Methods
in
Molecular Biology™
Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK
For further volumes: http://www.springer.com/series/7651
wwwwwwwwwwwwwww
Permeability Barrier Methods and Protocols Edited by
Kursad Turksen Regenerative Medicine Program, Sprott Centre for Stem Cell Research, The Ottawa Hospital Research Institute, Ottawa, ON, Canada
Editor Kursad Turksen, Ph.D Regenerative Medicine Program Sprott Centre for Stem Cell Research The Ottawa Hospital Research Institute Ottawa, ON Canada
[email protected]
Please note that additional material for this book can be downloaded from http://extras.springer.com ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-61779-190-1 e-ISBN 978-1-61779-191-8 DOI 10.1007/978-1-61779-191-8 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2011935456 © Springer Science+Business Media, LLC 2011 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Humana Press is part of Springer Science+Business Media (www.springer.com)
Preface When I approached Dr. John Walker, the Editor-in-Chief of the Methods in Molecular Biology series, with the idea of putting together a protocol volume on the permeability barrier, his immediate response was “What a great idea. We have not covered this topic!” John: Thanks for the great support for this volume from the beginning. Although the topic of the permeability barrier is diverse, the authors herein cover many critical areas, making this volume valuable to almost anybody who is interested in this field. I hope that the many valuable protocols included will be useful for both novices and experts alike. I thank all the contributors for very graciously providing their protocols for this volume. Without them and their willingness to share protocol details, this new volume would not have materialized. Patrick Marton, the Editor of the Methods in Molecular Biology series at Springer, has been a great support and is always available to answer my questions and listen to my suggestions. A very special thanks goes to David Casey for his invaluable help during the production stages of this volume. Ottawa, Ontario, Canada
Kursad Turksen
v
wwwwwwwwwwwwwww
Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
v xi
1 Computational Modeling of the Skin Barrier . . . . . . . . . . . . . . . . . . . . . . . . . . . . Arne Naegel, Michael Heisig, and Gabriel Wittum 2 In Vitro Human Skin Segmentation and Drug Concentration–Skin Depth Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ana Melero, Tsambika Hahn, Ulrich F. Schaefer, and Marc Schneider 3 Transcriptional Regulation of Epidermal Barrier Formation . . . . . . . . . . . . . . . . . Ambica Bhandari, Michael L. Salmans, William Gordon, and Bogi Andersen 4 Epidermal Permeability Barrier Measurement in Mammalian Skin . . . . . . . . . . . . Arup Kumar Indra and Mark Leid 5 Assessment of Permeability Barriers to Macromolecules in the Rodent Endometrium at the Onset of Implantation . . . . . . . . . . . . . . . . . . Brent M. Bany and G. Scot Hamilton 6 Assessment of Intestinal Permeability in (Premature) Neonates by Sugar Absorption Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Willemijn E. Corpeleijn, Ruurd M. van Elburg, Ido P. Kema, and Johannes B. van Goudoever 7 Analysis of Epithelial Cell Shedding and Gaps in the Intestinal Epithelium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Carrie A. Duckworth and Alastair J. Watson 8 Studying Permeability in a Commonly Used Epithelial Cell Line: T84 Intestinal Epithelial Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Rino P. Donato, Adaweyah El-Merhibi, Batjargal Gundsambuu, Kai Yan Mak, Emma R. Formosa, Xian Wang, Catherine A. Abbott, and Barry C. Powell 9 Optimization of the Caco-2 Permeability Assay to Screen Drug Compounds for Intestinal Absorption and Efflux . . . . . . . . . . . . . Barry Press 10 Ouabain Modulates Cell Contacts as well as Functions that Depend on Cell Adhesion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Isabel Larre, Ruben G. Contreras, and Marcelino Cereijido 11 Monitoring of the Dynamics of Epithelial Dome Formation Using a Novel Culture Chamber for Long-Term Continuous Live-Cell Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Judith Lechner, Daniel Hekl, Helmut Gatt, Markus Voelp, and Thomas Seppi
1
vii
33
51
73
83
95
105
115
139
155
169
viii
Contents
12 Measuring Permeability in Human Retinal Epithelial Cells (ARPE-19): Implications for the Study of Diabetic Retinopathy . . . . . . . . . . . . . . . . . . . . . . . Marta Garcia-Ramírez, Marta Villarroel, Lídia Corraliza, Cristina Hernández, and Rafael Simó 13 Analysis of Epithelial Barrier Integrity in Polarized Lung Epithelial Cells . . . . . . . Monika Strengert and Ulla G. Knaus 14 Permeability of Differentiated Human Urothelium In Vitro . . . . . . . . . . . . . . . . . Peter Rubenwolf and Jennifer Southgate 15 Phenotyping the Claudin 11 Deficiency in Testis: From Histology to Immunohistochemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Séverine Mazaud-Guittot, Alexander Gow, and Brigitte Le Magueresse-Battistoni 16 An In Vitro System to Study Sertoli Cell Blood-Testis Barrier Dynamics . . . . . . . Dolores D. Mruk and C. Yan Cheng 17 Analysis of Endothelial Barrier Function In Vitro . . . . . . . . . . . . . . . . . . . . . . . . . Yuping Wang and J. Steven Alexander 18 Role of Endothelial Cell–Cell Junctions in Endothelial Permeability . . . . . . . . . . . Armelle Le Guelte and Julie Gavard 19 In Vitro Analyses of Endothelial Cell Permeability . . . . . . . . . . . . . . . . . . . . . . . . Elizabeth Monaghan-Benson and Erika S. Wittchen 20 Mechano-Transduction and Barrier Regulation in Lung Microvascular Endothelial Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kristina Giantsos, Mark Cluff, and Randal Dull 21 Role of Caveolin-1 in the Regulation of Pulmonary Endothelial Permeability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yu Sun, Richard D. Minshall, and Guochang Hu 22 Assessment of Endothelial Permeability and Leukocyte Transmigration in Human Endothelial Cell Monolayers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Andreas Ludwig, Anselm Sommer, and Stefan Uhlig 23 Permeability of Endothelial Barrier: Cell Culture and In Vivo Models . . . . . . . . . . Alexander N. Garcia, Stephen M. Vogel, Yulia A. Komarova, and Asrar B. Malik 24 Size-Selective and In Vitro Assessment of Inner Blood Retina Barrier Permeability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Matthew Campbell and Peter Humphries 25 Assessment of Permeability in Barrier Type of Endothelium in Brain Using Tracers: Evans Blue, Sodium Fluorescein, and Horseradish Peroxidase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Mehmet Kaya and Bulent Ahishali 26 In Vitro and In Vivo Methods for Assessing FcRn-Mediated Reverse Transcytosis Across the Blood–Brain Barrier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Nadia Caram-Salas, Eve Boileau, Graham K. Farrington, Ellen Garber, Eric Brunette, Abedelnasser Abulrob, and Danica Stanimirovic 27 Evaluation of VEGF-Induced Vascular Permeability in Mice . . . . . . . . . . . . . . . . . Sara M. Weis
179
195 207
223
237 253 265 281
291
303
319 333
355
369
383
403
Contents
ix
28 In Vivo Measurement of Glioma-Induced Vascular Permeability . . . . . . . . . . . . . 417 Jisook Lee, Andrew Baird, and Brian P. Eliceiri 29 In Vivo Optical Imaging of Ischemic Blood–Brain Barrier Disruption . . . . . . . . . 423 Abedelnasser Abulrob, Eric Brunette, Jacqueline Slinn, Ewa Baumann, and Danica Stanimirovic Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441
wwwwwwwwwwwwwww
Contributors Catherine A. Abbott • School of Biological Sciences, Flinders University, Adelaide, SA, Australia Abedelnasser Abulrob • Institute for Biological Sciences, National Research Council of Canada, Ottawa, ON, Canada J. Steven Alexander • Department of Molecular and Cellular Physiology, Louisiana State University Health Sciences Center, Shreveport, LA, USA Bulent Ahishali • Departments of Physiology and Histology and Embryology, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey Bogi Andersen • Departments of Medicine and Biological Chemistry, University of California, Irvine, CA, USA Andrew Baird • Department of Surgery, University of California San Diego, San Diego, CA, USA Brent M. Bany • Department of Physiology and Obstetrics and Gynecology, Southern Illinois University School of Medicine, Carbondale, IL, USA Ewa Baumann • Cerebrovascular Research Group, Institute for Biological Sciences, National Research Council of Canada, Ottawa, ON, Canada Ambica Bhandari • Departments of Medicine and Biological Chemistry, University of California, Irvine, CA, USA Eve Boileau • Institute for Biological Sciences, National Research Council of Canada, Ottawa, ON, Canada Eric Brunette • Cerebrovascular Research Group, Institute for Biological Sciences, National Research Council of Canada, Ottawa, ON, Canada Matthew Campbell • UCD School of Biomolecular Biomedical Sciences, Conway Institute, University College Dublin, Dublin, Ireland Nadia Caram-Salas • Institute for Biological Sciences, National Research Council of Canada, Ottawa, ON, Canada Marcelino Cereijido • Department of Physiology, Biophysics and Neurosciences, Center for Research & Advanced Studies (CINVESTAV), Av. Instituto Politécnico Nacional 2508, Col. San Pedro Zacatenco, Del. GAM C.P. 07360 México C. Yan Cheng • Center for Biomedical Research, Population Council, New York, NY, USA Mark Cluff • Department of Anesthesiology, University of Utah School of Medicine, Salt Lake City, UT, USA Ruben G. Contreras • Center for Research & Advanced Studies (CINVESTAV), Mexico City, México Willemijn E. Corpeleijn • Division of Neonatology, Erasmus MC – Sophia Children’s Hospital, Rotterdam, The Netherlands
xi
xii
Contributors
Lídia Corraliza • Diabetes and Metabolism Research Unit, Institut de Recerca Hospital Universitari Vall d’Hebron, Universitat Autònoma de Barcelona and CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Instituto de Salud Carlos III (ISCIII), Barcelona, Spain Rino P. Donato • Women’s and Children’s Health Research Institute, North Adelaide, SA, Australia Carrie A. Duckworth • Gastroenterology Research Unit, School of Clinical Sciences, University of Liverpool, Liverpool, UK Randal Dull • Department of Anesthesiology, University of Utah School of Medicine, Salt Lake City, UT, USA Ruurd M. van Elburg • Department of Pediatrics, Division of Neonatology, Vrije Universiteit Medical Center, Amsterdam, The Netherlands Brian P. Eliceiri • Department of Surgery, University of California San Diego, San Diego, CA, USA Adaweyah El-Merhibia • Women’s and Children’s Health Research Institute, North Adelaide, SA, Australia Graham K. Farrington • Institute for Biological Sciences, National Research Council of Canada, Ottawa, ON, Canada Emma R. Formosa • Women’s and Children’s Health Research Institute, North Adelaide, SA, Australia Ellen Garber • Institute for Biological Sciences, National Research Council of Canada, Ottawa, ON, Canada Alexander N. Garcia • Department of Pharmacology, University of Illinois, Chicago, IL, USA Marta Garcia-Ramírez • Diabetes and Metabolism Research Unit, Institut de Recerca Hospital Universitari Vall d’Hebron, Universitat Autònoma de Barcelona and CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Instituto de Salud Carlos III (ISCIII), Barcelona, Spain Helmut Gatt • Department of Therapeutic Radiology and Oncology, Innsbruck Medical University, Innsbruck, Austria Julie Gavard • Institut Cochin, Université Paris Descartes, UMR-CNRS 8104, Paris, France Kristina Giantsos • Department of Anesthesiology, University of Utah School of Medicine, Salt Lake City, UT, USA William Gordon • Departments of Medicine and Biological Chemistry, University of California, Irvine, CA, USA Johannes B. van Goudoever • Division of Neonatology, Erasmus MC – Sophia Children’s Hospital, Rotterdam, The Netherlands Alexander Gow • Center for Molecular Medicine and Genetics, Wayne State University, Detroit, MI, USA Armelle Le Guelte • Institut Cochin, Université Paris Descartes, UMR-CNRS 8104, Paris, France Batjargal Gundsambuu • A Women’s and Children’s Health Research Institute, North Adelaide, SA, Australia Tsambika Hahn • Biopharmaceutics and Pharmaceutical Technology, Saarland University, Saarbrucken, Germany
Contributors
xiii
G. Scot Hamilton • Department of Physiology and Pharmacology, University of Western Ontario, London, ON, Canada Daniel Hekl • Department of Therapeutic Radiology and Oncology, Innsbruck Medical University, Innsbruck, Austria Michael Heisig • Goethe-Center for Scientific Computing, Goethe-University, Frankfurt am Main, Germany Cristina Hernández • Diabetes and Metabolism Research Unit, Institut de Recerca Hospital Universitari Vall d’Hebron, Universitat Autònoma de Barcelona, Barcelona, Spain Guochang Hu • Departments of Anesthesiology and Pharmacology and Center for Lung and Vascular Biology, University of Illinois at Chicago, Chicago, IL, USA Peter Humphries • UCD School of Biomolecular Biomedical Sciences, Conway Institute, University College Dublin, Dublin, Ireland Arup Kumar Indra • Department of Pharmaceutical Sciences, College of Pharmacy, and Environmental Health Sciences Center, Oregon State University, Corvallis, OR, USA Mehmet Kaya • Departments of Physiology & Histology and Embryology, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey Ido P. Kema • Department of Laboratory Medicine, Division of Clinical Chemistry and Hematology, University Medical Center Groningen, Groningen, The Netherlands Ulla G. Knaus • Conway Institute, University College Dublin, Dublin, Ireland Yulia A. Komarova • Department of Pharmacology, University of Illinois, Chicago, IL, USA Isabel Larre • Center for Research & Advanced Studies (CINVESTAV), Mexico City, México Judith Lechner • Division of Physiology, Department of Physiology and Medical Physics, Division of Physiology, Innsbruck Medical University, Innsbruck, Austria Jisook Lee • Department of Surgery, University of California San Diego, San Diego, CA, USA Mark Leid • Department of Pharmaceutical Sciences, College of Pharmacy, and Environmental Health Sciences Center, Oregon State University, Corvallis, OR, USA Andreas Ludwig • Institute of Pharmacology and Toxicology, Faculty of Medicine, RWTH Aachen University, Aachen, Germany Brigitte Le Magueresse-Battistoni • Inserm U870, Oullins, Oullins, France; 2INRA, UMR1235, Oullins, France; 3INSA-Lyon, RMND, Villeurbanne, France Kai Yan Maka • A Women’s and Children’s Health Research Institute, North Adelaide, SA, Australia Asrar B. Malik • Department of Pharmacology, University of Illinois, Chicago, IL, USA Séverine Mazaud-Guittot • Inserm U870, Oullins, France, 2INRA, UMR1235, Oullins, France; 3INSA-Lyon, RMND, Villeurbanne, France Ana Melero • Biopharmaceutics and Pharmaceutical Technology, Saarland University, Saarbrucken, Germany
xiv
Contributors
Richard D. Minshall • Departments of Anesthesiology and Pharmacology and Center for Lung and Vascular Biology, University of Illinois at Chicago, Chicago, IL, USA Elizabeth Monaghan-Benson • Department of Cell and Developmental Biology and Lineberger Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA Dolores D. Mruk • Center for Biomedical Research, Population Council, New York, NY, USA Arne Naegel • Goethe-Center for Scientific Computing, Goethe-University, Frankfurt am Main, Germany Barry C. Powell • Women’s and Children’s Health Research Institute, North Adelaide, SA, Australia Barry Press • MaRS Centre, Ontario Institute for Cancer Research, Toronto, ON, Canada Peter Rubenwolf • Department of Paediatric Urology, Regensburg University Hospital, Klinik St. HedwigRegensburg, Germany Michael L. Salmans • Departments of Medicine and Biological Chemistry, University of California, Irvine, CA, USA Ulrich F. Schaefer • Biopharmaceutics and Pharmaceutical Technology, Saarland University, Saarbrucken, Germany Marc Schneider • Pharmaceutical Nanotechnology, Saarland University, Saarbrucken, Germany Thomas Seppi • Department of Therapeutic Radiology and Oncology, Innsbruck Medical University, Innsbruck, Austria Rafael Simó • Diabetes and Metabolism Research Unit, Institut de Recerca Hospital Universitari Vall d’Hebron, Universitat Autònoma de Barcelona and CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Instituto de Salud Carlos III (ISCIII), Barcelona, Spain Jacqueline Slinn • Cerebrovascular Research Group, Institute for Biological Sciences, National Research Council of Canada, Ottawa, ON, Canada Anselm Sommer • Institute of Pharmacology and Toxicology, Faculty of Medicine, RWTH Aachen University, Aachen, Germany Jennifer Southgate • Jack Birch Unit of Molecular Carcinogenesis, Department of Biology, University of York, York, UK Danica Stanimirovic • Cerebrovascular Research Group, Institute for Biological Sciences, National Research Council of Canada, Ottawa, ON, Canada Monika Strengert • Conway Institute, University College Dublin, Dublin, Ireland Yu Sun • Departments of Anesthesiology and Pharmacology and Center for Lung and Vascular Biology, University of Illinois at Chicago, Chicago, IL, USA Stefan Uhlig • Institute of Pharmacology and Toxicology, Faculty of Medicine, RWTH Aachen University, Aachen, Germany Marta Villarroel • Diabetes and Metabolism Research Unit, Institut de Recerca Hospital Universitari Vall d’Hebron, Universitat Autònoma de Barcelona and CIBER de Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM), Instituto de Salud Carlos III (ISCIII), Barcelona, Spain
Contributors
xv
Markus Voelp • Innovative Technologie-Voelp (IT-V), Innsbruck, Austria Stephen M. Vogel • Department of Pharmacology, University of Illinois, Chicago, IL, USA Xian Wang • Women’s and Children’s Health Research Institute, North Adelaide, SA, Australia Yuping Wang • Department of Obstetrics and Gynecology, Louisiana State University Health Sciences Center, Shreveport, LA, USA Alastair J. Watson • Gastroenterology Research Unit, School of Clinical Sciences, University of Liverpool, Liverpool, UK Sara M. Weis • Moores UCSD Cancer Center, University of California, San Diego, La Jolla, CA, USA Erika S. Wittchen • Department of Cell and Developmental Biology and Lineberger Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA Gabriel Wittum • Goethe-Center for Scientific Computing, Goethe-University, Frankfurt am Main, Germany
wwwwwwwwwwwww
Chapter 1 Computational Modeling of the Skin Barrier Arne Naegel, Michael Heisig, and Gabriel Wittum Abstract A simulation environment for the numerical calculation of permeation processes through human skin has been developed. In geometry models that represent the actual cell morphology of stratum corneum (SC) and deeper skin layers, the diffusive transport is simulated by a finite volume method. As reference elements for the corneocyte cells and lipid matrix, both three-dimensional tetrakaidecahedra and cuboids as well as two-dimensional brick-and-mortar models have been investigated. The central finding is that permeability and lag time of the different membranes can be represented in a closed form depending on model parameters and geometry. This allows a comparison of the models in terms of their barrier effectiveness at comparable cell sizes. The influence of the cell shape on the barrier properties has been numerically demonstrated and quantified. It is shown that tetrakaidecahedra in addition to an almost optimal surface-to-volume ratio also has a very favorable barrier-to-volume ratio. A simulation experiment was successfully validated with two representative test substances, the hydrophilic caffeine and the lipophilic flufenamic acid, which were applied in an aqueous vehicle with a constant dose. The input parameters for the simulation were determined in a companion study by experimental collaborators. Key words: Skin, Stratum corneum, Cell shape, Geometry models, Cuboid, Tetrakaidecahedra, Permeability, Lag time, Mathematical modeling, Numerical simulation, Homogenization, Drug diffusion, Concentration–depth profiles
1. Introduction The skin represents an area of about 2 m2. It is the largest organ of the human body and is used to separate the entire organism from the relatively cold and dry environment. From a pharmaceutical view, the skin is very attractive for the application of drugs, e.g., compared with the oral dosage form the first-pass effect in the liver is bypassed (1, 2). The human skin consists of three layers, epidermis (outer skin), dermis, and subcutaneous tissue (hypodermis). The outermost layer is the epidermis. This layer is avascular. The supply of
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_1, © Springer Science+Business Media, LLC 2011
1
2
A. Naegel et al.
nutrients is carried by the underlying dermis. In this layer, the hair follicles and sweat and sebaceous glands are anchored. The adipose tissue of the subcutis is the thermal barrier and provides a mechanical cushion. In the healthy epidermis, a continuous process of renewal proceeds, which lasts approximately 4 weeks in adults: the cells are formed at the border to the dermis in the stratum basale (basal layer), migrate through the stratum spinosum and the stratum granulosum to the outside and die, forming the stratum corneum (horny layer, SC). The SC is morphologically clearly distinguished from the rest of the epidermis, called the living epidermis. The thickness varies depending on the individual and the body part, but is typically 10–20 mm. Despite its low thickness, the SC represents the main barrier from the environment. The reason for this is the geometrically and chemically complex internal structure. With the death, the initially water-filled keratinocyte cells of the epidermis differentiate into dry, compact, keratin-filled cells called corneocytes. These cells are interwoven with a fibrous network of keratin filaments. Water can penetrate into the cells, but it is assumed that diffusion processes run in the corneocytes significantly slower than in an aqueous medium. The space between the cells is filled with a lipid matrix. This matrix has a bilayer structure that results from a parallel orientation of the head groups of the lipids. This laminar structure is the reason why diffusion across the bilayers is more difficult than along the head and tail groups. Because of this orientation of the head and tail groups, diffusion pathways for both lipophilic and hydrophilic molecules are available.
2. Modeling Section This section introduces three different geometry models for the SC. All have in common a prototype for a corneocyte cell that is embedded in a lipid matrix. The SC membrane is then obtained by agglomeration and allows to distinguish the corneocyte phase Wcor from the lipid phase Wlip. The parameterization of the cells is discussed for each geometry model individually, before a comparison of the models is provided. Finally, a literature survey on realistic dimensions is presented. 2.1. Model Geometries 2.1.1. Cuboid Model in Two Dimensions
A common approach to modeling of the SC is the use of geometries of two-dimensional cross sections, e.g., the reduction of a cuboid model in two dimensions in Fig. 1. For historical reasons, geometries of this type are usually referred to as brick-and-mortar models. Owing to the neglection of the third spatial dimension, it is, however, more precise, to speak of a reduced two-dimensional cuboid model or a ribbon-type model. Characteristic parameters are
1 Computational Modeling of the Skin Barrier
3
Fig. 1. Cuboid membrane (2D): Corneocyte cells in lipid matrix. Shown is a stack with N = 10 cell layers. Parameters: edge length w, height h, lipid channel thickness d, overlap w = scell/wcell.
the cell thickness h, the width w, and the lipid channel diameter d. Between two adjacent layers of cells, the horizontal overlap w plays an important role. The overlap is defined as the ratio w = scell/wcell. The shortest edge is defined as scell and the total edge length as wcell = w + d. 2.1.2. 3D Models Cuboid Model in Three Space Dimensions
Tetrakaidecahedron Model in Three Space Dimensions
A three-dimensional cuboid model is presented by Rim and coworkers (3, 4), Wagner (5) as well as Goodyer and Bunge (58). As shown in Fig. 2, the overlap of the cells in two directions can be varied. However, morphologically an identical overlap in both directions of the xy-plane is reasonable. Already in the nineteenth century, Lord Kelvin discovered that tetrakaidecahedra (TKD) provide a dense spatial packing with a nearly optimal balance between surface and volume (6). The use of these tetrakaidecahedra, shown in Fig. 3, as a basic element in the cell geometries of SC was suggested more than 30 years ago (8–10). The parameters for the characterization of the corneocytes are shown in Fig. 3. The cell thickness h is the height, and the quantities a and w control the planar extension of the cell. The edge length a defines the hexagonal shape of the (top or bottom) surface. The width w is the distance between two parallel edges of the two adjacent separated hexagonal faces. The values b and s are implicitly defined by the other three parameters. The addition of a surrounding lipid matrix of thickness d/2 results in a single basic element T. The arrangement in a cluster of three basic elements as in Fig. 4 leads to a base cell P, which allows
4
A. Naegel et al.
Fig. 2. Cuboid membrane (3D): Schematic top view of two adjacent cell layers (modified from ref. 4).
Fig. 3. Parameterization of a tetrakaidecahedron by height h, edge a and width w (redrawn from ref. 7).
a periodic continuation within the xy-plane. A stack of N layers of this type describes a membrane of finite thickness with infinite transverse extent. Using the parameters b and s, as defined in Fig. 3, the ratio
5
1 Computational Modeling of the Skin Barrier
a
Corneocyte C
Lipid matrix L
TKD base cell T
b Z
X
Y
Periodic cell ensemble P (single layer)
Fig. 4. Membrane of the TKD-type: (a) The basic TKD-element T consists of parameterized corneocytes C embedded in a lipid matrix L according to Fig. 3. The embedding is such that the respective side faces have the distance d/2. (b) Three basic TKD-elements T form a periodic base cell B. The simulation domain consists of a stack with N layers.
w = scell / bcell =
1 a 2 + 3 cell scell
<
1 2
provides a measure for the horizontal overlap. The quantities with subscript (acell, bcell, scell) refer to the cell T embedded in the lipid matrix, cf. Fig. 4 (7, 11). 2.1.3. Dimensions in the Literature
The dimensions of individual corneocytes were determined experimentally and are available in the literature (see below). These measurements are afflicted with inaccuracies. Hereafter, the used methods and results are discussed. In all cited papers water- induced swelling usually has only a small influence on cell expansion in a horizontal direction. The effect is mainly in the vertical z-direction. The experimental work of Richter et al. (12) studied the swelling of corneocytes in distilled water using tapping mode scanning force microscopy. Samples were removed at the inside of the forearm near the wrist. Only the topmost layers of cells were taken. For dried cells, the authors reported a diameter of 30–40 mm
6
A. Naegel et al.
at an average height of 0.2–0.3 mm. Cell surface and volume amount to approximately 1,000 mm2 or 300 mm3. By swelling the height and the volume grow by about 50%. In a subsequent study (13), the cell thickness was measured between 0.5 and 1.5 mm. Bouwstra et al. (14) conducted a study with samples from the abdominal area using cryo-scanning electron microscopy. The observed cell diameter was between 20 and 30 mm, the height varied between 0.3 mm for dried skin and 3.0 mm for hydration over distilled water. Samples from the upper arm and cheek area were used by Kashibuchi et al. (15). Surface sizes were in the range of 900–1,000 mm2. Depending on the body site the cell thickness can be as low as 0.15 mm, with a corresponding volume of 200 mm3. In this case, however, the corneocytes were first treated with xylene and then air-dried. Mihara (16) documented thicknesses of corneocytes between 0.5 and 2 mm and diameters in the range 30–45 mm. From the cell volume and the lipid content the dimension of the lipid matrix can be estimated. Based on the mass, it amounts approximately 20–30% for dried SC (17), or approximately 15% under natural hydration conditions (18). Correspondingly, the choice of parameters differs in previously published papers (Table 1). We have taken a cell width of w = 30 mm, and a cell thickness of h = 1.0 mm. The lipid channel thickness is fixed to d = 0.1 mm. Usually N = 10 (N = 16 in the experimental section) layers of the SC are assumed. A comprehensive survey of microscopic “brick-and-mortar” diffusion models of SC microstructure is also provided by Wang et al. (2006) (cf. Table 1). 2.1.4. Comparison of Geometric Models
The horizontal overlap w is consistently defined and compared for all three models. Nevertheless, this value provides only limited information about the actual effective path length and tortuosity because of the neglection of the three-dimensional structure. The shape of the cells in the SC is the result of a differentiation
Table 1 Choice of parameters published for brick-and-mortar models References
a (mm)
h (mm)
d (mm)
w
Heisig et al. (19)
30
1.0
0.1
Variable
Johnson et al. (20)
40
0.8
0.07
1/9
Barbero, Frasch (21)
44
3.5
0.1
12/49
Swollen
Wang et al. (22)
30 31.2
0.8 2.8
0.08 0.08
Variable Variable
partially swollen Swollen
Rim et al. (4)
40
0.8
0.08
1/2
In our 2D and 3D models, the parameterization from ref. 19 is used
Remarks
1 Computational Modeling of the Skin Barrier
7
process, for which only a limited amount of tissue is available. It is, therefore, equally important to consider the cell volume as well. The question arises whether barriers of cells with equal volume have equivalent barrier properties. A summary of the investigated model geometries is shown in Table 2. For a basic element, i.e., a single cell, the lipid volume Vlip, the corneocyte volume Vcor, and the corneocyte surface Acor is shown. In addition, the relative volume fractions are qlip = Vlip/Vcell and qcor = Vcor/Vcell, which can be calculated from the cell volume Vcell = Vlip + Vcor. 2.2. Model Equations
Transport processes of a substance a in a medium M are usually described in models by the spatial and temporal evolution of the concentration density cMa . Through the concept of density, it is suggested that a volume VM contains a amount of substance mMa which results from mMa = ò cMa dV . In this sense, the concentration VM is the amount of an imaginary, infinitesimally small volume, i.e., cMa =
dmMa dVM
(1)
The unit of concentration is éëcMa ùû = mg/ml. As described in the introduction, transport in the corneocytes takes place in a water-surrounded protein network. In this network, the carrier medium is made up of two phases: In the mobile liquid phase f, the actual transport takes place, while the immobile solid phase s is available for adsorption and reaction processes. For substance a, we must distinguish between the concentration in free form (c af , M ) and adsorbed form (csa, M ). Both are defined by Eq. 1, with volumes (and masses) restricted to Vf,M and Vs,M, respectively. Formally, two separate conservation equations are given:
¶ f ,M a (2) qM c f , M )+ Ñ·(-qMf , M D af , M Ñc af , M )= R af , M ( ¶t ¶ s,M a (3) (qM cs,M )= Rsa,M ¶t Accordingly, changes in concentration in a given test volume are only due to diffusive fluxes over the edge of the volume respectively through sources and sinks that are in this volume. D af , M denotes the diffusion coefficient in the fluid phase and R af , M , Rsa, M the reaction and source terms. The units are [ D af , M ] = cm2/s, and [R.a,M] = mg/ml/s. The relative volume fractions are fixed in space and time, i.e., qMf , M :=
V f ,M VM
and qsM, M :=
Vs , M VM
30
30 25 20 40a
30 30 30 30
Ribbon (2D)
Cuboid (3D)
TKD (3D)
5 8 10 13
a (mm) 9.39 9.69 9.81 9.99 8.91 18.67 15.46 13.5 10.83
96.61 68.01 44.41 125.25 81,18 76.23 72.10 64.86
qlip (%)
–
Vlip (mm3)
a
Geometry from ref. 4 with h = 0.8 mm and d = 0.075 mm. All other cases use h = 1 mm and d = 0.1 mm
w (mm)
Geometry
1,517.45 1,450.82 1,389.69 1,276.49 81.33 84.54 86.50 89.17
353.63 416.85 461.80 533.76
1,920.00 1,350.00 880.00 3,328.00 90.31 90.19 90.01 91.07 900.00 625.00 400.00 1,280.00
Acor (mm2) –
qcor (%) 90.61
Vcor (mm3)
Table 2 Comparison of the model geometries. Given are the lipid and the corneocyte volume Vlip, Vcor, the relative volume fractions qlip, qcor and the corneocyte surface area Acor for a single corneocyte cell, embedded in the lipid matrix
8 A. Naegel et al.
9
1 Computational Modeling of the Skin Barrier
Obviously, the identity cMa = qMf , M c af , M + qsM, M csa, M
holds so that in cases where csa, M is explicitly given, e.g., as a function of c af , M , the conservation equation for substance a can be written as
¶ f ,M a q M c f , M + q sM, M csa, M + Ñ· -q Mf , M D af , M Ñc af , M = R Ma ¶t
(
) (
)
(4)
where R Ma summarizes the reaction terms from Eqs. 2 and 3. 2.2.1. Transmission Conditions
The description is so far made for a single medium M. The skin membranes described in the introduction, however, distinguish clearly between lipids and corneocytes. In general, the simulation domain W consists of nonoverlapping subdomains Mi, i = 1,…k, i.e.,
W = Èik=1 M i
with M i Ç M j = Æ for i ¹ j.
In addition to boundary conditions on the outer boundary ¶W, additionally boundary conditions on the inner boundaries, i.e., the interfaces Gij := ¶M i Ç ¶M j between Mi and Mj, are prescribed. This coupling between two subdomains is guaranteed by transmission conditions. In addition to the conservation of flux f ,M q Mf ,iM i D af , M i Ñc af , M i - q M j j D af , M j Ñc af , M j ·n = 0 on G ij
a discontinuity in the form of the Nernst distribution law is allowed: c af , M i = K ija c af , M j (5)
(
)
coefficient of substance a Here K ija > 0 denotes the partition -1 between Mi and Mj with K aji = (K ija ) . Reformulation for Constant Coefficients
In the most common situations, the partition coefficients K ija are constants. In this case, the transformation
c af , M i = qaM i c fa, M i , csa, M i = qaM i csa, M i with constants qaM i > 0 yields, that the discontinuity in Eq. 5 is replaced by the continuity condition c f , M i = c f , M j , i ¹ j . To this end, we choose, for example, qaM 0 := 1 and qaM i := K ia0 for i > 0. If we accordingly define
f ,q
s ,q
qMf ,iM i , a := qaM i qM i Mi
and qMs , Mi i , a := qaM i qM i Mi
Equation 4 is restated in the following form:
¶ f ,M i ,a a qM c f + qMs , Mi i , a csa, M i + Ñ · -qMf ,iM i D af , M i Ñc fa, M i = R Ma ¶t
(
)
(
)
(6)
10
A. Naegel et al.
2.2.2. Flux, Permeability, and Lag Time
Let us now state a description of a setup for an experimental diffusion cell under infinite dose conditions. Neglecting contributions of adsorption and reaction, we seek a solution c–, which satisfies ¶ ( qc ) + Ñ·(-qDÑc ) = 0 in W (7) ¶t with the initial value c = 0 and constant boundary values in time and space: cG+ = c + , cG- = c - (8) Let us further assume that the boundary ∂W allows to identify two disjoint surfaces G+ and G−. These must be equal in size and should specify areas for inflow and outflow, respectively. The remaining boundary ¶W\(G+ È G−) should allow a periodic identification. Flux, permeability, and lag time are defined as follows:
Flux
For a membrane M = (W, G) and the test problem Eqs. 7 and 8, the term 1 J M (t ) := -qDÑc ·n dx (9) ò Gò dx G-
denotes the flux through the membrane. Here, n is the outward normal. By integration in time, the (released) mass results in
t
m M (t ) := ò J M ( s )ds
0
(10)
Both quantities are defined per unit area. They have the physical units [mg/cm2/s] and [mg/cm2]. The flux in a steady state is denoted by J M := lim J M (t ) . t ®¥
Permeability
Permeability is defined as the constant of proportionality between the flux JM and the concentration difference (c + - c - ) adjacent to the membrane J M = PM (c + - c - ) The unit is that of a velocity: [PM] = cm/s.
Lag Time
2.2.3. Model Membrane with Constant Coefficients
The lag time TM of a membrane for constant boundary values in the absence of sources and sinks is given by æ J t - m M (t ) ö TM = lim ç M ÷ t ®¥ JM è ø In the geometries introduced in Subheading 2.1 subdomains for lipids and for corneocytes are distinguished. For these membranes, piecewise constant diffusion coefficients Dlip and
1 Computational Modeling of the Skin Barrier
11
Dcor := eDlip with e > 0 and a constant partition coefficient Kcor/ := K > 0 are assumed. Equation 7 is equivalent to the piecewise lip formulation
¶ clip + Ñ·(- Dlip Ñclip ) = 0 ¶t K
¶ ccor + Ñ·( -eKDcor Ñccor ) = 0 ¶t
(11) (12)
where ccor = K cor /lip ccor and clip = clip denote the actual concentrations. The coupling between time and length scales is expressed by the diffusion coefficient Dlip. Since this can be chosen arbitrarily, the case Kcor/lip = 1, e = 1, Dlip = constant. is regarded as the (homogeneous) reference configuration. By defintion JSC (and consequently also PSC) is linear in Dlip and furthermore depends on a dimensionless quantity only: x: = eK
(13)
Analogously, the relative lag time TSC is represented as a function of K and x. For reasons of greater convenience, we define the relative permeability
aSC :=
PSC J SC = (hom) (hom) PSC J SC
and the relative lag time
tSC :=
TSC TSC(hom)
(14)
(hom) , TSC(hom) > 0 , which denote permeability, flux, relative to PSC(hom) , J SC and lag time for the homogeneous reference configuration, respectively. Note that the latter quantities are often found easily. For the rectangular membrane (with thickness hSC), the homogeneous case corresponds, e.g., to a one-dimensional problem, which yields
(hom) J SC = Dlip Dclip / hSC
PSC(hom) = Dlip / hSC 2 TSC(hom) = hSC / (6 Dlip )
2.2.4. Numerical Methods
1. Equation 7 is solved numerically in an appropriate software environment. In addition to commercially available tools, such as COMSOL (23), several solutions from the academic sector exist as well. Our method of choice is UG (24, 25). 2. Following Rothe’s method for the spatiotemporal discretization, the time variable is discretized first. Owing to the parabolic character of the equation, an implicit time-stepping
12
A. Naegel et al.
scheme should be preferred. The implicit Euler scheme uses the approximation
c (t + t , x ) - c (t , x ) + Ñ. - qD Ñ c (t + t , x ) = 0 t to obtain the solution c– at time t + t given a positive step size t > 0. More complicated schemes can be used as well: The Crank-Nicolson scheme, e.g., provides a higher order approximation w.r.t. t. The fractional-step-q-scheme is additionally also stable with respect to oscillations.
(
)
3. The resulting system is semidiscrete, i.e., discrete with respect to time, but still continuous with respect to space. In order to exploit local conservation of mass, the spatial discretization is preferably performed using a vertex-centered finitevolume(FV)-scheme (26–28). The main idea can be summarized as follows: (a) The computational domain is approximated by a suitable grid. Supported element types are triangles and quadrilaterals (2D) and tetrahedra, pyramids, prisms, and hexahedra (3D). For each vertex xi in the grid, subsequently, a control volume Bi is constructed using a dual mesh, cf. Fig. 5. (b) The solution is then sought to be a continuous function, ch (t , x ) := S{iŒV (Wh )} c hi (t ) jhi (x ), which is defined by the basis functions jhi (x ) . These are (multi-) linear on each element i and satisfy jh (x j ) = d ij , which means the value is 1 at x = xi
Bi
xi
Fig. 5. Example of a grid consisting of triangles and quadrilaterals. The control volume box Bi for one vertex xi is shown in gray. Dotted lines indicate the dual mesh.
1 Computational Modeling of the Skin Barrier
13
and 0 at vertices x = xj where j ¹ i. On each box Bi we approximate the following local conservation property: ò ch (t + t , x) - ch (t , x) dx + t ò -qD (x)— ch (t + t , x) n ds = 0 (15)
(
)
¶Bi
Bi
4. Quadrature (integration) methods for the evaluation of fluxes and masses can readily be implemented using standard FV components. 2.2.5. Homogenization
The fine-scale model in Eqs. 11 and 12 distinguishes between corneocytes and lipid channel. In some cases, however, it is desirable to neglect structural information and to consider a homogenized membrane only. Mathematical techniques for such a homogenization are given by the method of asymptotic expansion (29). Applications to SC membranes have been given for brick-and-mortar models (30), for cuboid models (3, 4), and for the TKD model (31). Instead of a fine-scale model (Eqs. 11 and 12), a coarse scale model is considered: ¶ ˆ (qSC cˆSC ) + Ñ·éë - Dˆ SC ÑcˆSC ùû = 0 ¶t The hats now indicate averaged versions of the corresponding quantities. In addition to the effective mass coefficient
cor lip qˆ SC := qSC K cor /lip + qSC
this formula contains an effective diffusion tensor Dˆ SC . As different directions must be reflected Dˆ SC must be expressed as a matrix, e.g., Dˆ SC
æ DSC, x ç =ç 0 ç 0 è
0 DSC, y 0
0 ö ÷ 0 ÷ DSC, z ÷ø
for the three-dimensional models (in appropriate coordinates), with DSC,x = DSC,y due to the symmetry of the problem. 2.3. Related Work in the Literature
In this section, we give an overview of related publications on diffusion models listed chronologically (see Table 3). A review also covering QSPR and compartment models was published recently by Mitragotri et al. (59). One of the first mathematical models of the SC is from Michaels et al. (32). They described the SC as a two-phase protein–lipid heterogeneous membrane (in which the lipid phase is continuous), which correlates the permeability of the membrane to a specific penetrant with the water solubility of the penetrant
Dimension
1D
1D
1D
1D
1D
2D
1D
1D
1D/2D
1D
2D
1D
2D
References
Michaels et al. (32)
Albery and Hadgraft (33)
Gienger et al. (34)
Tojo (35)
Edwards and Langer (36)
Heisig et al. (19)
Lee et al. (37)
Johnson et al. (20)
Manitz et al. (38)
Anissimov and Roberts (39)
Charalambopoulou et al. (40)
Anissimov and Roberts (41)
Frasch (42)
FV
Analytical
Analytical
FEM
LT
Analytical
Method
SC
Vehicle/skin
SC
Vehicle/skin
Vehicle/skin
SC
Random walk
LT
Neutron scattering
LT
FD
Analytical
SC/epidermis Numerical
SC
SC
SC
Vehicle/skin
Epidermis
SC
Part of skin
Steady state
Transient
Steady state
Transient
Transient
Steady state
Transient
Steady state
Transient
Transient
Steady state
+
Deff, leff
Finite dose
Deff
Flow rate
Concentration profiles
PSC
Concentration profiles
PSC, tlag
Transport coeffecient
PSC
+
+
+
+
+
+
+
+
+
+
Drug transport +
−
−
−
−
−
+
−
−
−
−
+
−
+
Theory Experiment
Drug transport +
Permeability
Time dependence Focus
Table 3 Overview of published diffusion models of the skin barrier listed chronologically
Comparison with Flynn data
Concentration– depth profiles
Variable geometry
Constant donor concentration
Two-component model
Bilayer-scale transport
Two pathways
Variable Dcor/lip, Kcor/lip, w
Comparison with experiment
Three pathways
Time-variable D, K
Interfacial barriers
2-Phase model
Remarks
14 A. Naegel et al.
Dimension
1D
1D
1D/2D
1D
2D
2D
2D
2D
3D
2D
2D
1D
2D/3D
2D
References
Kubota et al. (43)
Mitragotri (44)
Frasch and Barbero (45)
Anissimov and Roberts (46)
George et al. (47)
George (48)
Rim et al. (49)
Frasch and Barbero (50)
Feuchter et al. (7)
Barbero and Frasch (21)
Wang et al. (22)
Mollee and Bracken (51)
Rim et al. (3)
Chen et al. (52)
SC
SC
SC
SC
SC
SC
SC
Vehicle/skin
Full skin
Full skin
SC
SC
SC
Vehicle/skin
Part of skin
Transient
Transient
Transient
Transient
Transient
Transient
Transient
Transient
Steady state
Steady state
Transient
FD
Transient
Concentration profiles
Homogenization Deff
JSS, DSC
+
+
+
+
+
PSC, tlag PSC, DSC
+
+
+
+
+
+
+
+
+
Transcellular pathway
Variable Dcor/lip, Kcor/lip, w
Variable D and K
Multi-component diffusion
Dual-sorption; A(t), J(t)
A(t), J(t)
Variable D and K
Lipid pathway; Deff, leff
Pfv, Plateral, Ppore, Pshunt
Repeated application
Remarks
Cells impermeable
Capture and release
(continued)
− (1 Comparison with component) experiment
−
−
− (4 Corneocyte components) transport
− (Database)
−
−
+
−
−
−
−
−
−
Theory Experiment
TKD model
PSC, tlag
Finite dose
Concentration profiles
Concentration profiles
Concentration profiles
JSS, tlag
Four pathways
A(t), J(t)
Time dependence Focus
Analytical, LT Transient
FD
FEM
FV
FEM
FEM
FD
FD
LT
FEM
Analytical
FD
Method
1 Computational Modeling of the Skin Barrier 15
2D/3D
2D
1D
3D
2D/3D
3D
Rim et al. (4)
Naegel et al. (53)
Anissimov and Roberts (55)
Rim et al. (56)
Naegel et al. (57)
Muha et al. (31)
SC
Vehicle/skin
Vehicle/skin
SC
Vehicle/skin
SC
Part of skin
FV
MD
LT
FV
Method
Slow binding
Concentration profiles
PSC, DSC
Homogenization Deff
Transient
Homogenization Deff
Transient
Transient
Homogenization Deff
Time dependence Focus
+
+
+
+
+
+
−
−
+ (Fentanyl)
TKD model
Comparison of models
Multiscale modeling
Desorption kinetics
Comparison with experiment
- (54) +
Cuboid cells permeable
Remarks
−
Theory Experiment
TKD tetrakaidecahedron, w overlap, LT Laplace transform, FD finite difference method, FEM finite element method, FV finite volume method, MD molecular dynamics, PSC permeability coefficient of the SC, Pfv contribution of free volume diffusion to overall skin permeability, Plateral contribution of lateral lipid diffusion to overall skin permeability, Ppore contribution of pores to overall skin permeability, Pshunt contribution of shunts to overall skin permeability, DSC diffusion coefficient of the SC, Dlip diffusion coefficient of the lipids, Dcor diffusion coefficient of the corneocytes, Kcor/lip corneocyte–lipid partition coefficient, tlag lag time, Deff effective diffusivity, leff effective pathlength, A(t) amounts of drug per unit area at time t, J(t) Flux of drug per unit area at time t, JSS flux of drug in steady state
Dimension
References
Table 3 (continued)
16 A. Naegel et al.
1 Computational Modeling of the Skin Barrier
17
and with its lipid–protein partition coefficient. Several years later, Albery and Hadgraft (33) derived equations to describe the percutaneous absorption of a substance through the epidermal barrier. The treatment includes interfacial barriers and allows for the depletion of the substance in the external phase. Both the drug diffusion through the keratinized cells (transcellular route) and the diffusion in the interstitial channels around the cells (intercellular route) were considered. Gienger et al. (34) modeled the drug transport from a transdermal patch across the skin. They used a commercial software package (DISPL) to obtain a numerical solution for this model, including time variable diffusion coefficients and interface conditions in case of nonhomogeneous media. In their model, they assumed a drug reservoir (patch) placed on the skin surface and a linear concentration gradient of the drug has been assumed inside the skin layer. Tojo (35) developed a model for drug permeation across the SC to predict the effect of the physical configuration and chemical composition of the SC on the skin permeability of drugs. Several years later, Edwards and Langer (36) developed a theory of charge, fluidmass, and solute (including macromolecular) transport through porous media and applied it to describe transport phenomena across the external layer of mammalian skin. In our first publication on this topic in Heisig et al. (19), we developed a numerical method for a brick-and-mortar SC-geometry, enabling a numerical solution for time-dependent drug concentration within corneocyte and lipid phases. We calculated the lag time and permeability and demonstrated how the barrier property of this model membrane depends on relative phase permeability, corneocyte alignment, and corneocyte–lipid partition coefficient. The calculations suggested that beside the intercellular pathway the corneocyte-phase transport plays a major role. In a further study, Lee et al. (37) presented a mathematical model of percutaneous absorption, which considers the simultaneous penetration of drug by transcellular and intercellular pathways, as well as movement of drug between these two pathways. Johnson et al. (20) developed a mathematical model to describe the macroscopic SC permeation via the interkeratinocyte lipid domain in terms of the structure and dimensions of the SC, and the microscale lipid bilayer transport properties, which include the bilayer–water partition coefficient, the lateral diffusion coefficient, the interfacial transbilayer mass transfer coefficient, and the intramembrane transbilayer mass transfer coefficient. Manitz et al. (38) considered several penetrating substances formulated within a vehicle for modeling the case of an applied drug and some penetration modifiers (enhancers and reducers, respectively). A coupling via concentration-dependent diffusivities between the diffusion equations of the involved substances was used to model the dependencies between them. The model predicts the concentration profile
18
A. Naegel et al.
depending on time on a two-dimensional multilayered domain representing a cross section through human skin. Anissimov and Roberts (39) developed a diffusion model for the percutaneous absorption of a solute through the skin for the specific case of a constant donor concentration with a finite removal rate from the receptor due to either perfusion rate or sampling. The model was developed to include a viable epidermal resistance and a donor−SC interfacial resistance. Charalambopoulou et al. (40) quantitatively determined the most appropriate geometry of porcine SC’s lipid and protein phases in a “brick-and-mortar” configuration and correlated it with the barrier properties (diffusivity) of the SC model structures. Anissimov and Roberts (41) developed a diffusion model for percutaneous absorption for the specific case of delivery to the skin being limited by the application of a finite amount of solute. Frasch (42) modeled diffusion as a two-dimensional random walk through the biphasic (lipid and corneocyte) SC. This approach permitted calculations of diffusion phenomena in a morphologically realistic SC structure. The model has been provided insight into the contributions of SC diffusivity and effective path lengths to overall skin permeability. Kubota et al. (43) developed a mathematical model for percutaneous absorption with regular applications of the drug. The linear partial differential equations (PDEs) of the model were solved using a finitedifference method that is second-order accurate in space. The solutions of these PDEs gave the concentrations of the drug in the vehicle and the skin at a given time. Mitragotri (44) compiled fundamentally based analytical expressions that can be used to predict skin permeability to hydrophilic as well as hydrophobic solutes. Solute permeation through four possible routes in SC including free-volume diffusion through lipid bilayers, lateral diffusion along lipid bilayers, diffusion through pores, and diffusion through shunts was analyzed. The model yielded a series of equations to predict skin permeability based on solute radius and octanol–water partition coefficient. Frasch and Barbero (45) presented finite element model (FEM) solutions of the diffusion through two-dimensional representations of the SC lipid pathway. It was assumed that diffusion occurs only within the SC lipids and the lipids are isotropic. The steady-state flux and lag time are solved and compared with the corresponding values for a homogeneous membrane of the same thickness consisting of lipid material. By SC desorption experiments Anissimov and Roberts (46) calculated steady-state fluxes that were larger than those obtained by epidermal penetration studies. A possible explanation of this result is a variable diffusion or partition coefficient. They, therefore, developed the diffusion model for percutaneous penetration and desorption to study the effects of either a variable diffusion coefficient or variable partition coefficient in the SC over the diffusion path length. George et al. (47) developed a two-dimensional mathematical model for percutaneous absorption of a drug, which
1 Computational Modeling of the Skin Barrier
19
may be used when the diffusion of the drug in the direction parallel to the skin surface must be examined, as well as in the direction into the skin. Further, George (48) developed a two-dimensional dual-sorption model for percutaneous absorption of a drug, which shows nonlinear kinetic behavior in the permeation process. Rim et al. (49) developed a finite element method to simulate two-dimensional (axisymmetric) drug diffusion from a finite drug reservoir into the skin. The numerical formulation is based on a general mathematical model for multicomponent nonlinear diffusion which takes into account the coupling effects between the different components. The implemented finite element framework is suitable for modeling both linear and nonlinear diffusions, single- and multicomponent diffusions, in heterogeneous media where the diffusivities and partition coefficients may vary in each subregion. Barbero and Frasch (50) used a finite element method to model diffusion in the skin’s outermost layer, the SC. The SC is considered to be a finite two-dimensional composite having different diffusivity values in each medium as well as a partition coefficient at the interfaces between media. Feuchter et al. (7) presented the first three-dimensional geometry model with tetrakaidecahedra as basic units (corneocytes, lipid matrix) for the SC membrane. By means of this concept the nonsteady-state problem of drug diffusion within a biphasic model SC-membrane having homogeneous lipid and corneocyte phases were solved numerically by a multigrid method. In a further study Barbero and Frasch (21) compared experimental measurements of permeability and lag time with the predictions of a finite element model. A database of permeability and lag time measurements of hydrophilic compounds was compiled from the literature. Transcellular and lateral lipid diffusion pathways were modeled within a brickand-mortar geometry representing fully hydrated human SC. Wang et al. (22) presented a two-dimensional microscopic transport model of the SC incorporating corneocytes of varying hydration and permeability embedded in an anisotropic lipid matrix. Results were expressed in terms of a dimensionless permeability. Model calculations were exemplified by characterizing the skin permeability of four representative permeants: water, ethanol, nicotinamide, and testosterone. The calculations again confirmed that corneocyte-phase transport plays a major role for all four permeants. Mollee and Bracken (51) presented a one-dimensional model of solute transport through the SC. Solute is assumed to diffuse through lipid bi-layers surrounding impermeable corneocytes. Transverse diffusion (perpendicular to the skin surface) through lipids separating adjacent corneocytes is modeled in the usual way. Longitudinal diffusion (parallel to the skin surface) through lipids between corneocyte layers is modeled as temporary trapping of solute, with subsequent release in the transverse direction. Rim et al. (3) obtained the macroscopic diffusion coefficients of the SC through homogenization of the diffusion equation
20
A. Naegel et al.
using the method of asymptotic expansion. Homogenization allowed the calculation of the 3D effective diffusivity tensor. Assuming the corneocytes to be impermeable, they studied the effects that different geometric arrangements of the corneocytes within the lipid matrix have on the continuum diffusion coefficients. Chen et al. (52) presented a computer model for predicting transdermal permeation of solutes in the heterogeneous SC including both the tortuous lipid pathway and the transcellular corneocytes pathway. The partition and diffusion properties of solutes in SC lipid matrix and corneocytes were calculated from the fundamental physical chemical properties of octanol–water partition coefficient, molecular size, and diffusion coefficients in water and lipid, using equations established elsewhere. In a further paper, Rim et al. (4) used the method of homogenization to obtain the macroscopic diffusion tensor of the SC allowing the corneocytes to be permeable and considering the partitioning between the corneocytes and the lipid phases. Naegel et al. (53) presented a mathematical model of drug permeation through SC and viable epidermis/dermis. The underlying geometry for the SC was of brick-and-mortar character. All phases were modeled with homogeneous diffusivity. The partition coefficients and diffusion coefficients were determined experimentally or derived consistently with the model. The corneocyte diffusivity was estimated based on an approximation, which used the apparent SC- and lipid-diffusion coefficients as well as corneocyte–lipid partition coefficients. The quality of the model was evaluated by a comparison to experimentally obtained concentration–SC-depth profiles. Anissimov and Roberts (55) studied the desorption of water from SC and obtained profiles for amount desorbed versus time profiles, which were more consistent with water transport occurring in a heterogeneous membrane. Analysis of the resulting profiles yielded a model which was consistent with a slow equilibration/ slow binding of water within SC as well as its permeation through the SC. Diffusion model solutions were used to derive the steadystate flux, lag time, and mean desorption time for water in SC. Rim et al. (56) proposed a multiscale framework of modeling the multicomponent transdermal diffusion of molecules. The problem was divided into subproblems of increasing length scale, i.e., microscopic, mesoscopic, and macroscopic. First, the microscopic diffusion coefficient in the lipid bilayers of the SC was found through molecular dynamics (MD) simulations. Then, a homogenization procedure was performed over a model unit cell of the heterogeneous SC, resulting in effective diffusion parameters. Naegel et al. (57) presented a diffusion model to compare the permeability of the SC membrane of three different geometry concepts, i.e., ribbon, cuboid, and tetrakaidecahedral type, in two and three space dimensions. The results confirmed that tetrakaidecahedral cells with an almost optimal surface-to-volume ratio
1 Computational Modeling of the Skin Barrier
21
provide a barrier, in which a minimal amount of mass is used very effectively. Muha et al. (31) showed how the method of asymptotic expansion can be used to homogenize membranes consisting of tetrakaidecahedral-shaped cells and calculate the effective diffusivity. Furthermore, they confirmed by numerical results that the resulting tensor is of diagonal shape, and the transversal and lateral diffusivity can be described uniformly with different coefficients.
3. Results and Discussion The mathematical and geometric models from the previous section are now further investigated. Descriptions of the permeability and the lag time for the different geometry models are presented in Subheadings 3.1 and 3.2. In Subheading 3.3, a virtual diffusion cell is presented which allows a comparison with an in vitro permeation experiment. 3.1. Approximation of the Permeability
The relative permeability of the investigated SC membranes depends only on the dimensionless quantity x = eK defined in Eq. 13. If we use the approach
aSC (x) » a
(lip) SC
+
(a (a
(¥) SC
(lip) - aSC )(1 - aSC(lip) )x
(¥) SC
(lip) - 1)+ (1 - aSC )x
(16)
we can make very good predictions. The constants on the righthand side (lip) (¥) 0 < aSC £ 1 £ aSC describe two important limit cases: ●●
●●
(lip) aSC is the relative permeability in the case of impermeable corneocytes, i.e., drug diffusion occurs only in the lipid channel. (¥) aSC is the relative permeability in the idealized case in which the drug diffusion in the corneocytes is infinitely fast.
Both quantities depend on the used geometry only. 3.1.1. Comparison of Geometric Models
For all geometry models introduced in Subheading 2.1, the permeability can be described by Eq. 16. Corresponding comparisons for the reduced cuboid model (2D), the full cuboid model (3D), and the TKD model (3D) are shown in Figs. 6–8. Each model is shown with different overlaps 0 £ w £ 1/2. The results are based on the standard parameterization described in Subheading 2.1.3. Qualitatively, similar results are obtained for other edge lengths, heights, and cell thicknesses representative for the SC.
22
A. Naegel et al.
1
10
0
Rel. Permeability αSC
10
−1
10
10−2
10−3
Overlap ω = 1/2 ω = 1/4 ω = 1/8 ω=0
10−4
10–10
10–5
100 ζ=αcor Kcor/lip
105
1010
Rel. permeability αSC
Fig. 6. Relative permeability aSC for the reduced rectangular membrane (2D) depending on the model parameter x = eK and the horizontal overlap w.
10
1
10
0
10
–1
10
–2
10
–3
10
–4
Overlap ω =1/2 ω =1/4 ω =1/8 ω =1/16 ω =1/32
10
–10
–5
10
0
10
5
10
10
10
ζ=αcor Kcor/lip Fig. 7. Relative permeability aSC for the cuboid membrane (3D) depending on the model parameter x = eK and the horizontal overlap w.
3.1.2. Barrier Effectiveness of Membrane
To evaluate the effectiveness of the membrane three different criteria are used: the degree of horizontal overlap, the volume of the cells, i.e., the amount of necessary tissue, and the influence of the number of cell layers are investigated.
Rel. Permeability αSC
1 Computational Modeling of the Skin Barrier
10
1
10
0
10
23
−1
10−2
10
−3
10
−4
Side length a = 8, ω = 0.27 a = 10, ω = 0.21 a = 13, ω = 0.09 10−10
10−5
100
105
1010
ζ=αcor Kcor/lip Fig. 8. Relative permeability aSC for the TKD-membrane (3D) depending on the model parameter x = eK and the horizontal overlap w resp. edge length a. 0.0045 Ribbon (2D)
Relative permeabiliyy αSC,0
0.004
Cuboid (3D) TKD (3D)
0.0035 0.003 0.0025 0.002 0.0015 0.001 0.0005 0
0
0.1
0.2
0.3
0.4
0.5
Horizontal overlap ω
Fig. 9. Relative permeability aSC as a function of the horizontal overlap w. (lip)
Horizontal Cell Overlap
To summarize the previous results, the minimal relative permea(lip) bilities aSC of the different geometry models of SC are compared, cf. Fig. 9. Already an overlap of about 20% yields a relatively small permeability value for all geometry models. It is important
24
A. Naegel et al.
that for small overlaps the TKD model is similar to the rectangular model (3D). With increasing overlap, the TKD model approaches the reduced cuboid model (2D). From a biological perspective, it is relevant how a barrier with low permeability is obtained with a limited amount of tissue. Therefore, the impact of the cell volume on the permeability must be taken into account. Hence, the three-dimensional TKD model is compared with the three-dimensional cuboid model. For the latter edge lengths of w = 20, 25, 30 mm and a full horizontal overlap w = 0.5 are used. In the study also the rectangular geometry of Rim et al. (4) (w = 40 mm, h = 0.8 mm, d = 0.075 mm) is considered. The graphical representation is shown in Fig. 10. As one might expect, TKD membranes have a significantly larger barrier-tovolume ratio than geometries with cuboids. The TKD geometry with a = 8 mm (Vcor = 416.85 mm3) is almost two times less permeable than the rectangular geometry with w = 30 mm (Vcor = 900 mm3). With a comparable cell volume the TKD geometry is almost three times less permeable than the cuboid geometry (w = 20 mm, Vcor = 400 mm3, w = 25 mm, Vcor = 625 mm3). Comparable permeabilities are obtained for the parameters (w = 40 mm, Vcor = 1,280 mm3) used by Rim et al. (4), the cell volume in this case is approximately three times larger (see Table 2). It is, in particular, interesting that in Fig. 10 the permeability decreases with decreasing TKD cell volume.
Cell Volume
1 Cuboid
0.6
3
Vcor = 416.85 µm
3
Vcor = 353.16 µm
a=13
a=10
a=8
a=5
Vcor = 461.88 µm
3
3
w=30 w=40*
Vcor = 533.76 µm
Vcor = 1280 µm
w=25
3
w=20
Vcor = 900 µm
3
0
Vcor = 625 µm
0.2
3
3
0.4
Vcor = 400 µm
Permeability PSC,0 [10−4 cm/h]
TKD 0.8
Geometry parameter [µm] (lip) Fig. 10. Minimum permeability PSC for three-dimensional geometric models at Dlip = 10−8 cm2/s. The cell volume for the geometries with cell diameter w and side length a is from Table 2.
1 Computational Modeling of the Skin Barrier
25
Fig. 11. Relative permeability aSC for cuboid geometries (3D) and TKD geometries (3D) dependent on the layer number N of the membrane. The lines mark the limit for the homogenized problem.
Number of Layers
The barrier properties of the membrane change significantly with the number of layers. To examine this more closely, the relative permeability aSC for different model membranes is evaluated depending on the layer number N. The base cells of rectangular and TKD-type are considered (see Fig. 11). Qualitatively, both graphs show the same behavior: For small N aSC changes relatively strongly, for N >> 1 aSC is constant. These values correspond to the results for cuboid membranes, which were calculated by homogenization (see Subheading 2.2.5). The same holds for the TKD models (31).
3.2. Approximation of the Lag Time
Theoretical consideration for the lag time: For the defined model membrane in Subheading 2.2.3, the following conclusions hold: (a) If the parameters K = x = 1 are used, the result is a homoge(hom) neous membrane with lag time TSC . (b) For 1 << K the lag time TSC is maximal and unbounded w.r.t. K. Intuitively, this corresponds to an infinitely large sponge in the corneocytes. The smaller the x and also the diffusion constant e, the larger is TSC. (c) Conversely, for K << 1 << x the lag time is minimal. However, since the system is limited by the diffusion through the lipids, the corresponding value TSC(min) > 0 differs from zero for the considered geometric models.
26
A. Naegel et al.
Table 4 Limit cases in the description of the lag time TSC, shown as a function of the model parameters K and x T (K, x)
Stationary case (K → 0)
Slowly x ® 0
(e) Impermeable corneocytes T(lip)
Identical x = 1
(d) Harmonic extension in corneocytes
Fast x ® ¥
(c) Minimal lag time T(min) > 0
Equal time scales (K = 1)
Maximal retardation (K → •) (b) M aximal sponge effect, T ® ¥
(a) Homogeneous membrane with T(hom) = h2/6Dlip
(d) For K << 1 and x = 1, we obtain a model with decoupled scales for K ® 0. In the corneocytes, an equilibrium adjusts infinitely fast. Therefore, the concentration values are continuous at the corneocyte boundaries. (e) Additionally, if x ® 0, we obtain a membrane with Neumann boundary conditions at the interface between the corneocytes and the lipids. Formally, this corresponds to a membrane with impermeable corneocytes and with the lag time TSC(lip) For a better illustration, these observations are summarized in Table 4. 3.2.1. Quantitative Description
For the homogenized equations of Subheading 2.2.5 the lag time
2 cor lip TSC = (qSC K + qSC )6hDSC SC
is derived. Even for membranes for which the homogenization condition (hcell << hSC) is not satisfied,
TSC (x, K ) »
(T1 K + T2 ) aSC (x)
(17)
constitutes an useful approach. The parameters T1, T2 are eometry-dependent time constants, which are determined by g the theoretical considerations from above. Owing to (a) and (e) we obtain TSC
(T (x, K ) »
(hom) SC
(lip) - TSC(lip) aSC )K + TSC(lip) aSC(lip)
aSC (x)
(18)
1 Computational Modeling of the Skin Barrier
27
or, equivalently, for tSC as defined in Eq. 14:
(lip) (lip) (lip) (lip) tSC (x, K )aSC (x) » (1 - tSC aSC )K + (tSC aSC )
The results in Fig. 12 show that this model is applicable for the rectangle (a) and the cuboid (b) membranes. In each case the ten-layer standard model with full overlap w = 1/2 was used.
Fig. 12. Relative Lag time tSC as a function of K and x: (a) Reduced cuboid model (2D), (b) Cuboid model (3D).
28
A. Naegel et al.
3.3. Virtual Diffusion Experiment
In the following section, the proposed model (without adsorption and metabolism) is used to simulate a permeation experiment by Hansen et al. (53, 54) in a diffusion cell. A comparison with the concentration–SC-depth profiles is made for the model substances caffeine (hydrophilic) and flufenamic acid (lipophilic). The used geometry is the rectangular membrane with N = 16 cell layers. Since experimentally full skin membrane is used, we expand our model by a further compartment (height: 1.5 mm) for the deeper layers of the skin (deeper skin layers, DSL), i.e., living epidermis and dermis. In addition to the diffusion coefficient Dlip, Dcor, and the partition coefficient Kcor/lip, the diffusion coefficient DDSL, and the partition coefficients Klip/don and KDSL/lip are required. For donor and acceptor, temporally and spatially constant Dirichlet boundary conditions (infinite-dose conditions) are adopted: Whereas the donor has a constant concentration cdon, the boundary to the acceptor is modeled as a perfect sink, i.e., with a vanishing concentration.
3.3.1. Concentration– SC-Depth Profiles
The model is driven by experimentally determined partition and diffusion coefficients of Hansen et al. (54), which are listed in Table 5. The model is validated by a comparison of the concentration–SC-depth profiles for the three incubation times t = 1 h, 2 h, and 6 h. The illustrations for flufenamic acid are shown in Fig. 13 and for caffeine in Fig. 14. In all cases, a good agreement between experiment and simulation is achieved. In the qualitative assessment of the curve for flufenamic acid, it must be noted that
Table 5 Experimental input parameters from ref. 53, 54 Substance Parameter
Flufenamic acid
Caffeine
Dlip (cm2/h)
1.1 × 10−4
2.1 × 10−4
DSC (cm2/h)
1.7 × 10−7
1.4 × 10−7
DDSL (cm2/h)
4.9 × 10−3
2.3 × 10−3
Klip/don
20.32
2.15
cdon (mg/ml)
1.0
12.5
KSC/don
5.96
4.78
Kcor/lip
0.21
2.31
0.10
KDSL/lip Dcor (cm /h) 2
0.07
5.1 × 10
−8
1.7 × 10−9
1 Computational Modeling of the Skin Barrier
Fig. 13. Concentration–SC-depth profiles of experiment and simulation for flufenamic acid at t = 1 h, 2 h, and 6 h.
Fig. 14. Concentration–SC-depth profiles of experiment and simulation for caffeine at t = 1 h, 2 h, and 6 h.
29
30
A. Naegel et al.
apparently the steady state in the simulation adjusts much faster than that in the experiment. A possible cause is that the model does not take into account the fact that flufenamic acid can be adsorbed onto keratin.
Acknowledgments The authors thank Steffi Hansen, Claus-Michael Lehr, Dirk Neumann and Ulrich Schaefer for conducting the experiments and for providing experimental input parameters. Further, the authors thank Dirk Feuchter, Yu-Hong Liu and Christine Wagner for providing the software TKD Modeller and the Cuboid Modeller, respectively. Parallel computations were performed on the SGI Altix 4700 system at the Leibniz-Rechenzentrum, Munich. Parts of this work were funded by the ZEBET division of the Federal Institute for Risk Assessment, Berlin under Contract No. BfR-ZEBET-1328-177. References 1. Scheuplein, R.J., Blank, I.H. (1971) Permeability of the skin. Physiol. Rev. 51, 702–747 2. Barry B.W. (1991) Modern methods of promoting drug absorption through the skin. Mol. Aspects Med. 12, 195–241 3. Rim, J.E. , Pinsky, P.M. , van Osdol, W.W. (2007) Using the method of homogenization to calculate the effective diffusivity of the stratum corneum. J. Membr. Sci. 293, 174–182 4. Rim, J.E., Pinsky, P.M., van Osdol, W.W. (2008) Using the method of homogenization to calculate the effective diffusivity of the stratum corneum with permeable corneocytes. J. Biomech. 41, 788–796 5. Wagner, C. (2008) Dreidimensionale digitale Rekonstruktion des humanen stratum corneum der Haut in Kombination mit Simulation substantieller Diffusion durch das stratum corneum, Veterinary University of Hannover, PhD thesis (in German). 6. Thomson, W., Lord Kelvin (1887) On the division of space with minimum partitional area. Phil. Mag. 24, 503 7. Feuchter, D., Heisig, M., Wittum, G. (2006) A geometry model for the simulation of drug diffusion through the stratum corneum. Comp. Visual. Sci. 9, 117–130 8. Christophers, E., Wolff, H.H., Laurence, E.B. (1974) The formation of epidermal cell columns. J. Invest. Dermatol. 62, 555–559
9. Menton, D.N. (1976) A liquid film model of tetrakaidecahedral packing to account for the establishment of epidermal cell columns. J. Invest. Dermatol. 66, 283–291 10. Allen, T.D., Potten, C.S. (1976) Significance of cell shape in tissue architecture. Nature 264, 545–547 11. Feuchter, D. (2008) Geometrie- und Gittererzeugung fuer anisotrope Schichtengebiete, University of Heidelberg, PhD thesis (in German). 12. Richter, T., Mueller, J.H., Schwarz, U.D., Wepf, R., Wiesendanger, R. (2001) Investigation of the swelling of human skin cells in liquid media by tapping mode scanning force microscopy. Appl. Phys. A 72, 125–128 13. Richter, T., Peuckert, C., Sattler, M., Koenig, K., Riemann, I., Hintze, U., Wittern, K-P., Wiesendanger, R., Wepf, R. (2004) Dead but highly dynamic – The stratum corneum is divided into three hydration zones. Skin Pharmacol. Physiol. 17, 246–257 14. Bouwstra, J.A., de Graaff, A., Gooris, G.S., Nijsse, J., Wiechers, J.W., van Aelst, A. C. (2003) Water distribution and related morphology in human stratum corneum at different hydration levels. J. Invest. Dermatol. 120, 750–758 15. Kashibuchi, N., Hirai, Y., O’Goshi, K., Tagami, H. (2002) Three-dimensional analy-
1 Computational Modeling of the Skin Barrier ses of individual corneocytes with atomic force microscope: morphological changes related to age, location and to the pathologic skin conditions. Skin Res. Technol. 8, 203–211 16. Mihara, M. (1988) Scanning electron microscopy of skin surface and the internal structure of corneocyte in normal human skin. An application of the osmium-dimethyl sulfoxide-osmium method. Arch. Dermatol. Res. 288, 293–299 17. Anderson, R.L., Cassidy, J.M. (1973) Variation in physical dimensions and chemical composition of human stratum corneum. J. Invest. Dermatol. 61, 30–32 18. Raykar, P.V., Fung, M.C., Anderson, B.D. (1988) The role of protein and lipid domains in the uptake of solutes by human stratum corneum. Pharm. Res. 5, 140–150 19. Heisig, M., Lieckfeldt, R., Wittum, G., Mazurkevich, G., Lee, G. (1996) Non steady-state descriptions of drug permeation through stratum corneum. I. The biphasic brick-and-mortar model. Pharm. Res. 13, 421–426 20. Johnson, M.E., Blankschtein, D., Langer, R. (1997) Evaluation of solute permeation through the stratum corneum: lateral bilayer diffusion as the primary transport mechanism. J. Pharm. Sci. 86, 1162–1172 21. Barbero, A.M., Frasch, H.F. (2006) Transcellular route of diffusion through stratum corneum: Results from finite element models. J. Pharm. Sci. 95, 2186–2194 22. Wang, T.-F., Kasting, G.B., Nitsche, J.M. (2006) A multiphase microscopic diffusion model for stratum corneum permeability. I. Formulation, solution, and illustrative results for representative compounds. J. Pharm. Sci. 95, 620–648 23. COMSOL is a multiphysics software package for performing finite-element-method (FEM) simulations. See COMSOL AB, http://www. comsol.com/. 24. Bastian, P., Wittum, G. (1994) Robustness and adaptivity: The UG concept. In: Hemker, P., Wesseling, P. (eds.): Multigrid Methods IV, Birkhäuser, Basel. 25. Lang, S., Wittum, G. (2005) Large scale density driven flow simulations using parallel unstructured grid adaptation and local multigrid methods. Concurrency Computat. 17, 1415–1440 26. Bank, R.E., Rose, D.J. (1987) Some errorestimates for the box method. SIAM J. Numer. Anal. 24(4), 777–787 27. Cai, Z. (1991) On the finite volume element method. Numer. Math. 58, 713–735
31
28. Michev, I.D. (1996) Finite Volume and Finite Volume Element Methods for Nonsymmetric Problems. PhD thesis, Texas A&M Univ., Inst. for Scientific Computation, 612 Blocker, College Station, Texas 77843–3404, USA. Also available as Technical Report ISC-96-04MATH. 29. Bensoussan, A., Lions, J. L., Papanicolaou, G. (1978) Asymptotic analysis for periodic structures. North-Holland Publishing. Amsterdam, New York, Oxford. 30. Neuss, N. (1996) Homogenisierung und Mehrgitter, University of Heidelberg, PhD thesis (in German). 31. Muha, I., Naegel, A., Stichel, S., Grillo, A., Heisig, M., Wittum, G. (2011) Effective diffusivity in membranes with tetrakaidekahedral cells and implications for the permeability of human stratum corneum. J. Membr. Sci. 368, 18–25 32. Michaels, A.S., Chandrasekaran, S.K., Shaw, J.E. (1975) Drug permeation through human skin: theory and in vitro experimental measurement. AIChE J. 21, 985–996. 33. Albery, W.J., Hadgraft, J. (1979) Percutaneous absorption: theoretical description. J. Pharm. Pharmacol. 31,129–139. 34. Gienger, G., Knoch, A., Merkle, H.P. (1986) Modeling and numerical computation of drug transport in laminates: model case evaluation of transdermal delivery system. J. Pharm. Sci. 75 (1) 9–15. 35. Tojo, K. (1987) Random brick model for drug transport across stratum corneum. J. Pharm. Sci. 76, 889–891. 36. Edwards, D.A., Langer R. (1994) A linear theory of transdermal transport phenomena. J. Pharm. Sci. 83, 1315–1334 37. Lee, A.J., King, J.R., Barrett D.A. (1997) Percutaneous absorption: a multiple pathway model. J. Control. Rel. 45, 141–151 38. Manitz, R., Lucht, W., Strehmel, K., Weiner, R., Neubert, R. (1998) On mathematical mode ling of dermal and transdermal drug delivery. J. Pharm. Sci. 87, 873–879 39. Anissimov, Y.G., Roberts, M.S. (1999) Diffusion modeling of percutaneous absorption kinetics. 1. Effects of flow rate, receptor sampling rate, and viable epidermal resistance for a constant donor concentration. J. Pharm. Sci. 88, 1201–1209 40. Charalambopoulou, G.Ch., Karamertzanis, P., Kikkinides, E.S., Stubos, A.K., Kanellopoulos, N.K., Papaioannou, A.Th. (2000) A study on structural and diffusion properties of porcine stratum corneum based
32
A. Naegel et al.
on very small angle neutron scattering data. Pharm. Res. 17, 1085–1091 41. Anissimov, Y.G., Roberts, M.S. (2001) Diffusion modeling of percutaneous absorption kinetics: 2. Finite vehicle volume and solvent deposited solids. J. Pharm. Sci. 90, 504–520 42. Frasch, H.F. (2002) A random walk model of skin permeation. Risk Analysis 22, 265–276 43. Kubota, K., Dey, F., Matar, S.A., Twizell, E.H. (2002) A repeated-dose model of percutaneous drug absorption. Appl. Math. Modelling 26, 529–544 44. Mitragotri, S. (2003) Modeling skin permeability to hydrophilic and hydrophobic solutes based on four permeation pathways. J. Contr. Rel. 86, 69–92 45. Frasch, H.F., Barbero, A.M. (2003) Steadystate flux and lag time in the stratum corneum lipid pathway: results from finite element models. J. Pharm. Sci. 92, 2196–2207 46. Anissimov, Y.G., Roberts, M.S. (2004) Diffusion modeling of percutaneous absorption kinetics: 3. Variable diffusion and partition coefficients, consequences for stratum corneum depth profiles and desorption kinetics. J. Pharm. Sci. 93, 470–487 47. George, K., Kubota, K., Twizell, E.H. (2004) A two-dimensional mathematical model of percutaneous drug absorption. BioMedical Engineering OnLine 3,18 48. George, K. (2005) A two-dimensional mathematical model of non-linear dual-sorption of percutaneous drug absorption. BioMedical Engineering OnLine 4,40 49. Rim, J.E., Pinsky, P.M., van Osdol, W.W. (2005) Finite element modeling of coupled diffusion with partitioning in transdermal drug delivery. Ann. Biomed. Eng. 33, 1422–1438 50. Barbero, A.M., Frasch, H.F. (2005) Modeling of diffusion with partitioning in stratum corneum using a finite element model. Ann. Biomed. Eng. 33, 1281–1292 51. Mollee, T.R., Bracken, A.J. (2007) A model of solute transport through stratum corneum
using solute capture and release. Bull. Math. Biol. 69, 1887–1907 52. Chen, L., Lian, G., Han, L. (2008) Use of “bricks and mortar” model to predict transdermal permeation: model development and initial validation. Ind. Eng. Chem. Res. 47 (17), 6465–6472 53. Naegel, A., Hansen, S., Neumann, D.,, Lehr, C.M., Schaefer, U.F., Wittum, G., Heisig, M. (2008) In-silico model of skin penetration based on experimentally determined input parameters. Part II: Mathematical modelling of in-vitro diffusion experiments. Identification of critical input parameters. Eur. J. Pharm. Biopharm. 68, 368–379 54. Hansen, S., Henning, A., Naegel, A., Heisig, M., Wittum, G., Neumann, D., Kostka, K.H., Zbytovska, J., Lehr, C.M., Schaefer, U. F. (2008) In-silico model of skin penetration based on experimentally determined input parameters. Part I: Experimental determination of partition and diffusion coefficients. Eur. J. Pharm. Biopharm. 68, 352–367 55. Anissimov, Y.G., Roberts, M.S. (2009) Diffusion modeling of percutaneous absorption kinetics: 4. Effects of a slow equilibration process within stratum corneum on absorption and desorption kinetics. J. Pharm. Sci. 98, 772–781 56. Rim, J.E., Pinsky, P.M., van Osdol, W.W. (2009) Multiscale modeling framework of transdermal drug delivery. Ann. Biomed. Eng. 37, 1217–1229 57. Naegel, A., Heisig, M., Wittum, G. (2009) A comparison of two- and three-dimensional models for the simulation of the permeability of human stratum corneum. Eur. J. Pharm. Biopharm. 72, 332–338 58. Goodyer, C.E., Bunge, A.L. (2009) Comp arison of numerical simulations of barrier membranes with impermeable flakes. J. Membr. Sci. 329, 209–218 59. Mitragotri, S., Anissimov, Y.G., Bunge, A.L., Frasch, H.F., Guy, R.H., Hadgraft, J., Kasting, G.B., Lane, M.E., Roberts, M.S. (2011) Mathematical models of skin permeability: An overview. Int. J. Pharm. in press doi:10.1016/j.ijpharm.2011.02.023
Chapter 2 In Vitro Human Skin Segmentation and Drug Concentration–Skin Depth Profiles Ana Melero, Tsambika Hahn, Ulrich F. Schaefer, and Marc Schneider Abstract Highly optimized methods for skin segmentation are provided using tape stripping in combination with infrared absorption measurements for stratum corneum (SC) and cryosectioning for deeper skin layers. Furthermore, an example is calculated for demonstration of the respective procedures. Key words: Tape stripping, Cryosectioning, Skin segmentation, IR-D measurements
Nomenclature AT c EXn
Area of tape stripping (e.g. 1.767 cm2) Drug concentration in the extract of n-th pool of tape strips cl Drug concentration in skin for l-th pool of cuts cn Drug concentration in skin for n-th pool of tape strips dnn Stratum corneum (SC) depth after nn pools of tape strips dn* Absolute SC thickness of the n-th pool drel(n)* Relative SC thickness removed per pool n dtot Total SC thickness dx Middle of segment h First strip belonging to the n-th pool i Tape strip number IR-D Infrared densitometry j Last strip belonging to the n-th pool Lk* Depth in the deeper skin layers after pool k of cuts
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_2, © Springer Science+Business Media, LLC 2011
33
34
A. Melero et al.
Ll* Absolute thickness of pool l of cuts Ll Depth into the deeper skin layers after pool l of cuts Ltot Total thickness of the stripped skin mEXn Extracted amount of drug of pool n Nmax Total number of pools of cuts nmax Total number of tape strips Q Cumulative drug amount in SC without strip 1 SC Stratum corneum VEx Volume of the extracting agent (e.g. 2.0 ml) wk Weight of the pool of cuts/rest of the skin xi Absorption of a tape strip measured by IR-D and corrected for absorption of empty tape xi* Absorption measured by IR-D non-corrected value X N max Absorption of total number of tapes nmax x0 Absorption of an empty tape measured by IR-D xpool(n) Sum of absorption of strips measured by IR-D in pool n containing strips h − j measured by IR-D WN max Weight of total number of pools of cuts
1. Introduction Over the last few decades, many methods have been designed to study drug transport through the skin (permeation) and drug distribution within the different human skin layers (penetration) (1, 2). This information is essential for pharmaceutical and medical applications with respect to drug invasion, drug bioavailability, and safety aspects, as well as for risk assessment of chemicals (3). Here, we present an optimized in vitro method to evaluate drug distribution within the stratum corneum (SC) and the deeper skin layers (DSL) of human skin samples (4, 5). This method is based on the physical segmentation of skin incubated with a formulation containing the drug under study at preselected time intervals. The drug amount in thin slices of skin, taken parallel to the skin surface, is analyzed and plotted versus skin depth to obtain drug concentration–skin depth profiles. Additionally, different time points allow the determination of drug penetration kinetics. These profiles provide information about the following: ●●
●●
●●
●●
The pathway followed by a drug after its application onto the skin’s surface, The influence of the incubation time and the vehicle on drug distribution, Drug partition, and Saturation of the SC as well as possible formation of depots in the different layers.
2 In Vitro Human Skin Segmentation and Drug Concentration–Skin Depth Profiles
35
2. Materials 2.1. Measurement of the Skin Thickness
1. Two pairs of forceps.
2.1.1. Complete Skin Thickness
3. Caliper gauge.
2.1.2. SC Thickness
1. Punch biopsy of 4 mm diameter (Kai Europe GmbH, Solingen, Germany).
2. Glass slide with coverslip.
2. Freezing block. 3. Tissue-Tek® (Sakura Finetek, Torrance, CA, USA). 4. Cryomicrotome (Slee, Mainz, Germany) with knife type C (Slee, Mainz, Germany). 5. FluorSave™ (Calbiochem, San Diego, CA, USA). 6. Conventional light microscope (Leica, Wetzlar, Germany) equipped with a graduate scale bar, a digital camera (Jenoptik, Jena, Germany), and image software to analyze the images. (Carl Zeiss Axio Vision Rel. 4.7, Carl Zeiss Imaging Solutions GmbH, Göttingen, Germany). Visualization is performed at 400× magnification. 2.2. Tape Stripping
1. Self-made tape stripping apparatus. It consists of three pieces (see Fig. 1): (a) Stamp that can be fixed in a high position with screws or be placed directly on the skin. (b) Aluminum block. It has a 25 mm diameter × 20 mm depth hole in the middle. The hole is filled with cork disks. (c) Weight of 2 kg. 2. Needles (Prym, Stolberg, Germany) to fix the skin on cork disks. 3. A self-made Teflon mask with a 15-mm diameter hole (stripping area). 4. Tape strips: the tape used is Tesa® Film kristall-klar Nr. 5733000000 (Tesa AG, Hamburg, Germany) cut into pieces (19 × 19 mm) with scissors. 5. Stopwatch. 6. Two pairs of forceps. 7. Infrared densitometer: Squame Scan™ 850A (Heiland electronic GmbH, Wetzlar, Germany) including sample holders and standard filters. 8. Glass pearls 2 mm diameter (Merck, Darmstadt, Germany). 9. Glass scintillation vials (20 mL, neoLab Migge LaborbedarfVertriebs GmbH, Heidelberg, Germany).
36
A. Melero et al.
Fig. 1. Self-made tape stripping device.
2.3. Cryosectioning of the Deeper Skin Layers
1. Carbon dioxide gas bottle. 2. 13 mm diameter punch and rubber mallet. 3. Self-made aluminum tube, closed on one side, where the skin is placed and that can be fixed in the cryomicrotome. It has a ring that can be moved to different positions (see Fig. 2). 4. Glass slide. 5. Two pairs of forceps. 6. One dissection pin. 7. Purified water prepared by Millipore synthesis device (Millipore GmbH, Schwalbach, Germany). 8. Cryomicrotome and knife; see Subheading 2.1.2. 9. Glass test tubes (e.g. 98 × 17 mm, Assistent-Präzision, Glaswarenfabrik Karl Hecht GmbH & Co KG, Sondheim, Germany). 10. Analytical balance (Sartorius, Göttingen, Germany).
2.4. Extraction of Drug from Tape Strips and Sections of Deeper Skin Layers
1. Samples, collected in scintillation vials and test tubes. 2. Extraction medium. 3. Pipette (Eppendorff, Hamburg, Germany). 4. Shaker. 5. Centrifuge.
2 In Vitro Human Skin Segmentation and Drug Concentration–Skin Depth Profiles
37
Fig. 2. Self-made cryo block.
3. Methods Human skin is obtained from abdominal plastic surgery and prepared for storage. For performing the experiments, 25 mm diameter pieces of skin are used. After incubation for different time intervals with the formulation under study, the SC is removed by successive application of adhesive tapes onto the surface of the skin. The SC cells (corneocytes) and intercellular lipids adhere to the tape and can be easily removed by this technique. Two parameters are essential for the correct investigation of skin depth profiles: first, the amount of skin removed with each sample is required and second the amount of drug contained in each sample. From these parameters, the SC depth and the normalized drug concentration (drug amount per skin volume of the sample) can be calculated. A first step toward generating concentration–depth profiles is the measurement of both, the total thickness of the whole skin and also of the thickness of the intact SC. Traditionally, the stripping depth has been determined by the tapestrip number (4). It was assumed that repeated stripping may remove a constant amount of SC, independent of the number of strips already performed. The momentary depth may thus be calculated by dividing the mean thickness of the overall SC by the total number of tapes. However, it has been observed that the first tape strips remove a higher amount of SC than the successive strips (6, 7). Therefore, the validity of this assumption has been questioned and alternative methods have been introduced, which determine the amount of SC individually on each tape strip, such as differential weighing (6) or protein assays (8). These methods are time-consuming and (in the latter case) destructive. Thus, the drug amount and the depth in the SC cannot be determined on the identical tape strip.
38
A. Melero et al.
A fast and nondestructive method is infrared densitometry (IR-D), which measures the optical absorption through a tape strip and the SC on its surface (9). The reduction in light intensity of the IR-D is linearly related to the SC amount on the tape and is expressed in terms of percentage of absorption (%). Recently, a correlation between the SC depth determined by infrared densitometry and BCA protein assay has been established for in vitro human skin with Tesa® Film kristall-klar (5). In addition, it could be clearly demonstrated that only negligible amounts of SC were left on the skin’s surface when reaching the lower limit of quantification (LLOQ) of the instrument. As the cumulative absorption of all tape strips is related to the total SC thickness, the depth into the SC can be easily calculated from the sum of optical absorption until the tape strip under investigation (see Subheading 3.5 Eqs. 1–5). After removal of the complete SC by tape stripping, the stripped skin is frozen by expanding carbon dioxide, and 25 mm thickness slices are cut by means of a cryomicrotome. The drug is extracted from all samples and hereafter quantified. Normalized drug concentration can be plotted versus skin depth to see the drug distribution within the different layers after the selected incubation times. 3.1. Measurement of the Skin Thickness 3.1.1. Complete Skin Thickness
1. The caliper gauge is calibrated to 0 using a glass slide and a coverslip. 2. Place the skin disk on a glass slide and put a coverslip on top. 3. Measure the whole skin thickness with a tactile probe. The total skin thickness is the mean of five measurements on different positions on the skin (see Fig. 3).
Fig. 3. Skin thickness measurement positions.
2 In Vitro Human Skin Segmentation and Drug Concentration–Skin Depth Profiles 3.1.2. SC Thickness
39
The mean SC thickness for each skin donor is calculated from microscopic pictures of histological cross-sections on at least ten different regions (see Fig. 4). Skin samples are prepared as follows: 1. Punch out a 4 mm biopsy of frozen full-thickness skin. 2. Temperate an aluminum tube (see Fig. 2) in the cryomicrotome. 3. Put a drop of Tissue-Tek® on the metal block and wait until it starts to freeze. 4. Embed the biopsy in the Tissue-Tek® with the SC arranged perpendicular to the metal block surface and place the metal block in the cryomicrotome to allow complete freezing (see Note 9). 5. Cut the skin perpendicular to the surface of the skin in 10 mm thick slices and mount these slices directly on glass slides (the slides need to be warmer than the slices, e.g. room temperature) by melting the frozen slices to the slides. 6. Staining of the skin, e.g. with haematoxylin will improve the contrast. Afterward, a drop of FluorSave™ is applied to the sample and a coverslip is placed onto it to protect the sample. 7. Measure the SC thickness using a light microscope and a scaled slide. With help of the software and a graduated scale bar, the exact thickness of the SC can be determined.
Fig. 4. Histological cross-section of human skin at 400× magnification.
40
A. Melero et al.
3.2. Tape Stripping
1. Cut the Tesa® Film tape in strips of 19 mm length each with scissors. 2. Autozero the Squame Scan™ 850A with the provided 0% standard filter and measure the provided gray filter, which should give 33.8% absorption. 3. Place an empty tape strip on the sample holder of the SquameScan™ 850A with the adhesive side facing up and measure the optical density. This will be taken as background and will be later subtracted from the values obtained for the samples. 4. Fix the skin to the aluminum block filled with cork disks with needles using forceps. Stretch the skin properly to reduce wrinkles. 5. Cover the skin with a Teflon mask with the hole of 15 mm diameter centered on the skin, to define the stripping area. Fix it by tightening the screws. 6. Place one tape strip centrally on the stripping area with the forceps and press slightly with the back side of the forceps to assure full contact between the tape and the skin. 7. Place the aluminum block under the stamp of the stripping device and charge the stamp with a weight of 2 kg for exactly 10 s. 8. Remove the adhesive tape rapidly with forceps in one swift move. 9. Directly after stripping, transfer each strip on the sample holder with the adhesive side facing up and place the sample holder with the tape strip under the measuring head. The optical absorption of each tape is directly displayed. 10. Strip the skin with the Teflon mask present until the SC is removed completely. This end point has to be determined previously, for Tesa® Film kristall-klar this is the case, when the lower limit of quantification (LLOQ) of the IR-D is reached. The LLOQ can be calculated from the fivefold background noise of an empty tape. Reaching this end point, no further strips are applied (5). 11. Cover the sticky side of the tape strips with some glass pearls and transfer them into scintillation vials with forceps with the sticky side facing up. 12. Depending on the nature of the analyzed drug it may be necessary to combine several tapes into pools (e.g. “pool 1”: tape n° 1. “pool 2”: tape n° 2. “pool 3”: tapes n° 3–5. “pool 4”: tapes n° 6–10. “pool 5”: tapes n° 11–15. “pool 6”: tapes n° 16–20).
2 In Vitro Human Skin Segmentation and Drug Concentration–Skin Depth Profiles
3.3. Cryosectioning of the Deeper Skin Layers
41
1. After complete removal of the SC freeze the rest of the skin, still fixed on the aluminum block, in a stream of expanding carbon dioxide until achieving a pale appearance. 2. Take a 13-mm diameter (cryosection area) biopsy and allow thawing. 3. Use some drops of water to assure the contact between the closed part of the cryoblock and the skin. 4. Mount the biopsy (Epidermis up) on the drops. 5. Place the spacer ring in the upper position, at the same height as the skin and cover the skin with a glass slide. 6. Freeze the skin using expanding carbon dioxide again while retaining the glass slide in the same position to obtain a flat surface. 7. Remove the glass slide, put the ring in the lowest position to uncover the skin and fix the tube with the frozen skin to the cryomicrotome. 8. Perform the cuts parallel to the skin’s surface (thickness of the skin cuts: 25 mm). 9. Collect the cuts in 12 preweighed centrifuge tubes and combine them using the following scheme: incomplete cuts, 4 × 2 cuts, 4 × 4 cuts, 2 × 8 cuts, and skin rest. Depending on the drug and the incubation time, different schemes may be applied. 10. Weigh the centrifuge tubes containing the samples. The weight of the skin samples is later related to the skin thickness (see Subheading 3.5).
3.4. Extraction of Drug from Tape Strips and Deeper Skin Layers
The selection of the extraction media has to be carefully evaluated for each drug. To assure the recovery effectiveness, a mass balance should be carried out. For doing so, the whole experiment is done as usual and every part of the setup in contact with the formulation is extracted with the medium, including the rest of the applied formulation. After extraction and analysis of all these samples, a mass balance should be performed. The recovery should always be between 85 and 115% of the applied amount of drug. The selected solvent should not remove the glue of the tapes or skin components, as they might interfere in the chromatogram or damage the HPLC column, if the latter is selected as analytical method. A suggestion for an extraction procedure is described below: 1. Cover each sample with an appropriate volume of extraction medium and shake it for 2 h. 2. Centrifuge the skin cuts at 5,000 rpm and 800 × g for 30 min. 3. Collect the supernatant and analyze the drug amount.
