solution from plastic syringe to a glass syringe using a 2 mm inner diameter PTFE tube to connect the two syringes. Ensure that no bubbles are produced during the transfer. 6. Prepare water bath in large glass beaker using ultrapure water (dd-water). 7. Fill syringe mixing system with dd-water and violently expel any trapped air bubbles into a water bath. Eject all water from the attached syringe and keep the mixing system submerged. Attach the glass syringe containing collagen to the empty Luer adapter on the mixing system, take care not to introduce bubbles. Place assembled mixing system on ice (see Note 22). 8. Dissolve DMTMM to 20% w/w in H2O. Sterile filter through a 0.2 μm syringe filter. 9. Inject dissolved DMTMM through the septum of the mixing system; use a volume equivalent to 0.7 times the molar amount of primary amines in exColA (see Note 23). 10. Mix the solution by alternating pressing the two plungers of the mixing system; pass the solution through the central t-piece 40 times to ensure sufficient homogeneity. 11. Eject 150 μL of exColA/DMTMM solution to each cornea mold. Assemble molds and jigs, expelling any surplus CLP/DMTMM solution around the edges of the molds. Place in a hydrated chamber overnight at room temperature. 12. Open the jigs and place the mold assemblies in PBS overnight at 4  C. 13. Carefully pry the jigs open and incubate the open molds overnight in PBS at 4  C. 14. Gently lift the corneal implants out of the molds once fully hydrated. Wash in PBS at 4  C for 7 days, changing the buffer daily.



Elle Edin et al.

Notes 1. When an appropriate CLP has been designed or selected, codon optimization for the desired expression system must be performed. This service can be performed by commercial entities such as GenScript. We suggest incorporation of an N-terminal 6x His tag and TEV endoprotease cutting site. The design should take into account the nucleases that will be used for future enzymatic cloning, ensuring that these are avoided within the coding sequence. 2. Cleavage sites other than the TEV site can be used. TEV endoprotease was chosen due to high specificity and due to the rarity of the motif. 3. Endonuclease deficient bacterial strains are required for longterm stability of cloning strains. This protocol uses a cold shock competent bacterial strain. If different competency bacteria is used, the reader should follow the protocols supplied together with that cloning strain for transformation. We use ClearColi®, an E. coli strain that was modified to have diminished or non-existent activation of LPS response in mammalian cells. A distinction should be made between endotoxin pathway activation and lack of immunogenicity. LPS is not the only bacterial constituent that can trigger inflammation or rejection [34]. 4. Microwaving will sterilize the solution sufficiently for no spontaneous growth to occur for several weeks on properly stored agar plates, even in the absence of antibiotics. 5. Dialysis tubing made from nitrocellulose can generally be used. Mw cutoff needs to be based on target protein Mw. Dialysis tubing should be purchased with a pore size that is at least 5 kDa smaller than a single subunit of your protein of interest. The dialysis tubing width should be chosen so that a standard batch results in tubing that is the correct length for your dialysis chamber. If necessary, use two shorter lengths of dialysis tubing for the sample to fit the beaker, so it floats free and unencumbered within the chamber. 6. The dialysis buffer container should be sufficiently large to hold a minimum of 50 the sample volume. A large graduated cylinder may be ideal as it allows for longer lengths of dialysis tubing for large samples than a large beaker. 7. Optional: at this point sequencing can be used in place of restriction digestion to ensure that no mutations have been introduced into the sequence. 8. We recommend use of SYBR Safe in place of toxic and mutagenic ethidium bromide.

Synthesis and Application of Collagens for Assembling a Corneal Implant


9. Oxygenation is of critical importance during the initial growth of the bacterial colony. The culture conditions outlined here assumes a large aerated space in the shaking incubator, or a ventilated/actively oxygenated incubator. 10. Binding capacity is heavily dependent upon the geometry and surface chemistry of the electrophoresis matrix. If a different matrix is used, the reader must reference manufacturer instructions and adjust bed volume to facilitate a sufficient protein binding capacity. 11. Depending on the predisposition of the protein used, a higher ionic strength might be needed for wash/bind buffer and elution buffer. If the protein is noted to form insoluble particles when exposed to the wash/bind buffer, 5 PBS can be used instead of 1. 12. Ensure that the column manufacturer-specified max delta pressure is not exceeded in this step. If column becomes visibly compressed or the delta pressure exceeds manufacturer recommendations, a lower pump speed should be used. 13. If this is a routine run, the flow-through in this step can be run to waste. If this is an early optimization run, the sample should be collected by running it to an outlet valve with a clean collection vessel. 14. Average peak base width using HisPrep™ FF 16/10 columns is 1 CV, which is why 20 mL fractions are used. 15. If yields are poor the imidazole absorption at 280 nm can make resolving the exact range of target elusion difficult; in these cases, a “dry run” with all the same parameters but without protein in the loading buffer can supply a baseline that can be deducted from the absorbance chromatogram. 16. The first dialysis stage should be no more than 3 h before the buffer is changed. 17. The stepwise dialysis is necessary to avoid protein falling out of solution. 18. This flow-through contains your target protein. Only bacterial His rich proteins will bind the column. 19. This step is a cleaning step, removing His rich bacterial proteins, TEV protease, and digested His-tags from the column. 20. Samples should be frozen and lyophilized according to the instructions provided by the lyophilizer manufacturer. These instructions are based on protocols for most research lyophilizers. 21. If the protein cannot be kept soluble in water at room temperature, other buffers can be used. Note that many collagens do not lyophilize well in phosphate-based buffers; PB and PBS should be avoided in this step.


Elle Edin et al.

22. Depending on the size and inter-strand interaction strength, the viscosity of different CLPs at any given concentration will vary. A mechanical syringe mixer can facilitate mixing of solutions that are not possible to safely mix by hand. 23. DMTMM has a MW of 276.72 Da. The volume of exColA is 0.7 mL. The concentration of exColA is 0.2 g/mL. The concentration of DMTMM is 0.2 g/mL. To calculate the volume to inject use the formula: V DMTMM ¼ ð0:7  ð0:2=MW exColA Þ  ðNumber of primary amines per exColAÞ  MW DMTMM Þ=0:2

References 1. Montes GS, Junqueira LC (1982) Biology of collagen. Rev Can Biol Exp 41(2):143–156 2. Brodsky B, Persikov AV (2005) Molecular structure of the collagen triple helix. In: Advances in protein chemistry, vol 70. Academic Press, Cambridge, pp 301–339 3. Kadler KE, Holmes DF, Trotter JA, Chapman JA (1996) Collagen fibril formation. Biochem J 316(1):1. bj3160001 4. Islam MM, Buznyk O, Reddy JC, Pasyechnikova N, Alarcon EI, Hayes S, Lewis P, Fagerholm P, He C, Iakymenko S, Liu W, Meek KM, Sangwan VS, Griffith M (2018) Biomaterials-enabled cornea regeneration in patients at high risk for rejection of donor tissue transplantation. NPJ Regen Med 3:2. 5. Zeugolis DI, Paul RG, Attenburrow G (2008) Factors influencing the properties of reconstituted collagen fibers prior to self-assembly: animal species and collagen extraction method. J Biomed Mater Res A 86A(4):892–904. 6. Lynn AK, Yannas IV, Bonfield W (2004) Antigenicity and immunogenicity of collagen. J Biomed Mater Res B Appl Biomater 71B (2):343–354. b.30096 7. Jenkins CL, Bretscher LE, Guzei IA, Raines RT (2003) Effect of 3-hydroxyproline residues on collagen stability. J Am Chem Soc 125 (21):6422–6427. ja034015j 8. Yu Z, An B, Ramshaw JAM, Brodsky B (2014) Bacterial collagen-like proteins that form triple-helical structures. J Struct Biol 186 (3):451–461. 2014.01.003

9. Peng YY, Yoshizumi A, Danon SJ, Glattauer V, Prokopenko O, Mirochnitchenko O, Yu Z, Inouye M, Werkmeister JA, Brodsky B, Ramshaw JA (2010) A streptococcus pyogenes derived collagen-like protein as a non-cytotoxic and non-immunogenic crosslinkable biomaterial. Biomaterials 31 (10):2755–2761. biomaterials.2009.12.040 10. An B, Kaplan DL, Brodsky B (2014) Engineered recombinant bacterial collagen as an alternative collagen-based biomaterial for tissue engineering. Front Chem 2(40). https:// 11. Cosgriff-Hernandez E, Hahn MS, Russell B, Wilems T, Munoz-Pinto D, Browning MB, Rivera J, Hook M (2010) Bioactive hydrogels based on designer collagens. Acta Biomater 6 (10):3969–3977. actbio.2010.05.002 12. Xu C, Yu Z, Inouye M, Brodsky B, Mirochnitchenko O (2010) Expanding the family of collagen proteins: recombinant bacterial collagens of varying composition form triplehelices of similar stability. Biomacromolecules 11(2):348–356. bm900894b 13. Shoulders MD, Raines RT (2009) Collagen structure and stability. Annu Rev Biochem 78:929–958. annurev.biochem.77.032207.120833 14. Fallas JA, Gauba V, Hartgerink JD (2009) Solution structure of an abc collagen heterotrimer reveals a single-register helix stabilized by electrostatic interactions. J Biol Chem 284 (39):26851–26859. 1074/jbc.M109.014753 15. Nassa M, Anand P, Jain A, Chhabra A, Jaiswal A, Malhotra U, Rani V (2012) Analysis of human collagen sequences. Bioinformation