42
A. Melero et al.
3.5. Data Treatment and Plotting
A step-by-step calculation of the depth in the SC and in the viable skin layers is presented below. Please note that the first tape strip also removes rests of formulation that can be left on the skin surface after cleaning. Therefore, the first strip is usually not plotted in drug concentration–skin depth profiles (7). However, it has to be taken into account for determining the SC depth as well as for mass balance. ●●
Corrected absorption value for a single tape strip xi = xi* - x 0
(1)
where, xi is the absorption of a tape strip measured by IR-D and corrected for absorption of empty tape, xi* absorption measured by IR-D, noncorrected value, x0 is the absorption of an empty tape measured by IR-D, i is the tape strip number. ●●
Absorption value for pooled strips j
x pool(n ) = å xi
(2)
i =h
where, xpool(n) is the sum of absorption of strips in pool n containing strips h − j measured by IR-D, xi: absorption of i-th tape strip measured by IR-D and corrected for empty tape, i is the tape strip number, h is the first strip belonging to the n-th pool, j is the last strip belonging to the n-th pool. ●●
Relative thickness of SC removed per pool j
* drel( n ) = å x i / X n max
X nmax =
with
i =h
n max
åx i =1
i
(3)
*
where, drel(n ) is the relative SC thickness removed per pool n, xi is the absorption of i-th tape strip measured by IR-D and corrected for empty tape, h is the first strip belonging to the n-th pool, j is the last strip belonging to the n-th pool, nmax is the total number of tape strips, Xmax is the absorption of total number of tapes nmax. ●●
Absolute SC thickness of the n-th pool * dn* = dtot ·drel( n)
(4)
*
where dn is the absolute SC thickness of the n-th pool, dtot is * the total SC thickness, and drel( is the relative SC thickness n) removed per pool n. ●●
SC depth after nn pools of tape strips nn
dnn = å dn* n =1
(5)
2 In Vitro Human Skin Segmentation and Drug Concentration–Skin Depth Profiles
43
dnn
* is the SC depth after nn pools of tape strips and dn is the thickness of the n-th pool. ●●
Absolute thickness of pool l of cuts l
L*l = L tot ·å wk / WN max
with WN max =
k =1
N max
åw k =1
k
(6)
Ll* is the absolute thickness of pool l of cuts, Ltot is the total thickness of the stripped skin, wK is the weight of the pool of cuts/rest of the skin, k is the lower number of pools of cuts, l is the upper number of pools of cuts, Nmax is the total number of pools, and WN max is the weight of total number of pools. ●●
Depth of Deeper Skin Layers after pool l of cuts l
Ll = å L*k
k =1
(7)
Ll is the depth into the deeper skin layers after pool l of cuts and Lk* is the thickness of pool k of cuts. ●●
Drug amount in pool n mEXn = c EXn *V EX
(8)
where mEXn is the extracted drug amount of pool n, c EXn is the drug concentration in the extract of n-th pool of tape strips, and VEx is the volume of the extracting agent (e.g. 2.0 ml). ●●
Normalized drug concentration in SC for n-th pool of tape strips cn =
c Exn ·V Ex
AT ·dn*
(9)
where cn is the normalized drug concentration in skin for n-th pool of tape strips, c Exn is the drug concentration in the extract of n-th pool of tape strips, VEx is the volume of the extracting agent (e.g. 2.0 ml), AT is the area of tape stripping (e.g. 1.767 cm2), and d * is the thickness of SC removed by n tape strips of pool n. ●●
Normalized drug concentration in Deeper Skin Layers for l-th pool of cuts cl =
c Exl ·V Ex
AC ·L*l
(10)
where cl is the normalized drug concentration in skin for l-th pool of cuts, c Exl is the drug concentration in the extract of
44
A. Melero et al.
l-th pool of cuts, VEx is the volume of the extracting agent (2.0 ml), AC is the area of cryocutting (1.327 cm2), and L*l is the thickness of viable skin removed by cuts of pool l. ●●
Calculation of the middle of segment dx =
(dh + d j ) 2
(11)
dx is the middle of segment, dh is the depth before first strip of the pool and dj is the depth after last strip of the pool. ●●
Cumulative drug amount in SC Q =
n max
åm i =2
EXn
(12)
where Q is the cumulative drug amount in SC without strip 1, mEXn is the extracted drug amount of pool n, and nmax is the total number of tapes. On the following pages, an example for the calculation to determine drug concentration skin depth profiles is given (see Fig. 5; Tables 1–3). For the calculations the following values are used: Stripping area (AT): 1.767 cm². Area of cryocuts (AC): 1.327 cm². Extraction volume for samples (mEx): 2 mL.
Stratum corneum
6000
Drug concentration [µg/cm3]
Drug concentration [µg/cm3]
Plots of drug concentration-depth profiles
5000 4000 3000 2000 1000 0
0
2
4
6 8 10 depth [mm]
12
14
16
Deeper Skin Layers
40 30 20 10 0
0
500
1000 1500 2000 2500 3000 depth [µm]
Fig. 5. Example of drug concentration–skin depth profiles for stratum corneum (SC) and deeper skin layers as obtained from the example values.
21.2
18.8
16.9
19
18.9
22.3
24.1
20.1
21.7
23.6
18.4
17.8
15.3
12.3
13.2
2
3
4
5
6
7
8
9
10
11
12
13
14
15
6.7
xi* measured by IR-D [%]
1
Empty strip
i
6.5
5.6
8.6
11.1
11.7
16.9
15
13.4
17.4
15.6
12.2
12.3
10.2
12.1
14.5
xi corrected (from IR-D) [%]
#5; (h = 11; j = 16)
#4; (h = 6; j = 10)
#3; (h = 3; j = 5)
#2; (h = j=2)
#1; (h = j=1)
Pool
i =h
j
x pool (n ) = å x i
43.5
78.3
34.7
12.1
14.5
sum of absorption of strips in pool n measured by IR-D [%]
Table 1 Determination of the SC thickness removed with the tape strips
21.7
39.0
17.3
6.0
7.2
relative amount of SC removed per pool [%]
i =h
j
* d rel (n ) = å x i / X n max
3.25
5.85
2.59
0.90
1.08
(continued)
thickness of segment [mm]
* dn* = d tot ·d rel (n )
2 In Vitro Human Skin Segmentation and Drug Concentration–Skin Depth Profiles 45
10.4
9.3
8.1
18
19
20
i =1
åx
334.6
11.4
17
i
11.8
16
n max
xi* measured by IR-D [%]
i
Table 1 (continued)
200.6
1.4
2.6
3.7
4.7
5.1
xi corrected (from IR-D) [%] #6; (h = 15; j = 20)
Pool 17.5
sum of absorption of strips in pool n measured by IR-D [%]
i =h
j
x pool (n ) = å x i
8.7
relative amount of SC removed per pool [%]
i =h
j
* d rel (n ) = å x i / X n max
1.31
thickness of segment [mm]
* dn* = d tot ·d rel (n )
46 A. Melero et al.
2 In Vitro Human Skin Segmentation and Drug Concentration–Skin Depth Profiles
47
Table 2 Determination of the depth into the viable skin layers l
L*l = L tot ·å wk / WN max k =1
Pool
wk weight of viable skin removed by pool [mg]
thickness of segment [mm]
#7
3.06
60.36
#8
3.07
60.55
#9
7.46
147.14
#10
9.29
183.24
#11
27.16
535.71
178.53
3,521.38
228.57
Cumulative weight of viable skin removed by all pools
#12 (rest of skin) i = #12
åw
i =# 7
i
Table 3 Calculation of concentration–depth profiles in the SC and the viable skin layers j
m
=c
Exn EXn cEx experimental extracted Pool [mg/ml] amount [mg]
*VEX
dn* = d tot ·å x i / X nmax
(dn + dn -1 ) / 2
i =h
thickness of segment [mm]
dn depth into middle of the SC [mm] segment [mm]
#1
0.5334
1.0668
1.08
1.08
0.54
#2
0.2268
0.4536
0.90
1.99
1.54
#3
0.2742
0.5484
2.59
4.58
3.29
#4
0.1377
0.2754
5.85
10.44
7.51
#5
0.0645
0.1290
3.25
13.69
12.07
#6
0.0220
0.0440
1.31
15
14.35
n max
Vn = AT ·d tot ·å x i / X nmax
cn =
volume of segment [cm3]
concentration [mg/cm3]
Q / AT normalized for i =2 amount in SC (without stripping area first strip) [mg] [mg/cm2]
1.916E-04
5,568.24
1.4504
1.599E-04
2,837.21
4.585E-04
1,196.11
1.035E-03
266.20
5.748E-04
224.44
2.312E-04
190.29
j
i =h
c Exn ·VEx AT ·dn*
Q =
åm
EX n
0.8208
(continued)
48
A. Melero et al.
Table 3 (continued) mEX n = c EX n * VEX
L*l = L tot ·wk / WN max
Pool
cEx experimental [mg/ml]
extracted amount [mg]
thickness of segment [mm]
#7
0.1352
0.2704
60.36
75
#8
0.0678
0.1356
60.55
136
#9
0.0654
0.1308
147.14
283
#10
0.0589
0.1178
183.24
466
#11
0.0532
0.1064
535.71
1002
#12
0.0504
0.1008
3521.38
d tot + (li + li -1 ) / 2
cl =
dtot + li depth inside the skin [mm]
4523
c Exl ·V Ex * l
AC ·L
#12
åm
i =# 7
#12
EX l
åm
i =# 7
EX l
/ AC
volume of segment [cm3]
concentration [mg/cm3]
amount in normalized for viable skin cutting area [mg] [mg/cm2]
45
8.009E-03
33.76
0.8618
106
8.035E-03
16.88
209
1.953E-02
6.70
375
2.432E-02
4.84
734
7.109E-02
1.50
2,763
4.673E-01
0.22
middle of segment [mm]
Vk = AC ·Ltot ·w K / WN max
0.6494
4. Notes 1. Commonly, abdominal skin from female patients is used for skin absorption experiments. Informed consent forms must be signed by the patients. 2. Safety aspects concerning biological materials have to be considered. 3. Skin should not be used after 6 months storage at −26°C, as the permeability characteristics and the SC thickness might be changed (10, 11). 4. The subcutaneous fatty tissue should never come into contact with the SC because its lipids could mix with the SC lipids and change its penetration characteristics. 5. Avoid repeated freezing and thawing of skin as this procedure changes the skin’s permeability characteristics.
2 In Vitro Human Skin Segmentation and Drug Concentration–Skin Depth Profiles
49
6. A detailed incubation procedure is not provided because it depends on the type of formulation and the drug. Basically, the formulation is applied onto the skin and incubated at 32°C temperature for the selected incubation time. Afterward, the excess formulation is removed; the surface of the skin is cleaned with cotton, and the skin sample is treated as explained. 7. Always handle the tape strips with forceps and only touch the corners of the tape strip that will not be measured by the infrared densitometer. 8. It has been observed that, especially with aqueous donor media, the generation of concentration–skin depth profiles is limited to a few hours of incubation. In this case, the epidermis might be torn off in large flaps instead of the SC being removed with successive tapes. 9. Tissue-Tek® gets a cloudy white appearance and gets more viscous when starting to freeze. 10. Depending on the drug, tape stripping and cryosection samples can be stored at −26°C for several days before extraction of the drug. 11. The glass pearls prevent the strips from sticking to each other or to the vials. 12. Carbon dioxide is normally used because it is the quickest way to freeze skin, as it is important to stop drug diffusion. This method is not suitable for volatile drugs. References 1. OECD. (2004) Guideline for the Testing of Chemicals. Skin Absorption: in vitro Method, 428. 2. Howes D, Guy R, Hadgraft J, Heylings J, Hoeck U, Kemper F, Maibach H, Marty JP, Merk H, Parra J, Rekkas D, Rondelli I, Schaefer H, Täuber U, Verbiese N. (1996) Methods for Assessing Percutaneous Absorption: The Report and Recommendations of ECVAM Workshop 13. ATLA Alternatives to Laboratory Animals. 24:81–106. 3. REGULATION (EC) No 1907/2006 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL concerning the Registration E, Authorisation and Restriction of Chemicals (REACH). (2006). 4. Wagner H, Kostka KH, Lehr CM, Schaefer UF. (2000) Drug Distribution in Human Skin Using Two Different In Vitro Test Systems: Comparison with In Vivo Data. Pharmaceutical Research. 17:1475–1481.
5. Hahn T, Hansen S, Neumann D, Kostka KH, Lehr CM, Muys L, Schaefer UF. (2010) Infrared Densitometry: A Fast and NonDestructive Method for Exact Stratum Corneum Depth Calculation for in vitro TapeStripping. Skin Pharmacology and Physiology. 23:183–192. 6. Kalia YN, Pirot F, Guy RH. (1996) Homogeneous Transport in a Heterogeneous Membrane: Water Diffusion Across Human Stratum Corneum In Vivo. Biophysical Journal. 71:2692–2700. 7. Lademann J, Jacobi U, Surber C, Weigmann HJ, Fluhr JW. (2009) The tape stripping procedure – evaluation of some critical parameters. European Journal of Pharmaceutics and Biopharmaceutics. 72:317–323. 8. Dreher F, Modjtahedi BS, Modjtahedi SP, Maibach HI. (2005) Quantification of stratum corneum removal by adhesive tape stripping by total protein assay in 96-well
50
A. Melero et al.
microplates. Skin Research and Technology. 11:97–101. 9. Voegeli R, Heiland J, Doppler S, Rawlings AV, Schreier T. (2007) Efficient and simple quantification of stratum corneum proteins on tape strippings by infrared densitometry. Skin Research and Technology. 13:242–251.
10. Harrison SM, Barry BW, Dugard PH. (1984) Effects of freezing on human skin permeability. Journal of Pharmacy and Pharmacology. 36:261–262. 11. Schaefer U. (1996) An ex-vivo model for the study of drug penetration into human skin. Pharmaceutical Research. 13 (Suppl.).
Chapter 3 Transcriptional Regulation of Epidermal Barrier Formation Ambica Bhandari, Michael L. Salmans, William Gordon, and Bogi Andersen Abstract The mammalian epidermis is a self-renewing stratified squamous epithelium. Its basal cell layer contains proliferating keratinocytes that exit the cell cycle when they move into the suprabasal compartment. These cells activate a gene differentiation program aimed at building a protective epidermal barrier as they move toward the surface, successively going through the spinous and granular layers. At the completion of this process, the keratinocytes become enucleated and form the cornified layer, the surface layer of the skin. The highly cross-linked protein–lipid envelope and extracellular lipids in the cornified layer along with cell–cell adhesions in the granular layer are required for an effective epidermal barrier. Transcriptional mechanisms are critical for the formation of the epidermal barrier, and in this chapter, we describe methods to evaluate the role of a transcription factor (TF) in epidermal differentiation. To identify direct target genes of a TF, we propose a combination of bioinformatics and experimental approaches. The ultimate goal of these approaches is to understand the mechanisms whereby a TF regulates epidermal barrier formation. Key words: Epidermal barrier, Keratinocytes, Time course microarray analysis, Electromobility shift assays, Chromatin immunoprecipitation, Luciferase
1. Introduction The mammalian epidermis consists of multiple layers of epithelial cells at different stages of differentiation. The proliferating progenitor cells are found in the basal layer; these cells are constantly moving toward the skin’s surface, progressively differentiating as they move along, ultimately reaching the outermost cornified layer which is continuously sloughed off. The cornified envelope and lipids of the cornified layer as well as the junctions between keratinocytes in the deeper layers constitute key elements of the
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_3, © Springer Science+Business Media, LLC 2011
51
52
A. Bhandari et al.
Computational Approaches
epidermal barrier; their disruption results in several skin diseases (1). This differentiation process is tightly regulated by a battery of TFs that either suppress the differentiation program, thus maintaining the undifferentiated state of basal cell layer keratinocytes, or activate it, thus promoting differentiation in the suprabasal layers (2, 3). Here, we describe selective methods to investigate the role of a transcription factor (TF) in epidermal differentiation (Fig. 1). An early step to determine the role of a TF is to mutate the encoding gene in mice. Once a knockout mouse is available, a range of Identification of genes whose expression mirrors expression of the TF
Identification of genes differentially expressed in TF -/- skin
Whole genome expression arrays in WT developmental time course
Whole genome expression arrays in WT vs TF-/-
ANOVA performed to identify DEG during the time course
Group genes with similar expression pattern using KMeans clustering
Cyber-T analysis to identify differentially expressed genes (DEG)
Correlation of TF expression with expression of other DEG
Potential TF target gene list
Experimental Approaches
ConSite and BEARR used to define potential TF binding sites in target genes
Test in-vitro binding of TF with putative binding sites using EMSA
Test in-vivo binding of TF with putative binding sites using ChIP assays
Test function of TFBS using reporter assays
Fig. 1. A schematic outline of methods to evaluate potential gene targets of a transcription factor in skin epidermal differentiation. TF transcription factor, DEG differentially expressed genes, BEARR Batch extraction and analysis of cisregulatory regions, EMSA electromobility shift assay, ChIP chromatin immunoprecipitation, TFBS transcription factor binding site.
3 Transcriptional Regulation of Epidermal Barrier Formation
53
experiments can be performed to elucidate the gene program regulated by the TF. 1. To identify genes that are regulated by the TF in skin, gene expression microarray experiments can be performed, comparing knockout and wild-type (WT) mouse skin. This data can be analyzed using Cyber-T (4), a publicly available program that defines statistically significantly differentially expressed genes in the knockout versus wild-type skin. Some of the differentially regulated genes may be direct targets of the TF. 2. To help identify genes directly regulated by a TF, a developmental time course microarray experiment can be performed using back skin from wild-type mice sampled over the differentiation process (5). Similar time course experiments can be done with in vitro cell models for epidermal differentiation, and we have used this approach to study transcriptional regulation in other epithelia such as hair follicles (6, 7) and bladder (5). The data are imported in Multiple Experiment Viewer (MeV) (8, 9) and analyzed by ANOVA to identify those genes whose expression changes significantly over the time course. The significantly altered genes can then be grouped using K-Means clustering to define which genes share a similar pattern of expression over the time course. Genes that cluster with the TF in question are more likely to be its targets, especially if they are also differentially regulated in the TF-mutated skin. Alternatively, one can use correlation analysis to identify those genes that show a strong positive or negative correlation in expression with the TF; if the TF acts as a repressor, target genes may show complementary patterns of expression. 3. The candidate target genes identified as described above are then analyzed for the presence of transcription factor binding sites (TFBS) using two publicly available programs ConSite (10) and BEARR (Batch extraction and analysis of cis-regulatory regions) (11). This approach assumes that the binding site for the TF has been determined. Presence of a conserved TFBS in the upstream region or first intron of a gene enhances the probability that the gene in question is a direct target of the TF. 4. The candidate TFBS can then be tested for TF binding in vitro using electro mobility shift assays (EMSA). This method will also determine the in vitro binding affinities of each site. 5. To determine whether the TF binds the candidate site in vivo, chromatin immunoprecipitation (ChIP) assays are performed using primary human or mouse keratinocytes and a specific antibody against the TF.
54
A. Bhandari et al.
6. To demonstrate that these sites are functional, the gene regulatory region in question is linked to a reporter gene, and the ability of the TF to activate transcription from a WT and TFBS-mutated reporter is evaluated. A more rigorous in vivo test of a TFBS is to generate BAC transgenic mice expressing GFP under the control of a wild-type and mutated promoter (12, 13). This is not discussed here, as it is beyond the scope of this chapter. While this series of experiments will not comprehensively identify all target genes of a TF, the proposed approach should allow the identification of multiple direct TF targets, thus providing insights into the mechanisms whereby a TF regulates epidermal differentiation and barrier formation.
2. Materials 2.1. Gel Shift Assays
1. ATP, [g 32P]-EasyTide, 250 mCi (Perkin Elmer; Cat. No. BLU502Z250UC). 2. T4 polynucleotide Kinase and buffer (Promega; Cat. No. M4101) Store at −20°C. 3. Loading Dye: 10 ml of 100% Glycerol, 2 ml of 0.5 M EDTA, 1% bromo-phenol blue. 4. 50 mM NACl in TE buffer. 5. 5× Gel shift Buffer: 50 mM HEPES (pH 7.4), 250 mM KCl, 25% glycerol. Aliquot and store at −20°C. 6. 10 mg/ml Bovine Serum Albumin (BSA). Store at −20°C. 7. 100 mM Dithiothreitol (DTT). Aliquot and store at −20°C. Avoid freeze–thawing. 8. Poly (dI-dC)∙Poly(dI-dC) (1 mg/ml) (EMD chemicals; Cat. No. 71294-125ul) Store at −20°C. 9. 10% native gel: 50 ml Protogel, 50 ml 1× TBE, 600 ml 10% APS, 60 ml TEMED. 10. 6% native gel: 20 ml Protogel, 50 ml 1× TBE, 10 ml 50% Glycerol, 20 ml Water, 600 ml 10% APS, 60 ml TEMED. 11. Scintillation Fluid – Cytoscint (Fisher; Cat. No. BP458-4).
2.2. Chromatin Immunoprecipitation Assays
1. Aprotinin-5 mg (Sigma; Cat. No. A11535MG). Make stock solution at 10 mg/ml in 0.01 M HEPES pH 8.0, aliquot as 10-ml volumes in PCR tubes, and store at −20°C. Dilute at 1:100 prior to use. 2. Leupeptin-5 mg (Sigma; Cat. No. L2884-5MG). Make stock solution at 10 mg/ml in water. Aliquot as 10-ml volumes
3 Transcriptional Regulation of Epidermal Barrier Formation
55
in PCR tubes and store at −20°C. Dilute at 1:1,000 prior to use. 3. PMSF-1G (Sigma; Cat. No. 78830-1G). Make stock solution at 100 mM in ethanol. Aliquot as 10-ml volumes in PCR tubes and store at −20°C. Dilute at 1:1,000 prior to use. 4. Pansorbin cells, lyophilized (Staph A cells). Reconstituted in distilled water (Calbiochem; Cat. No. 507862). 5. IgG from Rabbit serum, lyophilized (Sigma; Cat. No. I500610MG). 6. RNA polymerase II 8WG16 Monoclonal Antibody (Covance; Cat. No. MMS-126R). 7. GenElute PCR cleanup kit (Sigma; Cat. No. NA1020-1KT). 8. 37% Formaldehyde. Use at a final concentration of 1% formaldehyde. It can be directly added to the media. 9. 2.5 M Glycine. Store at RT. Use at a final concentration of 0.125 M. 10. 10× PBS (Phosphate-buffered saline). 11. 5 M NaCl. Store at RT. 12. 10 mg/ml of Salmon sperm DNA (Invitrogen; Cat. No. 15632-011). 13. Bioruptor-sonicator (Diagenode). 14. Denville Taq Polymerase and buffer (Denville; Cat. No. CB4050-1). 15. 2.5 mM dNTP Mix (Fermentas; Cat. No. R0192). 16. Dialysis Buffer (DB):2 mM EDTA, 50mM Tris–Cl pH 8.0. 17. SDS Lysis Buffer: 50 mM Tris–Cl pH 8, 10 mM EDTA, 1% SDS. 18. ChIP Dilution Buffer: 0.01% SDS, 1.1% Triton X 100, 1.2 mM EDTA, 16.7 mM Tris–Cl pH 8.0, and 167 mM NaCl. 19. Low-Salt wash buffer: 0.1% SDS, 1% Triton X 100, 2 mM EDTA, 20 mM Tris–HCl pH 8.1, 150 mM NaCl. 20. High-Salt wash buffer: 0.1% SDS, 1% Triton X 100, 2 mM EDTA, 20 mM Tris–HCl pH 8.1, 500 mM NaCl. 21. LiCl wash buffer: 0.25 M LiCl, 1% IGEPAL-CA630, 1% deoxycholic acid (sodium salt), 1 mM EDTA, and 10 mM Tris pH 8.1. 22. TE: 10 mM Tris–HCl, 1 mM EDTA pH 8.0. 2.3. Reporter Assays
1. Lipofectamine 2000 transfection reagent (Invitrogen, Cat. No. 11668-019). 2. Opti-MEM® I Reduced-Serum Medium (Invitrogen, Cat. No. 31985-070).
56
A. Bhandari et al.
3. E-Low calcium medium (14). Add 1.2 mM calcium to this media if you need to use differentiated cells for ChIP assays. 4. Dual-Luciferase Reporter System (Promega; Cat. No. E1910). 5. pGL3-Basic vector (Promega; Cat. No. E1751). 6. BD falcon round bottom tube (Cat. No. 352052)-For use in the luminometer. 7. 1× PLB: Add 1 volume of 5× Passive Lysis Buffer (PLB) to 4 volumes of distilled water. Mix well. Store at 4°C (£1 month). 8. LAR II: The lyophilized Luciferase Assay Substrate (LAR) is resuspended in 10 ml of Luciferase Assay Buffer II (for Cat. No. E1910, E1960). Store at −20°C (£1 month) or −70°C (£1 year). 9. Stop & Glo Reagent: Dilute the 50× reagent to 1× using the Stop & Glo Buffer. Make smaller amount as needed. Store at –20°C for 15 days.
3. Methods 3.1. Computational Approaches 3.1.1. Differential Gene Expression Analysis Using Cyber-T
Cyber-T is a statistical program with a Web interface (http://cybert. microarray.ics.uci.edu) that can be conveniently used on microarray data for the identification of statistically significant differentially expressed genes (4). The statistical analyses carried out by Cyber-T is based on either simple t-tests that use the observed variance of replicate gene measurements across replicate experiments, or regularized t-tests that use a Bayesian estimate of the variance among gene measurements within an experiment. Cyber-T is also able to compute posterior probability of differential expression (PPDE) based on the modeling of p-value distributions which is an indicator of an experiment-wide false discovery rate. It has been shown to provide a higher confidence list of differentially regulated genes compared to many other programs used for analyzing expression microarray data (15). Cyber-T input: It is recommended to include at least three replicates of control and test samples to allow for statistical inference. In order to run the data through Cyber-T, the Probe ID and the expression values must be separated from the additional annotation data which is provided with your array results and stored as a tab delimited text file (the annotation data is placed back in the Cyber-T output file). Absent genes (raw expression values below ~100 for Affymetrix chips) can be removed in Excel using the “IF” command and scoring for each gene in each sample as 1 or 0 if its expression value is present or absent, respectively. A gene is
3 Transcriptional Regulation of Epidermal Barrier Formation
57
included as present if it has expression values over ~100 in all three replicates in either the control or experimental samples. The tab delimited text file of Probe IDs and raw expression values of present genes are used as the input text file for this program. Since most microarrays are single dye experiments, you should choose the statistical analysis for “Control versus Experimental Data” and upload the input text file. After filling in the column information in the input file, make sure to check the button to perform PPDE analysis. A default value of “10” can be chosen for Bayesian standard deviation estimation. Cyber-T output: Once the Cyber-T is run, the output provides a score of all genes which can then be exported into an Excel sheet. For easier processing of data, you can delete most of the columns in the file which shows the output of Cyber-T algorithm’s various calculations. The columns you want to keep are the Probe set IDs, all experimental and control values, Bayes P or Bayes lnP, Fold, and PPDE values. After this is completed, the additional data you removed before running the Cyber-T program, including gene symbol and Gene annotation, can be added back (make sure that your sorting is correct to keep the appropriate information with its correct Probe ID or you can use Office Access to combine the two lists). All genes that are present and have a p-value £ 0.01 are the statistically significant differentially expressed genes. You can take advantage of the PPDE value and also use more stringent p-values depending on the confidence you wish to have in the differentially expressed genes (see Note 1). 3.1.2. Time Course Gene Expression Analysis and Clustering Using Multiple Experiment Viewer
Time course gene expression analysis is a useful tool for understanding the role of TFs across time. By analyzing this type of data it is possible to discover new connections between TFs and the downstream pathways on which they act. Multiple Experiment Viewer (MeV) is a program available publicly at http://www.tm4. org/ (8, 9). Within this software package are a number of useful tools for the analysis of time course gene expression data.
Analysis of Variance
Analysis of variance (ANOVA) is an extension of the student’s t-test to more than two experimental conditions making it a good choice for analyzing a time course experiment. ANOVA identifies genes with significant differences in their mean expression across three or more time points by computing the F statistic (the ratio of the variance among replicates for a single time point in relation to the variance in mean across all the time points) and producing an output p-value. A gene expression change is considered significant if the p-value is lower than the user specified critical p-value. ANOVA Input: MeV can take input from a variety of sources (.tav, Affymetrix, and .gpr), however the simplest method to input your gene expression data is to group your replicate experiments in Excel with a gene identifier column as your first column (like
58
A. Bhandari et al.
probe ID or gene symbol) and save it as a tab delimited text file. This file can then be opened in MeV by using the Load Data function (see Note 2). Along the top of the window, go to the Statistics button and select One-way ANOVA. The first step in running the ANOVA is selecting the number of experimental groups and then grouping your data (minimum of two replicate experiments per group) in the window. A critical p-value, which determines the cutoff for significance, is entered by default at p £ 0.01. ANOVA Output: Once the ANOVA is run, the data is output into two lists: genes that met the significance cutoff and those that did not. It is now possible to perform additional analysis on these genes within the MeV software or to export the data from the Table View and work with it elsewhere. K-Means Clustering
K-Means Clustering is a method by which a number of inputs (n), in our case genes, are grouped into a number (k) of clusters based on their similarity (nearest mean). This is a widely used algorithm to identify groups of genes that show similarity in expression over time. While there are exceptions, it is likely that genes that cohabit a cluster with a given TF are enriched for targets of this TF (5). K-Means input: In order to perform clustering, the values of the time course gene expression data must first be normalized. The first step is to move the significant gene expression outputs from the ANOVA back into Excel and generate the average expression value of your replicates for each gene at each time point of your dataset. Next, this average data is log2 normalized using the Log(base) function in Excel. Generate the average expression value of the log normalized data for all time points for each gene, and the resulting number is subtracted from each log normalized time point. The resulting log- and mean-normalized data can then be saved in Tab Delimited Text format and loaded back into MeV for K-means clustering. Clustering in MeV has multiple options, however the most straightforward method is to select the Clustering button, then choose K-Means, then provide a number of potential clusters (the default is 10) (see Note 3), a number of iterations (the default is 50, preferably ~500 to ensure the best fit for the data), and the distance metric (default is euclidean). This algorithm runs very quickly so it is possible to choose different starting conditions (e.g., cluster number) and view the results of each run independently. K-Means Output: After performing the clustering, the resulting analysis includes the following: expression images, which provide heat maps of the clusters showing the expression levels of the genes that were grouped together, centroid graphs, which show the average expression value across your time course for each cluster, expression graphs, which show the individual expression
3 Transcriptional Regulation of Epidermal Barrier Formation
59
values for each gene within a cluster, table views, which show the genes and their annotation data for each cluster, and cluster information, which shows the statistics for each cluster and what portion of the population they represent. Time Course Gene Expression Correlated with Expression of a TF
Pearson’s product moment correlation coefficient: This is a statistical method that can be applied to gene expression data as a measurement of the linear dependence between two genes (or arrays). Its use is based on the assumption that the expression values of a TF and the genes it regulates are similar (positive correlation) or complementary (negative correlation). Correlation coefficient values range from +1 to −1, where +1 suggests a strong correlation, 0 suggests no correlation, and −1 suggests negative correlation. As an example, the transcriptional coregulator LMO4 was suggested to regulate the expression of BMP7 based on correlation coefficients calculated from the expression values of the two genes in several microarray data sets (16). For analysis of microarray data, correlation coefficients can be calculated to compare a single TF’s expression with the rest of the genes from the time course data set: 1. Create an Excel spreadsheet with the log2 normalized expression values for each probe set that meets your present gene criteria, with each array replicate as an individual column and each probe set as an individual row (Make sure to use a gene identifier in the first column). 2. Create a new column for calculation of correlation coefficients. To calculate each correlation coefficient use the “CORREL” function in Excel, with the log2 expression value of your transcription factor from each microarray as “Array1” (For example, if you have five replicates in your experiment, there will be five values selected for Array1) and the log2 expression value of your probe set of interest from each microarray as “Array2”. 3. Repeat this for each probe set in your spreadsheet, keeping the same values for your transcription factor in “Array1” and replacing the values for each “Array2” with those from each probe set (see Note 4). 4. Use two additional columns in your spreadsheet to calculate the t-test statistic for each correlation coefficient and the p-values for each t-test statistic (use TDIST function in Excel to calculate the p-value). The list of probe sets can then be sorted by correlation coefficient and p-value. Plotting the distribution of correlation coefficients between the expression of a single TF and each probe set on the array should result in a normal distribution with a mean centered around zero and maximum and minimum values near +1
60
A. Bhandari et al.
and −1, respectively. A cutoff value for relevant correlation coefficients can be selected (an ideal cutoff value is usually between 0.8 and 0.9), and a list of genes that lie above the cutoff value can be generated for downstream analysis as potential targets for your TF. This process can also be executed with multiple TFs. Lists of transcriptional regulators in your microarray data set to be used in correlation analysis can be generated using probe sets with annotations such as Gene Ontology Biological Process “regulation of transcription, DNA-dependent” (GO ID 0006355) or Gene Ontology Cellular Component “nuclear” (GO ID 0005634). Calculate correlation coefficients for each transcription factor in the list compared to each gene on the array as described above if using only a small list of transcription factors. To perform a largescale correlation analysis of multiple TFs, it is recommended to use numerical analysis software such as MATLAB (http://www.mathworks.com/) or R (http://www.r-project.org/) (see Note 5). 3.1.3. Transcription Factor Binding Site Analysis
We have used two programs to identify transcription factor binding site (TFBS) in candidate target genes. The first is BEARR (Batch extraction and analysis of cis-regulatory regions) (11) and the second is ConSite (10). Both these programs work on the premise that the TF consensus binding site is known (experimentally identified, for example by performing Casting experiments). As a negative control, this analysis is also performed on the genes that are not differentially expressed. BEARR extracts defined sequences upstream and downstream of transcription start sites from several genes in a batch manner and gives a score to each site based on the provided transcription factor consensus binding sequence. However, as this program does not take into account the evolutionary conservation of sites or the region in which the site is, there are a high number of false positive results. Based on our experience, a site having a score over 6 is likely to be detected by EMSA. ConSite overcomes this problem by scoring sites based on across species conservation of the site and the region. The ConSite program has the ability to modify a search, making it stringent or lenient based on your requirements. We typically use 70–75% TFBS cutoff and 70–75% conservation cutoff. Limitations to using ConSite include the inability to extract sequences in batches and the requirement to manually paste the two sequences (usually human and mouse) being compared (see Note 6).
3.2. Competitive Electromobility Shift Assays
Electromobility shift assays (EMSA) are performed to demonstrate the actual binding of the transcription factor to the putative site. A g 32P-labeled probe (binding site) is incubated with or without an unlabelled perfect consensus sequence and the TF or protein lysate. The amount of unlabeled probe required to quench the signal is a measure of the binding affinity of the factor to the sequence.
3 Transcriptional Regulation of Epidermal Barrier Formation 3.2.1. Probe Preparation
61
1. The probe is designed to be ~20–25 bp with the binding site at the center. Single stranded complementary oligonucleotides can be custom ordered for annealing. Dilute the primer stock (2 mg/ml) 1:10 (to a final concentration of 200 ng/ml). Use 250 ng of each oligonucleotide per annealing reaction. Set up the annealing reaction: 1.25 ml sense oligonucleotide, 1.25 ml antisense oligonucleotide, and 17.5 ml water to bring the reaction volume to 20 ml. 2. Heat at 80°C for 10 min in a heat block. Take the metal block of the heat block and leave it on the bench to slowly cool down at RT (overnight if necessary). 3. Kinase reaction: 20 ml of annealing reaction, 5 ml 10× Polynucleotide kinase (PNK) buffer, 2 ml g 32P ATP, 1 ml T4 PNK, and 22 ml water to bring it to a final volume of 50 ml (see Note 7). 4. Incubate at 37°C for 1 h (up to 2 h). 5. During this incubation time make a 10% native polyacrylamide gel using 1.5-mm spacers and a comb with thick wells to set up the gel (see Note 8). 6. Once the incubation is finished add 5 ml of loading dye to each kinase reaction and load on the gel. Use 1× TBE to run the gel. 7. Run the gel for ~2 h at 350 V (the dye should reach the bottom of the gel just above the buffer). 8. Disassemble the gel by lifting the notched plate (this should be done behind a protective screen. Use a Geiger counter frequently to make sure there is no radioactive contamination elsewhere). 9. Cover the gel along with the plate on which it is placed with saran wrap and expose it for 30 s to a film. Develop the film. This film will then be aligned back to the gel and you can visualize the probe. 10. The film can be lined up with the gel to help cut out the band containing the annealed oligos. Cut it into small pieces and transfer them to a tube (the double-stranded annealed oligo moves slower than the single-stranded oligo). 11. Elute the probe by adding 400–500 ml of 50 mM NaCl in TE buffer. Incubate overnight at 4°C. For best results, the probe should be used within 5 days, since it will start decaying. 12. Measure the amount of probe to use for gel shift assay: Make 1:100 dilution of probe. Take 3 ml of each probe and spot it on a filter paper, allowing it to dry. Cut out the filter paper and place in the scintillation vial with 5 ml of scintillation fluid. Cap the vial and measure radioactivity in a scintillation counter. Calculate the amount necessary for each reaction. You will need ~200 counts per second (cps) of labeled probe per gel shift reaction.
62
A. Bhandari et al.
3.2.2. Gel Shift Reaction
1. Prepare a 6% native polyacrylamide gel (or can be commercially bought). It takes over an hour to polymerize. Use 1-mm spacers and a comb with 12 wells and plates to set up the gel (see Note 9). 2. To set up the reaction add the following, mix gently and incubate on ice for 10–20 min (allow 30 s to 1 min interval between samples). Gel shift reaction: 4 ml 5× Gel shift buffer, 4 ml 10 mg/ml BSA, 1 ml 100 mM DTT, 1 ml poly dIdC (1 mg/ml) (optional), X ml, Competitive oligo (Unlabelled consensus sequence) (Use 1–100 molar excess), (10–500 ng)ml DNA binding protein, X ml 200cps probe, and bring the reaction volume to 20 ml with water. 3. Load the samples to the gel without a dye in the same sequence as you set them up. Make sure to run a probe only lane as a negative control with 20 ml loading dye. Run the gel for 2.5 h at 250 V (or till the dye reaches the top of the bottom buffer container). Once the gel is done place the Whatman paper on the gel and remove from the plate. Dry it for ~20–30 min in a gel dryer and expose it to a film. If the signal is weak, the cassette with the film and the gel can be placed in −80°C freezer to develop overnight. 4. To compute relative affinity increasing amounts of cold consensus probe (competitor) is added to the gel shift reaction while keeping the protein or nuclear extract and the radiolabeled probe as constant. The DNA–protein complexes are quantified by cutting out the Whatman paper with the dried gel and measuring the counts using a scintillation counter (the exposed film can be lined up with the Whatman paper to ensure you are cutting the right bands). The percent binding of the TF to the labeled probe in presence of a competitor relative to that in its absence is used to make a competitive binding curve. This binding curve can then be used to estimate the concentration of unlabeled oligo (IC50) needed for 50% of inhibition of binding. The relative affinity can be calculated by dividing the IC50 of a test gene or TFBS to the IC50 of a known gene or TFBS (17).
3.3. Chromatin Immuno--precipitation Assay
Chromatin immunoprecipitation (ChIP) assays can be performed on either undifferentiated or differentiated primary human keratinocytes (Invitrogen) depending on when the TF being tested is expressed. If using undifferentiated keratinocytes, cells should be in the growth phase. The number of cells required per ChIP is to some extent dependent on how good the antibody is. In most cases, 1 × 107 should be enough; however, the number of cells can be reduced to 2 × 106 per ChIP (or 1.5 10 cm plate per ChIP for undifferentiated cells and 1 plate per ChIP for differentiated cells).
3 Transcriptional Regulation of Epidermal Barrier Formation 3.3.1. Staph A Cell Preparation
63
(a) Wash Staph A Cell: (1) Add 10 ml of 1× Dialysis Buffer (DB) to resuspend 1 g of lyophilized Staph A cells (You can use a P1000 Pipetman or allow 30¢ to rehydrate, which helps resuspension); (2) Transfer the resuspended cells to a 15-ml tube and centrifuge at 2,300 × g for 10¢ at 4°C, pour off supernatant; (3) Wash the pellet again with 1× DB, spin and pour off supernatant; (4) Resuspend in 3 ml of PBS, 3% SDS, 10% BME (To 2.25 ml of 1× PBS, add 450 ml of 20% SDS and 300 ml of beta-mercaptoethanol) (5) Boil for 30¢ in fume hood and pellet the cells down at 2,300 × g for 10¢ at RT, pour off supernatant into chemical waste; (6) Wash with 10 ml of 1× DB and pellet cells down at 2,300 × g for 10¢ at RT, pour off supernatant (7) Repeat step 6 and resuspend in 4 ml of 1× DB (8) Make 100-ml aliquots in 0.5-ml tubes, snap-freeze and store at −80°C for indefinite time. (b) Blocking Staph A Cells: The Staph A cells must be blocked before they can be used for ChIP assays. Once blocked they can be stored at 4°C for up to 14 days. (1) Thaw one tube (100 mL each) of Staph A Cells for four IPs. (2) Add 10 ml of each of 10 mg/ml salmon sperm DNA and 10 mg/ml of BSA, mix by pipette. (3) Incubate overnight on a rotating platform at 4°C (You can incubate them on a rotating platform for 2 h at RT or ³3 h at 4°C if necessary) (4) The next day, the Staph A cells are transferred to a 1.5 ml tube and spun at 20,000 × g for 3¢ at 4°C and supernatant discarded. (5) The pellet is resuspended in 1 ml 1× DB, centrifuged at 20,000 × g for 3¢ at 4°C, and supernatant discarded (6) Repeat step 5 (7) Resuspend the pellet in 100 mL of 1× DB with 1 mM PMSF (use 1 ml of 100 mM PMSF).
3.3.2. Day 1: Preparation of Cross-Linked Cells
1. To fix cells, add formaldehyde at a final concentration of 1% directly to the media in plates and rock gently for 10 min at RT. 2. The cross-linking reaction is stopped by adding 2.5 M Glycine at a final concentration of 0.125 M for 5 min. 3. Wash cells with cold 1× PBS at least twice. Keep plates on ice while washing. After the last wash, leave about 1 ml of PBS in the plate. 4. Scrape cells from the plates and transfer the PBS with the cells into a 50-ml tube. You can scrape enough cells for four to six ChIPs in one 50-ml tube. 5. Spin the tube at 590 × g for 10¢. Carefully remove PBS and wash the cells again with 40 ml PBS. Pellet again and aspirate out the supernatant. 6. You can proceed to sonication step immediately or freeze the cells at −80°C for later use.
64
A. Bhandari et al.
3.3.3. Day 2: Sonication, Preclear, and Antibody Incubation
1. The cell pellet is resuspended in SDS lysis buffer (Add fresh protease inhibitors: 1 mM PMSF, 0.1 mg/ml Aprotinin and Leupeptin). We use 400 ml lysis buffer per ChIP and sonication is performed (see Note 10). 2. Take out 30 ml of sonicated material to do a sonication check before you proceed with the ChIP. Bring it up to 50 ml with water. Add 2 ml of 5 M NaCl and place the tube in boiling water for 15¢. Perform a PCR-purification using Sigma PCRpurification kit. Elute in 30 ml water and run on a 2% gel. It is best to get majority of the chromatin to an average of ~500 bp. 3. The samples are then centrifuged at 20,000 × g for 10¢ at 4°C and supernatant transferred to a new tube. All the sonicated material can be combined and realiquoted into 400 ml per tube. This is done to ensure that sonicated chromatin for all ChIPs is the same (the sonicated chromatin can be frozen at this stage). 4. Preclear chromatin by adding 10 ml of blocked/washed Staph A cells (freshly added 1 ml of 100 mM PMSF to 100 ml of blocked Staph A cells) per 1 × 107 cells (Staph A cells are used in place of Protein A agarose beads, as they reduce nonspecific binding, are easier to work with and fairly inexpensive). Incubate on rotating platform for no longer than 15¢ at 4°C. Save the remaining Staph A cells for the next day at 4°C. 5. Microfuge at 20,000 × g for 5¢ at 4°C. Transfer supernatants from all samples to a new tube, measure volume and divide equally into new 1.5-ml eppendorf tubes (can use 2 ml tubes, if you have higher volume). 6. Make twice the volume of ChIP dilution buffer (800 ml of buffer for 400 ml of sonicated chromatin). Add fresh protease inhibitors to the same final concentration as step 1. 7. Add 1–10 mg of antibody (see Note 11) to each sample. Incubate on a rotating platform at 4°C overnight. We use IgG as a negative control, RNA polymerase II as a positive control and one or two different concentration of your test antibody.
3.3.4. Day 3: Washing and Reverse Cross-Linking
1. To the sample add 12.5 ml of Staph A cells (add fresh 1 ml of 100 mM PMSF per 100 ml of blocked and washed Staph A cells). Incubate on rotating platform for 15¢ at RT. 2. Microfuge samples at 20,000 × g for 4¢ at 4°C. Save all the supernatant from IgG IP for “Input”; it will be later used at the reverse cross-linking step. Pour off the supernatant of the other samples (see Note 12). 3. Wash pellets with milliliters of Low-Salt buffer, invert tubes 20–30 times by hand at RT, spin down at 11,000 × g for 4¢ and pour off the supernatant.
3 Transcriptional Regulation of Epidermal Barrier Formation
65
4. Repeat step 3 and then wash twice with High-Salt buffer, once with LiCl and twice with TE in a similar manner (see Note 13). Remove final TE wash buffer by aspirating. 5. Prepare fresh elution buffer (1% SDS, 5 mM NaHCO3) at room temperature (do not put on ice). You will need 100 ml of elution buffer per IP. 6. Add 50 ml of elution buffer to each tube and shake on vortexer for 15¢ at RT. 7. Pellet down the cells at 20,000 × g for 3¢ at RT. Transfer the supernatant to a new tube. Repeat steps 6 and 7 combining the supernatant. 8. Microfuge the combined supernatant again at 20,000 × g for 5¢ to remove any traces of staph A cells. Transfer the supernatant to a new tube. 9. Add 4 ml of 5 M NaCl (0.2 M NaCl final concentration) to each IP. For “10% Input” transfer 10% volume of the “Input” saved from step 2 to a new tube. Bring the volume to 100 ml and add 4 ml of 5 M NaCl. If the volume of the 10% Input is higher than 100 ml, increase the NaCl accordingly. 10. Reverse cross-link all samples and the “10% Input” by placing them at 65°C for 4 h or overnight. The samples can be frozen at −20°C after this stage if necessary, but should not be left for more than a day or two. 3.3.5. Day 4: Colum Purification and PCR Analysis
1. Each sample is column purified using Sigma PCR cleanup kit (Can also use Invitrogen Purelink PCR micro kit, if you want to elute in a much smaller volume – down to 10 ml). Elute in 30–50 ml of water. 2. It is crucial to have primers with good primer efficiency for PCR. Before using your primers for ChIP, do a PCR with dilutions of your input alone (1:10, 1:100, and 1:1,000) to check for primer efficiency. The primers have a good efficiency if band intensity is relative to the dilution of the input sample. Also, it is important to determine what annealing temperature works best for each primer. It is much easier if primers that are being used have the same annealing temperature. 3. Setup PCR in a 20 ml reaction volume: 2 ml of 10× buffer, 1 ml of 2.5 mM dNTP, 0.5 ml of each forward and reverse 10 mM primers, 0.5–1 ml Taq Polymerase (We use one from Denville, but it can be substituted with Taq from other companies) (see Note 14). 4. Controls are very important in a ChIP experiment. For IP, IgG serves as a negative control and RNAPolII as a positive control. For PCR, the positive control primer will be against a region where RNAPolII is sure to bind like the RNAPolII
66
A. Bhandari et al.
promoter region. This same primer however will give no band with IgG IP. Presence of PCR product in IgG IP indicates nonspecific binding. A negative control primer is designed in a region away from the promoter region that should have no RNAPolII binding. We use primer in the DHFR 3¢ UTR region. Negative control primer can also be designed in a region far away from the TF binding site. 3.4. Luciferase Reporter Assays
3.4.1. Plating and Transfection
To test if the validated TF binding site is functional, reporter assays are performed. HaCaT cells are easily transfectable and are therefore ideal for this assay, but primary mouse and human keratinocytes can be used as well. In some cases it is most useful to use a nonkeratinocyte cell line that does not express the TF in question. The cells are transfected with reporter plasmid expressing the luciferase gene downstream of a promoter containing the relevant TF binding site. The TF expression plasmid is cotransfected with the reporter plasmid, cells are collected and luciferase activity measured. A mutation is also made in the TF binding site and this reporter plasmid is used as a negative control. In addition to transfecting the pGL3 Firefly-luciferase reporter plasmid, Renilla luciferase plasmid is cotransfected for normalization purpose. Below is the experimental design with all the controls (Table 1) (see Note 15). 1. A day prior to trasfection plate HaCaT cells grown in low calcium medium (undifferentiated state) at 1 × 105 cells per well in a 12-well plate. This will give a confluency of 80–90%,
Table 1 Experimental design of reporter assay Amount of plasmid per well
Sr.
Plasmids to be transfected
1
pGL3 reporter-Empty + Empty expression plasmid + Renilla plasmid
1.6 mg
2
pGL3 reporter-Promoter + Empty expression plasmid + Renilla plasmid
1.6 mg
3
pGL3 reporter-Promoter + TF Expression plasmida + Renilla plasmid
1.6 mg
4
pGL3 reporter-Promoter-mutb + TF Expression plasmid + Renilla plasmid
1.6 mg
5
pGL3 reporter-Promoter-mut + Empty expression plasmid + Renilla plasmid
1.6 mg
Titration of the expression plasmid is important to determine the best concentration The TFBS in the promoter region is mutated using site-directed mutagenesis kit (Stratagene). If the site being tested is functional, this mutation will result in abolishing or significantly reducing the luciferase activity
a
b
3 Transcriptional Regulation of Epidermal Barrier Formation
67
which is ideal for transfection using Lipofectamine 2000 (Invitrogen). 2. The following day take polypropylene tubes (1.5 ml eppendorf or 15 ml conical tube depending on the volume) and prepare the DNA-lipofectamine mix. 3. For all transfections make a master mix since each transfection is done in triplicate. Calculation must be done for 3.5 wells. 4. To the first tube add plasmid DNA to 100 ml of optiMEM. Make master mix for 3.5 wells. 5. To the second tube add 4 ml of Lipofectamine 2000 to 100 ml of optiMEM. Flick the tube or mix gently and let it sit. Again to prepare master mix for 3.5 wells you will add 14 ml Lipofectamine to 350 ml optiMEM. Let it sit for 5¢. 6. Add the DNA-optiMEM mix to the lipofectamine-optiMEM. Let it sit for 20¢ in the hood. 7. While waiting for the incubation to complete, aspirate the media from the cells and replace it with fresh low calcium media. 8. After 20¢ add the DNA-Lipofectamine complex to each well. To mix, rock the plate back and forth a couple of times and place it back in the 37°C incubator. 9. Change media 12–16 h after transfection. Collect cells for luciferase assay after 48 h. 3.4.2. Luciferase and Renilla Activity Measurement
Luciferase activity is measured by the Dual-Luciferase Reporter Assay System (Promega) where two reporter assays are performed sequentially from the same sample, one from Firefly luciferase and the second from the Renilla luciferase. The latter is used to normalize for the number of cells that are actually being transfected. 1. Cells are collected for luciferase activity measurement 48 h after transfection. Aspirate the media from the cells and add 200 ml of 1× passive lysis buffer (Materials). Freeze the plate with the lysis buffer in −80°C for at least an hour. Freeze thawing helps lyse the cells. Take out the plate and thaw at RT. During this time also start thawing the LARII reagent and the Stop & Glo reagent. 2. Scrape cells from each well and place in a 1.5 ml eppendorf tube. Vortex for 10 s and spin down for 10¢ at 20,000 × g at 4°C. Transfer the supernatant to a new tube (see Note 16). 3. To a luminometer tube add 50 ml of LARII. Add 5 ml of the transfected HaCaT cell lysate. Mix and measure firefly luciferase activity. To the same tube add 50 ml of Stop & Glo reagent and measure Renilla luciferase activity. Background activity is measured by adding the LARII and Stop & Glo in the same order without the lysate.
68
A. Bhandari et al.
4. For analysis, first subtract the background reading from all readings, then, divide the Renilla luciferase measurement from the Firefly luciferase measurement, thus normalizing to the number of cells that are being transfected. These Firefly luciferase measurements are then graphed normalized to cells transfected with empty plasmids. Include standard error measurements based on the triplicate wells readings of each transfection.
4. Notes 1. You can use a certain fold value as a cutoff to make a list more manageable. We use a fold cutoff of 1.5–2.0 depending on the datasets. 2. The array type and species can be selected when loading your data into MeV if present within the software providing you with gene annotation data. If your array information is not present in the software you can also manually load the annotation data that is provided with your array platform. 3. Since it is possible to run numerous K-means cluster analyses quickly, it is helpful to choose a variety of possible clusters and compare the results between the different runs to look for repeated and unique patterns. The number of clusters you decide upon will depend directly on your data. 4. In the case where a single TF is represented with multiple probe sets on an array, one probe set can be randomly selected for comparison to all genes, while the other can be included with the list of all the other probe sets on the array. This allows for a nice positive control, where one would expect the correlation value between the two probe sets of the same TF to be close to 1. 5. The Pearson’s product moment correlation coefficient can also be used for global gene expression analysis of multiple arrays. For example, it can be used to compare the global similarity in gene expression between two tissue types at a specific time or stage in development. In this case the correlation coefficients are calculated across all probe sets for each pair of sample types, and then averaged to create a single correlation coefficient between the pair of tissues. Another useful application for correlation is to compare the expression level of a transcriptional regulator of interest to a potential target gene of interest across many arrays. From each array plot the log2 expression value of the transcriptional regulator versus the log2 expression level of the target gene of interest, and from the plot that was generated calculate the Pearson’s product moment correlation coefficient. Again, correlation coefficients
3 Transcriptional Regulation of Epidermal Barrier Formation
69
closer to 1 have a high correlation and those closer to 0 have a low correlation. 6. In case the TF of interest is not in the pull down list, you can upload the Position Weight Matrix or Raw counts matrix of your TF using the “User defined profile”. It is important to note here that the example given to the right of the sequence of ACTG is incorrect. It is ACGT instead. 7. To label the probe g 32P ATP is used, which is a high energy isotope with a short half life of 14 days. It is at the labeling step that maximum care should be taken, keeping the tube with radioactive 32P in acrylic case and using an acrylic shield for protection. Both the labeling reaction and running the gel should be done behind acrylic shields, especially when preparing the probe. Be extremely careful while working with radioactive 32P using a Geiger counter constantly to avoid contaminating any part of your clothing and other areas in the lab. Follow all rules specified by your university or institution. 8. It is important to use the long vertical gel apparatus at this step, so that there is enough separation in case annealing is incomplete and you can cleanly excise the gel band with the annealed probe. 9. Commercially available gels (Invitrogen-Native PAGE Novex 4–16% Bis–Tris gels; Cat. No.BN1002BOX) can be used at this stage to get cleaner results. 10. Optimizing sonication is important for a successful ChIP experiment, and to get consistent results. It is best to do this with the exact number of cells with which you want to do the ChIP. Using the Bioruptor, 1 × 107 keratinocytes in 800 ml lysis buffer (enough for two ChIPs, with a good antibody) are sonicated to average 500 bp by setting it on High and sonicating 15 s on and 15 s off for 30¢. 11. The amount of antibody required for ChIP varies widely and should be tested empirically. Make sure to run the positive and negative controls to ensure that there is no nonspecific binding. 12. Supernatant of the IgG IP is used as input, since that is the total amount of DNA that you are starting out with and there should be no DNA binding to IgG in a specific manner. 13. The LiCl wash buffer has the highest amount of salt and hence is the most stringent. Washing once with this buffer is usually enough. However, if you are getting nonspecific binding with IgG, wash the Staph A cells twice with LiCl buffer. 14. You can also perform quantitative RT-PCR and calculate quantitative levels of enrichment in antibody ChIP as compared to IgG as a percentage of input.
70
A. Bhandari et al.
15. As a positive control, you can use pGL3-promoter vector (Promega; Cat. No. E1761). 16. The lysates can be stored at −20°C for a couple of days, although the luciferase activity may reduce slightly.
Acknowledgments This work was supported by TRDRP dissertation award 17DT0192 (to A.B.), NIH-NLM Biomedical Informatics Training grant 5T15LM007743 (to MS) and NIH grant AR44882 (to BA). The authors acknowledge the contributions of other laboratory members to the approaches described in this chapter. We thank Amy Soto and Suman Verma for reading the manuscript. We acknowledge the Expression Analysis Core at UC Davis for ChIP training. References 1. Nickoloff, B. J. (2006) Keratinocytes regain momentum as instigators of cutaneous inflammation. Trends Mol Med 12, 102–6. 2. Blanpain, C., and Fuchs, E. (2009) Epidermal homeostasis: a balancing act of stem cells in the skin. Nat Rev Mol Cell Biol 10, 207–17. 3. Koster, M. I., and Roop, D. R. (2007) Mechanisms regulating epithelial stratification. Annu Rev Cell Dev Biol 23, 93–113. 4. Long, A. D., Mangalam, H. J., Chan, B. Y., Tolleri, L., Hatfield, G. W., and Baldi, P. (2001) Improved statistical inference from DNA microarray data using analysis of variance and a Bayesian statistical framework. Analysis of global gene expression in Escherichia coli K12. J Biol Chem 276, 19937–44. 5. Yu, Z., Mannik, J., Soto, A., Lin, K. K., and Andersen, B. (2009) The epidermal differentiation-associated Grainyhead gene Get1/Grhl3 also regulates urothelial differentiation. Embo J 28, 1890–903. 6. Lin, K. K., Chudova, D., Hatfield, G. W., Smyth, P., and Andersen, B. (2004) Identification of hair cycle-associated genes from time-course gene expression profile data by using replicate variance. Proc Natl Acad Sci USA 101, 15955–60. 7. Lin, K. K., Kumar, V., Geyfman, M., Chudova, D., Ihler, A. T., Smyth, P., Paus, R., Takahashi, J. S., and Andersen, B. (2009) Circadian clock genes contribute to the regulation of hair follicle cycling. PLoS Genet 5, e1000573.
8. Saeed, A. I., Bhagabati, N. K., Braisted, J. C., Liang, W., Sharov, V., Howe, E. A., Li, J., Thiagarajan, M., White, J. A., and Quackenbush, J. (2006) TM4 microarray software suite. Methods Enzymol 411, 134–93. 9. Saeed, A. I., Sharov, V., White, J., Li, J., Liang, W., Bhagabati, N., Braisted, J., Klapa, M., Currier, T., Thiagarajan, M., Sturn, A., Snuffin, M., Rezantsev, A., Popov, D., Ryltsov, A., Kostukovich, E., Borisovsky, I., Liu, Z., Vinsavich, A., Trush, V., and Quackenbush, J. (2003) TM4: a free, opensource system for microarray data management and analysis. Biotechniques 34, 374–8. 10. Sandelin, A., Wasserman, W. W., and Lenhard, B. (2004) ConSite: web-based prediction of regulatory elements using cross-species comparison. Nucleic Acids Res 32, W249–52. 11. Vega, V. B., Bangarusamy, D. K., Miller, L. D., Liu, E. T., and Lin, C. Y. (2004) BEARR: Batch Extraction and Analysis of cis-Regulatory Regions. Nucleic Acids Res 32, W257–60. 12. Chi, X., Zhang, S. X., Yu, W., DeMayo, F. J., Rosenberg, S. M., and Schwartz, R. J. (2003) Expression of Nkx2-5-GFP bacterial artificial chromosome transgenic mice closely resembles endogenous Nkx2-5 gene activity. Genesis 35, 220–6. 13. Decker, T., Pasca di Magliano, M., McManus, S., Sun, Q., Bonifer, C., Tagoh, H., and Busslinger, M. (2009) Stepwise activation of enhancer and promoter regions of the B cell
3 Transcriptional Regulation of Epidermal Barrier Formation commitment gene Pax5 in early lymphopoiesis. Immunity 30, 508–20. 14. Wells, J., and Dai, X. Using siRNA knockdown in HaCaT cells to study transcriptional control of epidermal proliferation potential. Methods Mol Biol 585, 107–25. 15. Choe, S. E., Boutros, M., Michelson, A. M., Church, G. M., and Halfon, M. S. (2005) Preferred analysis methods for Affymetrix GeneChips revealed by a wholly defined control dataset. Genome Biol 6, R16. 16. Wang, N., Lin, K. K., Lu, Z., Lam, K. S., Newton, R., Xu, X., Yu, Z., Gill, G. N., and
71
Andersen, B. (2007) The LIM-only factor LMO4 regulates expression of the BMP7 gene through an HDAC2-dependent mechanism, and controls cell proliferation and apoptosis of mammary epithelial cells. Oncogene 26, 6431–41. 17. Yu, Z., Lin, K. K., Bhandari, A., Spencer, J. A., Xu, X., Wang, N., Lu, Z., Gill, G. N., Roop, D. R., Wertz, P., and Andersen, B. (2006) The Grainyhead-like epithelial transactivator Get-1/Grhl3 regulates epidermal terminal differentiation and interacts functionally with LMO4. Dev Biol 299, 122–36.
wwwwwwwwwwwww
Chapter 4 Epidermal Permeability Barrier Measurement in Mammalian Skin Arup Kumar Indra and Mark Leid Abstract A defective skin epidermal permeability barrier (EPB) is responsible for a high mortality rate in premature infants and is an important risk factor in inflammatory skin diseases such as eczema. We report here fast and accurate methods for measurement of EPB in animal models or in human patients using simple techniques that monitor diffusion of dyes (X-Gal or Lucifer Yellow) through the upper epidermis and measure transepidermal water loss (TEWL) resulting from a defective skin barrier. Accurate diagnosis and early detection of EPB defects in human patients are critical for effective treatment of certain classes of inflammatory skin diseases. Key words: Skin, EPB, TEWL, X-Gal diffusion, Lucifer yellow
1. Introduction The skin, which consists of the epidermis and underlying dermis, is a very attractive tissue to study the in vivo functions of genes that regulate the expression of proteins involved in the control of cellular proliferation and differentiation. It is the largest organ in the body comprising approximately 10% of body weight and protects the body from dehydration and environmental insults through establishment of the protective epidermal permeability barrier (EPB). During embryonic development, the ectodermal cell layer covering the body develops into a stratified epidermis that is essential at birth, when the organism confronts the arid and toxic postnatal environments. Keratinocytes, an ectodermally derived cell type, form the proliferative, basal layer of the epidermis. Keratinocytes periodically withdraw from the cell cycle and
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_4, © Springer Science+Business Media, LLC 2011
73
74
A.K. Indra and M. Leid
commit to terminal differentiation while migrating through the suprabasal layers. The outermost layer of the skin (stratum corneum) is composed of mechanically tough, dead, cornified cells (squames), which develop as a result of a complex terminal differentiation program and provide vital physical and permeability barriers to vertebrates (1, 2). The epidermal barrier, which is composed of the cornified envelope, the cornified lipid envelope, and extracellular lipids, is formed in a highly reproducible pattern during late stages of embryogenesis (3–5). Formation of the EPB requires the delivery of lipids and proteins, which are contained in lamellar granules (keratinosomes) present in keratinocytes of the granular layer, to the stratum corneum interstices, as well as the formation of highmolecular-weight polymers through the cross-linking of corneocyte envelope proteins (loricrin, involucrin, filagrin, and other peptides) and packing of corneocytes by corneodesmosomes. Postnatally, the epidermal barrier is maintained by a complex epidermal differentiation program, which results in the constant production of the cellular and lipid components of the barrier (reviewed in (6)). Epidermal homeostasis relies on a tightly regulated balance between keratinocyte proliferation and differentiation; the alteration of this leads to various skin diseases (7, 8). A defective EPB accounts for high mortality rate (>40%) in premature infants, and is an important feature of many inflammatory skin diseases such as eczema and psoriasis, affecting nearly 10% of the world population. Compromised barrier function in prematurely born infants or in patients with certain inherited skin diseases often results in dehydration and increased susceptibility to infections (9–12). Defects in protective skin barrier function(s) and minor lesions of the skin result in an increased transepidermal water loss (TEWL), altered skin pH, and hydration. Hardman et al. (13) have developed an elegant and qualitative, whole-mount assay for skin permeability and showed that the assay measures the first stage of barrier formation. The barrier forms first at distinct epidermal sites and then spreads across the epidermis as a moving front. It has been demonstrated that late stages of cornified envelope assembly accompany movement of the front. Hence the whole-mount permeability assays record developmental acquisition of a known, essential component of the adult barrier. The authenticity and utility of the assays was further validated by monitoring barrier formation after hormonal treatment (maternal glucocorticoid therapy) known to accelerate fetal barrier development. Similar patterned skin barrier acquisition in additional species confirmed that patterned change is probably a ubiquitous mode of epidermal differentiative change during mammalian development (13). Several laboratories, including us, have used those assays to
4 EPB Monitoring in Mammals
75
demonstrate that barrier formation is indeed highly patterned during development (14–16). Dye diffusion assays provide a rapid and cost-effective means of assessing EPB function in rodent models without the need of investing in additional equipment. Dye diffusion assays are most useful when the magnitude of EPB disruption is large. However, owing to the enhanced sensitivity of the technique, measurement of TEWL has become the preferred means for quantitative assessment EPB function, particularly when EPB dysfunction is subtle, as is likely the case in human dermatological diseases. For example, both Ctip2ep−/− and Gata-3ep−/− (conditional deletion of transcription factors Ctip2 and GATA-3, respectively, in epidermis) mice are able to exclude X-gal dye at ~E18.0 (before birth), suggesting normal EPB function. However, both lines of mice exhibit a significantly increased rate of TEWL at that stage compared to the control littermates (16, 17). In these cases, dye diffusion assays and TEWL measurements appear at odds simply as a result of the differential sensitivities of these two techniques. In the present article, we briefly describe the various methodologies of determining EPB formation using mouse as a model system.
2. Materials 1. ICR or CD1 mice can be purchased from Jackson Laboratory (http://jaxmice.jax.org/) or Charles River Laboratories (http://ftp.criver.com). Ctip2L2/L2 and Brg1L2/L2 mice can be obtained from Mark Leid (Mark.Leid@oregonstate. edu) and Pierre Chambon (
[email protected]), respectively. 2. PBS (Dulbecco’s phosphate-buffered saline) (Cat. No. D5652), potassium ferricyanide (Cat. No. 244023), potassium ferrocyanide (Cat. No. P3289), Lucifer Yellow CH (Cat. No. L0259), and toluidine blue O (Cat. No. T3260). 3. Hematoxylin solution A (Cat. No. 32897) are from SigmaAldrich; Diamidino-2-phenylindole dihydrochloride (DAPI) (Cat. No. 236276) and X-Gal (5-bromo-4-chloro-3-indolylb-d-galactopyranoside) (Cat. No. 745740) are from Roche Diagnostics. 4. Tewameter® (Acaderm Inc., USA). 5. Zeiss Axio Scope fluorescent microscope with X-cite120 for four channels (green, red, blue, Hoecht) attached with an Axiocam monochromatic camera running with Axiovision 4.7.1 system software.
76
A.K. Indra and M. Leid
3. Methods The X-Gal and Lucifer Yellow diffusion assays (described below), as well as the measurement of the TEWL, are routinely performed in laboratories around the world to detect skin barrier defects in mouse models. Using these techniques, we have successfully established the roles of a transcriptional regulatory protein, Ctip2, and a member of the chromatin remodeling complex, Brg1, in epidermal terminal differentiation and skin barrier formation (14–16). Diagnostic TEWL is also routinely performed by dermatologists on patients with skin disorders, such as eczema and psoriasis, and the technique is also useful for evaluation of the efficacy of topical drug treatment on barrier recovery in the skin of these patients. 3.1. Embryos
1. ICR or CDI mice were time-mated within a 5-h mating window (permeability assays, and TEWL and EM studies). 2. The midpoint of the mating window designated gestational age zero. 3. Embryos were derived from random matings within a 4-day period and categorized according to gross morphology and barrier status. 4. Fetal gestational age was calculated (see Note 1), and pregnant dams were sacrificed by CO2 asphyxiation prior to recovery of the embryos.
3.2. Skin Permeability Assay
Assay 1: This assay depends on barrier-dependent access of 5-bromo-4-chloro-3-indolyl-b, d-galactopyranoside (X-gal) to untreated skin. 1. Unfixed, untreated, freshly isolated E18.5 embryos were rinsed in phosphate-buffered saline (PBS) and dried briefly. 2. Embryos were immersed in standard X-gal reaction mix (18), with pH adjusted to 4.5. 3. Embryos were incubated at 37°C for 8–10 h, washed in PBS for 1–2 min and photographed. 4. EPB defect was assessed functionally depending on the diffusion of X-Gal dye into the embryos (Fig. 1a, see Notes 2 and 5). Assay 2: This assay modifies skin to permit barrier-dependent penetration by histological dyes such as toluidine blue or hematoxylin. 1. Unfixed, untreated embryos were incubated for 1–5 min in methanol and rinsed in PBS.
4 EPB Monitoring in Mammals
77
Fig. 1. Epidermal permeability barrier defects in Ctip2−/− (germline deletion of gene encoding transcriptional regulator Ctip2), and CTIP2ep−/− (selectively lacking the gene for Ctip2 in epidermal keratinocytes) fetuses. (a) X-gal diffusion assay performed on Ctip2+/+, CTIP2−/−, CTIP2L2/L2 and CTIP2ep−/− fetuses at E17.5 and E18.5 as indicated. (b) TEWL measurements from dorsal and ventral skin of CTIP2+/+, CTIP2−/−, CTIP2L2/L2 and CTIP2ep−/− mice at E17.5 and E18.5. (Plotted are mean measurements of three independent mice per genotype ± S.E.M.). * – p < 0.05, # – not statistically significant. Ctip2+/+ (wild type littermates) and CTIP2L2/L2 (with floxed Ctip2 allele) fetuses were used as controls. (Courtesy of Arup Kumar Indra and Mark Leid; adapted with permission from Golonzhka et al., 2009).
2. Embryos were incubated in 0.5% hematoxylin or 0.1% toluidine blue. 3. Embryos were embedded in agarose and photographed using a Zeiss Stemi SV11 microscope with transmitted and surface illumination. 4. Scanned images were processed with Adobe Photoshop and the agarose background was removed. 5. EPB defects were evaluated depending on degree of dye penetration (see Notes 3–5). 3.3. In Vivo Transdermal Absorption of the Fluorescent Dye Lucifer Yellow
1. E18.5 fetuses/embryos were restrained in Petri dishes with their backs in contact with 1 mM Lucifer Yellow (Sigma) in PBS (pH 7.4) at 37°C, as described (19). 2. After 1 h of incubation, fetuses were sacrificed, frozen, and cryosectioned dorsoventrally at a thickness of 5 mm. 3. Sections were counterstained with 10 mg/ml DAPI (4¢,6-diamidino-2-phenylindole dihydrochloride), dehydrated, and mounted with DPX mounting medium.
78
A.K. Indra and M. Leid
Fig. 2. Impaired skin barrier function in Brg1ep−/−(c) mutant fetuses selectively lacking the gene for Brg1 (a member of the mammalian chromatin remodeling protein complex). (A) Gross morphology of E18.5 Brg1L2/L2 (control; a) and Brg1ep–/–(c) fetuses (B) In vivo Lucifer yellow diffusion in (a) Brg1L2/L2 and (b) Brg1ep–/–(c) E18.5 fetuses. (Courtesy of Arup Kumar Indra and Pierre Chambon; reproduced and adapted with permission from Indra et al., 2005b).
Fig. 3. Measurement of TEWL using a Tewameter®. The Tewameter® probe is touched on the skin surface and the readings are noted on the screen of a multiprobe adaptor [21] system.
4. Sections were analyzed by fluorescence microscopy and photographed using a Zeiss fluorescent microscope (see Fig. 2 and Note 6). 3.4. Assessment of TEWL
TEWL is the most important and sensitive parameter for evaluating the integrity of the EPB in all conditions in laboratory mice, rats, and humans (see Notes 7 and 8). 1. Dorsal and ventral skin TEWL were determined on E17.5 and E18.5 embryos, neonatal or adult mice (after shaving to remove the hair) with a Tewameter® (Acaderm Inc., USA) equipped with a TEWL probe (see Fig. 3).
4 EPB Monitoring in Mammals
79
2. Embryos, neonatal or adult mice were restrained on their dorsal or ventral side and TEWL on the skin surface was directly measured with the Tewameter®, which is the most accepted and best-selling TEWL measurement device worldwide [(Acaderm Inc., USA); see Note 9]. 3. Mean values of six measurements per animal were determined. 4. Data were expressed in g/m2-h, as means ± s.e.m. from three to four animals [(Figs. 1b and 3) see Note 10]. 3.5. Summary
In summary, assessment of EPB function is an extremely important technique for scientists, clinicians, and the cosmetological industry. It seems likely that the technology behind EPB measurement will continue to evolve and provide more sophisticated methods for the enhanced detection of compromised EPB function in mice and man. In addition, further research in this area will identify the biophysical properties of the skin barrier that regulate the relative permeabilities of small molecules, infectious agents, water, and gases across this surface, all of which are still poorly understood.
4. Notes 1. Estimated fetal gestational age (EGA, #98) was calculated from the time designated zero. For example, 16 days/5 h after time zero was termed E16.5 or 16.5 days EGA. 2. X-Gal diffusion assay depends on barrier-dependent access of 5-bromo-4-chloro-3-indolyl-b, d-galactopyranoside (X-gal) to untreated skin. At low pH, skin contains abundant endogenous b-galactosidase-like activity, which cleaves X-gal to produce a colored precipitate (13–16). 3. The basis of the skin modification during dye diffusion (Assay 2) is unknown but is likely to involve extraction of polar lipid (20). 4. Skin permeability Assays 1 and 2 gave the same staining pattern, inferring that those techniques measure similar skin characteristics. Comparisons were performed on sagittal sides of a single embryo to eliminate variation arising from differences in developmental stage. 5. The EPB begins to develop on the dorsal aspect of developing fetuses at approximately E16.5 and spreads ventrally, resulting in complete dye impermeability by E17.5 (13–16). Mice with impaired EPB development and/or function display increased staining, which was assessed by either assay 1 or 2.
80
A.K. Indra and M. Leid
6. When visualized by fluorescence microscopy at E18.5 the fluorescent dye Lucifer yellow labels the upper layer of the stratum corneum of control fetuses (19). In contrast, diffused staining was seen throughout the stratum corneum which extends down to the dermis and hypodermis of mutant fetuses with compromised EPB function (see Fig. 2 (15, 19)). 7. The skin constantly loses water in form of vapor and that TEWL is accelerated by minor skin lesions and other factors that disrupt the EPB function. This has rendered the technique of measuring TEWL invaluable for both dermatological and cosmetological applications. In addition, TEWL measurment is widely used in occupational medicine, medical consultancy, observation of the newborn, and the food industry. 8. TEWL measuring principle and methodology: The measurement of water evaporation is based on the diffusion principle in an open chamber. The open chamber measurement method is the only method available for assessing TEWL continuously, and in a noninvasive manner. Stable TEWL measurements can be made rapidly and with ease using a variety of instruments on the market such as a Tewameter® (Acaderm Inc., USA) or a vaporimeter (Delphin Inc., USA). 9. The TEWL probe of the Tewameter® consists of two pairs of sensors to measure the humidity and temperature gradients in two different spacings. On the basis of the resulting humidity and temperature gradients, the TEWL is automatically calculated and shown. 10. E18.5 control fetuses with normal skin barrier function display a TEWL value in the range of 1–2 g/m2-h on the dorsal side, and 5–10 g/m2-h on the ventral side using a Tewameter® or a vaporimeter (14–16). The difference in TEWL value (between the dorsal and ventral side) could be attributed to patterned barrier acquisition during embryogenesis when barrier formation is completed earlier on the dorsal side compared to the ventral side (13). E16.5–E17.5 fetuses in the early developmental stages of skin maturation and barrier formation, as well as E18.5 fetuses with an impaired EPB are expected to display significant higher values both on the dorsal and ventral side under identical conditions.
Acknowledgments These studies were supported by grant AR056008 (AI) from the National Institutes of Health and by a NIEHS Center grant (ES00210) to the Oregon State University Environmental Health
4 EPB Monitoring in Mammals
81
Sciences Center. Acknowledgments are due to the copyright permission department of Development and Journal of Investigative Dermatology for allowing the reproduction and adaptation from previously published papers. References 1. Byrne, C., Hardman, M. and Nield, K. (2003) Covering the limb – formation of the integument. J Anat, 202, 113–123. 2. Kalinin, A.E., Kajava, A.V. and Steinert, P.M. (2002) Epithelial barrier function: assembly and structural features of the cornified cell envelope. Bioessays, 24, 789–800. 3. Koch, P.J., Zhou, Z. and Roop, D.R. (2004) In P.M. Elias, a. K. R. F. (ed.), Skin Barrier. Marcel Dekker, Inc., New York, pp. 97–110. 4. Elias, P.M. and Feingold, K.R. (1992) Lipids and the epidermal water barrier: metabolism, regulation, and pathophysiology. Semin Dermatol, 11, 176–182. 5. Koster, M.I. (2009) Making an epidermis. Ann N Y Acad Sci, 1170, 7–10. 6. Segre, J. (2003) Complex redundancy to build a simple epidermal permeability barrier. Curr Opin Cell Biol, 15, 776–782. 7. Watt, F.M. (2000) Epidermal stem cells as targets for gene transfer. Hum Gene Ther, 11, 2261–2266. 8. Watt, F.M. and Hogan, B.L. (2000) Out of Eden: stem cells and their niches. Science, 287, 1427–1430. 9. Cartlidge, P. (2000) The epidermal barrier. Semin Neonatol, 5, 273–280. 10. Elias, P.M., Hatano, Y. and Williams, M.L. (2008) Basis for the barrier abnormality in atopic dermatitis: outside-inside-outside pathogenic mechanisms. J Allergy Clin Immunol, 121, 1337–1343. 11. Elias, P.M. and Steinhoff, M. (2008) “Outsideto-inside” (and now back to “outside”) pathogenic mechanisms in atopic dermatitis. J Invest Dermatol, 128, 1067–1070. 12. Elias, P.M., Williams, M.L., Holleran, W.M., Jiang, Y.J. and Schmuth, M. (2008) Pathogenesis of permeability barrier abnormalities in the ichthyoses: inherited disorders of lipid metabolism. J Lipid Res, 49, 697–714. 13. Hardman, M.J., Sisi, P., Banbury, D.N. and Byrne, C. (1998) Patterned acquisition of skin barrier function during development. Development, 125, 1541–1552.
14. Indra, A.K., Mohan, W.S., 2nd, Frontini, M., Scheer, E., Messaddeq, N., Metzger, D. and Tora, L. (2005a) TAF10 is required for the establishment of skin barrier function in foetal, but not in adult mouse epidermis. Dev Biol, 285, 28–37. 15. Indra, A.K., Dupe, V., Bornert, J.M., Messaddeq, N., Yaniv, M., Mark, M., Chambon, P. and Metzger, D. (2005b) Temporally controlled targeted somatic mutagenesis in embryonic surface ectoderm and fetal epidermal keratinocytes unveils two distinct developmental functions of BRG1 in limb morphogenesis and skin barrier formation. Development, 132, 4533–4544. 16. Golonzhka, O., Liang, X., Messaddeq, N., Bornert, J.M., Campbell, A.L., Metzger, D., Chambon, P., Ganguli-Indra, G., Leid, M. and Indra, A.K. (2009) Dual role of COUP-TFinteracting protein 2 in epidermal homeostasis and permeability barrier formation. J Invest Dermatol, 129, 1459–1470. 17. de Guzman Strong, C., Wertz, P.W., Wang, C., Yang, F., Meltzer, P.S., Andl, T., Millar, S.E., Ho, I.C., Pai, S.Y. and Segre, J.A. (2006) Lipid defect underlies selective skin barrier impairment of an epidermal-specific deletion of Gata-3. J Cell Biol, 175, 661–670. 18. Bonnerol, C. and Nicolas, J.-F. (1994) In P.M. Wasserman, a. M. L. D. (ed.), Guide to techniques in mouse development. Academic Press, San Diego, pp. 451–469. 19. Matsuki, M., Yamashita, F., Ishida-Yamamoto, A., Yamada, K., Kinoshita, C., Fushiki, S., Ueda, E., Morishima, Y., Tabata, K., Yasuno, H. et al. (1998) Defective stratum corneum and early neonatal death in mice lacking the gene for transglutaminase 1 (keratinocyte transglutaminase). Proc Natl Acad Sci USA, 95, 1044–1049. 20. Wertz, P.W. and Downing, D.T. (1987) Covalently bound omega-hydroxyacylsphingosine in the stratum corneum. Biochim Biophys Acta, 917, 108–111. 21. Steiner, M., Aikman-Greed, S. and Dick F.D. (2010). Side-by-side comparison of open chamber (TM 300) and closed chamber (Vapometer) TEWL. Skin Research and Technology. 16, 489–490.
wwwwwwwwwwwww
Chapter 5 Assessment of Permeability Barriers to Macromolecules in the Rodent Endometrium at the Onset of Implantation Brent M. Bany and G. Scot Hamilton Abstract In rodents, embryo implantation is an invasive process, which begins with its attachment to the uterine wall and culminates in the formation of the definitive placenta several days later. It is critical that the endometrium provide a supportive environment for the implanting embryo during this process, as the placenta is not yet established. The concept of changing permeability barriers to macromolecules between different extracellular compartments in the rodent uterus at the onset of implantation has been established. This chapter provides protocols that can be used to assess this changing permeability barrier and the associated redistribution of macromolecules during the early phases of implantation in rodents. An increased permeability of the endometrial vasculature to plasma proteins occurs in areas adjacent to the implanting blastocyst. In addition, alterations in the extracellular matrix enhance the accumulation of fluid and extravasated macromolecules. We describe several protocols proven to be effective in studying and quantifying early vascular and extravascular responses to natural and artificial “implantation stimuli.” The first three protocols represent qualitative and quantitative methods to assess the early endometrial “vascular permeability” response. On the contrary, the fourth protocol addresses the onset of decidualization and the arising permeability barrier, which restricts the movement of macromolecules through the extracellular space. This barrier is believed to provide transient protection for the implanting embryo against potentially harmful maternal serum proteins. This protocol describes assessment of resistance of the primary decidual zone to the movement of macromolecules across the compartments of the extracellular space. Key words: Uterus, Endometrium, Decidualization, Embryo implantation, Vascular permeability
1. Introduction Highly restrictive barriers to the passage of macromolecules have long been accepted for brain and testis. These barriers involve various intercellular junctions involving endothelial cells, astrocytes, and Sertoli cells (1, 2). Ordinarily, the permeability of these
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_5, © Springer Science+Business Media, LLC 2011
83
84
B.M. Bany and G.S. Hamilton
barriers is only increased as a result of pathological conditions. By contrast, the uterus of many mammals contains a barrier to macromolecules which is influenced by physiological processes including embryo implantation. Undoubtedly, the most studied permeability barrier to macromolecules in the rodent uterus during implantation is that between the vascular blood space and the extracellular tissue. At the onset of implantation, there is a dramatic increase in the endometrial vascular permeability barrier in areas immediately adjacent to the implanting blastocyst. This was first described by the work of Psychoyos over 50 years ago (3) and is the earliest macroscopically observable sign of the onset of implantation in rodents (4, 5) as well as some nonrodent species (6–10). A similar response may also be elicited by applying a suitable artificial “deciduogenic” stimulus to the uterine lumen in place of an implanting embryo. These artificial stimuli require specific temporal and concentration-dependent exposure to hormonal fluctuations (“sensitization”) very similar to those which precede embryo implantation (11–14). Previous studies have indicated that the “permeability response” comprises a number of vascular and extravascular responses including: increased permeability of the endometrial vessels, increased uterine blood flow, and increased extracellular fluid volume (15–22). The permeability of the endometrial vessels can be assessed qualitatively through the accumulation of albumin-dye (Evans Blue) complexes in the uterine wall after intravenous injection of the dye (see Fig. 1). The dye complexes cannot leave the uterine vessels in areas which have not exhibited the “vascular permeability” response. By the same mechanism, the extent of extravasation can be quantitatively evaluated using 125I-labeled albumin or other proteins (19). Finally, tissue distribution of iv-injected 51Cr-EDTA has also been used to calculate changes in the size of the extravascular space (15, 16). Much like the use of Evans Blue dye, MRI has been used to demonstrate sites of embryo implantation and enhanced endometrial fluid accumulation (17, 18, 23). The first three protocols described in this chapter provide the basis for qualitative and quantitative assessment uterine vascular permeability (Vp) and extracellular fluid volume (ECFV) changes observed in rodents at the onset of implantation or after artificial deciduogenic stimulation. During the early phase of implantation, another permeability barrier comes into play. The endometrial tissue immediately surrounding the implanting embryo begins to differentiate into decidual tissue just after the onset of implantation. This results in the formation of a distinct zone of tissue within 24 h after the onset of implantation in rodents. The tissue, called the primary decidual zone, has the unique characteristic of being quite avascular and the extracellular space is highly impermeable to macromolecules (24–28). This impermeability is due to the presence of
5 Assessment of Permeability Barriers to Macromolecules in the Rodent Endometrium¼
85
Fig. 1. Increased uterine vascular permeability in the mouse uterus as determined by the Evans Blue dye method. (a) Representative pregnant mouse uterus on Day 4.5 of pregnancy showing areas of bluing in segments of the uterus containing an implanting conceptus. (b) Representative mouse uterus where one horn was exposed to an artificial deciduogenic stimulus while the contralateral horn served as a control. Ovariectomized mice were hormonally sensitized and artificially induced (intraluminal injection of sesame oil) to undergo decidualization as previously described [27]. The Evans Blue bluing reaction was determined 6 h after application of the artificial deciduogenic stimulus (sesame oil).
numerous cell–cell junctions such as adherens, tight, and gap junctions (26, 28–30) and may also be related to changes in the extracellular matrix. The fourth protocol described in this chapter is used to visualize this relative impermeability of the primary decidual zone to macromolecules. Although some models of artificially inducing decidualization do not result in the formation of a similar primary decidual zone (28), we have recently shown that other models do (27) using similar methods.
2. Materials (See Notes 1–3) 2.1. General Solutions and Materials
1. Normal saline (0.9%): 156 mM NaCl. Autoclave sterilize (see Note 4). 2. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4, 1.47 mM KH2PO4, pH 7.4 (see Note 4). 3. Tris-buffered saline (TBS): 150 mM NaCl, 50 mM Tris, pH 7.4 (see Note 4). 4. Latex gloves.
86
B.M. Bany and G.S. Hamilton
2.2. Animals
1. Animals: Keep mice or rats under controlled light conditions (lights on 6 a.m. to 8 p.m.) with free access to food and water. Obtain pregnancies by placing females with males and the day a vaginal plug (mouse) or sperm in vaginal smears (rats) is found at 9 a.m. is considered Day 0.5 of pregnancy. By this convention, implantation in mice begins between approximately 12 and 5 a.m. in the dark phase between Days 3.5 and 4.5 of pregnancy (4) while in rats begins later on day 4.5 of pregnancy (3). 2. Ketamine and xylazine: use at a dose of 80 mg/g and 15 mg/g, respectively (Henry Schein, Melville, NY). 3. Dissection instruments: forceps, fine plus coarse scissors, and scalpel handle with blades, surgical silk sutures.
2.3. Qualitative Assessment of Vp
1. Evans Blue dye solution (1% w/v) is made by dissolving 0.5 g of Evans Blue dye powder in 50 ml of normal saline. The solution is filter-sterilized using a syringe and 0.45-mm filters (Millipore) into sterile 1.5-ml tubes (see Note 5). 2. 1-ml syringe (Fisher Scientific). 3. 21 gauge needles (Fisher Scientific).
2.4. Quantitative Assessment of Vp
1. 125I-labeled albumin: Iodinate 50 mg of bovine albumin (Roche) using IODO-GEN as recommended by the manufacturer (Thermo Scientific) using 0.7 mCi Na125I (Perken Elmer, Waltham, MA). Subject the reaction mixture to Biogel P6 (Bio-RAD) gel exclusion chromatography (10 ml) to separate the labeled protein from free 125I (see Note 6). 2. Radioactivity waste bags and containers. 3. Freshly heparinized 1-ml syringe with 21 gauge needles, 1.5-ml microtubes, centrifuge capable of 10,000 × g with microtube rotor. 4. Pipettes and tips. 5. 1.5-ml microtubes. 6. Gamma counter and sample vials.
2.5. Quantitative Assessment of ECFV
1. 51Cr-EDTA (PerkinElmer). 2. 0.9% NaCl. 3. 1-ml syringe (Fisher Scientific). 4. 21 gauge needles (Fisher Scientific). 5. 1.5-ml microtubes. 6. Pipettes and tips. 7. Gamma counter and sample vials. 8. Radioactivity waste bags and containers.
5 Assessment of Permeability Barriers to Macromolecules in the Rodent Endometrium¼
2.6. Visualizing PDZ Impermeability to Macromolecules
87
1. Biotinylated albumin solution: Prepare a 5 mg/ml solution of biotinylated albumin (A8549, Sigma) in sterile PBS (see Note 7). 2. Fixative solution: 4% (w/v) paraformaldehyde in PBS (see Note 8). 3. Glass slides and coverslips (Fisher Scientific). 4. TBS-tween solution (TBST): 0.1% Tween-20 in TBS. Store at room temperature. 5. PAP Pen (ScyTek Laboratories, Logan, UT). 6. Humid chamber. 7. Blocking buffer: 2% (v/v) of equine serum (Equitech-Bio Inc., Kerrville, TX) in TBST (see Note 8). 8. Streptavidin–alkaline phosphatase conjugate: Prepare at a concentration of 0.5 mg/ml (Jackson Immunoresearch, West Grove, PA) in TBST (see Note 8). 9. Alkaline phosphatase buffer (AB): 100 mM Tris–HCl, 100 mM NaCl, 5 mM MgCl2, 0.05% Tween 20, 5 mM levamisole, pH 9.5 (see Note 8). 10. BCIP (5-Bromo-4-Chloro-3-Indolyl-Phospate-P-Toluidine Salt) stock solution (50 mg/ml): dissolve BCIP (Research Products International Corp., Mt. Prospect, IL) in 100% dimethylformamide (DMF) (see Note 9). 11. NBT (2,2¢-Di-p-nitrophenyl-5, 5¢-diphenyl-3, 3¢-(3,3¢dimethoxy-4,4¢-diphenylene)-ditetrazolium chloride) stock solution (50 mg/ml): dissolve NBT (Research Products International Corp.) in 70% DMF (see Note 9). 12. Color development solution (CDS): For each ml of AB, add 6.7 ml and 5.2 ml of BCIP and NBT stock solutions, respectively (see Note 8). 13. Coplin jars. 14. Nuclear fast red staining solution (ScyTek Laboratories). 15. Citramount (Polysciences Inc., Warrington, PA).
3. Methods 3.1. Qualitative Assessment Vp Using Evans Blue Dye
1. Inject Evans Blue solution (mice, 0.1 ml; hamster, 0.25 ml; rat, 0.5 ml) into a lateral tail vein while the animal is under ketamine anesthetic. Evans Blue is a diazo dye that has a very high affinity for serum albumin (31). 2. After allowing the dye to circulate for 15 min, kill the animal and dissect out uteri.
88
B.M. Bany and G.S. Hamilton
3. Areas of “bluing” represent implantation site areas of increased vascular permeability to Evans Blue–albumin complexes (see Fig. 1). 3.2. Quantitative Assessment of Vp
1. Before autopsy on Day 4.5 (mice) or 5.5 (rats) of pregnancy, the animals are given two intravenous injections (tail vein) while under ketamine anesthetic: (a) Inject 125I-albumin (2–3 mCi) into the tail vein. For experiment 1, this is at 5, 10, 20, 40, 80, or 160 min before autopsy on Day 4.5 of pregnancy. For experiment 2, the injections are given at 6, 12, or 24 h (see Note 10). (b) Inject Evans Blue dye solution at 15 min prior to autopsy as described above. 2. Plasma collection: Upon autopsy, inject animal with ketamine– xylazine mixture intraperitoneally and collect blood slowly from the heart or abdominal vein into heparinized syringes. Place blood into 1.5-ml centrifuge tubes and centrifuge at 10,000 × g for 2 min. Collect plasma sample (top clear phase) into another 1.5-ml tube and prepare duplicate 0.1 ml plasma samples for radioactivity determination. 3. Tissue collection: Dissect uterine horns and trim off mesentery. After washing quickly in PBS the implantation (as revealed by the intense bluing by Evans Blue) and nonimplantation sites need to be separated using a scalpel and pooled into preweighed tubes. Weigh the tube containing the tissues to obtain tissue weight. 4. Measure radioactivity levels of the plasma and tissue samples using a gamma counter. 5. Calculations: To assess differences in vascular permeability in the implantation versus nonimplantation segments of the uterus: (a) Calculate the tissue volumes of distribution (VD) of 125 I-albumin for both samples from each mouse by dividing tissue radioactivity (cpm/g) by plasma radioactivity (cpm/ml). (b) Steady-state volume of distribution (VDss) for 125I-albumin is an estimate of the extracellular fluid space of the tissue and is calculated for implantation (VDssI) and nonimplantation segment (VDssN) tissue. This is achieved by calculating the overall average volume of distribution for the 6, 12, and 24 h time points (experiment 2). (c) Calculate the slopes and coefficients of determination of the line defined by – ln(1 − (VDt/VDss)) as a function of time (hours) for each tissue type of Experiment 1 by the leastsquares method of linear regression. For pregnant uteri,
5 Assessment of Permeability Barriers to Macromolecules in the Rodent Endometrium¼
89
Fig. 2. Determination of KI and KN by linear regression. The slopes of the lines (K) defined by ln(1 − (VDt/VDss)) versus time for implantation (I) compared to nonimplantation (N) segments are a measure of the differences in vascular permeability to 125I-albumin. A similar difference in vascular permeability constants between artificially induced compared to contralateral control uterine horns can be made in animals undergoing artificially induced decidualization.
the slopes are equal to the rate constant K for implantation (KI) and nonimplantation (KN) tissue (Fig. 2) (see Note 11). 3.3. Quantitative Assessment of ECFV
1. While under ketamine plus xylazine, a midventral incision is made in the abdomen. After carefully exposing each kidney, the renal pedicles are ligated with surgical silk and the kidneys are removed with scissor cuts made distal to the ligation (see Note 12). 2. Using a 1-ml syringe and a 25 gauge needle, 5 mCi 51Cr-EDTA in saline (0.2 ml mouse; 0.5 ml rat) as well as Evans Blue dye (as described above) is injected into a lateral tail vein. 3. After 15 min, plasma samples are collected as above and the animals are killed. Prepare duplicate 0.1 ml plasma samples for radioactivity determination. 4. Tissue collection: Dissect uterine horns and trim of mesentery. After washing quickly in PBS the implantation (as revealed by the intense bluing by Evans Blue) and nonimplantation sites need to be separated using a scalpel and pooled into preweighed tubes. Weigh the tube containing the tissues to obtain tissue weight. 5. Measure radioactivity levels of the plasma and tissue samples using a gamma counter. 6. The tissue volume of distribution of 51Cr-EDTA represents the ECFV. This is determined by dividing the uterine tissue radioactivity (cpm/mg) by the plasma radioactivity (cpm/ml).
90
B.M. Bany and G.S. Hamilton
The ECFV (in ml) can be compared in implantation sites versus nonimplantation sites or in uterine horns subjected to artificial deciduogenic stimulation versus unstimulated horns (see Note 13). 3.4. Visualizing PDZ Impermeability to Macromolecules
1. While under anesthesia, inject 1 mg of biotinylated albumin into a lateral tail vein (see Note 14). 2. After 1–2 h, collect the uteri and place in fixative solution. 3. Fix tissue in fixative solution for 24 h at 4°C with gentle shaking. 4. Wash in 70% ethanol twice. 5. Place in 70% ethanol for 24 h with shaking at 4°C. Replace the 70% ethanol. You can store the tissue at 4°C indefinitely. 6. Process the tissues into paraffin blocks then prepare cross sections (4–10 mm) and mount onto glass slides using standard techniques. 7. De-paraffinize and hydrate sections: (a) Heat slides to 80°C for 5 min to melt wax. (b) Place in xylene for 10 min. Repeat twice. (c) Place in 100% ethanol for 5 min. Repeat twice. (d) Place in 95% ethanol for 5 min. Repeat once. (e) Place in 70% ethanol for 5 min. Repeat once. (f) Place in water for 5 min. Repeat once. 8. Place in TBST for 5 min. Repeat once. 9. Encircle the section(s) with a hydrophobic barrier using a PAP Pen and incubate sections in an appropriate volume of blocking buffer for 1 h at room temperature in a humid chamber. 10. Wash sections in TBST for 5 min. Repeat once. 11. Incubate sections in humid chamber in a 1:500 dilution of streptavidin–alkaline phosphatase solution for 30 min (see Note 15). 12. Wash for 15 min in TBST. Repeat twice. 13. Incubate for 5 min in AB buffer. Repeat once. 14. Incubate sections with CDS until desired color precipitate develops. 15. Wash in TBST for 10 min. 16. Wash twice in distilled water for 10 min. 17. Immerse slides in nuclear fast red for 1–5 min to stain the nuclei a reddish color. 18. Place in two changes of tap water for 30 s. 19. Dehydrate slides in alcohol series:
5 Assessment of Permeability Barriers to Macromolecules in the Rodent Endometrium¼
91
Fig. 3. Representative photomicrographs of biotinylated-albumin histochemistry of sections prepared from Day 5.5 pregnant mice. Tissues were collected 1 h after intravenous injection of biotinylated albumin. Values above scale bars represent microns. Sections from the uteri of mice that were not injected with biotinylated albumin showed no purple staining (not shown). PDZ primary decidual zone of the endometrium, C implanting conceptus.
(g) Place in 70% ethanol for 5 min. (h) Plave in 95% ethanol for 5 min. Repeat once. (i) Place in 100% ethanol for 5 min. Repeat once. 20. Place slides in xylene or appropriate xylene replacement for 5 min. Repeat once. 21. Place coverslip over section using Citramount mounting medium. 22. Examine slides under a light microscope. Nuclei are stained red and a blue/purple precipitate is localized to the intravenously injected biotinylated albumin (see Fig. 3).
4. Notes 1. Unless otherwise stated, all materials were purchased from Sigma-Aldrich (St. Louis, MO). 2. Users need to inform themselves of the dangers of the reagents, especially the radioactive materials, and potential or known carcinogens used in these protocols. Ensure proper personal protective gear and methods of waste disposal are used as recommended by Federal and local regulatory agencies.
92
B.M. Bany and G.S. Hamilton
3. All solutions should be prepared in water that has a resistivity of 18.2 MW-cm and total organic content of less than five parts per billion, unless otherwise stated. 4. It can be stored at room temperature. 5. These tubes can be stored indefinitely but over time a precipitate may form, which needs to be centrifuged out. 6. This can be used immediately or store at 4°C. Do not store for more than 48 h. 7. Store in usable aliquots indefinitely at −20°C but avoid repeated freeze–thaw cycles. 8. Prepare fresh on the day of use. 9. Store in aliquots at −20°C protected from light. 10. This is carried out in two separate experiments using at least five to six animals per time point. The shorter time points (5–160 min) are before the volume of distribution for 125 I-albumin reaches a steady-state, while those of experiment 2 (6–24 h) are after a steady-state is reached. 11. To determine if the slopes of the lines between implantation and nonimplantation segments differ, use statistical methods described by Sokal and Rohlf (32). 12. When using 51Cr-EDTA for assessment of ECFV, animals must be nephrectomized to allow equilibration of the tracer among fluid compartments. In intact animals, the kidney eliminates the EDTA fairly rapidly. 13. Perform a statistical analysis using repeated-measures analysis of variance methods to determine differences between implantation versus nonimplantation segments or stimulated versus control uterine horns for pregnant mice or mice undergoing artificially induced decidualization, respectively. 14. For negative controls, refrain from injecting the animals with the biotinylated albumin. Alternatively, these control animals can be injected with the same amount of nonbiotinylated albumin. 15. Make sure not to have serum in this incubation, as serum can have appreciable amounts of biotin or biotin-like molecules that can bind the streptavidin.
Acknowledgment This work was supported, in part, by RO1 HD049010 from NIH (Eunice Shriver Kennedy National Institute of Child Health and Development) to B.B.
5 Assessment of Permeability Barriers to Macromolecules in the Rodent Endometrium¼
93
References 1. Maddocks S, Setchell BP 1988 The physiology of the endocrine testis. Oxf Rev Reprod Biol 10:53–123 2. Ueno M 2009 Mechanisms of the penetration of blood-borne substances into the brain. Curr Neuropharmacol 7:142–149 3. Psychoyos A 1960 La reaction deciduale est precede de modifications prcoces de la permeabilite capillaire de l’uterus. C R Seances Soc Biol Fil 154:1384–1387 4. Finn CA, McLaren A 1967 A study of the early stages of implantation in mice. J Reprod Fertil 13:259–267 5. Orsini MW, Donovan BT 1971 Implantation and induced decidualization of the uterus in the guinea pig, as indicated by Pontamine blue. Biol Reprod 5:270–281 6. Hoffman LH, DiPietro DL, McKenna TJ 1978 Effects of indomethacin on uterine capillary permeability and blastocyst development in rabbits. Prostaglandins 15:823–828 7. Boshier DP 1970 The pontamine blue reaction in pregnant sheep uteri. J Reprod Fertil 22:595–596 8. Guillomot M, Flechon JE, Wintenberger-Torres S 1981 Conceptus attachment in the ewe: an ultrastructural study. Placenta 2:169–182 9. Keys JL, King GJ, Kennedy TG 1986 Increased uterine vascular permeability at the time of embryonic attachment in the pig. Biol Reprod 34:405–411 10. Mead RA, Bremner S, Murphy BD 1988 Changes in endometrial vascular permeability during the periimplantation period in the ferret (Mustela putorius). J Reprod Fertil 82: 293–298 11. Kennedy TG 1979 Prostaglandins and increased endometrial vascular permeabiltiy resulting from the application of artificial stimulus to the uterus of the rat sensitized for the decidual cell reaction. Biol Reprod 20:560–566 12. Psychoyos A 1961 [Capillary permeability and uterine deciduation.]. C R Hebd Seances Acad Sci 252:1515–1517 13. Lundkvist O, Nilsson BO 1982 Endometrial ultrastructure in the early uterine response to blastocysts and artificial deciduogenic stimuli in rats. Cell Tissue Res 225:355–364 14. Milligan SR, Mirembe FM 1984 Time course of the changes in uterine vascular permeability associated with the development of the decidual cell reaction in ovariectomized steroid-treated rats. J Reprod Fertil 70:1–6
15. McRae AC, Heap RB 1988 Uterine vascular permeability, blood flow and extracellular fluid space during implantation in rats. J Reprod Fertil 82:617–625 16. Hamilton GS, Kennedy TG 1993 Uterine extracellular fluid volume and blood flow after artificial uterine stimulation to rats differentially sensitized for the decidual cell reaction. Biol Reprod 48:910–915 17. Hamilton GS, Kennedy TG, Karlik SJ 1994 Early identification of sites of embryo implantation in rats by means of gadolinium-enhanced MR imaging. J Magn Reson Imaging 4: 481–484 18. Hamilton GS, Kennedy TG, Norley CJ, Karlik SJ 1993 Gadolinium-DTPA enhanced MRI demonstrates uterine vascular changes associated with artificially induced decidualization and ovoimplantation in rats. Magn Reson Med 29:817–821 19. Bany BM, McRae AC 1992 Uterine uptake of alpha 2-macroglobulin and alpha 1-proteinase inhibitor from the blood during early implantation in the mouse. Biol Reprod 47:514–519 20. McRae AC, Kennedy TG 1982 Evidence for a blood-uterine lumen permeability barrier in rats treated with hormones to mimic early pseudopregnancy. Can J Physiol Pharmacol 60:1630–1635 21. McRae AC, Kennedy TG 1983 Selective permeability of the blood-uterine lumen barrier in rats: importance of lipid solubility. Biol Reprod 29:886–894 22. McRae AC, Kennedy TG 1983 Selective permeability of the blood-uterine lumen barrier in rats: importance of molecular size. Biol Reprod 29:879–885 23. Plaks V, Kalchenko V, Dekel N, Neeman M 2006 MRI analysis of angiogenesis during mouse embryo implantation. Magn Reson Med 55:1013–1022 24. Rogers PA, Murphy CR, Rogers AW, Gannon BJ 1983 Capillary patency and permeability in the endometrium surrounding the implanting rat blastocyst. Int J Microcirc Clin Exp 2:241–249 25. Parr MB, Parr EL 1986 Permeability of the primary decidual zone in the rat uterus: studies using fluorescein-labeled proteins and dextrans. Biol Reprod 34:393–403 26. Parr MB, Tung HN, Parr EL 1986 The ultrastructure of the rat primary decidual zone. Am J Anat 176:423–436 27. Herington JL, Underwood T, McConaha M, Bany BM 2009 Paracrine signals from the
94
B.M. Bany and G.S. Hamilton
mouse conceptus are not required for the normal progression of decidualization. Endocrinology 150:4404–4413 2 8. Wang X, Matsumoto H, Zhao X, Das SK, Paria BC 2004 Embryonic signals direct the formation of tight junctional permeability barrier in the decidualizing stroma during embryo implantation. J Cell Sci 117:53–62 2 9. Tung HN, Parr MB, Parr EL 1986 The permeability of the primary decidual zone in the rat uterus: an ultrastructural tracer
and freeze-fracture study. Biol Reprod 35:1045–1058 30. Paria BC, Zhao X, Das SK, Dey SK, Yoshinaga K 1999 Zonula occludens-1 and E-cadherin are coordinately expressed in the mouse uterus with the initiation of implantation and decidualization. Dev Biol 208:488–501 31. Freedman FB, Johnson JA 1969 Equilibrium and kinetic properties of the Evans blue-albumin system. Am J Physiol 216:675–681 32. Sokal RR, Rohlf FJ 1981 Biometry. Second ed. New York: W.H. Freeman and Company
Chapter 6 Assessment of Intestinal Permeability in (Premature) Neonates by Sugar Absorption Tests Willemijn E. Corpeleijn, Ruurd M. van Elburg, Ido P. Kema, and Johannes B. van Goudoever Abstract Infants born prematurely have an enhanced intestinal permeability compared to healthy term infants. This enhanced permeability might be a contributing factor in the development of Necrotising Enterocolitis. The assessment of intestinal permeability in premature neonates with sugar absorption tests has been proven to be safe and of minimal burden to the infant. After enteral administration of a test solution containing lactulose and mannitol, the excretion of these sugars is measured in urine, and the ratio is calculated. The lactulose and mannitol concentrations in urine can be measured by the use of a gas chromatograph after pre-purification and derivatisation of the sample. Non-invasive assessment of intestinal permeability can be useful in monitoring the effects of experimental (nutritional) therapy. Key words: Intestinal permeability, Premature neonate, Lactulose, Mannitol, Sugar absorption test, Gas chromatography
1. Introduction The wall of the small intestine serves as a barrier. This barrier needs to be selective; uptake of nutrients and water has to be facilitated, but at the same time, pathogens need to be repulsed. To absorb nutrients and water from the lumen, intestinal epithelial cells are equipped with various selective transporters and ion channels. To prevent unwanted entities (e.g. bacteria and parasites) from entering via the paracellular spaces, epithelial cells are sealed together by structures called tight junctions. There are several diseases that are associated with an enhanced intestinal permeability, such as coeliac disease (1) and inflammatory bowel disease (2). Sugar absorption tests measuring intestinal
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_6, © Springer Science+Business Media, LLC 2011
95
96
W.E. Corpeleijn et al.
permeability can be helpful not only in establishing diagnoses of these diseases but also in monitoring the effectiveness of therapy. Enhanced intestinal permeability is physiologic in a newborn infant and may even play an important role in the uptake of larger (nutritional) molecules from breast milk and the induction of systemic tolerance. However, preterm infants have an even higher intestinal permeability compared to healthy term newborns (3). The enhanced permeability in this group may be caused by incomplete expression of the junctional proteins that form the tight junctions, and this may lead to increased translocation and a systemic spread of bacteria (4, 5). This might contribute to the development of Necrotising Enterocolitis (NEC) and sepsis. It has been shown that premature infants receiving a majority of their feeding as human milk have a significantly lower intestinal permeability compared to infants who receive formula (6).The assessment of intestinal permeability by the sugar absorption test has been proven to be safe in preterm newborns (3) and has been useful in scientific research on the effects of experimental therapy (7, 8). The sugar absorption test is based on the theory that, in a healthy intestine, monosaccharides (<0.5 nm in diameter, e.g. mannitol) are readily absorbed through the transcellular pathway, but larger disaccharides (>0.5 nm in diameter, e.g. lactulose) are only absorbed through the paracellular pathway. The paracellular pathway is a small population of larger pores located in the tight junctions, permitting the passage of both lactulose and mannitol. After absorption through the intestinal wall, the unchanged mono- and disaccharides will be quickly excreted by the kidneys. Therefore, the lactulose–mannitol ratio (L/M ratio) in urine is a measure of intestinal integrity. In disease states, the permeability of lactulose can be increased and/or the permeability of mannitol decreased. A decrease in the permeability of mannitol is more likely to be the result of enteropathy, resulting in a decrease of absorptive surface, whereas an increased permeability is more likely to be the result of inflammation resulting in decreased functioning of the tight junctions. Mannitol has a natural occurrence in urine; however, the concentration is negligible and is unlikely to influence test results. In older infants and adults, sugar absorption tests are usually performed with a hyperosmolar solution of lactulose, mannitol, and sucrose (1,560 mosm) because it discriminates better between patients with and without a condition such as coeliac disease (9). However, the use of hyperosmolar solutions in neonates has been associated with NEC. For this protocol in neonates, sucrose was omitted from the solution to create a slightly hyperosmolar solution (375 mosm) that can be safely used in premature neonates.
6 Assessment of Intestinal Permeability in (Premature) Neonates¼
97
2. Materials 2.1. Materials for Preparation of 100 ml of Sugar Solution
1. Mannitol, 2 g. 2. Crystalline lactulose, 50% syrup, 10 g (Dulphalac, Solvay Pharma BV, The Netherlands). 3. Sterile demineralised water. 4. Inert plastic bottles (content > 100 ml).
2.2. Materials for Pre-purification and Urinary Analysis
1. Gas chromatograph with flame ionisation detection equipped with an autosampler. 2. Cryobath. 3. Chromatography data system. 4. GLC column (60 m × 0.25 mm Zebron ZB-1, 100% methylpolysiloxane (film thickness 0.11 mm)). 5. Sodium sulphate (Na2SO4), stored in a cold chamber. 6. Bis(trimethylsilyl)trifluoroacetamide (BSTFA), derivatisation reagent (Sigma Aldrich, Zwijndrecht, The Netherlands), stored in a cold chamber (corrosive!). 7. Tri-sil-TBT (silylation reagent comprising TMSI, BSA, and TMCS) (Thermo Scientific), stored in a cold chamber (toxic/ flammable). 8. Chlorohexidine digluconate, 20% in water, stored in an acid cabinet. 9. Heptane, stored in a safety cabinet (flammable). 10. Hydrochloric acid solution, 0.1 mol/L, stored in an acid cabinet (corrosive). 11. Methanol, stored in a safety cabinet (flammable). 12. Urine for the preparation of a urine pool. 13. Method for urine creatinine content determination (e.g. Modular P 800 analyzer, Roche Diagnostics Nederland B.V. Almere, The Netherlands). 14. Mannitol (Sigma Aldrich, Zwijndrecht, The Netherlands), stored in a chemical cabinet. To prepare a standard stock solution of mannitol (concentration: ca. 10 mmol/L): 91 mg of mannitol is dissolved in 50 ml of methanol/H2O (20%/80%) (Store in a fridge, expires after 12 months). 15. Myo-inositol (Sigma Aldrich, Zwijndrecht, Netherlands), stored in a fridge. To prepare a standard stock solution of myoinositol (concentration: ca. 10 mmol/L): 90 mg of myo-inositol is dissolved in 50 ml of methanol/H2O (20%/80%). (Store in a fridge, expires after 12 months).
98
W.E. Corpeleijn et al.
16. Sucrose (Sigma Aldrich, Zwijndrecht, Netherlands), stored in a fridge. To prepare a standard stock solution of sucrose (concentration: ca. 5 mmol/L): 86 mg of sucrose is dissolved in 50 ml of methanol/H2O (20%/80%). (Store in a fridge, expires after 12 months). 17. Lactose (Sigma Aldrich, Zwijndrecht, The Netherlands), stored in a fridge. To prepare a standard stock solution of lactose (concentration: ca. 5 mmol/L): 90 mg of lactose is dissolved in 50 ml of methanol/H2O (20%/80%). (Store in a fridge, expires after 12 months). 18. Glucose (Sigma Aldrich, Zwijndrecht, The Netherlands), stored in a fridge. To prepare a standard stock solution of glucose (concentration: ca. 10 mmol/L): 90 mg of lactose is dissolved in 50 ml of methanol/H2O (20%/80%). (Store in a fridge, expires after 12 months). 19. Lactulose (Sigma Aldrich, Zwijndrecht, The Netherlands), stored in a chemical cabinet. To prepare a standard stock solution of lactulose (concentration: ca. 1 mmol/L): 34 mg of lactose is dissolved in 100 ml of methanol/H2O (20%/80%). (Store in a fridge, expires after 6 months). 20. Perseitol (Pfanstiehl Laboratories), stored in a fridge. To prepare an internal standard perseitol solution (0.4 mmol/L): 10.6 mg of perseitol is dissolved in 100 ml of methanol/H2O (20%/80%). (Store in a fridge, expires after 6 months). To prepare a standard mix solution of mannitol, myo-inositol, sucrose, lactose, and glucose (with a concentration of 4 mmol/L of each component): in a 100-ml flask, mannitol, myo-inositol, and glucose are diluted ×25. Lactose and sucrose are diluted ×12.5 and the flask is filled up to 100 ml with methanol/H2O (20%/80%). (Store in a fridge, expires after 6 months).
3. Methods 3.1. Preparation of the Sugar Solution
1. Two grams of mannitol are dissolved in 10 ml of demineralised sterile water. The lactulose syrup is added, followed by water up to 100 ml (see Note 1). The solution should be slightly hyperosmolar (375 mmol/L). The prepared solution should be stored in inert plastic bottles at −70°C. The expiration date is 6 months after preparation.
3.2. Sugar Absorption Test-Performance in Neonates
1. There is no need for the subjects to be fasted or to have an empty bladder before the start of the study. A nasogastric tube needs to be in situ.
6 Assessment of Intestinal Permeability in (Premature) Neonates¼
99
2. The test solution is instilled through the nasogastric tube at a dose of 2 ml/kg body weight. To make sure the complete dose of test solution reaches the stomach, the tube is flushed with 0.5 ml of air (see Note 2). 3. Urinary output is collected over 5 h. A piece of plastic is inserted in the infants diaper and is covered with soft gauzes. The plastic prevents the urine from being absorbed by the diaper. The wet gauzes are placed in a 20-ml syringe after removing the plunger. When the plunger is put back into place and gently pressed down, the urine will leave the syringe under the influence of the pressure. After collection, 0.1 ml of chlorohexidine digluconate (20%) is immediately added as a preservative. If the gauzes are partly smeared with faeces, these parts are cut out. 4. The total amount of urine is measured and 5 ml is stored at −20°C until analysis. Frozen urine can be stored for up to 1 year. At least 5 ml of urine needs to be collected. If after 5 h the volume collected is <5 ml, collection is continued as long as needed. 3.3. Preparation of the Urinary Pools
As a quality control, three pools of urine with an increasing lactulose–mannitol ratio (about 0.010, 0.075, and 0.150) are included in each series of urine analyses. These pools are prepared as follows: 1. Urine is collected from healthy volunteers and divided into three pools of 300 ml (2 ml of chlorohexidine digluconate (20%) is immediately added as a preservative). 2. It is presumed that urine naturally contains approximately 2 mmol mannitol/300 ml of urine. Therefore, the following amounts of lactulose and mannitol have to be added to the urine pools. Pool 1 (L–M ratio: 0.010): 6.8 mg lactulose, 364 mg mannitol. Pool 2 (L–M ratio: 0.075): 51.3 mg lactulose, 364 mg mannitol. Pool 3 (L–M ratio: 0.150): 102.7 mg lactulose, 364 mg mannitol. Divide the pools into small aliquots (4 ml) and store at −20°C until needed. These aliquots expire after 3 years.
3.4. Method of Derivatisation, Pre-purification, and Analysis
Sugars such as mannitol and lactulose are non-volatile compounds, which makes them unsuitable for measurement by gas chromatography. Therefore, these sugars, like many other organic compounds, need to be derivatised prior to analysis. During derivatisation, a hydrogen group is replaced by a silicon atom with three methyl groups attached to it, (–Si (CH3)3]. Attachment of trimethylsilyl groups makes molecules more volatile and, therefore,
100
W.E. Corpeleijn et al.
suitable for gas chromatography. Urine is a complex liquid that contains compounds such as organic salts, amino acids, and urea, in addition to sugars. As a consequence, the compound that one wants to analyse needs to be separated from the rest of the matrix. However, sugars do not contain functional groups that enable their selective isolation from urine or other body fluids, and prepurification methods are usually based on the removal of other (polar) substances. The method described here is based on the assumption that trimethylsilylated sugars are much more stable towards hydrolysis (by the addition of water and hydrochloric acid) than the TMS derivatives of other compounds such as amino acids. The intact sugar TMS derivatives are, after hydrolysis of the sample, extracted into hexane. The method of pre-purification described here is based on the method developed by Jansen et al. (10) and modified by van Elburg et al. Additionally, the creatinine content is measured so that the lactulose and mannitol concentrations can be expressed as mmol/mol creatinine. 1. First, the standard mix and the internal standard are prepared (see Subheading 2.2). See Note 3. 2. Urine is thawed before analysis. Creatinine content is measured. The pH of the urine is measured and adjusted to 5–7 if necessary by using a solution of 1 mol/L HCl. 3. The urine is centrifuged at room temperature for 10 min at 2,800 × g. 4. For the calibration line: 0, 50, 100, 200, 300, and 400 ml of the standard mix solution and the lactulose solution are pipetted into glass test tubes. 5. For the control samples: 90 ml of pools 1, 2, and 3 are pipetted into test tubes. For the fourth 90 ml urine sample, 100 ml of standard mix solution and 100 ml of lactulose stock solution are added. 6. From each urine sample (see Note 4), the equivalent of 0.7–0.9 mmol of creatinine is pipetted into a test tube and 100 ml of the internal standard is added;for example if the urine has a creatinine content of 12 mmol/L, then 70 ml of urine will be the equivalent of 0.84 mmol of creatinine. The amount of urine used is noted. 7. To all of the samples and standards, 2 ml of methanol is added. 8. The samples are placed under a stream of nitrogen and evaporated to dryness at 60°C. 9. The dry residues are derivatised by the addition of 300 ml of Trisil-TBT and heated for 30 min at 100°C. Addition of the Tri-sil-TBT should be done under a fume hood.
6 Assessment of Intestinal Permeability in (Premature) Neonates¼
101
10. The sample is allowed to cool and is subsequently diluted with 1 ml of distilled water. This should also be done under a fume hood. Addition of the water should make a sizzling sound; if not, the sample is discarded (see Note 5). 11. The derivatives are extracted into 1 ml of heptane. The sample is placed on a multivortex for 2 min and subsequently centrifuged at 2,711 × g for 5 min at room temperature. 12. The heptane layers are pipetted into a glass tube (alternatively, the bottom layer can first be frozen using a cryogenic bath). Steps 8 and 9 are then repeated. 13. The heptane layer is washed by the addition of 1 ml of 0.1 mol/L hydrochloric acid solution and placed on a multivortex for 1 min. 14. The samples are centrifuged at 2,711 × g for 5 min. 15. The heptane layers are removed from the samples and a small scoop of anhydrous sodium sulphate is added to the heptane layer. 16. The samples are centrifuged at 2,711 × g for 5 min and the heptane layer is subsequently poured off into a new urine tube. 17. One drop (ca. 25 ml) of bis(trimethylsilyl)trifluoroacetamide is added in heptane to re-install potentially lost trimethylsilyl groups on the anomeric hydroxy group. The sample has now been pre-purified and derivatised, and the sample is ready for injection into the gas chromatograph. 18. Start each series with injection of the standard mix solution and the pooled urine samples. 19. Inject 0.5 ml portions of the trimethylsilylether derivate (TMS). Use the following settings: a helium flow rate of 35 cm/s, detector temperature of 300°C, and injector temperature of 280°C. 20. The following oven program can be used: start at 120°C, increase by 3°C/min to 240°C, increase at 2°C/min to 270°C, hold for 10 min at 270°C, raise at 20°C/min to 300°C, and hold for 2 min at 300°C. This will take 68.5 min. 3.5. Quantification of Lactulose and Mannitol
Figure 1 shows a chromatogram of the standard mix. Figure 2 shows a typical chromatogram of a urine sample of a patient after ingestion of the test solution. The area under a peak is proportional to the amount of analyte (in this case, mannitol and lactulose) present in the chromatogram and represents the concentration in the original sample. The area under the peak can be calculated by using a chromatography data system.
102
W.E. Corpeleijn et al.
Fig. 1. Chromatogram of the standard mix.
Fig. 2. Typical chromatogram of a urine sample of a patient after ingestion of the test solution.
3.6. Calculations
The following calculations are used: Mannitol, myo-inositol, and glucose concentrations (in mmol/L):
X /M ´ 1 / 50 ´ 1 / 25 ´ Vstandard ´ [Area int.stand. / Area stand ] ´ ëé Area sample / Area int.stand. ûù ´ 1 / Vsample ´ 103 ´ 103 = concentration in mmol / L
6 Assessment of Intestinal Permeability in (Premature) Neonates¼
103
where X is the amount of standard used for preparation of the stock in mg, M is the mol mass of the sugars, 1/50 is the dilution of compound X in 50 ml, 1/25 is the dilution from the stock solution to the standard mix solution, Vstandard is the injected volume of the standard, area is the calculated area under the curve, Vsample is the injected amount of sample, ×103 converts mmol to mmol and ×103 converts ml to L. Sucrose and lactose:
[
X /M ´ 1 / 50 ´ 2 / 25 ´ Vstandard ´ Area int.stand. / Area stand
]
´ ëé Area sample /Area int.stand. ûù ´ 1 / Vsample ´ 103 ´ 103 = concentration in mmol / L
where X is the amount of standard used for preparation of the stock in mg, M is the mol mass of the sugars, 1/50 is the dilution of compound X in 50 ml, 2/25 is the dilution from the stock solution to the standard mix solution, Vstandard is the injected volume of the standard, area is the calculated area under the curve, Vsample is the injected amount of sample, ×103 converts mmol to mmol and ×103 converts ml to L. Lactulose:
[
X /M ´ 1 / 100 ´ Vstandard ´ Area int.stand. / Area stand
]
´ ëé Area sample / Area int.stand. ûù ´ 1 / Vsample ´ 10 ´ 103 = concentration in mmol / L 3
where X is the amount of standard used for the preparation of the stock in mg, M is the mol mass of lactulose, 1/100 is the dilution of lactulose in the stock solution, Vstandard is the injected volume of the standard, area is the calculated area under the curve, Vsample is the injected amount of sample, ×103 converts mmol to mmol and ×103 converts ml to L. To convert the concentration to mmol per mol creatinine: mmol/L/creatinine mmol/L = mmol/mol creatinine. To calculate the lactulose–mannitol absorption rate: mmol/mol lactulose/mmol/mol mannitol ratio.
mannitol = lactulose–
4. Notes 1. Owing to the vulnerability of the study population, it is advised to work under sterile conditions during preparation of the test solutions. 2. When performing the sugar absorption test (SAT), keep an aliquot of the test solution in a −80°C freezer until all analyses
104
W.E. Corpeleijn et al.
have been performed. The lactulose and mannitol concentrations of the used sugar solution can then be checked in the case of unexpected results. 3. It is important to work with clean tubes and materials. Disturbances caused by contaminated lab materials can be avoided by cleaning tubes and other equipment with methanol and pentane. 4. It is advisable to analyse samples in duplicate. 5. If the addition of water in Subheading 3.4 step 11 does not produce a sizzling sound, discard the sample. Start over with half of the amount of urine used in the first attempt (to a maximum of 350 ml) and dry two times after the addition of 1 ml of methanol. References 1. van Elburg R M, Uil J J, Mulder C J et al. (1993) Intestinal permeability in patients with coeliac disease and relatives of patients with coeliac disease. Gut. 34(3): p. 354–7. 2. Pearson A D, Eastham E J, Laker M F et al. (1982) Intestinal permeability in children with Crohn’s disease and coeliac disease. Br Med J (Clin Res Ed). 285(6334): p. 20–1. 3. van Elburg R M, Fetter W P F, Bunkers C M, et al. (2003) Intestinal permeability in relation to birth weight and gestational and postnatal age. Arch Dis Child Fetal Neonatal Ed 88(1): p. F52–5. 4. Kelly D, Coutts A G (2000) Early nutrition and the development of immune function in the neonate. Proc Nutr Soc. 59(2): p. 177–85. 5. Teitelbaum J E, Walker W A (2005) The development of mucosal immunity. Eur J Gastroenterol Hepatol. 17(12): p. 1273–8. 6. Taylor S N, Basile L A, Ebeling M, et al. (2009) Intestinal permeability in preterm infants by feeding type: Mother’s milk versus formula. Breastfeed Med.; Mar (4)1:11–5
7. van den Berg A, Fetter W P F, Westerbeek E A M et al. (2006) The effect of glutamine-enriched enteral nutrition on intestinal permeability in very-low-birth-weight infants: a randomized controlled trial. JPEN J Parenter Enteral Nutr. 30(5): p. 408–14. 8. Corpeleijn W E, van Vliet I, de Gast-Bakker D A et al. (2008) Effect of enteral IGF-1 supplementation on feeding tolerance, growth, and gut permeability in enterally fed premature neonates. J Pediatr Gastroenterol Nutr. 46(2): p. 184–90. 9. Uil, J J, van Elburg R M, Janssens P M W et al. (2000) Sensitivity of a hyperosmolar or “low”-osmolar test solution for sugar absorption in recognizing small intestinal mucosal damage in coeliac disease. Dig Liver Dis. 32(3): p. 195–200. 10. Jansen G, Muskiet F A J, Schierbeek H, et al. (1986) Capillary gas chromatographic profiling of urinary, plasma and erythrocyte sugars and polyols as their trimethylsilyl derivatives, preceded by a simple and rapid prepurification method. Clin Chim Acta. 157(3): p. 277–93.
Chapter 7 Analysis of Epithelial Cell Shedding and Gaps in the Intestinal Epithelium Carrie A. Duckworth and Alastair J. Watson Abstract The intestinal barrier is formed by a monolayer of columnar epithelial cells. This barrier is effectively maintained despite the high turnover of epithelial cells in the gut. Defects in the mechanism by which barrier function is maintained are believed to play a central role in the pathogenesis of inflammatory bowel disease (IBD). Proinflammatory cytokines such as TNF-a and IFN-g are often elevated in inflamed tissue of patients with IBD. In fact, anti-TNF-a therapy is routinely administered to patients with Crohn’s disease. We have previously demonstrated that intestinal epithelial cells are shed from the intestine leaving a ‘gap’ in the epithelium that is able to maintain barrier function. The rate of cell shedding and barrier permeability is substantially increased by the administration of TNF-a. Loss of barrier function at the site of a gap may provide a site of entry for disease-causing bacteria. Key words: Small intestine, Confocal, Cell shedding, Epithelium, TNF-a, Gap, Barrier function
1. Introduction The single layer of columnar epithelial cells lining the small intestine and colon is replaced every 3–5 days, making the intestine one of the most rapidly regenerating tissues of the body. Approximately 1,400 epithelial cells are shed from each villus every 24 h (1). Barrier function is, nonetheless, normally maintained despite the high rate that epithelial cells turnover. Cells arise from the stem cell region at the base of the crypt and migrate along the crypt–villus axis in the small intestine and along the crypt in the colon. Epithelial cells are then shed mainly from the villus tip in the small intestine and from the crypt table in the colon (1). Defects in this mechanism are believed to play a role in disease pathogenesis of the intestine.
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_7, © Springer Science+Business Media, LLC 2011
105
106
C.A. Duckworth and A.J. Watson
We have observed the cell shedding process using confocal and multiphoton microscopy in vivo and have confirmed the presence of gaps in the epithelium left behind by a shed cell (2, 3). The nuclear stain Hoechst 33258 is used to observe shedding nuclei and to assess whether a cell is present or not by confocal microscopy. An image stack is created by taking confocal images at several focal planes through the tissue to ascertain that the apparent gap is not due to a displaced nucleus. Confirmation that epithelial gaps were not being confused with goblet cells containing mucin came from the study of Math1−/− mice that lack goblet cells in approximately 90% of the small intestine and colon and, however, display the normal complement of epithelial gaps (3). We have previously demonstrated that epithelial cells leave perpendicular to the plane of the epithelium at a rate of 0.83 ± 0.6 mm/min (2), which is faster than the 0.3 mm/min rate of epithelial cell migration along the villus axis (4, 5), suggesting that shedding cells are pushed out of the epithelium, rather than being passively expelled. The lack of cellular components within epithelial gaps has been investigated. BCECF-AM did not stain gaps, indicating that the cytosolic esterases required to cleave BCECF-AM to fluorescent BCECF were not present. The membrane dye DiI highlighted discontinuities in the brush border membrane at the site of epithelial gaps and YC3.0 calcium chameleon mice demonstrated epithelial gaps despite fluorescent protein expression in the cytosol of all intestinal epithelial cells. Lucifer yellow, a highly polar fluorescent probe that is not able to permeate cell membranes, was observed in approximately 3% of gaps; however, it did not extend deeper than the line of adjacent nuclei, suggesting that there was some impermeable substance filling the gap. Despite the absence of cellular components in epithelial gaps, confocal reflectance imaging confirmed that gaps contained nonfluorescent material that prevented Lucifer yellow from penetrating deeper (2). The administration of murine recombinant TNF-a causes shrinkage of villi, shedding at the villus tip, and enterocyte detachment in the small intestine (3, 6). TNF-a-induced cell shedding leads to a decrease in the integrity of local barrier function and an increase in intestinal permeability. We have shown that Lucifer yellow penetrates approximately 20% of TNF-a-induced shedding events which is substantially increased from that observed during physiological cell shedding (3). Epithelial cell shedding and subsequent gap formation is a dynamic process and is, therefore, best assessed in real-time. Time-lapse confocal detection of fluorescent dyes administered in vivo, in combination with standard laboratory techniques provides a valuable tool for assessing this dynamic process.
7 Analysis of Epithelial Cell Shedding and Gaps in the Intestinal Epithelium
107
2. Materials 2.1. Surgical Preparation of Mice
1. Ketamine (20 mg/ml; Vetalar, Pfizer, Germany) and medetomidine (0.16 mg/ml; Domitor, Pfizer, Germany) are made up in water for injection. 2. 150 mM sodium chloride (Sigma, Dorset, UK). 3. Surgical instruments (scissors, forceps). 4. Heat pad (Vet Tech Solutions, UK). 5. Cauterizer. 6. 1 mm external diameter glass tube or Pasteur pipette bent into a right angle and tip blunted by Bunsen burner (Fig. 1). 7. Suture.
2.2. Confocal Imaging in Mice
1. Leica DM IRE2 Confocal microscope (Leica Microsystems (UK), Milton Keynes, UK). 2. 35 mm round glass bottom culture dish with 10-mm microwell covered by 0.16–0.19 mm coverglass (MatTek corporation, Ashland, MA).
Fig. 1. Example of Pasteur pipette bent to form an ‘L’ shape. Arrows indicate where a Bunsen burner has been used to bend pipette and to blunt the end, thus helping to reduce damage caused to small intestine upon cauterizing.
108
C.A. Duckworth and A.J. Watson
3. 3.85 mM acriflavine hydrochloride made up in 150 mM sodium chloride (Sigma, Dorset, UK). 4. Hank’s Balanced Salt Solution (HBSS; Sigma, Dorset, UK). 5. Hoechst 33258 is dissolved at 1 mg/ml in 150 mM sodium chloride and stored in aliquots at −20°C (Sigma, Dorset, UK). 6. Lucifer yellow is dissolved at 100 mM and stored in aliquots at −20°C (Invitrogen, Paisley, UK). 7. Dextran MW 10,000 conjugated to Alexa Fluor 647 is dissolved at 2 mg/ml and stored in aliquots at −20°C (Invitrogen, Paisley, UK). 8. Insulin syringe (Vet Tech Solutions Ltd, UK). 9. Murine TNF-a (PeproTech EC, London, UK) is dissolved in 0.1% bovine serum albumin (BSA, Sigma, Dorset, UK) at a concentration of 40 mg/ml. 2.3. Rigid Pen Confocal Probe Microscopy in Mice
1. Pen confocal probe (Optiscan, Melbourne, Australia). 2. 3.85 mM acriflavine hydrochloride made up in 150 mM sodium chloride (Sigma, Dorset, UK). 3. Hank’s Balanced Salt Solution (HBSS; Sigma, Dorset, UK). 4. Murine TNF-a is dissolved in 0.1% BSA at a concentration of 40 mg/ml (PeproTech EC, London, UK). 5. 20 mm × 30 mm × 3 mm piece of cork board with a 10 mm × 5 mm hole made in the middle. 6. Fine gauge pins. 7. Clamp and stand to hold confocal probe in position.
2.4. Confocal Endoscopy in the Human
1. Conscious sedation of patient with midazolam hydrochloride, fentanyl citrate, and propofol. 2. N-Butyl-scopolamine used as an antispasmolytic (Buscopan, Boehringer, Ingelheim, Germany). 3. Miniature confocal microscope inside the distal tip of a conventional video endoscope (EC 3830K, Pentax, Tokyo, Japan). 4. Acriflavine made up in 150 mM sodium chloride at a concentration of 1.93 mM.
2.5. Image Analysis of Cell Shedding and Gap Identification
1. Image analysis is carried out with Leica Confocal software, Leica Applications Software, and Image J. 2. Image analysis of confocal endoscopy and Optiscan pen confocal probe is undertaken with proprietary Pentax and Optiscan software and Photoshop.
7 Analysis of Epithelial Cell Shedding and Gaps in the Intestinal Epithelium
109
3. Methods 3.1. Surgical Preparation of Mice
1. C57BL/6 male and female mice are used from 8 to 12 weeks of age and are anesthetised with 75 mg/kg ketamine and 1 mg/kg medetomidine intraperitoneally and placed on a heat mat. A body temperature of 37°C is maintained throughout the study and 0.2 ml of subcutaneous saline is administered. 2. A midline incision of approximately 1 cm is carried out and a small segment of small intestine is exteriorized. 3. Being careful to maintain a healthy blood supply at all times, the segment of intestine is flushed out with 150 mM sodium chloride. 4. The segment is then isolated by tying sutures around each end to prevent intestinal contents coming out of the intestine. 5. A cauterizer is used to make a small hole at each end of the isolated segment on the anti-mesenteric axis of the intestine. A blunted glass rod is thread through the intestine through the two holes and cautery is carried out from one hole to the other along the antimesenteric axis (see Note 1). 6. The intestine is carefully opened out and kept moist by the application of 150 mM sodium chloride.
3.2. Confocal Imaging in Mice
1. Drug treatments if needed are administered; for instance, 10 mg TNF-a is placed within the peritoneal cavity (see Note 2). 2. Hoechst 33258 is administered at a dose of 2 mg/kg via lateral tail vein or retro-orbital injection. 3. A 35-mm round glass bottomed culture dish is placed on top of the opened out intestine with the area of intestine to be imaged over the coverslipped area. The mouse is carefully transferred from the supine position to the prone position and placed on an inverted confocal microscope stage that is surrounded by an incubator (see Note 3). A small volume of HBSS is administered directly into the dish to keep the intestine moist once on the microscope. Luminal dyes such as Lucifer yellow may be introduced into the culture dish at this stage (see Note 4). 4. A suitable field of view is found where there is not too much peristaltic movement but a continuous blood supply can be observed. A 40× objective is ideal for directly studying gaps whereas a 20× objective is more suitable for the assessment of cell shedding (see Note 5).
110
C.A. Duckworth and A.J. Watson
5. Z-stacks at 1-mm intervals are taken every 2.5 min for up to 3 h using excitation and emission wavelengths suitable to the fluorescent dyes being used (see Note 6). 3.3. Rigid Pen Confocal Probe Microscopy in Mice
1. Drug treatments if needed are administered following surgical preparation; for instance, 10 mg TNF-a is placed within the peritoneal cavity. 2. The exteriorized intestine is pulled through a small hole in a cork board and pinned out flat, giving a solid surface for the confocal probe. It is essential that the blood supply is not inhibited by this process. 3. A few drops of the acriflavine solution are placed directly onto the exposed intestinal mucosa (see Note 7). 4. The mouse is left in the supine position and the confocal probe is placed directly in contact with the intestinal mucosa (see Note 8). 5. Z-stacks are taken every 2.5 min for up to 3 h (see Note 9). 6. The mucosa must be kept moist at all times throughout the procedure by dropping 20 ml volumes of HBSS onto the exposed mucosa as required.
3.4. Confocal Endoscopy in Human
1. Standard endoscopical techniques are used to locate a suitable clean area of intestine for imaging. Two fluorescent probes can be used. Fluorescein (5 ml of 10% sodium fluorescein (Alcon, Pharma, Gmbh, Freiburg, Germany)) gives good views of mucosal blood vessels. Alternatively acriflavine (3.85 mM) gives high-resolution images of mucosal nuclei. A spraying catheter is used to administer a small amount of acriflavine in the region chosen for imaging and the endoscope is placed in a gentle fashion on the mucosal surface. 2. A z-stack of images 7 mm apart is taken from villi in the en face orientation up to a depth of 70 mm. 3. Images are assessed for gaps if they are part of three or more image planes within a stack of images and of sufficient technical quality to discern individual cells.
3.5. Image Analysis of Cell Shedding and Gap Identification
1. Goblet cells and gaps have a different microscopic appearance. When stained with acriflavine, goblet cells display a “target”-like appearance. The center of the target appears as a small dark area which contains mucin and is unstained, this is surrounded by a lighter colored area that contains mucin and cytoplasm which does stain (Fig. 2). 2. Gaps do not stain with acriflavine; however, they do reflect laser light, showing that they do contain a plug. 3. Luminal administration of Lucifer yellow has shown that on rare occasions gaps are infiltrated by Lucifer yellow (Fig. 3);
7 Analysis of Epithelial Cell Shedding and Gaps in the Intestinal Epithelium
111
Fig. 2. Goblet cells have a distinct appearance from gaps. (a) Diagram illustrating a goblet cell and gap in the epithelial layer. (b) Diagrammatic representation of the “targetlike” appearance of a goblet cell and a gap in the en face view. (c) En face view of goblet cells in the human small intestinal epithelium (white arrows) and an example of an epithelial gap (black arrow ) stained with acriflavine.
however, in the majority of cases, barrier function is maintained despite gaps in the epithelial layer. 4. Cell shedding is a dynamic process and can be quantified by assessing the movement of Hoechst 33258 or acriflavine stained nuclei out of the epithelial layer (Fig. 4). These can be counted by close inspection of movies created at each z position. Confirmation that a cell has shed leaving a gap is carried out by the observation of the same cell shedding at various z positions. Apoptotic fragmentation of the Hoechst stained nuclei of a shed cell can also be observed in the lumen.
Fig. 3. Confocal section across villi showing all epithelial cell nuclei stained by Hoechst 33258 (blue). Time-course showing infiltration of an epithelial gap (white arrow) by Lucifer yellow over a 15 min period and restitution by 25 min.
Fig. 4. Acriflavine-stained villous epithelial cells. Single cell shedding from the small intestinal epithelium over time (white arrow).
7 Analysis of Epithelial Cell Shedding and Gaps in the Intestinal Epithelium
113
4. Notes 1. The glass rod is used to protect the mucosa of the intestine beneath from being damaged by cautery. A Pasteur pipette bent in a Bunsen burner is easy to make (Fig. 1). 2. TNF-a administration causes an increase in cell shedding and gap formation. In order to study this process, it is beneficial to administer TNF-a in some cases. 3. It is important that the mouse is kept warm at all times throughout the procedure. A heated stage or heated incubator where the temperature can be modified to keep the body temperature of the mouse at 37°C is essential. 4. Hoechst 33258 and alexa-dextran 647 are administered intravenously. Acriflavine and Lucifer yellow are topically administered to the mucosal surface. 5. Confocal reflectance images may also be taken to confirm that gaps contain an impermeable substance that is not stained by other dyes. 6. Sufficient anesthesia is maintained throughout by intramuscular injections of ketamine and medetomidine. Continual assessment of anesthesia and body temperature is required throughout the procedure. 7. Acriflavine stains both nuclei and cytoplasm and has a very broad emission spectrum. It is currently one of only two dyes that are licenced for use in humans, allowing a direct comparison between mouse and man. Both the confocal probe and confocal endoscope are currently only able to detect dyes excited by a 488-nm solid-state laser (Fig. 4). 8. The pen confocal probe is used in a very similar fashion to that used by the human endoscope and in fact has the same optics. 9. The pen confocal probe images provide an ideal way to observe epithelial gaps, which are readily distinguishable from goblet cells (Fig. 2).
Acknowledgments The authors would like to thank Dr Ralph Kiesslich from The University of Mainz, Germany for adapting the confocal endoscopy technique in the human.
114
C.A. Duckworth and A.J. Watson
References 1. Potten, C. S., and Loeffler, M. (1990) Stem cells: attributes, cycles, spirals, pitfalls and uncertainties. Lessons for and from the crypt, Development 110, 1001–1020. 2. Watson, A. J., Chu, S., Sieck, L., Gerasimenko, O., Bullen, T., Campbell, F., McKenna, M., Rose, T., and Montrose, M. H. (2005) Epithelial barrier function in vivo is sustained despite gaps in epithelial layers, Gastroenterology 129, 902–912. 3. Kiesslich, R., Goetz, M., Angus, E. M., Hu, Q., Guan, Y., Potten, C., Allen, T., Neurath, M. F., Shroyer, N. F., Montrose, M. H., and Watson, A. J. (2007) Identification of epithelial gaps in human small and large intestine by confocal endomicroscopy, Gastroenterology 133, 1769–1778.
4. Kaur, P., and Potten, C. S. (1986) Effects of puromycin, cycloheximide and noradrenaline on cell migration within the crypts and on the villi of the small intestine. A model to explain cell movement in both regions, Cell Tissue Kinet 19, 611–625. 5. Potten, C. S. (1998) Stem cells in gastrointestinal epithelium: numbers, characteristics and death, Philos Trans R Soc Lond B Biol Sci 353, 821–830. 6. Piguet, P. F., Vesin, C., Guo, J., Donati, Y., and Barazzone, C. (1998) TNF-induced enterocyte apoptosis in mice is mediated by the TNF receptor 1 and does not require p53, Eur J Immunol 28, 3499–3505.
Chapter 8 Studying Permeability in a Commonly Used Epithelial Cell Line: T84 Intestinal Epithelial Cells Rino P. Donato*, Adaweyah El-Merhibi*, Batjargal Gundsambuu, Kai Yan Mak, Emma R. Formosa, Xian Wang, Catherine A. Abbott, and Barry C. Powell Abstract The integrity, or barrier function, of the intestinal epithelium is of paramount importance in maintaining good health. This is largely imparted by a single layer of epithelial cells linked by the transmembrane tight junction protein complex near their apical surface. Disruption of epithelial permeability via the tight junctions can contribute to disease progression. The cytokine IFNg is involved in many inflammatory processes and has been shown to dramatically increase permeability via changes at the tight junction in experimental models. One of its key effectors is the transcription factor, IRF-1. In our studies of the role of IRF-1 in barrier function using the human T84 intestinal epithelial cell monolayer model, we have found that induction of IRF-1 alone is insufficient to change permeability and that if IRF-1 is involved in mediating the permeability effects of IFNg, then other factors must also be required. Key words: T84 monolayer, Transepithelial electrical resistance, FITC-dextran flux, Immuno fluorescence, IFNg, IRF-1, Cytokine, Transfection
1. Introduction The intestinal epithelium plays two crucial roles in intestinal physiology. It mediates the regulated absorption of nutrients from the gut lumen and, at the same time, provides a highly selective barrier to prevent invasion of underlying tissues by pathogens and toxins in the gut. These actions are carried out by a single-layered epithelium largely comprised of absorptive cells and a smaller number of secretory cells; enteroendocrine cells, goblet cells, * These authors contributed equally to this work. Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_8, © Springer Science+Business Media, LLC 2011
115
116
R.P. Donato et al.
and Paneth cells. Its barrier function capability is remarkable, considering the dynamic state of the epithelium; with the exception of the longer lived Paneth cells, the other epithelial cells are renewed every 3–6 days. The epithelial cells are linked to their neighbours near their apical surface by a transmembrane protein complex known as the tight junction, which provides a paracellular barrier that is selectively permeable to ions and macromolecules (1). Its properties are subject to modification by a variety of cytokines, other biological and chemical agents, and pathogens (2–4). An understanding of the molecular regulation of the tight junction is important, as its disruption can lead to an increase in permeability and contribute to disease progression (5, 6). From a different perspective, the tight junction and epithelial permeability are important in determining the efficacy of oral drug administration (7). Model epithelial cell lines derived from the intestine can be cultured as monolayers to mimic the intestinal epithelium and can provide insights into the physiological and molecular characteristics of the tight junction and epithelial permeability. T84 cells are a widely used cell line derived from a human colon carcinoma that differentiate spontaneously at confluence to form polarized monolayers with well-formed tight junctions. They resemble adult colonic crypt cells in their morphology, tight junctions, and ion transport characteristics (8–10). Unlike Caco-2 cells, which are another well-studied intestinal cell line from a permeability perspective, T84 cells are less prone to differentiate into sublines with altered characteristics. Barrier function can be readily studied in T84 cells using a permeable transwell support system in which the T84 cells are grown as monolayers. This enables measurement of permeability by a variety of techniques, of which the most common and simplest are transepithelial electrical resistance (TER) and FITCdextran flux, which measure transport of ions and macromolecules, respectively (11). Agents that alter permeability may act by a variety of mechanisms, for example by altering protein synthesis and degradation, interaction, relocalization, or phosphorylation status. Western blotting and confocal microscopy are two powerful complementary techniques that can be used to investigate these processes. Another powerful approach in studying the functional importance of specific proteins is to manipulate their activities in stable cell lines. This chapter describes methods to study the effect of cytokines on the permeability of intestinal epithelial cell monolayers using, by way of example, the cytokine IFNg, the early response effector transcription factor, IRF-1 (12), and the tight junction protein occludin. Methods for the establishment of stably transfected T84 cell lines are also provided.
8 Studying Permeability in a Commonly Used Epithelial Cell Line…
117
2. Materials 2.1. Cell Culture
1. T84 cells can be obtained from the American Type Culture Collection (Manassas, VA, USA) http://www.atcc.org/. 2. Dulbecco’s Modified Eagle’s Medium (DMEM) and Ham’s F12 Nutrient Mixture (1:1) supplemented with 10% (v/v) foetal bovine serum (SAFC Bioscience, Lenexa, KS, USA) and penicillin–streptomycin stabilized solution (100 U/mL penicillin and 100 mg/mL streptomycin final concentrations (Sigma-Aldrich, St Louis, MO, USA)) is used as the primary growth medium for T84 cells. 3. Dulbecco’s Phosphate-Buffered Saline (DPBS) and 10× Trypsin solution (0.5% Porcine trypsin and 0.2% EDTA: SAFC Bioscience, Lenexa, KS, USA) are used to passage the cells. 4. Trypan blue solution (0.4% w/v) and a Haemocytometer (Hirschmann Em Techcolor, Eberstadt, Germany) are used to count the cells. 5. Dimethyl sulphoxide (DMSO) (Sigma-Aldrich, St Louis, MO, USA) is used to freeze cells down. 6. Mr. Frosty cryofreezing container (Nalgene, Thermo Fisher Scientific, Waltham, MA, USA). 7. Human recombinant interferon-g (IFNg) expressed in E. coli (Sigma-Aldrich, St Louis, MO, USA) is reconstituted in DPBS to a concentration of 500 mg/mL. Store as aliquots at −20°C and discard after thawing. Final concentration of IFNg, diluted in media, that is used in permeability experiments is typically 10–100 ng/mL.
2.2. Measuring Paracellular Permeability
1. A Millicell-ERS volt-ohm meter with “chop stick” electrodes (Millipore, Billerica, MA, USA). Transwells (Fig. 1) with polyester (PET) membrane clear inserts with a pore size of 0.4 mM (6.5 mm and 24 mm insert diameter: Corning Life Sciences, Acton MA, USA). 2. 10 kDa FITC-dextran (Sigma-Aldrich, St Louis, MO, USA), Hank’s Balanced Salt solution (HBSS: SAFC Bioscience, Lenexa, KS, USA), black F96 MicroWellTM Plate (Nunc, Thermo Fisher Scientific, Waltham, MA, USA), and Wallac Victor-2 1420 Multilabel Counter (excitation at 485 nm and emission at 535 nm) (Perkin-Elmer, Waltham, MA. USA).
2.3. Transfection of T84 Cells
1. Lipofectin, Plus reagent, Opti-Mem reduced serum media, Blasticidin, and Geneticin (Invitrogen, Carlsbad, CA, USA). 2. Control DNA: GFP vector (Invitrogen, Carlsbad, CA, USA).
118
R.P. Donato et al. Electrodes
Transwell Insert
Upper media compartment Epithelial cell monolayer Microporous membrane Lower media compartment
Fig. 1. Schematic of transwell. Tight junctions between cells in the monolayer are represented by = symbol.
3. Growth medium: DMEM:Hams F12 (Sigma, St. Louis, MO, USA). FBS (SAFC Bioscience, Lenexa, KS, USA) and penicillin–streptomycin stabilized solution (100 U/mL penicillin and 100 mg/mL streptomycin final concentrations (Sigma-Aldrich, St Louis, MO, USA)) (see Note 1). 4. Trypsin (Sigma, St. Louis, MO, USA). 5. Dulbecco’s Phosphate-Buffered Saline (DPBS; Sigma, St. Louis, MO, USA). 6. Cloning ring (Sigma, St. Louis, MO, USA). 2.4. Cell Lysis and Sample Preparation for Western Blotting
1. RIPA Lysis buffer: 50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 0.1% (w/v) SDS, 1 mM EDTA, and 1% Triton X-100. To 10 mL of lysis buffer, add one protease inhibitor tablet (Roche Diagnostics GmbH, Mannheim, Germany) (see Note 2). Aliquot and store at −20°C. 2. Protein determination: Pierce BCA Protein Assay kit (Thermo Fisher Scientific, Waltham, MA, USA). 3. SDS-PAGE Loading buffer (5×): prepare a solution of 125 mM Tris–HCl, pH 6.8, 40% (v/v) Glycerol, 5% (w/v) SDS, 0.1% (v/v) Bromophenol blue, and 200 mM b-mercaptoethanol. Store in aliquots at 4°C. 4. IRF-1 Recombinant protein (P01; Abnova, Neihu district, Taiwan).
2.5. SDSPolyacrylamide Gel Electrophoresis
1. 1.5 M Tris–HCl, pH 8.8 for separating buffer. Store at room temperature. 2. 0.5 M Tris–HCl, pH 6.8 for stacking buffer. Store at room temperature.
8 Studying Permeability in a Commonly Used Epithelial Cell Line…
119
3. Thirty percent bis-acrylamide solution (Bio-Rad, Hercules, CA) and N,N,N,N ¢-Tetramethyl-ethylenediamine (TEMED; Sigma, St. Louis, MO, USA) (see Note 3). 4. Ammonium persulphate (APS; Sigma, St. Louis, MO, USA): 10% (w/v) solution made in Milli-RO water. Make 10% APS fresh weekly and store at 4°C. 5. SDS (Sigma, St. Louis, MO, USA): 10% (w/v) solution made in Milli-RO water. Store at room temperature (see Note 4). 6. 10% Tween 20: 10% (v/v) solution made in Milli-RO water. 7. Running buffer (1×): 25 mM Tris base, 192 mM Glycine, 0.1% (w/v) SDS. Store at room temperature. 8. Pre-stained molecular weight marker: Precision Plus Protein Kaleidoscope Marker (Bio-Rad, Hercules, CA, USA). 2.6. Western Blotting: Detection of IRF-1
1. Transfer buffer (1×): 25 mM Tris base, 192 mM Glycine, and 20% (v/v) methanol. Store at 4°C until use. Place in transfer apparatus with cooling device (see Note 5). 2. PVDF HybondTM-P membrane (Amersham; GE Healthcare Bio-sciences AB, Uppsala, Sweden) and Whatman blotting paper (Whatman plc, Kent, UK). 3. Tris-buffered saline (TBS): prepare a 10× stock solution of 1.486 M NaCl; prepare a second 10× stock solution of 0.5 M Tris–HCl, pH 7.6. For a working solution, combine 100 mL of 10× NaCl stock, 100 mL of 10× Tris stock, and 800 mL Milli-RO water. 4. Tris-buffered saline with Tween (TBS-T): prepare TBS as above. For every 1 L of TBS, add 5 mL 10% Tween 20. 5. Blocking buffer: prepare a 12% (w/v) skim milk powder solution using TBS (see Note 6). 6. Antibody dilution buffer: primary antibody; TBS-T. Secondary antibody; TBS-T with 0.5% (w/v) skim milk powder. 7. Primary antibody: rabbit anti-IRF-1 (C-20; polyclonal antibody; Santa Cruz Biotechnology, Santa Cruz, CA, USA). 8. Secondary antibody: goat anti-rabbit, peroxidase conjugated (Calbiochem, EMD Chemicals, Gibbstown, NJ, USA). 9. Developing kit: Supersignal West Femto maximum sensitivity substrate kit (Thermo Fisher Scientific, Waltham, MA, USA). 10. Syngene G:Box (Syngene, Cambridge, UK) or similar equipment for recording chemiluminescent images.
2.7. Stripping and Reprobing Blots: b-Tubulin
1. Stripping buffer: prepare a solution consisting of 2% (w/v) SDS, 62.5 mM Tris–HCl, pH 6.7 and 100 mM b-mercaptoethanol. Store at room temperature (see Note 7). 2. Wash buffer: TBS and TBS-T.
120
R.P. Donato et al.
3. Antibody dilution buffer: for primary antibody; TBS-T with 5% (w/v) skim milk powder: for secondary antibody; TBS-T with 1% (w/v) skim milk powder. 4. Primary antibody: mouse anti b-tubulin (monoclonal antibody; Santa Cruz Biotechnology, Santa Cruz, CA, USA). 5. Secondary antibody: goat anti-mouse, peroxidase conjugated (Calbiochem; EMD Chemicals, Gibbstown, NJ, USA). 2.8. Immuno cytochemical Detection: Occludin
1. Fixing Solution: 1:1 (v/v) of methanol and acetone. Store at −20°C (see Note 8). 2. Blocking buffer: 3% (w/v) skim milk powder in 1× PBS. 3. Antibody dilution buffer: 1% (w/v) skim milk powder in 1× PBS. 4. Permeabilization buffer: 0.1% (v/v) Triton X-100 in 1× PBS. Store at room temperature. 5. Primary antibody: mouse anti-occludin (polyclonal antibody; Invitrogen, Carlsbad, CA, USA). 6. Secondary antibody: goat anti-mouse Alexa 555 (Invitrogen, Carlsbad, CA, USA). 7. Mounting medium: Dako, Glostrup, Denmark. 8. Confocal microscope. We used a BioRad Radiance 2100 Confocal Microscope (BioRad, Hercules, CA, USA).
3. Methods In studying barrier function in T84 monolayers, a number of factors need to be taken into account. In the IFNg cytokine model, for example, the timing, duration, and dose of cytokine all affect the permeability status of the monolayer (13–16). These parameters need to be selected at the start of the study. At the doses typically used, IFNg affects monolayer permeability in the absence of increased apoptosis or necrosis (15); however, depending on the experimental circumstances, it may be desirable to test these parameters and this can be easily performed with commercially available kits (17). Cell characteristics and responses can change over time with increasing passage number. Studies should be conducted using cells over a defined, narrow range of passage (e.g. 5–10 passages). TER and FITC-dextran flux are two different measures of permeability that measure different parameters. They can be regulated independently (11). Treatments that alter permeability may act by a variety of mechanisms, changing protein characteristics in the cell, including protein synthesis and degradation,
8 Studying Permeability in a Commonly Used Epithelial Cell Line…
121
interaction, relocalization, and phosphorylation status. Western blot and confocal microscopy are two powerful and complementary techniques that can be used to investigate these processes. Whereas the western blot provides semi-quantitative assessment of protein abundance, the confocal microscope is capable of very fine spatial resolution of proteins in a cell. By capturing images at incremental submicron steps through a cell, it is possible to build up a model of protein localization in three dimensions. Key requirements in implementing these techniques successfully are good antibodies and methods for preserving the protein epitopes detected by them. An important consideration in studying the effect of specific genes on permeability in T84 cells is that it takes 1–2 weeks to establish a monolayer with good permeability characteristics. To avoid effects on formation of the monolayer and the permeability barrier, studies should be performed using an inducible gene expression system where the gene can be switched on when the barrier is established. Before choosing an inducible system, any reported leakiness of the system should be investigated. A system with tight regulation of gene expression is essential for analysis of inducible protein function on permeability. In creating the first cell line with the regulatory gene, for example the Tet repressor gene of the T-REx system (17), a number of independent lines should be established and compared to select the highest level of regulatory gene expression. In our experience in T84 cells, a 50-fold difference in expression levels of inserted genes is not uncommon. Additionally, the strengths of various gene promoters should be tested in the T84 cells, as some promoters may not work as well as others. This may impact on the choice of inducible system, requiring re-engineering of the promoters or choice of a different system. Mycoplasma infection is a perennial risk in cell culture that can distort cellular responses. Regular testing is an essential requirement when conducting cell culture studies (18). It is common to add N-terminal or C-terminal tags to track a protein of interest, particularly when there is a need to distinguish it from the endogenous protein. Positioning of tags (e.g. the FLAG or myc tag) may affect protein function and this should be investigated prior to choosing a vector or an inducible system. Note that in our experience anti-6HIS antibody binds to two high molecular weight bands on a western blot of native T84 protein and thus the 6HIS tag should be used with caution in these cells. Because of the slow growth of T84 cells, it is advisable, where practical, to first test protein modifications in a more easily transfectable cell line, such as the MDCK kidney epithelial cell line, which is also used to study permeability and in which a regulated inducible gene expression system such as the T-REx tetracycline system is effective (17).
122
R.P. Donato et al.
3.1. Cell Culture: Establishment of Monolayers for Permeability Studies
1. Thaw cryopreserved T84 cells in a 37°C water bath and immediately seed into a tissue culture flask containing pre-warmed media. Maintain cells in 25 cm2 or 75 cm2 tissue culture flasks, containing 5 or 10 mL normal growth or appropriate media, respectively. Change media every 2 days. 2. For routine passaging, split T84 cells at ~70% confluence. (a) To split cells, remove the media by aspiration and rinse cells twice with Dulbecco’s PBS solution. Add 1× trypsin solution to the cells (2 mL for a 75 cm2 flask, 1 mL for a 25 cm2 flask) followed by incubation in a 37°C, 5% CO2 humidified incubator until all cells are detached (approximately 10–20 min). (b) Once cells are detached, add four volumes of fresh media to the cells to inactivate the trypsin. (c) Pipette cells to resuspend and split equally into three 75 cm2 tissue culture flasks. To split cells grown on tissue culture plates, add a sufficient volume of 1× trypsin solution to cover the cells. Reseed the cells in fresh media. Gently swirl the flask/wells to obtain even coverage and return to the incubator. 3. To seed cells at specific densities, detach cells as described above. (a) To determine cell density, remove 25 mL of the cell suspension and mix with 25 mL of 0.08% trypan blue solution. (b) Load 10 mL of this cell suspension onto a haemocytometer and view under a phase contrast microscope. Count the number of cells in 5 × 5 fields. Cells with intact membranes, indicated by trypan blue dye exclusion, are considered the viable count. Trypan blue stained cells are considered non-viable and are not counted. (c) To obtain a cell suspension at a desired density, centrifuge the suspension at 250 × g force for 5 min in a sterile 10-mL or 50-mL tube and resuspend the pellet in an appropriate volume of cell growth medium. 4. For cryopreservation, detach cells as described above. (a) Transfer the cell suspension to a sterile 10 mL or 50 mL centrifuge tube and pellet the cells by centrifugation at 250 × g force for 5 min. (b) Resuspend the cell pellet in an appropriate volume of freezing medium (95% cell culture medium and 5% DMSO (v/v)). For a T75 flask of cells resuspend in 3–5 mL.
8 Studying Permeability in a Commonly Used Epithelial Cell Line…
123
(c) Transfer the cell suspension into cryovials (1 mL per vial) and place in a cryofreezing container (Mr. Frosty) at −70°C overnight to cool the cells slowly. (d) Place the cryovials into liquid nitrogen for long-term storage (e.g. years). 5. Establishing monolayers. Seed T84 cells onto 6.5 mm or 24 mm transwell membranes at a density of 1 × 106 cells/cm2 (see Note 9). Typically, add 0.5 mL of media to the basolateral compartment and 150 mL to the apical compartment. Change media every 2 days. Perform experiments on monolayers once a high stable TER, indicative of established tight junctions is reached. For T84 monolayers this is approximately 1,000–3,500 W∙cm2 (ohms per square centimetre) which may take 7–14 days to attain after initial seeding (see Note 10). 6. For maximal effect of IFNg on permeability of T84 monolayers grown on transwells, add IFNg at final concentration of 100 ng/mL (1,000 U/mL) to the media in the basolateral transwell compartment (see Note 11). The IFNg receptor is located on the basolateral surface. Change media every 24 h, adding fresh IFNg (see Note 12). 1. Measure transepithelial electrical resistance (TER) in triplicate in transwells using a Millicell-ERS volt-ohm meter with “chopstick” electrodes. Equilibrate transwell plates at room temperature for 20 min before taking measurements (Fig. 2a). Express TER as W∙cm2 (see Note 13).
TER (% of control)
a
b
TER 120 100 80 60 40 20 0
24 hr *
IFNγ
48 hr
*
72 hr Tet
Flux (nM/cm2/hr)
3.2. Measuring Paracellular Permeability in T84 Monolayers
10 kDa FITC-Dextran Flux 350 300 250 200 150 100 50 0
* 24 hr 48 hr * * Untreated
IFNγ
72 hr Tet
Fig. 2. Effect of IFNg and tetracycline (Tet) on permeability in T84 monolayers. (a) IFNg (100 ng/mL) causes a timedependent decrease in permeability, whereas tetracycline has no effect. (b) IFNg (100 ng/mL) causes a small but significant increase in the flux of 10 kDa FITC-dextran after 24 and 48 h, but a large increase at 72 h. Tetracycline has no effect. Tetracycline is often used in inducible gene expression systems and was tested here to show that it does not affect permeability in T84 monolayers. T84 monolayers were grown until a stable TER was attained, then treated with 100 ng/mL of IFNg for up to 72 h. TER was measured daily. N = minimum of three replicates. One-way ANOVA (Tukey’s test) was used to determine statistical significance. Asterisk denotes statistical significance (p < 0.05) from untreated monolayers (set at 100% TER in (a)).
124
R.P. Donato et al.
2. Measure paracellular flux of 10 kDa FITC-dextran as follows (Fig. 2b): (a) Wash monolayers on 6.5 mm diameter transwell inserts once with HBSS, adding 200 mL of HBSS to the apical compartment, 500 mL to the basolateral compartment. (b) Replace with 200 mL of HBSS in the apical compartment and 500 mL in the basolateral compartment and place the cells in a 37°C humidified incubator with 5% CO2 for 30 min to equilibrate. (c) Remove the HBSS by vacuum aspiration. (d) Add 500 mL of fresh HBSS to the basolateral transwell compartment and 200 mL of 1 mg/mL 10 kDa FITCdextran solution in HBSS to the apical compartment. (e) Place transwells in the incubator for 1 h, then mix the HBSS in the lower chamber by pipetting and take 200 mL of that solution and transfer to a black F96 MicroWell™ Plate. (f) Remove the remaining HBSS in the basolateral compartment by aspiration and replace with 500 mL of fresh HBSS. (g) Return the transwell plate to the incubator and repeat the process at the end of each hour for the next 3 h so that four 200 mL samples are collected. (h) Measure the fluorescence of the samples using a Wallac Victor-2 1420 Multilabel Counter (excitation at 485 nm and emission at 535 nm). Convert fluorescence values to concentrations of 10 kDa FITC-dextran using a standard curve of twofold serially diluted 10 kDa FITC-dextran. Set up the standard curve in duplicate, starting at 2 mg/mL and diluting twofold seven times. Use three wells containing HBSS alone for a blank. The average concen tration of the four time samples, divided by the surface area gives the transepithelial flux, which is expressed in nM/cm2/h following the method of Sanders et al. (19). 3.3. Transfection of T84 Cells: Establishing Stable Cell Lines
1. Transfect T84 cells in suspension to establish stable cell lines (see Note 14). The methodology described is for seeding transfected cells into six well plates which are used to obtain monoclonal stable cell lines. This method may be scaled up or down using the guidelines indicated on the Lipofectin and Plus reagent product inserts. 2. Wash cells in T75 flask with 2 × 5 mL PBS. 3. Add 2 mL pre-warmed 1× trypsin and return to the incubator for 10 min. 4. Using a pipette, transfer cells into a 10-mL tube.
8 Studying Permeability in a Commonly Used Epithelial Cell Line…
125
5. Pellet cells at 300 × g force/5 min. 6. Resuspend cells in 2–5 mL of Opti-Mem reduced serum media (the volume will depend on how confluent the T75 flask is) and count cells as described in Subheading 3.1 step 3b (see Note 15). 7. Plate cells at a density of 2 × 106 cells/well in a final volume of 800 mL Opti-Mem reduced serum media and return to the incubator. 8. Per reaction, dilute 20 mL Lipofectin in 100 mL Opti-Mem reduced serum media and incubate at room temperature for 15 min. 9. During this incubation period dilute DNA (2.5–5 mg) in 100 mL Opti-Mem reduced serum media. 10. After the 15 min incubation period, add 20 mL Plus reagent to the diluted DNA and incubate at room temperature for 15 min (at this stage Lipofectin incubation = 30 min). 11. Combine the diluted Lipofectin with the DNA/Plus reagent complex and incubate at room temperature for 15 min. 12. Once the incubation is complete add the transfection cocktail (Lipofectin/DNA/Plus reagent complex) to the cells. 13. Incubate the cells at 37°C/5% CO2 for 5 h. During this incubation, return twice to the cells and very gently pipette up and down to disperse them (see Note 15). 14. After the 5 h incubation period, add 2 mL normal growth medium to the cells and gently mix by pipette. Return the cells to the incubator (37°C/5% CO2). 15. Do a full media change with normal growth medium 24 h post transfection. Add 2 mL normal growth medium per well. Assess GFP controls for fluorescence using an inverted microscope with the appropriate filter. 16. Add antibiotics 48 h post transfection to select for transfected clones (see Note 16). For example, for selection of T84 TetR transfectants using the T-REx system (Invitrogen) start blasticidin at 2.5 mg/mL for 2 days, then increase to 4 mg/mL then 6 mg/mL. For selection of T84 transfectants with the gene of interest, start geneticin at 200 mg/mL for 2 days and then increase to 300 mg/mL, and then 400 mg/mL. Stable cell lines should be maintained with selective antibiotics in the media. 17. Blasticidin works faster than geneticin. However, by 2 weeks small colonies should be evident. 18. To make monoclonal cell lines, colonies must be observed regularly to prevent merging and the forming of a mixed population. If necessary, scrape smaller colonies away from larger colonies with a sterile tip, wash cells with 2 mL PBS/ well then change media.
126
R.P. Donato et al.
19. Once colonies are large enough (see Note 17), cloning rings may be used to pick colonies for growth in a 96-well plate. 20. To pick colonies, first wash cells twice with 2 mL PBS/well. 21. Apply a sterile cloning ring with sterile Vaseline on the base around the colony. 22. Add 100 mL 1× trypsin and incubate at 37°C/5% CO2 for ~20 min (see Note 18). 23. Remove cells and place in a 10-mL tube. Using 100 mL PBS rinse inside the cloning ring transferring the resultant solution to the 10-mL tube. 24. Add 2 mL media to the 10-mL tube and spin cells down at 300 × g force/5 min (see Note 19). 25. Remove all media except ~50 mL, as a pellet will not be evident. Gently resuspend cells in this remaining volume and plate in a 96-well plate. Add an additional 150 mL normal growth media to the well. Do not add selective antibiotics at this stage as the cells will struggle to recover when antibiotics are in the media. 26. 24 h post-seeding, change media on the cells, add selective antibiotics and check under the microscope that cells have adhered. 27. Transfer cells in the following order: 96 well, 48 well, 24 well, 6 well, T25, T75 (see Note 20). 28. Cells can be frozen down at the 24-well stage by transferring cells from the 48-well plate to 2 wells in a 24-well plate. One well can be continued for growth and the other frozen down. Cells should also be frozen down at the T25 and T75 stage as a precaution against subsequent loss. 29. It can take 3–4 months to create stable cell lines using T84 cells, as they grow very slowly when present at low density, and 7–8 months if an inducible system is required where the cell line is doubly transfected; first with the regulatory gene construct then that line is transfected with the gene of interest. Figure 3 shows examples of normal and abnormal T84 colonies. 30. Test cell lines for expression of gene of interest by quantitative real-time RT-PCR and by western blot (see Note 21). 3.4. Cell Lysis and Sample Preparation for Western Blotting
1. Harvest cells using RIPA lysis buffer. The volume to use will depend on the size of the wells/flasks and the density of cells. For example, use 100 mL for a well in a 96-well plate. 2. Determine protein concentration using a BCA kit. 3. Combine 30 mg of protein with the required volume of 5× loading buffer.
8 Studying Permeability in a Commonly Used Epithelial Cell Line…
127
Fig. 3. Images of antibiotic-resistant IRF-1 transfected colonies (in TetR/T84 cell line) after several weeks of selection compared to normal untransfected T84 cells. (a) Subconfluent normal T84 cells; (b) Subconfluent IRF-1 #10 C cell line; (c) “Normal” IRF-1 colony, 100× magnification; (d) “Normal” IRF-1 colony, 40× magnification; (e) “Abnormal” IRF-1 colony, 100× magnification; (f) Another “abnormal” IRF-1 colony, 40× magnification. Morphology of cells in (b)–(d) is similar to normal T84 cells; the cells in these colonies are densely packed and have well-defined edges. (e) The cells in this abnormal colony are spread, are less well defined, and are irregular in appearance. These types of colonies do not last long in culture. (f) The centre of this colony appears to have normal looking cells; however, the cells on the outer edges are irregular in appearance. This will impact on the growth/spread of the colony. A colony of this appearance generally dies off.
4. Incubate samples at 100°C for 5 min (see Note 22). Spin samples briefly to collect at the bottom of the tube. 5. Whilst protein samples are incubating, prepare the IRF-1 standard curve using the IRF-1 recombinant protein (twofold dilutions). 6. The samples are ready for SDS-polyacrylamide gel electrophoresis (SDS-PAGE).
128
R.P. Donato et al.
3.5. SDS-PAGE
1. The methodology which follows uses a Bio-Rad Mini Protean tank set-up and equipment, however, this method is easily adaptable to other systems. 2. Clean both glass plates with 70% EtOH (see Note 23). Assemble plates in casting mould and attach to gel holding stand. Check for leakage using Milli-Q water. 3. Prepare a 1-mm thick 10% separating gel by mixing, in the following order: 3.965 mL Milli-Q water, 2.5 mL 1.5 M Tris–HCl buffer, pH 8.8, 3.33 mL 30% bis-acrylamide solution, 100 mL 10% SDS, 100 mL 10% APS, and 5 mL TEMED (final volume is 10 mL – increase for more than one gel). 4. Pipette between 6 and 7 mL of gel solution between the glass plates leaving room for the stacking gel. Top with a few millilitres of Milli-Q water so the gel sets without drying out. Allow the gel to set at room temperature (~30 min). 5. Once the gel has set remove the water and make a 4% stacking gel by mixing, in the following order: 2.969 mL Milli-Q water, 1.26 mL 0.5 M Tris–HCl buffer, pH 6.8, 666 mL 30% bis-acrylamide solution, 50 mL 10% SDS, 50 mL 10% APS and 5 mL TEMED per gel. 6. Pipette the stacking gel on top of the separating gel and insert the comb between the plates (see Note 24). Allow the gel to set at room temperature (~30 min). 7. Whilst the gel is setting prepare the Running buffer (as described in Subheading 2.5 item 7). 8. Once the gel is set, remove from the casting and place into tank. Pour 1× SDS running buffer into the running tank. Fill both the inner and outer chambers. 9. Gently remove the comb from the gel and, using a syringe and needle, flush the wells before loading samples. 10. Load samples into the wells. Use 5 mL of marker and 2 mL of loading buffer to fill in any blank wells. 11. Place the lid onto the tank and attach to the BIORAD Power PAC 200. Run at 100 V (voltage constant); amps = 2 (highest); ~1–1.5 h or until dye front has run to the end of the gel.
3.6. Western Blotting: IRF-1
1. Before the dye front reaches the end of the gel prepare the PVDF membrane for transfer. Cut out a piece of membrane (~9 × 7 cm). PVDF membranes need to be pre-wet before using in transfer procedures. Soak the membrane in 100% methanol for 10 s, rinse in Milli-Q water for 5 min and then soak the membrane in transfer buffer for 10 min. It is now ready to use.
8 Studying Permeability in a Commonly Used Epithelial Cell Line…
129
2. During the running time also pre-soak Whatman blotting paper and sponges in transfer buffer. You will need three pieces of blotting paper for each side. 3. Once the gels have run, transfer the proteins to a membrane. Rinse the gel in Milli-Q water for 5 min followed by transfer buffer for 5 min before setting up the transfer “sandwich”. 4. Fill the transfer tank with transfer buffer and add an ice pack and magnetic flea. 5. Place the plastic grid white side down and layer in the following order: sponge, three pieces of blotting paper, prepared PVDF membrane (make sure there are no air bubbles – roll out if necessary); polyacrylamide gel (make sure that there are no air bubbles – roll out if necessary), three pieces of blotting paper, and sponge. Close the grid and clamp. Place into the holder in the transfer tank and top up the tank with transfer buffer. 6. Place the lid on the tank. Ensure that black is facing black and white is facing red for correct migration of protein. 7. Sit the tank on a stirrer and allow to transfer for 1 h at 100 V (maximum amps = 2). 8. After the transfer has finished, depending on time constraints, store the membranes in TBS at 4°C or probe immediately with antibodies. 9. If probing with antibodies rinse in TBS (2 × 5 min/shaking). 10. Add Blocking Buffer (12%) for 40 min/shaking. 11. Rinse membrane in TBS-T (2 × 5 min/shaking). 12. Add primary antibody (IRF-1; 1:200) and incubate shaking/ rotating at room temperature for 3 h or 4°C overnight (see Note 25). For the secondary-only control add the same volume of TBS-T. 13. Rinse membrane in TBS-T (2 × 5 min/shaking). 14. Add secondary antibody to all membranes (test and control; goat anti-rabbit; 1:2,000) and incubate shaking/rotating for 40 min. 15. Rinse membrane in TBS-T (2 × 5 min/shaking). 16. Rinse membrane in TBS (2 × 5 min/shaking). 17. Develop membrane using Supersignal West Femto maximum sensitivity substrate kit and Syngene G:Box (Syngene, Cambridge, UK). Figure 4a, b show examples of tetracycline induction of IRF-1 in monolayers of two different IRF-1 stable cell lines compared to IFNg induction of IRF-1 in normal T84 monolayers. Figure 4c shows the effect of the different doses of IFNg used in Fig. 4a on TER responses of normal T84 monolayers.
130
R.P. Donato et al.
a
T84
b
IRF-1 #10C
IFNγ (ng/mL) UT
1
2.5
5
T84 10
UT
+Tet
IRF-1#1
UT
UT
+Tet
IRF-1 β-tubulin
c TER (percent of control)
120 100 *
80
* *
60
*
* *
24hr 48hr 72hr
* *
*
40 20 0
1
2.5
5 IFNγ ng/mL
10
100
Fig. 4. Induction of IRF-1 in untransfected and transfected T84 monolayers. (a) Normal T84 monolayers were treated for 24 h with increasing doses of IFNg (1–10 ng/mL) or, a monolayer of a clonal IRF-1 stable cell line (#10C), was induced for 24 h with tetracycline (+Tet). Protein was extracted and equal amounts electrophoresed on SDS-PAGE. After transfer the blot was first probed for IRF-1 (top panel) stripped and reprobed for b-tubulin (bottom panel ). The IRF-1 was not completely stripped (= lower band in bottom panel). Note that low levels of IFNg are detectable in untreated monolayers (UT) and that a dose as low as 1 ng/mL IFNg causes robust induction of IRF-1 at 24 h. IRF-1 induction in the IRF-1#10C line is minimal after 24 h tetracycline treatment. (b) Tetracycline causes robust induction of IRF-1 at 2 h in a polyclonal IRF-1 cell line (IRF-1#1). (c) Effect of IFNg dose on permeability in T84 monolayers. No significant change in TER was observed until after 24 h exposure to IFNg and, for the lowest dose, 1 ng/mL, not until after 48 h. As robust induction of IRF-1 was observed at 1 ng/mL at 24 h (a), this suggests that if IRF-1 is involved in the IFNg effect on TER, that other factors induced by IFNg must also be required. In all these studies, T84 monolayers were grown until a stable TER was attained, and then they were treated with varying doses of IFNg for 72 h. TER was measured daily. N = minimum of three replicates. One-way ANOVA (Tukey’s test) was used to determine statistical significance. Asterisk denotes statistically significant difference (p < 0.05) to untreated monolayers.
3.7. Stripping and Reprobing Blots: b-Tubulin
1. In a fume hood, pour stripping buffer onto the membrane until submerged. 2. Place tray on a shaker and incubate for 15 min/shaking at room temperature. 3. Rinse membrane in TBS-T/ (2 × 5 min/shaking). 4. Rinse membrane in TBS (2 × 5 min/shaking). 5. Reblock membrane with Blocking Buffer for 40 min. 6. The membrane is ready to be reprobed as previously described.
8 Studying Permeability in a Commonly Used Epithelial Cell Line…
131
a
b
IRF-1 recombinant protein standard curve
Band volume
25000000 y = 159005x-577899 R2= 0.995
20000000 15000000 10000000 5000000 0 0
20
40
60
80
100
120
140
Protein (ng) Fig. 5. IRF-1 recombinant protein standard curve. (a) Serial dilutions of IRF-1 recombinant protein were blotted and detected with IRF-1 antibody. (b) Standard curve of linear range of immunodetection of IRF-1 recombinant protein.
7. Primary antibody (b-tubulin; 1:200); incubate shaking/ rotating at room temperature for 2 h. 8. Secondary antibody (goat anti-mouse; 1:10,000); incubate shaking/rotating for 40 min. 9. Band intensity is measured using the Syngene GeneTools software that is provided with the Syngene Gel Doc. 10. Open western blot files in the above program. 11. Select manual quantification. 12. Select automatic background substitution. 13. Measure the volume of bands for the standard protein, loading control, and protein of interest following directions outlined in the program (see Note 26). 14. Prepare a standard curve based on the values obtained for the standard protein in Microsoft excel. 15. Normalize the protein of interest using the average value obtained for the loading control. 16. Determine the concentration of the protein of interest using the equation obtained from the standard curve. Figure 5 shows an IRF-1 dilution western blot and the standard curve. 3.8. Immuno cytochemical Detection: Occludin
1. T84 cells are seeded onto transwells and media changed every 2 days. Cells are maintained for ~14 days until steady-state TER is achieved (see Subheading 3.1 step 5).
132
R.P. Donato et al.
2. Once TER reaches the appropriate level (or depending on the purpose of the experiment, the cells are treated with the relevant agent), aspirate the cell culture media from both upper and lower compartments of the transwell (see Note 27). 3. Wash the transwells twice with ice-cold 1× PBS to remove the culture media completely (see Note 28 and 29). 4. Add 100 mL of Fixing Solution into the transwell insert only and incubate for 15 min at room temperature (see Note 29). 5. Discard the Fixing Solution (into hazardous waste container) and rinse cells once with 1× PBS. 6. Permeabilize cells with PBS/Triton X-100 solution for 3 min at room temperature (see Note 30). 7. Wash cells with 1× PBS (2 × 5 min). 8. Add 100 mL of blocking buffer for either 1 h at room temperature or 4°C overnight. 9. Wash the cells with 1× PBS (2 × 5 min). 10. If required, the membrane can be cut into smaller pieces to test different antibodies on the same monolayer (see Note 31). 11. Add 100 mL of diluted primary antibodies (see Note 32) in Antibody Dilution Buffer and incubate for 1 h at room temperature or overnight at 4°C. For the secondary only control, add the same volume of 1× PBS. 12. Wash the cells with 1× PBS (4 × 5 min). 13. Add 100 mL of diluted fluorescent-conjugated secondary antibody, for example Alexa 555 goat anti-rabbit/mouse or Alexa 488 goat anti-rabbit/mouse (Invitrogen, Carlsbad, CA, USA) to all transwell inserts (test and control) and incubate for 1 h at room temperature (see Note 33). 14. Wash the cells with PBS (4 × 5 min). 15. For nuclear staining, add Hoechst Stock Solution to a final concentration of 1 mg/mL (Invitrogen, Carlsbad, CA, USA) and incubate for 10 min at room temperature (see Note 34). 16. Cut out the insert membrane from the transwell with a scalpel blade (see Note 35). 17. Put the membrane on a microscopic slide, with cell side up. 18. Drop the mounting medium onto the membrane and put the coverslip on top of it (see Note 36). 19. Visualize staining by confocal microscopy. An example of occludin staining in T84 monolayers is shown in Fig. 6.
8 Studying Permeability in a Commonly Used Epithelial Cell Line…
133
Fig. 6. Localization of the tight junction protein occludin in a monolayer of normal T84 cells by confocal microscopy. (a) En face (x–y plane) image showing occludin sharply localized at cell boundaries and presenting a typical “honeycomb” staining pattern. (b) Occludin staining (arrows) at the cell periphery in the vertical (z) dimension showing continuous vertical “stripes” of occludin staining at the tight junction. The composite image shows a section of the cell spanning about 4 m. Note that occludin staining ceases (arrowhead ) towards the basal surface of the cell. Composite images were made from multiple z-sections using the public domain Java image processing program, ImageJ (http://rsb.info.nih.gov/ij/).
4. Notes 1. If using a tetracycline regulated inducible gene system (e.g. T-REx: Invitrogen, Carlsbad, CA, USA), tetracycline-free FBS is recommended to prevent any leaky expression. 2. If interested in detecting the phosphorylation status of proteins also include a phosphatase inhibitor such as PhosSTOP (Roche Diagnostics, Mannheim, Germany). 3. Bis-acrylamide is highly toxic. Proper precautions (personal protective equipment) should be adhered to when using this chemical. 4. SDS may precipitate out. If this happens warm solution to redissolve SDS. 5. A cooling device is required to prevent over heating of the gel and apparatus.
134
R.P. Donato et al.
6. We use a commercially available brand of skim milk powder (Black and Gold) found in South Australia, however, any brand should suffice. 7. Do not make large volumes of stripping buffer at any one time as b-mercaptoethanol will decompose over time. 8. The most suitable fixing solution can vary depending on the molecules to be detected and needs to be established for the protein under study. Some fixing solutions can work for a variety of epitopes. The specified conditions have been used in our hands for staining of tight junction molecules, occludin, claudin-2, ZO-1 and E-cadherin, and b-catenin (17). Treatment with 100% ice-cold ethanol for 15 min at 4°C, followed by incubation with acetone for 3 min at room temperature has also proved effective in our hands in detecting these proteins. 9. Observe the membranes of transwells closely. Ill-fitting membranes may compromise permeability measurements such as TER and FITC-dextran flux. 10. Reasons for monolayers not establishing a high TER may include slow growing cells, unhealthy cells, apoptosis, necrosis, ill-fitting membranes, suboptimal incubator conditions, old media, or cells in need of a media change. 11. Maximal increase in permeability generally occurs 72 h after start of treatment. 12. Once a brand of IFNg has been chosen it is advisable not to switch between brands. Record when a new batch is being used, as this may impact on the results obtained. Lower concentrations of IFNg can also be used, albeit with some reduced effect on permeability (Fig. 4c) (16). 13. Take care not to damage the monolayers when inserting the electrode into the upper chamber. 14. Cells are transfected in suspension due to the low transfection efficiency when transfecting adherent T84 cells. It is not necessary to use subconfluent cells with this method. Once transfected, T84 cells do not like to be split too severely. When creating stable monoclonal cell lines, it is recommended that transfections are done in six-well plates. If polyclonal cell lines are required, 24-well or 48-well plates should suffice. 15. You will need to work quickly as the cells will begin to clump when there is no serum present or in reduced serum media (i.e. Opti-Mem). 16. A dose response/kill curve should be performed for each antibiotic against the cells. Invitrogen provide a method for performing a geneticin dose response/kill curve on their website: (http://www.invitrogen.com/site/us/en/home/
8 Studying Permeability in a Commonly Used Epithelial Cell Line…
135
support/FAQ.html – search for Geneticin dose response/kill curve). This method also gives antibiotic dose ranges for blasticidin and zeocin. Control untransfected cells should be treated with antibiotic alongside transfected cells to ensure effective killing. 17. Colonies can be sized based on field of view under a 4× objective. If they are ~1/2 to 3/4 of the field of view, colonies can be picked. 18. Because the cells in the colonies are denser than plated cells, it will take longer for the trypsin to work. 19. Trypsin needs to be removed, given the cells will eventually be plated in only 200 mL in a 96-well plate. 20. T84 cells like company; if the density of the cells in the well is too low (<10%), they will slow down in growth or may not grow at all. 21. At least two cell lines exhibiting the same characteristics must be studied to preclude integration-specific effects. 22. Place a heavy weight over the Eppendorf tubes to prevent them from popping during the incubation. To prevent tubes from popping when taking them off the heating block, take them off a few at a time and hold lids down firmly for a few seconds. 23. Glass plates must be thoroughly cleaned after use so that no acrylamide gel is left on the plates to dry, as this will interfere with the running of the proteins through the gel. 24. Avoid air bubbles; otherwise, the wells will be compromised. 25. The IRF-1 antibody produces a good signal when the membrane is probed at room temperature. If there are time constraints the membrane may be probed at 4°C overnight. 26. If bands are not clearly visible, invert the image so bands are black and the background is white. 27. Extra care has to be taken while solutions are being removed from the transwell insert. Do not touch the microporous membrane during aspiration because that could damage the cell monolayer and the membrane. 28. At the end of the ice-cold PBS wash, the transwell insert can be taken out from the well to remove the rest of the PBS. Place it in a new 24-well plate. 29. From this point, the cells need only be treated from the upper compartment of the transwell. 30. This step is optional; permeabilization is performed when cytosolic molecules like F-actin are co-stained with transmembrane proteins.
136
R.P. Donato et al.
31. Remove the membrane completely from the transwell insert and put on a microscope slide, cell side up. Hold the scalpel blade perpendicular to the membrane and press it down to cut. Do not move it back and forward as there is a risk of damaging the monolayer. Place the membrane pieces in the wells of a 96-well plate for staining with different antibodies. Make sure that the cell side of the membrane faces up. 32. The following primary antibodies have proved effective for immunocytochemistry and confocal microscopy in epithelial cell monolayers in our hands (17, 20): anti-occludin, antiZO-1, anti-claudin 2 (Zymed Laboratories, San Francisco, CA, USA), anti-E-cadherin, and b-catenin (BD Biosciences, San Jose, CA, USA). Antibodies were diluted in Antibody Dilution Buffer according to the manufacturers’ instructions. 33. Various dilutions (from 1:100 to 1:2,000) are typically tested to determine the optimal dilution of fluorescently labelled secondary antibodies. 34. Care should be taken in handling and disposal of Hoescht staining solution, as it can bind to DNA and disrupt replication. Consequently, it is potentially mutagenic and carcinogenic. 35. Use a new scalpel blade to remove the membrane from the transwell insert. Take the transwell insert from the plate and hold upside down with one hand and make a smooth cut around the edge of the membrane. Do not take the glued rim of the membrane on the transwell insert as the edge is usually rough and folded and could cause problems when the coverslip is applied. 36. Seal the coverslips on the slide with nail polish for long-term storage.
Acknowledgements This study was supported in part by Dairy Australia and the Women’s and Children’s Hospital Foundation. R. Donato was supported by a Flinders University Faculty of Science and Engineering Research award. References 1. Tsukita, S., Furuse, M., and Itoh, M. (2001) Multifunctional strands in tight junctions, Nat Rev Mol Cell Biol 2, 285–293. 2. Capaldo, C. T., and Nusrat, A. (2009) Cytokine regulation of tight junctions, Biochim. Biophys. Acta 1788, 864–871.
3. Al-Sadi, R., Boivin, M., and Ma, T. (2009) Mechanism of cytokine modulation of epithelial tight junction barrier, Front Biosci 14, 2765–2778. 4. Schulzke, J. D., Ploeger, S., Amasheh, M., Fromm, A., Zeissig, S., Troeger, H., Richter,
8 Studying Permeability in a Commonly Used Epithelial Cell Line…
5.
6.
7.
8.
9.
10.
11.
12.
13.
J., Bojarski, C., Schumann, M., and Fromm, M. (2009) Epithelial tight junctions in intestinal inflammation, Ann N Y Acad Sci 1165, 294–300. Clayburgh, D. R., Shen, L., and Turner, J. R. (2004) A porous defense: the leaky epithelial barrier in intestinal disease, Lab Invest 84, 282–291. Marchiando, A. M., Graham, W. V., and Turner, J. R. (2010) Epithelial barriers in homeostasis and disease, Annu Rev Pathol 5, 119–144. Deli, M. A. (2009) Potential use of tight junction modulators to reversibly open membranous barriers and improve drug delivery, Biochim. Biophys. Acta 1788, 892–910. Dharmsathaphorn, K., McRoberts, J. A., Mandel, K. G., Tisdale, L. D., and Masui, H. (1984) A human colonic tumor cell line that maintains vectorial electrolyte transport, Am J Physiol 246, G204–208. Dharmsathaphorn, K., and Madara, J. L. (1990) Established intestinal cell lines as model systems for electrolyte transport studies, Methods Enzymol 192, 354–389. Hillgren, K. M., Kato, A., and Borchardt, R. T. (1995) In vitro systems for studying intestinal drug absorption, Med Res Rev 15, 83–109. Harhaj, N. S., and Antonetti, D. A. (2004) Regulation of tight junctions and loss of barrier function in pathophysiology, Int. J. Biochem. Cell Biol. 36, 1206–1237. Schroder, K., Hertzog, P. J., Ravasi, T., and Hume, D. A. (2004) Interferon-gamma: an overview of signals, mechanisms and functions, J Leukoc Biol 75, 163–189. Adams, R. B., Planchon, S. M., and Roche, J. K. (1993) IFN-gamma modulation of epithelial barrier function. Time course, reversibility,
14.
15.
16.
17.
18.
19.
20.
137
and site of cytokine binding, J Immunol 150, 2356–2363. Colgan, S. P., Parkos, C. A., Matthews, J. B., D’Andrea, L., Awtrey, C. S., Lichtman, A. H., Delp-Archer, C., and Madara, J. L. (1994) Interferon-gamma induces a cell surface phenotype switch on T84 intestinal epithelial cells, Am J Physiol 267, C402–410. Bruewer, M., Luegering, A., Kucharzik, T., Parkos, C. A., Madara, J. L., Hopkins, A. M., and Nusrat, A. (2003) Proinflammatory cytokines disrupt epithelial barrier function by apoptosis-independent mechanisms, J Immunol 171, 6164–6172. Youakim, A., and Ahdieh, M. (1999) Interferon-gamma decreases barrier function in T84 cells by reducing ZO-1 levels and disrupting apical actin, Am J Physiol 276, G1279–1288. Donato, R., Wood, S. A., Saunders, I., Gundsambuu, B., Yan Mak, K., Abbott, C. A., and Powell, B. C. (2009) Regulation of epithelial apical junctions and barrier function by Galpha13, Biochim. Biophys. Acta 1793, 1228–1235. Drexler, H. G., and Uphoff, C. C. (2002) Mycoplasma contamination of cell cultures: Incidence, sources, effects, detection, elimination, prevention, Cytotechnology 39, 75–90. Sanders, S. E., Madara, J. L., McGuirk, D. K., Gelman, D. S., and Colgan, S. P. (1995) Assessment of inflammatory events in epithelial permeability: a rapid screening method using fluorescein dextrans, Epithelial Cell Biol 4, 25–34. Sander, G. R., Cummins, A. G., Henshall, T., and Powell, B. C. (2005) Rapid disruption of intestinal barrier function by gliadin involves altered expression of apical junctional proteins, FEBS Lett 579, 4851–4855.
wwwwwwwwwwwww
Chapter 9 Optimization of the Caco-2 Permeability Assay to Screen Drug Compounds for Intestinal Absorption and Efflux Barry Press Abstract In vitro permeability assays are a valuable tool for scientists during lead compound optimization. As a majority of discovery projects are focused on the development of orally bioavailable drugs, correlation of in vitro permeability data to in vivo absorption results is critical for understanding the structural– physicochemical relationship (SPR) of drugs exhibiting low levels of absorption. For more than a decade, the Caco-2 screening assay has remained a popular, in vitro system to test compounds for both intestinal permeability and efflux liability. Despite advances in artificial membrane technology and in silico modeling systems, drug compounds still benefit from testing in cell-based epithelial monolayer assays for lead optimization. This chapter provides technical information for performing and optimizing the Caco-2 assay. In addition, techniques are discussed for dealing with some of the most pressing issues surrounding in vitro permeability assays (i.e., low aqueous solubility of test compounds and low postassay recovery). Insights are offered to help researchers avoid common pitfalls in the interpretation of in vitro permeability data, which can often lead to the perception of misleading results for correlation to in vivo data. Key words: Caco-2, Permeability, P-gp, Efflux, Absorption, Bioavailability, Enterocyte, Epithelial monolayer
1. Introduction The Caco-2 model is one of the most extensively utilized in vitro assays for permeability and efflux screening within the pharmaceutical industry (1, 2). For chemical series in which transporter substrate issues are not of immediate concern, less laborious methods such as artificial membrane assays and/or computationalbased predictions have also been gaining in popularity in recent years (3, 4). Oral delivery is without question the most desirable route of drug administration. Following oral administration, a drug can be absorbed into the systemic circulation via passive and
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_9, © Springer Science+Business Media, LLC 2011
139
140
B. Press
Apical
P-gp 1
2
3
4
5
Basolateral Fig. 1. Multiple pathways and mechanisms for intestinal absorption of a compound across the intestinal epithelium. Following oral administration, a drug can be absorbed by passive diffusion via paracellular routes between cells (1), transcellular transport across the cell membrane (2), or via active carrier-mediated (receptor) uptake (3), and transcytosis mechanisms (4). Apical-localized efflux pumps such as P-gp and MRP-2 serve as a protective barrier by recycling xenobiotics back out to the intestinal lumen (5), thereby decreasing absorptive potential of a molecule. (Intestinal, first-pass metabolism can also act to decrease absorption via enzymes such as CYP3A4).
active transport routes across the intestinal wall. Lipophilic compounds can traverse the plasma membrane unrestricted, allowing the transcellular pathway to dominate (see Fig. 1). For hydrophilic drugs and most peptides, the paracellular (intercellular) pathway becomes the dominant transport route, as these polar classes of compounds are relatively impermeable across the cell membrane (5). In addition to the passive types of diffusion, active (energy-requiring) mechanisms also exist at the intestinal epithelia. Caco-2 cells possess many primary-like qualities of enterocytes, both functional and morphological. Importantly, Caco-2 cells are able to fully polarize into differentiated monolayers displaying brush border (microvilli) regions and cell–cell (tight junctions) upon culturing over a 21-day period (6). The tight junctions and lipophilic makeup of the intestinal epithelium serve as a barrier to the absorption of drugs that are administered orally. Although the intestinal epithelium in vivo is composed of a variety of different cell types (including endocrine, exocrine, and tuft cells), it is the enterocyte cell that dominates at the intestinal wall. As a result, cell-based systems such as Caco-2 have been shown to be predictive of transcellular-based absorption (7). In addition to passive transport routes, the intestinal epithelium can transport substances via mechanisms that require energy expenditure. These energy-dependent mechanisms include active transport and a recycling process known as efflux (8, 9). Understanding both the strengths and limitations of the Caco-2 assay is essential for making the best use of these screening data during lead optimization and compound advancement.