Synthesis and Application of Collagens for Assembling a Corneal Implant 8(1):26–33. 97320630008026 16. Boudko SP, Engel J, B€achinger HP (2012) The crucial role of trimerization domains in collagen folding. Int J Biochem Cell Biol 44 (1):21–32. 2011.09.009 17. Sundaramoorthy M, Meiyappan M, Todd P, Hudson BG (2002) Crystal structure of nc1 domains. Structural basis for type iv collagen assembly in basement membranes. J Biol Chem 277(34):31142–31153. 1074/jbc.M201740200 18. Paramonov SE, Gauba V, Hartgerink JD (2005) Synthesis of collagen-like peptide polymers by native chemical ligation. Macromolecules 38(18):7555–7561. 1021/ma0514065 19. Olsen D, Yang C, Bodo M, Chang R, Leigh S, Baez J, Carmichael D, Perala M, Hamalainen ER, Jarvinen M, Polarek J (2003) Recombinant collagen and gelatin for drug delivery. Adv Drug Deliv Rev 55(12):1547–1567 20. Bagley D (2019) Declaration of source material - vitricol. https://www.advancedbiomatrix. com/wp-content/uploads/2011/09/Declara tion-of-Source-Material-9-3-15.pdf. Accessed 10 June 2019 21. Tomita M, Munetsuna H, Sato T, Adachi T, Hino R, Hayashi M, Shimizu K, Nakamura N, Tamura T, Yoshizato K (2003) Transgenic silkworms produce recombinant human type iii procollagen in cocoons. Nat Biotechnol 21 (1):52–56. 22. Shoseyov O, Posen Y, Grynspan F (2014) Human collagen produced in plants: more than just another molecule. Bioengineered 5 (1):49–52. 26002 23. Adachi T, Wang X, Murata T, Obara M, Akutsu H, Machida M, Umezawa A, Tomita M (2010) Production of a non-triple helical collagen alpha chain in transgenic silkworms and its evaluation as a gelatin substitute for cell culture. Biotechnol Bioeng 106 (6):860–870. 22752 24. O’Leary LER, Fallas JA, Bakota EL, Kang MK, Hartgerink JD (2011) Multi-hierarchical selfassembly of a collagen mimetic peptide from triple helix to nanofibre and hydrogel. Nat Chem 3:821. nchem.1123. cles/nchem.1123#supplementary-information


25. Islam MM, Ravichandran R, Olsen D, Ljunggren MK, Fagerholm P, Lee CJ, Griffith M, Phopase J (2016) Self-assembled collagen-likepeptide implants as alternatives to human donor corneal transplantation. RSC Adv 6 (61):55745–55749. 1039/C6RA08895C 26. Samarawickrama C, Samanta A, Liszka A, Fagerholm P, Buznyk O, Griffith M, Allan B (2018) Collagen-based fillers as alternatives to cyanoacrylate glue for the sealing of large corneal perforations. Cornea 37(5):609–616. 0000000000001459 27. D’Este M, Eglin D, Alini M (2014) A systematic analysis of dmtmm vs edc/nhs for ligation of amines to hyaluronan in water. Carbohydr Polym 108:239–246. 1016/j.carbpol.2014.02.070 28. Voytas D (2000) Agarose gel electrophoresis. Curr Protoc Mol Biol 51(1):2.5A.1–2.5A.9. mb0205as51 29. Islam MM, Griffith M, Merrett K (2013) Fabrication of a human recombinant collagenbased corneal substitute using carbodiimide chemistry. Methods Mol Biol 1014:157–164. 30. Plasmid cloning by restriction enzyme digest. addgene. protocols/subcloning/. Accessed 10 June 2019 31. Struhl K (1991) Subcloning of DNA fragments. Curr Protoc Mol Biol 13 (1):3.16.11–13.16.12. 1002/0471142727.mb0316s13 32. Diagnostic restriction digest (2018) addgene. tic-digest/. Accessed 10 June 2019 33. Gallagher SR (2006) One-dimensional sds gel electrophoresis of proteins. Curr Protoc Mol Biol 75(1):10.12.11–10.12A.37. https://doi. org/10.1002/0471142727.mb1002as75 34. Mamat U, Wilke K, Bramhill D, Schromm AB, Lindner B, Kohl TA, Corchero JL, Villaverde A, Schaffer L, Head SR, Souvignier C, Meredith TC, Woodard RW (2015) Detoxifying escherichia coli for endotoxin-free production of recombinant proteins. Microb Cell Factories 14(1):57. https://

Chapter 13 Development and Validation of a 3D In Vitro Model to Study the Chemotactic Behavior of Corneal Stromal Fibroblasts Evrim Ceren Kabak, Julia Ferna´ndez-Pe´rez, and Mark Ahearne Abstract Chemotaxis plays a pivotal role in crucial biological phenomena including immune response, cancer metastasis, and wound healing. Although many chemotaxis assays have been developed to better understand these multicomplex biological mechanisms, most of them have serious limitations mainly due to the poor representation of native three-dimensional (3D) microenvironment. Here, we describe a method to develop and validate a novel 3D in vitro chemotaxis model to study the migration of corneal fibroblasts through a stromal equivalent. A hydrogel was used that contained gelatin microspheres loaded with platelet-derived growth factor-BB (PDGF-BB) in the inner section and corneal fibroblasts in the outer section. The cell migration toward the chemical stimuli over time can be monitored via confocal microscopy. The development of this in vitro model can be used for both qualitative and quantitative examinations of chemotaxis. Key words Chemotaxis assays, Chemoattractant, Corneal wound healing, Tissue engineering, 3D microenvironment, In vitro model, Corneal stromal fibroblasts, Growth factor delivery system, Plastic compression


Introduction Chemotaxis involves the movement of cells in response to chemical stimuli. This phenomenon is involved in many physiological and pathological processes including immune response, wound healing, and cancer metastasis by allowing immune cells to reach infection sites, facilitating tissue repair by supporting the migration of cells to close a wound and enabling the cancer cells to relocate at different sites in the body, respectively [1, 2]. The ability to study chemotaxis is important in the development of novel therapeutics and treatments. During corneal wound healing cells respond to chemical signals produced by other cells resulting in chemotaxis [3, 4]. Corneal stromal cells residing in the stromal layer of cornea become activated and migrate toward the wound site to remodel and repair the damages tissue [5–7].

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Evrim Ceren Kabak et al.

In order to study the chemotactic behavior of cells, many assays have been developed. One of the widely used assays is a Boyden chamber, which is based on the quantification of a cell movement from one compartment to another through a porous filter membrane toward the chemoattractant gradient [8]. Despite its robust structure and ability to screen large numbers of cells, it has serious limitations mainly due to poor representation of the 3D microenvironment and inability to maintain a well-defined chemoattractant gradient [9]. Although microfluidic-based chemotaxis assays are able to generate a well-defined chemoattractant gradient with an accurate quantitative analysis capacity, they are unsuitable to be routinely used in biological and biomedical labs, due to the expense and complexity of the system, mainly during the setup and data analysis processes [10]. In order to overcome the limitations in the current models, we developed a method to fabricate a novel in vitro 3D chemotaxis model based on utilizing tissue engineering and biomaterial fabrication techniques, such as fabrication of a growth factor delivery microspheres [11] and plastic compression [12]. The chemotaxis platform developed by the given protocol below will provide a wide range of applicability to a different type of cells and chemotactic agents with distinct concentrations. Moreover, the versatile feature of the model allows a variety of analyses including real-time imaging, immunocytochemical and biochemical analyses to be performed on the model.



2.1 Fabrication and Characterization of Gelatin Microspheres

1. Deionized (DI) water.

2.1.1 Medium and Solution

4. Acetone.

2. Olive oil. 3. Gelatin from bovine skin, type B. 5. Tween 80. 6. Glutaraldehyde. 7. 3.73 mg/mL Glycine solution.

2.1.2 Plastic and Glass

1. Pasteur pipettes (3 mL). 2. Serological pipettes (2 mL, 5 mL, 10 mL, 25 mL, 50 mL). 3. 50 μm cell microsieve. 4. 70 μm cell strainer. 5. Syringe. 6. 18G needles. 7. 2-cm-long flea magnet.

Development and Validation of a 3D In Vitro Model to Study the Chemotactic. . .


8. Round-bottom flasks. 9. Rubber or cork bases. 10. Centrifuge tubes (5 mL, 15 mL, 30 mL, 50 mL). 11. Glass and plastic Petri dishes. 12. Glass and plastic beakers (50 mL, 100 mL, 250 mL, 500 mL). 2.1.3 Equipment

1. Pipettor. 2. Magnetic stirrer. 3. Laboratory ice machine. 4. Freeze dryer. 5. Vacuum oven. 6. Epi-fluorescence microscope. 7. Fume hood.


Cell Culture

2.2.1 Medium and Solution

1. Culture medium for cell growth: Low glucose Dulbecco’s Modified Eagle Medium (DMEM), 10% (v/v) fetal bovine serum (FBS), 100 U/mL penicillin, 100 mg/mL streptomycin, 250 ng/mL amphotericin B. 2. Culture medium for chemotaxis studies: DMEM/F12 (1:1), 50 μg/mL ascorbic acid, 1 insulin-transferrin-selenium solution (ITS). 3. Sterile phosphate buffered saline (PBS), pH: 7.4. 4. Trypsin-EDTA solution (0.25%). 5. Ethanol 70% (v/v) in water. 6. Trypan blue.

2.2.2 Plastic and Glass

1. Microcentrifuge tubes (0.5 mL, 1 mL, 1.5 mL). 2. Cryogenic vials (1 mL, 2 mL). 3. T75 and T175 angled neck cell culture flasks. 4. Micropipette tips 100–1000 μL).

(1–10 μL,





5. Hemocytometer. 2.2.3 Equipment

1. Micropipettes (1–10 μL, 2–20 μL, 20–200 μL, 100–1000 μL). 2. Water bath. 3. Cell culture CO2 incubator. 4. Inverted phase contrast microscope. 5. Class II biosafety cabinet. 6. Refrigerated centrifuge. 7. Refrigerator (+4  C).


Evrim Ceren Kabak et al.

8. Deep freezer (20  C). 9. Ultralow temperature freezer (80  C). 10. Nitrogen tank. 2.3 Development of 3D Chemotaxis Model: The Outer and Inner Ring Gel Model

1. Corneal fibroblasts.

2.3.1 Cells, Medium, and Solution

5. Sterile deionized water (dH2O).

2. High concentration rat tail collagen type I in acetic acid. 3. 1 N Sterile NaOH solution. 4. 10 PBS. 6. Trypsin-EDTA (0.25%). 7. 1 PBS, pH: 7.4 8. 100 μg/mL recombinant human platelet-derived growth factor-BB (PDGF-BB).

2.3.2 Plastic and Glass

1. 24-Well plates. 2. RAFT™ 3D cell culture system. 3. Aid kit box.

2.3.3 Equipment

1. Tweezers. 2. Scissors. 3. Mini and microcentrifuge.

2.4 Validation of 3D Chemotaxis Model

1. 2 μM Calcein acetoxymethyl (Calcein AM).