9 Optimization of the Caco-2 Permeability Assay to Screen Drug Compounds…
141
2. Materials 2.1. Cell Culture and Monolayer Maintenance
1. Caco-2 cells purchased from American Type Culture Collection (ATCC, Rockville, MD) should be utilized between passage numbers 25 and 35. 2. Caco-2 aliquots should be frozen for liquid nitrogen storage within a narrow passage range in freeze media containing 95% fetal bovine serum and 5% DMSO. Aliquots are stable at −80°C for up to 1 month before transfer to liquid nitrogen vapor phase. 3. Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 20% fetal bovine serum (FBS), 2 mM l-glutamine, 100 U/mL penicillin, 100 mg/mL streptomycin, and 2 mg/ml Amphotericin B, all purchased from Invitrogen Corp., Gibco (Grand Island, NY). DMEM should be stored at 4°C, and all media supplements are stored at −20°C, preferably in aliquots (to minimize freeze–thaw cycles). 4. Trypsin-EDTA (0.25% (w/v)-1 mM EDTA), and Hank’s balanced salt solution containing MgCl2 and CaCl2 (HBSS, cat. #10-527F, or equivalent) can be purchased from Lonza Corporation (Basel, Switzerland). 5. Cell culture transwell inserts (24-well) were purchased from BD BioSciences (PET membrane, 1.0 mm, cat. #351181) or Costar (polycarbonate membrane, 0.4 mm, cat. #3396). 6. Enhanced recovery plates (for basolateral wells) can be purchased from BD BioSciences (cat. #453600) to reduce nonspecific binding of highly lipophilic compounds.
2.2. Caco-2 Assay Reagents
1. Permeability Assay Buffer (PAB): Hank’s Balanced Salt Solution (HBSS) containing 1.0 mM MgCl2, 1.0 mM CaCl2, 10 mM d(+)glucose and 20 mM HEPES (pH 7.4 ± 0.1), and stored at room temperature (or, 4°C) for up to 1 month. 2. PAB containing Bovine Serum Albumin (BSA): Use 4% BSA (w/v) in HBSS with 0.22 mm filtering (use of prefilter also recommended), pH 7.4 ± 0.1. 3. Wash buffer: HBSS containing 1.0 mM MgCl2 and 1.0 mM CaCl2. 4. Voltmeter: EVOMX voltammeter with an STX100F electrode (World Precision-Instruments, Sarasota Florida), or equivalent, to monitor transepithelial electrical resistance (TEER) readings to check monolayer integrity.
2.3. Reference Control Compound and Inhibitor Stocks
1. Stock solutions of control compounds metoprolol, digoxin, propranolol, naproxen, caffeine, taurocholic acid, furosemide, ranitidine, nadolol, lisinopril, and enalaprilat can all be purchased
142
B. Press
from Sigma, (St. Louis, MO) and prepared in DMSO at a final concentration of 10 mM. Atenolol stock solutions are prepared in DMSO at 100 mM, and all DMSO stock solutions should be stored at −20°C for up to 1 month. 2. For a 100× stock solution of Lucifer Yellow (20 mg/ml), LY is prepared in HBSS and filter-sterilized using a 0.2-mm syringe filter after heating to 37°C for 30 min. Lucifer Yellow stocks should be stored at −20°C for up to 3 months. 3. Positive control inhibitors verapamil, cyclosporin A (CsA), MK-571, quercetin, fumitremorgin C (FTC), ketoconazole, and leukotriene C4 were also purchased from Sigma and prepared in DMSO at a final concentration of 10 mM (or 20 mM) and stored at −20°C for up to 1 month. 2.4. HPLC and Mass Spectrometry Conditions
1. Generic conditions for UPLC analysis, an Acquity UPLC BEH C18 2.1 × 50 mm × 1.7 mm column and a XBridge C18 2.1 × 50 mm × 5 mm column (Waters Corporation, Milford, MA) are used. 2. Generic Conditions: Injection volumes are 5 mL for the Acquity UPLC. The mobile phase flow rate is 1.0 ml/min with a 3:1 post column split. 3. Both HPLC-MS/MS systems use the same mobile phase and gradient conditions. The HPLC mobile phase gradient was initialed at 2% eluent B, followed by a linear increase to 100% eluent B over 0.3 min, held at 100% B for 0.6 min, and reconditioned at 2% eluent B for 0.1 min. Eluent A was 0.1% formic acid in water for +ESI. Eluent B was 0.1% formic acid in acetonitrile for +ESI. UPLC column temperatures are maintained at 50°C.
3. Methods Since large differences have been reported in the quantitative measurements of apparent permeability (Papp) coefficients from different laboratories, it is critical to analyze Papp values of new compounds in relation to a reference calibration curve, established in-house with commercial drugs (of known human absorption) exhibiting a wide range of permeability coefficients (see Fig. 2). In our experience, intraday variability in Caco-2 permeability measurements can be as high as 20% (percent coefficient of variation). Apparent permeability rate coefficients are generally expressed as: Papp (cm/s) = Amount transported/(Area × Initial concentration × Time), not taking into account recovery levels that may be low as a result of membrane accumulation and/or
% Human Absorption
9 Optimization of the Caco-2 Permeability Assay to Screen Drug Compounds…
propranolol
100 80
verapamil
caffeine
metoprolol naproxen
ketoprofen
60
143
40 20 0
0
5
10
15
20
25
Caco-2 Papp Rate (x10
30 -6
35
40
45
50
cm/sec)
Fig. 2. Relationship between Caco-2 permeability and human intestinal absorption. It is important to interpret Papp rate coefficients of new drug compounds in relation to a reference calibration curve, established with commercial drugs of known human absorption (and exhibiting a wide range of permeability). The shallow slope seen above for highly permeable compounds (Papp > 10) and the steep slope observed for drugs with lower permeability (Papp < 2) are typical for in vitro permeability assays (see Note 1). Taurocholic acid exhibits a Papp value of 5.5 in the above graph, and the remaining six compounds, not labeled in the above graph, all represent lower permeable compounds: atenolol, furosemide, ranitidine, nadolol, lisinopril, and enalaprilat.
nonspecific binding. To better predict correlation to in vivo conditions, the calculation of Papp values using mass terms (mass balance) is recommended for highly lipophilic compounds, allowing for differences in postassay recovery levels. Low permeability rate measurements usually indicate that the absorptive potential of a test compound is not optimal. However, such low permeability data may also suggest that the compound was simply not recovered upon completion of the assay. The presence of bovine serum albumin (BSA) in the basolateral (serosal) assay buffer can mimic the physiological levels of human albumin within the capillary lumen. Since plasma protein binding can significantly influence the ability of lipophilic compounds to permeate cellular monolayers, the inclusion of BSA at levels up to 4% can markedly increase the permeability and postassay recovery levels of these compounds by providing a more physiological ‘sink’ condition (driving force) for partitioning through the cell monolayer. In this manner, BSA can reduce the accumulation of hydrophobic compounds in the cell membrane and may better reflect the in vivo permeability rates of these drugs. As a result, the mass balance of many lipophilic compounds can be improved in the presence of BSA, thereby rendering permeability measurements more reliable for screening programs. However, such protein sink conditions do not allow for efflux ratio measurements for test compounds exhibiting high levels of protein binding. In addition, LC/MS sample preparation is considerably more tedious for buffers containing high levels of BSA. 3.1. Seeding Monolayers
1. Cells between passages 25 and 35 (from ATCC original passage number 18) are cultured in 75 cm2 flasks at 37°C in a humidified atmosphere containing 5.0% CO2 (see Note 1).
144
B. Press
2. Caco-2 cultures were passaged every week at a split ratio of 1:10 or seeded in the range of 0.25–1.0 × 106 cells in a T-75 cm2 flask. 3. Following an initial washing step with 10 mL PBS, cells were harvested from the flasks with a 0.25% trypsin-EDTA/HBSS solution for 10 min at 37°C. 4. Caco-2 cells were seeded on Day 1 at a density of 75,000 cells/cm2 (a range of 50,000–100,000 cells/cm2 is acceptable) on high density PET membrane inserts or Transwell polycarbonate inserts, (0.4 or 1.0 mm pore size, 0.33 or 0.31 cm2 surface area, respectively), see Note 2. 5. Culture media was changed every other day (or every third day) for 21 days. Cell monolayers are suitable for use between 21 and 28 days post seeding. 3.2. Standard Assay Design
1. To certify cell monolayers before each experiment, monolayer integrity needs to be determined by measuring the transepithelial electrical resistance (TEER). TEER values should be measured in the permeability assay buffer, PAB, using an EVOMX voltammeter with an STX100F electrode, or equivalent. 2. The potential difference is expressed as the TEER (W*cm), after subtraction of the intrinsic resistance of the model (i.e., the resistance obtained over the cell-free inserts). Wells containing monolayers with TEER values >200 Ω*cm2 were included in the permeability study, and those with TEER values below 200 Ω*cm2 were not used. A monolayer with a low TEER was assumed to exhibit extensive leakage through imperfect occluding junctions or holes in the monolayer. 3. For each drug studied, both “apical to basolateral” (A–B) and “basolateral to apical” (B–A) bidirectional transport routes are routinely investigated in order to determine if carrier-mediated transport or carrier-mediated efflux mechanisms are involved. Compounds are diluted in PAB to achieve a final concentration of 1–10 mM (often depending on the sensitivity of the bioanalytical instrumentation), and a final solvent concentration < 1.0% dimethylsulfoxide (DMSO). Lower drug concentrations should be used when aqueous solubility is a problem (or to measure efflux substrate potential), and changes must be noted in the Papp rate calculations (see Note 3). 4. When using [14C)- or [3H)-radiolabeled compounds, final specific activity should be approximately 1.0 mCi/mL. The “apical to basolateral” assay is initiated by adding 0.25 mL drug solution to the apical side of the monolayer and 1.0 ml PAB (pH 7.4) to the basolateral side of the monolayer. The “basolateral to apical” assay is performed in a similar
9 Optimization of the Caco-2 Permeability Assay to Screen Drug Compounds…
145
manner, except that 0.25 mL PAB (pH 7.4) is added to the apical side and 1.0 mL drug solution is added to the basolateral (serosal) side. 5. When screening predominantly basic or acidic compounds, it is best to run the Caco-2 assay under nongradient pH conditions (i.e., pH 7.4 buffer should be used in both the apical and basolateral buffers). The use of a pH gradient can overestimate efflux for basic compounds, and underestimate efflux for acidic compounds, due to differential ionization of the compound in acidic (pH 6.0–6.5) versus neutral buffers (10). 6. For Caco-2 assays utilizing a pH gradient, metoprolol cannot be included, since it is completely ionized under acidic conditions resulting in an artificially low A–B Papp value and a correspondingly high (artificial) efflux ratio. When the test compound in question is expected to be a substrate for protoncoupled transporters (such as the oligopeptide, lactic acid, and short-chain fatty-acid transporters), the use of a pH gradient is necessary for proper functioning of these transporters. 7. Permeability experiments should be conducted in triplicate (n = 3), and aliquots of dosing solution from the donor well should be taken at the beginning and end of the assay, to allow for recovery measurements. 8. The cells were allowed to preincubate in permeability assay buffer for 20 min at 37°C (atmospheric CO2) before initiating the assay. 9. Performing the Caco-2 assay on an orbital shaker or rocker (approximately 30 rpm) is also recommended to reduce the influence of the unstirred water layer (UWL) which can be present under static conditions. In the absence of stirring, the UWL in vitro can be up to ten times the thickness of the UWL measured at the epithelium in vivo (i.e., 1,000 mm vs. 100 mm). 10. For one-time point analysis, 90 min is commonly used. Time points of 30, 60, and 120 min are also acceptable, being careful to adjust Papp calculations accordingly. It is advised not to exceed 120 min for this type of rate assay, to ensure that the final time point is well below the time at which equilibrium may be approached, and to ensure monolayer integrity in the absence of serum. 11. Sample aliquots taken from the donor solution are typically diluted approximately tenfold in acetonitrile, before running LC/MS analysis. Samples taken from the receiver wells at the end of the assay are routinely diluted in an equal volume of acetonitrile before initiating LC/MS analysis. 12. It is critical to check for postassay recovery levels of each new chemical entity (NCE) that is screened in the Caco-2 system.
146
B. Press
To avoid erroneously classifying a low-recovery compound as having poor permeability, Papp rate measurements should only be considered reliable for post-assay recovery levels >60% (with some laboratories assigning a minimum cutoff criteria of 65% or 70%). 13. Recovery is calculated based upon the percent remaining of total compound in the donor well at the end of the assay (plus the receiving well time point, “T ”), expressed as: {(Cf + T)/Ci × 100} and following correction for sample volumes in each chamber. 14. Apparent Permeability Calculations: The apparent permeability rate coefficients (Papp) are calculated according to the following equation: Papp = (V/(A × Ci))×(Cf/T ), where V is the volume of the receptor chamber (1.0 ml), A is the area of the membrane insert (0.31 or 0.33 cm2), Ci is the initial dosing concentration of drug (1 mM or 10 mM), Cf is the final concentration of drug in the receiving well (mm), and T is the assay time (seconds). To convert from units of “10−6 cm/s” to units of “nm/s”, simply multiply the Papp coefficients by a factor of 10. 3.3. Assay Controls
1. The quality control (QC) checks of assay plates used in permeability studies are typically conducted in two steps. The first step consists of measuring the suitability (TEER value) of each assay well of the plate (as described above). TEER values are dependent not only on the surface area of the monolayer, but also on the type of semi-porous support well (i.e., polyethylene terephthalate or polycarbonate membrane) and cell passage number. 2. The second QC step consists of testing both high and low permeability reference compounds. Common reference compounds used to assess the suitability are atenolol, Lucifer Yellow, or mannitol (for low permeability) and metoprolol, caffeine, or naproxen (for high permeability). It is recommended to include a negative control (internal standard) in every dosing sample at the start of the assay to quantitatively confirm monolayer integrity (see Note 4). 3. For each assay plate tested, the permeability coefficient for atenolol or mannitol needs to be within a specified range to be certified as “acceptable”. Additionally, Papp values for low and high permeability control compounds must meet their respective permeability classification to be certified as “acceptable”. 4. Following the study, the Papp coefficient of atenolol (or mannitol) must be less than 1 × 10−6 cm/s (“low”), and metoprolol (or naproxen) must exhibit a Papp value greater than 10 × 10−6 cm/s (“high”) to ensure that monolayer integrity was maintained throughout the course of the assay (see Notes 5 and 6).
9 Optimization of the Caco-2 Permeability Assay to Screen Drug Compounds…
3.4. Screening Molecules as Substrates for P-gp, BCRP, and MRP-2 Efflux Pumps
147
1. Perform the initial Caco-2 screening assay as described above (Subheading 3.2). Drug compounds exhibiting moderate or high efflux ratios (B–A Papp/A–B Papp ratio >3 or >10) should be retested for potential efflux liability (see Note 7). 2. P-gp has broad substrate requirements (neutral or positive charge with higher degrees of hydrophobicity). MRP-2 complements the substrate specificity of P-gp, since MRP-2 preferentially binds negatively charged (acidic) molecules (11). Breast cancer resistance protein (BCRP) is related to the MRP family and has overlapping substrate requirements similar to P-gp and MRP-1. 3. For secondary (efflux) screening of drug compounds, testing should be conducted in the presence of a tenfold excess of a positive control inhibitor (12). The positive control inhibitor should be included in both the apical and basolateral buffers, for both A–B and B–A transport routes (see Fig. 3).
Vinblastine Papp (10E-6 cm/s)
a
MRP-2 (Vinblastine Permeability) A-B B-A
20 15 10 5 0 Control
b
MK-571
BCRP (Mitoxantrone Permeabililty) 15
Mitoxantrone Papp (10E-6 cm/s)
Leukotriene C4
A-B B-A
10
5
0 Control
Fumitremorgin C
Quercetin
Fig. 3. Effect of efflux pump inhibitors on probe substrate permeability. Drug compounds exhibiting high efflux ratios can be retested for potential efflux pump substrate liability in the presence of a tenfold excess of a known inhibitor. The percentage change in the efflux ratio (± inhibitor) is then determined. (a) For MRP-2 screening, vinblastine (1 mM) is a model substrate, together with the leukotriene D4 receptor antagonist MK-571 or leukotriene C4 as inhibitors (10 mM). (b) For BCRP screening, mitoxantrone (1 mM) is a model substrate, together with fumitremorgin C (FTC) or the flavonoid quercetin as inhibitors (10 mM).
148
B. Press
4. Preincubate the cell monolayers for 15 min (at 37°C) in permeability assay buffer (containing the inhibitor), to observe the maximal potential effect of the efflux pump being tested. 5. The percentage change in the efflux ratio of the test compound (± inhibitor) is then determined. For screening purposes and rank ordering, a change of 50% or greater in the efflux ratio is usually considered significant, especially if a concomitant increase in A–B (absorptive) permeability is observed for the test compound. 6. Decrease in efflux ratio (% Inhibition) = (1−(RI − 1)/ (R − 1)) × 100; where RI equals the efflux ratio of the marker P-gp substrate (or test compound) in the presence of inhibitor, and R equals the efflux ratio of the marker P-gp substrate (or test compound) in the absence of inhibitor. 7. For P-gp screening, the substrates digoxin and saquinavir (1 mM) are typically used as substrate positive controls. Verapamil is a well-characterized inhibitor of P-gp as well as cyclosporin A (13). 8. For MRP-2 screening, the substrates vinblastine, furosemide, or methotrexate (1 mM) can be used as positive controls. The leukotriene D4 receptor antagonist MK-571 is a commonly utilized inhibitor of MRP-2, as well as leukotriene C4. (Note that many lots of Caco-2 cells express low levels of MRP-1, so the inhibitory effects of MK-571 on MRP-1 are often minimal). 9. For BCRP screening, the substrates mitoxantrone and sulfasalazine are commonly utilized as substrate positive controls. Fumitremorgin C (FTC) and the flavonoid quercetin are used as inhibitors of BCRP. 3.5. Screening Molecules as Inhibitors of P-gp, BCRP, and MRP-2 Efflux Pumps
1. Perform the initial Caco-2 screening assay as described above (Subheading 3.2), and also see Notes 8–10. 2. For secondary (“Tier 2”) testing to determine the potential of a new compound to inhibit P-gp, BCRP, or MRP-2 transporters, the positive control substrates discussed in the previous section should be used as marker (probe) substrates (14). 3. Depending on the aqueous solubility of the compound being tested for efflux pump inhibition, it is recommended to test the compound at two concentrations (at a tenfold increase or higher over the probe substrate). Typically, the compound being tested should be assayed at 10 mM and 50 mM (or 100 mM), if aqueous solubility permits. 4. Preincubate the cell monolayers for 15-min (at 37°C) in permeability assay buffer containing both concentrations of the compound being tested, to observe the maximal potential effect for the compound to inhibit the efflux transporter in question.
9 Optimization of the Caco-2 Permeability Assay to Screen Drug Compounds…
149
Table 1 Digoxin permeability as a marker substrate to screen drug compounds as inhibitors of P-gp efflux across Caco-2 monolayers. Assessing the change in the permeability profile of digoxin (1 mM) in the presence of a known P-gp inhibitor (verapamil, CsA, KTZ, as shown above) or varying concentrations of a test compound is performed to determine the potential of a molecule to be an inhibitor of P-gp. Typically, a tenfold higher concentration (10 mM) of compound (or positive control) is utilized for initial screening Inhibitor
Digoxin Papp (A–B)
Digoxin Papp (B–A)
Efflux ratio
% Inhibition
None (control)
0.5
12.4
24.8
N/A
Verapamil
5.0
6.1
1.2
95%
Cyclosporin A(CsA)
6.4
6.5
1.0
100%
Ketoconazole (KTZ)
5.0
7.2
1.4
94%
Papp: Apparent Permeability Coefficient (× 10 cm/s), n = 3. Values of permeability are expressed as mean, and the coefficient of variation in the permeability data (both directions) is less than 20% among replicates −6
5. The percentage change in the efflux ratio of the marker substrate (± test compound) is then determined. For screening purposes and rank ordering, a change of 50% or greater in the efflux ratio–permeability profile of the marker substrate is considered significant, especially if a concomitant increase in A–B (absorptive) permeability is observed for the marker substrate (see Table 1). 3.6. Permeability Screening Utilizing BSA (Protein-Sink Conditions)
1. Perform the initial Caco-2 screening assay as described above (Subheading 3.2), using the following modifications: 2. Include 4% BSA (w/v) in the basolateral permeability assay buffer (PAB), and see Notes 11 and 12. 3. To ensure a consistent matrix across all samples being analyzed, two standard curves need to be generated. For apicaldosed samples, the standard curve should be generated from a solution of 4% BSA/PAB, while recovery samples should be measured against a standard curve generated from PAB without BSA (see Note 13). 4. Once sampling is complete, add an equal volume of acetonitrile followed by vortexing and then centrifugation at 10,000 rpm (15,000 × g) for 10 min to pellet the precipitated protein. 5. Being careful to avoid the protein pellet, transfer the top half of sample volume to a separate plate/vial for LC/MS analysis. 6. Only A–B (absorptive) transport rates should be measured (see Table 2). Efflux ratios should not be determined for Caco-2 assays performed under protein sink conditions since protein binding of the test compound to BSA can result in
150
B. Press
Table 2 Inclusion of BSA in the basolateral buffer significantly improves permeability and postassay recovery levels of lipophilic compounds in a Caco-2 Assay. BSA was included in the basolateral buffer at a final concentration of 4% to act as a protein ‘sink’. The inclusion of BSA does not affect the A–B Papp rate coefficient of the P-gp substrate digoxin. Postassay recovery levels of both Felodipine and Chlorpromazine increased from <20% (without BSA) to levels of 65% and 72%, respectively, in the presence of 4% BSA. Secretory, B–A (bidirectional) transport Papp rates are not measured for compounds screened using basolateral BSA buffers, since protein binding can result in an underestimation of actual efflux Compound
Non-BSA Papp
Permeability
BSA Papp
Permeability
Digoxin (P-gp substrate)
0.5
Low
0.7
Low
Chlorpromazine (clog P: 5.2)
0.4
Low
11.4
High
Progesterone (clog P: 4.0)
5.2
Medium
25.9
High
Felodipine (clog P: 4.8)
1.0
Low
14.1
High
Papp (A–B, only absorptive transport route measured): Apparent Permeability Coefficient (× 10 cm/s), n = 3. Values of permeability are expressed as the mean −6
a severe underestimation of secretory transport in the basolateral-to-apical transport direction for compounds exhibiting high levels of protein binding. (Additional protocol modifications to increase compound solubility and recovery are discussed in Notes 14 and 15). 3.7. HPLC and Mass Spectrometry Analysis
1. Typically, compounds are detected by HPLC-MS/MS using multiple reaction monitoring (MRM) transitions, and automated determination of MRM transitions were generated with the manufacturer’s software. 2. As an alternative, narrow-window mass extraction HPLC and Mass Spectrometry (LC/MS) analysis can be performed for all samples if exact mass determination is possible (via the use of a quadrupole time-of-flight (QTof) mass spectrometer). 3. Linear calibration curves should be obtained in the range of approximately 1–500 nM. Sample concentrations below the lowest standard curve point are typically reported as BLOQ (below the limit of quantification). 4. The concentration of each sample should be determined using a calibration curve for the test compound and the integrated peak area – excluding points from the calibration curve may be acceptable if such point(s) fall outside the concentration range of the sample.
9 Optimization of the Caco-2 Permeability Assay to Screen Drug Compounds…
151
4. Notes 1. It is recommended to generate a new calibration curve (using similar reference control standards) each and every time a new lot/passage of Caco-2 cells is thawed. Caco-2 passage numbers should be monitored closely, and it is advisable to keep cultures actively growing for no more than 3 months, as cell population drift could affect comparisons to historical data. 2. From work in our laboratory, utilizing an initial seeding density at the lower range noted in the literature (~0.5–1 × 105 cells/cm2) and maintaining 20% fetal bovine serum (FBS) throughout the 21-day culture period have demonstrated more consistent TEER readings and high levels of P-gp expression during the course of the assay. Although accelerated (3–5 day) Caco-2 screening kits have shown design improvements over the years, the use of the 21-day system is still preferred for routine, in-house screening and rankorder purposes. 3. Screen compounds no higher than 10 mM, with optimal concentrations in the 1–10 mM range to minimize the possibility of efflux pump saturation. This issue is particularly important for systems such as Caco-2, in which multiple (endogenous) transporters are expressed, often at levels considered low relative to stable cell lines. In addition, DMSO concentrations should not exceed 1%. When DMSO levels approach 1.5–2.0%, TEER values (monolayer integrity) become compromised. 4. For in-house, rank-ordering classification, the following parameters are recommended: Papp values < 2 × 10−6 cm/s (or, <1 × 10−6) predict low, transcellular absorption in vivo (0–20% fraction absorbed); Papp values ~ 2–10 (or, 1–10) predict moderate, transcellular absorption in vivo (20–70% fraction absorbed), and Papp values > 10 typically predict high, transcellular absorption in vivo (70–100% fraction absorbed). The definition of high permeability is based in part on published data for in vivo absorption of known reference compounds and is also based on the Biopharmaceutical Classification System (BCS), which states that compounds with greater than 70% absorption are well absorbed and considered highly permeable. 5. To help determine if a test compound is exhibiting passive transport (the major mechanism for drug uptake across the intestine), Papp values should be independent of transport direction, independent over a wide range of concentrations (i.e., 1–100 mM), and independent of active transport inhibitors (i.e., low temperature, verapamil, MK-571, etc.). For active transport, Papp values should be dependent on the transport direction, dependent on concentration (i.e., saturable),
152
B. Press
dependent on the effects of efflux inhibitors, and significantly decreased at low temperature (i.e. 18–20°C). 6. For interlab data variability concerns, an emphasis should be placed on finding similar trends and rank orders for known reference compounds, even if the absolute data is shown to be different. Although variability in Caco-2 data from different labs has long been observed, it is most important to deal with data discrepancies that often come up within the same lab. 7. Cassette dosing should be avoided for this assay. While efflux transporter saturation can be minimized when cassette dosing multiple compounds at low concentrations (i.e., 1 mM each) the potential of a compound to inhibit an efflux pump at low concentrations justifies the need to avoid cassette dosing during the assay (although it should be noted that cassette LC/MS analysis is often acceptable). 8. For lead compounds, quantification of Caco-2 data should be performed against a standard curve covering approximately one order of magnitude above and below the dosing concentration of the compound. Generating Papp coefficients based on relative LC/MS peak areas can be done for rank ordering, noting that such data may be less reliable if the test compound in question displays a nonlinear profile (or exhibits very high or very low permeability). 9. When possible, test lead compounds for the absolute, kinetic transport by taking measurements at several time points, instead of a single time point measurement. 10. For compounds showing acceptable (or high) permeability even with low postassay recovery levels, Papp coefficients can be reported as an underestimation of actual permeability rates, due to the low recovery. 11. In an effort to deal with the problem of low postassay sample recovery, the use of hydrophilic, covalent-coated plastics for the receiver wells (e.g., BD Gentest enhanced recovery plates) can reduce the nonspecific adsorption of lipophilic compounds, thereby reducing the frequency of false negative data. 12. For assays performed in the presence of BSA, if one accounts for the unbound fraction ( fu) of the drug in the Papp calculations (in the form of using unbound concentration, Cu, rather than the dosing concentration, Co), and expressed as Cu = fuCo, it has been shown that the adjusted transport rates may not be significantly different than those in the absence of BSA. As a result, efflux could still be accurately predicted (even for high protein binding compounds) using this method (15, 16). 13. For data generated from assays using protein-sink conditions, confirm that BSA does not affect the absorptive permeability rate for compounds undergoing efflux (17). In addition, it is
9 Optimization of the Caco-2 Permeability Assay to Screen Drug Compounds…
153
recommended to generate a new calibration curve, including known P-gp and MRP substrates (see Table 2). 14. Solubility can greatly influence the ability of a drug to be absorbed across the intestinal wall. Most screening programs utilize aqueous media devoid of any solubility-enhancing surfactants (18). For some lipophilic drugs, inclusion of bile acids (such as taurocholic acid or glycocholic acid) at concentrations of 10 mM may have a beneficial effect in helping to solubilize these compounds. The drug excipient sodium lauryl sulfate (SLS) has been shown to clearly affect the integrity of Caco-2 monolayers by causing leakiness of the cell–cell junctions. Additional studies have revealed that the polyethoxylated excipients Tween-80, PEG-300, and the nonionic Cremophor EL (routinely used for in vivo formulations) can inhibit active P-gp efflux mechanisms in vitro. Additional work using bile salts in complex formulations to mimic fasted state simulated intestinal fluid (SIF) has shown promise to improve the solubility of highly lipophilic drugs without causing notable toxicity to the Caco-2 monolayers (19). 15. Caco-2 polarized monolayers are oversensitive to excipient effects on drug permeation, and attention should be placed on the integrity of the monolayer when using such modified buffers. Surfactants can incorporate into the lipid bilayers, thereby changing the properties and fluidity of the membrane. Such an effect may alter the functioning of transmembrane proteins such as P-gp, MRP-2, and BCRP. As a result, it is recommended to include appropriate efflux pump substrate controls when employing excipients in this assay.
Acknowledgments The author would like to thank Deanna Di Grandi for her contributions and engaging discussions in optimizing and troubleshooting various aspects of the assay techniques described herein. References 1. Artursson P, Palm K, and Luthman K (2001) Caco-2 monolayers in experimental and theoretical predictions of drug transport. Adv Drug Deliv Rev 46:27–43 2. Li AP (2001) In vitro approaches to evaluate ADMET drug properties. Curr Top Med Chem 4(7):701–706 3. Smith DA, van de Waterbeemd H (1999) Pharmacokinetics and metabolism in early drug discovery. Curr Opin Chem Biol 3(4):373–378
4. van de Waterbeemd H, Smith DA, Beaumont K, Walker DK (2001) Property-based design: optimization of drug absorption and pharmacokinetics. J Med Chem 44(9): 1313–1333 5. Matsson P, Bergstrom C, Nagahara N, Tavelin S, Norinder U, Artursson P (2005) Exploring the role of different drug transport routes in permeability screening. J Med Chem 48 (2): 604–613
154
B. Press
6. Balimane PV, Chong S (2005) Cell culturebased models for intestinal permeability: a critique. Drug Discov Today 10(5):335–343 7. Yee S (1997) In vitro permeability across Caco-2 cells (colonic) can predict in vivo (small intestinal) absorption in man – fact or myth. Pharm Res 14(6):763–766 8. Chan S, Lowes S, Hirst BH (2004) The ABCs of drug transport in intestine and liver: efflux proteins limiting drug absorption and bioavailability. Eur J Pharm Sci 21:25–51 9. Hoffmann U, Kroemer HK (2004) The ABC transporters MDR1 and MRP2: multiple functions in disposition of xenobiotics and drug resistance. Drug Metab Rev 36(3–4): 669–701 10. Palm K, Luthman K, Ros J, Grasjo J, Artursson P (1999) Effect of Molecular Charge on Intestinal Epithelial Drug Transport: pH-Dependent Transport of Cationic Drugs. J Pharmacol Exp Ther 291(2):435–443 11. König J, Nies AT, Cui Y, Leier I, and Keppler D (1999) Conjugate export pumps of the multidrug resistance protein (MRP) family: localization, substrate specificity, and MRP2mediated drug resistance. Biochim Biophys Acta 1461(2):377–394 12. Makhey VD, Guo A, Norris DA, Hu P, Yan J, Sinko PJ (1998) Characterization of the regional intestinal kinetics of drug efflux in rat and human intestine and in Caco-2 cells. Pharm. Res 15(8):1160–1167
13. Gao J, Murase O, Schowen RL, Aubé J, Borchardt RT (2001) A functional assay for quantitation of the apparent affinities of ligands of P-glycoprotein in Caco-2 cells. Pharm Res 18(2):171–176 14. Balimane PV, Han YH, Chong S (2006) Current industrial practices of assessing permeability and P-glycoprotein interaction. AAPS J 8(1):E1–E13 15. Fisher JM, Wrighton SA, Calamia JC, Shen DD, Kunze KL, and Thummel KE (1999) Midazolam metabolism by modified Caco-2 monolayers: effects of extracellular protein binding. J Pharmacol Exp Ther 289(2):1143–1150 16. Neuhoff S, Artursson P, Zamora I, Ungell AL (2006) Impact of extracellular protein binding on passive and active drug transport across Caco-2 cells. Pharm Res 23(2):350–359 17. Saha P, Kou JH (2002) Effect of bovine serum albumin on drug permeability estimation across Caco-2 monolayers. Eur J Pharm Biopharm 54(3):319–324 18. Lakeram M, Lockley DJ, Pendlington R, Forbes B (2008) Optimization of the caco-2 permeability assay using experimental design methodology. Pharm Res 25(7):1544–1551 19. Saha P, Kou JH (2000) Effect of solubilizing excipients on permeation of poorly watersoluble compounds across Caco-2 cell monolayers. Eur J Pharm Biopharm 50(3): 403–411
Chapter 10 Ouabain Modulates Cell Contacts as well as Functions that Depend on Cell Adhesion Isabel Larre, Ruben G. Contreras, and Marcelino Cereijido Abstract Ouabain, a toxic of vegetal origin used for centuries to treat heart failure, has recently been demonstrated to have an endogenous counterpart, most probably ouabain itself, which behaves as a hormone. Therefore, the challenge now is to discover the physiological role of hormone ouabain. We have recently shown that it modulates cell contacts such as gap junctions, which communicate neighboring cells, as well as tight junctions (TJs), which are one of the two differentiated features of epithelial cells, the other being apical/ basolateral polarity. The importance of cell contacts can be hardly overestimated, since the most complex object in the universe, the brain, assembles itself depending on what cells contacts what other(s) how, when, and how is the molecular composition and special arrangement of the contacts involved. In the present chapter, we detail the protocols used to demonstrate the effect of ouabain on the molecular structure and functional properties of one of those cell–cell contacts: the TJ. Key words: Ouabain, Cell contacts, Cell adhesion, Tight junctions, Communicating junctions, Differentiation, Epithelia, Claudin, Connexin
1. Introduction The affinity and specificity of Na+, K+-ATPase for inhibitor ouabain are so high, that led to suspect that there might exist endogenous analogues. In keeping with this possibility, Hamlyn et al. (1) demonstrated the presence in plasma of a substance that, so far, cannot be distinguished from ouabain even by specific antibodies and mass spectrometry. This endogenous ouabain was found to increase during exercise (2) and pathological conditions such as arterial hypertension (3–6) and eclampsia (7), raising the possibility that ouabain would constitute a new hormone (in the sense that its status was only recently recognized), and prompting
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_10, © Springer Science+Business Media, LLC 2011
155
156
I. Larre et al.
efforts to discover its physiological role. We have recently found that ouabain specifically modulates cell–cell and cell–substrate contacts, as well as cell processes that depend on cell adhesion: 1. Effect of ouabain on gap junctions. Wild MDCK cells (epithelial from dog kidney) (W-MDCK) exposed to toxic concentrations of ouabain (above 1.0 mM) detach from each other and from the substrate, and if kept in suspension detached cells die in a few hours. However, there are MDCK that can proliferate and differentiate in the presence of high concentrations of ouabain (R-MDCK) (8–12). Monolayers of MDCK cells only communicate through gap junctions for a brief period between 8 and 15 h following plating at confluence, as evidenced by the transference of Lucifer Yellow injected into a given cell using an impaling glass microelectrode, and by the typical image of these junctions in freezefracture replicas (9). A given cell is connected – in the average – to three neighbors. The same phenomenon is observed with confluent monolayers of mixed W and R cells. Yet upon treatment with ouabain, cells connect to each other and pass an intracellular marker (in this case neurobiotin) freely to an average of five cells although sometimes to 30 or more nearest and distant neighbors (Fig. 1a, b). This increase of cell– cell connection caused by ouabain is due to an enhancement
Fig. 1. Ouabain modulates cell communication. Cocultures of wild and ouabain-resistant MDCK cells in confluent monolayers. (a) Confocal micrograph of a cell injected with Dextran (asterisk) (the dye is green, yet the overlapping of both colors makes the mixture to appear yellow.), plus the intracellular marker neurobiotin. This marker is not visible by itself; therefore, once fixed with 4% paraformaldehyde, the monolayer is treated with avidin coupled to TRITC fluorophore (red ). Avidin has a very high and specific affinity for neurobiotin and reveals that this marker has diffused to three nearest neighbors cells (located at 10, 5, and 7th hour). (b) A similar monolayer treated with ouabain overnight, where a cell injected with neurobiotin (asterisk) transfers the dye to all nearest neighbors, plus several layers of secondarily connected cells. (c) Western blot analysis of connexins 43, 32, and 26 in lanes 1–2, 3–4, and 5–6 respectively. Staining with an antiactin antibody is used to normalize the amount of protein loaded in each lane. Ouabain increases the density of connexin 32, but not those of connexins 43 and 26 (Taken with the permission of PNAS).
10 Ouabain Modulates Cell Contacts as well as Functions that Depend on Cell Adhesion
157
of the synthesis of connexin (cnx) 32 (but not cnx-26 nor cnx-43, Fig. 1c). Thanks to this interconnection, W-MDCK cells are rescued by R ones from the toxic effects of ouabain and do not detach (metabolic cooperation). A measurement of the K, Na, and Cl content of hundreds of individual cells using an electron probe (9) indicates that the mixed populations of W and R cells constitutes now a single ion compartment, reflecting the fact that ouabain has enhanced cell–cell communication (12). 2. Effect of ouabain on tight junction (TJ). Another very special form of cell–cell contact is the TJ that by sealing the outermost end of the space between neighboring epithelial cells transforms epithelia into effective permeability barriers. The degree of sealing of an epithelium can be estimated through the transepithelial electrical resistance (TER) (Fig. 2a, open circles). The value of TER of a given epithelial is proportional to the difference in composition between the two compartments it separates. Thus, the epithelium of the proximal tube of the kidney separates interstitial fluid from just-filtrated plasma; to withstand this week gradient, the proximal tube has a TER of merely 5–10 W cm2. The colon mucosa separates, instead, the lumen that contains a flora of single-celled microorganisms (13) immersed in the rests of processed food and digestive juices, and the interstitial fluid that bathes the basolateral side of the epithelial cells. This is a really sharp gradient, and it comes as no surprise that TER of the colon mucosa is well above 5,000 W cm2. While this correspondence between the difference of composition between the two side of an epithelium and the value of TER is teleologically sound, we are just beginning to learn the factors and mechanism that modulate TER. In an effort study of this correlation between TER and the sharpness of the gradient across epithelia, we have made two
Fig. 2. Ouabain modulates the molecular composition and degree of tightness of the TJ. (a) TER, an electrical parameter reflecting the degree of sealing of the TJ, increases in the presence of 10 and 50 nM ouabain. (b) Unidirectional flux of 3 kDa Dextran (JDEX) through the monolayer under control condition or treated with 10 nM ouabain. (c) Western blot of claudin 2 and actin performed at the second day of exposure to 10–100 nM ouabain.
158
I. Larre et al.
Fig. 3. 10 nM ouabain increases claudin 2 mRNA. (a) Total RNA extracted from monolayers of MDCK cells under control and 10 nM ouabain treatment. (b) mRNA of cln-1 and -2 mRNA. GAPDH is used to normalize results. (c) MDCK cells transfected with firefly luciferase under cln-1 promoter, and pRL-CMV driving Renilla luciferase (to normalize results) under control and 10 nM ouabain stimulation.
ifferent headways. One was the observation that the lumen of d the nephron is bathed by a fluid rich in epithelial growth factor (EGF) that modulates TER in MDCK monolayers in a concentration-dependent manner (14–16); the second is the demonstration that also ouabain is able to modulate this TER in a concentration-dependent manner. Both EGF and ouabain trigger complex signaling routes that result in the incorporation o new molecular components to the TJ. The initial observation indicating that ouabain acts on the TJ was that at 1.0 mM, ouabain binds to Na+, K+-ATPase and triggers a P → A mechanism (from pump/adhesion) that causes retrieval of molecules belonging to the tight, adherent, and focal junctions (17–19). The effect of ouabain on the TJ is not a linear one. Thus, at low, nontoxic concentrations (i.e., that neither inhibit the Na+, K+-ATPase nor distort the K+ balance of the cell), ouabain increases TER in a concentration-dependent manner and decreases the unidirectional flux of 3-kDa Dextran in a concentration- independent manner (Fig. 2). In the rest of the present chapter, we describe the protocol to study the effect of ouabain at hormonal concentrations (10–50 nM) on the function and molecular structure of the TJ (Fig. 3).
2. Materials 2.1. Cell Culture
1. Starter MDCK-II cultures are obtained from the American Type Culture Collection (MDCK, CCL-34, subculture 60–65). These cells have low basal transepithelial electrical resistance (~300 W cm2). 2. Plastic bottles (Costar 3151). 3. CO2 incubator (Forma Scientific CO2, Steri-Cult 200).
10 Ouabain Modulates Cell Contacts as well as Functions that Depend on Cell Adhesion
159
4. Biological safety cabinet (Forma Scientific, 1284). 5. Phosphate-buffered saline (PBS, cat. num. 21300-058, GIBCO-Invitrogen). Prepare according to manufacturer’s instructions. 6. Ca2+-free phosphate-buffered saline (Ca-free PBS; GibcoBRL 21300-058). Prepare according to manufacturer’s instructions but omit adding CaCl2 powder. 7. Trypsin solution: 0.5% Trypsin-0.5 mM EDTA in Ca- and Mg-free Hank’s balanced salt solution (HBSS) (In Vitro). 8. Trypsin inhibitor solution: 5 mg/ml trypsin inhibitor in PBS (In Vitro). 9. Dulbecco Eagle’s Medium (DMEM, cat. num. 12100-061, GIBCO-Invitrogen, Carlsbad, CA). Prepare according to manufacturer,s instructions (To prepare 10 L of DMEM dissolve the content of one bottle in 9.6 L of deionized water, add 37 g of NaHCO3 and 59.57 g of HEPES, adjust pH to 7.4 with 10 N NaOH, and bring the volume to 10 L). 10. CDMEM: DMEM supplemented with 10% donor bovine serum (cat. num. 10371-029, GIBCO-Invitrogen) and penicillin–streptomycin 10,000 U-mg/ml (In Vitro, Acayucan, Mexico, DF). 11. Transwells (inserts) and multiplates are purchased from Corning (Transwell Corning Costar, Cambridge, MA (cat. nums. 3472 or 3413, 3506 6 well and 3338 48 well)). 12. Ouabain stock solution: Ouabain (O-3125, Sigma) is dissolved with vortex agitation in PBS, to obtain a 10 mM solution. Store in aliquots at −20°C. 2.2. Measurement of the Transepithelial Electrical Resistance
1. EVOM system (World Precision Instruments, Sarasota, FL). 2. EndOhm-6 systems (World Precision Instruments, Sarasota, FL). 3. Tweezers.
2.3. SDSPolyacrylamide Gel Electrophoresis
1. Buffer protein extraction (RIPA): Tris 10 mM pH 7.5, 1 mM sodium orthovanadate, 1% SDS, supplemented with Complete® protease inhibitor cocktail (Roche Applied Science, Germany). 2. Stacking buffer: Tris–HCL 1.0 M pH 6.8. Store at room temperature. 3. Separating buffer: Tris–HCL 1.5 M pH 8.8. Store at room temperature. 4. 10% SDS. Store at room temperature. 5. Acrylamide/bisacrylamide solution 30% (Bio-Rad, cat. num. 161-0156, this is a neurotoxin when unpolymerized, and therefore, care should be taken not to get exposed to it).
160
I. Larre et al.
6. Ammonium persulfate (APS): prepare 10% solution in water and immediately freeze in aliquots at −20°C. 7. N,N,N,N ¢-Tetramethyl-ethylenediamine (TEMED, Bio-Rad, Hercules, CA). 8. Running buffer 10×: 0.25 M Tris, 1.92 M glycine, 1% SDS. 9. Precision plus protein standards (Bio-Rad, cat. Num. 161-0374). 2.4. Western Blotting
1. Transfer buffer 1×: 48 mM Tris, 39 mM glycine, 0.037%, 20% methanol. Store at 4°C. 2. Laemmli sample buffer 1× (cat. Num. 161-0737 Bio-Rad, CA). 3. Total protein content was measured by BCA™ protein assay (Thermo scientific Rockford, USA). 4. Polyvinylidene difluoride (PVDF) membrane (Amersham Biosciences). 5. Tris-buffered saline with Tween (TBS-T): Prepare 10× stock with 1.37 M NaCl, 27 mM KCl, 250 mM Tris–HCl, pH 7.4, and 2% Tween-20. Dilute 100 ml in 900 ml water just before use. 6. Blocking buffer: 5% (w/v) nonfat dry milk in TBS-T. 7. Primary and secondary antibody dilution buffer: TBS-T supplemented with 5% (w/v) fraction V bovine serum albumin (BSA). 8. Rabbit polyclonal anti-Cldn-8, Horseradish peroxidase (HRP)-conjugated anti-rabbit and mouse IgG, FITCconjugated anti-rabbit antibodies were purchased from Zymed-Invitrogen Laboratories Inc (San Francisco, CA). 9. Enhanced chemiluminescent (ECL) reagents from Amersham (RPN2132).
2.5. Paracellular Flux of FITC-Dextran
1. P buffer: 10 mM HEPES, pH 7.4, 1 mM sodium pyruvate, 10 mM glucose, 3 mM CaCl2, and 145 mM NaCl. 2. FITC Dextran solution: add 10 mg of FITC-dextran per 1 ml of buffer P (3 kDa, Pharmacia Fine Chemicals) (see Note 1). 3. Spectrofluorometer (Perkin Elmer). 4. Transwell inserts 0.33 cm2 (Corning Costar, Cambridge, MA, cat. Num.3472).
2.6. Claudin-1 Promoter Assay Using Calcium Phosphate Precipitation
1. HEPES buffer 2×: 3 M NaCl, 0.5 M HEPES pH 7.1, 1 M NaPO4 pH 7.1. Filter this solution and make it fresh each time. 2. Plasmids coding for firefly luciferase are driven by claudin-1 (cln-1) promoter (−748 to +252), and those coding for Renilla luciferase are driven by promoter pRL-CMV.
10 Ouabain Modulates Cell Contacts as well as Functions that Depend on Cell Adhesion
2.7. Isolation of RNA and RT-PCR
161
1. DNAase-RQ1 (Promega). 2. Distilled water, DNase, and RNase free (cat. num. 10977015,GIBCO). Primer for Claudin-2 forward: 5¢-ACGACAAGCAAACAGGCTCCGAAG-3¢ and reverse: 5¢-GCGGGATCCCACATACCCAGTCAGG-3¢ (20) GAPDH: Forward: 5¢-TGACCTGCCGCCTGGAGAAA-3¢ and Reverse: 5¢-GGCGTCGAAGGTGG-AAGAGTG-3¢ (21) purchased from Invitrogen. 3. Solution D mix: 4 M guanidinium thiocyanate 25 mM sodium citrate 0.5% sarcosyl and 0.1 M 2-mercaptoethanol. The stock without 2-mercaptoethanol is stable for 3 months at room temperature, but when you add 2-mercaptoethanol it is stable only for 1 month (22). 4. Buffer TE: Tris–HCL 10 mM pH 8.0, EDTA 1 mM. 5. RNasin plus RNase inhibitor (Promega), AMV-reverse transcriptase (Promega), and Taq DNA polymerase (Roche) (see Note 8). 6. Thermocycler Veriti (AB applied Biosystems).
3. Methods 3.1. Cell Culture
1. In a biological safety cabinet, wash cells growing in plastic bottles three times with Ca-free PBS prewarmed at 37°C (5 ml for a 75 cm2 bottle). PBS must be poured directly to the walls of the bottle and not to the cells (see Note 2). 2. Remove Ca-free PBS. 3. Add 1–2 ml per 75 cm2 bottle of trypsin-EDTA solution and distribute over the complete surface of growing cells. Remove the excess to avoid cell damage (see Note 3). 4. Introduce the closed bottle into the CO2 incubator for 2 min; observe if there is any detachment. A light stroke helps to resuspend cells and detect cell detachment. If there are still cells attached to the plastic incubate for a longer time. 5. Once most cells are detached, add 10 ml of CDMEM prewarmed at 37°C with a glass pipette, resuspend five times to obtain a homogeneous suspension, and with the help of a sterile micropipette, plate 100 ml of the suspension onto the apical compartment of a 0.33-cm2 Transwell insert placed in a 24-well multiplate with 800 ml of CDEM in each well. This will give you a density about 2.25 × 105 cell/cm2 or 30% of confluency. Incubate for three days in CDMEM. 6. Then, rinse cells on Transwells twice with DMEM (without serum) and incubate them for 24 h in DMEM supplemented with 1% bovine serum.
162
I. Larre et al.
7. To incubate cells with ouabain, dilute the necessary volume of ouabain stock solution in sterile DMEM with 1% bovine serum, aspirate the media from the dish, and add ouabain solutions both to the apical and basolateral compartments. Incubate for required time. 3.2. Measurement of the Transepithelial Electrical Resistance
1. Assemble the EndOhm and EVOM systems as indicated by the manufacturers (see Note 4). 2. Fill the EndOhm chamber with 800 ml of medium or buffer (DCMEM, Ca-free MEM, PBS, etc., 37°C, see Note 5). 3. Make sure that the Transwell contains at least 100 ml of the media in the apical compartment. Take the Transwell (0.33 cm2) with tweezers and position it carefully in the EndOhm chamber. 4. Place the cover of the EndOhm system over the chamber and push the measurement button for several seconds (from 5 to 30, until displayed value stabilizes). The displayed value corresponds to the transepithelial electrical resistance in W cm2.
3.3. Polyacrylamide Gel Electrophoresis
1. After 24 h of serum deprivation, TER is measured, and this is considered as TER at time zero. Then, monolayers are treated with ouabain and TER is measured as a function of time up to 3 days. Once the TER is registered, keep Transwells with cell in the 24-well plates and remove all medium by aspiration or with the help of a micropipette. 2. Rinse the Transwells with cells twice with PBS. Remove excess PBS. 3. Add 30 ml of RIPA, supplemented with Complete® protease inhibitor cocktail, in the apical compartment and incubate at −20°C for 30 min. 4. Recover the RIPA from the apical compartment. This is a critical step. First scrape gently the bottom filter, either with a yellow tip attached to the micropipette or with the yellow tip held with the hand. Deposit the recovered volume in a 400-ml Ependorff tube. 5. Centrifuge extracts for 10 min at 4°C and 17,000 × g. Take 4 ml to measure protein content by BCA™ protein assay (Thermo scientific Rockford, USA) and mix the remaining volume (about 26 ml) with and equal volume of Laemmli sample buffer. Boil samples for 10 min. 6. Prepare 10 ml of acrylamide separating gel. For a single 15% acrylamide/bisacrylamide, 1.5-mm thick separating gel, mix 2.3 ml of water with 5 ml of acrylamide/bisacrylamide solution, 2.5 ml of Tris–HCl pH 8.8, 100 ml of SDS 10%, 100 ml of APS 10%, and 4 ml of TEMED. Pour the gel solution in the gel cast system and overlay with distilled water. The gel polymerizes in about 30 min.
10 Ouabain Modulates Cell Contacts as well as Functions that Depend on Cell Adhesion
163
7. Pour off the water. 8. Prepare 4 ml of stacking gel solution (for a single gel) by mixing 2.7 ml of water, 670 ml acrylamide/bis solution with 500 ml of Tris pH 6.8, 40 ml of 10% SDS, 40 ml of 10% APS, and 4 ml of TEMED. Add the solution to the top of the separating gel and insert the comb. The stacking gel should polymerize within 30 min. 9. Prepare the running buffer 1× by diluting 100 ml of 10× running buffer in 900 ml of water. Once the stacking gel has set up, carefully remove the comb and use a 3-ml syringe fitted with a 22-gauge needle to wash the wells with running buffer. 10. Add the running buffer to the upper and lower chambers of the gel unit and load the 50 ml of each sample in a well. Include one well for prestained molecular weight markers. 11. Complete the assembly of the gel unit and connect to a power supply. Run the gel at 100 V for approximately 1 h. As soon as the molecular weight markers close to the protein of interest spread out and can be easily observed, stop running. 3.4. Western Blot Analysis
1. Perform transference to polyvinylidene difluoride membrane (Amersham Biosciences) according to the manufacturer’s description of semidry chamber (Bio-Rad) at 420 mA for 40 min in transfer buffer. 2. The PVDF membrane is then incubated in 20 ml with 5% milk for 1 h at room temperature on a rocking platform. 3. The milk is discarded and the membrane is quickly rinsed prior to addition of a 1:1,000 dilution of claudin-8 antibody in TBST/ 5% BSA for 1 h at room temperature on a rocking platform. 4. The primary antibody is then removed and the membrane washed four times for 10 min each with 50 ml TBS-2% Tween. 5. The secondary antibody is freshly prepared for each experiment as 1:5,000-dilution in TBS-5% BSA and added to the membrane for 1 h at room temperature on a rocking platform. 6. The secondary antibody is discarded and the membrane washed four times for 10 min each with TBS. 7. During the final wash, 2 ml aliquots of each portion of the ECL reagent are warmed separately to room temperature and the remaining steps are done in a dark room under safe light condition. Once the final wash is removed from the blot, the ECL reagents are mixed together and then immediately added to the blot, which is then rotated by hand for 1 min to ensure even coverage.
164
I. Larre et al.
8. The membrane is placed between the leaves of an acetate sheet protector that has been cut to the size of an X-ray film cassette. Place the assembly on an X-ray film exposure cassette. 9. Lay a hyper-film sheet (Amersham) over the protected membrane, close the cassette and expose several times, from seconds to minutes. 10. Develop the film. Once a satisfactory exposure for the result has been obtained, the membrane is then stripped of the signal and then reblotted with an antibody that recognizes actin. An example of the results obtained is shown in Fig. 2. 11. Bands are visualized on Kodak X-OMAT LO fixer and replenisher (Eastman Kodak Co.). All Western blot experiments are performed at least three times. Representative data are presented in Fig. 2c. 3.5. Paracellular Flux of FITC-Dextran
1. Culture cells in Transwell according Subheading 3.1. 2. TER is measured in the control and ouabain treated monolayer. If TER changes according to your expectancy, then proceed to the next step. 3. Prepare P buffer and FITC-dextran solution. 4. Wash twice with P buffer. 5. Add 10 mg/ml of FITC-dextran and ouabain solution in the apical compartment and incubate 37°C for 1 h. 6. Then, the basal medium is collected and the fluorescence of the transported FITC-dextran (3 kDa) is measured with a fluorescence spectrometer LS-3B (Perkin-Elmer) exiting FITC at 492 nm and measuring the emission fluorescence at 520 nm. The quantity of FITC is calculated by comparing with a standard curve. 7. Calculate the unidirectional (apical → basolateral) flux as follows: divide the fluorescence intensity of a given sample of the bottom solution (arbitrary units) by the corresponding value of the upper solution diluted as indicated. The figure obtained corresponds to the flux (JDEX) in ng/cm2/h.
3.6. Claudin-1 Promoter Transfection by Calcium Phosphate Precipitation
1. In order to be efficiently transfected, a cell should spread and reach its maximum area. In order to achieve this area, plate at most 60,000 cells per well in a 48-well plate. 2. Next day, prepare fresh 2× HEPES buffer. Sterilize by filtering. 3. Prepare fresh 0.25 M CaCl2 from 1.0 M stock and sterilize by filtering. 4. To test precipitate formation, add 0.25 ml CaCl2 dropwise to 0.25 ml 2× HEPES while stirring in a vortex. Cloudiness should appear in a few minutes. If so, go on with the protocol.
10 Ouabain Modulates Cell Contacts as well as Functions that Depend on Cell Adhesion
165
5. Add 20–50 mg DNA to 2× HEPES buffer. Firefly luciferase and Renilla luciferase plasmids were driven by promoters claudin-1 and pRL-CMV, respectively. The second one should not react in the presence of ouabain and can be used to normalize values. 6. Add 0.25 M CaCl2 dropwise while shaking vigorously after each drop (keep tube on vortex all the time). 7. Allow 30 min for the precipitate to form at room temperature. 8. Vortex and add to cell culture plates (split the day before) without culturing medium for 30 min at 37°C and shake every 10 min. 9. Add 1.0 ml of medium with 1% serum. Stir and incubate overnight at 37°C. 10. Next day remove media. Add 12.5% sterile glycerol containing 10 mM Hepes, pH 7.4 (“glycerol shock”). Incubate for 1 h at room temperature. 11. Remove glycerol and rinse twice with PBS. Add 1.0 ml of medium with 1% serum and incubate overnight at 37°C. 12. Add 10 nM ouabain and wait for 10–12 h. 13. Remove the medium and start using Dual-Luciferase kit reporter assay system (Promega) according to manufacturer directions. Read the cell suspension of each well in a Fluorskan at the wavelength recommended by the Dual-Luciferase kit. You will obtain two different values: one for firefly luciferase, which measures the activity of a given promoter, and the other for Renilla luciferase, which controls the transfection process. 14. Divide the value for firefly luciferase by the one for Renilla luciferase (transfection control). 3.7. mRNA Isolation
1. Plate cells in a six-well plate. 2. Scrape Monolayers in 750 ml of solution D and transfer suspension into a 1.5-ml Eppendorf tube. 3. Add sequentially the following and shake: 75 ml of 2 M sodium acetate pH 4.0, 705 ml of buffer saturated Phenol, and 50 ml of chloroform. 4. Vortex the solution and chill on ice for 15 min. Then, spin the solution at 10,000 × g for 20 min at 4°C. 5. Take 700 ml of the aqueous phase and precipitate with 700 ml of isopropanol (1:1 proportion) at 20°C for at least 1 h. 6. Spin at 22,000 × g for 20 min at 4°C. 7. Resuspend the pellet in 300 ml of solution D and precipitate with 1 ml of isopropanol at −20°C for at least 1 h.
166
I. Larre et al.
8. Spin at 12,000 × g for 20 min 4°C. 9. Wash the RNA pellet with 1 ml of 70% ethanol. Spin for 5 min and discard ethanol. Leave it open at room temperature and let it dry. Dry the pellet and dissolve in pure water (50 ml). 10. Treat the resuspended pellet with DNAse (2.5 mg/ml) at 37°C, 1 h. 11. Run a sample in an agarose gel to check that samples are not degraded (see Note 6). 3.8. RT-PCR
1. Prepare mix 1: In an Eppendorf tube mix the following: 500 ng to 1 mg of RNA with 1 ml of 50 mM antisense primer and adjust to 13 ml with water. 2. Leave samples in the thermocycler at 65°C for 5 min, then at 95°C for 1 min. Leave it on ice at 4°C. 3. Prepare mix 2: In an Eppendorf tube, mix the following: 6 ml of buffer 5×, 6 ml of 10 mM dNTP, RNasin 30 m/ml, 3 ml DMSO y 1 ml reverse transcriptase (AMV-RTase) and bring to 50 ml with water. 4. Add carefully mix 2 into mix 1. Incubate for 1 h AMV-Rtase at 40ºC (see Note 7). This reaction yields cDNA.
3.9. PCR
1. To prepare the reaction, mix: 5 mL of buffer 10×, 8 mL of dNTP-mix (1.25 mM each one), 1 ml of sense primer (50 mM), 1 ml of antisense primer (50 mM), 1–5 ml of cDNAs, 5 ml of DMSO and bring to 50 ml of water. 2. Leave samples in the thermocycler at 95°C for 3 min. Run 25 cycles at 95°C for 1 min, 62°C for 1 min, 72°C for 1 min, and 4°C. 3. Take a 1 mg sample of reacting solution, add it to the 2% agarose gel, run it, stain it, and take a picture.
3.10. Concluding Remarks
Cell contacts are crucial in processes as fundamental as cell cycling, proliferation, differentiation, cancer, metastasis, metabolic cooperation, wound healing, etc. The importance of finding factors such as ouabain that modulate cell contacts opens new avenues to understand these processes. We hope that with the protocol described above, which we routinely use to explore the role of ouabain in the biology of the TJ, you will be able to explore many fascinating areas of biology and pathology. All methods are straightforward and have no secrets. Nevertheless, we will be glad to furnish any additional information that you might need and will be ready to troubleshoot:
[email protected]
10 Ouabain Modulates Cell Contacts as well as Functions that Depend on Cell Adhesion
167
4. Notes 1. Use a fresh solution. Dextran tracers of different molecular weights are available with several different fluorophores. For a 24-well insert culture system, use 100 ml in the upper chamber and 600 ml in the bottom chamber. 2. Cells strongly attached to the substrate can be easily detached if the last wash is prolonged for 30–60 min in the CO2 incubator. 3. Use trypsin solution cautiously; keep it frozen, warm it just before use, and use minimal amounts. 4. TER may be measured in the safety cabinet, if you wish to follow TER changes along the time, or in the laboratory table. TER is very sensitive to temperature (see (23)), so measure quickly. 5. The basolateral media of control cells may be used for this purpose. pH must be continuously monitored by checking the red color of the media. To deliver the media, use a micropipette with sterile tips. 6. You can also use Trizol and follow the manufacturer’s recommendations. 7. Pipette tips and flasks should never be handled with naked hands: always use gloves. 8. Each enzyme comes with its buffers and oligo-dT. References 1. Hamlyn JM, Blaustein MP, Bova S, DuCharme DW, Harris DW, Mandel F, Mathews WR, and Ludens JH. Identification and characterization of a ouabain-like compound from human plasma. Proc Natl Acad Sci U S A 88: 6259–6263, 1991. 2. Bauer N, Muller-Ehmsen J, Kramer U, Hambarchian N, Zobel C, Schwinger RH, Neu H, Kirch U, Grunbaum EG, and Schoner W. Ouabain-like compound changes rapidly on physical exercise in humans and dogs: effects of beta-blockade and angiotensinconverting enzyme inhibition. Hypertension 45: 1024–1028, 2005. 3. Delva P, Degan M, Capra C, Fallo F, Mantero F, Menegazzi M, Lechi C, Steele A, and Lechi A. Plasma ouabain-like activity in essential hypertensive patients and in subjects with primary aldosteronism. Miner Electrolyte Metab 15: 315–320, 1989. 4. Hamlyn, J. M., Ashen, M. D., Forrest, B., Rogowski, A. C., and White, R. J. Species
5.
6.
7.
8.
s ensitivity of the sodium pump to a circulating ouabain-like inhibitor in acute hypervolemia and DOCA hypertension: comparison with ouabain. Prog Biochem Pharmacol 23, 22–34, 1988. Hamlyn JM, Ringel R, Schaeffer J, Levinson PD, Hamilton BP, Kowarski AA, and Blaustein MP. A circulating inhibitor of (Na+ + K+) ATPase associated with essential hypertension. Nature 300: 650–652, 1982. Moreth K, Kuske R, Renner D, and Schoner W. Blood pressure in essential hypertension correlates with the concentration of a circulating inhibitor of the sodium pump. Klin Wochenschr 64: 239–244, 1986. Lopatin DA, Ailamazian EK, Dmitrieva RI, Shpen VM, Fedorova OV, Doris PA, and Bagrov AY. Circulating bufodienolide and cardenolide sodium pump inhibitors in preeclampsia. J Hypertens 17: 1179–1187, 1999. Bolivar JJ, Lazaro A, Fernandez S, Stefani E, Pena-Cruz V, Lechene C, and Cereijido M.
168
I. Larre et al.
Rescue of a wild-type MDCK cell by a ouabain-resistant mutant. Am J Physiol 253: C151–161, 1987. 9. Cereijido M, Bolivar JJ, and Lazaro A. A ouabain resistant epithelial cell that protects the wild type in co-cultures. Pflugers Arch 405 Suppl 1: S147–151, 1985. 10. Contreras RG, Lazaro A, Bolivar JJ, FloresMaldonado C, Sanchez SH, GonzalezMariscal L, Garcia-Villegas MR, Valdes J, and Cereijido M. A novel type of cell-cell cooperation between epithelial cells. J Membr Biol 145: 305–310, 1995. 11. Contreras RG, Lazaro A, Mujica A, GonzalezMariscal L, Valdes J, Garcia-Villegas MR, and Cereijido M. Ouabain resistance of the epithelial cell line (Ma104) is not due to lack of affinity of its pumps for the drug. J Membr Biol 145: 295–300, 1995. 12. Larre I, Ponce A, Fiorentino R, Shoshani L, Contreras RG, and Cereijido M. Contacts and cooperation between cells depend on the hormone ouabain. Proc Natl Acad Sci U S A 103: 10911–10916, 2006. 13. Eckburg PB, Bik EM, Bernstein CN, Purdom E, Dethlefsen L, Sargent M, Gill SR, Nelson KE, and Relman DA. Diversity of the human intestinal microbial flora. Science 308: 1635–1638, 2005. 14. Flores-Benitez D, Rincon-Heredia R, Razgado LF, Larre I, Cereijido M, and Contreras RG. Control of tight junctional sealing: roles of epidermal growth factor and prostaglandin E2. Am J Physiol Cell Physiol 297: C611–620, 2009. 15. Flores-Benitez D, Ruiz-Cabrera A, FloresMaldonado C, Shoshani L, Cereijido M, and Contreras RG. Control of tight junctional sealing: role of epidermal growth factor. Am J Physiol Renal Physiol 292: F828–836, 2007.
16. Gallardo JM, Hernandez JM, Contreras RG, Flores-Maldonado C, Gonzalez-Mariscal L, and Cereijido M. Tight junctions are sensitive to peptides eliminated in the urine. J Membr Biol 188: 33–42, 2002. 17. Contreras, R. G., Flores-Beni Tez, D., FloresMaldonado, C., Larre, I., Shoshani, L., and Cereijido, M. Na+, K+-ATPase and hormone ouabain:new roles for an old enzyme and an old inhibitor. Cell Mol Biol (Noisy-le-grand) 52, 31–40, 2006. 18. Contreras, R. G., Flores-Maldonado, C., Lazaro, A., Shoshani, L., Flores-Benitez, D., Larre, I., and Cereijido, M. Ouabain binding to Na+, K+-ATPase relaxes cell attachment and sends a specific signal (NACos) to the nucleus. J Membr Biol 198, 147–58, 2004. 19. Contreras, R. G., Shoshani, L., FloresMaldonado, C., Lazaro, A., and Cereijido, M. Relationship between Na(+),K(+)-ATPase and cell attachment. J Cell Sci 112 ( Pt 23), 4223–32, 1999. 20. Furuse, M., Fujita, K., Hiiragi, T., Fujimoto, K., and Tsukita, S. Claudin-1 and -2: novel integral membrane proteins localizing at tight junctions with no sequence similarity to occludin. J Cell Biol 141, 1539–50, 1998. 21. Medina-Contreras O, Soldevila G, PatinoLopez G, Canche-Pool E, Valle-Rios R, and Ortiz-Navarrete V. Role of CRTAM during mouse early T lymphocytes development. Dev Comp Immunol 34: 196–202, 2010. 22. Chomczynski P and Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156–159, 1987. 23. Gonzalez-Mariscal, L., Chavez de Ramirez, B., and Cereijido, M. Effect of temperature on the occluding junctions of monolayers of epithelioid cells (MDCK). J Membr Biol 79, 175–84, 1984.
Chapter 11 Monitoring of the Dynamics of Epithelial Dome Formation Using a Novel Culture Chamber for Long-Term Continuous Live-Cell Imaging Judith Lechner, Daniel Hekl, Helmut Gatt, Markus Voelp, and Thomas Seppi Abstract Epithelial tissue guarantees proper performance of many organs, e.g., the kidneys, the gastrointestinal organs, and endocrine glands. Epithelial layers are responsible for the formation and maintenance of separate compartments with distinct solute composition. This is achieved by epithelial layers forming a barrier between the two compartments and concomitantly allowing site-directed transepithelial transport, uptake or secretion of electrolytes, energy substrates, proteins, and other solutes. Research on epithelial tissue functions has highly profited from the establishment of tissue culture technologies allowing to cultivate primary epithelial cells or established epithelial cell lines. A property of transporting epithelia cultured in vitro that has long been noted is the formation of the so-called domes on solid growth supports, which represent fluid filled blisters between the solid growth surface and the cell layer. Formation of domes is regarded as a sign of active transport processes and an intact epithelial barrier function due to functional tight junctional cell–cell contacts. A novel methodology for long-term live-cell light microscopy is described in the present article, which allows the monitoring of the dynamic nature of structures, such as epithelial domes over days to weeks of tissue culture (“under the microscope”). Key words: Epithelium, Dome, Cell culture, Live-cell imaging, Renal proximal tubulus
1. Introduction Epithelial tissue guarantees proper performance of many organs in the mammalian body, e.g., organs of the gastrointestinal tract, the urogenital tract, or endocrine glands. Epithelial cell layers form barriers and concomitantly allow site-directed transepithelial transport, thus being responsible for the formation and maintenance of separate fluid compartments. Prerequisite for selective barrier function is a continuous belt of tight (TJ) Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_11, © Springer Science+Business Media, LLC 2011
169
170
J. Lechner et al.
and adherens junctional (AJ) strands linking together individual epithelial cells. TJ complexes function as “gates” in the paracellular transport route restricting and controlling the passage of solutes and small ions. The driving forces for paracellular transports are transepithelial electrochemical gradients generated by transcellular transports via apical or basolateral domain-restricted membrane transporters and energy-consuming pumps. The TJ complexes also prevent membrane lipids and proteins of the apical and basolateral plasma membrane domains to mix (“fence” function of the TJ). Thereby, they play a pivotal role in establishment and maintenance of epithelial cell surface polarization, a prerequisite for site-directed transport processes (1–4). Research on epithelial tissue functions has highly profited from the establishment of tissue culture technologies allowing culturing of primary epithelial cells or established epithelial cell lines. Epithelial cells in vitro establish confluent cell layers resembling epithelial barriers as they occur in vivo. Established cell lines of renal tubular cells or intestinal epithelial cells have been used for long time in order to study epithelial barrier properties and transepithelial transport representing renal or intestinal epithelial tissue (5, 6). A specific feature of transporting epithelia grown on tissue culture dishes in vitro is the formation of fluid-filled blister-like structures (7). These structures were called “domes.” They form on solid growth support material like standard tissue culture plastic dishes or glass surfaces after the cultured cells have reached optical confluency. Epithelial domes are thought to originate from ongoing apical to basolateral transepithelial transport processes that lead to the accumulation of fluid between the epithelial cell layer and the growth support. Besides active transcellular transport mechanisms, prerequisite for the formation of domes are intact tight junctional strands, since they prevent the backleak of fluid through the paracellular route. Therefore, the formation of “domes” on solid growth supports is regarded as a sign of a highly differentiated, transporting epithelia forming an intact epithelial barrier via functional tight junctional cell–cell contacts coupling individual cells together. In the present article, we describe a novel method allowing the observation of the dynamic nature of dome formation by a permanent cell culture technique. The patented device enables live-cell imaging directly on the stage of an inverted microscope during days or even weeks.
2. Materials 2.1. Cell Culture
1. Porcine renal proximal tubular LLC-PK1 cells obtained from the American Tissue Culture Collection, ATCC (see Note 1). 2. Cell culture medium: Dulbecco’s modified Eagle’s medium (DMEM;GIBCO, Austria) containing 5.5 mM glucose, 2 mM
11 Monitoring of the Dynamics of Epithelial Dome Formation Using a Novel Culture...
171
glutamine, 7% FCS, 100 U/ml penicillin, and 100 mg/ml streptomycin (all Sigma, Austria). 3. Cell incubator to allow standard cell culture in a humidified 5% CO2 atmosphere at 37°C. 4. Phosphate buffered saline (PBS): 8 g/l NaCl, 0.2 g/l KCl, 1.15 g/l Na2HPO4, 0.2 g/l KH2PO4, pH 7.4. 5. Trypsin/EDTA: 0.25% Trypsin; 2 g/l EDTA solution from Sigma (see Note 2). 6. 100 mm surface-treated tissue culture dishes (Cellstar®; Greiner Bio-One). 2.2. Live-Cell Imaging
1. Translight™ (inventive-ts, Innsbruck, Austria): patented culture device for long-term live-cell imaging in a controlled humidified atmosphere at constant temperature (±0.1°C at 36.6°C), which is compatible with inverted microscopes of most manufacturers. The Translight™ base-unit for static cell culture incubation can be upgraded for perfusion culture. Figure 1 shows a representation of the Translight™ culture device (for further information see: www.inventive-ts.com).
Fig. 1. Translight™. Live-cell imaging unit for static long-term incubation of adherent cell cultures on the stage of inverted microscopes. Copyright: inventive-ts, Innsbruck, Austria, published with permission.
172
J. Lechner et al.
2. Hemocytometer (Neubauer chamber). 3. Inverted Microscope (Eclipse TE-300; NIKON, Austria). 4. Digital camera (Digital Sight DS-U1, NIKON, Austria). 5. Temperature feedback-control unit (Julabo-MP, Seelbach, Germany). 6. Gas supply (humidified bio-ariacarb gas mixture; 95% air/5% CO2; SOL, Mantova, Italy).
3. Methods 3.1. Standard Cell Culture
1. LLC-PK1 cells are routinely cultured on 100 mm surfacetreated tissue culture dishes. The cell culture medium is replaced with fresh medium (7 ml) every 48 h. 2. The cells are subcultured when reaching highly confluent density as indicated by the appearance of epithelial domes. Usually, we passage the cultures once a week at a split ratio of about 1:10. 3. For removal from the culture dish, confluent cell layers on 100-mm plastic dishes are washed twice with 10 ml PBS and then treated with 1.5 ml Trypsin/EDTA solution at 37°C for about 5–10 min. Treatment with trypsin is stopped by the addition of 10-ml culture medium after all the cells have shed from the growth surface (see Note 3). 4. In order to obtain a cell suspension containing mainly single cells, the solution is carefully pipetted up and down several times by carefully avoiding the formation of air bubbles. 5. 1 ml of the cell suspension is added to 6 ml of culture medium transferred into a new 100-mm culture dish.
3.2. Permanent Cell Culture on the Stage of the Microscope
1. The Translight™ incubation wells equipped with glass platelets or other polymeric transparent growth surfaces (area 3 cm2) are mounted in a five-well incubation revolver and sterilized by autoclaving at 121°C and 1 atm overpressure. 2. Trypsinized and gently singularized LLC-PK1 monolayer cells are counted in a hemocytometer. Cells are then seeded directly into the Translight™ incubation wells at confluent density (2 × 105 cells/cm2) under sterile conditions (see Notes 4 and 5). A total of 0.4 ml of culture medium is used per well. The inoculated Translight™ wells in the incubation revolver are covered by a vented lid and put in a standard cell culture incubator for at least 4 h (or overnight) to allow the cells to settle and attach to the growth surface (see Note 6). 3. When the culture is ready to be monitored, the Translight™ base-unit is preassembled on the stage of the inverted microscope by connecting it to the temperature and gassing control
11 Monitoring of the Dynamics of Epithelial Dome Formation Using a Novel Culture...
173
units in order to allow constant temperature (see Note 7) and gassing with humidified 95% O2/5% CO2. Equilibration of incubation conditions is reached within 10 min. Finally, the five-well incubation revolver is transferred from the incubator to the pre-equilibrated Translight™ base-unit for continuous live-cell imaging. 4. After adjusting focus and region of interest, the camera is set to repeatedly capture single light microscopic graphs for the generation of time-lapse motion pictures by a specialized software (DAcq-ProLive™, inventive-ts, Innsbruck, Austria). Micrograph capture intervals were ranging from 30 s to 5 min. Supplementary file Movie 1 (published in the electronic version) represents a time-lapse motion picture series of a highly confluent monolayer of LLC-PK1 forming domes and allows monitoring of the dynamics of domes. The movie visualizes how many smaller domes merge into one big dome, which shows pulsating up and down movements of the dome area over time. Selected images of the series used for the movie shown in Movie 1 are represented by Fig. 2. Figure 2a shows
Fig. 2. Dynamics of dome formation monitored by cell culture under the microscope. The sequence of light microscopic images shows LLC-PK1 cells in culture on borosilicate
174
J. Lechner et al.
Fig. 2. (continued) glass as growth support after having reached confluence in cell culture. The graphs represent a selection of single pictures derived from the time-lapse motion picture series published as supplementary file Movie 1 (available in the electronic version). The images in part (a) show how a number of smaller domes merge into a very large dome. Domes are marked by D and their margins are labeled by a dashed line. In part (b) apoptotic cell extrusion can be monitored at the margins of domes, which occurred at low frequency by spontaneous apoptosis. Monitored cells are labeled by arrows, the first appearance of apoptotic cell extrusion is labeled by A. A 20× objective was used giving a total magnification of 200-fold, and the obtained images were used for (a). Part (b) represents in more detail the lower right corner of the original pictures electronically magnified by a factor of 3 with respect to the original captures. The graphs in (a) were selected to represent single captures 100 min apart from each other, in (b) they were obtained 20 min apart. The focal plane is set to sharply represent cells that are well attached to the growth surface. Areas of dome formation comprise cell groups that detach from the surface forming fluid-filled blisters. Domes can be recognized by groups of cells moving out of the focal plane showing less sharp cellular structures and cell–cell interfaces.
how a number of smaller domes merge into a very large dome. Figure 2b shows in more detail a number of spontaneously occurring apoptoses at the margins of a dome. 5. The Translight™ base-unit is equipped with five separate well holders. Thus, up to five parallels or differing treatments are
11 Monitoring of the Dynamics of Epithelial Dome Formation Using a Novel Culture...
175
recordable during the same experimental run by using a motorized stage for precise automated repositioning (e.g., Märzhäuser Scan-series 100 × 100 mm; Wetzlar, Germany). 6. Incubation media are replaceable directly on stage of the microscope at any time during long-term culture and simultaneous live-cell imaging by inserting sterile syringe needles into the sterile filler-ports on the top cover plate unit of the Translight™. Similarly, drugs, inhibitors, or toxins are added (see Note 8). Supplementary file Movie 2 (published in the electronic version) shows a time-lapse motion picture series of cells after attachment, which start to divide and fill up the whole growth area. In addition, the last part of the movie shows an intoxication experiment leading to apoptotic cell death. 7. Via introducing syringe needles into the filler-ports, it is also possible to manually obtain aliquots of the cell culture medium for measurements of medium-released metabolites or enzymes (see Notes 9 and 10). 8. At the end of the microscopic observation period, various end-point assays may be performed (see Note 11). 9. In addition to monitoring dome formation and dynamics in cultures of confluent transporting epithelia using phase contrast or differential interface contrast microscopy, the live-cell imaging device can be used for several other applications requiring long-term monitoring of cell cultures directly on the stage of a microscope (see Note 12).
4. Notes 1. Other cell lines widely used for studying epithelial barrier properties and forming domes on solid growth support are, for example, renal tubular MDCK cells, which show properties of distal tubulus and collecting duct cells (8), opossum kidney proximal tubular OK cells (9), and human intestinal epithelial Caco-2 cells (10). 2. We aliquot the Trypsin/ EDTA solution at first thawing to portions of 6 ml, which are stored at −20°C, in order to avoid repeated freeze/ thaw cycles, and thus preserve the maximal enzymatic activity of trypsin. 3. Optionally, cells may be centrifuged at 100 × g and room temperature for 5 min in order to completely remove the trypsin. A washing step with 10 ml PBS can be introduced followed by centrifugation. The pellet is carefully resuspended in 10 ml fresh culture medium. 4. Culturing the cells at confluent density allows reproducibility of the experiments and decreases the time necessary for
176
J. Lechner et al.
e pithelial dome formation. Further increasing the seeding density is counterproductive, since the cells are too crowded and establishment of a monolayer is hindered. 5. If monitoring of the proliferative phase of the culture is also of interest, cells may be diluted and seeded at 1 × 104 cells/cm2 or similar density instead (see supplementary file Movie 2 published in the electronic version). 6. The cups can also be directly mounted to the holding device and placed under the microscope. In this case, microscopic recordings should be started only after the cells have settled and attached to the growth surface. Within the first hours of culturing, repeated adjustment of the focus plane or use of an automated device for auto-focusing is necessary. 7. Temperature of the water bath is set at a slightly higher temperature than the desired culture medium temperature. We usually operate the water bath at 38°C in order to obtain a culture temperature of 36.8°C. The exact settings have to be adjusted individually to compensate for differing laboratory room climate. 8. Treatments with specific cytokines (11–15), drugs (16), or heavy metals (17) are known to affect epithelial barrier permeability. 9. By regularly removing aliquots of the supernatant medium, it is, for example, possible to measure the enzymatic activity of lactate dehydrogenase in the culture medium, which is a good indicator of cellular necrosis (e.g., by the “Cytotoxicity Detection Kit LDH” of Roche Diagnostics). 10. Alternatively to manually remove aliquots, mountings can be purchased to use the Translight™ device also for the simultaneous quantification of gaseous or media-dissolved compounds by coupling the live-cell imaging unit to gas or liquid chromatography devices. 11. We performed immunofluorescence microscopy analysis, Western Blot analyses, cell viability assays, apoptosis assays, etc. 12. If equipped with a fluorescent microscope, the same protocol can be used with cells transfected with adherens or tight junctional proteins labeled with green fluorescent protein, for example, in order to monitor the dynamics of junctional complex assembly during establishment of the confluent monolayer. Other possible applications are treatments with fluorescent dyes suitable for living cells, for example, FLICA apoptosis detection kits (Immunochemistry Technologies, LLC), which can be used to monitor the timing of caspase activation during an apoptotic process.
11 Monitoring of the Dynamics of Epithelial Dome Formation Using a Novel Culture...
177
Acknowledgments The presented work was part of scientific projects supported through grant No. P17583-B13 of the Austrian Science Fund (FWF, Dr. J. Lechner) and grant No. UNI-0404/229 of the Tyrolean Science Fund (Tiroler Wissenschaftsfond, Dr. J. Lechner). The live-cell imaging device was developed by Dr. T. Seppi, H. Gatt, and M. Voelp (patent AT 500473 B1 2006). Their work was supported through grants NanoBII-815828: PolyCell and NanoBII-0353: GEMID of the Austrian NANO Initiative (a thematic program of the FFG – Austrian Research Promotion Agency). References 1. Shin K, Fogg VC, and Margolis B (2006) Tight Junctions and Cell Polarity. Annu Rev Cell Dev Bi 22: 207–235. 2. Van Itallie CM, and Anderson JM (2004) The Molecular Physiology of Tight Junction Pores. Physiology 19: 331–338. 3. Schneeberger EE, and Lynch RD (2004) The tight junction: a multifunctional complex. Am J Physiol Cell Physiol 286: C1213-1228. 4. Tsukita S, Furuse M, and Itoh M (2001) Multifunctional strands in tight junctions. Nat Rev Mol Cell Biol 2: 285–93. 5. Misfeldt DS, Hamamoto ST, and Pitelka DR (1976) Transepithelial transport in cell culture. P Natl Acad Sci USA 73: 1212–1216. 6. Cereijido M, Robbins E, Dolan W, Rotunno C, and Sabatini D (1978) Polarized monolayers formed by epithelial cells on a permeable and translucent support. J. Cell Biol. 77: 853–880. 7. Leighton J, Brada Z, Estes LW, and Justh G (1969) Secretory Activity and Oncogenicity of a Cell Line (MDCK) Derived from Canine Kidney. Science 163: 472–473. 8. Lever JE (1979) Inducers of mammalian cell differentiation stimulate dome formation in a differentiated kidney epithelial cell line (MDCK). P Natl Acad Sci USA 76: 1323–1327. 9. Rizzoli R, and Bonjour J (1987) Effect of dexamethasone on parathyroid hormone stimulation of cyclic AMP in an opossum kidney cell line. J Cell Physiol 132: 517–23. 10. Grasset E, Pinto M, Dussaulx E, Zweibaum A, and Desjeux JF (1984) Epithelial properties of human colonic carcinoma cell line Caco-2: electrical parameters. Am J Physiol Cell Physiol 247: C260-267.
11. McKay DM, and Baird AW (1999) Cytokine regulation of epithelial permeability and ion transport. Gut 44: 283–289. 12. Nusrat A, Turner JR, and Madara JL (2000 Nov) Molecular physiology and pathophysiology of tight junctions. IV. Regulation of tight junctions by extracellular stimuli: nutrients, cytokines, and immune cells. Am J Physiol Gastrointest Liver Physiol 279: G851-7. 13. Lechner J, Malloth N, Seppi T, Beer B, Jennings P, and Pfaller W (2008) IFN-{alpha} induces barrier destabilization and apoptosis in renal proximal tubular epithelium. Am J Physiol Cell Physiol 294: C153-C160 [Epub 2007, Nov 21]. 14. Lechner J, Malloth NA, Jennings P, Hekl D, Pfaller W, and Seppi T (2007) Opposing roles of EGF in IFN-{alpha}-induced epithelial barrier destabilization and tissue repair. Am J Physiol Cell Physiol 293: C1843-1850 [Epub 2007, Oct 3]. 15. Lechner J, Krall M, Netzer A, Radmayr C, Ryan M, and Pfaller W (1999) Effects of interferon alpha-2b on barrier function and junctional complexes of renal proximal tubular LLC-PK1 cells. Kidney Int 55: 2178–91. 16. Martin-Martin N, Ryan G, McMorrow T, and Ryan MP (2010) Sirolimus and cyclosporine A alter barrier function in renal proximal tubular cells through stimulation of ERK1/2 signaling and claudin-1 expression. Am J Physiol Renal Physiol 298: F672-682. 17. Forti E, Bulgheroni A, Cetin Y, Hartung T, Jennings P, Pfaller W, and Prieto P (2010) Characterisation of cadmium chloride induced molecular and functional alterations in airway epithelial cells. Cell Physiol Biochem 25: 159–68.
178
J. Lechner et al.
Electronic Supplementary Material Movie 1 Dynamics of dome formation monitored by cell culture under the microscope Light microscopic images of LLC-PK1 cells in culture on borosilicate glass as growth support were obtained after the cells have reached confluence in cell culture. Phase contrast light microscopy was performed using a 20× objective (giving a total magnification of 200-fold). For the timelapse motion picture series, single light microscopic images taken automatically every 2 min were combined. The focal plane was set to sharply represent the cells that were well attached to the growth surface. Areas of dome formation comprise cell groups that detach from the surface forming fluid-filled blisters. Domes can be recognized by groups of cells moving out of the focal plane showing less sharp cellular structures and cell–cell interfaces. At the beginning of the sequence, small regions with dome formation are already present, which merge into a very large dome in the course of the movie. It is noteworthy to see pulsating up and down movements of the whole dome area after the big dome has formed. In the first few hours, sporadic apoptotic cell extrusion can be monitored at the margins of domes, which are highlighted in Fig. 2b. (Visit http://extras.springer.com/ to view the movie.) Movie 2 Cell growth to confluence and apoptotic cell death induced by a toxin A sequence of microscopic images of LLC-PK1 cells in culture on borosilicate glass as growth support is shown as a motion picture series. Single light microscopic images were taken every 2 min. Part 1 of the movie shows cells after attachment, which start to divide and fill up the growth support area using a 10× microscope objective, part 2 shows in more detail cells forming a confluent monolayer using a 20× objective, part 3 shows cellular movements and cell death after the addition of an apoptosis-inducing toxin (100 mM cisplatin) to the culture medium monitored by Nomarski differential interference contrast microscopy using a 40× objective. Part 1 and 2 were recorded using phase contrast microscopy. Copyright: inventive-ts, Innsbruck, Austria, published with permission. (Visit http://extras.springer.com/ to view the movie.)
Chapter 12 Measuring Permeability in Human Retinal Epithelial Cells (ARPE-19): Implications for the Study of Diabetic Retinopathy Marta Garcia-Ramírez, Marta Villarroel, Lídia Corraliza, Cristina Hernández, and Rafael Simó Abstract The retinal pigment epithelium (RPE) is a specialized epithelium lying in the interface between the neural retina and the choriocapillaris where it forms the outer blood–retinal barrier (BRB). The tight junctions (TJ)s expressed in the outer BRB control fluids and solutes that enter the retina and this sealing function, which is essential for the retinal homeostasis, is impaired in diabetic retinopathy. In this chapter, we provide the methods to explore the function of the RPE barrier by measuring Transepithelial electrical resistance (TER) and paracellular permeability to dextran in cultures of ARPE-19 cells (an immortalized RPE cell line). A method for inducing a lesion mimicking which occurs in diabetic retinopathy is described. In addition, methods for assessing mRNA expression and protein content of the main TJ proteins (occludin, zonula occludens-1 [ZO-1]) are detailed. Finally, we provide the methods required for confocal immunofluorescence detection of the TJ proteins, as well as for assessing the capacity of ARPE-19 cells to retain their functional properties. Key words: ARPE-19 cells, Retinal pigment epithelium, Tight junctions, Blood–retinal barrier, Diabetic retinopathy, Transepithelial electrical resistance, Dextran permeability
1. Introduction The retinal pigment epithelium (RPE) is a monolayer of pigmented cells lying in the interface between the neuroretina and the choroids. The RPE is of neuroectodermal origin and is therefore considered to be part of the retina. The apical membrane of the RPE faces the photoreceptors’ outer segments and its basolateral membrane faces Bruch’s membrane, which separates the RPE from the fenestrated endothelium of
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_12, © Springer Science+Business Media, LLC 2011
179
180
M. Garcia-Ramírez et al.
Fig. 1. Retinal section of normal retina stained with hematoxilin–eosin showing the location of the retinal pigment epithelium (RPE). GCL ganglion cell layer, INL inner nuclear layer, ONL outer nuclear layer, PR photoreceptors. Scale bar, 10 mm.
the choriocapillaris (Fig. 1). The RPE constitutes the outer blood–retinal barrier (BRB). The inner BRB is mainly constituted by endothelial cells. Tight junctions (TJ)s between neighboring RPE cells and neighboring endothelial cells are essential for the strict control of fluids and solutes that cross the BRB, as well as to prevent the entrance of toxic molecules and plasma components into the retina. Therefore, this sealing function is essential for the integrity of the retina (1). Apart from the barrier function, RPE participates in (1) the absorption of light and protection against photooxidation; (2) the reisomerization of all-trans-retinal into 11-cis-retinal, which is a key element of the visual cycle; (3) the phagocytosis of shed photoreceptor membranes; (4) the secretion of various factors essential for the structural integrity of the retina; (5) the immunoprivileged status of the eye (1–3). With these different complex functions, the RPE is essential for visual function. A failure of any one of these functions can lead to degeneration of the retina, loss of visual function, and blindness. Diabetic retinopathy (DR) remains the leading cause of blindness among working-age individuals in developed countries (4). Whereas proliferative diabetic retinopathy (PDR) is the commonest sight-threatening lesion in type 1 diabetes, diabetic macular edema (DME) is the primary cause of poor visual acuity in type 2 diabetes. Because of the high prevalence of type 2 diabetes, DME is the main cause of visual impairment in diabetic patients (5). In addition, DME is almost invariably present when PDR is detected
12 Measuring Permeability in Human Retinal Epithelial Cells (ARPE-19)…
181
in type 2 diabetic patients (6). Neovascularization due to severe hypoxia is the hallmark of PDR, whereas vascular leakage due to the breakdown of the BRB is the main event involved in the pathogenesis of DME (7, 8). Most of the research on the physiopathology of DR has been focused in the impairment of the neuroretina and the breakdown of the inner BRB. By contrast, the effects of diabetes on the RPE have received less attention. In this chapter, we provide the methods to explore the function of the RPE barrier by measuring transepithelial electrical resistance (TER) and paracellular permeability to dextran in cultures of ARPE-19 cells. This is a spontaneously immortalized cell line that has been commonly used as a model for the outer BRB because it has been demonstrated to have structural and functional properties characteristic of in vivo RPE cells (9). The procedures indicated above have been performed in standard conditions and after inducing a lesion by using high glucose concentrations and IL-1b, thus mimicking what occurs in the diabetic milieu (10). In addition, methods for assessing mRNA expression and protein content of the main TJ proteins (occludin, zonula occludens-1 [ZO-1], claudin-1) are described. Finally, methods required for confocal immunofluorescence detection of the TJ proteins mentioned above have been detailed. This is useful not only to quantify the expression and spatial distribution of the TJ proteins but also to demonstrate the establishment of a differentiated monolayer and provide evidence that ARPE-19 cells in culture retain the functionally polarized characteristics of the RPE. This latter condition is demonstrated by showing the apical localization of both TJ proteins and Na+/K+ ATP-ase activity (Fig. 2).
2. Materials 2.1. Human RPE Cell Culture
1. ARPE-19, a spontaneously immortalized human RPE cell line, obtained from the American Type Culture Collection (CRL-2302; ATCC; Manassas, VA, USA). 2. Dulbecco’s Modified Eagle’s Medium (DMEM) and Ham’s F12 medium with 2.50 mM l-glutamine supplemented with 10% fetal bovine serum (FBS; Hyclone; Thermo Fisher Scientific Inc, MA, USA) and 1% penicillin/streptomycin (Hyclone; Thermo Fisher Scientific Inc, MA, USA). Commercial medium without glucose, supplemented to final 5.5 mM or 25 mM d-Glucose in order to mimic the euglycemic and hiperglycemic medium, respectively (see Note 1). 3. Dulbecco’s PBS (1×) without Ca and Mg (PAA Laboratories GMBH; Pasching, Austria).
182
M. Garcia-Ramírez et al.
Fig. 2. Immunohistochemical characterization of the ARPE-19 monolayer maintained in 25 mM d-glucose 21 days. Confocal images showing the expression of ZO-1 (red)/Claudin-1 (green ) (a, b); Na+/K+ ATPase (red)/Occludin (green) (c, d), and DAPI (blue). IL-1b treatment (48 h) induces disruption of TJ organization (b, d). At the bottom of each panel Z-projection, the apical location of TJ proteins or Na+/K+ ATP-ase is revealed.