2.4.1 Cell Viability Analysis

3. Confocal laser scanning microscope (CLSM).

2.4.2 Enzyme-Linked Immunosorbent Assay

1. ABTS ELISA Buffer Kit.

2. 4 μM Ethidium homodimer-1 (EthD-1). 4. ImageJ software.

2. Human PDGF-BB Mini ABTS ELISA Development Kit. 3. Disposable reagent reservoir. 4. 8-Channel micropipette (10–100 μL, 30–300 μL).

2.4.3 Immunocytochemistry

1. Fixation solution: 4% paraformaldehyde (PFA) in PBS. 2. F-Actin staining solution: Tetramethylrhodamine B isothiocyanate (TRITC)-conjugated phalloidin in PBS with a ratio of 1:2000. 3. DAPI nuclear staining solution: 40 ,6-diamidino-2-phenylindole (DAPI) in PBS with a ratio of 1:2000.

2.4.4 Live Cell Imaging

Live cell imaging solution: Reconstitute one 50 μg vial of CellTrace Oregon Green in 20 μL DMSO.

Development and Validation of a 3D In Vitro Model to Study the Chemotactic. . .




3.1 Fabrication of Gelatin Microspheres

Gelatin microspheres are fabricated by using water-in-oil emulsion technique, crosslinked, freeze-dried, and sterilized by dehydrothermal (DHT) treatment (Fig. 1a). 1. Pour 100 mL olive oil into a round-bottom flask (labeled flask 1) with 2-cm-long flea magnet. 2. Place the flask on the rubber base in a large beaker surrounded with warm water and allow the oil to heat up to 42  C (see Note 1) on the magnetic stirrer. 3. Weigh out 2 g of gelatin and pre-warm DI water in the 30 mL tube while stirring on the magnetic stirrer (see Note 2). 4. Once the gelatin is fully dissolved, add the gelatin solution to the olive oil dropwise approximately 1 cm from the center of the flask. 5. Turn off the heat and continue to stir for 10 min. 6. Remove the warm water surrounding the flask and place the flask into the beaker with enough ice to surround it, while continuing to stir for 30 min. 7. Add 40 mL prechilled acetone on ice to the flask and continue to add ice if it starts to melt. 8. Pour the contents of the round-bottom flask through a 50 μm cell microsieve in a funnel into bottle marked waste. 9. Wash all the olive oil through the sieve using chilled acetone into a waste bottle (see Note 3). 10. Once the microspheres (MS) have been fully washed, leave them in a glass Petri dish for 5 min to be completely dried off (see Note 4). 11. Prepare 0.1% (v/v) solution containing Tween 80 and glutaraldehyde in 30 mL tube surrounded with ice. Add the dry MS into the solution and stir for 1 h. 12. Pour the solution through a new 50 μm cell microsieve into a waste bottle and wash with approximately 100 mL DI water. 13. Add the MS into a prechilled glycine solution, prepared by dissolving glycine in DI water to a concentration of 3.73 mg/mL in a round-bottom flask and stir the solution on ice for 1 h. 14. Pour the solution through a new 50 μm cell microsieve into a waste bottle and wash with approximately 100 mL DI water. 15. Store the MS at 4  C overnight in 40 mL glycine solution. 16. Pour through a new 50 μm cell microsieve and repeat steps 13 and 14 twice.


Evrim Ceren Kabak et al.

Fig. 1 Illustration of work scheme for the development of the 3D chemotaxis model. (a) Fabrication and characterization of gelatin microspheres. (b) Fabrication and characterization of outer and inner ring gel model

Development and Validation of a 3D In Vitro Model to Study the Chemotactic. . .


17. Place the MS into a Petri dish with approximately 10 mL DI water (see Note 5). 18. Freeze-dry the samples by initially freezing at 30  C for 1 h and then freeze-drying under 200 mbar vacuum at 1  C overnight (see Note 6). 19. Aliquot 6 mg freeze-dried MS into 1.5 mL labeled centrifuge tubes. 20. Place tube into a vacuum oven and set the temperature and pressure to 110  C and 50 mbar, respectively, and leave for 24 h. 21. Once the treatment procedure is complete, leave the samples to cool. 22. Collect and store the samples in a cool, dry place (preferably at 4  C). 3.2 Fabrication of Collagen Hydrogels

1. Calculate the required volume of collagen using the stock collagen concentration (SCC) and the required collagen concentration (RCC) (see Note 7). 2. Determine the volume of each reagent required for preparing outer and inner collagen hydrogels based on the given formulas: Volume of 10  PBS ¼ Volume of stock collagen ¼

Total volume 10

Total volume  RCC SCC

Volume of NaOH ¼ Volume of stock collagen  0:023 Volume of dH2 O ¼ Total volume  10  PBS  collagen  NaOH 3. Keep all reagents on ice to prevent the early polymerization. 3.3 Fabrication of Cell-Laden Hydrogel: Outer Gel

1. Remove the culture medium in the cell culture flask containing corneal fibroblasts and wash cells with PBS. 2. Add trypsin-EDTA to the flask and incubate for 5 min at 37  C to detach the cells from the surface. 3. Add medium to halt the trypsinization and collect the cell suspension into a tube. 4. Centrifuge the tube for 5 min to obtain a pellet of cells. 5. Remove the supernatant from the tube and resuspend in 2 mL of medium. 6. Mix 10 μL of the cell suspension and 10 μL of trypan blue and count the cells in each 1 mm2 quadrant of a hemocytometer. Calculate the number of cells in the solution (see Notes 8).


Evrim Ceren Kabak et al.

7. Suspend cells in 10 PBS prior to the hydrogel to being formed with a concentration of 15,000 cells per 90 μL 10 PBS. 8. Mix the 10 PBS and cells with the other collagen reagents (NaOH, dH2O, and stock collagen). 3.4 Fabrication of Microsphere Embedded Hydrogel: Inner Gel

1. Add 15 μL PDGF-BB with 100 μg/mL stock concentration to the 1.5 mL tube with 6 mg microspheres for a final concentration of 300 ng per 300 μL gel and store the loaded microspheres at 4  C overnight. 2. Wash the microspheres with 1 mL PBS to remove unbound PDGF-BB, centrifuge for 5 min, and remove the supernatant. 3. Once the loading step is completed, suspend the microspheres in 150 μL, 10 PBS (see Note 9). 4. Add 10 PBS and microspheres into the other collagen reagents.

3.5 3D Chemotaxis Model

The outer and inner ring gel model is composed of two parts: outer ring gel with the cells and inner ring gel with the microspheres (Fig. 1b). The preparation of the model includes several steps that should be carried out in a sterile container that allows the inverted micropipette tips placed in a 24-well plate to fit when closed (Fig. 2). 1. Place an inverted P1000 micro-pipette tip in the inner part of a well of 24-well plate using sterile tweezers (Fig. 2a). 2. Add 900 μL cell-laden collagen solution into the region between the micro-pipette tip and the well (Fig. 2b) and place the aid box in the incubator at 37  C for 15 min. 3. Remove the micro-pipette tip gently from the well (Fig. 2c). 4. Pipette 300 μL microspheres embedded collagen solution directly into the inner area of the well (Fig. 2d) and incubate at 37  C for another 15 min to allow polymerization of the inner collagen gel. 5. Apply a RAFT absorber to the top of the gel to plastically compress the gels (see Note 10). 6. Add the culture medium to each well to initiate chemotaxis studies.

3.6 Validation of Model

1. Mix 4 μm EthD-1 and 2 μm Calcein with PBS in a 15 mL falcon tube and vortex for 3 min.

3.6.1 Cell Viability Analysis

2. Wash the gels with PBS and then add approximately 1 mL freshly prepared Live/Dead solution to each well. 3. Incubate the samples for 1 h at 37  C.

Development and Validation of a 3D In Vitro Model to Study the Chemotactic. . .


Fig. 2 Fabrication steps of the outer and inner ring gel model. (a) Placement of an inverted micropipette tip. (b) Addition of the collagen solution for the outer gel. (c) Removal of micropipette tip after the gel polymerization. (d) Addition of collagen solution for the inner gel and (e) placement of RAFT™ absorber on the top of the gel model

4. Wash the gels with PBS three times to remove unbound reagents and examine using a CLSM. 5. Cell viability can be quantified by counting the number of green (live) and red (dead) cells present. 3.6.2 ELISA

1. Once the gelatin microspheres are loaded with various concentration of PDGF-BB, place the microspheres in culture medium and maintain them in the incubator, at 37  C (see Note 11). 2. Change the medium every 3–4 days, collect and store the aspirated medium at 20  C. 3. Quantify the PDGF-BB released into the medium by utilizing a suitable ELISA Kit (e.g., ELISA Kit including ABTS ELISA Buffer Kit and Human PDGF-BB Mini ABTS ELISA Development Kit).

3.6.3 Immunocytochemistry

1. Fix the prepared outer and inner ring gel model samples with 4% PFA for 15 min at room temperature and wash the samples with PBS to remove the PFA residuals. Keep the samples at 4  C. 2. Mix the TRITC-conjugated phalloidin and DAPI dyes with PBS in a ratio of 1:2000.


Evrim Ceren Kabak et al.

Fig. 3 Confocal microscopy Z-stack image of the overall outer and inner ring gel model. Phalloidin and DAPI staining display F-actin and nucleus of cells, respectively, in the outer gel. Microspheres are embedded in the inner gel. The green dashed line indicates the interface of the two gels (magnification: 10, scale bar: 500 μm)

3. Stain the samples with the two dyes for 45 min at room temperature (see Note 12). 4. Examine the samples and acquire Z-stack images of the samples using LSCM (Fig. 3). 3.6.4 Live Cell Imaging

1. Trypsinize the desired concentration of corneal fibroblasts and collect the cells from the cell culture flask to the falcon tube (see Note 13). 2. Prepare 2.5 mg/mL stock solution of CellTrace by reconstituting the content of one vial (50 μg) of CellTracein 20 μL of DMSO in the meantime. 3. Add 1 μL of freshly made stock solution in DMSO to each mL of cell suspension for a final working concentration of 2.5 μg/ mL, cover the tube with an aluminum foil, and incubate for 20 min at 37  C. 4. Centrifuge the suspension for 5 min, remove the supernatant, and resuspend the pellet in the fresh medium for chemotaxis studies. 5. Mix the cell suspension with 10 PBS to prepare the cell-laden collagen hydrogels. 6. Once the outer and inner gels are prepared, examine the samples by imaging them using LSCM.