4. 0.05% Trypsin, 0.02% EDTA solution (Hyclone; Thermo Fisher Scientific Inc, MA, USA). 5. IL-1b (Preprotech; Rock Hill, NJ, USA). 6. Tissue culture dishes (75 cm2) (Costar; Corning Inc., NY, USA). 7. Refrigerated centrifuge PR4i (Thermo Electron Corporation, MA, USA). 8. Cell incubator IGO150 with control temperature and CO2. 37°C, 5% CO2 level, and humidity >95%, (Thermo Electron Corporation, MA, USA). 9. Sterile Bio-II-A. Class II Cabinet (Telstar, Bristol, PA, USA). 2.2. Measurement of Paracellular Epithelial Electrical Resistance
1. Polyester Membrane Transwell Inserts (HTS, Costar; Corning Inc., NY, USA) with a 0.4-mM pore size and 0.33 growth surface area (cm2). 2. Epithelial voltmeter (MILLICELL-ERS; Millipore, Billerica, MA, USA) with the STX100C (suitable for transwells of 24-well plates) electrode (World Precision Instruments, Sarasota, FL, USA) (see Note 2).
12 Measuring Permeability in Human Retinal Epithelial Cells (ARPE-19)…
2.3. Measurement of Permeability to Dextran
183
1. Polyester Membrane Transwells Inserts (HTS, Costar; Corning Inc., NY, USA) with a 0.4-mM pore size and 0.33 growth surface area (cm2). 2. Fluorescein isothiocyanate (FICT) dextran (40,000, 70,000 Da) (Sigma, St. Louis, MO, USA) (see Note 3). 3. Microplate reader (SpectraMax Gemini; Molecular Devices, Sunnyvale, CA, USA).
2.4. SDSPolyacrylamide Gel Electrophoresis
1. Resolving buffer (4×): 1.5 M Tris, pH 8.54, 0.4% SDS. Store at 4°C. 2. Stacking buffer (4×): 0.5 M Tris, pH 6.8, 0.4% SDS. Store at 4°C. 3. Thirty percent acrylamide/bis solution (29:1 with 3.3%C) (Acrylamide/Bis, Bio-Rad Laboratories, Hercules, CA, USA) and N,N,N,N¢-Tetramethyl-ethylenediamine (TEMED, Sigma; St. Louis, MO, USA). 4. Ammonium persulfate: prepare 20% solution in water and immediately freeze in single use (200 mL) aliquots at −20°C. 5. Running buffer (10×): 0.25 M Tris, 1.92 M glycine, 1% (w/v) SDS. Store at 4°C. Dilute 100 mL with 900 mL water for use. 6. Prestained molecular weight markers: Kaleidoscope markers (Bio-Rad, Hercules, CA, USA).
2.5. Western Blotting for Tight Junctions
1. Lysis buffer RIPA (Sigma; St. Louis, MO, USA) in 1 mM PMSF, 2 mM Na3VO4, 100 mM NaF containing 1× Protease Inhibitor Cockail (Sigma; St. Louis, MO, USA) (see Note 4). 2. Transfer buffer (10×): 0.25 M Tris, 1.92 M glycine, 10% (v/v) methanol (add prior to use). Store at 4°C. 3. Nitrocellulose membrane from GE Healthcare (Amersham Hybond™ ECL™) (GE Healthcare Bio-Sciences Corp., Waukesha, WI, USA), sponges from Bio-Rad and QuickdrawTM Blotting Paper from Sigma (St. Louis, MO, USA). 4. Tris-buffered saline (TBS): 100 mM NaCl, 100 mM Tris in water (see Note 5). 5. Tris-buffered saline with Tween (TBS-T): 100 mM NaCl, 100 mM Tris, 0.05% Tween in water (see Note 5). 6. Blocking buffer: 10% (w/v) nonfat dry milk in TBS-T (see Note 6). 7. Primary antibody dilution buffer: 10% (w/v) nonfat dry milk in TBS-T (see Note 7). 8. Primary antibodies: rabbit anti-claudin-1, rabbit anti-occludin, and mouse anti-ZO-1. (Zymed Lab Gibco; Invitrogen, San Diego, CA, USA).
184
M. Garcia-Ramírez et al.
9. Secondary antibodies: goat anti-rabbit or mouse horseradish peroxidase-conjugated secondary antibody (Pierce; Thermo Scientific, Rockford, IL, USA). 10. Enhanced chemiluminescence detection system (Supersignal CL-HRP Substrate System; Pierce; Thermo Scientific, Rockford, IL, USA). 2.6. Real-Time PCR
1. RNeasy Mini kit with DNAase (Qiagen Distributors, IZASA, Barcelona, Spain). 2. Spectrophotometer NanoDrop ND-1000 (Thermo Fisher Scientific, Wilmington, DE, USA). 3. TaqMan Reverse Transcription Reagents kit (Applied Biosystems, Madrid, Spain). 4. TaqMan specific gene expression assays (Applied Biosystems, Madrid, Spain): b-actin (Hs9999903_m1; Applied Biosystems, Madrid, Spain), ZO-1 (Zona occludens-1, Hs00268480_m1), OCLN (Occludin 1, Hs00170162_m1), and CLN-1 (claudin-1 Hs00221623_m1). 5. Thermo-cycler ABI PRISM 7900 HT (Applied Biosystems, Madrid, Spain).
2.7. Confocal Immuno fluorescence for Tight Junctions
1. Microscope circle cover-slips of glass (12 mm of diameter) from Thermo scientific, (Menzel-Gläser; Braunschweig, Germany). 2. 24-Well plates (Nunc; Thermo Fisher Scientific, Roskilde, Denmark). 3. Dulbecco’s phosphate-buffered saline (PBS) 1× with calcium and magnesium (PAA Laboratories GmbH; Pasching, Austria). 4. Fixing solution: Methanol (cold −20°C). 5. Blocking solution and antibody dilution buffer: PBS BSA 2%, 0.05% Tween. 6. Primary antibodies: Rabbit anti-claudin-1 or occludin, mouse anti-ZO-1 (Zymed Lab Gibco; Invitrogen, San Diego, CA, USA). Mouse anti-Na+/K+ ATPase (Millipore; Billerica, MA, USA). 7. Secondary antibody: Alexa 488 goat anti-rabbit and Alexa 594 donkey anti-mouse (Invitrogen; San Diego, CA, USA). 8. Vectashield mounting medium for fluorescence with DAPI (Vector Laboratories; Burlingame, CA, USA). 9. Confocal laser scanning microscope FV1000 (Olympus; Hamburg, Germany). 10. Microscope software: Fluoview 1.7.2.2. (Olympus; Hamburg, Germany). 11. Image processing and analysis: ImageJ software (National Institutes of Health, Bethesda, MD, USA).
12 Measuring Permeability in Human Retinal Epithelial Cells (ARPE-19)…
185
3. Methods In vitro models of RPE have been established by several groups based on ARPE-19 cell line culture (9–12). To obtain such models, ARPE-19 cells are cultured during 21 days with the objective of obtaining a monolayer that retains the in vivo morphological and physiological characteristics of native RPE. The presence of a polarized monolayer is one of the most important features assuring the functional integrity of ARPE-19 cells (13). The expression and location of markers such as TJ proteins at the apical side of the monolayer, as well as Na+/K+ ATPase are commonly used to assess the polarization of the monolayer (14) (Fig. 2). It is important to note that ARPE-19 cells grown directly in plastic rather than in coating supports (i.e., fibronectin, collagen) better retain the characteristics of native RPE (15). Given that IL-1b plays an essential role in the development of DR and contributes to retinal neurodegeneration (16–19), decreases transepithelial electrical resistance (TER), and increases permeability with alteration of tight junction content (10), we use this cytokine together with high glucose concentrations (25 mM) in order to mimic the diabetic milieu (Fig. 2). 3.1. ARPE-19 Culture
1. The ARPE-19 cells are grown until confluence in 75 cm2 tissue flasks in 8 mL medium with 10% FBS. 2. For subculturing, the medium is removed and cells rinsed with the same volume of PBS and then trypsinized. Trypsinization is carried out with 2 mL of trypsine solution in the incubator for 3–5 min until cell detachment. Trypsinization is stopped by adding the medium. Cells are transferred to a centrifuge tube and recovered by centrifugation (240 × g × 5 min) in a refrigerated centrifuge at 4°C. The supernatant is removed and the cell pellet is resuspended in fresh growth media and seeded onto new flasks. Next cell passages are obtained by dilution 1:3. 3. ARPE-19 cells at passage 23 are used to seed cultures suitable for obtaining RNA and protein extracts. At this point, a cell suspension of 20,000 cell/mL is obtained and split in a sixwell plate or Petri dishes. Culture is performed in confluent conditions for 18 days in complete medium. 4. Damage to ARPE-19 monolayer mimicking diabetic conditions: The diabetic milieu is mimicked by culturing ARPE-19 cells in media containing 25 mM glucose. In the 19th day of the experiment, serum is starved in the upper compartment and IL-1b (10 ng/mL) is added for 48 h until the end of the experiment (two doses of IL-1b, each 24 h).
186
M. Garcia-Ramírez et al.
The common method of obtaining differentiated ARPE-19 monolayers that resemble the in vivo cell state of the RPE consists of attaching a high density cell suspension to a transwell support that allows the development of a monolayer with basal and apical surfaces. In the present method, we use polyester transparent inserts to attach the cells. ARPE-19 cells attached to plastic membranes maintain differentiated characteristics and, in addition, provide good visibility under phase contrast or fluorescent microscopy (Fig. 3).
3.2. Measurement of Permeability to Dextran
1. The inserts pack is opened in sterile conditions. Complete medium (0.6 mL) is added to the wells of a 24-well culture cell plate. Then HTS transwells-24 are placed in the wells. 2. ARPE-19 cells (at passage 23) are seeded at 400,000 cells/mL (80,000 cells/well) in the upside of the transwells. The plates are covered and incubated at 37°C (5% CO2) in a tissue culture incubator. The monolayer is formed in the following 48 h. The medium is replaced each 3 days (see Note 8). 3. The permeability of RPE cells is determined at 18 days by measuring the apical-to-basolateral movements of FICT dextran (40 kDa). The test molecule is added to the apical compartment of the cells in a concentration of 100 mg/mL. 4. The cells grown under 25 mM d-glucose are treated with IL-1b (10 ng/mL, 1 application/day) during the last 48 h of the experiment (days 19 and 20) in order to mimick the tight junction disruption provoked by the diabetic milieu.
a
b [FICT-Dextran] (ng/ml/cm2)
TER (Ohm-cm2)
200
*
160
* 120 80 40 0
0
10
20
30
Time (hours)
40
50
250
* 200
*
150 100 50 0
0
20
40
60
80
Time (min)
Fig. 3. (a) Results of TER. The vertical axis represents the TER, expressed in Ohm × cm2, and the horizontal axis represents the time after the addition of the treatment (IL-1b). (b) Results of 40 kDa dextran permeability. The vertical axis is the concentration of dextran and the horizontal axis is the time after the addition of the molecule. ( ) 25 mM d-glucose; ( ) 25 mM d-glucose + IL-1b (10 ng/mL) 48 h. Dextran permeability is measured at 10, 40, and 75 min. Results are expressed as the mean ± SD. *p < 0.05.
12 Measuring Permeability in Human Retinal Epithelial Cells (ARPE-19)…
187
a Transwell® insert Upper compartment Lower compartment
ARPE-19 cells monolayer
Microporous membrane
b
Fig. 4. (a) Section of a transwell insert. ARPE-19 monolayer is established in a porous membrane where the upper side resembles the apical part of RPE and the lower compartment resembles the basal part of the RPE. (b) Image of HTS-24 wells with the epithelial voltmeter used for TER measurement.
5. 200 mL samples are collected from the basolateral side at 10, 40, and 75 min after adding the molecules. The medium in the basolateral compartment is replaced by fresh medium after the collection of every sample (see Note 9). A minimum of three wells are used for each time measurement. Absorbance is measured at 485 nm of excitation and 528 nm of emission with a microplate reader (Fig. 4). 3.3. Measurement of Paracellular Epithelial Electrical Resistance
1. ARPE-19 three-week monolayer cells are obtained similarly as described above. 2. Cells (0.8 × 106 cells/mL) are plated on permeable-membrane inserts in the complete medium (10% FBS) and maintained for 3 weeks in culture. At day 21, the complete medium is replaced by a depleted medium (1% FBS) on the apical side. 3. Transepithelial electrical resistance (TER) is measured by using an epithelial voltemeter according to the manufacturer’s instructions and following the method described by Dunn et al. (9) (see Notes 10 and 11).
188
M. Garcia-Ramírez et al.
4. Resistance measurements after subtraction of the background (resistance of transwells without cells) are expressed in ohms-cm2. A baseline measurement of transepithelial electrical resistance is obtained and TER changes are monitored at the beginning of the treatments and after 24 h and 48 h. Each condition is assayed in quadruplicate and at least two independent experiments are performed (Fig. 4). 3.4. SDS-PAGE
1. These methods assume the use of Mini PROTEAN 3 System (Bio-Rad). It is critical that the glass plates for the gels be cleaned with ethanol 70% after use and rinsed extensively with distilled water. 2. Prepare a 1.5-mm thick, 10% acrylamide/bis solution gel for occludin and claudin by mixing 2.5 mL of 4× resolving buffer, 3.4 mL acrylamide/bis solution, 4.1 mL water, 50 mL ammonium persulfate solution, and 5 mL TEMED. Pour the gel, leaving space for a stacking gel, and overlay with water. The gel should polymerize in about 30 min. 3. Prepare a 1.5-mm thick, 7.5% acrylamide/bis solution gel for ZO-1 by mixing 2.5 mL of 4× resolving buffer, 2.5 mL acrylamide/bis solution, 5 mL water, 50 mL ammonium persulfate solution, and 5 mL TEMED. Pour the gel, leaving space for a stacking gel, and overlay with water. The gel should polymerize in about 30 min. 4. Prepare the stacking gel (the same for the three proteins) by mixing 1.3 mL of 4× stacking buffer with 0.7 mL acrylamide/bis solution, 3.3 mL water, 30 mL ammonium persulfate solution, and 15 mL TEMED. The stacking gel should polymerize within 30 min. 5. Prepare the running buffer by diluting 100 mL of the 10× running buffer with 900 mL of water in a measuring cylinder. Cover with Para-Film and invert to mix. 6. Once the stacking gel has set, carefully remove the comb and wash the wells with running buffer. 7. Add the running buffer to the upper and lower chambers of the gel unit and load each sample into a well. Include one well for prestained molecular weight markers. 8. Complete the assembly of the gel unit and connect to a power supply. The gel should be run firstly at 90 V until the samples reach the resolving part of the gel, and then the voltage can be raised to 150 V. The dye front can be run off the gel for ZO-1 but be careful to stop it before the 37-kDa marker is lost. This permit us to perform the b-actin staining in the same samples. For the other two proteins it is better preserve the front.
12 Measuring Permeability in Human Retinal Epithelial Cells (ARPE-19)…
3.5. Western Blotting for Tight Junctions
189
1. After treatment, ARPE-19 cells are washed with ice-cold Dulbecco’s PBS (1×) without Ca and Mg. Protein is extracted with Lysis buffer. 2. 20 mg of total protein is resolved by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to supported nitrocellulose membranes electrophoretically. 3. These methods assume the use of a Mini Trans-Blot Cell (Bio-Rad). A tray of setup buffer is prepared that is large enough to lay out a transfer cassette with the sponge and the blotting paper submerged on one side. A sheet of nitrocellulose cut just larger than the size of the separating gel is laid on the surface of a separate tray of transfer buffer (1×) to allow the membrane to become wet by capillary action. The membrane is then submerged in the buffer on top of the blotting paper. 4. The gel unit is disconnected from the power supply and disassembled. The stacking gel is removed and discarded. The separating gel is then laid on top of the nitrocellulose membrane. 5. One further sheet of sponge and blotting paper is wetted in the buffer and carefully laid on top of the gel, ensuring that no bubbles are trapped in the resulting sandwich. Then the transfer cassette is closed. 6. The cassette is placed in the transfer tank so that the nitrocellulose membrane is between the gel and the anode. It is vitally important to ensure this orientation or the proteins will be lost from the gel into the buffer rather than transferred to the nitrocellulose. 7. Add the Bio-Ice cooling unit to the tank and a magnetic stirbar to ensure that the heat generated will be absorbed. Set the tank upon a magnetic agitator. 8. The lid is put on the tank and the power supply activated. Transfers can be accomplished at either 15 V overnight or 0.4 A for 1 h. 9. Once the transfer is complete the cassette is taken out of the tank and carefully disassembled, with the top sponge, sheets of blotting paper and gel removed. The nitrocellulose membranes are laid on a glass plate so that a cut in the corner can be made to ensure the correct orientation. The colored molecular weight markers should be clearly visible on the membrane. 10. The nitrocellulose is then cleaned with 2 min immersion in TBS-T and then incubated in 10 mL blocking buffer for 1 h at room temperature or overnight at 4°C on a rocking platform.
190
M. Garcia-Ramírez et al.
11. The blocking buffer is discarded and 7 mL of a 1:1,000 dilution of the primary antibody have to be added to the membrane in antibody dilution buffer for 1 h at room temperature on a rocking platform. 12. The primary antibody is then removed and the membrane is washed two times with 30 mL of TBS-T and two times more with 30 mL of TBS for 15 min each. 13. 7 mL of the secondary antibody is freshly prepared for each experiment as 1:10,000-fold dilution in blocking buffer for the antimouse and 1:20,000-fold dilution in blocking buffer for the anti-rabbit and added to the membrane for 1 h at room temperature on a rocking platform. 14. The secondary antibody is discarded and the membrane is washed two times with TBS-T and two times with TBS for 15 min each. 15. During the final wash, 750 mL aliquots of each portion of the ECL reagent were warmed separately to room temperature and mixed just before removing the final wash from the blot. Then the ECL is immediately added to the blot, which is then rotated by hand for 5 min to ensure even coverage. 16. The blot is removed from the ECL reagents, and placed into a saran wrap paper envelope. The remaining steps are done in a dark room under safe light conditions. 17. The membrane is then placed in a X-ray film cassette with film for a suitable exposure time, typically no more than 5 min. 3.6. Real-Time PCR
1. The total RNA is extracted from monolayers with the RNeasy Mini kit with DNAase digestion. 2. Quantification and quality of total mRNA are determined with a spectrophotometer NanoDrop ND-1000. 3. Reverse transcription is carried out with 1 mg of total RNA. The cDNA (40 ng) is used as a template for Real-Time PCR with the specific TJ assays and master mix (20 ml of total volume). Real-Time reactions are conducted as follows: 95°C for 10 min and 50 cycles of 15 s at 95°C and 1 min at 60°C. Each sample is assayed in triplicate. Reactions are performed on ice (see Note 12). 4. Automatic relative quantification data (R.Q.) is obtained in an ABI Prism 7900 (SDS software; Applied Biosystems, Madrid, Spain) using b-actin gene as the endogenous reference gene.
3.7. Confocal Immuno fluorescence
Confocal immunflorescence is essential to demonstrate wellstructured TJs and the polarity of the formed monolayer. For this purpose, immunofluorescence for ZO-1, occludin, claudin-1, and Na+/K+ ATPase is performed.
12 Measuring Permeability in Human Retinal Epithelial Cells (ARPE-19)…
191
1. Cover-slips must be sterilized with ethanol 70% and washed with PBS. Once dried place them in a 24-well plate. 2. ARPE-19 cells are seeded at a concentration of 20,000 cells/mL and maintained in 5.5 or 25 mM of d-glucose for 21 days as described above. 3. The cells grown under 25 mM d-glucose are treated with IL-1b (10 ng/mL) during 48 h (1 application/day) the last 2 days of the experiment (days 19 and 20) in order to mimick the TJ disruption provoked by the diabetic milieu. 4. Culture media is removed and the cells are washed with PBS for 5 min. 5. Cold methanol (−20°C) is added for 10 min at room temperature to fix the cells. 6. The methanol is discarded and the samples are washed twice for 5 min each with PBS. 7. The cells are blocked by incubation in antibody dilution buffer (PBS BSA 2% 0.05% Tween) at 4°C overnight. 8. The blocking solution is removed and primary antibodies rabbit anti-claudin-1, rabbit anti-occludin, mouse anti-ZO-1, and mouse anti-Na+/K+ ATPase are added, all diluted 1:200 in antibody dilution buffer and incubated for 1 h at room temperature. 9. The primary antibody is removed and the cells are washed three times for 5 min each with PBS. 10. The samples are incubated for 1 h at room temperature and in the dark with secondary antibodies such as Alexa 488 goat anti-rabbit and Alexa 594 donkey anti-mouse diluted 1:200 in antibody dilution buffer (see Note 13). 11. The secondary antibodies are discarded and the cells are washed three times for 5 min each with PBS. 12. To mount the samples each cover-slip must be inverted carefully into a drop of mounting medium for fluorescence with DAPI on a microscope slide (see Note 14). The cover-slip can be sealed with nail varnish (see Note 15). Avoid air bubbles in the mounting medium (see Notes 16 and 17). 13. The slides are viewed under confocal microscopy and the images are acquired by sequential scanning using a ×60 oil objective and the appropriate filter combination. Serial (z) sections are captured with a 0.25-mm step size through the thickness of the ARPE-19 monolayer until profiles of the immunolabeled tight junctions are no longer detectable. Images are taken at a resolution of 800 × 800 pixels with the same exposure settings and saved as TIFF files. Fluoview 1.7.2.2 software is used to project the serial sections into one image. ImageJ, a freely available java-based public-domain image processing program can also be used.
192
M. Garcia-Ramírez et al.
4. Notes 1. Pay attention to DMEM/F12 culture mediums. Not all DMEM/F12 mediums have the same composition. Use recommended culture medium for ARPE-19 (ATCC, Hyclone, Gibco (Invitrogen)). 2. Use a suitable epithelial voltmeter (STX100C) for the kind of transwell support that you use. The present protocol assumes the use of HTS transwell-24 Ref: 3379. 3. Store FICT dextran and the samples collected from the lower (basolateral) compartment of the transwells in a place protected from light. 4. The lysis solution for proteins must be freshly prepared. PMSF must be well dissolved. Avoid using precipitated PMSF. 5. It is recommended that both washing solutions for western blot (TBS-T and TBS) are used within 2 weeks. 6. Be careful with the milk used for blocking the membrane. It must be nonfat dry milk without bifidus. Check the date of the product because out of date milk can damage the membrane. Use the blocking buffer within 2 days. 7. The primary antibody can be saved for subsequent experiments. Store the primary antibody dilution used at −20°C. 8. The ARPE-19 monolayer should be intact. Be extremely careful when aspirating the medium. 9. In the permeability study, two different methods are possible for sample collection: (a) collect samples (200 mL) from the lower chamber and replace immediately with the same volume of fresh medium to maintain equilibrium. (b) Collect samples (200 mL) from the lower chamber, remove completely the volume of the lower chamber and replace immediately with fresh medium (600 ml) (this is the method we follow in the present protocol). 10. Measurements of TER should be taken after changing the cell culture medium (200 ml in the upper compartment of the transwell and 600 ml in the lower compartment are the recommended volumes). Differences in the total volume can affect the readings. Remember to use a warm culture medium. 11. Try to maintain a fixed temperature when measuring TER. Temperature changes affect the readings. 12. When preparing the samples for PCR, use micropipettes kept for PCR and filter tips. Be careful to maintain the samples and the PCR plate on ice.
12 Measuring Permeability in Human Retinal Epithelial Cells (ARPE-19)…
193
13. Protect the samples from light during incubation with Alexa secondary antibodies in order to avoid a reduction of the fluorescence. 14. Avoid using too much volume of mounting medium because cell monolayers could be damaged. 3–8 ml is the recommended volume of mounting medium. 15. Sealing the cover-slip with a bright color of nail varnish is useful to preserve sample integrity in case of long-term storage. Keep the samples at 4°C and protected from light. 16. Be careful with air bubbles when mounting the samples. Slow and careful application of the cover-slip can reduce the number of air bubbles. 17. Let the samples settle for some time (approximately 2 h) before performing microscopic observation. References 1. Strauss O (2005) The retinal pigment epithelium in visual function. Physiol Rev 85:845–881 2. Holtkamp GM, Kijlstra A, Peek R, de Vos AF (2001) Retinal pigment epithelium-immune system interactions: cytokine production and cytokine-induced changes. Prog Retin Eye Res 20:29–48 3. Simó R, Villarroel M, Corraliza L, Hernández C, Garcia-Ramírez M (2010) The retinal pigment epithelium: something more than a constituent of the blood-retinal barrier. Implications for the pathogenesis of diabetic retinopathy. J Biomed Biotechnol 2010: 190724 4. Congdon N, Friedman DS, Lietman T (2006) Important causes of visual impairment in the world today. JAMA 290:2057–2060 5. Lightman S, Towler HM (2003) Diabetic retinopathy. Clin Cornerstone 5:12–21 6. Tong L, Vernon SA, Kiel W, Sung V, Orr GM (2001) Association of macular involvement with proliferative retinopathy in type 2 diabetes. Diabet Med 18:388–394 7. Simó R, Carrasco E, García-Ramírez M, Hernández C (2006) Angiogenic and antiangiogenic factors in proliferative diabetic retinopathy. Curr Diabet Rev 2:71–98 8. Joussen A, Smyth N, Niessen C (2007) Pathophysiology of diabetic macular edema. Dev Ophthalmol 39:1–12 9. Dunn KC, Aotaki-Keen AE, Putkey FR, Hielmeland LM (1996) ARPE-19 a human retinal pigment epithelial cell line with differentiated properties. Exp Eye Res 62:155–169
10. Abe, T, Sugano, E, Saigo, Y, Tamai, M (2003) Interleukin-1b and barrier function of retinal pigment epithelial cells (ARPE-19): aberrant expression of junctional complex molecules. Invest Ophthalmol Vis Sci 44:4097–4104 11. Phillips BE, Cancel L, Tarbell JM, Antonetti DA (2008) Occludin independently regulates permeability under hydrostatic pressure and cell division in retinal pigment epithelial cells. Invest Ophthalmol Vis Sci 49:2568–2576 12. Villarroel M, García-Ramírez M, Corraliza L, Hernández C, Simó R (2009) Effects of high glucose concentration on the barrier function and the expression of tight junction proteins in human retinal pigment epithelial cells. Exp Eye Res 89:913–920 13. Philp NJ, Wang D, Yoon H, Hjelmeland LM (2003) Polarized expression of monocarboxylate transporters in human retinal pigment epithelium and ARPE-19 cells. Invest Ophthalmol Vis Sci 44:1716–1721 14. Kannan R, Zhang N, Sreekumar PG, Spee CK, Rodríguez A, Barron E, Hinton DR (2006) Stimulation of apical and basolateral VEGF-A and VEGF-C secretion by oxidative stress in polarized retinal pigment epithelial cells. 12: 1646–1659 15. Tian, J, Ishibashi, K, Handa, JT ( 2004). The expression of native and cultured RPE grown on different matrices. Physiol Genomics. 17:170–182 16. Kowluru RA, Odenbach S (2004) Role of interleukin-1beta in the pathogenesis of diabetic retinopathy. Br J Ophthalmol 88: 1343–1347
194
M. Garcia-Ramírez et al.
17. Gerhardinger C, Costa MB, Coulombe MC, Toth I, Hoehn T, Grosu P (2005) Expression of acute-phase response proteins in retinal Müller cells in diabetes. Invest Ophthalmol Vis Sci. 46:349–357 18. Demircan N, Safran BG, Soylu M, Ozcan AA, Sizmaz S (2006) Determination of vitreous
interleukin-1 (IL-1) and tumour necrosis factor (TNF) levels in proliferative diabetic retinopathy. Eye 20:1366–1369 19. Vincent JA, Mohr S (2007) Inhibition of caspase-1/interleukin-1beta signaling prevents degeneration of retinal capillaries in diabetes and galactosemia. Diabetes. 56:224–230
Chapter 13 Analysis of Epithelial Barrier Integrity in Polarized Lung Epithelial Cells Monika Strengert and Ulla G. Knaus Abstract Epithelial surfaces of the body are a key component of host defense by providing a mechanical barrier against potentially harmful substances. The respiratory tract is constantly challenged by a wide range of airborne pathogens and particulates, and provides not only a mucosal barrier, but also an intricate innate immune defense system. Disruption of the alveolar epithelial barrier can lead to acute lung injury, pneumonia, and acute respiratory distress syndrome. Protection of lung epithelial integrity, or repair of hyperpermeability with keratinocyte growth factor or Hsp90 inhibitors, is crucial for combating permeability edema. Ex vivo-differentiated lung epithelium represents a physiologically relevant tool for analyzing the effect of pathogens, chemicals, or drugs on lung barrier function. The integrity of the lung epithelial layer can be determined by several approaches. By combining two of these techniques, transepithelial electrical resistance and paracellular flux of fluorescent molecules, information about barrier integrity can be obtained in a prompt and convenient manner. As example, the virus- or bacterial toxin-mediated disruption of an ex vivo-differentiated mucociliary lung epithelial barrier is used here for assessing advantages and limitations of these methods. Key words: Epithelial resistance, Epithelial permeability, Lung epithelium, Influenza A virus, FITC-dextran
1. Introduction The human airways are a highly specialized organ and play a fundamental role for the host organism. They provide a vast surface area to the ambient air, thereby allowing efficient gas exchange of oxygen and the metabolic waste product carbon dioxide. However, the pulmonary epithelium is also exposed to inhaled particles and pollutants, toxins, and a wide range of airborne pathogens. Therefore, airway integrity is crucial for normal host function and
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_13, © Springer Science+Business Media, LLC 2011
195
196
M. Strengert and U.G. Knaus
needs to provide a selective barrier, which permits the passage of solutes and ions through paracellular spaces, but prevents pathogen or pollutant migration from lumen to interstitium. This is achieved by multiple mechanisms, including tight junctions, adherence junctions, and desmosomes (1). Analysis of epithelial integrity can be performed by assessing morphological features of tight junctions and adherence complexes through imaging techniques like electron microscopy or indirect immunofluorescence. In addition, differential expression of the various components of multiprotein junctional complexes can be analyzed by transcriptomic and proteomic techniques. To characterize the functional state of the paracellular gate quickly, changes in size-selective molecule diffusion and ion-selective permeability can be monitored either by assessing the flux of labeled tracer molecules across the epithelium or by measuring transepithelial electrical resistance (TEER) (2). Here, we use viral- or toxin-mediated disruption of barrier function in ex vivo-differentiated mucociliary lung epithelium to demonstrate advantages and pitfalls of both methods. By measuring consecutively changes in TEER and in the flux of fluorescent hydrophilic dextrans of different molecular weight, the epithelial integrity in response to infection can be determined in a rapid and reliable manner.
2. Materials 2.1. Cell Culture
1. Small Airway Epithelial Cells (SAEC) or Normal Human Bronchial Epithelial (NHBE) cells passages 2–4 (Lonza, Walkersville, MD). 2. Cell culture media: SAGM small airway epithelial cell growth media (SABM) supplemented with SAGM SingleQuots for SAEC, and BEGM bronchial epithelial cell growth media (BEBM) supplemented with BEGM SingleQuots (both Lonza). 3. Differentiation media: SABM supplemented with SAGM SingleQuots (Lonza); Dulbecco’s Modification of Eagle’s Medium high glucose Glutamax (DMEM, Gibco BRL, Bethesda, MD); retinoic acid (Sigma-Aldrich, St. Louis, MO) is dissolved at 10 mM in dimethyl sulfoxide (DMSO) for stock solution; retinoic acid working solution is prepared at 50 mM in DMSO. Both solutions are stored at −20°C. 4. Reagent Pack (Lonza): Trypsin/EDTA, HEPES-buffered saline solution, trypsin neutralizing solution. 5. Transwell® insert (6.5 mm diameter, 0.4 mm pore polyester membrane) for 24-well plate (CC3470, Corning Life Sciences, Lowell, MA).
13 Analysis of Epithelial Barrier Integrity in Polarized Lung Epithelial Cells
197
6. Human placenta collagen type VI (Sigma-Aldrich). A 2 mg/mL solution is prepared by dissolving collagen in 0.25% glacial acetic acid in distilled water, which then is further diluted with distilled water to a final concentration of 0.15 mg/mL, and passed through a sterile filter with 0.22 mm pores. 2.2. Transepithelial Electrical Resistance Measurement
1. Voltmeter EVOM2 with STX2 electrode (World Precision Instruments, Sarasota, FL).
2.3. Reagents for Barrier Disruption
1. Polyinosine-polycytidylic acid (poly(I:C), InvivoGen, San Diego, CA). Dissolve at 1 mg/mL in ultrapure water, store at −80°C.
2. Basal differentiation media without supplements: 50% SABM/ BEBM (Lonza) and 50% DMEM high glucose (Gibco BRL).
2. Influenza A virus H1N1 Texas91 (a generous gift from Dr. Adolfo Garcìa-Sastre, Mount Sinai School of Medicine, New York City, NY), stored at −80°C. 3. Lethal factor (LF), Protective antigen (PA), both from List Biological Laboratories, Campbell, (CA). Dissolve in sterile PBS at 0.1–1 mg/mL, quick freeze and store in 10 mL aliquots at −80°C. 4. Triton X-100 (Sigma-Aldrich). Prepare a 1% (v/v) solution in HBSS, pass through a sterile filter with 0.22 mm pores, and store at room temperature. 2.4. Permeability Assay
1. Tracers and solutions for cellular permeability (see Note 1): Fluorescein isothiocyanate (FITC)-dextran 4 kDa (FD4), FITC-dextran 10 kDa (FD10S), FITC-dextran 70 kDa (FD70), FITC-BSA (see Note 2, A9771, Sigma-Aldrich). Dextrans are dissolved at 10 mg/mL in ultrapure water and passed through a sterile filter with 0.22 mm pores. 10 mg/mL dextran stock solutions are further diluted to dextran assay solutions (5 mg/mL) and dextran standard curve solutions (0.5 mg/mL) in HBSS. Store all solutions in small aliquots (1 mL) protected from light at 4°C (see Note 3). Hanks Buffered Saline Solution (HBSS) with Ca2+ Mg2+ (Gibco BRL); store in aliquots (50 mL) at 4°C for convenient prewarming to 37°C. 2. Fluorescence measurement Multidetection microplate reader for fluorescence and absorbance measurement (used here SynergyMax, Biotek, Winooski, VT, see Note 3), microtiter 96-well solid plate (white polystyrene, Cat no. 136101, Thermo Scientific Nunc, Rochester, NY), 96-well polystyrene microplate (clear, flat, bottom, Greiner bio-one, Cat no. 650001, Frickenhausen, Germany).
198
M. Strengert and U.G. Knaus
2.5. Data Analysis
Graph Prism (GraphPad Software, La Jolla, CA): available for both Windows and MacIntosh, or Microsoft Office Excel (Microsoft Corporation, Redmond, WA).
3. Methods The extent of barrier disruption is dependent on the biological or chemical agent encountered by the epithelial layer and on the duration of this event. In our example, passage of fluorescein isothiocyanate (FITC)-labeled hydrophilic dextran molecules with different molecular weight, applied to the apical side of a virus-infected lung epithelial cell layer, is compared to the treatment with the Toll-like receptor 3 stimulus dsRNA (poly(I:C)) and with Bacillus anthracis lethal toxin (LT: LF + PA). After incubation for a particular time period Relative Fluorescence Units (RFUs) are determined in media sampled from the outer chamber of the Transwell® insert, and are extrapolated against a fluorescence concentration standard curve for each dextran molecule used. Subsequently, Papp and Flux is calculated to evaluate changes in size-dependent permeability of the lung epithelium upon various treatments (Fig. 1). TEER is measured for comparison and to assess more subtle changes in the selective ion permeability of the cell layer. 3.1. Cell Culture
Primary human SAEC or NHBE cells should be used at passages 2–4 for mucociliary differentiation. Cells at early passage can be
Fig. 1. Model of a polarized human lung epithelium. Primary lung epithelial cells are grown in air–liquid culture for several weeks to give rise to multiple cell types, including basal cells, goblet cells, and ciliated cells. TEER and dextran permeability are used here for analyzing the effect of viral or bacterial pathogen exposure on lung barrier integrity. Changes in paracellular diffusion are determined by calculating Flux or the apparent permeability coefficient (Papp) for fluorescent dextrans, where Cdex t represents the total dextran concentration added to the inner chamber (Vi ) and Cdex d the diffused dextran concentration after a specified incubation period in the total volume (Vo) of the outer chamber.
13 Analysis of Epithelial Barrier Integrity in Polarized Lung Epithelial Cells
199
amplified by growing in serum-free basal medium (SABM/ BEBM) supplemented with SAGM/BEGM SingleQuots at 37°C in a humidified 5% CO2 incubator. For the differentiated lung epithelial model (three-dimensional), cells are detached with supplied trypsin/EDTA solution and seeded onto Transwell® inserts coated with human placenta collagen type VI (15 mg/cm2). Lung epithelial cells on inserts are grown in serum-free culture medium (SABM or BEBM, DMEM at 1:1 ratio) supplemented with defined growth factors supplied by SingleQuots and freshly prepared, 50 nM retinoic acid to induce differentiation. After 2–4 days in immersed culture conditions, the media on top of the cells is removed and cell culture is switched to an air–liquid interface system for 2–3 weeks (3). 3.2. TEER Measurement
1. Before measurement, the electrode is rinsed with 70% ethanol and placed under ultraviolet light for at least 10 min to avoid cross-contamination of cultured cells. Afterwards, the electrode is equilibrated for 2–3 min in basal differentiation media without supplements. Before measuring TEER, prewarmed basal differentiation media without supplements is added to the inner chamber (200 mL) and outer chamber (400 mL) to adjust fluid levels in the Transwell® inserts. 2. The cells are equilibrated for 15 min at 37°C before the electrode is placed in the Transwell insert. Extreme care should be taken to avoid damaging the epithelial layer with the electrode. The position of the STX2 electrode needs to be as identical as possible from insert-to-insert to avoid inaccurate measurements due to ankle and depth variance. The resistance reading is repeated for each filter three times (see Note 4). 3. After the measurement, both the inner (300 mL) and outer (600 mL) chamber of the Transwell® insert are washed with prewarmed HBSS before applying the FITC-dextran assay solutions (see Note 5). If applying permeability altering agents, such as Triton X-100 as control, these washes can be done with basal differentiation media without supplements. 4. The Transwell® inserts are now transferred to a beforehand prepared 24-well plate containing prewarmed media (according to the planned treatment) or 450 mL HBSS (see Note 6) for the permeability assay. TEER should be measured before and after the treatment with chemical or biological agents, and can be used as a tool to select inserts with similar resistance for the experiment.
3.3. Application of Permeability Changing Reagents (See Note 7)
After assessing TEER and visual inspection of the cultured lung epithelial cell layers, inserts with corresponding resistance values and similar morphological features are grouped to be either left nontreated, infected with Influenza A virus, or treated with poly(I:C), PA alone or LT (PA + LF). Before and after each apical
200
M. Strengert and U.G. Knaus
treatment, the top of the cell layer is washed three times with 200 mL basal differentiation media without supplements (see Note 8). Unless otherwise stated, reagent stock solutions are diluted in basal differentiation media without supplements. 1. Infection with Influenza A virus H1N1 Tx91 is performed by applying virus at a multiplicity of infection (MOI) of 2–5 in a total volume of 50 mL from the top of the layer. 90 min after infection, the virus is taken off (see Note 9). 2. poly(I:C) is used at a final concentration of 10 mg/mL in a total volume of 50 mL and is applied for 5 h from the top of the cell layer. Before further dilution, heating of the stock solution to 65°C for 10 min, mixing using a 20 gauge needle and bath sonication for 2 min is recommended (see Note 10). 3. LT and PA treatment is applied for 24 h in the outer chamber in a total volume of 600 mL at a final concentration of 1 mg/mL after a 3 h starvation period in basal differentiation media without supplements. 4. A 1% (v/v) Triton X-100 solution (in HBSS) is used directly before performing the assay as positive control to destroy the overall integrity of the cell layer. The solution is applied for 10 min at 37°C from both the top (150 mL) and bottom (450 mL) of the layer (see Note 11). Treated epithelial layers are subjected again to TEER measurement (see Subheading 3.2) and visual inspection to assess treatment-induced changes before application of the different fluorescent tracers occurs for permeability measurement. 3.4. Permeability Assay
1. Preparation of standard curves. In order to determine the final concentration of FITC-labeled dextran molecules that passed the epithelial cell layer into the outer chamber, a standard curve with known concentrations of various dextrans is prepared (Fig. 2). 0.5 mg/mL dextran solutions are further diluted to concentrations ranging from 0 to 16.67 mg/mL in a total volume of 450 mL HBBS. RFU of the standard curve samples is determined in triplicate (100 mL/well in a 96-well plate) with an excitation wavelength of 490 ± 9 nm and an emission wavelength of 520 ± 9 nm with a sensitivity setting of 40% measured in the top part of the plate. The resulting standard curves for the three dextrans with different molecular weight are linear with correlation coefficients R2 FITC-dextran 4 = 0.992, R2 FITC-dextran 10 kDa = 0.994, and R2 FITC-dextran 70 kDa = 0.996 (see Notes 12 and 13). These standard curves are kDa subsequently used to determine the unknown concentration of diffused dextran molecules in the outer chamber. 2. Application of tracers to monitor cellular permeability changes and fluorescence measurement.
13 Analysis of Epithelial Barrier Integrity in Polarized Lung Epithelial Cells
201
Fig. 2. Fluorescence standard curves for FITC-dextran molecules with average molecular weights of 4, 10, and 70 kDa are prepared at fixed concentrations for subsequent extrapolation of unknown FITC-dextran concentrations recovered from the outer chamber after diffusion through the lung epithelial cell layer. Fluorescence is measured in relative fluorescent unit (RFU).
(a) Inserts are moved to a fresh 24-well plate containing 450 mL of prewarmed HBSS in the outer chamber. FITCdextran solutions (5 mg/mL in HBBS) with an average molecular weight of 4, 10, or 70 kDa are added to the apical side of the nontreated or treated epithelial cell layer (inner chamber) that contains prewarmed HBSS to achieve concentrations of 0.5 mg/mL in a total volume of 150 mL (see Notes 14 and 15). (b) After adding the fluorescent tracers, the cell layers are incubated at 37°C in a tissue culture incubator for a fixed period of time (120 min) (see Notes 16 and 17). At the end of the incubation period, the inserts are transferred to a 24-well plate filled with prewarmed differentiation media for visual inspection after the removal of the remaining fluorescent tracer from the inner chamber. (c) In addition to the described treatment groups, the assay is performed for collagen-coated Transwell® inserts without cells to determine the contribution of the coated filter to the calculated apparent permeability coefficient, and with Triton X-100 (see Subheading 3.3). The detergent Triton X-100 destroys epithelial integrity by membrane permeabilization and is used to assess to which extent the tracer molecules bind to remaining cellular components and debris. (d) Three 100 mL aliquots are taken from the outer chamber and are transferred to a 96-well plate for determination of RFU in a precalibrated microplate reader (see Note 16). (e) After completion of the assay, Transwell® inserts can be used further for multiple downstream applications, such as quantitative real-time PCR or immunoblot.
202
M. Strengert and U.G. Knaus
3.5. Data Analysis
1. Permeability coefficient and Flux (see Note 18) The permeability coefficient (P) reflects the ability of any particulate molecule to cross a semipermeable membrane over time. A standardized equation to calculate P is DQ / Dt , where DQ / Dt is the amount of Papp (cm / s ) = A ´ Co substance that passed the cell layer over a fixed period of time, determined by taking samples after the chosen time of incubation (here 120 min) from the outer chamber of the Transwell® insert (growth area A 0.33 cm2) after adding the tracer to the inner chamber at a fixed concentration Co. The RFU of the samples is measured in triplicate for each Transwell® insert of a treatment group (n = 3) in a precalibrated microplate reader. Triplicate relative fluorescence values are averaged and tracer concentration is determined by extrapolation with the fluorescence standard concentration curve obtained for each tracer. Finally, extrapolated concentrations are multiplied with the total volume in the outer chamber to determine the absolute amount of tracer present. ∆Q/∆t is obtained for 0 min and 120 min and inserted into the equation to gain final Papp-t (Ptotal). The permeability coefficient (Papp) of the collagen-coated filter membrane alone is measured in parallel so that the input of the cell layer (Papp-c) can be distinguished from the input due to the membrane (Papp-m), according to the relationship Papp-c = Papp-t – Papp-m, where Ptotal is the total permeability coefficient for the system. The movement of molecules across a membrane can also be characterized by Flux or change in concentration, where Co and Ci represent the concentration of the specific compound in the inner and outer chamber, and Vo and Vi are the absolute volumes in the corresponding chamber (see Note 18). Flux can be calculated as: Flux (% ) = C o ´ V o ´ 100 C i ´ Vi 2. Transepithelial electrical resistance o obtain the accurate resistance of the polarized lung epiT thelium, the resistance of a collagen-coated Transwell insert is measured (blank) and subtracted from the resistance reading across the tissue itself according to the formula Rtrue tissue (Ω) = Rtotal − Rblank. Typically, a collagen-coated Transwell® insert (6.5 mm, 0.4-mm pore polyester membrane) has a resistance R of 200 Ω and a membrane area A of 0.33 cm2. Subsequently, TEER is calculated as Unit Area Resistance: Resistance (Ω) × Effective Membrane Area (cm2). In Fig. 3, after calculating the normalized average TEER for all experimental groups, TEER for each of those groups before treatment is set at 100% and differences due to the treatment are expressed as ∆TEER (%).
13 Analysis of Epithelial Barrier Integrity in Polarized Lung Epithelial Cells
203
Fig. 3. Analysis of lung epithelial barrier integrity. Polarized NHBE cells were either left nontreated (NT), infected with Influenza A virus H1N1 for 1 h (H1N1), treated with poly(I:C) for 5 h (poly(I:C)), treated with protective antigen (PA), or treated with lethal toxin (LT:PA + LF) for 24 h. The epithelial layer was subjected to TEER measurement before and after treatment. Normalized TEER (Ω cm2) before treatment is defined as 100% and changes after 24 h is shown in % of change (∆TEER). TEER measurement is followed by a 2 h incubation with FITC-dextran 4, 10, or 70 kDa, as indicated, applied to the apical side, for permeability analysis. Changes in permeability were calculated as apparent permeability coefficient (Papp-c) of the cell layer (cm/s) and Flux (%). Papp-m of FITC-dextran 4 kDa was chosen as reference for calculation. Data are represented as mean ± SEM for 3 inserts/per condition (n = 2 for H1N1 with FITC-dextran 70 kDa).
4. Notes 1. Alternative molecules for studying permeability changes are radiolabeled substances (tritiated mannitol or BSA (4)), other fluorescent molecules, such as Lucifer Yellow Salt (5), FluoSpheres® (polystyrene microspheres of various sizes coupled
204
M. Strengert and U.G. Knaus
to a fluorescent dye) (6), or proteins detectable with colorimetric assays like horseradish peroxidase (HRP) (7). Transport of spheres is often faster than that of highly branched dextrans. Moreover, attention to the appropriate size of the spheres in relationship to the filter pore size is necessary. A tenfold size difference is recommended. By using short incubation times with FITC-dextrans of varying molecular weight, additional information about the paracellular space can be obtained. 2. Standard concentration curves determined for FITC-BSA are sometimes not linear, which might point to irregularities in protein labeling or the formation of protein aggregates that cannot be broken by sonication. In addition, fluorescence wide field microscopy can reveal some nonspecific binding of FITC-BSA on top of the epithelial layer. 3. A sufficient volume of dextran assay solution should be prepared for comparison in subsequent experiments, as FITC fluorescence is strongly pH-dependent. The actual choice of the fluorophore coupled to a tracer molecule depends upon the available fluorescence microplate reader (monochromator vs. filter-based). In addition, labeled dextrans which differ in net charge are available. 4. Instead of using the chop-stick STX2 electrode, World Precision Instruments offers a STX100 electrode, which permits easier and more accurate measurement due to a modified shape. This leads to easier positioning of the electrode in the insert. In addition, several versions of STX100 electrodes exist, designed to fit into various HTS tissue culture plates from different vendors. 5. These washing steps are essential to remove traces of phenol red-containing differentiation media. Phenol red interferes with subsequent fluorescence measurements due to its similar absorbance spectrum. 6. The media used for the permeability assay and for dilutions of the different dextran stock solutions can be selected according to the specific needs of the cultured cells. It is only necessary to ensure that all of the components are nonfluorescent. 7. If work with an infectious agent is performed, ensure to wear appropriate protection, to work in a BSL Class II hood, to properly dispose of waste and to decontaminate the working area. 8. Never use a vacuum-driven aspirator for removing media from the top of ALI-cultured lung epithelia, as it will severely impair the integrity of the cell layer. Instead, use a pipette with a 20–200 mL tip to gently remove all remaining liquid from the cells.
13 Analysis of Epithelial Barrier Integrity in Polarized Lung Epithelial Cells
205
9. During preparation of Influenza A virus for infection, the virus stock needs to be resuspended vigorously to break up protein aggregates, which will interfere with infection rates. 10. The described treatment of poly(I:C) stock solutions is necessary to break up large polymers that might have formed during freezing. 11. If treatments with accurate timing are required, prepare in advance 24-well plates filled with the washing solutions so that the Transwell® inserts can be rapidly transferred from one plate to another. 12. It is crucial to assess the linear range of fluorescence for the standard curves. A linear slope is required for correctly extrapolating the unknown concentration of dextran molecules in the outer chamber. 13. As shown in Fig. 2, RFU for FITC-dextran molecules with an average molecular weight of 70 kDa are lower than for FITCdextran 10 kDa due to a different extent of labeling on glucose molecules. Therefore, check carefully information available regarding the degree of labeling and avoid purchasing dextrans from different vendors as labeling can differ among different suppliers. It is possible to correct for a varying extent of labeling by measuring absorbance spectra for the fluorescence standard curve samples in order to determine the corresponding OD maxima. Following the law of Beer-Lambert A = ε ´ l ´ c ( ε molar extinction coefficient, c concentration of absorbing molecules, l path length), the FITC concentration present in each sample of the different standard curves can be calculated. In this way, a correction factor can be determined, which is applied to each assay sample to correct for labeling differences. 14. FITC-dextran solutions are light sensitive. 15. Measuring the absorbance of both assay and standard curve samples is helpful as the appearance of a compound spectrum will indicate contamination of solutions during preparation. Transfer 50 mL of the samples into a translucent 96-well plate and read the absorbance spectra in a microplate reader ranging from 280 to 600 nm in 10 nm steps. 16. Wrap the tissue culture plates during incubation and the 96-well plates during transport in aluminum foil to avoid light exposure. 17. The optimal incubation time for the permeability assay depends on the biological or chemical agents used and should be determined in preliminary experiments. Short incubation times favor the paracellular route.
206
M. Strengert and U.G. Knaus
18. Both Flux and Papp are cell type-, treatment-, and time- dependent and can be zero in nontreated controls. TEER and tracer permeability measurement reflect a composite of paracellular and transcellular routes. The paracellular pathway exhibits stronger size selectivity for tracer molecules than transcytosis. Although in most instances, barrier disruption can be equally determined by TEER or flux, certain small molecule inhibitors or drugs (i.e., dexamethasone) alter TEER, but not paracellular transport of molecules. In these cases, permeability measurements reflect more closely barrier integrity. Electron microscopy can be used for visual confirmation of barrier structure. Flux values (%) for multicellular epithelial layers, analyzed after a short time period (here 2 h), can be very small. Flux determination might be more meaningful for endothelial permeability or permeability measurement after 12–24 h.
Acknowledgments The authors would like to thank Luis Alvarez for helpful discussions and graphical assistance, and Mandy Lehmann for the initial establishment of the lung epithelial model and permeability assay system. The work was supported by CDC grant CI000095 and by funding from UCD School of Medicine and Medical Science. References 1. Matter K, and Balda, MS (2003) Signalling to and from tight junctions. Nat Rev Mol Cell Biol 4:225–236 2. Matter K, and Balda, MS (2003) Functional analysis of tight junctions. Methods 30:228–234 3. Gray TE, Guzman K, Davis CW, Abdullah LH, and Nettesheim, P (1996) Mucociliary differentiation of serially passaged normal human tracheobronchial epithelial cells. Am J Respir Cell Mol Biol 14:104–112 4. Hayashi S, Takeuchi K, Suzuki S, Tsunoda T, Tanaka C, and Majima, Y (2006) Effect of thrombin on permeability of human epithelial cell monolayers. Pharmacology 76:46–52
5. Bao S, and Knoell, DL (2006) Zinc modulates cytokine-induced lung epithelial cell barrier permeability. Am J Physiol Lung Cell Mol Physiol 291:L1132–1141 6. Chandra A, Barillas, S Suliman A, and Angle, N (2007) A novel fluorescence-based cellular permeability assay. J Biochem Biophys Methods 70:329–333 7. Carr JM., Hocking H, Bunting K, Wright PJ, Davidson A, Gamble J, Burrell CJ, and Li P (2003) Supernatants from dengue virus type-2 infected macrophages induce permeability changes in endothelial cell monolayers. J Med Virol 69:521–528
Chapter 14 Permeability of Differentiated Human Urothelium In Vitro Peter Rubenwolf and Jennifer Southgate Abstract The urothelium plays a critical role in the bladder as a permeability barrier to urine. Whereas it was once considered a simple physical barrier, it is increasingly evident that urothelium has a regulatory role in maintaining the barrier both through self-repair and by mediating the transport of ions and small molecules across the transcellular and paracellular interfaces. The development of cell culture systems that replicate the morphological and differentiated features of human urothelium provides a versatile in vitro tool for exploring molecular and functional relationships in normal bladder physiology and for examining inherent changes in the urothelia of patients with dysfunctional bladder syndromes. In addition, it provides a useful platform to study the effect of pharmacological treatment on urothelial barrier function. In this review, we describe the development of differentiated urothelial cell constructs from in vitro-propagated normal human urothelial cells, and the application of methods to assess barrier function using transepithelial electrical resistance, water, urea, and dextran transport as objective and quantifiable parameters. Key words: Human urothelium, Urothelial cell culture, Barrier function, Permeability, Transepi thelial electrical resistance, Water and solute transport
1. Introduction The bladder and the associated urinary tract are lined by urothelium, a transitional epithelium with highly specialised features that form a urine-blood barrier. The primary physiological function of the urothelium is to minimise reabsorption of noxious urine components and to maintain the composition and final concentration of the urine as excreted by the kidneys. Hence, a tight urothelial barrier is critical to normal bladder function and metabolic homeostasis. The barrier function of the urothelium is dependent on features expressed at the molecular level and acquired during the process of cellular differentiation. This structural–functional relationship is exemplified in the expression of uroplakin proteins in
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_14, © Springer Science+Business Media, LLC 2011
207
208
P. Rubenwolf and J. Southgate
the apical membrane (1–3) and by the pattern of claudin isoform expression in the intercellular tight junctions (4), which contribute respectively, to transcellular and paracellular urinary barrier properties. Most contemporary data on urothelial function has been obtained using non-human tissue sources, primarily due to the lack of access to normal human urothelial tissue, which has hampered physiological investigation. Although methods to grow normal urothelial cells in culture have been available for some years (5–8), there has long been a failure to demonstrate that human urothelial cells grown in vitro can be induced to differentiate to form a functional differentiated urothelium. In 2005, Cross and colleagues reported a novel method to generate a differentiated “biomimetic” urothelium from normal human urothelial (NHU) cells propagated in vitro in a low calcium serum-free medium (9, 10). The development of differentiated constructs from NHU cell cultures involved supplementation of the medium with bovine serum for 4 days prior to cell harvest and reseeding onto permeable membranes in modified medium. In their study, transepithelial electrical resistance (TER) and the diffusive permeability of the biomimetic urothelium to water, urea, and dextran were investigated using a modified Ussing chamber system to provide objective and quantifiable measures of barrier function. The human urothelial tissue equivalents revealed differentiation-associated protein expression, including polarised apical membrane-restricted amiloride-sensitive sodium ion channels and basal expression of Na+K+ATPase. Barrier function, assessed by TER, was found to be comparable to that reported for intact urothelia from other mammalian species (9, 10). The approach has since been reproduced using abattoirderived porcine urothelial cells, which revealed subtle, but important species differences (11). The biomimetic urothelium represents a versatile tool to explore molecular and functional relationships in normal urothelial physiology (9, 10). The value of this tool in delivering new knowledge has been demonstrated: although it is generally assumed that human urothelium is impermeable to water and to the constituents of the urine, recent evidence derived from the biomimetic culture system indicates that human urothelium expresses a network of water- and urea-transporting channels that may constitute a molecular basis for transurothelial water and solute transfer (12). This finding challenges the traditional concept of the impermeable urothelial barrier, as it suggests that urothelium may be able to mediate water and solute transport and thus, modify the composition and final concentration of the urine. This raises the possibility that such transfer mechanisms may be involved in poorly understood urinary incontinence and related benign bladder dysfunctions and further, provides a means
14 Permeability of Differentiated Human Urothelium In Vitro
209
to identify new pharmacological targets for modulating urinary barrier function. Thus, the development of appropriate in vitro tissue models, alongside tools for objective assessment of urothelial barrier function, is of major interest to both scientists and clinicians. In addition to in vitro applications, the ability to generate differentiated urothelial tissues from normal urothelial cells propagated in vitro raises the possibility of using such cells in autologous bladder tissue reconstructions. In this case, the development of methods to monitor barrier function objectively may be important for optimising tissue engineering and transplantation strategies (11, 13). The purpose of this chapter is to provide a detailed description of the procedures required to establish a functional or biomimetic urothelium from cultured urothelial cells and the methodology used to assess barrier properties, by measuring TER and permeability to water, urea, and dextran. The development and assessment of a functional urothelial barrier are described in conjunction with potential problems that may be encountered, as well as a discussion of the deficiencies and limitations of interpreting the results.
2. Materials 2.1. Cell Culture Reagents
1. Growth medium for NHU cells. Keratinocyte serum-free medium (KSFM) containing recombinant epidermal growth factor (5 ng/ml) and bovine pituitary extract (50 mg/ml), as supplied by Gibco, and additionally supplemented with cholera toxin (30 ng/ml; Sigma). KSFM supplemented with the above factors is referred to as “KSFM complete” (KSFMc). Store at 4 to 7°C for up to 2 weeks. 2. Differentiation medium A. KSFMc supplemented with 5% bovine serum (BS). The BS may be fetal or adult, but should be pre-selected by batch testing and kept at −80°C for longterm storage. Use 3.5 ml per Snapwell™ construct. 3. Differentiation medium B. KSFMc supplemented with 5% BS and with a final calcium concentration of 2 mM through supplementation with sterile 1 M calcium chloride (CaCl2) solution. This should take account of the endogenous calcium concentration in KSFMc (0.09 mM) and in the serum, which will be batch-dependent.
2.2. Cell Harvesting Agents
1. 0.1% (w/v) EDTA in Dulbecco’s Phosphate-Buffered Saline (dPBS). Sterilise by autoclaving and keep at ambient temperature. Use 10 ml per 75-cm2 flask.
210
P. Rubenwolf and J. Southgate
2. Trypsin Versene (TV). Add 20 ml Trypsin (10× sterile solution, Sigma) to 4 ml 1% EDTA (sterilised by autoclaving) and 176 ml HBSS (without Ca2+ and Mg2+). Store at −20°C in 5 ml sterile aliquots, which once thawed, may be kept at 4°C for up to 2 weeks. Use 1 ml per 75 cm2 flask. 3. Trypsin inhibitor (necessary where there is no serum in the growth medium to inhibit trypsin). Dissolve 100 mg Trypsin Inhibitor (Sigma) in 5 ml dPBS. Filter-sterilise (0.2 mm) and store as 100 ml aliquots in 5 ml plastic bijoux at −20°C. One aliquot inhibits 1 ml of trypsin. For use, add 5 ml of KSFMc per bijou and use to collect trypsinised cells from a 75 cm2 flask. 4. Cultureware. 75 cm2 Primaria® tissue culture flasks (Becton Dickinson) for routine NHU cell culture. Snapwell™ permeable supports with a 0.4 mm polyester membrane in a nominal 12 mm insert, supplied with a six-well tissue culture plate (Corning, supplied by Costar, High Wycombe, UK). 2.3. Transepithelial Electrical Resistance Measurements
1. Electronic Volt-Ohmmeter (EVOM, World Precision Instruments DC1100), Ag–AgCl microelectrodes connected to a VCC MC2 voltage–current clamp (Physiologic Instruments) and vertical modified Ussing chambers U-2500 (Havard Apparatus) designed to accept the Snapwell™ permeable supports. 2. Krebs solution. Stock solution A. For 100 ml: add 6.896 g NaCl, 0.353 g KCl and 0.291 g KH2PO4 to polished H2O (dH2O). Stock solution B. For 100 ml: add 2.092 g NaHCO3 to dH2O. Stock solution C. For 100 ml: add 0.187 g CaCl2∙2H2O to dH2O. To make 100 ml of Krebs solution: warm all stocks to 37°C, add 10 ml each of stocks A–C and 0.1 g d-glucose and make up to volume with pre-warmed dH2O. Prior to use, equilibrate Krebs solution with humidified 95% O2:5% CO2 and maintain at a constant temperature of 37°C in a water bath. 3. Portable EVOM™ Epithelial Volt-Ohmmeter Precision Instruments) (alternative to 1).
2.4. Assessment of Dextran Permeability
(World
1. Dextran of molecular weight 4,400 and 9,500 conjugated to fluorescein isothiocyanate (FITC-Dextran; Sigma). For use, dilute tracer to 1 mg/ml in appropriate growth medium and warm to 37°C. Use 500 ml per Snapwell™ membrane. 2. Spectrofluorometer (e.g. Fluoromax-2, Jobin-Yvon).
2.5. Assessment of Water and Urea Permeability
1. [3H]-labelled medium: 2.5 ml of [3H] water (200 mCi/ml, Sigma) added to 497.5 ml medium (KSFMc) per Snapwell™ membrane giving final activity of 1 mCi/ml; heat to and maintain at 37°C. Caution: radiation hazard – follow local safety rules for handling and disposal.
14 Permeability of Differentiated Human Urothelium In Vitro
211
2. 14[C]urea-labelled medium: 2.5 ml of [14C]urea (200 mCi/ml, Sigma) added to 497.5 ml medium (KSFMc) per Snapwell™ membrane giving a final activity of 1 mCi/ml; heat to and maintain at 37°C. Caution: radiation hazard! 3. Scintillation. Liquid scintillation counter (Packard Tri-carb 2700TR) to determine the radioactivity of samples. Depending on the number of measurements to be performed, prepare 5 ml scintillation vials (Packard Biosciences) containing 4 ml Ultima Gold™ XR scintillation fluid (Packard Biosciences) and label appropriately for each sample (time point/apical/basal chamber of the Snapwell™, e.g. “15 min apical”). Caution: safety hazard from scintillant – follow local regulations for control of substances hazardous to health (COSHH).
3. Methods 3.1. Establishment of Human Urothelial Cell Cultures
We have previously developed a cell culture system to isolate and propagate normal human urothelial (NHU) cells in vitro (8). These methods are described in detail elsewhere (14). When grown in a serum-free, low-calcium medium as described, NHU cells acquire a proliferative, regenerative phenotype, but do not express markers of urothelial cytodifferentiation (15). This proliferative phenotype is considered to replicate the wound-healing phenotype acquired by regenerating urothelium in vivo and has been shown to be driven by an autocrine loop regulated through the epidermal growth factor receptor (16). 1. Finite normal human urothelial (NHU) cell lines established as detailed elsewhere (8, 14) are propagated in Primaria™coated 75-cm2 tissue culture flasks in KSFMc and subcultured at just-confluence. 2. To harvest cells for subculture, carefully aspirate medium from flask; add 10 ml of sterile 0.1% (w/v) EDTA in PBS to a 75-cm2 flask. Place the flasks in an incubator at 37°C until the cells appear rounded (5–10 min). Aspirate the EDTA, add 1 ml TV and rock the flask gently to ensure that the bottom of the flask is covered. Incubate at 37°C for 1–2 min or until cells detach completely when the side of the flask is tapped sharply. Add 5 ml KSFMc containing one aliquot of trypsin inhibitor. Use a pipette to ensure complete dispersal of cells into suspension and transfer into a Universal tube for centrifugation at 300 × g for 4 min. Carefully aspirate the supernatant avoiding contact with the pellet. Tap the pellet to aid cell suspension and resuspend in 15 ml of complete KSFM.
212
P. Rubenwolf and J. Southgate
Reseed cells at a split ratio of about 1:3. Check the flasks after 24 h to ensure the cells are healthy and there is no contamination. Replace the medium every 2–3 days. 3.2. Establishment of Differentiated Urothelial Constructs (“Biomimetic” Urothelium)
Cross and colleagues demonstrated that NHU cells propagated as finite cell lines in vitro could be reproducibly induced to form stratified urothelia consisting of basal, intermediate and superficial cells, with differential expression of cytokeratins and superficial tight junctions. Functionally, the neotissues showed characteristics of native urothelium, including high transepithelial electrical resistance of >2,500 Ω∙cm2, apical membrane-restricted amiloridesensitive sodium ion channels, basal expression of Na+K+ATPase, and low diffusive permeability to urea, water, and dextran (9, 10). 1. NHU cells are harvested as detailed in Subheading 3.1, resuspended in 15 ml KSFMc containing 5% BS and preconditioned by growth for 4 days in differentiation medium A (see Note 1). 2. Cells are detached from the flask with EDTA and TV as above (Subheading 3.1, step 2), counted, pelleted by centrifugation at 300 × g for 4 min, and resuspended in differentiation medium A at 1 × 106 cells per ml. Seed 0.5 ml cell suspension containing 5 × 105 cells into the apical compartment of a Snapwell™ supported in the supplied six-well culture plate. Add 4 ml of the same medium to the basal compartment (Fig. 1) (see Note 2). 3. After 24 h, replace the medium in both apical and basal compartments with differentiation medium B, to form stratified urothelial tissue constructs (see Note 3). Cultures are maintained in differentiation medium B for 7 days, with replenishment of medium on alternate days. Cultures should be inspected on an inverted microscope with phase-contrast optics (×10 objective).
Upper compartment Urothelial cell culture Culture medium Lower compartment Fig. 1. For functional studies, urothelial cell cultures are grown on Snapwell™ chambers incorporating a permeable membrane and providing access to both sides of the cell layer. These chambers provide a versatile means to study transport or other metabolic activities of cultured tissues in vitro.
14 Permeability of Differentiated Human Urothelium In Vitro
213
Fig. 2. Differentiated urothelial constructs grown on Snapwell™ membranes. Top: Haematoxylin and eosin-stained section of a paraffin-wax-embedded differentiated urothelial construct showing multilayered urothelium consisting of basal, intermediate and superficial cells. Bottom: Immunohistochemistry with an anti-AUM antibody (arrows) showed expression of differentiation-specific uroplakin protein limited to the apical membrane of superficial cells.
By 7 days, the urothelial tissue constructs should exhibit morphological, molecular, and functional properties of native urothelium, including expression of differentiation-associated urothelial markers (Fig. 2; (10, 12)) and are ready to use for TER assessment and permeability studies (see Note 4). 3.3. Assessment of TER
TER is a measure of ionic permeability of epithelial tissues and is calculated using Ohm’s law from the spontaneous transepithelial voltage and short circuit current readings. Fromter and Diamond divided epithelia into two categories: “leaky” and “tight” (17). This division is based upon both the magnitude of the TER and the relative resistance of the transcellular to paracellular pathways. Leaky epithelia are defined as having a TER < 500 W cm2, whereas
214
P. Rubenwolf and J. Southgate
Table 1 TER of urothelial constructs established on permeable Snapwell™ membranes in response to various culture conditions (reproduced from ref. 10) Cell culture Culture condition
KSFMc KSFMc KSFMc + 5% BS, KSFMc + 5% BS, 0.09 mM Ca2+ 2.0 mM Ca2+ 0.2 mM Ca2+ 2.0 mM Ca2+
TER (W cm2)
18.5 ± 2.4
49.4 ± 8.9
2,509 ± 172
3,023.4 ± 546
Number of cell lines tested
7
8
18
7
tight epithelia display resistances >500 W cm2 (17). According to their criterion, the urothelium is a tight epithelium with a very low ionic permeability. In fact, consistent with its barrier properties, the TER of the urothelium is one the highest recorded for any tissue (18, 19). Cross and associates demonstrated that the TER of human urothelial tissue constructs was significantly affected by the medium in which the NHU cells had been propagated (10) (Table 1). The barrier function of cultured urothelium may be assessed by TER measurements. Electrophysiological measurements of cultured cell sheets are performed in U-2500 vertical modified Ussing chambers and recorded using Ag–AgCl microelectrodes connected to a VCC MC2 voltage–current clamp. The apparatus also includes a DM_MC6 input module and dummy membrane to differentiate between electronic or experimental artefacts should they arise during an experiment. 1. All experiments are performed in modified Krebs solution prewarmed to 37°C and equilibrated with 5% CO2 in 95% O2. 2. The Ussing chamber is pre-warmed to 37°C using a thermocirculator (Harvard Apparatus). The instrument is calibrated before each experiment to compensate for both electrode asymmetry potentials and fluid resistance in the Ussing chamber. These compensation steps are performed with modified Krebs solution and no membrane in the chamber, according to the manufacturer’s protocol. Following calibration, an empty Snapwell™ membrane does not create a measurable resistance. 3. Cell sheets cultured on Snapwell™ membranes are placed in vertical Ussing chambers with voltage and current recording microelectrodes placed in both the apical and basal chambers and connected to the voltage–current clamp (see Note 5). As soon as in position, 5 ml of modified Krebs solution is added to apical and basal hemi-chambers. 4. The spontaneous transepithelial voltage potential (DV) across the membrane is recorded after a 5-min period of electrode
14 Permeability of Differentiated Human Urothelium In Vitro
215
stabilisation. The short circuit current (DI) is recorded as the current needed to create zero voltage potential across the membrane. From V and I readings, the TER of the cell sheet is calculated using Ohm’s law (V = IR) and expressed in W cm2, corrected for the surface area of the membrane, which is calculated to be 1.13 cm2 (see Note 6). 3.4. TER Monitoring Using a Portable EVOM (See Notes 7 and 8)
1. All TER measurements are carried out in an aseptic environment of a class 2 laminar flow cabinet. 2. Before each use, the electrodes are sterilised by immersion in 70% ethanol for 3 min, before air-drying and then rinsing in medium to ensure that no ethanol is left before use. 3. Self-tests and calibration of the instrument is performed according to the manufacturer’s instructions prior to each measurement. 4. The electrical resistance exerted by the Snapwell™ membrane and the culture medium is assessed for two blank (unseeded) Snapwell™ membranes. 5. The TER of cultured urothelium is measured by placing the longer external electrode into the basal compartment of the Snapwell™ insert whilst positioning the shorter inner electrode into the apical compartment without touching the tissue culture insert. 6. During measurement, the position of electrodes is kept constant until a steady reading is obtained. 7. The measured TER is corrected by subtracting the mean resistance of a blank Snapwell™ membrane.
3.5. Assessment of the Dextran Permeability of Cultured Urothelium
The permeability of cultured epithelial tissue to hydrophilic tracers can be measured by using fluorescently labelled (e.g. FITC-dextran) or radio-labelled (e.g. tritiated water or 14C urea) compounds. The flux of different sized tracers from apical to basal culture chambers through the urothelial cell sheet may be calculated as a correlate of barrier function. Cross and associates assessed the permeability of in vitro propagated urothelium to dextran using FITC-labelled dextran of molecular weights 4,400 and 9,500. Urothelial barrier function was determined by measuring the movement of dextran from the apical to the basal aspect of differentiated urothelial cell cultures. Compared to NHU cell cultures propagated in KSFMc, differentiated urothelial cultures had a significantly lower permeability to both molecular weights of dextran (10) (Fig. 3). Below, we describe a method for the quick and reliable analysis of size-selective paracellular tracer diffusion using fluorescent dextran of different molecular weights. The assay is robust and allows for the analysis of many samples in parallel.
216
P. Rubenwolf and J. Southgate
Fig. 3. Permeability of urothelial cell cultures to dextran.
1. Permeability assays are performed on urothelial cells propagated on Snapwell™ membranes (Subheading 3.2) using 4,400 and 9,500 molecular weight dextrans conjugated to fluorescein isothiocyanate (FITC). At the start, the medium in the apical compartment of the Snapwell™ membrane is replaced with 500 ml of the corresponding medium containing one of the tracers at 1 mg/ml. The basal compartment is replaced with 1,000 ml of tracer-free medium. 2. The urothelial tissue constructs are returned to the incubator for 3 h, then triplicate 350 ml samples are taken from the basal compartment. 3. Fluorescence is measured at an emission wavelength of 514 nm following excitation at 490 nm using a Fluoromax-2 spectrofluorometer. A standard curve of known tracer concentration (300 mg to 0.14 mg/ml in a threefold serial dilution) is plotted, and the concentration of FITC-dextran in the medium retrieved from the basal chamber is calculated following extrapolation from the standard curve. Results are expressed as mean concentration of dextran/cm2/h of triplicate cultures for each cell treatment. 3.6. Assessment of Water and Urea Permeability of Cultured Urothelium
It is generally assumed that the urothelium is almost impermeable to water and constituents of urine; however, numerous in vivo and in vitro permeability studies using radioisotopes have indicated a very low, but finite permeability to most substances present in the urine and blood (20). Negrete and colleagues assessed the diffusive water and urea permeabilities of dissected rabbit urothelium in vitro in an Ussing chamber. The measured
14 Permeability of Differentiated Human Urothelium In Vitro
217
water and urea permeability of dissected urothelium was 4.12 ± 0.29 × 10−5 cm/s and 4.3 ± 0.65 × 10−6 cm/s, respectively. The authors provided compelling evidence that the apical membrane of the urothelium represents the main bladder permeability barrier (18). Cross and co-workers assessed the water and urea permeabilities of NHU cell cultures grown on permeable Snapwell™ supports. The urothelial constructs revealed mean diffusive water and urea permeability coefficients (PD) ranging from 2.4 to 8.1 × 10−4 cm/s for water and from 0.3 to 1.5 × 10−4 cm/s for urea, depending upon the culture conditions (10). Even though it has been demonstrated that water molecules can traverse epithelia along various pathways, few studies have focused on potential water transport mechanisms. One mode by which water and urea could potentially traverse urothelium is via transmembrane pore-forming proteins, termed aquaporins (12, 21). The diffusive urea and water permeability coefficients of cultured urothelium are determined by measuring radio-isotopic fluxes from the apical compartment across the tissue to the basal compartment. All measurements are carried out in an aseptic environment using six-well plates containing the permeable Snapwell™ supports. 1. All radioactively labelled media are freshly prepared prior to the experiment and maintained at 37°C. 2. 2.5 ml of each tracer (at 200 mCi/ml) made up to 500 ml in medium (KSFMc) is required for one Snapwell™, i.e. for a six-well plate, 15 ml of [3H]water (200 mCi/ml) and/or 15 ml of [14C]urea (200 mCi/ml) are added to 2,985 ml medium. If simultaneous assessment of water and urea permeability is to be performed, 15 ml of each tracer is added to 2,970 ml of medium and mixed by careful vortexing for 3 s. 3. Remove the six-well plate from the incubator and carefully aspirate the medium from both compartments of each well. Add 3,000 ml of (unlabelled) medium to the basal chamber and add 500 ml of radioactively labelled medium to the apical compartment of the Snapwell™, avoiding spillage across the chambers. 4. During the next 60 min, 25 ml aliquots are taken from both the apical and basal hemi-chambers at 15 min intervals and placed into 5 ml scintillation vials containing 4 ml Ultima Gold™ XR scintillation fluid; (see Notes 9 and 10), mix thoroughly and wait for 5 min before counting, to allow dispersal of air bubbles. The number of counts of the individual isotopes within the samples is determined using a Packard Tri-carb 2700TR liquid scintillation counter (see Note 11).
218
P. Rubenwolf and J. Southgate
5. Disintegrations per minute (DPM) are used to calculate (a) the concentration gradient (DC) and the flux (F) of the isotope tracer across the urothelium and Snapwell™ membrane and (b) the diffusive water and urea permeability coefficients (PD) for each sampling time point (see Note 12). The measured diffusive water and urea permeabilities (PD) are calculated for each sampling time point using the formula.
PD =
F (unit : cm/s) [A ]´ [DC ]
F: flux of the isotope tracer across the urothelium and Snapwell™ membrane per sec and cm2 (unit: mmol/s cm2) A: area of the Snapwell™ membrane (1.13 cm2) DC: concentration gradient of the isotope across the urothelium and membrane (unit: mmol/cm3 = mmol/ml) 14 [ C]urea and [3H]water fluxes (F) are calculated using the formula:
Tracer flux =
Increase of tracer in basal hemichamber Time between samples (s) ´ area(cm 2 )
The increase in concentration (DC) of [14C]urea and [3H] water in the basal hemi-chamber is calculated from the sample DPM (where 2.22 × 106 DPM = 1 mCi) and the specific activity of the individual isotopes at the different time points. For example: where specific activity for [14C]Urea is 7.8 mCi/mmol and for [3H]Water is 0.022 mCi/mmol: ●●
●●
Concentration of [3H]water (mmol/ml) = [3H] DPM/ (2.22 × 106 × 0.022). Concentration of (2.22 × 106 × 7.8).
[14C]urea
(mmol/ml) = [14C]
DPM/
The water and urea diffusive permeabilities (PD) are calculated for each sampling time point using the isotope flux and the concentration of the corresponding tracer. The measured diffusive permeabilities for each individual urothelial cell culture are corrected for the presence of the Snapwell™ membrane:
1 / PD (Urothelium) = 1 / PD(Urothelium + Snapwell) - 1 / PD(Snapwell) ●●
●●
PD(Urothelium + Snapwell): measured diffusive permeability of urothelial cell culture including Snapwell™ membrane. PD(Snapwell membrane): measured diffusive permeability of only the Snapwell™ membrane.
The contribution of the filter and unstirred layers to the measured diffusive permeabilities is calculated by performing the experiment using blank Snapwell™ membranes that have been incubated with medium and tracer in the absence of cells (see Note 13).
14 Permeability of Differentiated Human Urothelium In Vitro
219
4. Notes 1. For most consistent results, use cells at or before passage 3. 2. After seeding of cells onto the Snapwell™ membranes, a suspension of non-adherent cells forms, which may be mistaken as infection. Non-adherent cells are removed at the first medium change. 3. Urothelial tissues grown on Snapwell™ membranes are prone to physical damage during medium change; particular caution is, therefore, required when aspirating medium from the apical chamber 4. These methods may be applied to urothelia from other large animal species, but may require adaptation before use (e.g. (11) for differentiated porcine urothelium). However, we have not found these culture methods suitable for rodent urothelial cell cultures. 5. To limit damage to the urothelial cell cultures during transfer, the basal hemi-chamber of the Ussing apparatus is initially positioned horizontally and the Snapwell™ is transported from the six-well culture tray to the chamber with the apical medium in situ. 6. As an alternative to performing TER measurements in a modified Ussing chamber, measurements may be taken directly from Snapwell™ chambers maintained in six-well culture plates by using a simple voltmeter with chopstick electrodes (EVOM, World Precision Instruments). This apparatus is designed specifically to perform non-destructive routine TER measurements in cell/tissue culture research and consists of two pairs of electrodes. The outer pair of electrodes (Ag/AgCl) is used to supply an alternating current at 12.5 Hz through the cell-seeded permeable membrane, whilst the inner electrodes (Ag/AgCl) are used to measure the voltage gradient across the membrane. The voltage reading is then converted to resistance using Ohm’s law and the instrument displays this resistance value (R) in ohms. 7. The Ussing chamber may be a more reliable method of measuring TER due to the fact that the electrode placement is uniform between measurements of different cell sheets. However, the advantage of chopstick electrodes is that TER can be determined whilst maintaining sterility, and thus, repeated measurements can be performed on the same urothelial tissue construct over a period of days; this may be particularly valuable if assessing the effects of treatment or damage on barrier properties. 8. Irrespective of the method used, the results are comparable, varying within a range of 5–10% comparing both techniques
220
P. Rubenwolf and J. Southgate
(unpublished data). However, the first method using Ussing chambers is by far more time-consuming given assembly and calibration of the apparatus and preparation of Krebs solution; furthermore, the urothelial sheets are almost invariably damaged and sterility compromised in the Ussing chamber; therefore, the tissue cannot be used for repeated or longitudinal measurements. We, therefore, prefer TER assessment using the portable apparatus, although it requires great caution not to touch the cell sheet with the inner electrode. Once this technical challenge has been overcome, the portable EVOM™ Volt-Ohmmeter enables easy-to-perform and reliable TER measurements. The striking advantage over the measurement in the Ussing chamber is that the TER can be monitored during urothelial cytodifferentiation, prior to and post exposure of the urothelium to various culture conditions, including pharmacological treatment. Table 2 provides a comparison of the two methods used to assess the TER of cultured urothelium.
Table 2 Comparison of the two methods to assess TER of urothelial tissue constructs Ussing chamber
Chopstick electrodes
• Measurement performed directly on tissue • Performed in vertical modified Ussing culture plates using a simple voltmeter and chambers using Ag–AgCl microelectrodes chopstick electrodes (EVOM, portable connected to a VCC MC2 voltage–current apparatus) clamp (not portable) • Cannot be performed directly on tissue culture • Enables measurement directly in tissue culture cabinet plates “Destructive” method • Urothelial sheets are damaged and sterility compromised in the Ussing chamber; tissue cannot be used for further measurements • Monitoring of TER of the same cell sheet not feasible
“Non-destructive” method • TER can be determined whilst maintaining sterility, and thus, repeated measurements can be performed on the same urothelial tissue construct • TER can be monitored during urothelial cytodifferentiation, prior to and post exposure of the urothelium to various culture conditions, including pharmacological treatment
• More time-consuming given assembly and calibration of the apparatus and preparation of Krebs solution
• Quick and easy to perform
• More reliable method of measuring TER as electrode placement is uniform between measurements of different cell sheets • Differentiates between electronic or experimental set-up problems during an experiment • Directly reads TER of cell sheet only
• Requires care not to touch and damage the cell sheet with the outer electrode • Provides less accurate results than measurement in Ussing chamber • Requires subtraction of TER exerted by blank Snapwell™ membrane
14 Permeability of Differentiated Human Urothelium In Vitro
221
9. Taking aliquots from the apical and the basal chambers requires particular caution as the urothelium is easily damaged by the tip; neither the bottom of the insert nor that of the basal chamber must be touched. The Snapwell™ insert is easily displaced whilst taking aliquots; this may falsify radioisotopic fluxes. 10. To avoid confusion when taking samples, ensure all vials are appropriately labelled at the start of the experiment (apical/ basal aliquot/time point)! 11. Under ideal conditions, the permeability coefficient (PD) of a given urothelial tissue produces a constant; hence, taking multiple (e.g. three 25 ml) samples at one time point (e.g. at 15 or 45 or 60 min) may be sufficient to calculate the water and urea permeability of the tissue, with the proviso that the permeability of the tissue is not being altered during the experiment. In this case, use the mean DPM for subsequent calculations of tracer flux and PD. 12. Measurements of permeability over 1 h are prone to artefacts that produce inconsistent results and should be avoided unless essential – e.g. if the aim is to study the impact of a given treatment on the permeability of the tissue over time. 13. It should be noted that where experiments are performed without stirring the medium in the apical/basal chamber of the Snapwell™ support, it will not take into account of the complex role of the unstirred layer in the assessment of transurothelial water and urea fluxes. This may affect the interpretation of permeability (and any in vitro/in vivo extrapolation).
Acknowledgements We are grateful to Mr. William Cross, PhD, Consultant Urologist and Mr Nicholas Smith, PhD, Specialist Registrar in Urology at St. James’s University Hospital, Leeds, for their input into the TER and dextran permeability sections. Lisa Clements and Ros Unwin are thanked for careful reading of the manuscript. References 1. Hu, P., Meyers, S., Liang, F. X., Deng, F. M., Kachar, B., Zeidel, M. L., and Sun, T. T. (2002) Role of membrane proteins in permeability barrier function: uroplakin ablation elevates urothelial permeability. Am J Physiol Renal Physiol 283, F1200–7.
2. Olsburgh, J., Harnden, P., Weeks, R., Smith, B., Joyce, A., Hall, G., Poulsom, R., Selby, P., and Southgate, J. (2003) Uroplakin gene expression in normal human tissues and locally advanced bladder cancer. J Pathol 199, 41–9.
222
P. Rubenwolf and J. Southgate
3. Wu, X. R., Manabe, M., Yu, J., and Sun, T. T. (1990) Large scale purification and immunolocalization of bovine uroplakins I, II, and III. Molecular markers of urothelial differentiation. J Biol Chem 265, 19170–9. 4. Varley, C. L., Garthwaite, M. A., Cross, W., Hinley, J., Trejdosiewicz, L. K., and Southgate, J. (2006) PPARgamma-regulated tight junction development during human urothelial cytodifferentiation. J Cell Physiol 208, 407–17. 5. Reznikoff, C. A., Johnson, M. D., Norback, D. H., and Bryan, G. T. (1983) Growth and characterization of normal human urothelium in vitro. In Vitro 19, 326–43. 6. Liebert, M., Wedemeyer, G., Chang, J. H., Stein, J. A., McKeever, P. E., Carey, T. E., Flint, A., Steplewski, Z., Buchsbaum, D. J., Wahl, R. L., and et al. (1990) Comparison of antigen expression on normal urothelial cells in tissue section and tissue culture. J Urol 144, 1288–92. 7. Hutton, K. A., Trejdosiewicz, L. K., Thomas, D. F., and Southgate, J. (1993) Urothelial tissue culture for bladder reconstruction: an experimental study. J Urol 150, 721–5. 8. Southgate, J., Hutton, K. A., Thomas, D. F., and Trejdosiewicz, L. K. (1994) Normal human urothelial cells in vitro: proliferation and induction of stratification. Lab Invest 71, 583–94. 9. Cross, W.R., and Southgate, J. (2004) Biomi metic Urothelium. Patent WO/2004/011630. 10. Cross, W. R., Eardley, I., Leese, H. J., and Southgate, J. (2005) A biomimetic tissue from cultured normal human urothelial cells: analysis of physiological function. Am J Physiol Renal Physiol 289, F459–68. 11. Turner, A. M., Subramaniam, R., Thomas, D. F., and Southgate, J. (2008) Generation of a functional, differentiated porcine urothelial tissue in vitro. Eur Urol 54, 1423–32. 12. Rubenwolf PC, Georgopoulos NT, Clements LA, Feather S, Holland P, Thomas DF, Southgate J. (2009) Expression and localisation
of aquaporin water channels in human urothelium in situ and in vitro. Eur Urol. 56, 1013–1024. 13. Fraser, M., Thomas, D. F., Pitt, E., Harnden, P., Trejdosiewicz, L. K., and Southgate, J. (2004) A surgical model of composite cystoplasty with cultured urothelial cells: a controlled study of gross outcome and urothelial phenotype. BJU Int 93, 609–16. 14. Southgate, J., Masters, J. R., and Trejdosiewicz, L. K. (2002) in “Culture of Epithelial Cells.” (Freshney, R. I., and Freshney, M. G., Eds.), pp. 381–400, J Wiley and Sons, Inc., New York. 15. Lobban, E. D., Smith, B. A., Hall, G. D., Harnden, P., Roberts, P., Selby, P. J., Trejdosiewicz, L. K., and Southgate, J. (1998) Uroplakin gene expression by normal and neoplastic human urothelium. Am J Path 153, 1957–67. 16. Varley, C., Hill, G., Pellegrin, S., Shaw, N. J., Selby, P. J., Trejdosiewicz, L. K., and Southgate, J. (2005) Autocrine regulation of human urothelial cell proliferation and migration during regenerative responses in vitro. Exp Cell Res 306, 216–29. 17. Fromter, E., and Diamond, J. (1972) Route of passive ion permeation in epithelia. Nat New Biol 235, 9–13. 18. Negrete, H. O., Lavelle, J. P., Berg, J., Lewis, S. A., and Zeidel, M. L. (1996) Permeability properties of the intact mammalian bladder epithelium. Am J Physiol 271, F886–94. 19. Lewis, S. A., and Diamond, J. M. (1976) Na+ transport by rabbit urinary bladder, a tight epithelium. J Membr Biol 28, 1–40. 20. Hicks, R. M. (1975) The mammalian urinary bladder: an accommodating organ. Biol Rev Camb Philos Soc 50, 215–46. 21. Spector, D. A., Wade, J. B., Dillow, R., Steplock, D. A., and Weinman, E. J. (2002) Expression, localization, and regulation of aquaporin-1 to -3 in rat urothelia. Am J Physiol Renal Physiol 282, F1034–42.
Chapter 15 Phenotyping the Claudin 11 Deficiency in Testis: From Histology to Immunohistochemistry Séverine Mazaud-Guittot, Alexander Gow, and Brigitte Le Magueresse-Battistoni Abstract The testis is a heterogeneous organ that comprises a number of cell types, including germ cells at different stages in their maturation, differentiated neighbor nursing cells, and endocrine somatic cells. Despite such cellular heterogeneity the testis is highly organized, with germ cell development and differentiation being compartmentalized into the interconnected tubular network of the seminiferous epithelium. Intratesticular scaffolds rely heavily on the basement membrane of the seminiferous tubules while germ cell development inside the seminiferous epithelium is critically dependent on the Blood Testis Barrier (BTB). The BTB is a macromolecular tight junction complex generated by somatic Sertoli cells within the seminiferous epithelium. The BTB divides the seminiferous epithelium into two compartments: the basal compartment, which delineates a niche for the proliferation and renewal of spermatogonia; and the adluminal compartment, where differentiating germ cells undergo meiosis and spermiogenesis. The BTB is unique in mammalian tissues because it is cyclically reconstructed during the spermatogenic cycle as preleptotene spermatocytes migrate from the basal compartment to the adluminal compartment and enter meiosis. In mouse, the loss of the BTB in the absence of the claudin 11 protein causes azoospermia and leads to infertility. Specifically, cldn11 deficiency results in sloughing of the cells of the seminiferous epithelium into the lumen. Understanding this pathophysiology has involved histological examination of the tissue defects as well as immunohistological characterization. Here, we present a comparative study of several modifications to the classical Hematoxylin–Eosin stain that may improve the diagnostic usefulness of this technique, as well as the use of several selective markers to identify testicular cell types. Key words: Claudin 11, Spermatogenesis, Sertoli cell, Blood testis barrier, Gene regulation
1. Introduction The Claudin (Cldn) family of tight junction (TJ) proteins comprises tetraspan transmembrane proteins that are the major selective diffusion barrier of the paracellular pathway between epithelial cells. The overall sieving properties of TJs are determined by
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_15, © Springer Science+Business Media, LLC 2011
223
224
S. Mazaud-Guittot et al.
the combination of individual Cldn family members incorporated into the intramembranous strands. Inactivation of the Cldn11 gene in mice (1) has clearly demonstrated the pivotal role of the Cldn family in TJ physiology, principally because cldn11 is often the only family member incorporated into the intramembranous strands; thus without cldn11, TJs are absent and the paracellular pathway is open. The characterization of Cldn11−/− mice has revealed the importance of cldn11 for spermatogenesis, central nervous system function, and normal hearing (1). Indeed, these mutants exhibit neurological deficits and deafness (2, 3), and males are sterile because of the failure of spermatogenesis beyond the stage of spermatocyte differentiation. The testes in these animals also exhibit a striking disorganization of the seminiferous tubules (1, 4). Understanding the relationship between cldn11 deficiency and male sterility requires a detailed examination of testis organization. The testis comprises multiple somatic cell types, including steroid-secreting Leydig cells located in the interstitial compartment and Sertoli cells in the seminiferous tubules, as well as germ cells located in the seminiferous tubules (5). Sertoli cells play several key roles in spermatogenesis: they are hormone target cells; they provide structural and nutritional support for developing germ cells; and more importantly, they synthesize the proteins that constitute the blood testis barrier (BTB), including cldn11. The BTB is assembled at the commencement of puberty between adjacent Sertoli cells and delineates the adluminal compartment of the seminiferous epithelium in which meiosis and spermiogenesis occur. The permeability of the BTB is hormonally regulated; thus, androgens regulate Cldn11 gene expression (6) and gonadotropins maintain Sertoli cell TJ integrity, including the trafficking and localization of cldn11 to functional TJs (7). In the primary study of the testicular phenotype in Cldn11−/− mice (1), cell clusters were observed in the lumen of many seminiferous tubules instead of spermatozoa. Although the cells in these clusters were not identified, histochemical analysis indicated that differentiated Sertoli cells were present. We hypothesized that the identity of the cells in these clusters would be instrumental to understanding the etiology and, to this end, we developed a methodology comprising a series of tests, including macroscopic observation, histology, and immunohistochemistry. Our approach initially involved staining of paraffin-embedded sections from wild type and Cldn11−/− mice with Hematoxylin– Eosin (H&E) and several other stains. We examined a broad range of ages throughout sexual development and in adults to develop a broad understanding of the tissue defects. Secondarily, we used a battery of selective antibodies to identify different cell types and stages of differentiation of these cells, using immunohistochemistry. This paper provides an in-depth demonstration of
15 Phenotyping the Claudin 11 Deficiency in Testis…
225
the full methods we have employed as well as a description of how the loss of Sertoli cell epithelial phenotype and azoospermia stem from the absence of cldn11 TJ integrity.
2. Materials 2.1. Tissue Processing and Sectioning
1. Phosphate-buffered saline (PBS). 2. Paraformaldehyde (Sigma, MO, USA) 4% in PBS, prepared immediately before use. Paraformaldehyde was dissolved by heating. 3. Modified Davidson’s fluid: 30% formaldehyde (Sigma), 15% ethanol, 5% acetic acid in water. 4. Bouin’s Fluid: 150 mL of saturated picric acid (Sigma), 50 mL of 37% formaldehyde, and 10 mL of glacial acetic acid. 5. Graded series of ethanols (70%, 95%, and 100%), xylene, and paraplast (Labonord). 6. Slides: Superfrost and polylysine-coated slides (Labonord).
2.2. Dewaxing, Rehydratation, Dehydratation, and Mounting 2.3. Histological Stains
1. Graded series of ethanol (50%, 70%, 95%, and 100%), xylene, and paraplast (Labonord). 2. Methanol-H2O2 (Sigma) 3%. 3. Eukitt mounting medium (Labonord). 1. Harris hematoxylin solution (#HHS16, Sigma). 2. Eosin Y solution 0.5% in water (#E4382, Sigma). 3. Periodic acid solution and Schiff’s reagent (#395-132 and #395-2016, respectively, Sigma). 4. Aniline blue (#1275, MERCK), acetic acid, phosphomolybdic acid (#20612, PROLABO), acidic fuchsin (#F8129, Sigma), Orange G (#GEPCO L01-61, Eurobio). 5. Tuchmann’s blue stock solution: 1% (w/v) aniline blue, 2% (w/v) orange G, 7% (v/v) acetic acid in water.
2.4. Immuno histochemistry
1. Bovine Serum Albumin (BSA) (Fraction V, Sigma) 10% in PBS. 2. Epitope retrieval solution: Sodium citrate stock, 100 mM: Dissolve 29.4 g of tri-sodium citrate (dihydrate) in 1 L of distilled water. Store this solution at room temperature for 3 months or at 4°C for longer storage. Citric acid stock, 100 mM: Dissolve 29.4 g of citric acid (.2H2O) into 1 L of distilled water. Store this solution at room temperature for 3 months or at 4°C for longer storage.
226
S. Mazaud-Guittot et al.
Citrate buffer, 10 mM: Prepare immediately before use by mixing 18 mL of 100 mM citric acid with 82 mL of 100 mM sodium citrate, adjusting to pH 6, and increasing the volume to 1 L with deionized water. 3. Antibody diluent (Dako Cytomation, #S2022) and antibody diluent with reducing background (Dako #S3022). 4. Primary antibodies: Ddx4 (diluted 1:750; kindly provided by Dr. T. Noce, Mitsubishi Kagaku Institute of Life Sciences, Tokyo, Japan and diluted 1:500; #Ab13840, Abcam) (8–10), TRA98 and TRA369 (calmegin) (both diluted 1:1,000; kindly provided by Dr. H. Tanaka, Osaka University, Osaka, Japan) (11, 12), Vimentin (diluted 1:100; LN-6 clone; DakoCytomation), and Gata4 and claudin11 (both diluted 1:100; sc-1237X and sc-2571, respectively, SantaCruz Biotech). 5. Secondary antibodies: Biotinylated anti-rabbit (diluted 1:500; #E0432, DakoCytomation) for rabbit primary antibodies (claudin11, vimentin), multilink swine anti-goat, -mouse, -rat IgG, biotinylated (diluted 1:500; #E0453, DakoCytomation) for goat primary antibodies (Gata4), biotinylated anti-rat IgG (diluted 1:200; #BA-4001, Vector) for rat primary TRA98 and TRA369 antibodies, Vectastain ABC kit (PK6100, Vector), DAB tablets (#D4168, Sigma).
3. Methods 3.1. Preparation of Samples
The choice of tissue fixative is of prime importance for characterization of tissue sections and diagnostic assessments. Immersion of whole testis in Bouin’s fixative provides excellent histology which is especially useful in visualizing germ cell chromatin to distinguish between different meiotic stages and permit a reliable assessment of the spermatogenic cycle in individual tubules (13, 14). Paraformaldehyde provides a milder fixation that typically preserves antigenicity and is preferred for immunohistochemistry with most antibodies that do not detect their target antigen in Bouin-fixed tissue. However, paraformaldehyde requires particular handling during testis fixation because it penetrates tissue slowly and because the testis is encapsulated by the dense connective tissue layer of the tunica albuginea. Thus, the testis is either rendered more permeable by puncturing the tunica albuginea once at each pole with an 18G needle or is fixed in situ using transcardiac perfusion. Finally, modified Davidson’s fluid has been proposed for testis fixation because it exhibits most of the useful properties of the Bouin’s and paraformaldehyde fixatives (15) (see Note 1).
15 Phenotyping the Claudin 11 Deficiency in Testis…
227
1. For Bouin’s fixation, wild type, Cldn11+/− and Cldn11−/− testes are dissected, immersed in Bouin’s solution for at least 24 h and washed several times with 70% ethanol until the yellow color of the picric acid largely disappears. Tissues can be stored in 70% ethanol and fixation should not exceed 48 h. Tissues immersed in Bouin’s fixative rapidly turn yellow. 2. For modified Davidson’s fluid, Cldn11+/− and Cldn11−/− testes are dissected, immersed in modified Davidson’s fluid for at least 24 h and washed several times with 70% ethanol. Tissues placed in Davidson’s fixative rapidly turn white and opaque. Tissue fixation should not exceed 48 h and 24 h is optimum. 3. For paraformaldehyde fixation, superior penetration of the fixative through the tunica albuginea is achieved by puncturing each pole of the testis once with an 18G needle. These holes are sufficiently small to allow fixative to enter the tissue but do not result in extrusion of the seminiferous tubules through the tunica albuginea. Tissue is then immersed in paraformaldehyde overnight at 4°C. 4. Testes are dehydrated through a graded series of ethanol into xylene and embedded in paraffin. Tissue blocks are cut in transverse orientation into 6 mm-thick sections, floated in a 45°C water bath to eliminate compression artifacts and mounted on poly-lysine or superfrost glass slides. 3.2. Histological Stain
For histological analysis, Bouin’s or Davidson’s fluid fixed organs are preferred to enable the accurate identification of germ cells from the fine fixation of their chromatin. The choice of histological stain has been instrumental in characterizing testicular pathology. H&E is commonly used in most laboratories, but Periodic Acid-Schiff-hematoxylin (PAS-H) staining is superior for distinguishing germ cell types. Indeed, PAS-H labels nuclei in blue and basement membranes and acrosomes in fuchsia pink and is the method of choice for histological evaluation of testis because it reveals the shape and orientation of the maturing acrosomes for staging of the spermatogenic cycle (16). Trichrome techniques, such as the Tuchmann-Duplessis stain, are also commonly used for epithelial tissues because they highlight the edges of epithelial units (17) (see Note 2).
3.2.1. Dewaxing and Rehydratation
For all histological stains, tissue sections are heated at 37°C for 30 min, deparaffinized in xylene (2 changes, 10 min each), rehydrated in graded ethanols (100%, 100%, 95%, 70%, and 50%, 5 min each) and briefly washed in distilled water.
3.2.2. H&E Stain
1. Immerse slides in HHS (diluted 1/3) for 30–60 s; 2. Wash in running tap water for 15 min and rinse in distilled water;
228
S. Mazaud-Guittot et al.
3. Counterstain in eosin Y solution for 1–3 min; 4. Wash slides briefly in running water and dehydrate through 95% ethanol, 2 changes of 100% ethanol (5 min each), clear in xylene and coverslip in a xylene-based mounting medium (Eukitt). 3.2.3. Periodic Acid-SchiffHematoxylin Stain
1. Immerse slides in periodic acid for 5 min at 25°C, wash in running water, and rinse in distilled water; 2. Immerse slides in Schiff’s reagent for 15 min at 25°C under a fume hood, then wash in running water for 5 min, and rinse in distilled water; 3. Counterstain sections with Harris hematoxylin diluted onethird for 30 s; 4. Wash slides in running water for 15 min, then dehydrate through 95% ethanol, 2 changes of 100% ethanol (5 min each), clear in xylene, and mount with Eukitt.
3.2.4. Tuchmann-Duplessis Trichrome Stain
1. Stain sections using the H&E stain (Subheading 3.2.2) and after eosin Y staining, wash briefly in running water; 2. Immerse slides in 1% acid fuchsin in water for 1 min, wash briefly in 1% acetic acid; 3. Immerse slides in 2% orange G solution in water for 1 min; then directly in 5% phosphomolybdic acid in water for 30 s, and immerse five to six times in running water; 4. Immerse sections in Tuchmann’s blue (diluted 1/4) for 4–10 min; then immerse five to six times in running water and quickly dehydrate, clear and mount. If the Tuchmann’s blue stain is too intense, let the slides in water for several minutes while surveying the dilution of the blue. 5. Dehydrate through 95% ethanol, 2 changes of 100% ethanol (5 min each), clear in xylene, and mount with Eukitt.
3.3. Immuno histochemistry
Although histology can generally be used to identify germ cell types if special attention is given to choosing the fixative and stain, this classical technique has limited application for characterizing the pathology in Cldn11−/− testes. In particular, identification of the cells sloughing from the seminiferous epithelium has not been possible. In the youngest animals, the composition of cell clusters is homogeneous, but the cells do not appear to be either germ cells or Sertoli cells (4). In older animals, cell clusters are heterogeneous, and identification of the various cell types has proven difficult (1). The choice of selective antibodies against a number of somatic and germ cell proteins, including Gata4 and vimentin (Vim) for Sertoli cells, Ddx4, TRA98, and TRA369 (Calmegin), has greatly improved our immunohistochemical characterization of cell clusters (see Note 3).
15 Phenotyping the Claudin 11 Deficiency in Testis…
229
3.3.1. Dewaxing, Rehydration, and Inactivating Endogenous Peroxidase
Sections are heated at 37°C for 30 min, deparaffinized in xylene (2 changes, 10 min each), rehydrated in a graded ethanol series (100%, 100%, 95%, 70%, and 50%, 5 min each), briefly washed in distilled water and rinsed in PBS. Endogenous peroxidases are inactivated by immersing slides in 0.3% H2O2 in methanol for 20 min following the rehydration step in 95% ethanol.
3.3.2. Epitope Retrieval
A major adverse effect of formalin or other aldehyde fixatives is the masking of tissue epitopes by formaldehyde, which cross-links the amino side chains of proteins and results in weak or falsenegative immunohistochemical detection of some proteins. The citrate-based unmasking buffer reverses the protein cross-links and increases the accessibility of epitopes in formalin-fixed, paraffin-embedded tissue sections for increased staining intensity by antibodies. 1. Slides are immersed in a Coplin jar containing the citrate buffer solution. The jar is covered and placed in a water bath, then microwaved twice for 2–3 min at maximum power with a 5 min delay between each treatment. 2. The jar is allowed to cool slowly for at least 30 min at room temperature and the slides are rinsed in PBS.
3.3.3. Blocking (Nonspecific) and Immunostaining
All steps are carried out at room temperature unless otherwise noted. Sections are blocked in PBS containing 10% BSA for at least 20 min in a moist chamber. After a rinse in PBS, sections are covered with primary antibody diluted appropriately in primary antibody diluent overnight at 4°C (optimal dilutions of primary antibodies are determined in pilot experiments). After several rinses in PBS, sections are incubated for at least 1 h with the appropriate secondary antibody diluted in PBS. After several rinses in PBS, sections are incubated with the avidin–biotin complex (ABC) for 30 min according to the manufacturer’s instructions. After several rinses in PBS, sections are incubated in diaminobenzidine substrate (DAB) and stain intensity is monitored using a microscope. After staining, sections are washed in PBS, counterstained in Harris hematoxylin (diluted 1/15) for 1 min, washed in water, dehydrated in a graded ethanol series, cleared in xylene, and mounted with Eukitt.
3.4. Conclusion
We have developed a step-by-step battery of histological and immunohistochemical tests that have enabled an accurate characterization of the testicular phenotype in Cldn11−/− mice. However, a deeper understanding of pathophysiology in the absence of the BTB will require ultrastructural analysis beyond that which has been performed (1), including transmission electronic microscopy and freeze-fracture, as well as immunocytochemistry to localize other TJ components. In this vein, signaling mechanisms
230
S. Mazaud-Guittot et al.
that underlie the transient opening of the TJs and the cyclical expression of TJ components during migration of spermatagonia into the adluminal compartment are poorly understood (18). Determining the full extent of BTB permeability in Cldn11−/− mice will require injection of a biotin tracer into the testicular artery and immunohistochemical analysis with antibodies against additional TJ components (6). Microarray analysis likely will be complementary to such in situ approaches for determining the mechanisms that regulate expression of the genes encoding BTB components (4).
4. Notes 1. Choice of histological stains. The H&E stain is a routine technique that is widely used for nuclear staining and general assessment of tissue morphology. Nuclei appear blue, and cytoplasm pink to red. However, H&E has limited use for characterizing complex organs, such as testis, particularly when attempting to distinguish between different stages of the spermatogenic cycle, because it only relies on the examination of chromatin. Because Periodic Acid-Schiff (PAS) stains glycoproteins, it has become the stain of choice for visualizing acrosome development, and especially the shape and orientation of the acrosome that permits accurate assessment of spermatid differentiation and staging of the seminiferous epithelium (16). In the case of the pathology of the Cldn11−/− testis, the PAS stain also reveals PAS-positive material within adluminal cell clusters (Fig. 1a, b, e). Periodic acid is a potent oxidant capable of breaking the covalent bond between two hydroxyl groups of glucopyranose. The two aldehyde groups generated reduce the leucoderivative of basic fuchsin Schiff, which becomes pink. Hematoxylin is used to enhance the contrast of nuclei. PAS stains in red fuschia aldehydes (sugars or polysaccharides) of the plasma membrane or nucleic acids, glycogen, and glycoproteins appear pink. Because PAS reacts with polysaccharides, glycoproteins, and acrosomes in the testis, we were able to identify acrosome remnants in the cell clusters (4). The Tuchmann-Duplessis trichrome stain provides good quality contrast of cells in a manner similar to H&E, but the inclusion of aniline blue in this technique also reveals matrix glycoproteins of the basement membrane, acrosomes, and nucleoli in pink (Fig. 1c, d, f). In Cldn11−/− testis, the particular flattened shape of cell nuclei in clusters is associated with a more intense blue staining of their cytoplasm, but the morphology remains difficult to interpret. Thus, the most useful stain for Cldn11−/− testes is PAS-H because of the capacity to determine
15 Phenotyping the Claudin 11 Deficiency in Testis…
231
Fig. 1. Comparison of the PAS-Hematoxylin and Tuchmann-Duplessis trichrome stains for testis histology. Transverse sections of P20 testis from control (a, c) and Cldn11−/− (b, d–f) mice were processed for either PAS-Hematoxylin histology (a, b, e) or Tuchmann-Duplessis trichrome stain (c, d, f). scale bars: 100 mm (a–d) and 50 mm (e, f ).
the stage of the spermatogenic cycle for each tubule and the labeling of small vesicles within the cell clusters. 2. Choice of antibodies for immunohistochemistry. Histological examination of the Cldn11−/− testicular pathology provided the initial insights into tissue structure, cell composition, and cell health; however, the use of several specific antigenic markers for Sertoli cells and germ cells enabled us to monitor the fate of each cell type and has permitted indisputable identification of cells in the adluminal clusters. These cells are positive for GATA4 and vimentin and negative for H3 phosphohistone, DDX4, CGLN/TRA369, TRA98, vitronectin, and cldn1 ((4) and Fig. 2). In the oldest animals, cell clusters mainly comprise Sertoli cells, but immature germ cells are occasionally observed.
232
S. Mazaud-Guittot et al.
Fig. 2. Immunohistochemical analysis of cell clusters in the adluminal compartment of Cldn11−/− mice. Immunolabeling for germ cell markers DDX4 (a, b) and calmegin (CGLN) recognized by the TRA369 antibody (c, d); Sertoli cell markers GATA4 (e, f) and Vimentin (VIM) (g, h). Sections from P20 (a, b) or P28 (c–h) control (a, c, e, g) and Cldn11−/− (b, d, f, h) testes show that cell clusters observed in the adluminal compartment of Cldn11−/− seminiferous tubules (arrow) comprise Sertoli cells. Scale bars: 100 mm.
15 Phenotyping the Claudin 11 Deficiency in Testis…
233
For Sertoli cell markers, the use of antibodies to label nuclear antigens provides more definitive identification than do cytoplasmic antigens (compare Fig. 2e–h) because cell shape is markedly variable as the seminiferous epithelium degenerates in older Cldn11−/− mice. We also paid particular attention to Sertoli cell markers with stage-specific expression through the seminiferous cycle (19, 20). Numerous proteins exhibit such cyclic expression during the development and maturation of the seminiferous epithelium or in association with testicular pathology, such as germ cell loss (20). We selected markers that continue to be expressed by Sertoli cells even in disease states. Thus, our immunohistochemical approach (4) increased the accuracy of identifying cell clusters compared to the earlier histochemical approach (1). Given the rarity of current mouse models of male infertility in which Sertoli cell sloughing from the seminiferous epithelium is prevalent, we consider that characterization of the cellular mechanism of Sertoli cell sloughing in Cldn11−/− mice is of prime importance. Finally, immunohistochemistry of several important BTB components revealed persistent expression of ZO-1 and occludin (1). Our data supports these findings and indicates that the expression of these proteins is induced in response to the inactivation of the Cldn11 gene (4). 3. Choice of tissue fixation. The use of testis tissue sections for both histological and immunohistochemical characterization requires that chemical fixation should simultaneously preserve morphological structure and protein antigenicity. A major problem in this effort is that high-quality preservation of chromatin architecture comes at the expense of preserving antibody epitopes. In histology, good morphology is crucial for developing insights into testis physiology. Fixatives such as Bouin’s or modified Davidson’s fluids that rapidly penetrate the testis and contain acetic acid to coagulate nucleic acids is preferred and recommended by the Society for Toxicologic Pathology (21). However, antigenicity is preserved to variable extents with these fixatives and may differ for each fixative. In Bouin’s fixed tissue, Gata4 and Vimentin are barely detectable (not shown) while DDX4, TRA98, and CLDN11 are readily detected. On the other hand, DDX4, TRA98, and CGLN are detected in modified Davidson’s fixed tissue (Fig. 3) but CLDN11 is not. Furthermore, immunohistochemical labeling of Bouin’s fixed tissue often shows a decreasing gradient of labeling intensity toward the center of the tissue, a pronounced side-effect due to the rate of fixative penetration into the organ. There is a clear difference in the compactness and staining intensity between Bouin’s and modified Davidson’s
234
S. Mazaud-Guittot et al.
Fig. 3. Comparison of immunocytochemical labeling using different methods of tissue fixation. Testes were fixed with either Bouin’s fixative (a, c, e, g) or modified Davidson’s fixative (b, d, f, h) and processed for immunohistochemistry with anti-cldn11 (CLDN11, a, b), anti-DDX4 (c, d), anti-TRA98 (e, f), or anti-CGLN antibodies recognized by the TRA369 antibody (g, h). Scale bars: 100 mm.
15 Phenotyping the Claudin 11 Deficiency in Testis…
235
fixed tissue that may be associated with shrinkage artifacts caused by the latter fixative (15). Paraformaldehyde fixation is the first-choice fixative for many tissues because it is a mild fixative that preserves the antigenicity of many antibody epitopes. Indeed, all antibodies described in the current study labeled paraformaldehyde-fixed testes (Fig. 2c–h). However, paraformaldehyde performs relatively poorly for preserving testis morphology because of its slow rate of tissue penetration. To maintain the integrity of seminiferous tubules and interstitial tissue, testis must be fixed whole. Unfortunately, the tunica albuginea surrounding the tissue acts as a diffusion barrier which slows fixative penetration even when a needle is used to puncture this dense conjunctive tissue at the poles. A reasonable compromise for paraformaldehyde is to use intracardiac perfusion and subsequent postfixation.
Acknowledgments We thank Drs. T. Noce and H. Tanaka for providing Ddx4 and Cgln antibodies, respectively. We thank Mr Michael Bradley, Center for Molecular Medicine and Genetics, WSU, for technical assistance in the collection of tissues from the Cldn11 mouse colony. This work was supported by Inserm, Inra, University Lyon I, and partly from grants to BLMB by ANR (ANR-06-PNRA-006) and AFSSET (EST-2006/1/33), and to A.G. by NIDCD, NIH (DC006262). References 1. Gow A, Southwood CM, Li JS, Pariali M, Riordan GP, Brodie SE, Danias J, Bronstein JM, Kachar B, Lazzarini RA. CNS myelin and sertoli cell tight junction strands are absent in Osp/claudin-11 null mice. Cell 1999; 99: 649–659. 2. Devaux J, Gow A. Tight junctions potentiate the insulative properties of small CNS myelinated axons. J Cell Biol 2008; 183: 909–921. 3. Gow A, Davies C, Southwood CM, Frolenkov G, Chrustowski M, Ng L, Yamauchi D, Marcus DC, Kachar B. Deafness in Claudin 11-null mice reveals the critical contribution of basal cell tight junctions to stria vascularis function. J Neurosci 2004; 24: 7051–7062. 4. Mazaud-Guittot S, Meugnier E, Pesenti S, Wu X, Vidal H, Gow A, Le Magueresse-
Battistoni B. Claudin 11 deficiency in mice results in loss of the Sertoli cell epithelial phenotype in the testis. Biol Reprod 2010; 82: 202–213. 5. Russell L. Observations on rat Sertoli ectoplasmic (“junctional”) specializations in their association with germ cells of the rat testis. Tissue Cell 1977; 9: 475–498. 6. Meng J, Holdcraft RW, Shima JE, Griswold MD, Braun RE. Androgens regulate the permeability of the blood-testis barrier. Proc Natl Acad Sci U S A 2005; 102: 16696–16700. 7. McCabe MJ, Tarulli GA, Meachem SJ, Robertson DM, Smooker PM, Stanton PG. Gonadotropins regulate rat testicular tight junctions in vivo. Endocrinology; 151: 2911–2922.
236
S. Mazaud-Guittot et al.
8. Fujiwara Y, Komiya T, Kawabata H, Sato M, Fujimoto H, Furusawa M, Noce T. Isolation of a DEAD-family protein gene that encodes a murine homolog of Drosophila vasa and its specific expression in germ cell lineage. Proc Natl Acad Sci U S A 1994; 91: 12258–12262. 9. Noce T, Fujiwara Y, Sezaki M, Fujimoto H, Higashinakagawa T. Expression of a mouse zinc finger protein gene in both spermatocytes and oocytes during meiosis. Dev Biol 1992; 153: 356–367. 10. Toyooka Y, Tsunekawa N, Takahashi Y, Matsui Y, Satoh M, Noce T. Expression and intracellular localization of mouse Vasahomologue protein during germ cell development. Mech Dev 2000; 93: 139–149. 11. Watanabe D, Sawada K, Koshimizu U, Kagawa T, Nishimune Y. Characterization of male meiotic germ cell-specific antigen (Meg 1) by monoclonal antibody TRA 369 in mice. Mol Reprod Dev 1992; 33: 307–312. 12. Yoshinaga K, Tanii I, Toshimori K. Molecular chaperone calmegin localization to the endoplasmic reticulum of meiotic and post-meiotic germ cells in the mouse testis. Arch Histol Cytol 1999; 62: 283–293. 13. Ahmed EA, de Rooij DG. Staging of mouse seminiferous tubule cross-sections. Methods Mol Biol 2009; 558: 263–277. 14. Borg CL, Wolski KM, Gibbs GM, O’Bryan MK. Phenotyping male infertility in the mouse: how to get the most out of a ‘nonperformer’. Hum Reprod Update 2010; 16: 205–224.
15. Latendresse JR, Warbrittion AR, Jonassen H, Creasy DM. Fixation of testes and eyes using a modified Davidson’s fluid: comparison with Bouin’s fluid and conventional Davidson’s fluid. Toxicol Pathol 2002; 30: 524–533. 16. Russell L, Ettlin R, Hikim A, Clegg E. Histological and histopathological evaluation of the testis. In: Cache River Press, Clearwater, FL, USA; 1990. 17. Tuchmann-Duplessis H. Une technique de coloration commode pour la numeration des cellules glandulaires de l’hypophyse. Bull Histol Appl Tech Microsc 1947; 24: 180. 18. Yan HH, Cheng CY. Blood-testis barrier dynamics are regulated by an engagement/ disengagement mechanism between tight and adherens junctions via peripheral adaptors. Proc Natl Acad Sci USA 2005; 102: 11722–11727. 19. Imai T, Kawai Y, Tadokoro Y, Yamamoto M, Nishimune Y, Yomogida K. In vivo and in vitro constant expression of GATA-4 in mouse postnatal Sertoli cells. Mol Cell Endocrinol 2004; 214: 107–115. 20. Sharpe RM, McKinnell C, Kivlin C, Fisher JS. Proliferation and functional maturation of Sertoli cells, and their relevance to disorders of testis function in adulthood. Reproduction 2003; 125: 769–784. 21. Creasy DM. Evaluation of testicular toxicology: a synopsis and discussion of the recommendations proposed by the Society of Toxicologic Pathology. Birth Defects Res B Dev Reprod Toxicol 2003; 68: 408–415.
Chapter 16 An In Vitro System to Study Sertoli Cell Blood-Testis Barrier Dynamics Dolores D. Mruk and C. Yan Cheng Abstract The use of an in vitro system based on primary cultures of Sertoli cells isolated from rat testes has greatly facilitated the study of the blood-testis barrier in recent years. Herein, we summarize the detailed procedures on the isolation of undifferentiated Sertoli cells from 20-day-old rat testes, the culture of these cells as a monolayer on Matrigel-coated bicameral units, the characterization of these cultured cells, and the use of the Sertoli cell epithelium for monitoring the integrity of the Sertoli cell blood-testis barrier. This information is based on the routine use of this system in our laboratory to study the Sertoli cell bloodtestis barrier in the past two decades, which should be helpful for investigators in the field. Key words: Testis, Sertoli cell, Blood-testis barrier, Tight junction, Ectoplasmic specialization, Anchoring junction, Spermatogenesis
1. Introduction The concept of the blood-testis barrier (BTB) was described more than a century ago when dyes injected into animals were found to stain all organs except the brain and seminiferous tubules in the testis (1, 2), illustrating the presence of a blood–tissue barrier in these two organs. Subsequent studies showed that the BTB, unlike other blood–tissue barrier [e.g., blood–brain barrier (BBB) and blood–retina barrier (BRB, also known as the blood-ocular barrier)] which are composed of tight junctions (TJs) between endothelial cells, is constituted almost exclusively by adjacent Sertoli cells whose function is to create an immunological barrier located near the basement membrane in the seminiferous epithelium (2–5). In other words, TJs between endothelial cells found
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_16, © Springer Science+Business Media, LLC 2011
237
238
D.D. Mruk and C.Y. Cheng
in the interstitium between seminiferous tubules play a relatively insignificant role in maintaining BTB function. Equally important, the BTB is not only constituted by TJs, but also by coexisting basal ectoplasmic specializations [basal ES, a testis-specific atypical adherens junction (AJ)], gap junctions, and desmosome junctions present between Sertoli cells (6–8) so that in this respect the BTB is a unique ultrastructure worthy of study. In recent years, there has been increasing interest in using Sertoli cells as a model to study BTB dynamics for a number of important reasons. First, it has been known since the 1980s that Sertoli cells cultured at high density in vitro have the ability to form a functional epithelium that closely mimics the BTB in vivo both structurally and functionally (9, 10). For instance, these Sertoli cells became polarized when seeded on extracellular matrix (e.g., Matrigel™) (9, 10) and created a functional TJ-permeability barrier (11–14). Moreover, ultrastructures corresponding to TJs, basal ES, gap junctions, and desmosome junctions that closely mimic the BTB in vivo were visible between adjacent Sertoli cells by electron microscopy (15, 16). Second, even though the BTB is one of the tightest blood–tissue barriers known to exist in mammals, surprisingly it is extremely susceptible to damage by environmental toxicants (e.g., cadmium) with its disruption often occurring before damage to the endothelial TJ-barrier is detected (17, 18). In fact, a recent study has shown that the developing BTB in immature rats is even more sensitive to toxicants, such as bisphenol A, than the animal as a whole which exhibited no signs of overt toxicity (19). Thus, this in vitro system provides a simple but suitable model that can be used in place of in vivo models [e.g., BTB integrity assay in which a dye is injected via the jugular vein to monitor barrier function similar to studies published in the 1970s (1, 20)] to assess the effects of different compounds on BTB function (21–23). Indeed, this in vitro system was recently used to examine the role of focal adhesion kinase (FAK) in BTB dynamics when the association of FAK with the occludin-ZO-1 protein complex was found to increase following cadmium treatment just prior to the disruption of the BTB (24, 25). It was shown that such an increase in association between FAK and the occludin-ZO-1 protein complex led to “unwanted” phosphorylation of occludin, causing occludin’s dissociation from ZO-1 (24), thereby disrupting cell adhesion at the BTB. These findings also suggest that cadmium-induced disruption of the BTB may be “kept in check” if FAK in the testis can be targeted therapeutically (24). Third, because of coexisting TJs, basal ES, gap junctions, and desmosome junctions which collectively contribute to the unique nature of the BTB, this in vitro system provides an interesting model to study the roles of these junctions in immunological barrier function. For instance, it was recently shown that gap junctions (26) and desmosomes (27) are crucial to maintaining the integrity of the BTB in that they mediate crosstalk not only among
16 An In Vitro System to Study Sertoli Cell Blood-Testis Barrier Dynamics
239
themselves but also between basal ES and TJs. Finally, the BTB is a very important structure in the seminiferous epithelium that sequesters meiosis I and II, and all subsequent events of postmeiotic germ cell development (e.g., spermiogenesis) in a specialized microenvironment known as the apical (or adluminal) compartment so that the host’s immune system cannot develop antibodies against sperm-specific antigens, some of which arise transiently during meiosis and spermiogenesis. If this were to occur, infertility would result. Thus, culturing Sertoli cells at high density in vitro provides a good system to study immunological barrier function. Herein, we provide a detailed protocol for the isolation of highly pure Sertoli cells from rat pups, as well as a protocol for the measurement of transepithelial resistance as a reliable means to assess barrier function in vitro. The protocol that we have been using for the past three decades to isolate Sertoli cells was adopted from Dr. Jennie Mather (28, 29) and published previously (30) with minor modifications (31), whereas the procedure that we have been using to lyse germ cells that may have inadvertently contaminated Sertoli cell cultures (it involves the use of a hypotonic buffer) was originally published by Dr. Mario Sefanini’s laboratory (32). We generally isolate Sertoli cells from 20-day-old rats for three important reasons. First, Sertoli cells isolated from rats at this age had almost negligible somatic (i.e., Leydig and peritubular myoid cells) and germ cell contamination with a reported Sertoli cell purity of >98% (33, 34) when cultures were hypotonically treated and cell purity was monitored by RT-PCR and/or immunoblotting using markers specific for different cell types in the testis (34). Second, Sertoli cell yield (~180– 200 × 106 Sertoli cells) has been shown to be excellent when ten rat pups at 20 days of age were used. Third, Sertoli cells isolated from 20-day-old rat testes are differentiated, and they have ceased to divide by this time (35). As such, they share similar physiological characteristics with adult Sertoli cells as reported in earlier studies (36, 37). It is also worth noting that a procedure is available to isolate Sertoli cells from adult rat testes using BSA gradients (38). However, the procedure for isolating Sertoli cells from adult rat testes is very tedious, time consuming, and expensive. Moreover, the purity of adult Sertoli cells isolated using this technique stands at ~85% owing to the fact that cultures are mostly contaminated with elongating/elongated spermatids, and multiple treatments with hypotonic buffer do not appear to increase cell purity to a more acceptable level (36, 37). This being the case, we suggest using 20-day-old rat testes for the isolation of Sertoli cells for most in vitro experiments. The following text is divided into two main sections. In the first section, we provide the protocol for the isolation of Sertoli cells from 20-day-old rat testes. In the second section, we describe the technique used to assess the integrity of the TJ-permeability barrier.
240
D.D. Mruk and C.Y. Cheng
2. Material 2.1. Isolation and Culturing of Sertoli Cells
The isolation of Sertoli cells from rat testes is a relatively straightforward procedure involving several enzymatic digestion and washing steps. Sterile, disposable 50 ml and 15 ml conicals (15 ml conicals must contain graduations from 0.1 to 1 ml which are needed to determine cell pack volume). Sterile, glass 125 ml Erlenmeyer flasks with caps. Sterile Pasteur pipettes. Sterile, individually wrapped, disposable 10 ml pipettes. Tissue culture-treated 100-mm dishes, multiwell plates (Corning, Corning NY) and/or glass coverslips (Thomas Scientific, Swedesboro, NJ). Matrigel™ (BD Biosciences, Bedford, MA). Approximately 3L of F12/DMEM prepared according to the specification sheet using Milli-Q-grade water (e.g., Millipore Advantage A10 or Siemens UltraGenetic Pure Lab water systems) with a conductivity of 18.2 MW cm. Media must be sterilized by filtration through a 0.2-mm filtering unit inside a certified cell/ tissue culture hood. Standard laboratory microscope. Soybean trypsin inhibitor (STI) Type IIS – (Cat. No. T9128 Sigma-Aldrich, St. Louis, MO); Prepare 1% (i.e., 1 gm/100 ml) in PBS (10 mM sodium phosphate, pH 7.4 at pH 22°C containing 0.15 M NaCl); Store in 0.4 ml aliquots at −20°C. Trypsin (from bovine pancreas); Type I – (Cat. No. T8003 Sigma-Aldrich); Store in 40 mg/0.4 ml aliquots at −20°C. Collagenase; Type I – (Cat. No. C0130 Sigma-Aldrich); Store in 20 mg/0.2 ml and 40 mg/0.4 ml aliquots containing 0.1% STI at –20°C. Hyaluronidase (from bovine testes); Type IS (Cat. No. H3506 Sigma-Aldrich); Store in 40 mg/0.4 ml aliquots containing 0.1% STI at −20°C. DNase (from bovine pancreas) (Cat. No. DN25 SigmaAldrich); Prepare 2 mg/ml in F12/DMEM. EGF 10 mg/ml stock. Human transferrin 10 mg/ml stock. Bovine insulin 20 mg/ml stock. Bacitracin 10 mg/ml stock. All enzymes and reagents listed above should be prepared in PBS unless otherwise specified and filtered through a 0.2-mm low protein binding filter (e.g., 0.2-µm Nalgene syringe filters, Thermo Scientific, Cat. No. 190-2520) to sterilize. Dulbecco’s Modified Eagle’s Medium Ham’s F12 Nutrient Mixture (F12/DMEM) (Cat. No. D2906 Sigma-Aldrich)
16 An In Vitro System to Study Sertoli Cell Blood-Testis Barrier Dynamics
241
supplemented with sodium bicarbonate (1.2 gm/L), gentamicin (20 mg/L), and phenol red (0.00863 gm/L) as instructed by the vendor. 2.2. To Assess Sertoli Cell TJ-Permeability Barrier Function 2.2.1. Materials
Millicell-HA culture plate inserts (Millipore, Billerica, MA) (e.g., Cat. #: PIHA01250, 12-mm diameter; ~1.1304 cm2 surface area). Millicell ERS system (Millipore, Cat. #:MERS 000 01): meter and Ag/AgCl electrodes (Millipore, Cat. #:MERSSTX01).
3. Methods 3.1. Pre-isolation set-up
3.2. Sertoli Cell Isolation
Use sterile scissors and forceps, or place them into 70% ethanol on the morning of the cell isolation. On the morning of the cell isolation, prepare 1 M glycine, 2 mM EDTA in F12/DMEM, pH 7.4 (~100 ml), sterilized by filtration through a 0.2-mm filter. Media must be warmed to 37°C. A centrifuge set at room temperature (~20°C), shaking water bath set at 37°C, and cell culture hood should be reserved for ~5–6 h of usage. Matrigel™-coated dishes, plates, and/or bicameral units should be prepared 24 h prior to cell plating by diluting Matrigel™ 1:7 in F12/DMEM. These should be incubated in a humidified CO2 incubator at 35°C in order to allow the Matrigel™ to completely dry. (Note: MatrigelTM should be removed from –20°C and kept at 4°C overnight for its liquefaction prior to its use for dilution in F12/DMEM for plating on culture dishes, bicameral units, and/or microscopic slides. 1. Ten male Sprague–Dawley rats (Charles River Laboratories, Kingston, NY) to be 20 days of age on the day of use. One can anticipate routinely isolating ~180–200 × 106 Sertoli cells from ten rat pups. Euthanize animals by asphyxiation using CO2 from a carbon dioxide tank with the regulator set at ~20 psi (~1.4 bars) for about 2–2.5 min using a setup similar to the one commercially available from Braintree Scientific, Inc. (Braintree, MA). 2. Rinse the scrotal area with 70% ethanol. 3. Remove testes, decapsulate to remove the tunica albuginea, and place all 20 testes into ~10 ml F12/DMEM in a 100-mm dish. From this step onward, all remaining culture steps will be performed inside a ventilated, certified cell culture hood with proper air flow located inside a specialized tissue culture room at room temperature unless otherwise specified.
242
D.D. Mruk and C.Y. Cheng
4. Transfer all testes (20 total) into another 100-mm dish containing ~10 ml F12/DMEM. Cut testes into ~1-mm pieces with surgical-grade, curved scissors, such as from Miltex. 5. Transfer testes into a 50-ml conical tube and add F12/ DMEM up to 50 ml. Wash testes by gentle agitation to remove contaminating blood cells and centrifuge at 800 × g for 2 min at room temperature. 6. Aspirate F12/DMEM with a Pasteur pipette. Add F12/ DMEM up to 50 ml again and wash until media is clear, usually two times. 7. Resuspend testes in 40 ml F12/DMEM containing 40 mg of trypsin and 0.8 mg DNase. Transfer testes into a sterile 125 ml Erlenmeyer flask with cap. Place in a shaking water bath at 60–90 osc/min for 30 min. This step will release Leydig cells and other interstitial cells (e.g., fibroblasts). 8. Transfer seminiferous tubules into a 50-ml conical. Add F12/ DMEM up to 50 ml and centrifuge at 800 × g for 2 min. 9. Aspirate F12/DMEM with a Pasteur pipette. 10. Resuspend seminiferous tubules in 40 ml of 1 M glycine, 2 mM EDTA, pH 7.4 containing 0.01% STI, and 0.8 mg DNase. Incubate at room temperature for 10 min with periodic gentle agitation. This step will lyse Leydig cells. 11. Transfer seminiferous tubules into a 50-ml conical. Add F12/ DMEM up to 50 ml and centrifuge at 800 × g for 2 min. 12. Aspirate F12/DMEM with a Pasteur pipette. 13. Wash cells by centrifugation at 800 × g two times or until F12/DMEM is clear. 14. Resuspend cells by gentle pipetting (~20–25 times) with a Pasteur pipette in 40 ml F12/DMEM containing 20 mg of collagenase and 0.2 mg DNase. Transfer cells into an Erlenmeyer flask and place in a shaking water bath at 60–90 osc/min for 5 min. 15. Transfer cells into a 50 ml conical. Add F12/DMEM up to 50 ml and centrifuge at 800 × g for 2 min. 16. Resuspend cells by gentle pipetting (as described above) with a Pasteur pipette in 40 ml F12/DMEM containing 40 mg of collagenase and 0.2 mg DNase. Transfer cells into an Erlenmeyer flask and place in a shaking water bath at 60–90 osc/min for 30 min. These two consecutive steps will remove peritubular myoid cells. 17. Transfer cells into a 50-ml conical. Add F12/DMEM up to 50 ml and centrifuge at 800 × g for 2 min. 18. Wash cells by centrifugation at 800 × g three times (with gentle resuspension of cells in fresh F12/DMEM between centrifugations).
16 An In Vitro System to Study Sertoli Cell Blood-Testis Barrier Dynamics
243
19. Resuspend cells by gentle pipetting (as described above) with a Pasteur pipette in 40 ml F12/DMEM containing 40 mg of hyaluronidase and 0.2 mg DNase. Transfer cells into an Erlenmeyer flask and place in the shaking water bath at 60–90 osc/min for 30 min. This step will break down hyaluronic acid, a major component of the extracellular matrix. 20. Transfer cells into a 50-ml conical. Add F12/DMEM up to 50 ml and centrifuge at 800 × g for 2 min. 21. Wash cells by centrifugation at 800 × g five times. 22. Transfer cells into a 15-ml conical. Add F12/DMEM up to 15 ml and centrifuge at 800 × g for 2 min. Depending on the cell pack volume and on experimental design, resuspend cells as follows (to obtain 0.5 × 106 cells/ml): Cell pack (ml)
Number of cells (106)
Volume of media (ml)
0.3
108
216
0.4
144
288
0.5
180
360
0.6
216
432
0.7
252
504
0.8
288
576
23. Resuspend cells in F12/DMEM containing 10 mg/ml insulin, 5 mg/ml human transferrin, 2.5 ng/ml EGF and 5 mg/ml bacitracin. Plate cells at low (5 × 104 cells/cm2) or high (0.5 × 106 cells/cm2) density. For high density, 100-mm dishes, multiwell plates, glass coverslips, or culture plate inserts should be coated with Matrigel™ to improve cell viability. To prepare cells for quantifying TJ permeability barrier function, a cell density of 1–1.2 ´ 106 cells/cm2 should be used. 24. Incubate cells in culture dishes/bicameral units in a humidified CO2 incubator at 35°C with 95% air and 5% CO2 (v/v). Sertoli cells will be in small clusters of five to ten cells as it is impossible to isolate single Sertoli cells. Once these cell clusters attach to the substratum, cells will quickly spread out to form an epithelium. 3.3. Hypotonic Treatment
This step is performed 36–48 h after plating Sertoli cells on 100mm dishes, multiwell dishes or glass coverslips to remove residual germ cells from Sertoli cells, so that these Sertoli cells will be contaminated with negligible germ cells and Leydig cells (see Note 4.1). For hypotonic treatment: 1. Prepare 20 mM Tris, pH 7.4 (100–500 ml), depending on the number of dishes/plates that need to be treated. 2. Filter the above buffer to sterilize.
244
D.D. Mruk and C.Y. Cheng
3. Buffer must be sterilized by filtration through a 0.2-mm filtering unit. (a) Aspirate media with a Pasteur pipette. (b) Add an appropriate volume of the above buffer per dish (e.g., ~10 ml per 100-mm dish; ~ 2 ml per well in a six-well dish) for exactly 2.5 min to lyse contaminating germ cells. (c) Aspirate buffer with a Pasteur pipette. (d) Wash two times with F12/DMEM to remove residual Tris buffer. (e) Add F12/DMEM supplemented with four factors as described above. (f) Incubate cells in a humidified CO2 incubator at 35°C with 95% air and 5% CO2 (v/v). (g) Feed cells every 3–4 days (low cell density) or every 1–2 days (high cell density) with growth factor supplemented-F12/DMEM. 3.4. Measurement of TER to Assess Barrier Function In Vitro
When Sertoli cells are plated on Matrigel™-coated bicameral culture units at 0.5–1.2 × 106 cells/cm2, they are capable of forming an intact epithelium with a functional TJ-permeability barrier, and these cells can resist the passage of an electrical current (i.e., ~2 s pulse of ~20 mA) sent across the epithelium (Fig. 1) with a Millipore Millicell-ERS meter with two electrodes which are placed in the apical and basal chambers of the bicameral unit (Fig. 2) (see also Note 4.2). TJ-barrier function is recorded in ohms (W) (i.e., the resistance of the cell epithelium to the flow of
Fig. 1. A typical experiment assessing the assembly of the TJ-permeability barrier in Sertoli cells cultured on Matrigel™-coated culture units by quantifying the TER across the epithelium. TER was recorded from quadruplet bicameral culture chambers (see Fig. 2A) in which Sertoli cells were plated at a density of 1 × 106 cells/cm2 as described in the text.
16 An In Vitro System to Study Sertoli Cell Blood-Testis Barrier Dynamics
245
Ag/AgCl electrodes
apical compartment
*
Nu basal compartment
Fig. 2. A schematic drawing illustrating Sertoli cells (Nu, nucleus of a Sertoli cell) forming an intact epithelium after being plated on bicameral culture units and used for recording TER. One chopstick (i.e., the longer end) of the Millicell ERS electrode is inserted into the basal compartment, whereas the other chopstick (i.e., the shorter end) is placed in the apical compartment of culture units. A short pulse of current is then sent across the Sertoli cell epithelium and resistance (in W) is recorded. This technique monitors the integrity of the TJ-permeability barrier. A cytoplasmic process on the apical side of the Sertoli cell is clearly visible (see asterisk) which is a typical characteristic of Sertoli cells when cultured in vitro.
current across the epithelium), and this value is multiplied by the surface area of the bicameral unit (cm2) to yield transepithelial electrical resistance (TER) (Fig. 1) (see Note 4.2 for precautionary notes to obtain reproducible and reliable data from this experiment). The TJ-barrier in Sertoli cells is formed by day 3 at which time TER values reach a peak (~50–70 W cm2), followed by a plateau (Fig. 1). The assembly of functional junctions between Sertoli cells can be further characterized by fluorescence microscopy, which in this case was used to visualize actin filaments (Fig. 3a), as well as occludin (a TJ-integral membrane protein) (Fig. 3b) and ZO-1 (a TJ-associated adaptor known to form a 1:1 stoichiometric ratio with occludin) (Fig. 3c) between adjacent Sertoli cells. Prior to their use, electrodes to be used to quantify the TER across the Sertoli cell epithelium (see below) should be placed into 70% ethanol (~5 ml) on the morning of the isolation, and then into F12/DMEM (~5 ml) 3 h before recording TER. 1. Sertoli cells being used for TER experiments must be first plated on Matrigel™-coated culture units at a density of 0.5–1.2 × 106 cells/cm before remaining Sertoli cells are plated on other substrates (i.e., 100-mm dishes, multiwell
246
D.D. Mruk and C.Y. Cheng
Fig. 3. (a–c) The functional characteristics of the Sertoli cell TJ-barrier are assessed by fluorescence microscopy. Sertoli cells were cultured at 0.025 × 106 cells/cm2 on Matrigel™-coated glass coverslips in order to visualize actin using FITC-conjugated phalloidin (a, Molecular Probes, Eugene, OR), occludin using an anti-occludin antibody (b, Invitrogen, San Francisco, CA) and ZO-1 using an anti-ZO-1 antibody (c, Invitrogen). For (b) and (c), occludin and ZO-1 were visualized using a secondary antibody conjugated to FITC. Sertoli cell nuclei were stained with DAPI. Bar = 10 mm in (a), which applies to (b) and (c).
plates, or glass coverslips). Each experiment must also include Matrigel™-coated culture units without cultured Sertoli cells (i.e., blank) for background subtraction. 2. When plating cells, caution should be taken to avoid trapping air bubbles on the inner side of culture units (i.e., where the nitrocellulose membrane meets the plastic). If this occurs, use
16 An In Vitro System to Study Sertoli Cell Blood-Testis Barrier Dynamics
247
a Gilson pipette with 1 ml attached pipette tip to resuspend Sertoli cells, in the process dislodging the air bubble. 3. Each bicameral culture unit is placed into one well of a 24 multiwell plate. Each culture unit should be positioned in the center of the 24 well. The maximum volume of media in each culture unit compartment should be 500 ml (both basal and apical chambers). Growth factor supplemented-F12/DMEM should be replaced daily. 4. It is important that Sertoli cells be plated uniformly on culture units. After plating Sertoli cells, allow the 24-well plate containing culture units to sit in the hood at room temperature for ~5–10 min before placing it into the CO2 incubator. Placing the plate into the incubator immediately after seeding tends to cluster cells in the center of the plate/ well/culture unit, leading to variations in TER within one culture unit. 5. The Millicell ERS meter, as well as the electrodes, should be first tested as instructed by the vendor. The range should be set to 2,000 W, and the mode should be set to RESISTANCE on the meter. 6. The first TER measurement should be ~24 h after plating cells and daily thereafter by placing one electrode in the apical chamber and the other electrode in the basal chamber. It is noted that one cannot obtain reliable TER readings immediately after plating Sertoli cells on culture units so that cells must be cultured for 24 h prior to obtaining the first reading. A total of four readings should be recorded at 12, 3, 6, and 9 o’clock positions for each bicameral culture unit, and these should be averaged into a single TER reading. 7. Multiwell plates containing culture units must be at room temperature before recording TER to avoid fluctuations in TER readings. Remove plates from the incubator and allow them to sit at room temperature for 20 min before recording TER. 8. When recording TER, Millicell ERS electrodes should stand upright in the well, and the culture unit should sit in the center of the 24-well. A ~2 s pulse of current (~20 mA) should be passed through the cell epithelium, and this number should be recorded. 9. F12/DMEM should be replaced after (NOT before) recording TER. 10. It is not necessary to rinse electrodes in between culture units. However, electrodes can be dipped into a well containing F12/DMEM to rinse the electrodes if the experiment involved treating Sertoli cells with different chemicals/compounds. Each treatment or control group should contain at least four replicate bicameral units to obtain sufficient data for statistical
248
D.D. Mruk and C.Y. Cheng
analysis. Each experiment should be repeated at least three times to assess reproducibility. 11. TER is recorded as follows: Resistanceunknown − Resistanceblank = Resistancecalculated × Effective surface area = W cm2.
4. Notes 4.1. Preparation of Sertoli Cell Cultures
It is noted that Sertoli cells isolated from 20-day-old rats and cultured in vitro are differentiated cells (35) since they cease to divide by postnatal day 15–17 at the time the blood-testis barrier is established in vivo. Furthermore, these cells establish the ultrastructures of both TJ, basal ectoplasmic specialization [a testisspecific atypical adherens junction type (39)], gap junction and desmosome junction when examined by electron microscopy within 48 h (15) which mimic the ultrastructural features of BTB in vivo. These morphological observations are also consistent with physiological data shown in Fig. 1, illustrating the establishment of a TJ-barrier within ~24–48 h after plating Sertoli cells onto the Matrigel-coated bicameral units. Listed below are a few precautionary notes for a successful experiment to monitor the Sertoli cell TJ-permeability barrier function. 1. A proper Sertoli cell density plated on Matrigel-coated bicameral units is critical to obtain reliable and reproducible TER reading since the entire surface area of the bicameral units must be covered with Sertoli cells to prevent a “leaky” TJ-barrier. We noted that using a Sertoli cell density at a range of 0.75–1.2 × 106 Sertoli cells/cm2 yielded the best data for this type of experiment as reported earlier (40) since at this cell density range, the cell epithelium on the bicameral unit was composed a single columnar-shaped Sertoli cell layer (41–43) similar to the Sertoli cells observed in vivo. However, when a higher Sertoli cell density was used, such as at 1.5–2 × 106 cells/ cm2, Sertoli cells began to stack into multiple layers (41). Even though this occurred in just a few patched areas on the bicameral units, necrosis was detected in these high cell density cultures (41), which in turn led to cell lysis, generating a small area without healthy Sertoli cells, causing the TJ-barrier to become “leaky” (22, 40). To prepare Sertoli cells for immunoblotting to assess changes of steady-state protein levels in treatment groups versus controls, we recommend the use of cell density at 0.5 ´ 106 cells/cm2, which would yield sufficient protein lysates for probing different marker proteins. 2. If one intends to use these Sertoli cell cultures to assess the effects of different treatments and/or drugs/chemicals/ reagents on the Sertoli cell TJ-barrier morphologically, such
16 An In Vitro System to Study Sertoli Cell Blood-Testis Barrier Dynamics
249
as the use of dual-labeled immunofluorescence analysis with fluorescence or confocal microscopy, a cell density at ~0.02– 0.05 × 106 cells/cm2 is recommended (27, 44–46) (see Fig. 3). If a higher cell density is used, Sertoli cells become too closely packed when nuclei were visualized by DAPI staining, making it difficult to visualize changes in protein localization, in particular protein redistribution, at the Sertoli–Sertoli cell interface. However, at this lower cell density, a few selected areas on the culture dish, glass slide, or coverslip may not be covered by Sertoli cells, but since the analysis is restricted on the use of fluorescent or confocal microscopy to assess changes in protein localization and/or redistribution at the Sertoli–Sertoli cell interface, this will neither affect data analysis nor the purpose of the experiment. 3. It is also important that from time to time, such as every 2–3 months, Sertoli cells from a given investigator who uses these Sertoli cell cultures for his/her experiments routinely, should be monitored for cell purity such as for contamination of germ cells and/or Leydig cells by RT-PCR using specific markers for spermatogonia, spermatocytes, spermatids, and Leydig cells as described earlier (33, 34). It is also necessary to monitor the presence of ultrastructures of TJ, basal ES, desmosome-like junction by electron microscopy, as well as the establishment of a functional TJ-permeability barrier by physiological techniques to assess TER across the Sertoli cell epithelium as described (15, 47, 48). 4.2. Assessing the Sertoli Cell TJ-Permeability Barrier by Recording TER Across the Sertoli Cell Epithelium
Below are some additional precautionary notes to be taken in order to obtain reproducible and reliable TER readings in a given experiment in order to assess Sertoli cell TJ-permeability barrier function. 1. As described above, a Sertoli cell density at 0.75–1.0 × 106cells/ cm2 is recommended to assess the Sertoli cell TJ-permeability barrier. At this range, the entire surface area of the Matrigelcoated bicameral unit will be covered with a layer of columnarshaped Sertoli cells. This will yield stable and steady TER readings across the cell epithelium. Caution should be taken to allow the Matrigel-coated units to dry overnight (at least ~8–10 h) before use to provide a scaffolding support to the Sertoli cell epithelium. 2. If a hypotonic treatment (32) is used to obtain Sertoli cell cultures with >98% purity, it is important that the cell epithelium on the Matrigel-coated bicameral unit be handled gently during the washing step with F12/DMEM to avoid perturbing the cell epithelium to create an artificial “leak,” causing a loss of TER across the Sertoli cell epithelium. We had also used collagen gel-coated bicameral units to substitute Matrigel,
250
D.D. Mruk and C.Y. Cheng
but these experiments did not yield results comparable to the use of Matrigel since the TER readings are less stable. 3. Since the Millipore ERS system will be used for several days (at least 5–7 days), it is important to check if the system is properly calibrated, the battery level is optimal, and the pair of electrodes is functioning properly. This is important since any unwanted changes to the ERS system and the electrodes will cause an artificial shift in the baseline TER reading, causing fluctuation in the TER readings between different time points within an experiment. 4. It is also important to have at least a triplicate set of bicameral units for each treatment group versus control. If an experiment is being handled by a less experienced investigator, one can consider to use five replicates per treatment group including control. References 1. Setchell B.P., Waites G.M.B. The blood-testis barrier. In: Hamilton D.W., Greep R.O. (eds.), The Handbook of Physiology. Section 7, Vol. V. Male Reproductive System. Washington, D.C.: American Physiological Society; 1975: 143–172. 2. Setchell B.P. (2008) Blood-testis barrier, junctional and transport proteins and spermatogenesis. In: Molecular Mechanisms in Spermatogenesis. Ed. Cheng C.Y., Austin, T.X., Landes Bioscience/Springer Science + Business Media, LLC, pp. 212–233. 3. Wong C.H., Cheng C.Y. (2005) The bloodtestis barrier: Its biology, regulation and physiological role in spermatogenesis. Curr Topics Dev Biol 71: 263–296. 4. Cheng C.Y., Mruk D.D. (2002) Cell junction dynamics in the testis: Sertoli-germ cell interactions and male contraceptive development. Physiol Rev 82: 825–874. 5. Mruk D.D., Silvestrini B., Cheng C.Y. (2008) Anchoring junctions as drug targets: Role in contraceptive development. Pharmacol Rev 60: 146–180. 6. Russell L.D., Peterson R.N. (1985) Sertoli cell junctions: morphological and functional correlates. Int Rev Cytol 94: 177–211. 7. Yan H.H.N., Mruk D.D., Cheng C.Y. (2008) Junction restructuring and spermatogenesis: The biology, regulation, and implication in male contraceptive development. Curr Top Dev Biol 80: 57–92. 8. Yan H.H.N., Mruk D.D., Lee W.M., Cheng C.Y. (2007) Ectoplasmic specialization: a friend or a foe of spermatogenesis? BioEssays 29: 36–48.
9. Byers S., Hadley M.A., Djakiew D., Dym M. (1986) Growth and characterization of epididymal epithelial cells and Sertoli cells in dual environment culture chambers. J Androl 7: 59–68. 10. Janecki A., Steinberger A. (1986) Polarized Sertoli cell functions in a new two-compartment culture system. J Androl 7: 69–71. 11. Janecki A., Jakubowiak A., Steinberger A. (1991) Regulation of transepithelial electrical resistance in two-compartment Sertoli cell cultures: in vitro model of the blood-testis barrier. Endocrinology 129: 1489–1496. 12. Janecki A., Jakubowiak A., Steinberger A. (1991) Effects of cyclic AMP and phorbol ester on transepithelial electrical resistance of Sertoli cell monolayers in two-compartment culture. Mol Cell Endocrinol 82: 61–69. 13. Okanlawon A., Dym M. (1996) Effect of chloroquine on the formation of tight junctions in cultured immature rat Sertoli cells. J Androl 17: 249–255. 14. Grima J., Pineau C., Bardin C.W., Cheng C.Y. (1992) Rat Sertoli cell clusterin, a2-macroglobulin, and testins: biosynthesis and differential regulation by germ cells. Mol Cell Endocrinol 89: 127–140. 15. Siu M.K.Y., Wong C.H., Lee W.M., Cheng C.Y. (2005) Sertoli-germ cell anchoring junction dynamics in the testis are regulated by an interplay of lipid and protein kinases. J Biol Chem 280: 25029–25047. 16. Byers S., Pelletier R.M., Suarez-Quain C. Sertoli cell junctions and the seminiferous epithelium barrier. In: Russell L.D., Griswold
16 An In Vitro System to Study Sertoli Cell Blood-Testis Barrier Dynamics
17.
18.
19.
20.
21.
22.
23. 24.
25.
26.
27.
M.D. (eds.), The Sertoli Cell. Clearwater: Cache River Press; 1993: 431–446. Setchell B.P., Waites G.M.H. (1970) Changes in the permeability of the testicular capillaries and of the “blood-testis barrier” after injection of cadmium chloride in the rat. J Endocrinol 47: 81–86. Wong C.H., Mruk D.D., Siu M.K.Y., Cheng C.Y. (2005) Blood-testis barrier dynamics are regulated by a2-macroglobulin via the c-Jun N-terminal protein kinase pathway. Endocrinology 146: 1893–1908. Li M.W.M., Mruk D.D., Lee W.M., Cheng C.Y. (2009) Disruption of the blood-testis barrier integrity by bisphenol A in vitro: Is this a suitable model for studying blood-testis barrier dynamics? Int J Biochem Cell Biol 41: 2302–2314. Li M.W.M., Xia W., Mruk D.D., Wang C.Q.F., Yan H.H.Y., Siu M.K.Y., Lui W.Y., Lee W.M., Cheng C.Y. (2006) TNFa reversibly disrupts the blood-testis barrier and impairs Sertoli-germ cell adhesion in the seminiferous epithelium of adult rat testes. J Endocrinol 190: 313–329. Janecki A., Jakubowiak A., Steinberger A. (1992) Effect of cadmium chloride on transepithelial electrical resistance of Sertoli cell monolayers in two-compartment cultures – a new model for toxicological investigations of the “blood-testis” barrier in vitro. Toxicol Appl Pharmacol 112: 51–57. Chung N.P.Y., Cheng C.Y. (2001) Is cadmium chloride-induced inter-Sertoli tight junction permeability barrier disruption a suitable in vitro model to study the events of junction disassembly during spermatogenesis in the rat testis? Endocrinology 142: 1878–1888. Siu E.R., Mruk D.D., Porto C.S., Cheng C.Y. (2009) Cadmium-induced testicular injury. Toxicol Appl Pharmacol 238: 240–249. Siu E.R., Wong E.W.P., Mruk D.D., Porto C.S., Cheng C.Y. (2009) Focal adhesion kinase is a blood-testis barrier regulator. Proc Natl Acad Sci USA 106: 9298–9303. Siu E.R., Wong E.W.P., Mruk D.D., Sze K.L., Porto C.S., Cheng C.Y. (2009) An occludinfocal adhesion kinase protein complex at the blood-testis barrier: a study using the cadmium model. Endocrinology 150: 3336–3344. Li M.W.M., Mruk D.D., Lee W.M., Cheng C.Y. (2009) Connexin 43 and plakophilin-2 as a protein complex that regulates bloodtestis barrier dynamics. Proc Natl Acad Sci USA 106: 10213–10218. Lie P.P.Y., Cheng C.Y., Mruk D.D. (2010) Crosstalk between desmoglein-2/desmocollin-2/Src kinase and coxsackie and adenovirus
28. 29. 30.
31.
32.
33.
34.
35. 36.
37.
38.
39.
251
receptor/ZO-1 protein complexes, regulates blood-testis barrier dynamics. Int J Biochem Cell Biol 42: 975–986. Mather J.P. (1980) Establishment and characterization of two distinct mouse testicular epithelial cell lines. Biol Reprod 23: 243–252. Mather J.P., Sato G.H. (1979) The use of hormone-supplemented serum-free media in primary cultures. Exp Cell Res 124: 215–221. Cheng C.Y., Mather J.P., Byer A.L., Bardin C.W. (1986) Identification of hormonally responsive proteins in primary Sertoli cell culture medium by anion-exchange high performance liquid chromatography. Endocrinology 118: 480–488. Mruk D.D., Siu M.K.Y., Conway A.M., Lee N.P.Y., Lau A.S.N., Cheng C.Y. (2003) Role of tissue inhibitor of metalloproteases-1 in junction dynamics in the testis. J Androl 24: 510–523. Galdieri M., Ziparo E., Palombi F., Russo M.A., Stefanini M. (1981) Pure Sertoli cell cultures: a new model for the study of somaticgerm cell interactions. J Androl 5: 249–259. Lee N.P.Y., Mruk D.D., Lee W.M., Cheng C.Y. (2003) Is the cadherin/catenin complex a functional unit of cell-cell-actin-based adherens junctions (AJ) in the rat testis? Biol Reprod 68: 489–508. Lee N.P.Y., Mruk D.D., Conway A.M., Cheng C.Y. (2004) Zyxin, axin, and Wiskott-Aldrich syndrome protein are adaptors that link the cadherin/catenin protein complex to the cytoskeleton at adherens junctions in the seminiferous epithelium of the rat testis. J Androl 25: 200–215. Orth J.M. (1982) Proliferation of Sertoli cells in fetal and postnatal rats: A quantitative autoradiographic study. Anat Rec 203: 485–492. Li J.C.H., Lee W.M., Mruk D.D., Cheng C.Y. (2001) Regulation of Sertoli cell myotubularin (rMTM) expression by germ cells in vitro. J Androl 22: 266–277. Lui W.Y., Lee W.M., Cheng C.Y. (2003) Transforming growth factor-b3 regulates the dynamics of Sertoli cell tight junctions via the p38 mitogen-activated protein kinase pathway. Biol Reprod 68: 1597–1612. Wright W.W., Zabludoff S.D., EricksonLawrence M., Karzai A.W. (1989) Germ cellSertoli cell interactions. Studies of cyclic protein-2 in the seminiferous tubule. Ann N Y Acad Sci 564: 173–185. Wong E.W.P., Mruk D.D., Cheng C.Y. (2008) Biology and regulation of ectoplasmic specialization, an atypical adherens junction type, in the testis. Biochem Biophys Acta 1778: 692–708.
252
D.D. Mruk and C.Y. Cheng
40. Chung N.P.Y., Mruk D.D., Mo M.Y., Lee W.M., Cheng C.Y. (2001) A 22-amino acid synthetic peptide corresponding to the second extracellular loop of rat occludin perturbs the blood-testis barrier and disrupts spermatogenesis reversibly in vivo. Biol Reprod 65: 1340–1351. 41. Wong C.C.S., Chung S.S.W., Grima J., Zhu L.J., Mruk D.D., Lee W.M., Cheng C.Y. (2000) Changes in the expression of junctional and nonjunctional complex component genes when inter-Sertoli tight junctions are formed in vitro. J Androl 21: 227–237. 42. Mruk D.D., Zhu L.J., Silvestrini B., Lee W.M., Cheng C.Y. (1997) Interactions of proteases and protease inhibitors in Sertoligerm cell cocultures preceding the formation of specialized Sertoli-germ cell junctions in vitro. J Androl 18: 612–622. 43. Lee N.P.Y., Cheng C.Y. (2003) Regulation of Sertoli cell tight junction dynamics in the rat testis via the nitric oxide synthase/soluble guanylate cyclase/3¢,5¢-cyclic guanosine monophosphate/protein kinase G signaling pathway: an in vitro study. Endocrinology 144: 3114–3129.
44. Lie P.P.Y., Chan A.Y.N., Mruk D.D., Lee W.M., Cheng C.Y. (2010) Restricted Arp3 expression in the testis prevents blood-testis barrier disruption during junction restructuring at spermatogenesis. Proc Natl Acad Sci USA 107: 11411–11416. 45. Wong E.W.P., Mruk D.D., Lee W.M., Cheng C.Y. (2010) Regulation of blood-testis barrier dynamics by TGF-b3 is a Cdc42-dependent protein trafficking event. Proc Natl Acad Sci USA 107: 11399–11404. 46. Wong E.W.P., Sun S., Li M.W.M., Lee W.M., Cheng C.Y. (2009) 14-3-3 protein regulates cell adhesion in the seminiferous epithelium of rat testes. Endocrinology 150: 4713–4723. 47. Lui W.Y., Lee W.M., Cheng C.Y. (2001) Transforming growth factor-b3 perturbs the inter-Sertoli tight junction permeability barrier in vitro possibly mediated via its effects on occludin, zonula occludens-1, and claudin-11. Endocrinology 142: 1865–1877. 48. Lui W.Y., Wong C.H., Mruk D.D., Cheng C.Y. (2003) TGF-b3 regulates the blood-testis barrier dynamics via the p38 mitogen activated protein (MAP) kinase pathway: an in vivo study. Endocrinology 144: 1139–1142.
Chapter 17 Analysis of Endothelial Barrier Function In Vitro Yuping Wang and J. Steven Alexander Abstract Increased microvascular solute permeability underlies many forms of pathophysiological conditions, including inflammation. Endothelial monolayer cultures provide an excellent model system which allows systemic and mechanistic study of endothelial barrier function and paracellular permeability in vitro. The endothelial-specific complexus adherens junction protein VE-cadherin and their intracellular complex form pericellular structures along the cell borders which are critical to regulate endothelial barrier function by controlling pericellular permeability of vasculature. Here, we describe methods for both visualizing and quantifying junctional permeability and barrier changes in endothelial monolayers in vitro. Key words: Endothelial permeability, Barrier function, VE-cadherin, TEER, Microcarrier beads, Cell-column chromatography
1. Introduction The vascular endothelium is a semipermeable barrier, which creates and separates the tissue and vascular compartments. Endothelial monolayers regulate the exchange of fluid and solutes between the blood and the interstitial space. The integrity of vascular endothelial junctions is perhaps the most important structural feature controlling bulk exchange; therefore, in vitro endothelial cultures provide a simple and powerful model of in vivo microvascular barrier that has been demonstrated to represent many of the inflammatory responses associated with clinical disease states. Increased microvascular solute permeability underlies the pathophysiology of many acute and chronic forms of inflammation, wound healing, and immune responses. In this chapter, we describe several well-characterized in vitro techniques commonly used to study endothelial permeability/barrier function. Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_17, © Springer Science+Business Media, LLC 2011
253
254
Y. Wang and J.S. Alexander
2. Materials 2.1. Immuno fluorescent Staining for VE-Cadherin
1. Vascular endothelial cells from different anatomic and species origins may be used in these experiments. 2. Medium can be either endothelial cell growth medium (EGM, Lonza Walkersville, Inc. Walkersville, MD), Medium199 supplemented with l-glutamine, 20% fetal bovine serum (FBS), 1% penicillin, streptomycin and amphotericin (Sigma, St. Louis, MO), or appropriate medium for the cell type being studied. 3. Dulbecco’s Phosphate Buffered Saline (DPBS) and Hank’s Balanced Salt Solution (HBSS) (Sigma). 4. Fibronectin is dissolved in HBSS at 25 mg/ml and is used to coat all cell-culture surfaces. Fibronectin is supplied by Biomedical Technologies Inc. (BTI, Boston, MA) at a stock concentration of 1 mg/vial. 5. Staining/incubation/wash buffer (0.1% evaporated milk in DPBS). 6. Full strength trypsin-EDTA from Cellgro, Mediatech Inc. (Manassas, VA) – typically diluted 1:5 with PBS for final working solution. 7. Glass coverslips (12 mm diameter and 1.0 mm thickness) (Fisher, Rochester, NY). 8. Corning Costar 24-well cluster cell culture plates, Petri dish (100 mm) (Fisher). 9. 95% ethanol (Fisher). 10. Primary antibody: monoclonal mouse anti-human VE-cadherin antibody (Immunotech, Westbrook, ME). 11. Cy3 donkey anti-mouse immunoglobulin (IgG) (H + L) fluorescent-labeled antibody (Jackson Immunoresearch Laboratories Inc. Westgrove, PA). 12. Coverslip forceps/tweezers. 13. Fluorescent/confocal microscope.
2.2. Endothelial Electric Resistance Measurement
1. See Subheading 2.1, items 1, 2, 4, and 8 for endothelial cells, cell growth medium, fibronectin, and 24-well cluster cell culture plates. 2. BD Falcon polycarbonate transwell inserts (8 mm pore size for 24 well/plate) are from VWR International (West Chester, PA). 3. Endohm EVOM endothelial ohmmeter (World Precision Instrument, Sarasota, FL).
17 Analysis of Endothelial Barrier Function In Vitro
2.3. Horseradish Permeability Assay
255
1. See Subheading 2.1, items 1, 2, 4, and 8 for endothelial cells, cell growth medium, fibronectin, and 24-well cluster cell culture plates. 2. See subheading 2.2 for polycarbonate transwell insert. 3. Horseradish peroxidase (HRP, VI-A type, 44,000 MW, Sigma): standard curve for HRP is prepared in a range of 0–50 mM in distilled/deionized H2O. 4. Sodium phosphate buffer (NaH2PO4 50 mM), guaiacol, and hydrogen peroxide (H2O2) (Sigma). 5. Glass tubes (Borosilicate disposable culture tube, 12 × 75 mm, Fisher). 6. Semimicro disposable plastic cuvettes (10 mm light path, VWR International). 7. Spectrophotometer.
2.4. Cell Column Chromatography
1. See Subheading 2.1, items 1 and 2 for endothelial cells and cell growth medium. 2. Cytodex-3 microcarrier beads (Sigma) in Ca2+/Mg2+ free PBS. 3. 0.5 L Bellco stir flasks and flask stirring device (Vineland, NJ). 4. 30 min graduations on: off 120VAC timer. 5. Column-injectate tracer mixture: 2% blue dextran (MW 2,000 kDa), 0.2% cyanocobalamin (Vit B12, MW 1.355 kDa), and 0.0125% sodium fluorescein (MW 0.376 kDa) in HBSS (see Note 1). 6. Gilson minipuls peristaltic pump (0.9–3 ml/min) (Gilson, Middleton, WI). 7. Omni water jacked 6.6 mm ID glass column (Omni, Minneapolis, MN) fitted with a 70 mm frit to restrict beads. 8. 50 ml in-line injector port (Rainin, Oakland, CA). 9. Gilson fraction collector (Gilson). 10. Flat-bottom 96-well plates (Nunc, Rochester, NY).
2.5. Trypan Blue Albumin Permeability Analysis
1. See Subheading 2.1, items 1 and 2 for endothelial cells, cell growth medium. 2. See Subheading 2.4, items 2–4 for Cytodex-3 microcarrier bead, Bellco stirring flasks, and on:off control timer. 3. Trypan blue (Sigma), bovine serum albumin (BSA), Cohn Frac V (fatty acid free) (Sigma). 4. Dibutyl phthalate and dioctyl phthalate (1:1 mixture) (Sigma) (see Note 2). 5. Narrow format stiletto plastic tubes (1 ml) (Fisher). 6. Microcentrifuge (Fisher). 7. 96-well plate reader.
256
Y. Wang and J.S. Alexander
3. Methods 3.1. Immunofluore scent Staining for VE-Cadherin
VE-cadherin is an endothelial complexus adherens-specific junction adhesion molecule located at interendothelial cell junctions. VE-cadherin plays a fundamental role in maintaining endothelial integrity, barrier function, and leukocyte extravasation. VE-cadherin also regulates cell proliferation, apoptosis, and modulates VEGF-receptor signaling. Loss of VE-cadherin expression from/or disorganized VE-cadherin distribution at cell junctions directly influences monolayer integrity and endothelial permeability. Consequently, immunostaining for VE-cadherin in endothelial cells is used intensively in endothelial biology to provide structural correlates to barrier changes. The method described below can be used to visualize VE-cadherin distribution in endothelial cells grown on glass coverslips in 24-well/plates in response to barrier modulators. 1. When endothelial cultures are near confluency, passage cells into 24-well/plates with glass coverslips preloaded in the plates (1). 2. To sterilize glass coverslips, put coverslips into a Petri dish with 95% ethanol (30 min). Transfer coverslips to a new sterile Petri dish and air dry coverslips under UV light in a biosafety cabinet; place one coverslip into each well of the 24-well/plate. 3. Coat coverslips with 100 ml of fibronectin at 25 mg/ml for 5 min (see Note 3). 4. Trypsinize ECs with Trypsin-EDTA, and resuspend cells in desired amount of endothelial growth medium. Seed the cells (1 × 105 cells/well) onto fibronectin-coated coverslips. 5. Place the plate in a 37°C incubator with 100% humidity and 5% CO2/air. Replace medium every other day. It usually takes 3–4 days for cells to reach confluency. 6. On the day of experiment, treat cells with appropriate stimuli for experiments. 7. Fix cells with 1 ml of cold (−20°C) 95% ethanol for 30 min in 4°C refrigerator (see Note 4). Discard ethanol and add 0.5 ml of 100% acetone to permeabilize cells for 1 min, discard acetone, and air dry coverslips for 3 min. 8. Incubate coverslips with 250 ml of primary anti-VE-cadherin antibody with a concentration of 1 mg/ml in DPBS with 0.1% milk for 2 h at RT. 9. Discard primary antibody solution and wash coverslips 3× with DPBS containing 0.1% milk. 10. Incubate coverslips with 250 ml of Cy3 donkey anti-mouse immunoglobulin G (H + L) (secondary antibody) at a dilution of ~1:200 for 2 h at RT.
17 Analysis of Endothelial Barrier Function In Vitro
257
Fig. 1. VE-cadherin staining in endothelial cells (ECs) grown on glass coverslips. (a) Control ECs; (b) cells treated with platelet-activating factor; and (c) cells after 24 h recovery. In control cells, VE-cadherin is only stained at cell borders, exhibits zigzag continuous expression pattern (a). In contrast, intercellular gaps are present at cell contact regions in ECs treated with PAF (b). Disorganized VE-cadherin distribution in endothelial cells is recovered after PAF was removed from the culture indicating that microenvironment condition is critical to maintain junction protein VE-cadherin expression and distribution and EC integrity.
11. Discard secondary antibody solution and wash coverslips 3× with DPBS containing 0.1% milk. 12. Move coverslips from the plate using coverslip forceps. Dry coverslips with cell side face up and mount coverslips with Vectashield mounting medium with cell side down on glass slide (see Note 5). 13. View stained cells by fluorescent microscopy. An example of VE-cadherin staining is shown in Fig. 1. The distribution of VE-cadherin is a continuous and zigzag pattern localized at the cell-contact region (cell junction). 3.2. Trans-Endothelial Electrical Resistance Measurements
Measurement of endothelial electrical resistance (ER) is a simple and direct method to study endothelial barrier function in vitro. The basic concept is that when cells grow to confluency and become tightly connected, they resist the passage of electrical current, and a property that can be easily measured using an Endohm EVOM endothelial ohmmeter. Trans-endothelial electrical resistance (TEER) measurement provides a quick and realtime evaluation of the integrity of endothelial monolayers. The method described below is intended specifically for measuring TEERs with endothelial cells grown in a 24-well cell inserts. 1. Place polycarbonate inserts in 24-well cluster plate and coat inserts with 0.2 ml of fibronectin (25 mg/ml) per insert. 2. Trypsinize ECs with Trypsin-EDTA, resuspend cells with endothelial growth medium with density of 1 × 105 cells/ml; seed 0.5 ml of cell suspension (5 × 104 cells/per insert) into an insert (upper chamber), and add 1.0 ml of medium per well (lower chamber). 3. Place the plate in a 37°C incubator with 100% humidity and 5% CO2/air. Change medium the next day. Cells usually reach
258
Y. Wang and J.S. Alexander
confluency in 3 days. Examine cells under microscope to confirm cell confluency by phase contrast. 4. On the day of experiment, remove medium from both upper and lower chambers, add 0.8 ml of prewarmed (37°C) fresh medium to the upper chamber and 1.0 ml of medium to the lower chamber (see Note 6). 5. Place the plate in a 37°C incubator for 1 h and let the cells to equilibrate and then measure the baseline electrical resistance with Endohm EVOM endothelial ohmmeter. In general, confluent endothelial cells grown in 8 mm pore size have a resistance reading between 400 and 420 W (see Note 7). 6. Add stimuli to the upper chamber. Measure electrical resistance (in triplicate) per well and record values at 1, 2, and 3 h, etc. The mean value is then used for final data calculation. 7. For TEER calculations, the values obtained from a blank insert (no cells) is subtracted to give the net resistance. This is then multiplied by the insert membrane area to give the resistance in area-corrected units and is expressed as W cm2, taking into account the surface area of the filter (see Note 8).
⋅
3.3. Horseradish Peroxidase Permeability Assay
There are a number of reagents that have been described as markers/tracers used for evaluating endothelial permeability, such as I125-labeled BSA, fluorescein-isothiocyanate (FITC)-conjugated dextran, 3H-mannitol, 14C-labeled dextran, and 14C-inulin, etc. Here, we describe a simple and sensitive method using HRP as a tracer to measure endothelial permeability. HRP activity can be inhibited by sodium azide (2). The enzymatic reaction can be initiated by adding hydrogen peroxide (H2O2) and a substrate (e.g., guiacol) in the testing samples. The changes of HRP activity in the samples can then be easily monitored by a UV spectrophotometer in vitro (2, 3) (see Note 9). The assay is carried out using 8.0 mm pore size polycarbonate cell culture inserts that fit in 24-well cluster cell culture plate (see Note 10). All reagents should be made fresh prior to the assay. 1. See steps 1–3 of Subheading 3.2 above for the preparation of fibronectin-coated inserts, passing cells, and culture methods. Endothelial electric resistance will help to determine the level of confluence (see Note 6). 2. On the day of the experiment, remove medium from both upper and lower chambers. Add 0.8 ml of prewarmed (37°C) medium containing HRP (0.126 mM) and permeability treatments to the upper chambers and 1.0 ml of medium in the lower chambers. 3. Place the plate in an incubator and at 1, 2, 3, and 6 h intervals, collect a 50 ml aliquot of medium from the lower chamber.
17 Analysis of Endothelial Barrier Function In Vitro
259
Store samples in borosilicate disposable tubes, and keep on ice until assaying. 4. Prepare a standard curve in the range of 0–50 mM in distilled/deionized H2O. 5. To measure HRP activity, add 25 ml of collected samples or standard to 860 ml of reaction buffer containing 50 mM NaHPO4 and 5 mM guaiacol in 2 ml glass tubes. Vortex samples. HRP standard curve should be 0, 0.5, 1.0, 2.5, 5.0, 10, and 25 mM. All samples should be assayed in duplicate. 6. Start the reaction by addition of 100 ml hydrogen peroxide reaction solution (0.6 mM in distilled/deionized H2O) to the sample and standard tubes and incubate the tubes at RT for 30 min (see Note 11). 7. After reaction, transfer samples to semimicro disposable cuvettes and read the sample absorbance at 470 nm by spectrophotometer, and record the OD number for data calculation. 8. Data are calculated as follows: DOD470 = sample OD470 − blank OD470. Data are expressed as ∆OD470 for permeation of HRP across transwell filters. Figure 2 shows endothelial permeability data of cells treated with platelet activating factor (PAF) simultaneously measured by both TEER and HRP permeability assays [4).
Electrical Resistance (Ω •cm2)
50 40
*
30
*
20
**
10 0
HRP Leakage (∆470nm)
0.20
**
0.15
**
0.10 0.05 0 0
2
5
10
PAF (µg/ml)
Fig. 2. Endothelial permeability simultaneously measured by TEER and HRP permeability assays in endothelial cells treated with platelet activating factor (PAF). Please note that reduced endothelial electrical resistance is correlated with increased HRP leakage in cells grown on cell culture inserts when cells are treated with PAF.
260
Y. Wang and J.S. Alexander
3.4. Cell-Column Chromatography Permeability Assay
Cell-column chromatography is a sensitive and high throughout method for evaluating endothelial monolayer permeability of cells grown on microcarriers (beads) to small molecular weight solutes using tracers, including blue dextran, cyanocobalamin, and sodium fluorescein (5) (see Note 12). The advantages of microcarrier cultures include exposure of cultured cells to continuous fluid shear, which may more closely mimic an in vivo environment. Secondly, defects of endothelial monolayers on individual microcarriers can be negligible on the total response. Importantly, the duration of each round of permeability analysis can be completed in ~3 min. Therefore, multiple tests can be performed simultaneously within the course of the experiment. Finally, repetitive testing on single batches of cells allows for powerful statistical approaches, e.g., repeated measures analysis of variance to be used. Endothelial cells derived from different organs have been cultured to confluence on microcarrier beads to study endothelial barrier function/permeability using this cell-column chromatography permeability assay (5–9). 1. Cytodex-3 microcarriers are cultured with endothelial cells by mixing 3 ml of gelatin-covered microcarrier beads with 50 ml of cell-culture medium. Cells, beads, and medium are placed in a BellcoTM stir flask at an endothelial cell density of 1:4 (assuming that each ml of Cytodex-3 beads has a total surface area of 328 cm2). 2. Cultures are stirred at 60 rpm intermittently (30:30 min, on:off) for 24 h. Medium is then adjusted to 100 ml, and then stirred continuously at 60 rpm. 3. Replace 50% of culture medium every 3 days. Cell confluence is examined visually by phase-microscopic inspection of microcarrier beads. 4. On the day of assay, cell-covered beads are transferred into a column volume of 0.67 cm3, which provides ~130 cm2 of endothelial cell culture surface (~6.5 × 106 cells). 5. Cells are washed and equilibrated with HBSS containing 0.5% BSA and 10 mM HEPES (pH 7.4). Constant pressure perfusion is maintained by a Gilson peristaltic pump at 0.9 ml/min (this rate was chosen to match the spontaneous column flow rate induced by gravity). 6. At time points, permeability measurements are made by injecting a 50 ml bolus of mixed tracers (blue dextran/sodium fluorescein/cyanocobalamin, see Subheading 2.4 item 5) by an in-line rotary 50 ml injection loop. The cell column and all perfusate solutions are maintained at 37°C. 7. A Gilson fraction collector (model 203) equipped with a drop counter is used to collect elutes into 96-well/plate: two drops of elutes per well in the initial 24 wells and then six drops per
17 Analysis of Endothelial Barrier Function In Vitro
261
well for the next 22 wells (see Note 13). A total of 46 elutes is collected per cell column. This approach not only provides higher resolution of tracers in the early (steep) portion of elution curves, but also collects enough samples to recover ~100% of all applied tracers. 8. Columns can then be washed out with HBSS containing 0.5% BSA and 10 mM HEPES (pH 7.4) for at least 5 min. 9. Repeat steps 6–8 for replicate testing of each column (see Note 14). 10. Read the plate at 620 nm for blue dextran, 540 nm for cyanocobalamin, and 492 nm for sodium fluorescein on a spectrophotometric plate reader and store the data on a computer for later analysis. Figure 3 shows an example of typical profiles of blue dextran, cyanocobalamin, and sodium fluorescein elution profiles under normal (a) and increased permeability (b) (reduced barrier) conditions and scanning microscopy of endothelial cells grown on microbeads (c and d).
Fig. 3. A representative data from cell-column chromatography permeability assay. (a) and (b) show typical profiles of blue dextran, cyanocobalamin, and sodium fluorescein eluate under normal (a) and increased permeability (b) conditions. (c) and (d) show scanning microscopy of endothelial cells reach confluence grown on microcarrier (beads).
262
Y. Wang and J.S. Alexander
3.5. Trypan-Blue Dye Extraction Permeability Assay
The extraction of trypan blue-albumin complexes by cell-covered microcarriers is another simple method for the measurement of in vitro barrier function. The original approach (10) has been modified by several laboratories (11, 12). It is estimated a stoichiometric maximal binding of 37 trypan blue molecules per albumin, and that >95% of trypan blue is bound to albumin as a 106 kDa complex (11). Thus, cell-covered microcarrier beads absorb trypan blue-albumin dye complexes proportionate with endothelial monolayer barrier properties. 1. Microcarrier cell cultures are accomplished as described in Subheading 3.4 (above). 2. Bellco stir flasks containing cell-covered microcarrier are removed from the incubator and allowed to settle. 3. An aliquot of 300 ml confluent cell-covered microcarrier beads in 2 ml of culture medium are transferred to a 3 ml plastic capped Eppendorf tube. Beads will fall to the tube bottom by gravity in 15 s. 4. Culture medium is replaced with test medium; alternatively, test agents can be added (“spiked”) into medium. 5. Tubes containing microcarrier beads are then continuously and gently agitated at 37°C. An aliquot of 200 ml of trypan blue/albumin complex (TBA: 0.4% trypan blue, 0.5% BSA in PBS) is then added to each tube, and the tubes are recapped. Trypan blue dye absorption by microcarrier beads progressively extracts TBA from the reaction mix. Maximal dye uptake is usually achieved within 60 min. If monolayer barrier function is decreased, the rate of dye uptake is increased, approaching that of naked (non-cell covered) microcarriers. 6. At time points, the tubes are rapidly opened and a 200 ml aliquot of mixed bead/dye is removed and centrifuged on a cushion of 400 ml of 1:1 dibutyl:dioctyl phthalate in narrow format Eppendorf tube (“stiletto” tube) effectively separating the beads from the dye in solution to end uptake of tracer dye by beads at each time point by centrifugation at 10,000 × g for 30 s (see Note 14). 7. The supernatant containing TBA is then diluted tenfold, and the absorbance of the sample is read at 620 nm by a spectrophotometer. 8. Trypan blue dye uptake permeability results are expressed as the inverse of the diluted supernatant absorbance (1/Abs vs. time). Appropriate controls include non-cell covered microcarriers that provide a measurement of maximal rate of dye uptake by microcarriers.
17 Analysis of Endothelial Barrier Function In Vitro
263
4. Notes 1. Do not filter sterilize column injectate. 2. Caution: Dibutyl: dioctyl phthalate is toxic! Avoid skin contact! 3. Fibronectin can be recovered and reused at least twice. 4. Fixed cells/coverslips can be stored in 95% ethanol at 4°C overnight and stained next day. 5. Use mounting medium with DAPI counterstain to visualize nuclear DNA in cells. 6. We usually perform endothelial electrical resistance measurements and HRP permeability assay simultaneously. 7. We do not recommend performing experiments with an initial baseline resistance reading less than 400 W. 8. Check the insert membrane area before calculation. The insert membrane area could be different if inserts are from different suppliers. 9. The advantages of using HRP over other markers/tracers (I125-labeled BSA, 14C-labeled dextran, 14C-inulin, and FITCconjugate dextran) are that no radio-labeled materials are involved in the assay and a visible range spectrophotometer is more common than fluorescent spectrometers. 10. We recommend using transparent inserts that allow visualization of live cells by light microscopy. Inserts with 3 or 0.4 mm pore size can also be used based on experimental needs. 11. H2O2 reaction solution must be prepared just before the assay. 12. To determine permeability in this model (5), it is assumed that elution profiles of permeant tracers (sodium fluorescein, cyanocobalamin) depend on the properties of the moving, fluid phase of the column, the endothelial junctional (pericellular) permeability properties of the endothelial monolayer and the diffusion characteristics of these permeant tracers within the microcarrier beads’ matrix. In contrast to the permeant tracers, the elution of blue dextran reflects only on the movement of the fluid phase through the column. A modified Marquardt iteration scheme is used to estimate monolayer permeability that best fits the experimental data. This “best fit” is determined by the minimization of the coefficient of variation between a computer-generated prediction of the permeant tracer’s elution profile and the experimentally observed elution profile.
264
Y. Wang and J.S. Alexander
13. The order of tracer elutes is based on the molecular weight of the tracers: (1) blue dextran with molecular weight of 2,000 kDa cannot penetrate the bead matrix. Therefore, it follows the fluid phase and is considered flow tracer; (2) sodium fluorescein with molecular weight of 0.376 kDa and (3) cyanocobalamin with molecular weight of 1.355 kDa can permeate the cell layer and diffuse into the bead matrix beneath the cells and exhibit delayed exit from the column if increased endothelial permeability occurs which is proportionate to monolayer barrier properties. 14. We recommend testing each sample in triplicate.
Acknowledgment We thank Loren Hoffmann at Vanderbilt University for the technical help of the endothelial imaging on microbeads by scanning electrical microscopy. This work is supported, in part, by the grant from National Heart Lung Blood Institute (NHLBI, HL65997). References 1. Wang Y, Gu Y, Granger DN, Roberts JM, Alexander JS. (2002) Endothelial junctional protein redistribution and increased monolayer permeability in HUVECs isolated during preeclampsia. Am J Obstet Gynecol 186: 214–220. 2. Ortiz de Montellano PR, David SK, Ator MA, Tew D. Mechanism-based inactivation of horseradish peroxidase by sodium azide. Formation of meso-Axidoprotoporphyrin IX. Biochemistry 27: 5470–5476, 1988. 3. Lampugnani MG, Resnati M, Raiteri M, Pigott R, Pisacane A, Houen G, Ruco LP, DeJana E. (1992) A novel endothelial-specific membrane protein is a marker of cell-cell contacts. J Cell Biol 118: 1511–1522. 4. Zhang Y, Gu Y, Lucas MJ, Wang Y. (2003) Antioxidant superoxide dismutase attenuates increased endothelial permeability induced by platelet-activating factor. J Soc Gynecol Investig 10: 5–10. 5. Haselton FR, Mueller SN, Howell RE, Levine EM, Fishman AP. (1989) Chromatographic demonstration of reversible changes in endothelial permeability. J Appl Physiol 67(5):2032–48. 6. Alexander JS, Hechtman HB, Shepro D. (1988) Phalloidin enhances endothelial barrier function and reduces inflammatory permeability in vitro. Microvasc Res 35(3):308–15.
7. Alexander JS, Blaschuk OW, Haselton FR. (1993) An N-cadherin-like protein contributes to solute barrier maintenance in cultured endothelium. J Cell Physiol 156(3):610–8. 8. Haselton FR, Dworska E, Evans SS, Hoffman LH, Alexander JS. (1996a) Modulation of retinal endothelial barrier in an in vitro model of the retinal microvasculature. Exp Eye Res 63(2):211–22. 9. Haselton FR, Woodall JH, Alexander JS. (1996b) Neutrophil-endothelial interactions in a cell-column model of the microvasculature: effects of fMLP. Microcirculation 3(3):329–42. 10. Boiadjieva S, Hallberg C, Högström M, Busch C. (1984) Methods in laboratory investigation. Exclusion of trypan blue from microcarriers by endothelial cells: an in vitro barrier function test. Lab Invest 50(2):239–46. 11. Bottaro D, Shepro D, Peterson S, Hechtman HB. (1986) Serotonin, norepinephrine, and histamine mediation of endothelial cell barrier function in vitro. J Cell Physiol 128(2):189–94. 12. Alexander JS, Patton WF, Christman BW, Cuiper LL, Haselton FR. (1998) Plateletderived lysophosphatidic acid decreases endothelial permeability in vitro. Am J Physiol 274(1 Pt 2):H115–22.
Chapter 18 Role of Endothelial Cell–Cell Junctions in Endothelial Permeability Armelle Le Guelte and Julie Gavard Abstract The endothelial barrier separates the inner blood compartment from the surrounding tissues. At the molecular level, adhesion molecules accumulate at the endothelial cell–cell junction and contribute to maintain vascular integrity. An increase in the endothelial permeability is frequently associated with the deregulation of junctional adhesion. Here, we review how to evaluate the in vitro functions of endothelial cell–cell contacts. We focus this chapter on cell imagery and biochemical analysis of VE-cadherin, the main constituent of adherens junction, and we also provide description of endothelial cell models and methods for studying tight junctions. Key word: Cadherin, Cell–cell contact, Adhesion, Microscopy, Internalization, Catenin, Immunoprecipitation, Phosphorylation
1. Introduction 1.1. Definition of the Endothelial Barrier
Endothelial cells (ECs), that formed the inner side of the vascular wall, control the infiltration of blood proteins and circulating cells to the underlying tissues within all organs. This endothelial barrier tightly regulates the passage of fluids, cells, and nutrients. Two pathways are known to mediate the exchange between the blood compartment and the irrigated tissues. Molecules and cells can pass through (transcellular) or between (paracellular) ECs. Transcellular passage requires either cell fenestration or a complex system of trafficking vesicles, called vesiculo-vacuolar organelles. The paracellular pathway by contrast demands the coordinated opening and closure of endothelial cell–cell junctions. This function must therefore be strictly coordinate to maintain vascular integrity.
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_18, © Springer Science+Business Media, LLC 2011
265
266
A. Le Guelte and J. Gavard
The formation of adhesive structures between adjacent cells, including adherens junctions (AJ) and tight junctions (TJ), contribute to cell polarity, differentiation and survival, and is ultimately required for tissue cohesion. Cadherins, the main constituent of AJ, belong to a conserved family of adhesion molecules that provide a molecular bond between cells (1). TJ form a dense ultrastructural organization, observable by electron microscopy, and involves several adhesive molecules, including occludin, junctional adhesion molecules (JAMs), and the claudin family of tetraspan transmembrane proteins (2). Whereas in epithelial cells, TJ are often located apically with respect to AJ, in ECs, both junctions are mixed throughout cell–cell contacts (3). The formation, maintenance, and remodeling of the intercellular contacts imply a functional interaction between these two structures and are suspected to assume both the endothelial barrier properties (4). 1.2. Physiological and Pathological Implications
The endothelial barrier function contributes to normal angiogenesis, blood pressure control, and immune responses (4). Angiogenesis is a tightly controlled sequential and multistep process, which mainly includes proliferation, migration and tubulogenesis of ECs, and finally maturation of vessels, allowing blood circulation. Furthermore, quiescent pre-existing vessels can become leaky, dilated, and destabilized upon exposure to various angiogenic factors such as VEGF (vascular endothelial growth factor), FGF (fibroblast growth factor), and angiopoietins (4, 5). This leads to the extravasation of plasma proteins, allowing the formation of a provisional matrix of fibrin, which in turn can facilitate EC migration and vascular remodeling. Such leaky features are intimately linked to the activity of the VE-cadherin adhesion molecule (4). Interestingly, deficiency of VE-cadherin prevented EC organization in vessel-like patterns in a model of embryoid bodies (6). More importantly, VE-cadherin knockout mice are embryonic lethal and exhibit multiple severe angiogenic defects (7). Such mice died at E9.5 due to vascular insufficiency. Moreover, many pathological conditions and human diseases exhibit an abnormal increase in vascular permeability, such as tumor-induced angiogenesis and metastasis, inflammation, macular degeneration, allergy, and brain stroke (8). For instance, a major cellular functional consequence of tumor-induced angiogenesis is the disruption of VE-cadherin-mediated endothelial cell–cell contacts and opening of the endothelial barrier (8). As a result, the vessels formed are irregular, leaky, and tortuous which is due, in part, to weakened VE-cadherin-mediated endothelial interactions. Moreover, tumor-induced angiogenesis increases the likelihood of metastasis, since tumor vascularization also provides a circulatory option for tumor cells to disseminate throughout the body (9). While tumor cells may passively enter the bloodstream, the majority of them actively breakdown the extracellular matrix
18 Role of Endothelial Cell–Cell Junctions in Endothelial Permeability
267
through protease secretion and subsequently penetrate the endothelial barrier. This has been exploited for the development of anti-angiogenic therapies. Indeed, two VE-cadherin antibodies have been shown to disrupt endothelial junctions in growing vessels without affecting the normal vasculature (10). Furthermore, the signaling mechanism leading to VE-cadherin destabilization and vascular permeability increase has been experimentally hampered in mice to reduce metastasis formation (9). 1.3. Molecular Mechanisms
Knock-out mice from VE-cadherin and claudin-5 exhibit serious defects of the vascular integrity (7, 11). Nonetheless, VE-cadherin function has been shown to directly control claudin-5 expression (12), suggesting that VE-cadherin adhesion might act upstream claudin-5 and TJ in the control of the endothelial barrier integrity. Of note, VEGF, first identified as a vascular permeability factor secreted by tumor cells, triggers the hierarchical activation of the Src tyrosine kinase, Vav2, a guanine-nucleotide exchange factor for the small Rho GTPases, Rac and its downstream effector PAK, the p21-activated kinase (13–16). This VEGF-initiated signaling axis culminates at the phosphorylation of VE-cadherin, on a highly conserved serine motif, which directs VE-cadherin to endocytosis, thereby removing VE-cadherin from the plasma membrane and subsequently disrupting the endothelial barrier (14). Based on these findings, it has been investigated whether antipermeability factors, such as Angiopoietin1 might impede on the phospho-serine-dependent internalization of VE-cadherin. Angiopoietin1 elicits a signaling pathway through Tie2, its cognate tyrosine kinase receptor, the small GTPase RhoA and mDia, one of RhoA downstream targets. mDia can in turn compete for Src activation by VEGF-R2, therefore halting the VEGF signaling to VE-cadherin internalization (17). Similar signaling mechanisms have later been demonstrated in response to Robo-4 to oppose VEGF-induced vascular permeability (18). Moreover, the inhibition of FGF signaling resulted in dissociation of the VE-cadherin/p120-catenin complex and disassembly of both AJ and TJ, which promoted severe impairment of the endothelial barrier function and dismantlement of the vasculature (19). Ultimately, the biochemical route by which VEGF, angiogenic factors, and oncogenes modulate VE-cadherin, cell–cell junctions, and vascular integrity may help identify new therapeutic targets for the treatment of many human pathological conditions that exhibit aberrant vascular leakage. Several options are available to explore the endothelial barrier function in vitro and the molecular mechanisms involved; including determination of the biochemical and structural organization of endothelial cell–cell junction. These approaches are, however, limited by the endothelial cell models themselves. Here, we
268
A. Le Guelte and J. Gavard
attempt to review the availability of endothelial model and the in vitro culture conditions, and, how to analyze the architecture of the intercellular junctions.
2. Materials 2.1. Cell Culture
Dulbecco’s Modified Eagle’s Medium + 4,5 g/L pyruvate (Invitrogen). Penicillin/streptomycin (Invitrogen). Trypsin–EDTA (Invitrogen). Collagenase (Invitrogen). PBS (Invitrogen). Endothelial Basal Medium EBM-2 (Lonza). Gold Fetal Bovine Serum (PAA Lab). Rat tail collagen I (RD Systems). Hydrocortisone (Sigma). FGF (Sigma). Chemically defined lipid concentrate (Invitrogen). MACS microbeads (Miltenyi).