Development and Validation of a 3D In Vitro Model to Study the Chemotactic. . .



Notes 1. In order to reach the heat of 42  C, we recommend setting the temperature control of magnetic stirrer to 100  C and monitoring the temperature of the oil. Once it reaches 42  C, switch off the heat. 2. The gelatin solution should be stirred until all the gelatin particles have been dissolved homogeneously. We recommend trying to avoid producing bubbles while stirring. The tube can be placed under warm water tap if it starts to get cell too quickly. Once the gelatin is dissolved uniformly in water, the large beaker with olive oil should be warmed again using a magnetic stirrer. 3. Washing the olive oil residuals off the microspheres on the 50 μm cell microsieve (no.1) requires approximately 700–800 mL acetone. 4. Microspheres should turn a lighter color when they are dry. 5. Prior to freeze-drying, the size uniformity of the microspheres should be inspected. A microscopy can be used to record images of the microspheres placed in DI water and their size distribution can then be quantified using appropriate software, e.g., ImageJ. 6. In the case that the freeze-dryer is unavailable at the time that the MS are ready, they can be stored in dH2O at 4  C. The ramp speed should be set at 1  C per minute in the freezedrying procedure. 7. We recommend the required RCC of collagen hydrogels should be 3.5 mg/mL and the SCC should be checked on the collagen bottle. 8. In order to calculate the cell number, the following equation should be utilized:  Total cell number ¼ Cell count  1  104  Volume  Dillution factor 9. In order for inner gel of the chemotaxis model to be differentiated from the outer gel, one of the reagents of the collagen solution—10 PBS—can be mixed with phenol before the inner gel preparation. 10. Once you place the RAFT™ absorber on the top of the chemotaxis models, we suggest leaving the 24-well plate for 15 min in the class II biosafety cabinet. During that time, a controlled amount of liquid from hydrogels will be taken up by the absorbers. Consequently, the concentration of hydrogels will be increased, and the thickness of them will reach approximately 100 μm.


Evrim Ceren Kabak et al.

11. ELISA should be conducted for two purposes: (1) to obtain the cumulative release curve of PDGF-BB with various concentrations of the growth factor, (2) to quantify the release of PDGF-BBs into medium directly from collagen hydrogel. 12. We suggest using TRITC-conjugated phalloidin and DAPI dyes to visualize F-actin and nucleus of cells, respectively. 13. The staining of cells with CellTrace should be conducted prior to the preparation of cell-laden collagen hydrogels so that when the cells are embedded in the gels, no additional staining steps are required.

Acknowledgement This research is supported by European Research Council starting grant [EYEREGEN-637460]. References 1. Kolsch V, Charest PG, Firtel RA (2008) The regulation of cell motility and chemotaxis by phospholipid signaling. J Cell Sci 121 (Pt 5):551–559. jcs.023333 2. Rees PA, Greaves NS, Baguneid M, Bayat A (2015) Chemokines in wound healing and as potential therapeutic targets for reducing cutaneous scarring. Adv Wound Care 4 (11):687–703. wound.2014.0568 3. Jester JV, Petroll WM, Cavanagh HD (1999) Corneal stromal wound healing in refractive surgery: the role of myofibroblasts. Prog Retin Eye Res 18(3):311–356 4. Fernandez-Perez J, Ahearne M (2019) Influence of biochemical cues in human corneal stromal cell phenotype. Curr Eye Res 44 (2):135–146. 02713683.2018.1536216 5. Matsuda H, Smelser GK (1973) Electron microscopy of corneal wound healing. Exp Eye Res 16(6):427–442. 1016/0014-4835(73)90100-0 6. Kratz-Owens KL, Hageman GS, Schanzlin DJ (1992) An in-vivo technique for monitoring keratocyte migration following lamellar keratoplasty. Refract Corneal Surg 8(3):230–234 7. Lee TJ, Wan WL, Kash RL, Kratz KL, Schanzlin DJ (1985) Keratocyte survival following a

controlled-rate freeze. Invest Ophthalmol Vis Sci 26(9):1210–1215 8. Cano PM, Vargas A, Lavoie JP (2016) A realtime assay for neutrophil chemotaxis. BioTechniques 60(5):245–251. 2144/000114416 9. Whitehead BC, Bezuidenhout D, Chokoza C, Davies NH, Goetsch KP (2016) Cast tube assay: a 3-d in vitro assay for visualization and quantification of horizontal chemotaxis and cellular invasion. BioTechniques 61(2):66–72. 10. Li J, Lin F (2011) Microfluidic devices for studying chemotaxis and electrotaxis. Trends Cell Biol 21(8):489–497. 1016/j.tcb.2011.05.002 11. Ahearne M, Buckley CT, Kelly DJ (2011) A growth factor delivery system for chondrogenic induction of infrapatellar fat pad-derived stem cells in fibrin hydrogels. Biotechnol Appl Biochem 58(5):345–352. 1002/bab.45 12. Cheema U, Brown RA (2013) Rapid fabrication of living tissue models by collagen plastic compression: Understanding threedimensional cell matrix repair in vitro. Adv Wound Care (New Rochelle) 2(4):176–184.

Chapter 14 Femtosecond Laser-Assisted Surgery for Implantation of Bioengineered Corneal Stroma to Promote Corneal Regeneration Neil Lagali and Mehrdad Rafat Abstract The femtosecond laser has achieved widespread use in ophthalmology owing to its ability to deliver focused high energy that is rapidly dissipated and thereby does not damage surrounding tissue outside the precise focal region. Extremely accurate and smooth cuts can be made by the laser, enabling a range of applications in anterior segment surgery. Minimally invasive corneal surgical procedures can be performed using the femtosecond laser, and here we describe the application of such procedures to improve implantation of bioengineered materials into the cornea. Bioengineered corneal tissue, including the collagenous corneal stroma, promises to provide a virtually unlimited supply of biocompatible tissue for treating multiple causes of corneal blindness globally, thereby circumventing problems of donor tissue shortages and access to tissue banking infrastructure. Optimal implantation of bioengineered materials, however, is required, in order to facilitate postoperative wound healing for the maintenance of corneal transparency and avoidance of postoperative complications such as scarring, inflammation, and neovascularization. Moreover, the avoidance of a detrimental physiological physiological wound healing response is critical for facilitating the corneal stromal regeneration enabled by the bioengineered stroma. Without proper implantation, the tissue response will favor inflammation and pathologic processes instead of quiescent keratocyte migration and new collagen production. Here we describe several procedures for optimized biomaterial implantation into the corneal stroma, that facilitate rapid wound healing and regenerative restoration of corneal transparency without the use of human donor tissue. A step-by-step methodology is provided for the use of the femtosecond laser and associated techniques, to enable seamless integration of bioengineered materials into the corneal stroma. Key words Femtosecond laser, Cornea, Biomaterial, Corneal transplantation, Corneal blindness, Artificial cornea


Introduction Bioengineered corneas, either the full cornea, or epithelial, stromal, and endothelial parts separately or in combination, have been the subject of intense research due to the worldwide shortage of suitable human donor tissue for transplantation [1]. Bioengineered or biosynthetic corneas, unlike keratoprostheses, are implanted

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Neil Lagali and Mehrdad Rafat

scaffolds designed to support endogenous corneal regeneration and repopulation of the implanted region with host cells and nerves [2]. The implantation technique itself, however, can represent an additional challenge for healing and regeneration, as the surgical technique may trigger inflammation and neovascular response. For example, in preclinical studies implanting a biomaterial in the alkaliinjured rabbit cornea [3], or in the rat cornea [4], the surgical sutures themselves trigger inflammation and neovascularization. Likewise, in a phase I human clinical trial conducted in Sweden, recombinant human collagen III (RHCIII) corneal substitutes implanted as deep anterior lamellar grafts were secured with overlying sutures [5]. The sutures delayed wound healing and caused inflammation and stromal melting in some patients [6]. There is therefore a need to develop less invasive implantation techniques for bioengineered tissue in the eye that minimize the surgical trauma and facilitate faster healing of the surgical wound. Using less invasive techniques, regeneration instead of inflammation-driven wound healing would be favored within the recipient tissue. Here, we focus on bioengineering of the corneal stroma, and the means by which stromal replacements can be surgically implanted into the cornea in a minimally invasive manner. Intra-stromal surgery has been facilitated by the advent of the femtosecond surgical ophthalmic laser [7], enabling precise incisions to be made within the cornea, at a selectable depth. The femtosecond laser technique is based on the principle of longwavelength, deep infrared light with a very short pulse duration (in the hundreds of femtoseconds regime), and high repetition rate (in kHz range). The light is focused to a very small spot size of a few μm in diameter, forming a localized plasma in the tissue that creates a cavitation bubble of carbon dioxide and water [8], which separates the layers (lamellae) of the surrounding corneal stromal tissue, as the bubble expands. By this method, a very smooth interface between cut surfaces can be obtained. The femtosecond laser is nowadays widely used in ophthalmic surgery, for laser in situ keratomileusis (Femtosecond Laser-Assisted LASIK), implantation of intracorneal ring segments, small incision lenticule extraction (SMILE) procedures, penetrating/lamellar keratoplasty, and cataract surgery [8]. We recently introduced the use of the femtosecond laser for the intrastromal implantation (femtosecond laser-assisted intrastromal keratoplasty; FLISK) of bioengineered corneal substitutes [9] in order to avoid the need for surgical sutures. This method can reduce suture-induced astigmatism in the transplant and may avoid an excessive epithelial and stromal wound healing response [6], thereby facilitating stromal regeneration following biomaterial implantation [9]. The minimally invasive FLISK procedure resulted in rapid wound healing due to retention of the recipient epithelium and Bowman’s layer [10], without major disruption to the corneal subbasal nerve plexus and associated

Laser-Assisted Biomaterial Implantation


postoperative complications. The procedure avoids inflammation and achieves a 100% implant retention rate without affecting the corneal transparency and structure. A similar procedure is starting to be adopted in a clinical setting for intra-stromal implantation of donor refractive lenticules [11]. Here we detail the method of femtosecond laser surgical implantation of laboratory-made biomaterials in the cornea. Such bioengineered stromal materials could avoid the need for use of human donor tissue and associated immune responses. This chapter describes preclinical implantation in animal models; however, the steps may be applicable (with appropriate modifications) to human implantation.