2.2. Antibodies
VE-cadherin (Santa Cruz). Claudin-5 (Santa Cruz). VE-cadherin BV6 (Millipore). Phospho-tyrosine 4G10 (Millipore). Tie 2 (Santa Cruz). Alpha-catenin (Sigma). Beta-catenin (Sigma). p120 catenin (BD Transduction Laboratories). CD31 (BD Transduction Laboratories). Occludin (Zymed Laboratories, Invitrogen). ZO-1 (Zymed Laboratories, Invitrogen). Claudin-1(Zymed Laboratories, Invitrogen). Claudin-3 (Zymed Laboratories, Invitrogen).
2.3. Biochemical Analysis
MicroBCA kit (Pierce). Biotinylation kit (Pierce). Slurry sepharose G protein (Invitrogen). Western-blot equipments (Nupage, Invitrogen).
d-glucose +
18 Role of Endothelial Cell–Cell Junctions in Endothelial Permeability
2.4. Immunostaining
269
HistoGRIP (Invitrogen). 12-mm diameter glass coverslips (Thermo Scientific). AlexaFluor conjugated secondary antibodies (Invitrogen). Mounting medium (Vectashield). AlexaFluor conjugated phalloidin (Invitrogen). BSA fraction V (Sigma). Triton (Sigma). Formaldehyde (Sigma). Glutaraldehyde (Sigma).
3. Methods 3.1. The Vascular System
The vascular system is constituted of a highly organized and branched network of arteries, capillaries, and veins that penetrates throughout body tissues, enabling efficient exchange of oxygen and nutrients and removal of waste products. Endothelial cells (ECs) cover the internal surface of blood vessels and are therefore important to establish a barrier between sometimes hostile external environments and the internal milieu, while allowing an organblood exchange, with tissue specificity that are exemplified below. The blood–brain barrier (BBB) is formed by brain ECs lining the cerebral microvasculature, and is crucial for protecting the brain from fluctuations in plasma composition, and from circulating agents such as neurotransmitters and xenobiotics capable of disturbing neural function. This functional unit, often called the neurovascular unit, is composed of ECs with extensive tight junctions, astrocytes, neurons, and a contractile apparatus of either smooth muscle cells (SMCs) or pericytes. BBB cells differ from the other ECs that cover the rest of the body mainly by the presence of dispersed pinocytic vesicular structures and more TJ, associated with the absence of fenestrations (20). In the lung, the ECs form a semiselective barrier between circulating blood and interstitial fluid. These cells are dynamically regulated by a counterbalance between protective and disruptive barrier. At the molecular level, increased permeability of lung EC is associated with activation of MAP kinases and tyrosine kinases that control actomyosin rearrangement, activation of contraction, and destabilization of AJ and gap formation. Furthermore, studies by many groups proposed an important role of cytoskeleton remodeling and regulation of EC barrier by the small GTPases Rho and Rac (21, 22). In the kidney, glomeruli, formed by ECs, together with a basement membrane, and specialized epithelial cells (podocytes) play a key role of filtration from the blood to the urinary space. Of
270
A. Le Guelte and J. Gavard
note, fenestrations present in kidney ECs allow the passage of macromolecules. On the other side of the basement membrane, foot processes from podocytes are associated with cellular extensions (slit diaphragms) that retain albumin and other macromolecules. A core of mesangial cells, smooth muscle-like cell, corroborates the filtration barrier. Both ECs and podocytes are decisive to preserve the filtration unit. It is noteworthy that VEGF is expressed in glomerular podocytes, and may be necessary for glomerulus development and to maintain its function (23). The liver receives blood from both the portal vein and the hepatic artery. Hepatocytes, resident macrophages a.k.a. Kupffer cells, and ECs are the three main liver cell types. Besides assuming a barrier function that restricts the access of blood-borne compounds, liver ECs are functionally specialized. Indeed, they have complex roles, including not only receptor-mediated clearance of endotoxin, bacteria and other compounds, but also the regulation of inflammation, leukocyte recruitment, and host immune responses to pathogens (24). 3.2. Primary Cell and Cell Lines
Primary cells can be isolated by rigorous protocols but they are difficult to obtain and to maintain in culture (see Note 1). In a working model, Marelli-Berg and her colleagues performed a rapid and reproducible method for the isolation of murine ECs (25). 1. Tissues are removed, washed, minced in 2 mm2, and digested with collagenase. 2. The cellular digest is filtered to remove undigested blocks, washed in PBS, and incubated with trypsin–EDTA solution to obtain single cell suspension. 3. Cells are first precleared with mouse immunoglobulins then incubated with anti-mouse CD31. They could be affinity purified by anti-mouse CD105 and biotinylated isolectin B4 as well. 4. Cells are counted, incubated with anti-mouse Ig- and streptavidin-conjugated microbeads, loaded in column with magnetically labeled cell suspension, and retained cells are washed and wash off. 5. Specific cells are resuspended in EC growth medium and plated out. 6. After overnight culture, nonadherent cells are removed. 7. Confluence culture is then allowed, and cells need to be further characterized for EC markers, such as CD31, VE-cadherin, and Tie2.
18 Role of Endothelial Cell–Cell Junctions in Endothelial Permeability
3.3. Cell Culture Condition
271
Immortalized cells are cultivated in specialized growth medium (see Note 2). 1. Immortalized HUVECs (Eahy926) are cultured in DMEM medium + GlutaMAX, 10% of fetal bovine serum (FBS), 100 U/mL penicillin, and 100 mg/mL streptomycin. 2. The HCMEC/D3 cells are maintained in EBM-2 basal medium, 5% of FBS Gold, 1.4 mM of hydrocortisone, 5 mg/mL of ascorbic acid, 1/100 of chemically defined lipid concentrate, 10 mM of HEPES, 100 U/mL penicillin, 100 mg/ mL streptomycin, and 1 ng/mL of bFGF and grown at 37°C in a humidified atmosphere of 5% CO2. 3. The optimal used of HCMEC/D3 requires a thin rat collagen coating (150 mg/mL) on the splitting plastic culture plates. 4. The thick rat tail collagen coating (4 mg/mL) promotes differentiation and formation of a basal lamina.
3.4. Cell Imaging of Adherens Junctions
The architecture of AJ could be visualized by immunofluorescence for its components; including VE-cadherin and catenins (a, b and p120). 1. ECs have to be cultured on glass coverslips at high density for at least 3 days (75,000 cells/cm²) (see Note 3). 2. Once the confluence has been reached and ECs form a uniform monolayer, cells could be fixed for further immunostaining or electron microscopy study, with paraformaldehyde and glutaraldehyde saline solutions, respectively (see Note 4). 3. PBS wash in order to remove cell debris, serum, and media containing molecules that might interfere with fluorescence analysis such as phenol red. 4. Fixation in PBS + 4% paraformaldehyde for 15 min RT, followed by a PBS wash to stop the fixation process. 5. Permeabilization (optional) in PBS + 5% Triton-X100 for 5 min RT, followed by a PBS wash to remove the detergent (see Note 5). 6. Blocking step in PBS with either high quality BSA fraction V or 1% serum for 30 min RT. 7. Incubation with the primary antibody in humid chamber, the antibody has to be diluted in the blocking buffer for 1 h RT. Extensive PBS washes (for instance, 5 × 2 min) are then performed. 8. Incubation with the secondary antibody in a humid dark chamber, the antibody has to be diluted in the blocking buffer for 1 h RT. Extensive PBS washes (for instance, 5 × 2 min) are then performed. 9. Cells can then be counterstained with nuclei probes (DAPI, Hoescht, PI) or fluorescent-labeled actin probes (phalloidin).
272
A. Le Guelte and J. Gavard
10. Samples are mounted on glass observation slides adapted to the microscope stage and embedded in a mounting medium that protects fluorescence, cell structure and do not interfere with light passage (see Note 6). 3.5. Cell Imaging of Tight Junctions
TJ are located at the areas of cell–cell contact in the apical forming impermeable barrier. This junction is important to maintain cell polarity and barrier function in epithelial and endothelial cells. Four integral membrane proteins have been described in TJ: claudin, occludin, the JAM, and the coxsackie and adenoviral serotype 2/5 receptor (CAR). Claudins belong to a multigene family of tetraspan membrane proteins and have a crucial structural and functional role in TJ assembly. The extracellular loops of claudins interact with the other cell to close up the cellular layer and regulate paracellular transport, while the intercellular loops interact with adaptators, such as ZO-1 and ZO-2. Various mouse models knocked-out for claudins indicate an important role in the maintenance of tissue integrity (26). Of interest, TJ present a dense ultrastructural organization observable by freeze-fraction electron microscopy (27). 1. Confluent cells are fixed with 2.5% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4) overnight. 2. After washes in the cacodylate buffer, samples are cryoprotected in 30% glycerol in cacodylate buffer and then frozen in liquid propane. 3. Frozen samples are fractured at −110°C and unidirectionally platinum-shadowed at an angle of 45°. Replicas are immersed in household bleach to dissolve cells, washed in distilled water, and then mounted on copper grids. 4. They are finally examined using an electron microscope. Immunofluorescence assay can also help in studying TJ architecture with similar protocol described for AJ.
3.6. Co-immunoprecipitation Analysis
The adhesive function of VE-cadherin, like for other classical cadherins, depends on its interaction between the intracellular domain and catenin proteins. An increase in endothelial permeability, linked to a diminution of VE-cadherin adhesion is associated with a reduced binding of catenins to VE-cadherin. This feature could be analyzed by co-immunoprecipitation. In addition to classical catenins, it has been suggested that VEGFR2 is associated with stable junctions and their association is reduced upon permeability (28). In contrast, b-arrestin, a clathrin binding protein involved in endocytosis is found to be recruited to VE-cadherin upon VEGF stimulation (14). Claudin can be immunoprecipitated to analyze claudin molecular complexes at TJ. Alternatively, TJ biochemical composition
18 Role of Endothelial Cell–Cell Junctions in Endothelial Permeability
273
can be explored through ZO-1 immunoprecipitation; but keeping in mind that Z0-1 is not a strict marker of TJ, especially in ECs where it could be linked to both AJ and TJ. Classical protocols can be adapted to this kind of study (see Note 7). 1. Culture between two and five millions of ECs at confluence and starved them overnight or at least 6 h in serum-free medium. Place all items required (tubes, scrappers, buffers) in ice. 2. Wash in ice-cold PBS extensively and add ice-cold lysis TIT buffer (use 400 mL for a 10-cm dish): Tris–HCl 50 mM pH 7.4, NaCl 150 mM, Igepal CA-630 1% Triton-X100 1%, EDTA 20 mM, and freshly added commercially available protease inhibitor cocktail, as well as phosphatase inhibitors: Na3VPO4 1 mM and NaF 1 mM. Work on ice. 3. Immediately, scrap the culture surface and transfer into icecold labeled tubes. Incubate on ice for 20 min. 4. Centrifuge for 5 min, 10,000 × g at 4°C to eliminate cell debris and DNA. Transfer supernatants to clean tubes and proceed to protein concentration estimation using Bradford test. Save 50 mg of total proteins for Western-blot analysis (input). 5. Incubate 1 mg of proteins with 30 mL G-protein slurry sepharose fast flow for 30 min on a circular rotor at 4°C. This preclear step eliminates background of binding to G-protein sepharose. Perform a quick centrifugation (30 s, 10,000 × g at 4°C) to recover precleared supernatants. 6. Incubate each supernatant with 1 mg of anti-VE-cadherin antibody for 1 h on a circular rotor at 4°C. Goat anti-VEcadherin from Santa Cruz is working nicely for this kind of analysis. 7. Add 30 mL G-protein slurry sepharose to the above mixture and incubate for 4 h on a circular rotor at 4°C. 8. Perform a quick centrifugation (30 s, 10,000 × g at 4°C) to recover the immunoprecipitated fractions in the pellet. The supernatant can be saved for analysis of the nonimmunoprecipitated fraction (output). 9. Wash five times by adding 500 mL of buffer to the beads. 10. Recover the immunoprecipitated fractions and proceed to Western-blot analysis for b-catenin, a-catenin, p120-catenin, and VE-cadherin using if possible immunoglobulins from different species. In addition, phosphorylation status can be investigated by probing with anti-phospho-tyrosine antibodies. Of note, anti-phospho-threonine/serine antibodies commercially available are still not entirely satisfactory (see Note 8).
274
A. Le Guelte and J. Gavard
3.7. Other Biochemical Assays
3.7.1. Trypsine Assay
Studying VE-cadherin function is also linked to its subcellular localization, specifically between plasma membrane where VE-cadherin is available for cell contacts and internal compartment, such as endosomes and lysosomes where its adhesive activity is blocked. Indeed, reduction of VE-cadherin stability at the plasma membrane might contribute to compromise endothelial barrier integrity. Interestingly, internalization of VE-cadherin has been observed in response to monocyte extrasavation, inhibition of FGF signaling, and activation of VEGFR2 by VEGF (14, 19, 29). 1. Confluent EC monolayer is starved overnight and subjected to permeability treatment. Prepare duplicate for each condition, as regular total cell protein extraction has to be performed (see Note 9). 2. Add trypsine 0.25%–EDTA 1 mM for 2 min 37°C. 3. Centrifuge cells (1,000 × g, 5 min, 4°C) and lyse the pellet in 100 mL of TIT buffer. 4. After protein concentration determination, assess VE-cadherin trypsine-resistant fraction as compared to total cell lysates by Western blot. An increase in this fraction would mean that the VE-cadherin cytosolic pool is enhanced by either upregulation of endocytosis and synthesis, or downregulation of exocytosis. Furthermore, VE-cadherin cell surface expression is frequently altered in hyper-permeability conditions such as inflammation, diabetes, and virus exposure (30–33).
3.7.2. Internalization
1. Confluent EC monolayer is cultured on glass coverslip and starved overnight. 2. Incubate with anti-VE-cadherin with an extracellular epitope (BV6 clone) for 1 h at 4°C. This step allows antibody binding with minimal cellular trafficking. 3. Wash cells with cold PBS. 4. Proceed to permeability stimulation at 37°C. It is recommended to first prepare a kinetic from 2 min to 1 h. 5. Wash cells with PBS–glycine 0.1 M pH 2.0 for 15 min at RT, to remove membrane-fixed antibody. It is recommended to prepare duplicate where this step is skipped in order to have a control staining for total VE-cadherin. 6. Wash in RT PBS and proceed for immunostaining protocol. Double staining and permeabilization are now possible, but one channel has to be saved for VE-cadherin labeling. Similarly, cytosol versus membrane fractions of VE-cadherin could be analyzed by biotinylation assay. There are commercially available excellent kits to perform such investigation.
18 Role of Endothelial Cell–Cell Junctions in Endothelial Permeability
275
4. Notes 1. The advantage of this method is a specific recognition of vasculature ECs by monoclonal antibodies against mouse CD31. Usually, the phenotype and morphology of these cultures could remain stable over 10–15 passages in culture, and no overgrowth of contaminating cells of nonendothelial origin is observed at any stage. This model thus offers the possibility to access ex vivo to virtually all transgenic mouse models. One example is the use of purified ECs from mouse bearing a Flox mark on the Rac1 gene, whose excision is embryonic lethal (16). 2. For long-term culture, immortalization can be used as an alternative method but transformed EC might lose the expression of important EC molecules. By contrast, they offer stable nonversatile cell culture model. In the case of the human cell lines, we will present two different endothelial cells used in many studies. The human umbilical vein endothelial cells (HUVECs) have been established under the name of Ea.hy926. This EC model was originated by fusing HUVEC with the human epithelial cell line A549 (34). The resulting line has retained virtually all functions from the macrovascular endothelium and none from the A549 cell line. The human brain endothelial cell line HCMEC/D3 (human cerebral microvascular EC) was generated by expressing in normal brain EC human telomerase and SV40 T antigen, transducted with a lentiviral vectors (35). This cell line has stable normal karyotype, expresses normal endothelial markers (e.g., CD31, VE cadherin), maintained contact inhibited monolayers in tissue culture, exhibited robust proliferation in response to endothelial growth factors, and formed capillary tubes in matrix but no colonies in soft agar. This cell line represents up to now the best monoculture model that recapitulates the BBB phenotype. 3. Most cells display weakened adhesion onto glass, therefore implying a coating with an adhesion substrate, carefully chosen not to interfere with experimental settings. Among them, we can mention unspecific adhesion substrate using electrostatic interactions such as poly-l-lysine or commercially available HistoGRIP. A specific coating of rat tail collagen or purified recombinant fibronectin would engage protein– protein interactions and might therefore impact on the endothelial biology. 4. Electron microscopy will further help in identifying structural changes at the contact zones, and would allow to accurately measure the intercellular spaces and visualize actin density, membrane dynamics, and cell-to-cell junctions.
276
A. Le Guelte and J. Gavard
5. Catenins staining requires a permeabilization step for antibodies to access their epitopes, whereas VE-cadherin could be directly probed using antibodies directed against the extracellular domain. 6. The acquired images should display both large fields in order to evaluate the overall monolayer organization and higher magnification of the contact zones. When the permeability is low, one could expect that most VE-cadherin staining accumulate in a discrete line that figures the cell–cell contact zone, no spaces between cells are observable (36). By contrast, increase in endothelial cell permeability is associated with opening of endothelial cell–cell junctions, that could lead to a zigzag, more punctuate pattern (Fig. 1). Borders between cells might also appear less defined. As ECs assembled in a thin monolayer, with a poor architectural polarity, confocal analysis is not mandatory, but will improve image resolution. 7. It is important to keep in mind that cadherins and associated molecules are found in plasma membrane associated fractions and therefore are not extracted as easily as typical cytosolic proteins. Furthermore, the cadherin–catenin complex dissociation is most likely a rapid, fast cycling, and transient process that depends on labile phosphorylation status. The protocol described here allows to explore catenin binding to VE-cadherin in ECs that could be adapted as well to the exogenous analysis of over-expressed VE-cadherin in an ectopic model. 8. Such protocol has been used to demonstrate the effect of permeability inducing factors (PIF), such as angiogenic factors
Fig. 1. VE-cadherin staining in endothelial cell monolayer. Human umbilical vein endothelial cells (HUVECs) were cultured at confluence for 3 days and starved overnight and left untreated (a) or stimulated with VEGF 50 µg/mL for 30 minutes (b). VE-cadherin staining was then performed as described and analyzed by confocal microscopy.
18 Role of Endothelial Cell–Cell Junctions in Endothelial Permeability
277
and cytokines on the VE-cadherin adhesion complex. Permeability increase is most of the time associated with decrease of catenin association with VE-cadherin and increased phosphorylation (36). Alternatively, decreasing the association between VE-cadherin and p120 catenin leads to clathrin-dependent VE-cadherin endocytosis (19, 37). Phosphorylation has been described early as a prominent mechanism by which cadherins and catenin interaction can be modulated, either positively or negatively, and therefore might decipher for overall adhesion forces. Thus, a time course experiment is a judicious choice to start the experiment to take into account VE-cadherin dynamics. To this regard, VE-cadherin bears nine putative phospho-tyrosine sites (38, 39). In addition, S665 has been identified as a pivotal regulator of VE-cadherin availability at the plasma membrane (14). 9. VE-cadherin, as other cadherins, is a calcium-dependent adhesion molecule, whose extracellular domain can be cleaved by proteases, such as trypsine. This feature has been exploited to analyze VE-cadherin localization between plasma membrane and intracellular compartments. References 1. Mege, R. M., Gavard, J., and Lambert, M. (2006) Regulation of cell-cell junctions by the cytoskeleton. Curr Opin Cell Biol 18, 541–548 2. Tsukita, S., Furuse, M., and Itoh, M. (2001) Multifunctional strands in tight junctions. Nat Rev Mol Cell Biol 2, 285–293 3. Dejana, E. (2004) Endothelial cell-cell junctions: happy together. Nat Rev Mol Cell Biol 5, 261–270 4. Gavard, J. (2009) Breaking the VE-cadherin bonds. FEBS Lett 583, 1–6 5. Carmeliet, P. (2005) Angiogenesis in life, disease and medicine. Nature 438, 932–936 6. Vittet, D., Buchou, T., Schweitzer, A., Dejana, E., and Huber, P. (1997) Proc Natl Acad Sci U S A 94, 6273–6278 7. Carmeliet, P., Lampugnani, M. G., Moons, L., Breviario, F., Compernolle, V., Bono, F., Balconi, G., Spagnuolo, R., Oostuyse, B., Dewerchin, M., Zanetti, A., Angellilo, A., Mattot, V., Nuyens, D., Lutgens, E., Clotman, F., de Ruiter, M. C., Gittenberger-de Groot, A., Poelmann, R., Lupu, F., Herbert, J. M., Collen, D., and Dejana, E. (1999) Targeted deficiency or cytosolic truncation of the VE-cadherin gene in mice impairs VEGFmediated endothelial survival and angiogenesis. Cell 98, 147–157
8. Weis, S. M., and Cheresh, D. A. (2005) Pathophysiological consequences of VEGFinduced vascular permeability. Nature 437, 497–504 9. Weis, S., Cui, J., Barnes, L., and Cheresh, D. (2004) Endothelial barrier disruption by VEGF-mediated Src activity potentiates tumor cell extravasation and metastasis. J Cell Biol 167, 223–229 10. Crosby, C. V., Fleming, P. A., Argraves, W. S., Corada, M., Zanetta, L., Dejana, E., and Drake, C. J. (2005) VE-cadherin is not required for the formation of nascent blood vessels but acts to prevent their disassembly. Blood 105, 2771–2776 11. Nitta, T., Hata, M., Gotoh, S., Seo, Y., Sasaki, H., Hashimoto, N., Furuse, M., and Tsukita, S. (2003) Size-selective loosening of the blood-brain barrier in claudin-5-deficient mice. J. Cell Biol. 161, 653–660 12. Taddei, A., Giampietro, C., Conti, A., Orsenigo, F., Breviario, F., Pirazzoli, V., Potente, M., Daly, C., Dimmeler, S., and Dejana, E. (2008) Endothelial adherens junctions control tight junctions by VE-cadherinmediated upregulation of claudin-5. Nat Cell Biol 10, 923–934 13. Eliceiri, B. P., Paul, R., Schwartzberg, P. L., Hood, J. D., Leng, J., and Cheresh, D. A.
278
A. Le Guelte and J. Gavard
(1999) Selective requirement for Src kinases during VEGF-induced angiogenesis and vascular permeability. Mol Cell 4, 915–924 14. Gavard, J., and Gutkind, J. S. (2006) VEGF controls endothelial-cell permeability by promoting the beta-arrestin-dependent endocytosis of VE-cadherin. Nat Cell Biol 8, 1223–1234 15. Stockton, R. A., Schaefer, E., and Schwartz, M. A. (2004) p21-activated kinase regulates endothelial permeability through modulation of contractility. J Biol Chem 279, 46621–46630 16. Tan, W., Palmby, T. R., Gavard, J., Amornphimoltham, P., Zheng, Y., and Gutkind, J. S. (2008) An essential role for Rac1 in endothelial cell function and vascular development. FASEB J. 22, 1829–1838 17. Gavard, J., Patel, V., and Gutkind, J. S. (2008) Angiopoietin-1 prevents VEGF-induced endothelial permeability by sequestering Src through mDia. Dev Cell 14, 25–36 18. Jones, C. A., London, N. R., Chen, H., Park, K. W., Sauvaget, D., Stockton, R. A., Wythe, J. D., Suh, W., Larrieu-Lahargue, F., Mukouyama, Y.-s., Lindblom, P., Seth, P., Frias, A., Nishiya, N., Ginsberg, M. H., Gerhardt, H., Zhang, K., and Li, D. Y. (2008) Robo4 stabilizes the vascular network by inhibiting pathologic angiogenesis and endothelial hyperpermeability. Nat Med 14, 448–453 19. Murakami, M., Nguyen, L. T., Zhang, Z. W., Moodie, K. L., Carmeliet, P., Stan, R. V., and Simons, M. (2008) The FGF system has a key role in regulating vascular integrity. J Clin Invest 118, 3355–3366 20. Ballabh, P., Braun, A., and Nedergaard, M. (2004) The blood-brain barrier: an overview: structure, regulation, and clinical implications. Neurobiol Dis 16, 1–13 21. Wojciak-Stothard, B., Potempa, S., Eichholtz, T., and Ridley, A. J. (2001) Rho and Rac but not Cdc42 regulate endothelial cell permeability. J Cell Sci 114, 1343–1355 22. Mehta, D., and Malik, A. B. (2006) Signaling mechanisms regulating endothelial permeability. Physiol Rev 86, 279–367 23. Rask-Madsen, C., and King, G. L. (2010) Kidney complications: factors that protect the diabetic vasculature. Nat Med 16, 40–41 24. Lalor, P. F., Lai, W. K., Curbishley, S. M., Shetty, S., and Adams, D. H. (2006) Human hepatic sinusoidal endothelial cells can be distinguished by expression of phenotypic markers related to their specialised functions
in vivo. World J Gastroenterol 12, 5429–5439 25. Marelli-Berg, F. M., Peek, E., Lidington, E. A., Stauss, H. J., and Lechler, R. I. (2000) Isolation of endothelial cells from murine tissue. J Immunol Methods 244, 205–215 26. Lal-Nag, M., and Morin, P. J. (2009) The claudins. Genome Biol 10, 235 27. Inai, T., Sengoku, A., Hirose, E., Iida, H., and Shibata, Y. (2009) Freeze-fracture electron microscopic study of tight junction strands in HEK293 cells and MDCK II cells expressing claudin-1 mutants in the second extracellular loop. Histochem Cell Biol 131, 681–690 28. Lampugnani, M. G., Orsenigo, F., Gagliani, M. C., Tacchetti, C., and Dejana, E. (2006) Vascular endothelial cadherin controls VEGFR-2 internalization and signaling from intracellular compartments. J. Cell Biol. 174, 593–604 29. Allport, J. R., Muller, W. A., and Luscinskas, F. W. (2000) Monocytes induce reversible focal changes in vascular endothelial cadherin complex during transendothelial migration under flow. J Cell Biol 148, 203–216 30. Gavard, J., Hou, X., Qu, Y., Masedunskas, A., Martin, D., Weigert, R., Li, X., and Gutkind, J. S. (2009) A role for a CXCR2/phosphatidylinositol 3-kinase gamma signaling axis in acute and chronic vascular permeability. Mol Cell Biol 29, 2469–2480 31. Alexander, J. S., Alexander, B. C., Eppihimer, L. A., Goodyear, N., Haque, R., Davis, C. P., Kalogeris, T. J., Carden, D. L., Zhu, Y. N., and Kevil, C. G. (2000) Inflammatory mediators induce sequestration of VE-cadherin in cultured human endothelial cells. Inflammation 24, 99–113 32. Dewi, B. E., Takasaki, T., and Kurane, I. (2008) Peripheral blood mononuclear cells increase the permeability of dengue virusinfected endothelial cells in association with downregulation of vascular endothelial cadherin. J Gen Virol 89, 642–652 3 3. Navaratna, D., McGuire, P. G., Menicucci, G., and Das, A. (2007) Proteolytic degradation of VE-cadherin alters the blood-retinal barrier in diabetes. Diabetes 56, 2380–2387 34. Edgell, C. J., McDonald, C. C., and Graham, J. B. (1983) Permanent cell line expressing human factor VIII-related antigen established by hybridization. Proc Natl Acad Sci U S A 80, 3734–3737 35. Weksler, B. B., Subileau, E. A., Perriere, N., Charneau, P., Holloway, K., Leveque, M.,
18 Role of Endothelial Cell–Cell Junctions in Endothelial Permeability Tricoire-Leignel, H., Nicotra, A., Bourdoulous, S., Turowski, P., Male, D. K., Roux, F., Greenwood, J., Romero, I. A., and Couraud, P. O. (2005) Blood-brain barrier-specific properties of a human adult brain endothelial cell line. FASEB J., 9(13), 1872–1874 36. Esser, S., Lampugnani, M. G., Corada, M., Dejana, E., and Risau, W. (1998) Vascular endothelial growth factor induces VE-cadherin tyrosine phosphorylation in endothelial cells. J Cell Sci 111 (Pt 13), 1853–1865 37. Xiao, K., Allison, D. F., Kottke, M. D., Summers, S., Sorescu, G. P., Faundez, V., and Kowalczyk, A. P. (2003) Mechanisms of VE-cadherin processing and degradation in
279
microvascular endothelial cells. J Biol Chem 278, 19199–19208 38. Potter, M. D., Barbero, S., and Cheresh, D. A. (2005) Tyrosine Phosphorylation of VE-cadherin Prevents Binding of p120- and {beta}-Catenin and Maintains the Cellular Mesenchymal State. J. Biol. Chem. 280, 31906–31912 39. Wallez, Y., Cand, F., Cruzalegui, F., Wernstedt, C., Souchelnytskyi, S., Vilgrain, I., and Huber, P. (2007) Src kinase phosphorylates vascular endothelial-cadherin in response to vascular endothelial growth factor: identification of tyrosine 685 as the unique target site. Oncogene 26, 1067–1077
wwwwwwwwwwwww
Chapter 19 In Vitro Analyses of Endothelial Cell Permeability Elizabeth Monaghan-Benson and Erika S. Wittchen Abstract Endothelial cells lining the vessels of the vasculature and the cell–cell junctions, which join them, provide the primary barrier to the passage of fluids, immune cells, and macromolecules between the bloodstream and the tissues. Appropriate and dynamic regulation of this barrier is required during normal physiological processes; however, if not tightly controlled, increased permeability of the endothelium can also contribute to many pathological situations, including chronic inflammatory diseases and edema. The development of in vitro methods to study endothelial barrier function has been key in the identification of molecular mechanisms underlying many of these disease states. In this chapter, we describe three complementary approaches to measure endothelial monolayer permeability and barrier function in vitro. Key words: Barrier function, Endothelium, Endothelial cell, FITC-dextran, Impedance-sensing, Permeability, Transendothelial resistance
1. Introduction The vascular endothelium lines the intima of blood vessels and provides a semipermeable barrier important in controlling the passage of macromolecules and fluid between the blood and the interstitial space. Loss of this appropriate barrier function and subsequent large-scale vascular leakage often result in tissue inflammation and edema, hallmarks of disease states such as heart disease, cancer, stroke, and diabetes. However, transient smaller changes in permeability are also required during normal vascular homeostasis, highlighting the need for dynamic regulation. Vascular permeability is regulated, in part, through changes in paracellular permeability. The paracellular pathway refers to the movement of substances across the endothelial barrier via the
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_19, © Springer Science+Business Media, LLC 2011
281
282
E. Monaghan-Benson and E.S. Wittchen
intercellular space between adjacent endothelial cells. Cell–cell junctional structures (adherens and tight junctions) and their protein constituents are important regulators of the permeability characteristics of a given endothelial vessel type. Regulation of endothelial cell junctions and, therefore, vascular permeability can be modulated by vasoactive compounds, such as inflammatory cytokines, thrombin, histamine, and VEGF (1–4). These compounds can induce transient gaps between endothelial cells that facilitate paracellular permeability. There are multiple assays that allow the detection of changes in endothelial permeability. In this protocol, we describe three different in vitro methods that can be used, together with examples of control treatments that either positively or negatively regulate endothelial permeability. These methods are (1) Transendothelial electrical resistance (TER), which measures the resistance to an electrical current passed across the endothelial monolayer as a measure of ionic permeability; (2) Tracer Flux Assay, which quantifies the movement of a labeled small-molecule tracer, such as dextran, across a confluent endothelial monolayer; and (3) Realtime cell electrical impedance sensing, which measures the impedance of an endothelial cell monolayer plated on gold-microelectrode coated wells as an indicator of barrier function.
2. Materials 2.1. Cell Culture and Examples of Various Treatments
1. Pooled Human Umbilical Vein Endothelial Cells (HUVEC) (Lonza CC-2519). HUVEC were routinely used between passage 4–6. 2. Manufacturer’s recommended media: EGM-2 (EBM-2 basal media plus the included singlequots of supplements: hEGF, hydrocortisone, GA-1000 (Gentamicin Amphotericin-B), FBS, VEGF, hFGF-B, R3-IGF-1, ascorbic acid, heparin). 3. Tissue culture incubator: 37°C, 5% CO2. 4. For positive and negative control conditions, cells can be treated as indicated with various compounds, such as: (a) EGTA, 4 mM in EGM-2, (b) Thrombin (Sigma, T6884), 1 U/ml, (c) 8-CPT-2¢-O-Me-cAMP (Biolog), 10–500 mM, (d) Anti-VE-cadherin (cl75) function-blocking Ab (BD- Transduction Labs), 12.5 mg/ml, and (e) VEGF (R&D Biosystems), 10 ng/ml.
19 In Vitro Analyses of Endothelial Cell Permeability
2.2. Transendothelial Electrical Resistance Measurements
283
1. Optimem media (Invitrogen 31985070). 2. Transwell filter inserts, i.e., Corning 3401 (12-mm diameter, 0.4-mm pore size). 3. Matrigel (growth factor reduced) for coating inner well surface (BD-Biosciences, 354234). Thaw stock solution at 4°C and do not allow undiluted Matrigel to warm or it will gel. Using prechilled pipet tips, quickly dilute 1:100 into cold PBS and store at 4°C. 4. Endohm-12 Transwell chamber connected to an EVOM voltohmmeter (World Precision Instruments). 5. 100% ethanol for sterilizing forceps and Endohm-12 chamber.
2.3. Tracer Flux Assays (FITC-Dextran)
1. Transwell filter inserts, i.e., Corning 3401 (12-mm diameter, 0.4-mm pore size). 2. FITC-dextran (Molecular Probes D1821, D1822). 3. Fluorescent plate LABTECH).
2.4. Real-Time Cell Electric Impedance Sensing
reader
(FluroStar
Optima,
BMG
1. Instrumentation: xCELLigence RTCA system (Real-Time Cell Analyzer) (Roche Applied Science in partnership with ACEA Biosciences Inc.). 2. E-plate 16 (Roche Applied Science).
3. Methods 3.1. Endothelial Cell Culture
1. HUVEC are grown and maintained in EGM-2 on 100-mm tissue culture dishes at 37°C and 5% CO2. 2. Cells are passaged at 90% confluence with 0.05% trypsin/ EDTA. 3. Cells are typically used for experiments between passage 4 and 6 and are fed with fresh EGM-2 media every second day. 4. To ensure health of the cells, never subculture at a dilution ratio of greater than 1:3.
3.2. Transendothelial Electrical Resistance Measurements
1. For all conditions, prepare Transwells in triplicate. Additionally, include one Transwell with no cells as a blank. 2. If desired, coat the inner well of 0.4-mm pore-size, 12-mm diameter Transwell filters with Matrigel: Add enough 1:100 Matrigel/PBS stock to completely cover the filter (~250 ml) and incubate at 37°C for 30–60 min. Aspirate excess. 3. Plate cells: Seed 500,000 endothelial cells per Transwell (area = 1.1 cm2). This is typically a confluent cell density (see
284
E. Monaghan-Benson and E.S. Wittchen
Note 1). Ensure that final volume of EGM-2 media is 0.5 ml in the inner well (i.e., the Transwell insert), and 1.5 ml in the bottom well (i.e., inside the well of the multi-well plate). 4. Feed cells by removing spent media and adding fresh EGM-2 daily (see Note 2). 5. Culture at least 2–3 days to allow complete maturation of endothelial cell junctions (see Note 3). 6. On day of measurement, sterilize the Endohm-12 chamber by filling with 100% ethanol and incubating under UV for at least 15 min. While sterilizing, prewarm Optimem to 37°C. 7. Equilibrate TER meter: Add 2.5 ml prewarmed Optimem media inside sterilized Endohm-12 chamber and replace lid with top electrode positioned just into the media. Connect chamber to EVOM voltohmmeter, and turn on to “Voltage” setting. After 10–15 min equilibration period, reading should be zero. If not, use adjusting screw to correct. 8. Take readings: Switch meter to “Resistance.” Before each measurement, add 0.5 ml fresh prewarmed Optimem into each Transwell. Using sterile forceps, remove Transwells one at a time and place into Endohm-12 chamber containing 2.5 ml 37°C Optimem media (see Note 4). Press and hold “Measurement” button to take reading. 9. Multiple readings can be taken concurrently from each well at appropriate timepoints after indicated treatments. 10. Net TER measurements are calculated by subtracting the value of the blank Transwell insert (no cells). 11. After use, rinse with ethanol, then water, and air dry. 12. Data can be represented graphically, either as raw TER values (W/cm2) or normalized i.e., percent change in TER from time zero. An example of a typical experiment is shown in Fig. 1, where HUVEC plated on Transwells were treated for various times with the junction-disrupting agent thrombin prior to measuring TER. Thrombin is used here as a positive control for junctional disruption and increased permeability (reviewed in (5)). 3.3. Tracer Flux Assays
1. HUVEC are grown and maintained as described in Subheading 3.1. 2. The bottom chamber of the Transwell plate is filled with 1.5 ml EGM-2. 500,000 HUVEC are plated into the top chamber of the Transwell in a final volume of 0.5 ml EGM-2 (see Note 1). All conditions should be set up and analyzed in triplicate. Additionally, include one Transwell with no cells as a blank. 3. Cells are allowed to grow for 3 days on the Transwell filters (see Note 3).
19 In Vitro Analyses of Endothelial Cell Permeability
285
Fig. 1. Transendothelial electrical resistance (TER) of HUVEC monolayer treated with thrombin. HUVEC were plated at confluent density for 2 days on Transwell filters. Steadystate TER values (with blank filter subtracted) reached 12.5 ± 1.5 W/cm2 (mean ± SEM of n = 4 Transwells). Thrombin (1 U/ml) was added for indicated times and TER was measured. Thrombin rapidly and significantly decreases TER (*p < 0.05).
4. The medium in the Transwells is changed on day 2 (see Note 2). 5. A working solution of FITC-dextran at 1 mg/ml is prepared fresh (see Notes 5 and 6). The working solution of FITC dextran is prepared in the medium necessary for the treatment being tested. A final FITC-dextran volume of 0.5 ml per Transwell is required. As a positive control, the FITCdextran solution should also be added to the Transwell containing no cells. 6. Label two sets of 15-ml tubes with the treatments being tested (in this example, we use VEGF) (6). Include all control treatments as well. In one set of tubes, add 0.5 ml of the FITC-dextran solution for each well receiving the treatment labeled on the tube. Add the appropriate treatment to this tube. This will be the solution added to the upper chamber of the Transwell (apical side of cell monolayer). 7. In the second set of tubes, add 1.5 ml of medium for each well receiving the treatment listed on the tube. Add the appropriate treatment to this tube. This solution does not contain FITC-dextran, and it will be added to the bottom well of the Transwell (basolateral side of cell monolayer). 8. Aspirate bottom and top chambers of all Transwells (see Note 2). For each treatment, add 0.5 ml of the FITC-dextran solution to the top chamber. Then, add 1.5 ml of the medium to the bottom chamber of the Transwell. 9. Incubate the Transwell plate at 37°C for an appropriate time for the treatment being tested (see Note 7).
286
E. Monaghan-Benson and E.S. Wittchen
10. To quantify passage of FITC-dextran across the monolayer, remove the Transwell plate from the incubator. Using a micropipetter, remove 50 ml from the bottom chamber of each Transwell and transfer to a 96-well plate. Record which treatment was placed in which well of the 96-well plate (see Note 8). 11. Place the 96-well plate into the fluorescent plate reader. 12. Set the excitation wavelength on the plate reader to 488 nm and the emission wavelength to 520 nm. Read the plate according to plate reader instructions. 13. Data can be represented graphically either as the raw data in the arbitrary fluorescence units given by the plate reader or calculated as a percent change from control values. An example of a typical experiment is shown in Fig. 2, where HUVEC plated on Transwells were treated for various times with the permeability enhancing agent VEGF (7). Both 10,000 kDa dextran (Fig. 2a) and 70,000 kDa dextran molecule (Fig. 2b) were used. VEGF treatment causes enhanced passage of FITCdextran across the monolayer. At 120 min, the passage of the
Fig. 2. FITC-dextran flux of HUVEC monolayers treated with VEGF. HUVEC were plated at confluent density for 3 days on Transwell filters. 1 mg/ml FITC-dextran (a) MW 10,000 or (b) MW 75,000 was placed in the upper well and cells were treated with 10 ng/ml VEGF for 15, 30, 60, and 120 min. A sample of medium from the lower chamber was taken after each timepoint, and the amount of FITC-dextran in the lower chamber was measured in a fluorescence plate reader.
19 In Vitro Analyses of Endothelial Cell Permeability
287
10,000 kDa dextran molecule is increased by ~50% in the VEGF treated cells compared to control untreated cells. However, passage of the 70,000 kDa dextran is only increased by ~25% in VEGF-treated cells compared to control cells. Particular treatments may cause the formation of different sized openings in the monolayer, and therefore several different dextran molecules may have to be tested to determine the optimum size for the treatment being tested. 3.4. Real-Time Cell Electric Impedance Sensing
This method utilizes instrumentation (RTCA-DP, Roche Applied Science/Acea Biosciences) which tracks electrical impedance signals, monitoring in real time the status of cells grown on microelectrode coated plates. The impedance readout is expressed in arbitrary units as “Cell Index,” which reflects changes in barrier function and permeability (8). 1. For each treatment/condition, plate enough wells to have at least triplicate conditions. 2. Add 100 ml EGM-2 to each well of the E-plate 16, insert plate into RTCA-DP instrument, and take background reading of selected wells using the RTCA software. 3. Seed appropriate number of endothelial cells into each well of the E-plate 16 (determined empirically, see Notes 1 and 9). In the case of HUVEC, 40,000 cells per well is sufficient to attain a completely confluent monolayer the next day. 4. Optional: Seed an equivalent density of cells (by area) on a coverslip for visual verification of monolayer (especially important if comparing different cell types). 5. Monitor impedance overnight, setting the software to take readings every 15–30 min. 6. On day 2, the impedance values will have reached a plateau, indicating a fully formed barrier. 7. At this point, one can add various compounds to appropriate wells and monitor changes in impedance over very short intervals, depending on the settings selected with the software (i.e., measurements every 1–15 min, for the desired total time interval). 8. Data can be represented graphically as Cell Index (CI) or Normalized Cell Index, which is useful when comparing effects upon the addition of drugs or compounds. To do this, select the timepoint immediately preceding the addition of compound as the reference value, which is set to “1.” Two examples are shown to validate this method as a way of detecting rapid changes in endothelial barrier function. Figure 3a shows increased barrier function in a dose-dependent manner upon the addition of a compound (8-CPT-2¢-O-Me-cAMP)
288
E. Monaghan-Benson and E.S. Wittchen
Fig. 3. Real-time impedance analysis of endothelial barrier function. (a) Impedance was measured in triplicate wells every 15 min following the addition of various doses of the Epac-activating cAMP analog 8-CPT-2¢-O-Me-cAMP, which is known to enhance barrier function. There is a dose-dependent increase in cell index, which is significant for all doses at t = 0.5 h and longer (p < 0.05). (b) After reaching steadystate index (t = 0), wells were treated with indicated junction-disrupting agents: EGTA (4 mM), clone75 Ab (VE-cadherin function-blocking AB), or thrombin (1 U/ml). Treatment with all compounds results in a rapid drop in cell index, indicating decreased barrier properties. Graph represents mean of triplicate wells ±SD. With the exception of the single data points thrombin and cl75 at t = 0.25 h, all remaining data points are significantly different from control (p < 0.05).
known to activate the Rap1 GTPase GEF, Epac (9), and positively affect the endothelial cell junction barrier (10, 11). Conversely, Fig. 3b illustrates the effects of various junction-disrupting treatments on endothelial barrier function. Treatments that can be used to disrupt endothelial cell junctions include calcium chelation by EGTA (12), thrombin treatment (5), and incubation with a VE-cadherin function-blocking antibody (13).
19 In Vitro Analyses of Endothelial Cell Permeability
289
4. Notes 1. It is important to plate the HUVECs at confluency. In order for these assays to successfully measure endothelial monolayer permeability, there can be no gaps between the cells. High plating density also assures that mature junctions will be established. 2. To avoid damage to the monolayer by uneven hydrostatic pressure, always aspirate medium from the bottom chamber first and the top chamber second. When aspirating medium from the top chamber take care not to touch the Transwell filter with the Pasteur pipet as this will damage the filter. When adding fresh medium, fill the top chamber first and the bottom chamber second. For the same reason, the amount of media in each chamber is very important, i.e., with the cap in place, the height of the fluid in the chamber should be at the same level as inside the Transwell in order to avoid hydrostatic pressure on the membrane. 3. Polycarbonate filter membranes are nontranslucent. Once the cells have been plated on the Transwell, they cannot be observed under the microscope to check growth and confluence. However, in parallel, an equivalent density (by area) can be plated into a single well of a 24-well plate to allow visual confirmation of monolayer formation. If it is desirable to directly observe cells plated on Transwells by phase-contrast light microscopy, you can alternatively use polyester Transwell inserts (Corning 3460) which are translucent. 4. To avoid erroneous readings, when placing insert into the Evohm-12 chamber, it is very important to ensure that no air bubbles are trapped between the media in the chamber and the underside of the Transwell filter. Note that the top electrode position can be adjusted by rotating the cap and using the locking nut. When the cap is in place, the clearance between the top electrode and the filter membrane should be 1–2 mm. 5. Stock solutions of FITC-dextran can be prepared according to the manufacturer’s instructions, aliquoted and frozen at −20°C. The working solution of FITC-dextran should be prepared fresh immediately before use. 6. The dextran molecules come in different sizes (as demonstrated in Fig. 2). Individual optimization is necessary to determine the appropriate-sized dextran molecule based on the conditions of the experiment. 7. The time point chosen must be optimized by the user. It is often helpful to perform a timecourse pilot experiment.
290
E. Monaghan-Benson and E.S. Wittchen
8. Try to leave one empty well between different treatment conditions. This will help prevent signal bleed over from well to well in the plate reader. 9. The optimal cell number for seeding onto the E-Plate 16 was determined by plating a range of cell densities in a 96-well plate. Keeping in mind the differences in area of a 96-well (~32 mm2) and the E-plate 16 (~20 mm2), number of cells per mm2 can be calculated. Optimal cell number was thus determined by visual assessment of each well, and choosing the seeding density that gave a tightly confluent monolayer after overnight plating.
Acknowledgments This work was supported, in part, by National Institutes of Health grant HL080166. E.S.W. is supported by an American Heart Association Scientist Development Grant (10SDG3430042). E.M-B. is supported by the American Cancer Society PF-09_119-01-CSM. References 1. Lum H, and Malik AB (1994) The involvement of VEGF in endothelial permeability: a target for anti-inflammatory therapy. Am J Physiol 3:L223–41. 2. Lee YC (2005) The involvement of VEGF in endothelial permeability: a target for antiinflammatory therapy. Curr Opin Investig Drugs 11:1124–30. 3. Mehta D, and Malik A (2006) Signaling mechanisms regulating endothelial permeability. Physiol Rev 1:279–367. 4. Aghajanian A, Wittchen ES, Allingham MJ, Garrett TA, and Burridge K. (2008) Endothelial cell junctions and the regulation of vascular permeability and leukocyte transmigration. J Thromb Haemost 6:1453–60. 5. Bogatcheva NV, Garcia JG, and Verin AD. (2002) Molecular mechanisms of thrombininduced endothelial cell permeability. Biochemistry (Mosc) 67:75–84. 6. Monaghan-Benson E, and Burridge K. (2009) The regulation of vascular endothelial growth factor-induced microvascular permeability requires Rac and reactive oxygen species. J Biol Chem 284:25602–11. 7. Senger DR, Galli SJ, Dvorak AM, Perruzzi CA, Harvey VS, and Dvorak HF. (1983) Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid. Science 219:983–5.
8. Atienza JM, Yu N, Kirstein SL, Xi B, Wang X, Xu X, and Abassi YA. (2006) Dynamic and label-free cell-based assays using the RealTime Cell Electronic Sensing System. Assay and Drug Development Technologies 4: 597–607. 9. Enserink JM, Christensen AE, de Rooij J, van Triest M, Schwede F, Genieser HG, Doskeland SO, Blank JL, and Bos JL. (2002) A novel Epac-specific cAMP analogue demonstrates independent regulation of Rap1 and ERK. Nat Cell Biol 4: 901–6. 10. Cullere X, Shaw SK, Andersson L, Hirahashi J, Luscinskas FW, and Mayadas TN. (2005) Regulation of vascular endothelial barrier function by Epac, a cAMP-activated exchange factor for Rap GTPase. Blood 105:1950–5. 11. Wittchen ES, Worthylake RA, Kelly P, Casey PJ, Quilliam LA, and Burridge K. (2005) Rap1 GTPase inhibits leukocyte transmigration by promoting endothelial barrier function. J Biol Chem 280:11675–82. 12. Shasby DM, and Shasby SS. (1986) Effects of calcium on transendothelial albumin transfer and electrical resistance. J Appl Physiol 60:71–79. 13. Hordijk PL, Anthony E, Mul FP, Rientsma R, Oomen LC, and Roos D. (1999) Vascularendothelial-cadherin modulates endothelial monolayer permeability. J Cell Sci 112:1915–23.
Chapter 20 Mechano-Transduction and Barrier Regulation in Lung Microvascular Endothelial Cells Kristina Giantsos, Mark Cluff, and Randal Dull Abstract Alterations in endothelial permeability are a hallmark of inflammation as well as the underlying cause of many clinical syndromes. Quantifying changes in endothelial barrier properties to water and macromolecules can be an important means of assessing the degree of cellular injury and, conversely, the effect of therapies to attenuate the inflammatory cascade. We use a combination of an isolated organ system and two cell culture models to investigate mechanisms of endothelial barrier regulation under variety of experimental conditions. Each assay has its own experimental strengths and limitations and must be used appropriately for the questions being asked. When used collectively, they can provide significant insight into the molecular regulation of lung endothelial permeability. Key words: Endothelium, Hydraulic conductivity, Filtration coefficient, Lung permeability coefficient, Lung, Mechano-transduction
1. Introduction Vascular endothelial cells are sensitive to both pressure and flow and alterations in local mechanical forces activate signaling pathways that can influence the permeability of cell–cell junctions. The lung is a complex organ and controlling factors that influence endothelial activation by mechanical forces or locally generated signals requires the use of both whole organ and cell culture models to elucidate pathways that are involved in mechanotransduction.
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_20, © Springer Science+Business Media, LLC 2011
291
292
K. Giantsos et al.
2. Materials 2.1. Lung Perfusion Media
1. Krebs-Ringer Bicarbonate Buffer (Sigma, St. Louis, MO) is dissolved in 900 mL of reverse osmosis water. Rinse original package with 5 mL of water to recover all traces of the powder. 2. To the solution, add 1.26 g (15 mM) sodium bicarbonate for each liter of final volume. 3. While stirring, dissolve 30 g of bovine serum albumin (BSA) into solution to make a 3% (w/v) concentration. We use albumin Fraction V, from Proliant Biologicals, Boone, IA. 4. Add 2 mM CaCl2 to your solution and bring up to a liter of final volume. 5. While stirring, adjust the pH of the medium to 7.4 with NaOH. Sterile filter with
2.2. Cell Culture Reagents
1. Bovine lung microvascular endothelial cells (BLMVEC; Vec Technologies, Rensselaer, NY). 2. Media I is Dulbecco’s modified eagle’s medium without phenol red (Sigma, St. Louis, MO) supplemented with 25 mM HEPES (Sigma, St. Louis, MO) and 0.1% penicillin/streptomycin (Invitrogen; Carlsbad, CA) at pH 7.4. 3. Media II is prepared from MCDB-131 (Sigma, St. Louis, MO) supplemented with 1% fraction V BSA (Here, Proliant is in Boone, IA), 25 mM HEPES (Sigma, St. Louis, MO), and 0.1% penicillin/streptomycin (Invitrogen, Carlsbad, CA), and adjusted to pH 7.4. 4. Media III is prepared from Media I with 1% fraction V BSA (Here, Proliant is in Boone, IA). 5. Gelatin (Sigma, St. Louis, MO) is dissolved to 0.4% (w/v) in deionized water and added to tissue culture dishes as required. 6. Bovine fibronectin (Sigma, St. Louis, MO) is dissolved in MCDB-131 to a final concentration of 0.1 mg/mL and added to Snapwell as required. 7. 1X Phosphate buffered saline (PBS; Invitrogen, Carlsbad, CA). 8. Formalin solution (Sigma, St. Louis, MO) is purchased as a 10% solution in a neutral buffer. 9. Ladd Multiple Stain (Ladd Research, Williston, VT) is a cellpermeant dye that stains various tissue elements in contrasting colors to aid in light microscopy. 10. Tetramethylrhodamine-conjugated BSA (TMR-BSA; Invitrogen, Carlsbad, CA) is reconstituted in 250 mL of Media I to working concentration of 20 mg/mL on the day of experiment.
20 Mechano-Transduction and Barrier Regulation…
2.3. Isolated Rat Lung System
293
The instrumentation in our system includes: 1. Pressure transducers (Model P75), Harvard Apparatus, Holliston, MA. 2. Transducer Amplifier Modules TAM-A 705/1 PLUGSYS. Harvard Apparatus, Holliston, MA.
HSE
3. Isometric Force Transducer (Model 159901A). Radnoti Corporation, Monrovia, CA. 4. Masterflex Roller pump with L/S 13 tubing. I.D. 1.6 mm O.D. Cole-Parmer, Vernon Hills, IL. 5. Data acquisition system developed in Labview 8.2. (National Instruments, Austin, TX). 6. Tracheal Cannula Polyethylene tubing PE 240 I.D. 1.67 mm (0.066) O.D. 2.42 mm (0.95) Intramedic-Clay Adams Brand, Cole-Parmer, Vernon Hills, IL. 7. A pressure-controlled rodent ventilator. Kent Scientific, Torrington, CT. 8. Heated, circulating water bath. 9. Water jacketed, glass reservoir (50 mL); stir plate. 2.4. In Vitro Hydraulic Conductivity System
1. Cell permeability chambers are custom built from two solid polycarbonate cylinders. The bottom half contains a small pressure-equilibrated reservoir (abluminal) that houses a Snapwell filter and seals with a rubber O-ring. The top half of the chamber contains a concavity that provides a 1-mL reservoir above the filters; it seals against the top rim of the Snapwell with a second O-ring. Two spring-loaded latches on the outside of the chamber lock the two halves of the cylinder together to form a water-tight housing for the cells and their supports. The top of the permeability chamber has two stopcock-controlled ports for inflow and outflow of tissue culture media. 2. A combination camera stand and glass capillary tube holder is custom built as one tightly fitted structure and machined from heavy aluminum. This design allows the camera to be calibrated only once for multiple uses of the equipment. The camera is a Sony RS-170 monochromatic video camera and programing of the video tracking and analysis software was done using LabVIEW (National Instruments, Austin, TX). 3. The pressure manifold is composed of repurposed disposable polystyrene serological pipettes (10 mL, Fisher Scientific, Pittsburgh, PA) connected to a five stopcock manifold (Qosina, Edgewood, NY). 4. Silastic tubing (Cole-Parmer, Vernon Hills, IL) connects the inflow port of the permeability chamber with the
294
K. Giantsos et al.
pressure manifold. The portion of the silastic tubing that runs beneath the camera and houses the air bubble is fitted with an 18-cm length of clear borosilicate capillary tubing (Wilmad Glass, Buena, NJ) with an internal diameter of 1.9913 mm ± 0.0004.
3. Methods We use a combination of methods to evaluate the effects of hydrostatic pressure on endothelial permeability. The isolated perfused lung preparation has been the gold-standard for assessing changes in lung vascular permeability. The development of cell culture techniques has allowed added control in terms of precise hydrostatic and oncotic pressure and the absence of other cell types that may release permeability-enhancing agents. 3.1. Isolated Lung Preparation
1. Sprague-Dawley rats (300–450 g) were anesthetized with an intraperitoneal injection of ketamine and xylazine (80/10 m/kg). 2. Surgically expose the trachea and perform a tracheotomy with PE 240 tubing and place two 4.0 silk sutures around the trachea and the tracheal cannula. The rat is mechanically ventilated at a rate of 60 breaths per minute with an FiO2 of 50%. 3. Make a midline laparotomy, with two additional perpendicular incisions down the sides just below rib cage to aid in exposure. 4. Make a midline sternotomy and retract the ribs to expose the heart and lungs. Inject heparin 500 USP units, directly into right ventricle and allow to circulate for 2 min. 5. Excise the thymus gland to expose the aorta and place a loose 4.0 suture, around the aorta and pulmonary artery trunk. 6. Exsanguinate the animal by cutting abdominal aorta and excising the apex of the left ventricle. 7. Make a small incision into the right ventricle near the pulmonary artery outflow tract; with the perfusion medium flowing, insert the pulmonary artery catheter into the pulmonary trunk and secure it with a 4.0 suture. Make sure that there are no bubbles in the tubing or at the tip of the catheter. 8. Place a purse string around the apex of the heart. Insert the left atrial cannula through the left ventricle and pass it through the mitral valve and into the left atrium. Secure this catheter with the purse string suture around the left ventricle. Fluid should now be draining out of the cannula.
20 Mechano-Transduction and Barrier Regulation…
295
9. Carefully excise the trachea, heart, and lungs en bloc, and hang the preparation from the force transducer in order to measure the continuous weight of the ex vivo preparation. Zero the pulmonary artery and left atrial pressure (Pla) transducers at the level of the left atrium. 10. At this point, the lungs are suspended from a force transducer and are being perfused with Krebs-Ringers Buffer containing 3% BSA that is being recirculated from the jacketed water bath. 11. Using a data acquisition system, collect: Continuous weight measurement Pulmonary artery pressure (Ppa) Left atrial pressure (PLA) Pulmonary artery inflow 12. The experiment starts with a 20- to 30-min isogravimetric period (steady state) for collecting baseline hemodynamic data and to ensure that the lungs have not been injured during the isolation procedure. During the isogravimetric period, the lung weight should remain constant. Any weight gain is indicative of an injured preparation. Normal pulmonary artery pressures are approximately 6.5–9.5 cmH2O when pulmonary flow rate is 6 mL/min and Pla is set to 3 cmH2O. 13. To derive a whole lung filtration coefficient (Kfc) for a given experimental condition, capillary pressure is increased to 7.5 cmH2O for 20 min with the use of a manometer placed on the left atrial outflow cannula. When PLA is increased, the lung preparation will gain weight rapidly due to an increase in vascular volume (fast phase) and a second slow continuous weight gain will occur, related to transvascular fluid filtration (1). 14. After the 20 min, PLA is returned to 3 cmH2O for another 20 min. 15. We use the Kfc value derived from the initial pressure pulse to derive “baseline” Kfc. During the second isogravimetric period, a test reagent can be delivered to the lungs via infusion into the pulmonary artery catheter or by addition to the reservoir. A second increase in capillary pressure is then performed for another 20-min period and second Kfc is derived. The Kfc ratio of the experimental period can be compared to baseline values in a manner that is appropriate for the questions being asked (e.g., Kfc2/Kfc1). 16. At the end of each experiment, a weight of known mass (usually a 2-mg weight) is hung from the preparation to ensure that the measured weight changes by the exact amount.
296
K. Giantsos et al.
17. Lastly, the prep is taken down and lungs are dissected from the heart and frozen in liquid nitrogen for future analysis. 3.2. Calculation of Kfc from DWeight/Time
1. Kfc has units of mL/min/cmH2O × 100 g of lung weight. The change in lung weight over the final 2 min during the increased PLA (18–20 min) is used to derive Kfc. The change in weight is expressed as ∆ mL/min, assuming that 1 mL of fluid = 1 g. 2. ∆ wt/time (s) is normalized to pulmonary capillary pressure (Pc), where Pc = (pulmonary artery pressure − Pla)/2. 3. Lastly, the predicted lung weight is derived from the following equation for rats: Plw = and used to derived a correction factor to normalize ∆ mL/min/Pc to 100 g of lung weight (2) (see Notes 1–4).
3.3. Preparation of BLMVEC Monolayers on Snapwell Permeability Chambers
1. In order to prepare an optimal growth surface for endothelial cells, 500 mL of 0.4% gelatin solution is added to the luminal chamber of the Snapwell culture filters and allowed to incubate for 1 h at room temperature. Gelatin is removed and replaced by 500 mL of bovine fibronectin solution for 1 h at room temperature. 2. Confluent BLMVEC monolayers grown in tissue culture flasks are rinsed one time with sterile PBS and incubated with trypsin (1 mL per 25 cm2 growth area) at 37°C for 5–7 min or until cells are no longer bound to support. Prewarmed MCDB-131 complete media is added to cells such that the ratio of media to trypsin is 1:4; this effectively halts trypsin activity. Cell suspension is gently aspirated in order to separate cells and homogenize the solution. 3. Cells are seeded at a density of 2.5 × 105 cells/cm2 onto luminal side of treated filters. A final luminal concentration of 500 mL is desired; therefore, MCDB-131 is used to bring up the volume. The abluminal chambers of the tissue culture dish are filled with 2.5 mL of MCDB-131 such that the top of the fluid contacts the bottom of the filter upon which cells are seeded. Cells may be used between days 7 and 10 and up to passage 15. Tissue culture media should be changed in both luminal and abluminal chambers on every third day and cells should be maintained at 37°C and 5% CO2 until their use (see Note 5).
3.4. Hydraulic Conductivity Measurement (see Notes 6 and 7)
1. To set up an experiment, Snapwell filters are removed from the support and gently washed twice with warm Media II. Enough media should remain in the filters to cover the cells. Filters are placed inside the custom permeability chamber, which remain in a water bath at 37°C for the duration of the experiment.
20 Mechano-Transduction and Barrier Regulation…
297
2. The pressure manifold, silastic tubing, and luminal aspect of the permeability chamber is filled with warm Media II such that no air bubbles are present in the system. Pressure differentials applied to cell monolayers are manifested by the height of the media in the pressure manifold relative to the cell monolayer in the permeability chamber; this pressure is denoted cmH2O. 3. An air bubble (approximately 1 mL volume) is introduced in the system and placed directly within the line of sight of the camera. Our computerized system is activated at this point to track the air bubble. 4. Once the permeability chamber is fully assembled, a hydrostatic pressure (DP) of 1 cmH2O is applied for 1 h as an equilibration period. 5. After 1 h, the pressure is increased step-wise by opening the stopcock on the pressure manifold that corresponds to the desired pressure applied to the cells. 6. For each experiment, the displacement of the air bubble, in cm, is collected every 5 s. The software tracks user-defined flow markers, records displacement versus time, and displays experimental data in real time for up to three independent data sets. 7. At the end of an experiment, the Snapwell filters are removed from the permeability chambers and placed in formalin solution for 15 min. Formalin is removed and cells are stained with Ladd Multiple Stain for 5 min then rinsed and evaluated under light microscopy for monolayer confluency and integrity. 8. The stored displacement versus time data is converted to Lp (cm/s/cmH2O) using a second custom written software program also created in LabVIEW that uses data to calculate and display the fluid flux rate versus time and the hydraulic conductivity coefficient versus time for three independent data sets according to the following relationship 3.5. In Vitro Albumin Diffusion Assay
Lp =
Jv . DP
1. Plastic assay block is custom built from a 1 × 12 × 12 in. block of polycarbonate with 9 wells drilled in a 3 × 3 matrix. Each well has a depth of 3 cm with a 1-cm radius capable of supporting a Transwell culture chamber. A side notch is cut into the lip of each well to facilitate sample removal from the abluminal chamber. 2. Bovine lung endothelial cells are grown on Transwell culture filters (12-mm diameter polycarbonate filters with 0.4-mm pore size, Corning, Lowell, MA). In order to prepare an
298
K. Giantsos et al.
ptimal growth surface for seeded endothelial cells, 500 mL o of 0.4% gelatin solution is added to the luminal chambers of Transwell culture filters and allowed to incubate for 1 h at room temperature. Gelatin is removed and replaced by 500 mL of bovine fibronectin solution for 1 h at room temperature. 3. Confluent BLMVEC monolayers grown in tissue culture flasks are rinsed one time with sterile PBS and incubated with trypsin (1 mL/25 cm2 growth area) at 37°C for 5–7 min or until cells are no longer bound to support. Prewarmed MCDB-131 complete media is added to cells such that the ratio of media to trypsin is 1:4; this effectively halts trypsin activity. Cell suspension is gently aspirated in order to separate cells and homogenize the solution. 4. Cells are seeded at a density of 2.5 × 105 cells/cm2, on to luminal side of treated filters. A final luminal concentration of 500 mL is desired, so MCDB-131 is used to bring up the volume. The abluminal chambers of the tissue culture dish are filled, with 2.5 mL of MCDB-131 such that the top of the fluid contacts the bottom of the filter upon which cells are seeded. Cells may be used between days 7 and 10 and up to passage 15. Tissue culture media should be changed in both luminal and abluminal chambers on every third day and cells should be maintained at 37°C and 5% CO2 until their use. 5. On the day prior to the experiment, the plastic block is warmed to 37°C in an incubator overnight. All media are warmed to 37°C approximately 20 min prior to the experiment. 6. To prepare for an experiment, tissue culture medium should be removed from the luminal chamber of the cells and replaced with 475 mL of Media I. The seeded Transwell filters are removed from their tissue culture dishes and placed gently in the plastic block. The plastic block is positioned on a multichannel stirplate in a temperature-controlled cabinet at 37°C. To the abluminal chambers of the plastic block, add a small stir bar and enough Media II to equilibrate the fluid levels in the luminal and abluminal chambers, thereby eliminating the hydrostatic pressure gradient. Allow the stir bars to gently swirl the media in the abluminal chambers as a preliminary equilibration. 7. To start an experiment, add 25 mL of 20 mg/mL solution of TMR-BSA such that the luminal BSA concentration becomes 1%; the final volume is 0.5 mL per chamber. Remove 30 mL of medium from the abluminal chamber of each sample well and place in a 96-well plate. It is necessary to maintain a constant volume in the abluminal chamber, therefore 30 mL
20 Mechano-Transduction and Barrier Regulation…
299
of Media II is replaced into each abluminal well directly following sample removal. Aliquots of 30 mL are continually sampled from the abluminal chambers at predetermined time intervals, usually 1 per hour (see Note 8). 8. In order to convert fluorescence intensity output to concentration units, a range of dilutions of the stock TMR-BSA solution is prepared. Beginning with 10 mL of the same 20 mg/mL TMR-BSA solution prepared earlier, this aliquot is added to 1 well of a 96-well plate and diluted with 290 mL of Media II. A 1:6 TMR-BSA:Media II dilution ratio is repeated ten times in the same 96-well plate as house the abluminal chamber samples. Each 30 mL sample is diluted with 120 mL Media II (5× dilution) to compensate for the sensitivity range of the fluorimeter, thus individual dilution factors will need to be adjusted for the sensitivity of each fluorimeter. 9. At the end of the experiment, the 96-well plate containing samples, the standards, and 3 blank wells consisting of Media II only is read by a fluorimeter using excitation/emission wavelengths of 541/572 corresponding to the excitation/ emission spectra for the TMR fluor. 10. From the standards data, a calibration plot may be obtained by plotting known concentration versus fluorescence intensity. The slope of this plot is used to convert sample fluorescence intensity to concentration in units of mg/mL. A permeability coefficient is derived from the formula:
P =
DCAvol , DtALconc
where P = diffusive permeability coefficient (cm/s) DC = change in tracer concentration in abluminal chamber (mg/mL) Avol = abluminal chamber volume (mL) A = monolayer surface area (cm2) Dt = time (s) Lconc = luminal concentration of tracer (mg/mL) 11. Postexperiment, the monolayer is visualized to confirm cell confluence and integrity. Experimental media is removed from the luminal cell chambers and an aliquot of 200 mL of formalin is added to each luminal chamber for 10–15 min at room temperature. After sufficient reaction time, the formalin is removed and a 5-mL aliquot of Ladd Multiple Stain is added to each cell monolayer for 5 min. The dye is removed
300
K. Giantsos et al.
and monolayers are washed once in deionized water. Filters may be examined under light microscopy for areas of cellular nonconfluence or damage to the filters.
4. Notes 1. The ex vivo lung preparation is modeled after the system developed by Stefan Uhling, Ph.D.(3) with minor modifications as required to meet the experimental questions being investigated. 2. There are numerous assumptions regarding the validity of the isolated, perfused lung preparation to accurately measure the whole Kfc that have been reviewed in detail by (1, 4, 5). 3. In order to study mechano-transduction in the lung, we vary Pla, pulmonary artery flow, and inspiratory pressure to alter the cumulative physical forces acting on the lung vasculature. 4. Pharmacological agents are added to the perfusate reservoir to modulate specific signaling pathways of interest. 5. For cell culture experiments, culture duration, growth media, filter preparation, cell type, and oncotic pressure all influence the permeability indices that are being measured and must be taken into account when optimizing the experimental conditions and hypotheses being tested. 6. Most clinically relevant edemagenic conditions involve processes that result in an increase in water permeability to a greater extent than an increase in macromolecular permeability. This is why lymph protein concentration typically falls during edema development, reflecting the greater increase in Lp compared to the change in the reflection coefficient for large molecules. We have developed an automated system for measuring the hydraulic conductivity of endothelial monolayers cultured on a porous support, typically, a polycarbonate filter with 0.4-mm pores. There are numerous vendors for this type of cell culture chamber, and we built our system around the Snapwell chamber produced by Costar, Inc. A complete description can be found in Hubert et al. (6). A simplified version of this instrumentation can be constructed without the digital camera or a computer program; the displacement of fluid meniscus in either the inflow or outflow capillary tube can be measured and used to calculate Lp. 7. The benefits of using a cell culture system to quantify Lp is that the surface is known (not true in the isolated perfused lung prep) and cell culture provides enhanced control of the
20 Mechano-Transduction and Barrier Regulation…
301
microenvironment (pressure, culture media, oncotic pressure, etc.) and the absence of other cell types that could have autocrine or paracrine effects on the endothelium. We have demonstrated many physiological similarities of cultured monolayers compared to the in vivo endothelium, including similar Lp values (7), similar response to shear stress via nitric oxide production(8, 9), and similar permeability response to pressure (10–12). 8. There are inflammatory states that result is significant endothelial barrier alterations that are sufficient in magnitude to allow enhanced passage of macromolecules out of the vascular compartment. Since albumin is the most abundant plasma protein and it is responsible for the majority of the plasma oncotic pressure, it is logical to use albumin as a tracer. Nonetheless, other proteins and macromolecular tracers can be used to suit the experimental conditions. The diffusion rate of the tracer across the endothelial monolayer should always be quantified as a permeability coefficient (P; cm/s). References 1. Uhlig, S. and von Bethmann, A. N. (1997) Determination of vascular compliance, interstitial compliance, and capillary filtration coefficient in rat isolated perfused lungs. J Pharmacol Toxicol Methods. 37, 119–127 2. Parker, J. C. and Ivey, C. L. (1997) Isoproterenol attenuates high vascular pressure-induced permeability increases in isolated rat lungs. J Appl Physiol. 83, 1962–1967 3. Uhlig, S. and Wollin, L. (1994) An improved setup for the isolated perfused rat lung. J Pharmacol Toxicol Methods. 31, 85–94 4. Bhattacharya, J. (2007) Interpreting the lung microvascular filtration coefficient. Am J Physiol Lung Cell Mol Physiol. 293, L9-L10 5. Parker, J. C. and Townsley, M. I. (2004) Evaluation of lung injury in rats and mice. Am J Physiol Lung Cell Mol Physiol. 286, L231–246 6. Hubert, C. G., McJames, S. W., Mecham, I. and Dull, R. O. (2006) Digital imaging system and virtual instrument platform for measuring hydraulic conductivity of vascular endothelial monolayers. Microvasc Res. 71, 135–140 7. Dull, R. O., Jo, H., Sill, H., Hollis, T. M. and Tarbell, J. M. (1991) The effect of varying albumin concentration and hydrostatic pressure on hydraulic conductivity and albumin
permeability of cultured endothelial monolayers. Microvasc Res. 41, 390–407 8. Florian, J. A., Kosky, J. R., Ainslie, K., Pang, Z., Dull, R. O. and Tarbell, J. M. (2003) Heparan sulfate proteoglycan is a mechanosensor on endothelial cells. Circ Res. 93, e136–142 9. Pahakis, M. Y., Kosky, J. R., Dull, R. O. and Tarbell, J. M. (2007) The role of endothelial glycocalyx components in mechanotransduction of fluid shear stress. Biochem Biophys Res Commun. 355, 228–233 10. Dull, R. O., Mecham, I. and McJames, S. (2007) Heparan sulfates mediate pressureinduced increase in lung endothelial hydraulic conductivity via nitric oxide/reactive oxygen species. American Journal of Physiology-Lung Cellular, and Molecular Physiology. 292, L1452–L1458 11. Kim, M. H., Harris, N. R. and Tarbell, J. M. (2005) Regulation of capillary hydraulic conductivity in response to an acute change in shear. Am J Physiol Heart Circ Physiol. 289, H2126–2135 12. Kim, M. H., Harris, N. R. and Tarbell, J. M. (2005) Regulation of hydraulic conductivity in response to sustained changes in pressure. Am J Physiol Heart Circ Physiol. 289, H2551–2558
wwwwwwwwwwwww
Chapter 21 Role of Caveolin-1 in the Regulation of Pulmonary Endothelial Permeability Yu Sun, Richard D. Minshall, and Guochang Hu Abstract Endothelial permeability measurements of intact vascular beds and monolayer cultures are used to describe transport of small molecules (ions, water, and nutrients), macromolecules and plasma protein across the vascular endothelia. Disruption of the endothelial barrier leads to vascular hyperpermeability and protein-rich edema which is a key hallmark of inflammation. Transport of the most abundant plasma protein, albumin, occurs by means of transcellular and paracellular pathways. In healthy, noninflamed vessels, endothelial cell–cell contacts significantly restrict the paracellular permeability of albumin, whereas its transcellular transport from the blood to the abluminal perivascular interstitium occurs via caveolae. Thus, caveolae-mediated transport is a primary determinant of the basal endothelial permeability properties. Increased paracellular permeability induced during inflammation is thought to be due to the opening of interendothelial cell–cell junctions and disruption of endothelial cell–matrix contacts within the vasculature. We recently demonstrated that caveolae-mediated transendothelial transport (transcytosis) of macromolecules through the microvascular endothelial barrier is also an important mechanism responsible for inflammation-evoked pulmonary vascular hyperpermeability and protein-rich edema formation. Moreover, caveolin-1, a structural and scaffolding protein required for caveolae formation and transcellular transport, also plays an important role in oxidant-induced paracellular hyperpermeability. This review highlights the methods used to assess transcellular and paracellular permeability properties of the intact mouse lung and cultured endothelial cell monolayers. Key words: Endothelial permeability, Transcellular, Paracellular, Caveolin-1, Lung
1. Introduction The vascular endothelium lining blood vessels functions as a semipermeable barrier between the blood and interstitial compartments and thus controls the transendothelial flux of fluid and macromolecules (1–4). Vascular leakage secondary to pulmonary endothelial hyperpermeability leads to extravasation of fluids and proteins from the capillary vessels into the tissues resulting in Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_21, © Springer Science+Business Media, LLC 2011
303
304
Y. Sun et al.
interstitial edema, a decrease in tissue perfusion, and organ dysfunction. Endothelial permeability to the primary plasma protein albumin can be classified as either transcellular (through cells) or paracellular (between cells). Transcellular permeability depends on the formation, fission, and trafficking of caveolae (1–4). Vesicular permeability via caveolae is a determinant of basal endothelial permeability. Paracellular permeability is regulated by protein complexes that make up adherens junctions (AJ) and tight junctions (TJ) between neighboring cells, and focal adhesions that link cells to substratum. During inflammation, paracellular endothelial permeability is increased due to intercellular gap formation resulting from activation of actin–myosin-based cell retraction and the loss of cell–cell adhesion (1–5). Caveolin-1 (Cav-1), the 22-kDa membrane-associated and cholesterol binding protein, is the principal structural and regulatory component of caveolae. An increase in transcellular permeability was shown to be dependent on Src-mediated phosphorylation of Cav-1 (6–13). We showed that pulmonary vascular hyperpermeability induced by activation of neutrophils adherent to the vessel wall is dependent on signaling via Cav-1 and increased caveolae-mediated transcytosis (14). Recent evidence also points to the important role of Cav-1 in the regulation of paracellular permeability. Downregulation of Cav-1 leads to a loss and redistribution of tight junction proteins (occludin and ZO-1) in brain microvascular endothelial cells (15) and induces an increase in the number of interendothelial gaps in pulmonary capillaries and veins (16). In contrast, deletion of Cav-1-attenuated protein kinase C induced interendothelial gap formation in myocardial microvascular endothelial cells (17). In this study, using genetic approaches, we investigated the role of Cav-1 in pulmonary microvascular permeability regulation through both transcellular and paracellular pathways. Our results demonstrate that phosphorylation of Cav-1 is an important mechanism mediating oxidant-induced vascular hyperpermeability by stimulating paracellular and caveolae-mediated transcellular permeability (18). Therefore, therapeutic inhibition of Cav-1 phosphorylation may be an effective target for limiting lung vascular hyperpermeability and injury.
2. Materials 2.1. Endocytosis of 125I-Albumin and Transendothelial 125 I-Albumin Transcytosis
1. Rat lung microvascular endothelial cells (RLMVECs) are obtained from Vec Technologies (Rensselaer, NY) and cultured on fibronectin-coated dishes in MCDB-131 complete medium supplemented with 10% fetal bovine serum (FBS). 2. 30% (v/v) Hydrogen peroxide (H2O2) is obtained from Fisher Scientific.
21 Role of Caveolin-1 in the Regulation of Pulmonary Endothelial Permeability
305
3. 1× Hank’s Balanced Salt Solution (HBSS): HBSS (powder; Sigma); dissolve in 900 mL ddH2O. 1.16 g Na-HEPES. 1.32 g HEPES. 0.35 g sodium bicarbonate (NaHCO3). Adjust pH to 7.4 and bring to 1,000 mL with ddH2O. Sterilize by passing through a 0.22-mm filter and store at 4°C. 4. 125I-albumin is purchased (Wellesley, MA).
from
PerkinElmer,
Inc.
5. Bovine serum albumin (BSA, fraction V, 99% pure, endotoxin free, cold alcohol precipitated) is obtained from SigmaAldrich. 0.1 and 100 mg/mL of BSA (unlabeled albumin) solutions are prepared in HBSS. 6. Acid wash buffer: 0.5 M NaCl and 0.2 M acetic acid, pH 2.5. 7. Tris–HCl buffer: 0.05 M Tris–HCl, 1% Triton X-100 and 0.5% SDS, pH 7.4. 8. Gamma counter (Packard Instruments, Downers Grove, IL). 9. Six-well tissue culture plates for endocytosis measurement are obtained from Becton Dickinson Labware (Franklin Lakes, NJ). 10. 12-mm polyester (PET) membrane transwell-clear filter inserts (12 wells, 1.12 cm2 growth area, 0.4-mm pore size) for transcytosis measurement are obtained form Corning Costar (Cambridge, MA). 2.2. Fluorescent Albumin Uptake and Confocal Imaging of Paracellular Gap Formation
1. 12-mm image quality cover glass from Fisher Scientific. 2. 2× HBSS: HBSS (powder; Sigma); dissolve in 400 mL ddH2O. 1.16 g Na-HEPES. 1.32 g HEPES. 0.35 g sodium bicarbonate (NaHCO3). Adjust pH to 7.4 and bring to 500 mL with ddH2O. Sterilize by passing through 0.22-mm filter and store at 4°C. Dilute 2× HBSS with equal volume of ddH2O to make 1× HBSS. 3. 4% paraformaldehyde: Dilute 16% paraformaldehyde (Electron microscopy sciences) to 8% with equal volume of 2× HBSS. Then dilute 8% paraformaldehyde to 4% with equal volume of 1× HBSS. 4. Quench solution: 100 mM glycine in 1× HBSS. 5. Permeabilization solution: 0.1% (v/v) Triton X-100 in HBSS. 6. Blocking solution and antibody dilution buffer: 5% (v/v) goat serum, 0.2% (w/v) BSA, 0.1% (w/v) sodium azide in HBSS.
306
Y. Sun et al.
7. Alexa 488-conjugated albumin, Alexa Fluor 546 phalloidin, and goat anti-mouse Alexa 488 were obtained from Molecular Probes; b-catenin monoclonal antibody is from BD Laboratories. 8. Nuclear stain: 1 mg/mL 4,6-diamidino-2-phenylindole (DAPI, Molecular Probes) in 1× HBSS. 9. Mounting medium: ProLong antifade mounting medium (Molecular Probes). 2.3. Transendothelial Electrical Resistance
1. ECIS cultureware (disposable electrode arrays, 8W10E) are from Applied Biophysics. 2. ECIS 1600R controller and CO2 incubator.
2.4. Albumin Permeability–Surface area (PS) Product Measurement
1. Caveolin-1 null (cav-1−/−) mice and wild-type B6/129SJ mice are purchased from Jackson Laboratory (Bar Harbor, ME). Mice are housed in microisolator cages under specific pathogen-free conditions, fed with autoclaved food, and used in experiments at 8–12 weeks of age. All studies using mice are approved by the University of Illinois Institutional Animal Care and Use Committee. 2. Modified Krebs–Henseleit solution: 118 mM NaCl, 4.7 mM KCl, 1.0 mM CaCl2, 1.0 mM MgCl2, 5.0 mM HEPES, 11 mM glucose, 0.025 mM EDTA (pH 7.35–7.45), and BSA (5 g/100 mL). 3. Surgical instruments (Fisher Scientific). 4. Peristaltic pump (Gilson Minipuls 3). 5. Isotemp 2100 Immersion circulating water bath (Fisher Scientific). 6. Water-jacketed bubble trap (Radnoti). 7. CP122 amplifier (Grass Instruments). 8. P23 XL pressure transducer (Grass Instruments). 9. FT03C force-displacement transducer (Grass Instruments). 10. Telefactor (Grass Instruments).
3. Methods Cultured endothelial cells represent an effective model of intact endothelium and can be analyzed for basal and stimulated changes in permeability. One well-established method utilizes albumin labeled with 125I as a tracer. The albumin transport protocol monitors the transfer of albumin tracer through the cell monolayer grown on filter inserts and into the abluminal chamber. In this
21 Role of Caveolin-1 in the Regulation of Pulmonary Endothelial Permeability
307
assay, the back-flux of tracer albumin from the abluminal compartment is minimal because of dilution of the volume in the abluminal well and similar hydrostatic pressure between two chambers, thus allowing a measure of unidirectional albumin flux (19). Excess unlabeled albumin (100 mg/mL) is used to compete with the tracer and thus specifically block transcellular transport, establishing the component of tracer albumin transport that is mediated by the vesicular permeability pathway and the paracellular or junctional permeability. Measurement of transcellular permeability is only possible when paracellular permeability (tracer leakage through interendothelial gaps that cannot be blocked by excess unlabeled albumin) is negligible. Monolayer integrity and formation of interendothelial gaps can also be confirmed by measuring transendothelial electrical impedance, a measure of interjunctional dimensions in confluent endothelial cell monolayers. To assess transcellular albumin permeability, it is also useful to measure caveolae-mediated endocytosis of 125I-albumin in endothelial cells, which is the initial step in albumin transport via transcytosis. Paracellular permeability of tracer albumin can be measured in the presence of a selective inhibition of caveolae-mediated transcytosis. Transvascular transport of 125I-albumin is more accurately used to determine vascular permeability in isolated lungs than in intact animals (20). Back flux of 125I-albumin from the pulmonary interstitium is considered negligible because of the brevity of the labeling period relative to the uptake time constant for albumin. Therefore, the uptake of 125 I-albumin is taken as a unidirectional protein flux across the endothelial barrier and will predominantly represent tracer albumin that has been transported via the transcellular pathway or has leaked into the tissue via the paracellular pathway. Transcellular permeability in isolated lung preparations can be determined when paracellular permeability is negligible as assessed by monitoring the flux of the small molecule paracellular tracer 3H-mannitol (14). Since in vitro methods may not accurately represent the in vivo condition, the most productive studies of lung permeability make use of at least two methods carefully selected according to their strengths and weaknesses. 3.1. Transcellular Permeability 3.1.1. Endocytosis of 125I-Albumin
1. Plate RLMVECs at a density of 1.6 × 105 cells/well in six-well plates. The cells will approach confluence after 48 h. At this time, the cultures should be rinsed twice and incubated for 5 h in serum-free basic MCDB-131 medium. 2. RLMVEC monolayers are pretreated with H2O2 at 37°C for 30 min before addition of 125I-albumin in the presence or absence of unlabeled albumin in HBSS. 3. The activity of 125I-albumin should be determined immediately before use. Dilute stock 125I-albumin solution to 2 × 105 cpm/mL (see Notes 1 and 2).
308
Y. Sun et al.
4. Unlabeled albumin in HBSS is added at a final concentration of 0.1 mg/mL in four wells of the six-well plate and 100 mg/mL in the other two wells. 5 mL of diluted 125I-albumin (1 × 106 cpm) is then added into each well and incubated at 37°C for 30 min. 5. Endocytosis is terminated by chilling the cells on ice and washing each well three times with ice-cold acid wash buffer to remove cell surface-bound 125I-albumin. 6. The cells are washed another three times with ice-cold HBSS and observed by phase contrast microscope. Washing the cells carefully to avoid removing cells from the well is critically important; areas of cell loss in each well should be recorded (see Note 3). 7. The adherent cells are then lysed with 1 mL Tris–HCl buffer for 5 min. 8. The cell lysate is transferred to labeled plastic test tubes and the radioactivity determined using a gamma counter within 1 h of collection (see Notes 1 and 2). 9. Specific uptake of 125I-albumin is estimated by subtracting the nonspecific cell-associated activity (determined in the presence of 100 mg/mL unlabeled albumin) from the total (determined in the presence of 0.1 mg/mL unlabeled albumin). 10. Specific uptake is normalized to total cell protein which can be determined by bicinchoninic acid (BCA) method and expressed in units of cpm/mg cell protein. An example of the results produced is shown in Fig. 1a. 3.1.2. Transendothelial 125 I-Albumin Permeability
1. Microporous polyester (PET) Transwell-clear membranes (1.12 cm2 growth area) are coated with 200 mL fibronectin for 2 h and rinsed with HBSS. 2. An initial equilibration period is used to improve cell attachment by adding 1.5 mL MCDB-131 complete medium to the multiple well plate well (lower chamber) and then 500 mL to the Transwell insert (upper chamber). The plates are then incubated for 2 h at 37°C. 3. Aspirate the medium in upper and lower chamber and then add fresh complete medium into lower chamber. Prepare RLMVECs at a density of 3.6–4.0 × 104 cells/mL and plate (in a total volume of 500 mL) drop-wise gently into each upper chamber (1.8–2.0 × 104 cells) and allow the cells to settle and attach for about 5–10 min. Do not swirl plates to avoid cell clustering. 4. Incubate the plated cells for 48 h or until confluent and then change the complete medium to serum-free basic MCDB131 medium for 5 h before measuring albumin permeability.
21 Role of Caveolin-1 in the Regulation of Pulmonary Endothelial Permeability
b 400
*
300 200
†
100
*
0 WT
c
Control H2O2 (0.2 mM)
Control
Y14F
Transendothelial 125I-Albumin Permeability (% of control)
125
I-Albumin Endocytosis (% of control)
a
309
H2O2 (0.05 mM)
400
*
Control H2O2 (0.2 mM)
300 200
†
100 0
* WT
Y14F
H2O2 (0.2 mM)
Fig. 1. Phosphorylation of caveolin-1 is required for H2O2-induced endocytosis and transendothelial albumin transport. (a, b) Effect of H2O2 on 125I-albumin endocytosis (a) and transendothelial albumin permeability (b) in cells stably expressing wild-type (WT) and phosphorylation-defective Y14F-Cav-1 mutant. n = 4–6 for each group. The baseline endocytosis value for WT control group was 2.1 ± 0.15 cpm/mg cell protein × 104 (a). The baseline permeability values for WT control group was 7.8 ± 1.2 mL/min/cm2 × 10−2 (b). *p < 0.05 compared with control groups, †p < 0.05 compared with respective control groups. (c) Confocal images showing H2O2-induced concentration-dependent increase in the uptake of Alexa 488-labeled albumin (green). The nucleus (blue) was stained with DAPI. Scale bars = 10 mm. Results are typical of three experiments ( ref. 18).
5. RLMVEC monolayers are pretreated with H2O2 at 37°C for 30 min before addition of 125I-albumin in the presence or absence of unlabeled albumin in HBSS. 6. The activity of 125I-albumin should be assessed before use. Dilute stock 125I-albumin solution to 2 × 105 cpm/mL and number new plastic test tubes (see Notes 1 and 2). 7. 0.5 mL of unlabeled albumin in HBSS is added at a final concentration of 0.1 mg/mL in four upper chambers and 100 mg/mL in the other two upper chambers. The lower chambers are filled with 1.5 mL of HBSS containing unlabeled albumin of the same osmolarity as the upper chambers, thus fluid levels and osmotic pressure in the “upper” and “lower” wells are equal to eliminate hydrostatic and osmotic pressure differences across the monolayer. 8. 5 mL of diluted 125I-albumin (total of 1 × 106 cpm) is then added into each upper chamber and incubated at 37°C for
310
Y. Sun et al.
30 min. Transcytosis is terminated by chilling on ice and aspirating the liquid in the upper chambers. 9. The cells on the upper chamber are observed optically to make sure that there is no cell detachment in the individual wells. 10. Aliquots of 200 mL are sampled from the lower chamber and transferred to the labeled test tubes. Leave the used pipette tips inside the tube. 11. Radioactivity of the samples is measured using a gamma counter within 1 h of collection (see Notes 1 and 2). 12. Specific transcytosis of 125I-albumin is estimated by subtracting the nonspecific cell-associated activity (determined in the presence of 100 mg/mL unlabeled albumin) from the total (determined in the presence of 0.1 mg/mL unlabeled albumin) (see Note 4). 13. Transendothelial 125I-albumin permeability is calculated from the flux of radiolabeled albumin across the cell monolayer and is expressed in units of mL/min/cm2. An example result is shown in Fig. 1b. 3.1.3. Fluorescent Albumin Uptake
1. Prepare sterilized 12-mm glass coverslips by autoclaving in a glass beaker with pieces of tissue paper. Using a forcep, pass individual coverslips through a flame once and place at the bottom of each well in a 24-well plate; perform this in the sterile hood. 2. The coverslips are then coated with 110 mL fibronectin and incubated for 1 h and then aspirated. Prepare RLMVECs at a density of 2.0–2.5 × 105 cells/mL and place a 110-mL drop of cell suspension (2.2–2.75 × 104 cells) onto each coverslip to allow to spread to, but not beyond, the edges of the coverslip. Allow cells to settle and attach for 20 min and then add 1 mL of MCDB-131 complete medium to each well. 3. The cells are cultured for 48 h in MCDB-131 complete medium to confluence and then incubated for 5 h in serumfree basic medium after rinsing with PBS. 4. RLMVEC monolayers are pretreated with H2O2 at 37°C for 30 min. 5. Monolayers are then incubated with fluorescently tagged albumin (Alexa 488-labeled bovine serum albumin; 50 mg/mL) containing 0.5 mg/mL unlabeled albumin in basic medium for 30 min at 37°C. From this step, the plates should be covered with aluminum foil and kept in dark. 6. The cells are then washed three times with ice-cold acid wash buffer and three times with HBSS to remove unincorporated tracer.
21 Role of Caveolin-1 in the Regulation of Pulmonary Endothelial Permeability
311
7. A sufficient volume (400 mL) of 4% paraformaldehyde is added to completely cover the coverslips and incubated for 30 min at room temperature on a rocking platform to fix the cells (see Note 5). 8. The paraformaldehyde is removed and the residual is quenched by rinsing in 100 mM glycine three times for 5 min each, followed by three washes with HBSS for 10 min each at room temperature on the rocking platform. 9. DAPI (1 mg/mL) is added for 20 min to stain the DNA and identify nuclei. Wash the samples three times for 10 min each with HBSS. 10. The samples are then ready to be mounted. Add one drop of mounting medium to a glass microscopy slide just before removing the coverslip from the plate. Invert each coverslip carefully onto a drop of mounting medium. Try to avoid trapping air bubbles. 11. The samples are stored in the dark to dry at room temperature overnight and then can be stored at −20°C in the dark for up to several months. 12. The slides are viewed by a laser-scanning confocal microscope (Zeiss LSM 510 META). Internalized fluorescent albumin (green) is imaged by confocal microscopy in optical sections midway through the cell (pinhole set to achieve 1 Airy unit). Examples of representative images are shown in Fig. 1a. 3.2. Paracellular Permeability 3.2.1. Assessment of Barrier Function by Transendothelial Electrical Resistance
1. Prepare RLMVECs at a density of 5.0–6.0 × 104 cells/mL and plate 400 mL (2.0–2.4 × 104 cells) as drops onto fibronectin-coated ECIS wells containing gold-plated microelectrodes. Do not move and let sit for approximately 10 min for cell attachment, and then transfer the cells to the incubator. 2. The cells are cultured for 48 h to reach confluence and then incubated in reduced serum MCDB-131 medium (0.5% FBS) for 1 h. 3. Change the medium to fresh reduced serum medium (280 mL). Place array slide on holder in an incubator and start ECIS model 1600R software. Check electrode; resistance values should be approximately 800–1,000 W for confluent cells with the one-electrode arrays (8W10E). 4. Run attachment measurements and set up the program. The small electrode and the larger counter electrode are connected to a phase-sensitive lock-in amplifier, and the impedance and resistance are continually monitored, stored, and processed
312
Y. Sun et al. H2O2 (mM)
a
Control WT+0.2 WT+0.6
1.0
H2O2 (mM) 1.1
0.9 0.8 0.7 0.6 0.5 0.0
b
TER (relative to baseline)
TER (relative to baseline)
1.1
Y14F+0.2 Y14F+0.6
0.5
1.0 Time (h)
1.5
2.0
0 0.2 Control WT
0.6 WT
1.0
†
0.9 0.8 0.7 0.6 0.5
0.2 0.6 Y14F Y14F
†
* *
Fig. 2. Phosphorylation of Cav-1 is required for signaling H2O2-induced endothelial barrier disruption. RLMVECs were grown to confluence, treated with H2O2 at the concentration indicated, and TER was recorded. (a) The original TER recordings are shown. (b) The mean value (±SEM) of the peak TER responses to H2O2 (relative to the starting value) is plotted. Phospho-defective Y14F Cav-1 mutant blocks the effect of H2O2 on TER. Overexpression of WT-Cav-1 in RLMVECs reduced the threshold of H2O2 needed to decrease TER. n = 4–6 for each group. *P < 0.05, compared to control (untreated) groups; † P < 0.05 compared with respective H2O2 groups (see ref. 18).
with a personal computer. The baseline transendothelial electrical resistance (TER) measurements are recorded. 5. TER is monitored for approximately 30 min to establish a stable baseline resistance. 6. Prepare different concentrations of H2O2 stock solutions; gently add 20 mL to each well (total volume of 300 mL/well). Subsequent changes in TER are monitored for about 5 h and then the data are saved to disc. 7. Results are presented as the change in the resistive (in-phase) portion of the impedance normalized to its initial value at time zero (Fig. 2). 3.2.2. Confocal Immunofluorescent Staining of F-Actin and b-Catenin to Assess Interendothelial Gap Formation
1. The first three steps are similar to the method for fluorescent albumin uptake. 2. The treated cells are rinsed twice rapidly with ice-cold HBSS. 400 mL of 4% paraformaldehyde solution is then added and cells are incubated for 30 min at room temperature on a rocking platform to fix the cells (see Note 5). 3. The paraformaldehyde is discarded and the residual quenched by rinsing the cells with 100 mM glycine three times for 5 min each, followed by three washes with HBSS for 10 min each at room temperature on the rocking platform. 4. The cells are permeabilized by incubation in permeabilization solution for 10 min at room temperature and then blocked with block solution for about 1 h.