Materials 1. Femtosecond laser. Here, we describe the use of the IntraLase iFs 150 kHz 5th Generation femtosecond laser manufactured by Abbot Medical Optics. Equivalent ophthalmic surgical femtosecond lasers are available from other manufacturers; however, the procedures may require some adaptation from the methods described here, in order to achieve the same effect. Femtosecond laser placement in a temperature and humidity controlled environment is required to ensure optimal operating conditions for the laser. Although not strictly required for animal studies, a clean room/sterile environment is recommended to avoid the possible contamination by airborne particles that could become trapped within the implant and/or cornea during the surgical procedures. 2. Single-use disposable patient interface module for the femtosecond laser (Fig. 1). This consists of a sterile “applanation cone” with flat glass plate that is fitted into the laser unit. The laser beam travels through the glass plate, which applanates the cornea. The interface module also contains a sterile plastic ring with suction that is designed to be placed on the cornea and under the eyelids and is designed to hold the glass ring and applanation cone in place during surgery (see Note 1). 3. Holder for eye globes. It may be desirable to practice the proposed laser procedures ex vivo in whole porcine or bovine eye globes, prior to performing the procedures in vivo. In this case a suitable holder for eye globes is required, which can stabilize the eye tissue and provide adequate outward pressure to simulate the intraocular pressure. An apparatus used for this purpose is shown in Fig. 2 below. 4. Eyelid speculum (blepharostat), for maintaining open eyelids during the procedure.


Neil Lagali and Mehrdad Rafat

Fig. 1 Patient interface module. This disposable single-use module consists of a suction ring connected to a syringe to achieve vacuum suction on the eye, and an applanation cone. The right image is a detail of the cone, where the glass plate is marked. The glass applanates the cornea by downward pressure, and also transmits the laser beam

Fig. 2 Eye globe holder. The ex vivo eye is placed into the holder and the syringe is used to hold the eye in place and ensure sufficient outward pressure from the eye globe

5. Topical anesthetic eye drops, to be instilled prior to blepharostat placement, prior to the laser procedure, and immediately postoperatively. 6. Surgical forceps for removing cut disc of tissue, handling and inserting of biomaterial. 7. Blunt or flattened surgical tool for dissecting tissue or lifting a flap of tissue after laser cutting. 8. Corneal button punch-type trephine for cutting the biomaterial into a circular button of a predetermined diameter. 9. Surgical marking pen for marking the center of the pupil prior to laser alignment.

Laser-Assisted Biomaterial Implantation


10. Postoperative analgesics, steroid eye drops, antibiotic eye ointment. 11. Anterior segment optical coherence tomography (OCT) equipment, to examine the cornea preoperatively and postoperatively.



3.1 Preparation of Laser

1. Start the laser and allow a warmup period after a cold start, usually 30–60 min. 2. After warm-up, the procedural details must be entered into the laser system, including identification of patient and eye and selection of type of procedure. It is important to specify the desired diameter of the operation zone in the cornea, taking into account the available diameter of biomaterial and the diameter of the cornea in the animal model used. Some femtosecond laser systems may have a minimum cutoff value for the diameter of the cutting region, such as 7 mm. If a smaller diameter is required (due to a smaller eye, or for testing implantation in only a specific zone of the cornea), then the laser may need to be used in a “research mode,” which may allow a more flexible choice of cutting diameter. Check the laser manual and contact the laser supplier for more details.

3.2 Choice of Laser Procedure(s)

1. Determine the laser procedures to be used for stromal implantation of bioengineered materials. Several options are available, depending on the specific application to be tested. The femtosecond laser provides the flexibility to combine various types of basic cuts to develop different surgical procedures: (a) A curved, arc-shaped cut of selectable arc length and depth is available to provide a limited-size incision for tissue insertion and excision (b) A full 360 circular cut of selectable depth and location (anterior or posterior) is available for anterior, posterior, or mid-stromal procedures (c) A flap-cut of selectable flap depth, arc length, and hinge position is available for the creation of an anterior corneal flap (d) A full circular lamellar cut of selectable depth and diameter is available to separate different layers of the stromal tissue 2. If required, combine the above cuts for implantation of biomaterials of different sizes by different methods. As an example, in Fig. 3, the procedure of intra-stromal keratoplasty [9] is illustrated. Here, an implant of 3 mm diameter and 150 μm thickness is to be implanted in a rabbit cornea of thickness 370 μm.


Neil Lagali and Mehrdad Rafat

Fig. 3 Femtosecond laser-enabled intra-stromal keratoplasty (FLISK), implemented in a rabbit cornea. A small 3 mm diameter biomaterial implant is inserted into an intra-stromal pocket created by three laser cuts (2 lamellar, 1 circular side cut). The removal of native tissue and insertion of the biomaterial is achieved by means of an access cut that is angled at 45 and provides access to the implant region. Note that no sutures are required to maintain the implant in place in the stroma

Fig. 4 Hybrid flap—anterior lamellar keratoplasty procedure for intra-stromal implantation of larger diameter and thicker implants, for replacing the majority of the corneal stroma (in cases of dystrophies or scars for example)

The procedure involves cutting and removing a disc of native tissue from the cornea and inserting a biomaterial with similar dimensions into the created pocket. The laser cuts used are two circular lamellar cuts (top and bottom interfaces), a 360 circular side cut, and a 60–90 arc-shaped side access cut, used to extract the native tissue and insert the biomaterial. The access cut is the only region where the epithelium is cut. Outside this access cut, the epithelium remains intact. 3. Determine if a hybrid intra-stromal procedure is required, to facilitate placement of larger and deeper stromal implants. An example of this is given in Fig. 4, where a LASIK flap is combined with an anterior lamellar keratoplasty procedure.

Laser-Assisted Biomaterial Implantation


The laser cuts used are a flap cut, a circular side cut, and a lamellar cut. To remove the native tissue, the flap is “peeled back” (held in place only by the hinge) and the underlying lamellar disc is removed. The biomaterial implant is then placed into the lamellar bed, and the flap is replaced and fixed in place using typically four superficial interrupted sutures around the circumference of the flap. 3.3 Laser Programming

1. Design the laser cutting procedure. When the laser is programmed to perform a series of cuts in succession, it must be kept in mind that the most posterior cuts will be made first, followed by the more anterior cuts. This procedure of deepestto-shallowest cuts is important, as one must avoid focusing the laser beam through a plane of the cornea that has already been cut. This is because the interface and refractive index changes in the cut plane can disrupt the laser beam. For a similar reason, cutting through dense scar tissue or other opacities should be avoided as these can disperse the laser beam and result in unpredictable results. 2. If possible, program multiple laser procedures ahead of time, with appropriate parameter values entered and saved into the system prior to starting the procedure (see Note 2). 3. Prior to programming the laser, it is recommended to first perform OCT imaging of the eye to be cut, to determine the pachymetric thickness of the cornea within the proposed cutting zone (see Note 3). Once the appropriate dimensions are known, the parameters for the laser cuts can be entered. For lamellar cuts, the desired lamellar depth can be entered. For deep lamellar cuts, the depth should be OCT guided (see Note 3). In general, more anterior lamellar cuts can be associated with a greater wound healing-fibroblast response and associated corneal haze, compared to more posterior cuts. 4. The diameter of the lamellar cut should be specified, and should exceed the desired diameter by 50 μm on each side (100 μm total). For example, if a 7 mm diameter lamellar cut is required, then specify a diameter of 7.1 mm. This is done to ensure that the subsequent circular side cut will intersect the lamellar cut, to aid in ease of tissue removal. 5. Specify the laser energy for the lamellar cut (0.6 is used in the iFs 150 system, but will need to first be tested with ex vivo tissue on other systems, to determine optimum energy for smooth separation of tissue). A raster pattern of the laser spot is chosen for cutting. Next, the laser spot separation is chosen to be 3 μm. 6. Specify the 360 anterior side cut. The diameter should be exactly the desired diameter (7 mm in the above example). The depth of the side cut is specified by two parameters, the


Neil Lagali and Mehrdad Rafat

posterior depth (at which the laser cutting starts) and the “depth in glass” at which the laser cutting stops. “Depth in glass” refers to the surface of the applanation cone glass plate in Fig. 1, which is in contact with the cornea. The stop depth of the side cut is referenced from this glass plane. If the anterior side cut should go through the entire epithelium (for instance in standard lamellar keratoplasty), then a “depth in glass” value of 50 μm should be used (which means the laser will actually cut 50 μm into the glass, ensuring that the full epithelium is also cut). Otherwise, for purely intra-stromal procedures, a negative value for “depth in glass” should be used, for example, 100 μm if the side cut should end 100 μm below the epithelial surface. As in the case of the lamellar cut, the depth of the anterior side cut should exceed the desired depth by 20 μm on each side (top and bottom), to ensure overlap with the lamellar cut. So if a stromal disc of 200 μm thickness is to be removed (and equivalent disc of biomaterial to be inserted), then the posterior depth should be 20 μm deeper than the posterior lamellar cut depth, and 20 μm shallower than the anterior lamellar cut depth. A higher energy (2.0) is used for the anterior side cut, and the full 360 arc length should be specified to obtain a circular/cylindrical cut. Usually for an applanated cornea, the angle of the side cut should be 90 (vertical cut). Laser spot separation is again 3 μm. 7. Specify the flap cut. A flap diameter should be chosen that is larger than the biomaterial to be implanted, typically 1 mm larger. This facilitates suturing of the flap without contact to the implanted biomaterial, and a wound healing response that is kept away from the biomaterial. Possible downgrowth of epithelium under the flap is also minimized when the flap size is larger than the implant size. Typical flap depth is 50–100 μm depending on the corneal thickness. The hinge can be placed at any location (nasal, temporal, superior, or inferior). 8. Specify the arc-shaped access cut. For this cut, an arc length of 60–90 can be used, to facilitate tissue excision and biomaterial insertion; however, typically 60–70 is recommended if a suture is to be avoided. With larger openings, one or two sutures need to be placed to ensure the biomaterial does not extrude through the opening. The arc length is also dependent on how pliable and elastic the biomaterial is; with biomaterials that can easily be folded or rolled, a smaller opening can be used, while more brittle materials will risk tearing if folded and thus require larger openings. The access cut angle is typically 45 , extending from the intra-stromal pocket and through the epithelium (see Note 4).

Laser-Assisted Biomaterial Implantation

3.4 Preparation of Biomaterial for Implantation


1. Using the corneal trephine punch for the desired diameter of implant, carefully cut the biomaterial button manually (see Note 5). 2. After trephination, if the biomaterial will not be directly implanted into the recipient cornea, it is important to keep the biomaterial immersed in PBS liquid prior to insertion into the cornea, to ensure it does not dry out and maintains its shape and optimal hydration state.

3.5 Laser Cut, Tissue Excision, and Biomaterial Insertion

1. Mark the pupil center of the eye to be cut with a surgical marking pen. 2. Apply several drops of topical anesthetic to the eye, and using a sponge tip, absorb the excess fluid. 3. Place the eye under the laser and align with the applanation cone (Fig. 5). Guidance LEDs or other laser features (such as a camera/video screen) can be used to assist in manual alignment. Once aligned, proceed to dock the applanation cone to the cornea, to achieve applanation on the eye. Make sure that the glass part of the cone is still centered after applanation, or else make small manual lateral adjustments with the joystick to center the cone on the pupil.

Fig. 5 Applanation cone after insertion into the femtosecond laser. The laser head is moved downward to make contact with the eye during the applanation procedure


Neil Lagali and Mehrdad Rafat

Fig. 6 Real-time monitoring of laser procedures using the video monitor. The left image indicates the posterior lamellar cut forming the lower interface of the intrastromal pocket, while the right image was taken after completion of the posterior lamellar and circular side cuts, and indicates the progress of the anterior lamellar cut

4. Start the laser cutting program using the software interface. When signaled by the software, press down the foot pedal and hold, to perform the programmed cutting sequence. The cutting can usually be followed in real time on the laser video monitor (Fig. 6). Once the sequence has completed, release the foot pedal. 5. Next, without moving the applanation cone, initiate any subsequent cutting procedures through the software interface, and complete these procedures (see Note 6). 6. Once the final laser procedure has been completed, raise the applanation cone from the eye, to return the eye to a normal curved shape. Remove and dispose of the applanation cone. 7. The operated eye should be brought under a surgical microscope located in the vicinity of the femtosecond laser. Using the surgical microscope and a blunt spatula-type surgical instrument, the corneal tissue should be separated gently by drawing the spatula across the cut lamellar surfaces. For a flap, the spatula can be used to open the flap and expose the underlying tissue. For an intra-stromal pocket, the spatula is used to enter the pocket via the access cut, and verify separation of the upper and lower lamellar interfaces. 8. Once the separation of stromal tissue has been achieved, for intra-stromal procedures surgical forceps are used. The forceps are inserted into the pocket via the arc-shaped angled access cut, and the forceps are placed such that the stromal tissue within the pocket is sandwiched between the forceps.

Laser-Assisted Biomaterial Implantation


9. With a gentle pulling motion, the native stromal tissue button is pulled out of the pocket and taken out of the cornea through the access cut. The cut disc of native stromal tissue should be excised in one piece. If, however, there is resistance or there are residual attachments of the button to the surrounding stromal tissue, use the blunt spatula to gently separate these attachment points. Do not forcefully pull out the stromal button. The optimal laser spot size, separation, and energy will ensure a smooth separation of the stromal button from the surrounding tissue. 10. In the reverse manner from the previous step, place the forceps around the biomaterial button such that the biomaterial is sandwiched between the forceps. The entire diameter of the biomaterial should be placed between the forceps, with the biomaterial ending at the tip of the forceps. 11. Next, carefully insert the forceps through the access cut and into the stromal pocket. Proceed forwards into the pocket with the forceps until the forceps reach the circular side cut of the pocket opposite the access cut. Then, slowly release pressure on the forceps and gently draw out the forceps from the pocket via the access cut (see Note 7). An example of the tissue excision and biomaterial insertion procedure in the rabbit eye is given in Fig. 7. 12. For lamellar procedures such as the flap—lamellar hybrid implantation, the blunt spatula is used to lift the flap to expose the underlying stromal tissue. Next, sharp-tipped forceps are used to gently separate the button from the underlying stroma. The forceps are used to lift the button from the residual corneal bed, leaving exposed a thin layer of posterior stroma, with underlying Descemet’s membrane, and endothelium. The biomaterial is gently placed on this stromal bed, flattened with a blunt spatula, and the flap is then drawn over the biomaterial, and flattened to remove wrinkles. Surgical sutures are then used to anchor the flap to the surrounding anterior stroma, without contact with the biomaterial. Typically 10-0 nylon sutures are used, and knots are buried in the stroma. Examples of suturing patterns are shown in Fig. 8 below. 3.6 Final Surgical Procedures

1. Following biomaterial implantation and suturing, OCT imaging (see Note 8) should be performed to verify the correct positioning of the biomaterial within the recipient stroma (Fig. 9). At this point, the native corneal stroma and biomaterial may appear swollen, yielding an appearance of a very thick cornea. This is a temporary intraoperative phenomenon and will subside within the first postoperative days.


Neil Lagali and Mehrdad Rafat

Fig. 7 Procedure for excision of native tissue cut by the femtosecond laser, followed by implantation of the bioengineered stromal implant. (a) Surgical forceps are used to gently pull the native stromal tissue out of the intra-stromal pocket. (b) Native tissue (white arrow) seen after removal from the femtosecond pocket (black arrow, points to access cut region). (c) Surgical forceps are used to grip the biomaterial, with the forceps holding the biomaterial across the entire diameter, in a sandwiched configuration. (d) The rabbit eye showing the stromal pocket (asterisk), native disc of stromal tissue (white arrow), and bioengineered stroma (black arrow) of identical thickness and diameter (3 mm). (e) Insertion of the bioengineered implant into the stromal pocket via the access cut. The forceps are drawn into the entire length of the pocket until the circular side cut opposite the access cut is reached

Fig. 8 Suturing pattern after biomaterial implantation in porcine eyes. (a) An anterior lamellar implant, where the biomaterial is held in place postoperatively by the use of overlying sutures. (b) An intra-stromal (FLISK) implant where two surgical sutures (arrow) are used to close the access cut region

Laser-Assisted Biomaterial Implantation


Fig. 9 Optical coherence tomography (OCT) images taken after biomaterial implantation in the porcine cornea. (a) Native porcine cornea. (b) Bioengineered stromal tissue (arrow) immediately after implantation intrastromally using the FLISK procedure. Note the postoperative swelling. (c) Bioengineered stromal implant (long arrow) as an anterior lamellar graft. Note the overlying sutures (short arrows)

2. Following OCT examination, animals should be given topical ophthalmic antibiotic ointment and topical steroid eye drops. If desired, a bandage contact lens can be placed over the operated cornea (Fig. 10).


Neil Lagali and Mehrdad Rafat

Fig. 10 OCT images of bandage contact lens (arrows) placed onto operated corneas, immediately postoperatively. The bandage lens helps protect the operated eye and facilitates smooth eyelid movement over the implanted eyes in the postoperative period. The bandage lens facilitates epithelial healing and allows topical eye drops to penetrate into the cornea 3.7 Postoperative Care and Follow-up

1. Postoperatively (and possibly even preoperatively—consult with a veterinarian) animals should be given systemic analgesics daily for at least the first few postoperative days. 2. Topical corticosteroids (e.g., dexamethasone) should be given in the immediate postoperative period and should be maintained at least twice daily for a period determined by the specific laser procedure and the general appearance of the eye postoperatively. For intra-stromal procedures where sutures are not used, the course of topical steroids may last for about 1 week, whereas anterior lamellar and flap procedures may require corticosteroids for 1 month after sutures are removed, typically 3–6 months postoperatively (see Note 9). 3. Minimize the use of general anesthesia during subsequent longitudinal follow-up to preserve the well-being of the animals. Utilize restraining procedures for animals while awake, where possible, to obtain ocular data.

Laser-Assisted Biomaterial Implantation



Notes 1. Different laser systems will have differing methods of corneal applanation and stabilization/alignment of the cornea to the laser beam. It is suggested to follow the laser manufacturer’s instructions to prepare the cornea prior to laser cutting. It is also possible and may be desirable to modify this procedure for animal experiments. For example, in large animals (rabbits and pigs), omitting the suction ring may be advantageous, as obtaining a vacuum may not be feasible in eyes of different sizes and anatomies. Omitting the suction ring is not recommended, however, in humans. 2. It may be advantageous to create several “hypothetical” patients and save the laser parameters for these procedures ahead of time. For example, if one will perform the hybrid flap-lamellar keratoplasty procedure in Fig. 4, the laser configurations for the deep lamellar procedure and for the flap procedure should be saved separately. Then, during surgery, once the deep lamellar procedure is completed, the flap procedure can be quickly started while the eye is still under applanation. Saving the laser parameters for the procedures ahead of time will minimize the time between procedures. There is a risk that the eye moves slightly between subsequent procedures if left too long. Such movement, even of a few micrometers, will bring the cuts out of alignment, resulting in suboptimal results. 3. OCT is essential for determining the corneal thickness prior to laser surgery. This will minimize the risk of corneal perforation with deeper stromal cuts, and result in biomaterial implantation in the desired region of the cornea. With animal corneas, it is important to keep in mind that these also have a variability in corneal thickness. For example, rabbit corneas of the same litter can vary from 320 μm to 420 μm in thickness, depending on the age and individual. For cases where opacities are to be removed, OCT will indicate the exact depth and extent of the opacity, which will aid in planning the laser procedures and parameters. 4. It is important to determine the orientation of the laser system relative to the programming interface. In some cases, the surgeon will have a view of the eye that is upside-down relative to the programming interface. It is important to know whether the surgeon is left- or right-handed, and which hand is preferred for holding the forceps during tissue excision and biomaterial insertion through the access cut. This will determine the exact location of the access cut around the circular implanted region.


Neil Lagali and Mehrdad Rafat

5. Depending on the size of the biomaterial button and the hydration properties, it may be desirable to use a diameter of biomaterial exceeding the diameter of the stromal pocket. This could partially compensate for de-swelling of biomaterials postimplantation and/or relaxation of the cornea after creation of the femtosecond pocket, resulting in a larger diameter pocket than initially intended. The diameter of the biomaterial can typically be increased by 250–500 μm relative to the stromal pocket diameter. 6. Depending on the animal model, some adjustments may be required. For example, rabbit eyes are typically small and require smaller implant sizes and/or omission of the suction ring. Porcine eyes and the anatomic bony plate around the eye may prevent proper alignment of the laser. These features should be checked prior to performing the laser procedures. It may, however, be unavoidable that the applanation to the animal eye is not possible to center on the center of the pupil. In such cases, the laser cuts will be de-centered and may enter the limbal region. This will trigger an inflammation and neovascularization postoperatively (see Fig. 11). 7. It is important that biomaterial insertion is performed using an operating surgical microscope. Under the microscope, it may be possible to see areas of unevenness and folding or wrinkling of the biomaterial. In this case, after biomaterial insertion, use a blunt spatula to flatten the biomaterial within the pocket. 8. An anterior segment OCT unit that is portable and can be configured to examine eyes on an operating table is ideal. Figure 12 below shows examination of a porcine eye using the Optovue iVue OCT. 9. One should aim to perform regular postoperative assessment of implanted eyes, ideally without general anesthesia. Additionally, procedures should be established to handle animals regularly, to facilitate postoperative eye drops to be given at least

Fig. 11 Effect of de-centration of biomaterial implant in the porcine eye. (a) The bony plate surrounding the porcine eye (arrows) may prevent proper alignment of the applanation cone to the pupil center. (b) De-centered cut region (arrow) close to the limbus on one side of the cornea. (c) The side nearest the limbus has limited vessel invasion (arrows) and the inflammation and vascular leakage results in an opaque implant

Laser-Assisted Biomaterial Implantation


Fig. 12 Examination of the implanted eye postoperatively using OCT with the intubated animal lying on the operating table. An OCT that can be placed both vertically and horizontally facilitates the examination

twice daily without use of anesthesia. Starting at 1 month postoperative, general anesthesia can be used to perform more extensive eye examinations (e.g., OCT, slit lamp photos) that typically require a greater degree of compliance than is possible while awake. Eye drops, fluorescein staining, and photography, however, can be performed in rabbits while they are awake.

Acknowledgements This work was supported by a Research Grant (Grant No. 667400) from the European Commission, under the Horizon2020 project “ARREST BLINDNESS.” References 1. Gain P, Jullienne R, He Z, Aldossary M, Acquart S, Cognasse F, Thuret G (2016) Global survey of corneal transplantation and eye banking. JAMA Ophthalmol 134:167–173 2. Lagali N, Fagerholm P, Griffith M (2011) Biosynthetic corneas: prospects for supplementing the human donor cornea supply. Expert Rev Med Dev 8:127–130

3. Hackett JM, Lagali N, Merrett K, Edelhauser H, Sun Y, Gan L, Griffith M, Fagerholm P (2011) Biosynthetic corneal implants for replacement of pathologic corneal tissue: performance in a controlled rabbit alkali burn model. Invest Ophthalmol Vis Sci 52:651–657 4. van Essen TH, Lin CC, Hussain AK, Maas S, Lai HJ, Linnartz H, van den Berg TJ, Salvatori DC, Luyten GP, Jager MJ (2013) A fish


Neil Lagali and Mehrdad Rafat

scale–derived collagen matrix as artificial cornea in rats: properties and potential. Invest Ophthalmol Vis Sci 54:3224–3233 5. Fagerholm P, Lagali N, Carlsson D, Merrett K, Griffith M (2009) Corneal regeneration following implantation of a biomimetic tissueengineered substitute. Clin Transl Sci 2:162–164 6. Fagerholm P, Lagali NS, Merrett K, Jackson WB, Munger R, Liu Y, Polarek JW, So¨derqvist M, Griffith M (2010) A biosynthetic alternative to human donor tissue for inducing corneal regeneration: 24 month follow-up of a Phase I clinical study. Sci Transl Med 2:46ra61 7. Ozulken K, Cabot F, Yoo SH (2013) Applications of femtosecond lasers in ophthalmic surgery. Expert Rev Med Dev 10:115–124 8. Roszkowska A, Urso M, Signorino A, Aragona P (2018) Use of the femtosecond lasers in ophthalmology. EPJ Web of Conferences 167:05004

9. Koulikovska M, Rafat M, Petrovski G, Vere´b Z, Akhtar S, Fagerholm P, Lagali N (2015) Enhanced regeneration of corneal tissue via a bioengineered collagen construct implanted by a nondisruptive surgical technique. Tissue Eng Part A 21:1116–1130 10. Lagali N, Germundsson J, Fagerholm P (2009) The role of Bowman’s layer in corneal regeneration after phototherapeutic keratectomy: a prospective study using in vivo confocal microscopy. Invest Ophthalmol Vis Sci 50:4192–4198 11. Moshirfar M, Shah TJ, Masud M, Fanning T, Linn SH, Ronquillo Y, Hoopes PCS (2018) A modified small-incision lenticule intrastromal keratoplasty (sLIKE) for the correction of high hyperopia: a description of a new surgical technique and comparison to lenticule intrastromal keratoplasty (LIKE). Med Hypothesis Discov Innov Ophthalmol 7:48–56

Chapter 15 The Use of Animal Models to Assess Engineered Corneal Tissue Robert Thomas Brady and Peter W. Madden Abstract Tissue-engineered corneal constructs offer the potential of readily available corneal substitutes for transplantation. As with all medical devices and implants, these constructs require rigorous safety assessments, combined with well-described analyses of the implant’s physical and biological characteristics. Although the constructs are developed in vitro, such studies are currently unable to fully emulate the complex biomechanical and biochemical conditions within living tissue, as well as the interplay between this environment and immunological factors. For these reasons, animal models remain essential to characterize such interactions. They form a stage where corneal implants can be tested for utility and survival in a living location to assess their ability to provide vision and avoid adverse event. Here, we examine the surgical considerations of animal models and we describe how the rabbit can be used for this purpose. This animal has been the routine model for ophthalmological studies and we set out methods to implant corneal constructs with this species. Key words Cornea, Animal surgery, Rabbit, Engineered tissue, Implant


Introduction It is difficult to mimic the complex environment of the living body in order to assess the ultimate value and utility of tissue-engineered corneal constructs. In vitro models, such as those involving tissue or organ culture, are an initial vital source of information with which to devise and optimize the most appropriate corneal construct. Such methods should always be used before contemplating implantation into any living animal or human. Nevertheless, animal surgery still remains the best method to imitate the complex environment of biophysiological processes, biomechanical forces, and immunological challenges that the implant will encounter. With the cornea, this is the most accepted proof of value to justify clinical human trials. The expected scientific benefit of any study must sufficiently justify the use of animals. Scientific justification has to form the

Mark Ahearne (ed.), Corneal Regeneration: Methods and Protocols, Methods in Molecular Biology, vol. 2145,, © Springer Science+Business Media, LLC, part of Springer Nature 2020



Robert Thomas Brady and Peter W. Madden

basis of plans to use animals and the scientific value must vindicate the harm that all animal studies incur. The seminal work of Russell and Burch [1], where animal use is Replaced, Reduced, and Refined, the so-called 3 R’s, should always be paramount, not only leading up to but also following animal surgery. If the required scientific outcome can be met without using live animals, then this path should be followed. Here we present an overview of factors that need to be considered prior to commencing an animal study. We highlight that all animal studies are an ethical balance of harm to benefit and that the selection of which model to use relies upon the scientific outcome to be achieved. The rabbit has been the conventional ophthalmic model and we describe the surgical stages of using this species to assess the survival of an engineered corneal construct.


Preliminary Considerations All animal research must be legal and comply with ethical requirements. Regulation and guidance may be available in your local institution, or through your governmental animal welfare regulating body. There are broad guideline documents available for such studies in Europe [2] and in the USA [3]. One, or more, of these guides should always be followed in the planning for any animal study and the 3R’s rigorously applied. There are also readily available organizational guidance documents, such as the ARVO statement on the use of animals in research [4]. Although the requirements for every study needs to be individually produced, they all have core aspects: (1) the animals should be sourced from a reputable breeder; (2) there must be a veterinarian physically available as necessary, having oversight of the animal’s health and housing; (3) there must be appropriate husbandry facilities and trained staff managing them; and (4) the surgery is to be performed by trained staff, working in adequate surgical facilities. A pilot study with a small number of animals can also often be gainfully used to optimize the procedures in order to minimize animal harm and maximize the scientific outcome.

2.1 Animal Model Selection

In the selection of an animal model, a review is required indicating which species have provided the best, most applicable data in the past and which will be the most useful for current and future research. The final choice should be at the lowest point on the phylogenetic tree to still allow the most scientifically accurate and interpretable results to be achieved, while requiring the fewest number of animals. Which animal model to be used depends upon the scientific requirement of the study. Corneal transplants have been performed with many animals, but there are advantages and disadvantages of

Corneal Surgical Animal Models


any model. Simians, with their closeness to humans, are the ultimate model before clinical trials, but the high intelligence of these species result in complex and expensive management, in addition to significant ethical boundaries. It is routine therefore to begin with a species lower on the phylogenetic tree, and historically the laboratory rabbit has been the routine animal to launch ophthalmic studies [5]. Here, we will further examine this value and explain how surgery can be performed with this species. An additional animal model after the rabbit can also be advantageous to gather further scientific value, such as using the rat [6] or mouse [7], where genetic factors can be more precisely varied to examine immunological events, or the pig [8], which has a cornea that is mechanically closer to that of a human. 2.2 Planning for Animal Studies

Prior to an animal study commencing, there is a requirement to review the suitability of the model and also to assess the confounding aspects of the particular model. The benefits and weaknesses of the rabbit as a model are shown in Table 1. Weaknesses need to be weighed against the particular scientific outcome that is being investigated. The breed of rabbit is also an important determinant, e.g., some breeds that grow significantly can make long-term handling and housing more complex. As a part of refining an animal study to cause the least harm, having the shortest experimental period with the least invasive procedures will clearly be advantageous. Many studies, however, will have set requirements to achieve the desired outcome. One mitigant that may be available is to perform implantation on only one eye of the animal. This allows the animal to retain many normal behaviors if the other eye is left untouched. Importantly, this also has the additional benefit of acting as an internal control. Rabbits are nonetheless troubled if one eye is not optimal; their eye position, one each side of the head, allows two very wide fields of view, but it also means there is a relatively narrow bilateral vision field. Sight loss in the operated eye may be ameliorated to an extent, if the implant can be placed off the visual axis, or by having an implant smaller than the pupil. All procedures should not only be documented with standard operating procedures of what is expected to occur, but there also needs to be procedures for adverse events that may occur. Importantly, this includes procedures to remove or add sutures, or perform lid reconstruction surgery. Consultation with a veterinarian is essential in all animal experiment planning, with specific regard to important determinants such as anesthesia, possible complications or harms of the surgery, and analgesic adjuncts. Seeking the view of a veterinary ophthalmologist who works with the specific species is of value; human ophthalmic surgeons may not be fully aware of the nuances of animal ocular physiology and behavior.


Robert Thomas Brady and Peter W. Madden

Table 1 Issues of rabbit corneal animal model of human transplants Issue


Historically, the model used for corneal studies

Vast accumulated data on the anatomy and physiology of the rabbit eye and its similarity with that of the human

Similar size of the eye to humans

Conventional human surgical instruments can generally be used

Historically, the rabbit has been used for the infamous Draize test

Vast information on the response of the eye to drug and chemicals on eye irritation [9]

Relatively large eye size to body weight

Smaller size makes for easy husbandry

Easy to breed and economical compared with larger Economy allows scientific outcome to be achieved animals at a lower cost Rabbits are docile

Makes for easy handling and examination

Difference in anterior chamber shape compared to Can make deep corneal constructs difficult to use a human, including shallow central depth [10] A nictitating membrane is present

This effectively acts as a third eyelid and so the mechanobiology of the corneal surface differs from humans. The membrane is also delicate and vascularized, and its damage can be problematic

All rabbit corneal endothelial cells retain the ability This may make it unsuitable as a model of human to divide even during adult life [11] penetrating keratoplasty as host endothelial cells may more readily repopulate a construct Limited range of genotypes

This makes genetic studies difficult

No vomit reflex

Fasting can be avoided and aspirating vomit is not an issue

Breath-holding with anaesthetic gases

Requires injectable anaesthesia for induction and careful monitoring

Difficult to intubate and anaesthesia balance can be Requires involved anaesthesia difficult Younger rabbits demonstrate a more active postoperative inflammatory response and there can be marked production of fibrin in the anterior chamber [12]

Such age-related differences can model paediatric humans and so this can be used to advantage. Alternatively, it may adversely interfere with a corneal implant

Lack of corneal rigidity

Constructs may be subject to differing mechanical forces than in humans

Corneal Surgical Animal Models




3.1 Animal Anesthesia and Physiological Monitoring

The anesthetic equipment will vary with the type of anesthesia, but oxygen and a mechanism to administer it will be required in every case. If gaseous anesthesia is used, then there must also be a scavenger unit to capture noxious waste gas, avoiding a risk of exposure to the surgeon and assistants. Intubation is generally the preferred method with many animals; however it can be problematic in rabbits because it is difficult to directly visualize the tracheal opening well, even with a laryngoscope, and so there has been much debate with regard to the best method [13]. We therefore use anesthesia by mask, but in the absence of intubation equipment, it can be more demanding to achieve stable anesthesia and provide assisted respirations where there may be drug-induced bradycardia or breath-holding episodes. Heart rate and oxygen saturation are required monitoring best provided by a pulse oximeter: many pulse meters designed for human use will not read heart rates above 250 beats per minute and are therefore unsuitable for use with the rabbit. Meters that also display waveform are advantageous to demonstrate heartbeats. Hypothermia can result from a lack of movement, or interruption of thermoregulatory mechanisms such as cutaneous vasodilation induced by anesthesia, a cold surgery table, or environment. For this reason, core body temperature should be measured and this can readily be achieved by the use of a rectal thermometer. Breathing needs to be monitored. This can be done visually, counting the breaths against time. If surgical drapes cover the animal, then a breathing monitor that displays physical chest movement is easier.



Operating microscope: A binocular microscope with appropriate lighting is essential for any corneal surgery. Foot controls can maintain sterility, or focus handles can be decontaminated with 70% v/v alcohol, or covered with sterile aluminum foil.


Tonometer: Significant changes from preoperative intraocular pressure can indicate eye perforation or be a sign of rejection. We advise that a contact tonometer is used, as non-contact methods may not measure at specific sites and air-puff methods may give noise that is disconcerting to rabbits.


Slit lamp: The cornea can best be examined with a handheld slit lamp for opacities or other pathologies.


Corneal thickness: A range of equipment can review corneal thickness and ideally implant depth: optical coherence tomography (OCT), confocal microscopy, or an ultrasonic pachymeter. While there are published values of corneal thickness of varying rabbit species, outliers exist. Surgical planning to detect an unusually thin cornea by one of these methods may therefore

Eye Examination


Robert Thomas Brady and Peter W. Madden

avoid perforation of the anterior chamber. As such, OCT or confocal microscopy give not only thickness, but also a structural evaluation and are to be preferred. With any equipment that is calibrated for the normally hydrated human cornea there will need to be calculated adjustment to provide true linear measurement of animal tissue. 3.3 Consumables and Medicaments


The desirable aim is to only use consumables that are certified for human clinical use. If this is not possible, then animal-certified items are acceptable without review. If only research grade items are available, then these should be reviewed as part of the project risk assessment planning. Medicament sources may need to be sourced by a different rank, with animal-certified, human-certified, and then research-certified, depending upon the regulatory control in place.

Operative Procedure in Rabbits Here we focus on anterior lamellar implants, as these involve many of the techniques used in all the procedures.

4.1 Selection Criteria and Acclimatization

Standard operating procedures need to include assessment by a supervising veterinarian of individual animals, as part of their admission to the surgical program. All animals should be reviewed against the animal selection criteria, such as sex, species, size, and weight and any required preexisting health issues for the particular scientific study. They should also have been given sufficient time to acclimatize to the facility conditions before any surgical procedures are undertaken. This is regularly taken to be 7 days, as a minimum. Prior to surgery, a general health review is required, ideally by a qualified veterinarian, plus a microscopic examination of the eye, ideally by an ophthalmologist. Thus, there is the starting point of an animal that is prepared for surgery, with good general and eye health, or with specific morbidities within the study selection criteria.

4.2 Preparation of Operating Theatre

No animal should be prepared for surgery unless the implant construct and the necessary staff, including animal welfare staff, and equipment are available. Surgical planning should start with having all surfaces prepared prior to procedure. Floors should be wet-mopped to reduce dust and operating benches disinfected. If immunocompromised animals are used, additional requirements will be indicated. It is advisable to have a full dry-run of the operating procedure, with the surgeon and all the ancillary workers present. The team can simulate how the animal will be moved through the procedure and where instruments and consumables are sourced.

Corneal Surgical Animal Models


4.3 Health Examination Including Preparatory Eye Examination

Regardless of whether the animal has previously appeared to be suitable for surgery, on the day of the procedure its health should be checked for any change. Changes in weight, temperature, or heart rate could indicate an underlying infection. Both eyes should be examined microscopically prior to anesthesia as corneal lesions are common [14]. Intraocular pressure should also be measured as this base level is valuable in monitoring eye response following construct implantation.

4.4 Anesthesia and Administration of Preoperative Drugs

Preoperative and operative procedures need to be out of sight, sound, and smell from other animals in a quiet area. Rabbits have a strong digestive tract cardiac sphincter [15], which precludes true vomiting and so there is no requirement for preoperative fasting, other than until one to two hours before surgery if intubation may be used, to ensure that the mouth does not contain ingested material. The anesthesia of rabbits can be difficult. Here we describe a technique that initially uses injectable anesthesia that is then supplemented with gaseous anesthesia to extend operating time. An approach of this type permits the long-term anesthesia that may be required to adequately position a construct, especially in the learning phase of surgery with a new device. Having weighed the animal, administer 0.01–0.03 mg/kg of buprenorphine subcutaneously and shave the marginal ear veins of both ears. Lignocaine 3% gel is also applied to these veins to lessen the sensation of needle insertion. The animal is placed in a holding cage alone for one hour without food or water. Buprenorphine can increase the duration of anesthesia and may provide beneficial analgesia for the first hours after surgery [6]. An ear vein is cleaned with an alcohol wipe and a butterfly catheter inserted. Its position and patency is confirmed by use of sterile injectable saline. If the patency is not adequate, then the other ear is used and direct light pressure applied to the needle site of the first until bleeding stops. Once a catheter is functional, it is taped into position on the ear using gauze padding and micropore tape. Xylazine at 3.0 mg/kg and ketamine, 10 mg/kg, mixed well in preprepared 1 ml Luer lock syringes are then administered using an injection port of the catheter, into which a butterfly needle has been inserted. Using a tubed system like this allows movement of the syringe to be decoupled from the catheter to ensure it is not dislodged during movement of the syringe, or the animal. Multiple syringes should be prepared, sufficient for the total anesthetic period. Similarly, 3 mL syringes of sterile 0.9% NaCl should also be readied. This saline should be used periodically throughout the anesthesia to flush the anesthetic line, and not only check that the catheter is patent, but also ensure that the small volume of anesthetic is fully dispensed from the tubing. If the surgery is greater


Robert Thomas Brady and Peter W. Madden

than one hour, then 5 mL of 0.9% NaCl should be given every hour peritoneally, as maintenance fluids. As soon as the animal is anesthetized with the injectable agents, it is put on the heated pad, and supplied with pure oxygen by face mask at a rate of 5 L/min. A rectal thermometer is put in place and a pulse oximeter applied over a prominent vein of the shaved ear. The pulse oximeter may be covered with aluminum foil to prevent the surgical lights interfering with the oximeter light sensor. Testing of anesthetic depth using a toe pinch and corneal reflex should be performed every 10 min and an additional 0.05–0.1 mL bolus anesthetic given if there is any reaction. Rabbits do have a very pronounced breath-holding response to gaseous agents [16]. As soon as anesthesia is confirmed, a very small concentration,