21 Role of Caveolin-1 in the Regulation of Pulmonary Endothelial Permeability
313
5. The samples are incubated with primary antibody (1 mg/mL) diluted in antibody dilution buffer overnight at 4°C. 6. The primary antibody is removed and the sample is washed three times for 10 min each with HBSS. 7. Cells are blocked again with blocking solution for 1 h. Secondary antibody is prepared at 1:250 in antibody dilution buffer and added to the samples for 2 h at room temperature. From this step, the sample should be covered with aluminum foil and kept in dark. 8. After incubation with secondary antibody for 1 h, add 5 mL Alexa Fluor 546 phalloidin and incubate samples for another 1 h. 9. Nuclei are stained with DAPI. The coverslips are finally mounted (see Subheading 3.1.3, steps 8–10 for details). 10. The slides are viewed by a laser-scanning confocal microscope [Zeiss LSM 510 META using Hg lamp and UV-filter set to detect DAPI (band pass (BP) 385–470 nm emission)], 488 nm excitation laser line to Alexa 488 (BP 505–550 nm emission) and 543 nm excitation laser line to excite Alexa 546, respectively. An example of results is shown in Fig. 3. 3.3. Albumin Permeability–Surface Area (PS) Product Measurement 3.3.1. Lung Preparation
1. Mice are anesthetized by intraperitoneal injection of 3 mg/kg xylazine and 75 mg/kg ketamine and ventilated with 7 mL/kg of tidal volume, a respiratory rate of 120 breath/min, 0 cmH2O end expiratory pressure, and FIO2 of 0.21, using a rodent ventilator (MiniVent, Harvard Biosciences; Holliston, MA). 2. A thoracotomy is performed and the pulmonary artery cannulated in situ, left atrium removed and the lungs perfused with a modified Krebs–Henseleit solution. The lung and heart (En Bloc) are removed and mounted on a perfusion apparatus. 3. The lung preparation is perfused via the pulmonary artery at a constant flow (2 mL/min) and venous pressure (3 cmH2O), and pulmonary artery pressures of 8 ± 2 cmH2O. 4. Pulmonary arterial pressure and lung weight are continuously monitored during experiments. Labtech software (Andover, MA) is used to control data acquisition and storage.
3.3.2. Albumin PS Product Measurement
1. The isolated lung is infused by a side-arm port with either saline or H2O2 (0.5 mM) for 30 min. 2. At the end of experiment, the isolated lung is infused by a side-arm port with 125I-albumin (~80,000 counts/mL) for 3 min (see Notes 1 and 2). 3. The lung is given a washout period of 6 min with Krebs solution containing unlabeled albumin (5 g/100 mL) to remove cell surface and circulating 125I-albumin.
314
Y. Sun et al.
Fig. 3. Immunofluorescent staining of b-catenin and F-actin in RLMVECs overexpressing WT vs. Y14F-Cav-1. Cells were grown to confluence and treated with H2O2 (0.2 mM) for 30 min. After fixation and permeabilization, cells were incubated with anti-b-catenin primary antibody followed by FITC-conjugated secondary antibody (green). The F-actin and nucleus were stained with Alexa 546 phalloidin (red) and DAPI (blue), respectively. In the third panel (WT + 0.2 mM H2O2), note gap formation (arrow). The fourth panel (Y14F-Cav-1 + 0.2 mM H2O2) shows no gap formation in response to H2O2. Scale bars = 10 mm. WT = wild type-caveolin-1 expressing cells; Y14F = Y14F-caveolin-1 mutant expressing cells (see ref. 18).
4. The lung is detached from the perfusion apparatus, and each lobe dissected, cleaned of connective tissue, rinsed, and blotted. Each sample is weighed and counted for gamma radioactivity.
Pulmonary Microvessel 125IAlbumin Permeability (µl·min−1·g−1)
21 Role of Caveolin-1 in the Regulation of Pulmonary Endothelial Permeability 70
*
315
*
60 50 40 30
*†
20
*†
*†
10 0
Cav-1 cDNA H2O2(0.5 mM)
− −
− +
cav-1+/+
− −
− +
WT
Y14F
+
+
cav-1 −/−
Fig. 4. Effect of H2O2 on pulmonary microvessel 125I-albumin permeability (PS product) in wild-type (cav-1+/+) and cav-1−/− lungs with or without rescue with WT-Cav-1 or Y14FCav-1 mutant cDNA. Cav-1−/− mice were injected intravenously with liposomes containing WT- or Y14F-Cav-1 cDNA. After 48 h, lungs were isolated and perfused with H2O2 (0.5 mM) for 30 min. H2O2 induced an increase in 125I-albumin permeability (PS product) in the isolated lung. In contrast, in Cav-1−/− mouse lungs, which exhibit reduced basal albumin PS, H2O2 did not induce an increase in albumin permeability. H2O2 induced an increase in albumin PS product in WT-Cav-1 expressing mouse lungs, whereas this effect was not reconstituted upon expression of Y14F-Cav-1 in Cav-1−/− lungs. n = 6/ each group. *P < 0.05 compared with control groups (cav-1+/+ mouse without H2O2 treatment); †P < 0.05 compared with cav-1+/+ mouse with H2O2 treatment group (see ref. 18).
5. PS product (in mL/min/g dry lung) is calculated with the formula A/(Cpt), where A and Cp are concentrations of tracer albumin in the tissue (in counts/g) and in the perfusate (in counts/mL), respectively, and t is the perfusion time for tracer albumin (3 min) (see Fig. 4).
4. Notes 1. When handling 125I: (a) Wear gloves and carry out experiment in a fume hood to avoid skin contamination and inhalation. (b) Use shielding (lead or leaded Plexiglas) to minimize radioactive exposure. (c) Cover test tubes used to count or separate fractions from iodination with parafilm or other tight caps to prevent release while counting or moving outside the fume hood. (d) Discard the radioactive solutions carefully in the sink and wash away tracer residue with tap water.
316
Y. Sun et al.
2. If the 125I-albumin is stored for a relatively long time, free 125I is generated and will affect the accuracy of the experimental measurement. Trichloroacetic acid precipitation analysis is used to confirm the purity; free 125I can be removed from 125I-labeled albumin with a Sephadex G25 column such that the contaminant-free 125I in the tracer contributes <0.3% of the total counts. 3. Equal number of cells in each well when measuring “Endocytosis of 125I-albumin” is very important for reproducibility of results. Before radioactivity of the cells is measured, the cells should be observed with the optical microscope and the area of cell loss in each well should be recorded. 4. For measurement of transcellular 125I-albumin permeability, the cells should reach confluence immediately prior to the experiment. If the nonspecific value (higher concentration of unlabeled albumin) is too high, it is possible that free 125I is generated or cells do not reach complete confluence. 5. Methanol can disrupt actin during the fixation process. Therefore, it is best to avoid any methanol containing fixatives.
Acknowledgments The authors thank David J. Visintine for technical assistance. This work was supported by American Heart Association Scientist Development Grant 0730331N (G.H.) and NIH NHLBI grants 1R01HL104092 (G.H.) and 5R01 HL071626 and P01 HL060678 (R.D.M). References 1. Hu, G. and Minshall, R.D. (2009) Cell Biology of Pulmonary Endothelial Permeability, in The Pulmonary Endothelium. Eds, S. Rounds and N. Voelkel, Wiley Inc., New York, pp113–127. 2. Mehta, D., Malik, A.B. (2006) Signaling mechanisms regulating endothelial permeability. Physiol Rev. 86:279–367. 3. Hu, G., Minshall, R.D. (2009) Regulation of transendothelial permeability by Src kinase. Microvasc Res. 77:21–25. 4. Hu, G., Place, A.T., Minshall, R.D. (2008) Regulation of endothelial permeability by Src kinase signaling: vascular leakage versus transcellular transport of drugs and macromolecules. Chem Biol Interact. 171:177–189. 5. Dejana, E. (2004) Endothelial cell-cell junctions: happy together. Nat Rev Mol Cell Biol. 5:261–270.
6. Minshall, R.D., Tiruppathi, C., Vogel, S.M., Malik, A.B. (2002) Vesicle formation and trafficking in endothelial cells and regulation of endothelial barrier function. Histochem Cell Biol. 117:105–112. 7. Minshall, R.D., Sessa, W.C., Stan, R.V., Anderson, R.G., Malik, A.B. (2003) Caveolin regulation of endothelial function. Am J Physiol. 285:L1179–L1183. 8. Shajahan, A.N., Timblin, B.K., Sandoval, R., Tiruppathi, C., Malik, A.B., Minshall, R.D. (2004) Role of Src-induced dynamin-2 phosphorylation in caveolae-mediated endocytosis in endothelial cells. J Biol Chem. 279:20392–20400. 9. Minshall, R.D., Tiruppathi, C., Vogel, S.M., Niles, W.D., Gilchrist, A., Hamm, H.E., et al. (2000) Endothelial cell-surface gp60 activates vesicle formation and trafficking via Gi-coupled
21 Role of Caveolin-1 in the Regulation of Pulmonary Endothelial Permeability Src kinase signaling pathway. J Cell Biol. 150:1057–1070. 10. Shajahan, A.N., Tiruppathi, C., Smrcka, A.V., Malik, A.B., Minshall, R.D. (2004) Gbgactivation of Src induces caveolae-mediated endocytosis in endothelial cells. J Biol Chem. 279:48055– 48062. 11. Drab, M., Verkade, P., Elger, M., Kasper, M., Lohn, M., Lauterbach, B., et al. (2001) Loss of caveolae, vascular dysfunction, and pulmonary defects in caveolin-1 gene-disrupted mice. Science. 293:2449 –2452. 12. Parton, R.G., Joggerst, B., Simons, K. (1994) Regulated internalization of caveolae. J Cell Biol. 127:1199 –1215. 13. Rothberg, K.G., Heuser, J.E., Donzell, W.C., Ying, Y-S., Glenney, J.R., Anderson, R.G. (1992) Caveolin, a protein component of caveolae membrane coats. Cell. 68:673– 682. 14. Hu, G., Vogel, S.M., Schwartz, D.E., Malik, A.B., Minshall, R.D. (2008) Intercellular adhesion molecule-1-dependent neutrophil adhesion to endothelial cells induces caveolae-mediated pulmonary vascular hyperpermeability. Circ Res. 102:e120–131. 15. Song, L., Ge, S., Pachter, J.S. (2007) Caveolin-1 regulates expression of junction
317
associated proteins in brain microvascular endothelial cells. Blood. 109:1515–1523. 16. Miyawaki-Shimizu, K., Predescu, D., Shimizu, J., Broman, M., Predescu, S., Malik, A.B. (2006) siRNA-induced caveolin-1 knockdown in mice increases lung vascular permeability via the junctional pathway. Am J Physiol. 290:L405–L413. 17. Waschke, J., Golenhofen, N., Kurzchalia, T.V., Drenckhahn, D. (2006) Protein kinase C-mediated endothelial barrier regulation is caveolin-1- dependent. Histochem Cell Biol. 126:17–26. 18. Sun, Y., Hu, G., Zhang, X., Minshall, R.D. (2009) Phosphorylation of caveolin-1 regulates oxidant-induced pulmonary vascular permeability via paracellular and transcellular pathways. Circ Res. 105:676-685. 1 9. John, T.A., Vogel, S.M., Tiruppathi, C., Malik, A.B., Minshall, R.D. (2003) Quantitative analysis of albumin uptake and transport in the rat microvessel endothelial monolayer. Am J Physiol. 284:L187–196. 20. Parker, J.C., Townsley, M.I. (2004) Evaluation of lung injury in rats and mice. Am J Physiol. 286:L231–246.
wwwwwwwwwwwww
Chapter 22 Assessment of Endothelial Permeability and Leukocyte Transmigration in Human Endothelial Cell Monolayers Andreas Ludwig, Anselm Sommer, and Stefan Uhlig Abstract Increased vascular permeability is the hallmark of inflammation. Here, we describe three methods to assess vascular permeability in cell culture: (1) Impedance measurements of endothelial cell monolayers that allow to monitor changes in cell shape in real time. (2) Diffusion of fluorescently labeled dextran across endothelial cell monolayers to identify openings large enough for bulky molecules. (3) Transmigration of neutrophils through confluent endothelial cell monolayers to study one major process that increases endothelial permeability in inflammation. Key words: Endothelial permeability, Impedance, Thrombin, Interleukin-8, Neutrophil, Dextran, Inmpedance, Neutrophil transmigration
1. Introduction Inflammation is a principal response of the organism to deterioration of its interior milieu (1). One hallmark of inflammation is increased vascular permeability leading to swelling (“tumor”), a classical symptom already observed by Cornelius Celsus more than 2,000 years ago. Probably, the swelling is the price that the organism has to pay for bringing leukocytes to the sites of injury. Many aspects of vascular leakage can be studied in confluent endothelial monolayers in culture, where critical parts of the machinery involved in the regulation of vascular permeability are maintained: the cell–cell junctions (tight junctions, adherens junctions, gap junctions) and the contractile elements (e.g., actin filaments, myosin-light chain kinase). Alterations in both systems, e.g., by thrombin, can cause cellular shape changes and paracellular gaps that can be studied in vitro (2, 3).
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_22, © Springer Science+Business Media, LLC 2011
319
320
A. Ludwig et al.
Changes in cell shape can be followed by impedance of cells that have contact with microelectrodes. This measurement is based on the fact that due to the insulation properties of their membranes cells behave like dielectric particles so that covering of electrodes increases impedance (4). This method is ideally suited to study time courses in a very high resolution, but does not reveal the cause for the change in impedance which may be due to changes in membrane permeability, but may also be caused by other factors, such as altered ion currents, cell adherence, or cell–cell contacts (see Note 1). Therefore, further methods, such as the assessment of the transfer of large-size molecules, are needed to demonstrate changes in vascular permeability. Ideally, these large size molecules are foreign to the cells under study because otherwise cell activation or transcytosis (as for albumin) cannot be excluded. In our experiments, we have made good experiences with fluorescently labeled dextran, which is not actively transported (5). Finally, endothelial cell monolayers can be used to study the transmigration of leukocytes. Leukocyte diapedesis is an active process that requires a concerted interaction between leukocytes and endothelial cells. This activation is carried out by adhesion molecules as well as by chemokines (6, 7). Leukocyte transendothelial migration can be measured using transwell inserts in which endothelial cells are cultured on a porous membrane (8). Subsequently, isolated leukocytes are seeded onto the endothelial cell layer and probed for migration through the cell layer and the underlying membrane.
2. Materials 2.1. Impedance Measurements
1. Clonetics® Human Lung Microvascular Endothelial Cells (HMVEC-L) (Lonza, Walkersville; USA). 2. Microvascular Endothelial Growth Medium-2 (EGM®-2MV) consisting of Endothelial Basal Medium-2 (EBM®-2) supplemented with growth factors, cytokines, and other supplements using the SingleQuots®-Kit as described by the manufacturer (Lonza, Walkersville; USA). 3. EBM®-2 without supplements (Lonza, Walkersville; USA). 4. xCELLigence RTCA SP System consisting of RTCA Analyzer, RTCA SP Station, RTCA Control Unit (Roche, Basel, Switzerland). 5. E-Plates 96 (Roche, Basel, Switzerland and ACEA Bioscience Inc., San Diego, USA). 6. The E-plates (Roche, Basel, Switzerland) can be put in an incubator, where the measurements take place.
22 Assessment of Endothelial Permeability and Leukocyte Transmigration in Human...
321
7. Labotect Hot Plate (Labotect, Goettingen; Germany). 8. Human a-Thrombin (MP Biomedicals, Solon; USA). 2.2. Permeability to Dextran
1. 12-well plates with transwell® inserts consisting of a polycarbonate membrane with a growth area of 1.12 cm2 and 0.4mm pore size (Costar Corning, Schiphol-Rijk; Netherlands). 2. Tetramethylrhodamine isothiocyanate-dextran, dextran (Sigma, Saint Louis; USA).
TRITC-
3. Minimal essential medium eagle (MEM) with Earle’s Salts and l-glutamine (PAA, Pasching; Austria). 4. Fetal bovine serum gold (PAA, Pasching; Austria). 5. Human a-thrombin (MP Biomedicals, Solon; USA). 6. Fluorescence plate reader GENios (Tecan, Salzburg; Austria). 2.3. Leukocyte Transmigration
1. 24-well polycarbonate membrane transwell® inserts (Costar Corning, Schiphol-Rijk, Netherlands) with a filter area of 0.33 mm2 and 5–8 mm pore size of (for the choice of pore size, see below). 2. PBS Dulbecco for cell culture, calcium and magnesium free (PAA Laboratories, Linz, Austria). 3. Full endothelial cell growth medium with supplement mix (Promocell, Heidelberg, Germany). 4. Collagen G solution (Biochrom, Berlin, Germany, 4 mg/ mL) or gelatine (Sigma, Munich, Germany). 5. RPMI (PAA Laboratories, Linz, Austria) with 0.1% w/v endotoxin-free BSA (Sigma, Munich, Germany). 6. Recombinant human IL-8 (R&D Systems, Wiesbaden, Germany). 7. Triton X-100 10% v/v in PBS. 8. Acetate buffer: 5 mM Na-acetate solved in H2O, adjust to pH 4 with HCl. 9. Substrate solution: 10 mM p-nitrophenyl-b-d-glucoronide (Calbiochem, Hamburg, Germany) solved in acetate buffer immediately before assay. 10. Glycine buffer: 0.4 M glycine, adjust to pH 10.3 with HCl.
3. Methods The phenotype of endothelial cells from different vascular beds differs substantially, and it is important to select endothelial cells appropriate for the question at hand. Human endothelial cells can
322
A. Ludwig et al.
be prepared from human umbilical vein using standard procedures (9). Preparation of human endothelial cells from other tissues is difficult because of the limited access to such tissues. Alternatively, cells can be obtained from commercial suppliers that offer arterial, venous, or microvascular endothelial cells from different organs, such as the HMVEC-L cells (passage 3–8) used in the measurement of impedance (Subheading 3.1) and permeability (Subheading 3.2). The transmigration procedure (Subheading 3.3) has been established for HUVECs (passage 3–5) and has also been successfully used for the HMVEC-L cells. For all methods, it is important that cells have grown to confluence before any measurements are made (see Notes 2–5). The xCELLigence Real-Time Cell Analyzing System measures the impedance using gold electrodes at the bottoms of the 96-well E-plates. The more cells are attached to the electrodes, the larger is the increase in electrode impedance. In addition, the impedance depends on the quality of cell interactions with the electrodes. For example, increased cell adhesions and tighter cell junctions lead to higher impedance values. To represent the cell status, a unitless parameter called Cell Index (CI) is derived to measure the relative change in impedance. The CI calculation is based on the following formula: CI = (Zi − Z0)/15, where Zi is the impedance at an individual time point during the experiment and Z0 is the impedance at the start of the experiment. The measurement of impedance is illustrated with thrombin that produces a very robust change of permeability in endothelial cell monolayers (Fig. 1). All solutions and media should be warmed up to 37°C, and all handlings should be done on a
3.1. Impedance Measurements
a 10
b
7. 8.
cell index [normalized to baseline]
6.
cell index
8 5.
6 4 2
2.
9.
0.0
−0.5
−1.0
0 0
20
40
60
Time [h]
80
100
0
5
10
15
20
Time [h]
Fig. 1. Impedance measurements in human lung microvascular endothelial cells. (a) Original tracing from a single experiment. The numbers refer to the numbering in Subheading 3.1.1. Briefly, cells were seeded at time point 0 (2.) and medium changed after 24 h (5.), 48 h (6.), and 72 h (7.) Thrombin (10 U/mL) was added 3 h later (8.). Solid line: thrombin; dashed line: control. (b) Normalized cell index in controls and thrombin-treated cells derived from original tracings. The cell index was normalized to the control at all time points starting with the time point immediately after the addition of thrombin (9.). Solid line: thrombin; dashed line: control. The gray area indicates the standard deviation from triplicate measurements of one experiment.
22 Assessment of Endothelial Permeability and Leukocyte Transmigration in Human...
323
eating plate at 37°C. All impedance measurements take place on h E-plates that are mounted in the RTCA station inside a standard incubator (5% CO2, 95% O2). 3.1.1. Measurements of Impedance in E-Plates
1. The RTCA System is prepared and the software is set up as described by the manufacturer. 2. In the E-plate, background impedance is measured with 200 mL medium/well without cells. All 96 wells have to be filled with medium, unused wells have to be filled with 200 mL PBS (Fig. 1a). 3. After background measurement in the RTCA station inside the incubator, the E-plate is put onto the heating plate, where the cells are seeded. 4,000 cells/well are seeded in 200 mL EGM2-MV medium. 4. The cells are allowed to settle down for 1 h, before the plate is transferred from the heating plate to the RTCA station with an impedance measurement every 15 min. 5. After 24 h, the plate is again transferred to the heating plate and half of the EGM-2MV medium is carefully changed without disturbing the attached cells. The E-plate is placed back in the RTCA station and the impedance measurements are continued. 6. After another 24 h, repeat step 5. 7. When the CI of all wells has reached a stable plateau of 7–9 (usually after 72 h), the complete medium is carefully replaced with 200 mL EBM-2 medium and the E-plate put back into the RTCA station for 3 h, where impedance is now measured every minute. In the analysis (Subheading 3.1.2), data are normalized to the last measurement during these 3 h. 8. After 3 h, when the artifacts in the cell index due to the medium changes have leveled off and the CIs have reached the same values they had before the E-plate was transferred to the heating plate, the agents under study are added in a small volume (we use 5 mL); we always prepare triplicates. This procedure should be done as fast as possible to reduce the time without measurement (see Note 6). 9. After transferring the plate in the RTCA station, impedance measurements are started again; now, the resolution may be as high as 1 measurement/15 s. These measurements are continued as long as needed.
3.1.2. Data Analysis and Interpretation
1. The initial steps of the analysis are done by the RTCA software itself. First, the mean and the standard deviation of the cell indices of the triplicate wells are calculated. These CI values are normalized to the last time point before addition of the agents (see step 7). This is necessary to compensate for
324
A. Ludwig et al.
slightly different growth curves and CIs in the different wells. 2. Usually, another normalization step (also offered by the proprietary software) is useful. The appropriate control is set as the baseline and all other values are then adapted accordingly. After this step, the control appears as a straight line (Fig. 1b). 3. Data interpretation: Values above the control line suggest an increase in endothelial barrier functions and values below a decrease. It should be noted again that changes in permeability are only one possible cause for changes in impedance. Therefore, the reasons for the impedance changes have to be validated with other methods. 4. The advantages of this method of impedance measurements are (1) it allows to identify the critical time points for further analysis; (2) it allows for broad and effective screening; (3) once the reasons for the change in impedance (e.g., increased endothelial permeability) have been identified, this method can also be used for mechanistic studies. 3.2. Permeability to Dextran
3.2.1. Measurements of Permeability in Transwell Plates
These measurements are based on the fact that confluent endothelial monolayers provide an efficient barrier to the transfer of large molecules. Special cell culture inserts are needed for these experiments to separate a top and a bottom chamber by a filter. Similar inserts are also used for the transmigration experiments (Subheading 3.3). In both cases, endothelial cells are grown on the filters. After the addition of dextran or leukocytes to the top chamber, permeability or transmigration is calculated as the amount of indicator (dextran, leukocytes) in the top chamber in relation to the amount in the bottom chamber. 1. The lower compartment of the wells is filled with 1.5 mL, the upper compartment with 0.3 mL EGM-2MV medium. 2. 200 mL of a cell suspension with 165 cells/mL is added to the upper chambers. Two wells of the plate are filled with 500 mL with EGM-2MV medium only; these wells without cells provide the maximum transfer of dye that is possible without active transport. 3. Because microscopic analysis of confluence is difficult on the transwell filter, it is advisable to seed cells in a conventional 12-well plate in the same density (cells/growth area) and analyze that one for confluency. 4. The cell culture plates are transferred to the incubator, and after 24 and 48 h half of the EGM-2MV medium is carefully changed without disturbing the attached cells.
22 Assessment of Endothelial Permeability and Leukocyte Transmigration in Human...
325
5. MEM medium is prepared and supplemented with 2% heat inactivated FBS. Heat inactivation occurs by heating up to 56°C for 30 min. 6. After cells are confluent (usually 72 h after seeding), the EGM-2MV medium is carefully replaced by the MEM with 2% heat inactivated FBS. The cells are starved for 1 h in the incubator. 7. After 1 h in MEM/2% FBS, the cells are treated with the agents under study, such as thrombin (10 U/mL). It is necessary that two wells remain untreated. These untreated wells represent the negative control (Fig. 2a). 8. 30 min after the addition of thrombin, 50 mL TRITC-dextran solution (20 mg/mL in MEM/2% FBS medium; vortex extensively), is added to the upper chamber of each well. TRITC-dextran may also be added at other time points, depending on the purpose of the study. 9. At various time intervals (e.g., 30, 60, 120, 180, 240, and 300 min after treatment), the inserts are carefully lifted, and the medium in the lower chamber is stirred and transferred in duplicates (each 50 mL) to a 96-well plate.
a 250
b 100
TRITC-Dextran [ng/ml]
Permeability Index [%]
10. Standard curves of TRITC-dextran are prepared: 250 mg/mL is serially (seven times) diluted in a ratio of 1:2. The last sample of the standard curve is pure MEM with 2% FBS. 50 mL of each standard curve sample is transferred in duplicates to the same 96-well plate as mentioned above.
200 150 100 50 0
80 60 40 20 0
0
1
2 3 Time [h]
4
0
1
2 3 Time [h]
4
Fig. 2. Permeability to dextran. Endothelial cells were seeded until confluence and treated at time 0 with thrombin (10 U/mL) as describe in Subheading 3.2. TRITC-Dextran was added 30 min after the addition of thrombin (black line). (a) Dextran measurements from wells without cells (defined as 100%) is shown as a gray solid line, and dextran measurements in wells with endothelial cell layers without thrombin are shown as a dashed gray line. (b) The same data as in panel (a) expressed as the permeability index. Data are the mean of duplicates in one typical experiment.
326
A. Ludwig et al.
11. The fluorescence of the samples is measured in a 96-well plate fluorescence reader. For TRITC-dextran, we use an excitation at 535 nm and emission at 595 nm wavelength. 3.2.2. Data Analysis and Interpretation
1. The concentrations of permeated TRITC-dextran are calculated from the standard curve. 2. Determination of Permeability Index (PI): permeated dextran concentrations of wells without cells (positive control; step 2) are defined as 100% permeability, the untreated wells (negative control, step 7) as 0% permeability. The results from the other wells are related to this reference frame. Since TRITC-dextran is not actively transported (5), the values in the treated cells are always between 0 and 100% (Fig. 2b). 3. If albumin is used instead of dextran, the permeability may exceed 100%. This is an indicator of active transport, probably transcytosis. It is, therefore, possible to combine TRITCdextran with FITC-albumin in one experiment, in order to assess the relative contribution of paracellular transport and transcytosis.
3.3. Transmigration 3.3.1. Endothelial Cell Culture on Transwell Inserts
1. 24-well polycarbonate membrane transwell inserts are coated with collagen [50 mL of collagen G solution (Biochrom, Berlin, Germany 40 mg/mL in PBS)], alternatively gelatine (Sigma, Munich, 40 mg/mL in PBS) can be used. 2. After 30 min incubation at 37°C, collagen solution is aspirated and the transwell inserts are washed with PBS. 3. Endothelial cells are harvested and suspended in full growth medium at a concentration of 1 × 105 cells/mL. 4. 100 mL of endothelial cell suspension (1 × 104 cells) is added to each transwell insert. 5. Cells are cultured in full growth medium for 48 h to reach full confluence (checked on an independent control plate as described in Subheading 3.2). 6. During this time, cells can be stimulated with proinflammatory stimuli to express adhesion molecules or chemokines (for induction of chemokines co-stimulation with IFNg and TNFa (R&D Systems, Wiesbaden, Germany), each 10 ng/ mL, for 24 h is recommended) (10). 7. Permeability of the endothelial cell layer can be tested using TRITC-dextran as described in this report.
3.3.2. Leukocyte Preparation and Culture
Different leukocyte populations can be used for transendothelial migration assays (Table 1). Neutrophils generally migrate within a short time period (1 h) while transmigration of monocytic cells takes longer (3 h). For neutrophil and T cell transmigration, a 5-mm pore size is recommended while monocytic cells migrate more efficiently through 8 mm pores. After transmigration, most
22 Assessment of Endothelial Permeability and Leukocyte Transmigration in Human...
327
Table 1 Transwell assays for leukocyte transendothelial migration Neutrophils
T cells
Monocytes
Pore size
3–5
5
8
Time period of migration
45 min–1 h Cells fall off
1–2 h Cells fall off
2–3 h Cells attach to filter
Quantification
Enzymatic assay, Flow cytometry
Enzymatic assay, Flow cytometry
Microscopic analysis of stained cells
Chemoattractant
IL-8/CXCL8
I-TAC/CXCL11, SDF-1/CXCL12
MCP-1/CCL2
monocytes tightly stick to the membrane and need to be stained on the membrane (this method is not described here). By contrast, the majority of transmigrated neutrophils and T cells falls off the membrane and can be counted in the lower compartment of the transwell insert. The transmigration assay described here has not only been established for neutrophils (8, 11), but can also be used for T cells, L1.2 cells, or THP-1 cells. 1. Neutrophils are prepared from citrated venous blood using standard protocols (12). 2. Cells are washed in sterile PBS Dulbecco. 3. Neutrophils are resuspended at a concentration of 2 × 106/ mL in RPMI containing 0.1% endotoxin-free BSA. It is important that the medium contains physiological concentrations of Ca2+ and Mg2+. 4. Cells are prewarmed for at least 15 min in polypropylene tubes. During this time, the cells should not sediment and a gentle agitation is recommended. 3.3.3. Transmigration Assay
Transmigration can either be induced by soluble chemokines added to the lower well or by induction of endogenous adhesion molecules and chemokines in endothelial cells. It is advisable to use attractants with high expression of the cognate chemoattractant receptor on the leukocyte subset analyzed. Here, we describe the classical chemotactic migration of neutrophils induced by soluble IL-8 (R&D systems, Wiesbaden, Germany) added to the lower well at a concentration range of 0.1–100 nM. For THP-1 cells, MCP-1 (0.1–100 nM) is recommended as a chemoattractant. 1. The confluent cell layers are carefully washed with PBS. 2. Chemotactic stimuli are diluted in RPMI supplemented with 0.1% BSA. To prepare triplicates more than 1.5 mL stimulus solution is needed.
328
A. Ludwig et al.
3. Neighboring wells on the 24-well plate that contain no inserts are filled with 500 mL stimulus solution. 4. Medium in the upper compartment of transwells is carefully aspirated. 5. Transwells are transferred into the wells containing stimulus solution and are immediately filled with 100 mL leukocyte suspension (2 × 105 cells) (see Note 7). 6. Leukocytes are allowed to transmigrate for 1–3 h at 37°C in an incubator. 7. Successful transmigration should be controlled by microscopy. 3.3.4. Measurement of Transmigrated Cells
Different methods for the quantification of transmigrated cells have been established. Cells can be counted using a hemacytometer. For this, transmigrated cells need to be carefully resuspended, and 10 mL is stained with trypan blue and counted (see Note 8). Here, we describe the quantification of transmigrated cells by measurement of endogenous b-glucoronidase activity (13, 14): 1. Transwell inserts are carefully removed. 2. Transmigrated cells within the lower compartment are lysed by the addition of Triton X-100 at a final concentration of 0.1% v/v. 3. A standard of defined cell number starting at 200,000 cells in 500 mL is serially diluted by 1 in 2 over eight steps and subsequently lysed by the addition of Triton X-100. From the standard curve cell numbers in the samples can be calculated. 4. 100 mL of cell lysate is transferred into a 96-well plate. It is recommended to perform duplicate determinations. 5. Endogenous b-glucoronidase activity is measured by the addition of 50 mL substrate solution per well. 6. After 14 h incubation at 37°C, the reaction is stopped by the addition of 50 mL glycine buffer. 7. Optical density is read at 405 nm in a 96-well photometer.
3.3.5. Data Analysis and Interpretation
Migration in the absence of a chemotactic stimulus is defined as random migration. Often the number of migrated cells is expressed as chemotactic index representing the fold increase over random migration. The dose–response curve typically has a bell shaped form (Fig. 3). Optimal migration is induced at concentrations between 10 and 30 nM. Within the optimum, the chemotactic index should reach a value of more than 5. At higher dosages, the response is again declining which is most likely due to complete occupation and desensitization of all receptors at higher dosages which would prevent sensing of the chemotactic gradient (14). Within the optimum, the chemotactic index should reach a value
22 Assessment of Endothelial Permeability and Leukocyte Transmigration in Human...
329
chemotactic index
8 6 4 2 0 0.1
1
10
100
1000
IL-8 [nM]
Fig. 3. Neutrophil transmigration through endothelial cell monolayer. HUVECs were grown on transwell inserts and subsequently assayed for transmigration of neutrophils in response to increasing dosages of IL-8. The data are expressed as chemotactic index representing the number of migrated cells in the presence of stimulus in relation to the number of migrated cells in the absence of stimulus. Data are shown as mean and SD from three experiments with neutrophils from different donors. IL-8 induced neutrophil migration is characterized by a typical bell-shaped dose–response curve.
of more than 5. For pharmacological studies, it may be important to choose a concentration around the EC50 value which normally is in the lower nanomolar range (0.5–3 nM). To define whether a stimulus induces directional migration, it is recommended to add the stimulus not only to the lower well, but also to the upper well. This would destroy the gradient and block directional chemotactic migration. To achieve best downregulation, it is recommended to pretreat neutrophils with the chemokine (for 10 min) prior to the transmigration assay.
4. Notes 1. An alternative to the impedance measurements is the determination of transendothelial electrical resistance (15, 16). The accuracy and the handling properties of this approach, however, appear to be inferior to measurements based on impedance. Ref. 4 lists several suppliers of commercially available hardware for impedance measurements. 2. It should be taken into consideration that endothelial cells in culture downregulate a number of genes that are expressed in situ (17). One critical factor that is missing in most cell culture system is shear stress, which upregulates several genes, including the PAF-receptor (18). This may explain the lack of response of some endothelial cell monolayers to PAF (19). In general, it appears that mechanical stress is healthy for endothelial cells and improves their barrier function (20).
330
A. Ludwig et al.
3. Endothelial cells in culture are also known to possess lower numbers of caveolae (21) and to produce more NO (22). Both factors affect vascular permeability (23). 4. Many cell culture media supplements contain growth factors, such as VEGF, that can also affect vascular permeability. Therefore, it is recommended to use basal media without supplements when vascular permeability is analyzed. 5. Cell preparations from commercial vendors may possess differing properties depending on the preparation. Therefore, it may be useful to make reservations for cell lots that are working. 6. Some mediators produce very rapid changes in vascular permeability that may be overlooked if the pipetting process takes too long. In the E-plate, substances with short-term effects can be analyzed by the use of special plate lid that allows to add substances while the plates stay in the RTCA station. Because of CO2 deprivation and cooling of the incubator, this procedure can only be done for a few wells. 7. Chemokines can also be loaded on the cell layer by preincubation of endothelial cells with chemokines. The chemokines are bound to surface proteoglycans on endothelial cells and then presented to the leukocytes and induce leukocyte migration. This process has been termed haptotaxis. 8. Alternatively, transmigrated cells can be counted using a flow cytometer. Transmigrated cells are resuspended and transferred to an FACS tube containing a defined amount of fluorescent beads. Subsequently, cells can be quantified in relation to the beads by flow cytometry (24). This method also allows to control whether only neutrophils have migrated through the pores (migration of endothelial cells could be possible when using 8 mm pores).
General Note All compounds should be warmed up to 37°C before usage.
Acknowledgments This work was supported by the IZKF Aachen and by the DFG, SFB 542, Projects A12 and C16.
22 Assessment of Endothelial Permeability and Leukocyte Transmigration in Human...
331
References 1. Medzhitov R (2008) Origin and physiological roles of inflammation. Nature 454:428–435 2. Dudek SM, Garcia JG (2001) Cytoskeletal regulation of pulmonary vascular permeability. J Appl Physiol 91:1487–1500 3. Vandenbroucke E, Mehta D, Minshall R, Malik AB (2008) Regulation of endothelial junctional permeability. Ann N Y Acad Sci 1123:134–145 4. Hu Z, Liu Q, Wang P. Electric cell substrate impedance sensor (ECIS) as cell-based biosensors. In: Liu Q, Wang P, editors. Cell-based biosensors: Principles and Applications. Boston, London: Artech House Inc., 2009: 151–178. 5. Shasby DM, Shasby SS (1985) Active transendothelial transport of albumin. Interstitium to lumen. Circ Res 57:903–908 6. Springer TA (1994) Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm. Cell 76:301–314 7. Ludwig A, Weber C (2007) Transmembrane chemokines: versatile ‘special agents’ in vascular inflammation. Thromb Haemost 97:694–703 8. Kuijpers TW, Hakkert BC, Hart MH, Roos D (1992) Neutrophil migration across monolayers of cytokine-prestimulated endothelial cells: a role for platelet-activating factor and IL-8. J Cell Biol 117:565–572 9. Jaffe EA, Nachman RL, Becker CG, Minick CR (1973) Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest 52:2745–2756 10. Abel S, Hundhausen C, Mentlein R, Schulte A, Berkhout TA, Broadway N, Hartmann D, Sedlacek R, Dietrich S, Muetze B, Schuster B, Kallen KJ, Saftig P, Rose-John S, Ludwig A (2004) The transmembrane CXC-chemokine ligand 16 is induced by IFN-gamma and TNF-alpha and shed by the activity of the disintegrin-like metalloproteinase ADAM10. J Immunol 172:6362–6372 11. Koenen RR, Pruessmeyer J, Soehnlein O, Fraemohs L, Zernecke A, Schwarz N, Reiss K, Sarabi A, Lindbom L, Hackeng TM, Weber C, Ludwig A (2009) Regulated release and functional modulation of junctional adhesion molecule A by disintegrin metalloproteinases. Blood 113:4799–4809 12. Boyum A (1968) Isolation of mononuclear cells and granulocytes from human blood.
Isolation of monuclear cells by one centrifugation, and of granulocytes by combining centrifugation and sedimentation at 1 g. Scand J Clin Lab Invest Suppl 97:77–89.:77–89 13. Creamer HR, Gabler WL, Bullock WW (1983) Endogenous component chemotactic assay (ECCA). Inflammation 7:321–329 14. Ludwig A, Petersen F, Zahn S, Gotze O, Schroder JM, Flad HD, Brandt E (1997) The CXC-chemokine neutrophil-activating peptide-2 induces two distinct optima of neutrophil chemotaxis by differential interaction with interleukin-8 receptors CXCR-1 and CXCR-2. Blood 90:4588–4597 15. Shasby DM, Shasby SS (1986) Effects of calcium on transendothelial albumin transfer and electrical resistance. J Appl Physiol 60:71–79 16. Lindner K, Uhlig U, Uhlig S (2005) Ceramide alters endothelial cell permeability by a nonapoptotic mechanism. Br J Pharmacol 145:132–140 17. Lacorre DA, Baekkevold ES, Garrido I, Brandtzaeg P, Haraldsen G, Amalric F, Girard JP (2004) Plasticity of endothelial cells: rapid dedifferentiation of freshly isolated high endothelial venule endothelial cells outside the lymphoid tissue microenvironment. Blood 103:4164–4172 18. Heydarkhan-Hagvall S, Chien S, Nelander S, Li YC, Yuan S, Lao J, Haga JH, Lian I, Nguyen P, Risberg B, Li YS (2006) DNA microarray study on gene expression profiles in co-cultured endothelial and smooth muscle cells in response to 4- and 24-h shear stress. Mol Cell Biochem 281:1–15 19. Uhlig S, Lindner K. Novel mechanisms of endothelial cell permeability in the lung. Lessons from studies with intact lungs and animals. In: Tooke J, Shore A, Whatmore J, editors. The microcirculation and vascular biology. Bologna: Monduzi Editore, 2002: 221–234. 20. Fujiwara K (2003) Mechanical stresses keep endothelial cells healthy: beneficial effects of a physiological level of cyclic stretch on endothelial barrier function. Am J Physiol Lung Cell Mol Physiol 285:L782–L784 21. Gratton JP, Bernatchez P, Sessa WC (2004) Caveolae and caveolins in the cardiovascular system. Circ Res 94:1408–1417 22. Hall CN, Garthwaite J (2009) What is the real physiological NO concentration in vivo? Nitric Oxide 21:92–103
332
A. Ludwig et al.
23. Kuebler WM, Yang Y, Samapati R, Uhlig S (2010) Vascular barrier regulation by PAF, ceramide, caveolae, and NO – an intricate signaling network with discrepant effects in the pulmonary and systemic vasculature. Cell Physiol Biochem 26:29–40
24. Molema G, Mesander G, Kroesen BJ, Helfrich W, Meijer DK, de Leij LF (1998) Analysis of in vitro lymphocyte adhesion and transendothelial migration by fluorescent-beads-based flow cytometric cell counting. Cytometry 32:37–43
Chapter 23 Permeability of Endothelial Barrier: Cell Culture and In Vivo Models Alexander N. Garcia, Stephen M. Vogel, Yulia A. Komarova, and Asrar B. Malik Abstract The methods for assessment of endothelial barrier permeability are vital tools of experimental biology. They allow us to measure permeability of endothelial monolayer in cell culture and in lung vessels or to determine formation of tissue edema resulting from increased permeability of vasculature. This chapter provides an overview of the most common protocols. Key words: Permeability, Endothelial barrier, Albumin flux, Transendothelial electrical resistance, Hydraulic conductance, Starling equation, Capillary filtration coefficient, Lung edema
1. Introduction The endothelium lining blood vessels forms a semipermeable barrier that separates the circulating blood from the underlying tissue. The basal permeability of the endothelium to macro molecules, fluids, and solutes is a finely regulated process that aims to supply surrounding tissue with nutrients and solutes, as well as to provide a set point for tissue homeostasis (1). Some pathological conditions such as inflammation, diabetes, and atherogenesis are associated with hyperpermeability of vessels, which leads to impaired tissue homeostasis, accumulation of fluid in the interstitium, and development of tissue edema (2). The understanding of how vessel permeability is regulated basally and in the context of disease is critical for development of therapies to treat the vascular leak.
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_23, © Springer Science+Business Media, LLC 2011
333
334
A.N. Garcia et al.
To date, numerous methods have been established to assess the permeability of the endothelial monolayer in cell culture and vessels in vivo and ex vivo using animal models (3). While some techniques provide direct measurements of the permeability of the endothelial layer or vessel wall to fluids and proteins (Table 1), others focus on accumulation of dye-labeled proteins in intersti tium and tissue edema (Table 2). The various protocols high lighted in this article showcase the most common techniques used in research laboratories to assess permeability of the endothelial barrier within a variety of different models.
Table 1 Overview of techniques for determining permeability of the endothelial barrier in cell monolayers and in lungs Physiological assessment
Unit of measurement
Description
Reference
Cell monolayer Transendothelial electrical resistance (TER) Transwell tracer 125 I-albumin permeability
Resistance (ohms) or normalized resistance (ohm-cm2) Palb (cm/s)
(4–6)
Hydraulic conductance
Lp (cm/min/cmH2O)
Resistive component of the impedance of a cell monolayer is measured Albumin flux is measured for a specified albumin concentration gradient across the upper and lower chambers of a transwell unit Fluid flux (Jv) across an endothelial monolayer of surface area S is measured at various hydrostatic pressures ( D P) (Lp = Jv/S D P)
Lungs Capillary filtration coefficient, Kfc
mL/min/cmH2O/g dry lung weight
Lung filtration rate is measured for a specified rise in pulmonary microvascular pressure. Kfc is a measure of lung liquid permeabil ity times area of filtration surface Rate of tracer-albumin uptake is measured for a specified traceralbumin concentration. Interstitial tracer concentration is assumed negligible. PS is a measure of tracer-albumin permeability times microvascular surface area Measures lung protein permeability independently of vascular surface area
(3, 25)
Albumin permea bility–surface area product, PS
PS in mL/min/g dry lung weight
Osmotic reflection coefficient for albumin, salb
Dimensionless
(9)
(12)
(26, 27)
(27)
Lp hydraulic conductivity; P permeability; Kf,c microvascular filtration coefficient; PS permeability–surface area product
23 Permeability of Endothelial Barrier…
335
Table 2 Overview of techniques for assessment of lung injury and edema Technique
Parameters (units)
Description
Reference
Wet-to-dry lung weight ratio
No units
The ratio of lung wet and dry weights is taken as a measure of lung edema
(14–17)
Evans Blue dye labeled Concentration Albumin extravasation from the lung albumin uptake (mg/mL) of Evans vasculature into the tissue is measured Blue in lung
(20, 21)
2. Materials 2.1. Transendothelial Electrical Resistance
1. Electric Cell-Substrate Impedance Sensing system (Applied Biophysics Inc., http://www.biophysics.com). 2. Specialized slides with eight individual wells (see Note 1). 3. CO2 incubator. 4. Computer and printer. 5. 2% Gelatin type B from bovine skin (Sigma, St. Louis, MO).
2.2. Permeability of the Endothelial Monolayer to 125 I-Albumin
1. Sephadex G-25 column (Pharmacia Inc., Clayton, NC). 2. 12-mm transwell support insert equipped with polyester membrane filter having a pore diameter of 0.4 mm (Corning, Lowell, MA) (see Note 2). 3. CO2 incubator. 4. Gamma counter (GMI, Minneapolis, MN). 5. Bovine serum albumin (Sigma, St. Louis, MO). 6. Phosphate-buffered saline (Invitrogen, Carlsbad, CA). 7. Na125I (GE Healthcare Biosciences, Pittsburgh, PA). 8. Iodination beads (Pierce Chemical, Rockford, IL). 9. Hanks balanced salt solution with CaCl2 and MgCl2 (Invitrogen, Carlsbad, CA). 10. 2% Gelatin type B from bovine skin (Sigma, St. Louis, MO).
2.3. Measurement of Hydraulic Conductance (Lp )
1. Perfusion pump (Gilson, Middleton, WI). 2. 12-mm transwell filters (0.4-mm pore size; Fisher Scientific, Pittsburgh, PA). 3. Ussing chamber (Harvard Apparatus, Hollinston, MA). 4. Volumetric pipette. 5. Eagle’s minimum essential medium (Sigma, St. Louis, MO). 6. HEPES (Sigma, St. Louis, MO).
336
A.N. Garcia et al.
2.4. Wet–Dry Weight Ratio
1. Balance (0.1-mg resolution). 2. Oven for tissue drying (65°C). 3. DU-64 Spectrophotometer (Beckman Instruments, Inc. Fullerton, California). 4. Ketamine. 5. Sodium lauryl sulfate (Sigma, St. Louis, MO). 6. Hemoglobin reagent set (Pointe Scientific, Canton, MI).
2.5. Permeability to Evans Blue-Labeled Albumin
1. Spectrophotometer. 2. Centrifuge. 3. 0.22-mm filter (Millipore, Billerica, MA). 4. 5-0 silk continuous suture (DemeTech Corporation, Miami, FL). 5. Evans Blue dye (Sigma, St. Louis, MO). 6. Bovine serum albumin (Sigma, St. Louis, MO). 7. Phosphate-buffered saline (Invitrogen, Carlsbad, CA). 8. Ketamine. 9. Formamide. 10. Solution 1 [preparation of Evans Blue labeled albumin (EBA)]. (a) Dissolve Evans Blue to a concentration of 5 mg/mL in phosphate-buffered saline (Calcium and Magnesium free). (b) Add BSA to a concentration of 4 g/100 mL. (c) Mix well by stirring with a magnetic bar. (d) Let solution stand at room temperature for 30 min. (e) Filter the solution through a sterile 0.22-mM filter. The aliquots of EBA solution can be stored at −80°C.
2.6. Measurement of Capillary Filtration Coefficient (Kf,c )
1. Pressure transducer. 2. Weight transducer. 3. Pressure-controlled ventilator (Kent Scientific, Torrington, CT). 4. Perfusion pump (Gilson, Middleton, WI). 5. Arterial bubble trap. 6. Water bath (40°C). 7. Catheters (Fisher Scientific, Pittsburgh, PA): arterial, PE60; venous, PE90. 8. RPMI 1640 medium (HEPES modification; Sigma) supple mented with 3% BSA (pH to 7.4), filtered. 9. Heparin. 10. Ketamine.
23 Permeability of Endothelial Barrier…
2.7. Measurement of Permeability × Surface Area Product ( PS )
337
1. Pressure-controlled ventilator (Kent Scientific Corporation, Torrington, CT). 2. Perfusion pump. 3. Arterial bubble trap. 4. Water bath. 5. Catheters (Fisher Scientific, Pittsburgh, PA): arterial, PE60; venous, PE90. 6. Gamma counter (GMI, Minneapolis, MN). 7. Sephadex G-25 column (Pharmacia Inc., Clayton, NC). 8. RPMI 1640 medium (HEPES modification; Sigma). 9. Bovine serum albumin (Sigma, St. Louis, MO). 10. Heparin. 11. Ketamine. 12. Na125I (GE Healthcare Biosciences, Pittsburgh, PA). 13. Iodination beads (Pierce Chemical, Rockford, IL). 14. Phosphate-buffered saline.
2.8. Measurement of Osmotic Reflection Coefficient ( s )
1. Pressure-controlled ventilator (Kent Scientific Corporation, Torrington, CT). 2. Perfusion pump. 3. Arterial bubble trap. 4. Water bath. 5. Catheters (Fisher): arterial, PE60; venous, PE90. 6. Gamma counter (GMI, Minneapolis, MN). 7. Sephadex G-25 column (Pharmacia Inc., Clayton, NC). 8. RPMI 1640 medium (HEPES modification; Sigma). 9. Bovine serum albumin. 10. Heparin. 11. Ketamine. 12. 125I- and 131I-labeled albumin. 13. Phosphate-buffered saline (Invitrogen, Carlsbad, CA).
3. Methods 3.1. Transendothelial Electrical Resistance
Electric Cell-Substrate Impedance Sensing system (4–6) or ECIS™ is a noninvasive biophysical approach to assess cell shape changes in real time. This method monitors time-resolved imped ance of a noninvasive current supplied by gold electrodes located on the bottom of the cell culture plate, through the cell culture
338
A.N. Garcia et al.
b
a
c
d
Current
Inhibition of current by cell monolayer
Insulating Film
Gold Electrode
Fig. 1. Assessment of endothelial barrier using Electrical Cell-Substrate Impedance Sensing system. (a) A top view of 8W1E cultureware consisting of eight wells. Each well contains a single circular active electrode (250 mm in diameter) and counter electrode as shown in enlarged inserts. (b) Schematic diagram of AC flow (arrows) between the active and counter electrodes. A top view. (c–d) Schematic diagram of AC flow from circular electrode though cell culture medium in the absence (c) and presence of cells (d). A side view. Cell monolayer inhibits AC flow, thus increasing resistance (d).
medium (Fig. 1). The system software extracts the resistive component of the total impedance from the raw data. The cells grown on the surface of gold electrodes impede AC flow between the small active electrode and the large counterelec trode according to the monolayer’s resistance. Disruption of the integrity of the endothelial monolayer as mediated by proinflam matory mediators (e.g., thrombin) results in cell contraction and shape change, thus leading to a decrease in monolayer resistance (4). In addition, ECIS™ technique can be used to monitor cell attachment, spreading, migration, and wound healing. Investigators usually find it convenient to simultaneously analyze the resistance of several control and treated monolayers. To do this, it is necessary to use a specialized slide that has eight individual wells. The measurement system can accommodate up to two 8-electrode arrays for a maximum of 16 individual mea surements in any given run.
23 Permeability of Endothelial Barrier…
339
Adopted from the original protocol by Tirrupathi et al. (4). 1. Coat the surface of each well of the slide with gelatin. Add ~200 mL 0.2% gelatin diluted in 1× phosphate-buffered saline (PBS) to each well and incubate for 15–20 min at 37°C. 2. Remove gelatin by aspiration. Add ~80,000–100,000 pri mary endothelial cells resuspended in 400 mL of cell culture medium (see Note 3). 3. Grow cells for 3–5 days until a confluent monolayer is estab lished. Control cell confluence by phase-contrast microscopy (see Note 4). 4. Change media 12 h before experiment. Data acquisition 5. Using the supplied connectors, connect the arrays of gold electrodes to the ECIS electronics and check electrical con nectivity with ECIS software by running “check electrodes” mode. The program will automatically detect if any electrode is not appropriately connected (see Note 5). 6. Set up frequency of AC at 50 Hz, period of data collection, and time interval between data points. Data points are auto matically recorded, displayed as a graph of absolute or nor malized impendance (resistance) vs. time and saved. 7. Record the basal impendance (resistance) for at least 30 min. 8. Add mediator (agent) while continuing data acquisition. Data analysis 9. Upon completion of experiment, export data to csv format and import data to Excel, SigmaPlot, etc. Figure 2 illustrates a graph of normalized resistance against time.
Fig. 2. Time-course of transendothelial electrical resistance (TER) change. (a) Phase contrast image of murine lung endothelial cells (ML-EC), which were used for measurements of TER presented in b. (b) Untreated ML-ECs (green line) form a stable monolayer over period of experiment. TER changes after thrombin treatment (blue line). Thrombin (time of stimulation is indicated by arrow) induces cell shape changes and disruption of endothelial barrier resulting in decreased resistance. Resistance was normalized to the basal value at 0 time.
340
A.N. Garcia et al.
3.2. Permeability of the Endothelial Monolayer to 125 I-Albumin
The measurement of monolayer permeability to radiolabeled albumin is a direct measure of protein permeability. Confluent monolayers of endothelium have low albumin permeability, whereas disruption of monolayer integrity leads to formation of intercellular gaps, allowing increased flux of albumin across the barrier (Fig. 3). Serum albumin, radiolabeled on tyrosine residues with 125I, has often been used to assess endothelial permeability (7). The method for assessing permeability is based on measuring albumin flux across the monolayer of endothelial cells grown on the membrane inserts of a transwell system. This technique uti lizes the Kedem–Katchalsky equation for determining the overall flux of a solute representing the sum of convective and diffusive components:
J s = J v (1 - s )C s + PS (DC ),
where Js is the solute flux, Jv is the volume flux of fluid; s is the osmotic reflection coefficient; Cs is the mean concentration of the solute within the pore; P is permeability; S is surface area; and ∆C is the difference in solute concentration across the monolayer. Because no hydrostatic or osmotic pressure gradient is imposed on the monolayer, fluid flux for the system is equal to zero and Js reflects the remaining diffusive component of solute flux. This allows permeability to be calculated as follows:
P = J s / S DC . Transwell Insert
Upper well
125-I albumin Compromised cell monolayer
Confluent monolayer 0.4 µm pore size membrane
Lower well
Non-labeled Albumin
Fig. 3. Assessment of permeability of endothelial monolayer to albumin. The endothelial monolayer grown on the transwell membrane prevents the passage of albumin across, resulting in low concentration of 125I-albumin in the lower chamber (left ). Disruption of endothelial barrier integrity following treatment with proinflammatory mediator such as thrombin results in 125I-albumin flux (right). Formation of gaps between cells allows the passage of 125I-albumin across the monolayer and results in increased concentration of 125I-albumin in the lower chamber.
23 Permeability of Endothelial Barrier…
341
Albumin Iodination (adopted from the original protocol by Du et al. (8)). 1. Wash iodination beads with 500 mL of PBS twice. 2. Dry beads on filter paper. 3. Add beads to 250 mL PBS containing 0.5 mCi Na125I. 4. Allow reaction to occur for 5 min. 5. Add 50 mg purified albumin dissolved in 250 mL PBS to the reaction mixture. 6. Allow reaction to occur for 5 min. 7. Stop the reaction by removing the solution from the iodina tion beads in the reaction vessel. 8. Separate radiolabeled proteins from free 125I using a 30 cm × 3.5 mm column packed with Sephadex G-25, which is saturated with unlabeled BSA and equilibrated with saline solution. Steps 9 through 16 (below) assess percentage of contaminant-free iodine (which should be <0.3% of the total radioactivity). 9. Count total radioactivity of a 250-mL sample in a Gamma counter. 10. Add TCA to an identical 250-mL sample to a final concentra tion of 10% to precipitate protein. 11. Incubate on ice for 30 min. 12. Centrifuge for 10 min at 15,000 × g at 4°C. 13. Remove supernatant and add an equal volume of 80% acetone to pellet to wash away residual TCA. 14. Vortex and centrifuge again as above. Repeat the process four times. 15. Aspirate the acetone and dry the pellet. 16. Measure counts of radioactivity and protein concentration. Compare with initial measurement to obtain purity of iodi nated protein solution relative to free 125I (see Note 6). Measurement of Transmembrane Labeled Albumin Flux (adopted from original protocol by John et al. (9)). 1. Add 1.5 mL of medium to lower chamber of 12-well plate. 2. Add 0.5 mL medium to the transwell insert. 3. Incubate at 37°C for 1 h in CO2 cell culture incubator. 4. Remove medium from transwell insert and add 0.2% gelatin in 0.5 mL PBS. 5. Incubate at 37°C for 30 min in CO2 cell culture incubator. 6. Remove gelatin and add 0.5 mL of cell suspension containing 1.0 × 105 cells/mL onto the transwell insert membrane.
342
A.N. Garcia et al.
7. Grow cells to a confluent monolayer. 8. Aspirate medium and wash cells with HBSS containing Ca2+ and Mg2+ twice to remove nonadherent cells. 9. Add warm HBSS buffer containing 68 nM 125I-albumin and 1.5 mM unlabeled albumin in the upper chamber. 10. Remove media from the lower chamber and add 1.5 mL HBSS buffer containing 1.5 mM unlabeled albumin. 11. Over the course of 90 min, take 50 mL samples from the lower chamber every 15 min and analyze for gamma radioactivity (see Note 7). 12. Plot the amount of 125I-albumin in the lower chamber against time of sample collection. 13. Fit the data to a linear regression 14. Calculate the slope of the best fitting linear regression to obtain the rate (flux) of 125I-albumin flux across the mono layer, Js, (mg/s). 15. Calculate the permeability P (cm/s) using the equation (10)
P = J s / S DC , where Js is 125I-albumin flux across the monolayer (mg/s); S is surface area (cm2); ∆C is initial concentration of 125I-albumin in the top well in (mg/cm3).
3.3. Measurement of Hydraulic Conductance (Lp )
Hydraulic conductance (Lp) is a key value in the Starling equation that describes net fluid flux per unit area across an endothelial monolayer. This is proven to be an important measurement as the filtration properties of the vascular endothelium determine the fluid homeostasis in the lung. The measurement of hydraulic con ductance has been originally developed for single-vessel prepara tions using the split-drop technique (11); however, it also been demonstrated to provide an accurate determination of this value in cell monolayer using a more straightforward procedure (12). Adapted from original protocol by Parker et al. (12). 1. Plate 80,000 cells/cm2 on 12-mm transwell filters of 0.4-mm pore size. 2. Grow cells to confluence. 3. Mount filters on a six-well plate and perfuse the system with Eagle’s Minimum Essential Medium containing 25 mM HEPES. 4. Circulate the perfusate between the upper chamber and a pressure reservoir supplying the upper chamber with the aid of a Gilson roller pump. 5. Control the perfusion pressure (P) by adjusting the height of the reservoir to 20, 30, 40, and 50 cmH2O.
23 Permeability of Endothelial Barrier…
343
6. Measure the filtration rate (Jv) by collecting filtered fluid from lower well and dividing by the time interval (s). 7. Normalize filtration rate by the area of the transwell filter (Jv/A) and plot resulting fluid flux against the perfusion pres sure (P). Repeat this step for a series of several different P values. 8. Fit a regression line to the data and calculate Lp (cm/s/ cmH2O) from the slope (13). 9. Similarly, Lp can be calculated form 3.4. Wet–Dry Weight Ratio
Lp = J v / A / P
Gravimetric methods for the assessment of pulmonary edema were first described by Hemingway in 1950 (14) and later modi fied by Pearce (15) and Fukuda (16). These methods determine the ratio between lung wet and dry weight (w/d) to estimate the water content in the lung outside of the pulmonary vasculature (development of edema). Since weight can be measured more readily than most other properties, gravimetric analysis has proven to be accurate and reproducible. This technique has the advan tage of being fairly simple to perform, as it does not require sig nificant surgical skill or sophisticated and expensive equipment. Despite its simplicity, this method suffers from a number of disadvantages. While the goal of these measurements is to deter mine the development of pulmonary edema, wet/dry ratios are complicated by inclusion of intravascular blood in the wet lung weight, which can account for significant error. This error can be corrected, however, through measurement of hemoglobin con centrations or radiolabeled tracers in the blood (17). The gravi metric assessment of pulmonary edema also has the disadvantage of being unable to address any regional heterogeneity in vessel permeability nor can it determine the contribution of hydrostatic pressure. The latter problem can be overcome by concomitant measurement of pulmonary vascular pressure, although it is not trivial technically for small animals such as mice. Adapted from the original protocol by Kobayashi et al. (17). 1. Anesthetize mice according to an approved protocol for animal use by the University Animal Care Committee (see Note 8). 2. Make a vertical midline incision along the sternum to provide access to the heart and lungs. 3. Make an incision in the left inferior vena cava and abdominal aorta to exsanguinate the mouse. 4. Remove the heart and lungs en bloc and quickly separate lung lobes from the heart and trachea. Mice have five lung lobes; ensure separation of all lobes.
344
A.N. Garcia et al.
5. Quickly weigh lung tissue to obtain wet weight. 6. Place lungs in an oven at 65°C for 2–3 days until a steady weight is obtained (see Note 9). 7. Present data as an average ratio (18) with a standard deviation
Wet:dry ratio = Weight wet / Weight dry Correction of lung wet weight for blood in the lung (mea surement of the hemoglobin concentration (15, 17)). 1. Homogenize 0.2 g of lung fragments in 0.4 mL of distilled water. 2. Centrifuge at 15,000 rpm for 30 min. 3. Take 0.4 mL of supernatant and add 0.05 mL of sodium lauryl sulfate (Sigma, St. Louis, MO). 4. Add 2 mL of total hemoglobin A1c reagent set (Pointe Scientific, Canton, MI) into tubes which will be used for preparation of standards and for sample analysis. 5. Place 10 mL of standards, lung samples (supernatants from homogenates), or blood from the abdominal aorta into respec tive tubes and incubate at room temperature for 3 min. 6. Measure absorption at 540 nm using a spectrophotometer. Use standards to obtain a standard curve; calculate concen tration of hemoglobin in lung tissue samples and blood. 7. Calculate the proportion of intrapulmonary blood (Qb) and subtract the weight of blood from the initial wet weight of the lung (19):
Qb = (Hb hs / Hb b ) ´ (Vbt ´ (1 - Htc)), where Hbhs is hemoglobin concentration in supernatant; Vbt is total blood volume; Htc is hematocrit; Hbb is hemoglo bin concentration in blood; Vbt × (1 − Htc) is the plasma volume
3.5. Permeability to Evans Blue-Labeled Albumin
Evans Blue dye exhibits high affinity to albumin (20) and Evans Blue-labeled Albumin (EBA) has been effectively used as an alter native to radioisotope labeling for assessment of albumin extrava sation in airways. Both Evans Blue and 125I-labeled albumin gave comparable results when albumin flux was measured in lavage fluid from isolated lungs and across endothelial monolayers (20). Adapted from original protocol by Moitra et al. (21). 1. Anesthetize mice according to an approved protocol for animal use by the University Animal Care Committee (see Note 10). 2. Expose the trachea and right internal jugular vein through neck incision (see Note 11).
23 Permeability of Endothelial Barrier…
345
3. Inject EBA through jugular vein at 30 mg/kg body weight. 4. Close the neck incision with a 5-0 silk continuous suture. 5. After 2 h, make a vertical inline incision along the sternum to provide access to the heart and lungs. 6. Make an incision in the left atrium and inject 6–10 mL of phosphate buffered saline through the right ventricle to remove intravascular EBA. 7. Remove the heart and lungs en bloc (22). 8. Separate lung lobes from the heart and trachea; weigh lung tissue. 9. Homogenize the lungs in formamide (1 mL forma mide/100 mg lung) (23). 10. Incubate for 24 h at 60°C. 11. Centrifuge samples at 12,000 × g for 30 min. 12. Remove supernatant. 13. Measure absorbance (OD) at 620 and 740 nm against a blank containing 50% formamide in PBS (see Note 12). Corrected absorbance (620 ) = Absorbance (620) - (1.426
´ Absorbance (740) + 0.03) 14. Generate an Evans Blue standard absorbance curve to obtain the concentration of Evans Blue in the tissue. The Evans Blue concentration read in mg/mL can then be converted to mg per g of wet weight of lungs by multiplying the Evans Blue con centration by the dilution factor of the homogenate (1 mL formamide/100 mg lung) (see Note 13).
3.6. Measurement of Capillary Filtration Coefficient (Kf,c )
The isolated perfused lung preparation is a well-established tech nique that can be used to assess permeability of lung vessels to fluids and solutes by measuring the capillary filtration coefficient (Figs. 4 and 5). This method can be used to address the changes in basal permeability in lungs of genetically modified mice, or in lungs challenged with receptor agonists (24) (e.g., PAR-1 agonist receptor, VEGF or with drugs), as well as in mice developing edema during lung inflammation (e.g., pneumonia or sepsis). The capillary filtration coefficient is derived from the Starling equation to express a sensitive measurement of transcapillary fluid conductance (25). Starling equation : J v = (L pS )[(Pc - Pi ) - s(p c - p i )] Where Jv equals the volume flux of fluid (mL/min); Lp is hydraulic conductivity (cm × per min × per mmHg); S is capillary surface area (cm2); Pc and Pi are capillary and interstitial fluid hydrostatic pres sures (mmHg); pc and pi are capillary and interstitial colloid oncotic
346
A.N. Garcia et al.
Fig. 4. Schematic diagram of murine-lung perfusion setup. Isolated lung preparation with arterial and left atrial (venous) cannula is perfused from a reservoir with the aid of peristaltic pump and is mechanically ventilated. Lung wet weight and arterial and venous pressures are continuously recorded [reproduced from Textbook of Pulmonary Vascular Disease Chapter 36: Animal Models of Lung Vascular Permeability figure 5A (Hanif et al., currently in press) with kind permission of Springer Science and Business Media].
Fig. 5. Stability of murine lung preparation. Lung preparation perfused with RPMI medium (t = 37°C) supplemented with 5% albumin remains isogravimetrically stable over 3 h after isolation. The liquid permeability of lung vessels is also nearly constant as determined by periodic Kfc measurements. Dg denotes ling weight change (in grams); Kfc is the capillary filtration coefficient (mL/min/cmH2O/dry lung). Data points represent mean values (± SE) from three lung preparations.
23 Permeability of Endothelial Barrier…
347
pressures; and s is the osmotic reflection coefficient of the vessel wall. Note that Kf,c is equal to the product LpS. Gravimetric Kf,c measurements are obtained by comparing the rate of lung weight gain during a baseline isogravimetric period with the rate after a change in hydrostatic pressure. Unlike other gravimetric assessments of pulmonary edema, edema formation due to increased vascular filtration pressure can be differentiated from edema resulting from increased vascular permeability. Although a widely used method, this technique suffers from the disadvantages of requiring difficult surgical techniques and a deli cate setup. Lung Preparation (Adopted from original protocol by Parker et al. (3), Fig. 6). 1. Warm RPMI 1640 Medium in water bath to ensure that medium entering the lung is at 37°C. 2. Anesthetize mouse with ketamine (150 mg/kg body weight) or 2.5% isoflurane in room air at a flow rate of 2 L/min inside
Fig. 6. Surgical procedures for perfused lung preparation. (a) Exposure of the trachea via neck incision. (b) Sternotomy of a mouse, exposing the heart and lungs. A vertical incision is carefully made in the chest of the mouse. A clamp holds the rib cage preventing any obstructions in the operating field. (c) Cannulation of the pulmonary artery. An incision is made in the right ventricle of the mouse heart. A PE-60 catheter is inserted into the pulmonary artery and is used to circulate 3% BSA–RPMI medium through the lung.
348
A.N. Garcia et al.
of an anesthesia chamber. In the latter case, anesthesia should be continued by means of a nose cone using a face mask. 3. Inject heparin (100 IU) into retroorbital venous sinus. 4. With animal lying on its back, make a small incision in neck to expose the trachea and jugular vein. 5. Make an incision in the trachea, being careful not to com pletely cut through (Fig. 6a). 6. Cannulate trachea (with 20-gauge catheter) and ventilate lungs with a gas mixture of 20% O2, 5% CO2, and 75% N2. Adjust to 125 breaths/min with a tidal volume between 200 and 300 mL. 7. Secure trachea and catheter together firmly using a suture. 8. Make an incision along sternum of mouse to expose heart and lungs (Fig. 6b; see Note 14). 9. Exsanguinate the mouse by cutting the inferior vena cava and abdominal aorta. 10. Make an incision into the right ventricle and insert PE60 catheter into pulmonary artery (Fig. 6c). 11. Pump the RPMI 1640 supplemented with 3% BSA medium through the pulmonary artery in a noncirculating manner with a peristaltic pump at a flow rate of 2 mL/min. 12. Tie a suture around aorta and pulmonary artery/catheter to secure it in place. 13. Loosely loop a suture around the left atrium. 14. Make an incision in the left atrium and insert a PE90 catheter to drain circulating media. 15. Tighten suture loop to secure catheter in place. 16. Remove the heart and lungs en bloc and suspend from a beam balance attached to a force transducer or calibrated displace ment transducer. Measurement of Kf,c (Fig. 7). 1. Allow lungs to remain in an isogravimetric state at a perfusion rate of 2 mL/min (see Note 15). 2. Increase the venous pressure by at least 6 cmH2O for a period of 20 min. 3. Simultaneously occlude arterial and venous lines for 5 s (before and immediately after the pressure change) to obtain the microvascular pressure Ppc (see Fig. 7, pressure recordings). 4. Monitor the rate of weight gain of the system (see Note 16).
23 Permeability of Endothelial Barrier…
349
Pressure (cm H2O)
25 20 15
Ppc 15
10 Ppc 8.6
5 0
Lung Weight Change (g)
15
20
25
30
35
40
45
0.4 0.3 0.2
Filtration rate 0.0032 ml/min
0.1 0 −0.1 15
20
25
30
35
40
45
Time (min) from beginning of perfusion
Fig. 7. Method for deriving capillary filtration coefficient (Kf,c) from experimental data. Murine lung is isolated, perfused, and ventilated (see text for details). Lung preparation undergoes a 20-min equilibration perfusion. Pulmonary capillary pressure (Ppc) is determined by double occlusion of arterial and venous lines at 5-min intervals. Venous pressure is elevated for 20 min (between 20th and 40th min). This causes capillary pressure to rise from 8.6 to 15 cmH2O (arrows) indicating a DPpc of 6.4 cmH2O. Lung wet weight increase is recorded and filtration rate of 0.0032 mL/min is determined from terminal 5-min slope of weight gain curve as indicated. Filtration rate is normalized by DPpc and lung dry weight (typically about 0.025 g in adult mice) to obtain the Kf,c value in mL/min/cmH2O/g.
5. Filtration rate is determined from the slope of the weight gain curve at 15–20 min. Kf,c is calculated as follows:
K f,c = (DW / Dt ) / DPpc , where W is weight; t is time, Ppc is pulmonary microvascular pressure. 6. Normalize the measured filtration rate by the microvascular pressure change and the lung’s dry weight to yield Kf,c in units of mL/min/cmH2O/dry lung g.
3.7. Measurement of Permeability × Surface Area Product ( PS )
Similar to the experimental procedure involved in measurement of the capillary filtration coefficient, measurement of permeability x surface area product (PS) can be used to assess permeability to plasma proteins in an intact lung. PS determines albumin clearance
350
A.N. Garcia et al.
in the lung under isogravimetric conditions; clearances of a wide variety of proteins with varying molecular radii can be determined in a similar manner to study the sieving properties of the pulmo nary vasculature. PS can be measured using a single sample method in which 125I-albumin is used as a tracer (26), or in a dual tracer method which also utilizes 131I-labeled albumin (27). Adopted from original protocol by Kern et al. (26). 1. Prepare a perfused lung as shown in the protocol for Lung Preparation (steps 1–17). 2. Perfuse lung with RPMI supplemented with 3% BSA media to remove residual blood from the vasculature. 3. Perfuse lung with 125I-labeled albumin (160,000 counts/mL prepared as shown in protocol Permeability of the endothelial monolayer to 125I-albumin) in RPMI media for 3 min. 4. Wash out lungs to remove 125I-albumin tracer by perfusing with RPMI containing 3% unlabeled BSA. 5. Remove lung preparations and blot off excess liquid. 6. Weigh samples and count for gamma radioactivity. 7. Calculate the PS in the units of mL/min/g (dry lung) using the equation:
A / (C p ´ t ),
where A is equal to the tissue concentration of tissue albumin (counts/g); Cp is equal to the tracer concentration in the per fusing liquid (counts/mL); t is equal to the exposure time (i.e., 3 min). 3.8. Measurement of Osmotic Reflection Coefficient (s)
The single-tracer method can be further modified and adopted for measurement of convective albumin clearance by using a dual tracer method (27). 125I is used to determine the diffusive flux of albumin, whereas the second tracer 131I measures total convective flux following an increase in pressure (28). This method assesses the osmotic reflection coefficient of the vessel wall, s, a dimen sionless number from 0 to 1 in which a value of 1 denotes imper meability to the solute. J s = PS DC + (1 - s) J vC s where Js is total solute flux of the tracer, Jv is the water flux across the endothelium (mL/min); P is the permeability (cm/s); S is the capillary surface area (cm2); ∆C is the concentration difference across the endothelium; and Cs is the average concentration of the solute across the endothelium. s can be expressed as:
s = 1 - (( J s - PS DC ) / ( J vC s ))
23 Permeability of Endothelial Barrier…
351
s can be adjusted for the time (t) yielding: s = 1 - [(A ¢ - PS DCt ) / (DV isC s )]
or
s = 1 - ((total tracer flux)(t ) - (diffusive tracer flux)(t ) / (water flux)(t )(C s )) where A¢ is a total flux of albumin tracer 2; ∆Vis – change in the interstitial volume due to water flux. Adopted from original protocol by Kern et al. (27). 1. Prepare a perfused lung as described in Lung Preparation protocol (steps 1–18). 2. Perfuse lung with RPMI media supplemented with 3% BSA to remove residual blood from the lung. 3. Perfuse lung with 125I-albumin tracer (160,000 counts/mL, prepared as shown in protocol Permeability of the endothelial monolayer to 125I-albumin) in RPMI for 3 min. 4. Wash out the lungs by perfusing with RPMI containing 3% unlabeled BSA. 5. Perfuse lung with 131I-albumin tracer (160,000 counts/mL) in RPMI and rapidly elevate venous pressure 3–6 cmH2O for 3 min to induce convective flux into the interstitium. 6. Quickly lower the vascular pressure to an isogravimetric level and wash out the lungs by perfusing with RPMI containing 3% unlabeled BSA for 3 min. 7. Remove lungs and blot off excess of liquid. 8. Homogenize lungs and count for gamma radioactivity of each tracer. 9. Calculate PS in units of mL/min/g (dry lung) using the equation:
A / (C p ´ t ) where A is equal to the tissue concentration of albumin (counts/g); t is the time of tracer 2 diffusion (3 min); ∆C is the concentration of tracer in the perfusing liquid. 10. Calculate s as
s = 1 - ((A ¢ - PS DCt ) / (DV isC s )) where A¢ is total albumin flux of tracer 2; ∆Vis is change in the interstitial volume due to water flux; ∆C is concentration dif ference across the endothelium (i.e., the vascular concentra tion); Cs is average concentration of the solute across the endothelium; t is the time that the endothelium is exposed to tracer 2.
352
A.N. Garcia et al.
Cs and ∆Vis are calculated as follow: ∆Vis (the increase in interstitial fluid volume) can be calculated from the second term of the double exponential that best fits the weight change (27). The rate of weight gain taken after the increase in venous pressure is multiplied by the time of pressure increase (3 min). Cs is unknown variable, and therefore the two extremes are used:
C s = Vascular concentration (C cap ) C s = (C cap / 2)
4. Notes 1. Slides are available with one (8W 1E ECIS Cultureware), ten (8W 10E) or two sets of 20 (8W 10E+) circular electrodes within the single well. 2. The membrane inserts can be obtained with a wide range of pore sizes, allowing the use of a variety of different tracers. However, pore sizes greater than 3 mm should not be used to avoid cell migration through pores. 3. Note that depletion or deletion of some genes might change cell proliferative characteristics. If this is the case, the number of cells plated on the electrodes should be adjusted to reflect the difference in proliferation rate. The density of the cul tured cells from both control and depleted cells should be determined in parallel experiments using light or fluorescent (DNA staining) microscopy. 4. Usually we allow an additional 1 or 2 days after cells reach confluence for establishing a stable monolayer before we address the effect of barrier disrupting agents on permeability change [thrombin, histamine, vascular endothelial growth factor (VEGF), tumor necrosis factor (TNFa)]. If barrierincreasing agents [Sphingosine-1 phosphate (S1P), 3¢-5¢-cyclic adenosine monophosphate (cAMP)] are tested, 80–90% confluent monolayers should be used. 5. The data acquisition is performed inside the cell culture incu bator to ensure an adequate environment. 6. A low efficiency of labeling of certain proteins with 125I can be observed in the presence of reducing agents (e.g., 2-mer captoethanol, dithiothreitol, glycerol, etc.) in the reaction solution or if there are no tyrosine residues present on the surface of the protein. The latter can be remedied through
23 Permeability of Endothelial Barrier…
353
addition of denaturing agents to make buried tyrosine resi dues more accessible for labeling. 7. After removing the sample, add back 50 mL of buffer to maintain fluid levels equally between the two chambers. 8. An intraperitoneal injection of ketamine (150 mg/kg body weight) or 2.5% isofluorane in room air at a flow rate of 2 L/min inside of an anesthesia chamber can be used. In the latter case, anesthesia should be continued by means of a nose cone during surgical no-survivor procedures. 9. If lungs are separated into different lobes after removal, the drying process may take as little as 1 day. 10. An intraperitoneal injection of ketamine (150 mg/kg body weight) or 2.5% isofluorane in room air at a flow rate of 2 L/min inside of an anesthesia chamber can be used. In the latter case, anesthesia should be continued by means of a nose cone during surgical no-survivor procedures. 11. This is a survival surgical procedure and must be performed in aseptic conditions and in accordance with requirements of the University Animal Care Committee. 12. Measurement of absorbance at 740 nm allows for correction of contaminating heme pigments. 13. The modification of this protocol includes injection of Evans Blue into the tail vein (23). This is to avoid a survival surgical procedure. Briefly, anesthetized mice are injected with 100 mL of 1% Evans Blue Dye in PBS into the tail vein and 40 min later, mice are perfused with PBS containing 2 mM EDTA via heart puncture for 20 min. Lung and other organs are removed and incubated in formamide and OD is measured as described above. 14. At this point, another injection of heparin can be given intra venously if there was a complication with the retroorbital injection. 15. Pressure in the system should be within the range of 5–8 cmH2O. Pressures exceeding this range might indicate an obstruction in the lung. 16. Measurements within the last 5 min of data collection are the most relevant. References 1. Komarova, Y., and A.B. Malik. (2010). Regulation of endothelial permeability via par acellular and transcellular transport pathways. Annu Rev Physiol 72:463–493. 2. Mehta, D., and A.B. Malik. (2006). Signaling mechanisms regulating endothelial permeabil ity. Physiol Rev 86:279–367.
3. Parker, J.C., and M.I. Townsley. (2004). Evaluation of lung injury in rats and mice. Am J Physiol Lung Cell Mol Physiol 286:L231–246. 4. Tiruppathi, C., A.B. Malik, P.J. Del Vecchio, C.R. Keese, and I. Giaever. (1992). Electrical method for detection of endothelial cell shape
354
A.N. Garcia et al.
change in real time: assessment of endothelial barrier function. Proc Natl Acad Sci U S A 89:7919–7923. 5. Wegener, J., C.R. Keese, and I. Giaever. (2000). Electric cell-substrate impedance sensing (ECIS) as a noninvasive means to monitor the kinetics of cell spreading to artifi cial surfaces. Exp Cell Res 259:158–166. 6. Ellis, C.A., C. Tiruppathi, R. Sandoval, W.D. Niles, and A.B. Malik. (1999). Time course of recovery of endothelial cell surface thrombin receptor (PAR-1) expression. Am J Physiol 276:C38–45. 7. Yalow, R.S., and S.A. Berson (1960) Immunoassay of endogenous plasma insulin in man J Clin Invest 39:1157–75. 8. Du, T., and M.J. Alfa. (2004). Translocation of Clostridium difficile toxin B across polar ized Caco-2 cell monolayers is enhanced by toxin A. Can J Infect Dis 15:83–88. 9. John, T.A., S.M. Vogel, C. Tiruppathi, A.B. Malik, and R.D. Minshall. (2003). Quantitative analysis of albumin uptake and transport in the rat microvessel endothelial monolayer. Am J Physiol Lung Cell Mol Physiol 284:L187–196. 10. Siflinger-Birnboim, A., P.J. Del Vecchio, J.A. Cooper, F.A. Blumenstock, J.M. Shepard, and A.B. Malik. (1987). Molecular sieving charac teristics of the cultured endothelial mono layer. J Cell Physiol 132:111–117. 11. Bhattacharya, J. (1988). Hydraulic conduc tivity of lung venules determined by split-drop technique. J Appl Physiol 64:2562–2567. 12. Parker, J.C., T. Stevens, J. Randall, D.S. Weber, and J.A. King. (2006). Hydraulic con ductance of pulmonary microvascular and macrovascular endothelial cell monolayers. Am J Physiol Lung Cell Mol Physiol 291:L30–37. 13. Solodushko, V., J.C. Parker, and B. Fouty. (2008). Pulmonary microvascular endothelial cells form a tighter monolayer when grown in chronic hypoxia. Am J Respir Cell Mol Biol 38:491–497. 14. Hemingway, A. (1950). A method of chemi cal analysis of guinea pig lung for the factors involved in pulmonary edema. J Lab Clin Med 35:817–822. 15. Pearce, M.L., J. Yamashita, and J. Beazell. (1965). Measurement of Pulmonary Edema. Circ Res 16:482–488. 16. Fukuda, N., H.G. Folkesson, and M.A. Matthay. (2000). Relationship of interstitial fluid volume to alveolar fluid clearance in mice: ventilated vs. in situ studies. J Appl Physiol 89:672–679. 17. Kobayashi, H., R. Hataishi, H. Mitsufuji, M. Tanaka, M. Jacobson, T. Tomita, W.M. Zapol, and R.C. Jones. (2001). Antiinflammatory
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
properties of inducible nitric oxide synthase in acute hyperoxic lung injury. Am J Respir Cell Mol Biol 24:390–397. Klinzing, S., T. Lesser, H. Schubert, M. Bartel, and U. Klein. (2000). Wet-to-dry ratio of lung tissue and surfactant outwash after one-lung flooding. Res Exp Med (Berl) 200:27–33. Boutoille, D., X. Marechal, M. Pichenot, C. Chemani, B. Guery, and K. Faure. (2009). FITC-albumin as a marker for assessment of endothelial permeability in mice: comparison with 125I-albumin. Exp Lung Res 35: 263–271. Patterson, C.E., R.A. Rhoades, and J.G. Garcia. (1992). Evans blue dye as a marker of albumin clearance in cultured endothelial monolayer and isolated lung. J Appl Physiol 72:865–873. Moitra, J., S. Sammani, and J.G. Garcia. (2007). Re-evaluation of Evans Blue dye as a marker of albumin clearance in murine models of acute lung injury. Transl Res 150:253–265. Dodd-o, J.M., M.L. Hristopoulos, N. Faraday, and D.B. Pearse. (2003). Effect of ischemia and reperfusion without airway occlusion on vascular barrier function in the in vivo mouse lung. J Appl Physiol 95:1971–1978. Standiford, T.J., S.L. Kunkel, N.W. Lukacs, M.J. Greenberger, J.M. Danforth, R.G. Kunkel, and R.M. Strieter. (1995). Macrophage inflammatory protein-1 alpha mediates lung leukocyte recruitment, lung capillary leak, and early mortality in murine endotoxemia. J Immunol 155:1515–1524. Vogel, S.M., X. Gao, D. Mehta, R.D. Ye, T.A. John, P. Andrade-Gordon, C. Tiruppathi, and A.B. Malik. (2000). Abrogation of thrombin-induced increase in pulmonary microvascular permeability in PAR-1 knockout mice. Physiol Genomics 4:137–145. Parker, J.C., M.N. Gillespie, A.E. Taylor, and S.L. Martin. (1999). Capillary filtration coef ficient, vascular resistance, and compliance in isolated mouse lungs. J Appl Physiol 87:1421–1427. Kern, D.F., D. Levitt, and D. Wangensteen. (1983). Endothelial albumin permeability measured with a new technique in perfused rabbit lung. Am J Physiol 245:H229–236. Kern, D.F., and A.B. Malik. (1985). Microvascular albumin permeability in iso lated perfused lung: effects of EDTA. J Appl Physiol 58:372–375. Rippe, B., and A. Taylor. (2001). NEM and filipin increase albumin transport in lung microvessels. Am J Physiol Heart Circ Physiol 280:H34–41.
Chapter 24 Size-Selective and In Vitro Assessment of Inner Blood Retina Barrier Permeability Matthew Campbell and Peter Humphries Abstract Assessment of tight junction integrity in vitro is fundamental when studying molecular processes that may be implicated in barrier dysfunction. At the blood brain and inner blood retina barrier (BBB and iBRB, respectively) adjacent endothelial cells lining the microvasculature have been shown to have very low rates of fluid phase transcytosis and high electrical resistances, due in part to the expression of tight junction proteins at the apical periphery of these cells. While these high electrical resistances are difficult to achieve in vitro, owing to complex interactions of endothelial cells in vivo with astrocytes and pericytes, it is possible to make an assessment of paracellular permeability when cells are analysed on a number of different fronts. In this regard, we will outline here a method for determining trans-endothelial electrical resistance, tracer molecule diffusion, and tight junction protein localization in primary cultures of bovine retinal microvascular endothelial cells. This system allows for the screening of a wide range of pro- and anti-angiogenic molecules in an in vitro model of the iBRB and can accurately assess the role individual tight junction proteins play in maintaining tight junction integrity in response to various cell stimuli. Key words: Tight junctions, Occludin, Blood retina barrier, Permeability
1. Introduction Determination of paracellular permeability in endothelial cells or indeed epithelial cells is a central physiological assay that can allow for the analysis of a wide range of effects in cells that express tight junctions. Tight junctions are formed at the apical periphery of endothelial cells or epithelial cells. They perform the dual role of creating a primary barrier to the diffusion of solutes through the paracellular pathway, while also maintaining cell polarity as a boundary between the apical and basolateral plasma
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_24, © Springer Science+Business Media, LLC 2011
355
356
M. Campbell and P. Humphries
membrane domains (1). Tight junctions are complex structures, which are composed of a series of integral and peripheral membrane proteins. The transmembrane proteins of the tight junction include occludin, junction adhesion molecule (JAM), and claudins, and they extend into the paracellular space, creating the seal characteristic of the tight junction (2). Of the known integral membrane proteins of the tight junction, the proteins Zonnula Occludens-1, -2, -3 (ZO-1, -2, -3), and cingulin may play an integral role in the scaffolding of the transmembrane proteins, while also creating a link to the perijunctional actin cytoskeleton (3). The 220 kDa phosphoprotein ZO-1, in particular, has three PDZ domains that could potentially bind to a wide variety of protein partners and allow for the control of tight junction assembly (4). At the ultra-structural level in freeze-fracture replicas, the tight junctions have been shown to appear as an intricate network of fibrils encircling the apical end of the lateral membrane in cells expressing tight junctions. These fibrils have been identified as the transmembrane proteins previously discussed (5). In particular, breakdown of the tight junctions lining the microvasculature of the retina, the inner blood retina barrier (iBRB), is a hallmark of many degenerative retinal diseases, including diabetic retinopathy, sickle-cell disease, and cystoid macula edema (6). Here, we describe the assessment of paracellular permeability in primary cultures of bovine retinal microvascular endothelial cells (RMECs) as a means of determining the effects of pro- and anti-angiogenic factors in an in vitro model of the iBRB as we have reported previously (7, 8). We also describe a Western blot protocol that allows for quantification of tight junction protein expression post-permeability assay and will allow for the correlation of physiological changes in barrier properties with molecular events at the tight junction.
2. Materials 2.1. Primary Cell Culture
1 M HEPES buffer (Sigma Aldrich). 70% ethanol (Sigma Aldrich). 85 mM gauze. 53 mM gauze. Isolation mixture: 10× Minimum Essential Medium stock (40 ml) (Gibco), sterile distilled H2O (340 ml), 1 M HEPES buffer (12 ml) (Sigma Aldrich), antibiotics/glutamine mixture (Gibco). Enzyme cocktail: 5 mg Pronase (Sigma Aldrich), 6 mg DNase (Sigma Aldrich), 3 mg collagenase (Sigma Aldrich). Growth medium: DMEM (Gibco) with 1 ml heparin (1 mg/ml) (Sigma Aldrich), 0.38 ml insulin (1 mg/ml) (Sigma Aldrich),
24 Size-Selective and In Vitro Assessment of Inner Blood Retina Barrier Permeability
357
10 ml antibiotics solution (streptomycin, penicillin, kanamycin) (Gibco), 10% porcine serum (Gibco), phosphate-buffered saline (PBS) (Gibco). 2.2. Permeability Assays
Gelatin (Sigma Aldrich). Costar® Transwell®-Clear inserts (Corning Incorporated, New York, USA) with a pore size of 0.4 mm. Endohm Chamber (World Precision Instruments, Aston, England), VEGF165 (Calbiochem). FD-70, or FD-4 (Sigma Aldrich). Endostatin (Abcam).
2.3. SDS-PAGE
Pre-stained protein marker (New England Biolabs). Atto electrophoresis equipment. BIO-RAD semi-dry electroblot. Nitrocellulose membrane (Whatman). Filter paper (Whatman). Rabbit anti-occludin (Invitrogen). Anti rabbit IgG-HRP (Abcam). BCA-assay (Pierce). 10× Running buffer: 30.3 g Tris, 144.2 g (Sigma Aldrich) glycine (Sigma Aldrich), 10 g SDS (Sigma Aldrich), and made up to 1 l, with a pH of 8.6. Tris-buffered saline (TBS): 50 mM Tris, 150 mM NaCl (Sigma Aldrich), pH 7.4, i.e. 6.05 g Tris-base (Sigma Aldrich), 8.766 g NaCl (Sigma Aldrich), adjust the pH with HCl, and bring the volume to 1 l with dH2O. 10% Acrylamide resolving gels: H2O (7.93 ml), 1.5 M Tris pH 8.8 (5 ml) (Sigma Aldrich), 10% SDS (0.2 ml) (Sigma Aldrich), bisacrylamide (30%) (6.67 ml) (Sigma Aldrich), 10% APS (0.2 ml) (Sigma Aldrich), TEMED (20 ml) (Sigma Aldrich). 4% Acrylamide stacking gels: H2O (6.8 ml), 0.5 M Tris pH 6.8 (2.5 ml) (Sigma Aldrich), 10% SDS (0.1 ml) (Sigma Aldrich), bisacrylamide (30%) (1.33 ml) (Sigma Aldrich), 10% APS (0.1 ml) (Sigma Aldrich), TEMED (10 ml) (Sigma Aldrich). 4× Sample buffer: 6.05 g Tris (Sigma Aldrich), 16 g SDS (Sigma Aldrich), 20 ml glycerol (Sigma Aldrich), 16 mg bromophenol blue (Sigma Aldrich), 10% mercaptoethanol (Sigma Aldrich), pH 6.6, in 100 ml. Transfer buffer: 3.03 g Tris (Sigma Aldrich), 14.42 g glycine (Sigma Aldrich), 200 ml methanol (Sigma Aldrich), and made up to 1 l. 5% Blocking solution: 5 g Marvell non-fat dry skimmed milk in 100 ml TBS. TBS/Tween: TBS containing 0.1% Tween-20 (Sigma Aldrich).
358
M. Campbell and P. Humphries
2.4. Developing Western Blots
PIERCE reagent kit. SARSTEDT tubes. Rabbit anti-b-actin (Abacm). FUJI X-ray film. Re-store Stripping solution (Pierce).
2.5. Immuno histochemistry
Normal goat serum (NGS) (Sigma Aldrich). Rabbit anti-ZO-1 (Invitrogen). Rabbit anti-claudin-5 (Invitrogen). Anti-rabbit IgG:Cy3 (Molecular Probes). Permeabilisation buffer: 0.05% Triton X-100 (Sigma Aldrich) in PBS (Gibco). Blocking solution: 5% NGS in permeabilisation buffer. DAPI: (1:5,000 dilution of a 1 mg/ml stock). Vectasheild (Vectorlabs).
3. Methods 3.1. Isolation of Primary Bovine Retinal Microvascular Endothelial Cells
1. Dispense ~10 ml of isolation medium into two separate sterile Petri dishes. 2. Take a large Petri dish, soak in 70% ethanol and place a bovine eye in the petri dish, wipe the outside of the eye with 70% ethanol and using a scissors and a forceps, strip away the excess fat and muscle tissue from the sides of the eye. 3. Gently make a 1-in. incision vertically at the side of the eye approximately above the neural retina in the region of the pars plana. When the vitreous is visible, use a scissors to cut around the eye horizontally in both directions, until the vitreous can be gently eased out. 4. Using a tweezers, gently remove the neural retina, and snip off at the optic nerve. Shake off any pigmented material before placing the retina in a Petri dish containing isolation medium. Place five to ten retinas in each Petri dish containing isolation medium. 5. Tip the contents of each Petri dish separately into a homogenising tube, and mash the retinas with a dounce homogenizer until vessels can be seen to be adequately homogenised. Place the homogenate into large centrifuge tubes, and spin at 500 × g for 10 min. Remove the supernatant, and re-suspend each pellet in 5 ml isolation medium. This re-suspension is then filtered through 85 mM gauze, using the blunt end of a sterile pipette. 6. The remnants left on top of the filter are washed with isolation medium, and then placed in a Petri dish containing 10 ml enzyme cocktail and left at 37°C for 27 min.
24 Size-Selective and In Vitro Assessment of Inner Blood Retina Barrier Permeability
359
7. After 27 min, the enzymatic reaction is stopped by the addition of 10 ml of ice cold isolation medium. This was then filtered through a 53-mM gauze, and the remnants left on top were placed into 10 ml of isolation medium. Cells are then centrifuged at 500 × g for 10 min. 8. The supernatant is removed, and the pellet is re-suspended in 10 ml of growth medium which will be used in all subsequent culturing of bovine retinal endothelial cells. The isolated endothelial cells are seeded in small Petri dishes, and microvessels should begin to grow in approximately 3–5 days. 3.2. Trans-endothelial Electrical Resistance Determination
1. RMECs are seeded at a cell density of 5 × 105 cells/ml and grown to confluence on 0.1% gelatin-coated Costar® Transwell®-Clear inserts with a pore size of 0.4 mm. 2. Trans-endothelial electrical resistance (TEER) is determined for RMECs using an EVOM resistance meter with Endohm Chamber and a Millicell-Electrical Resistance System. 3. For measurement of TEER, both the apical and basolateral sides of the endothelial cells are bathed in fresh growth medium at 37°C, and a current is passed across the monolayer with changes in electrical resistance measured in units of W/cm2. 0.1% SDS is added to the apical chamber of one group of wells 1 h prior to determination of TEER and will act as a positive control. 4. Electrical resistance for RMECs is measured in triplicate, and the inherent resistance of a blank filter coated with 0.1% gelatin is subtracted from the values obtained for the cells.
3.3. FD-70/FD-4 Permeability Assay
1. RMECs are seeded at a cell density of 5 × 105 cells/ml and grown to confluence on Costar® Transwell®-Clear inserts with a pore size of 0.4 mM. 2. Upon reaching confluence, cells are serum deprived for 6 h prior to treatment with pro- or anti-angiogenic factors of the experimenter’s choice. In this case, we have used DMEM containing 10 ng/ml VEGF165 or 20 ng endostatin or a combination of both. 3. Twenty-four hours post-treatment, 250 mg/ml of either FD-70, -40, or FD-4 (Sigma Aldrich) in 500 ml growth medium, is added to the apical chambers of the Transwell inserts. 4. The cells are placed on a shaker at 37°C, and at time “zero,” 150 ml growth medium is removed from each basolateral chamber, and replaced with new medium, bringing the basolateral volume to 1.8 ml. 5. Sampling aliquots are taken every 15 min for 120 min, and placed in a 96-well plate.
360
M. Campbell and P. Humphries
6. The fluorescence of FD-70, -40, or -4 (see Note 2) is determined in a fluorescence spectrofluorometer at an excitation wavelength of 490 nm and an emission wavelength of 515 nm. The apparent permeability coefficient (Papp) of FD-70, -40, or -4 is calculated using the following equation. 7. Papp (cm/s) = dQ/dt × 1/A × Co, where dQ/dt (mg/s) is the rate of appearance of FD-70/-40/-4 on the receiver side from 30 min to 120 min after application of FD-70/-40/-4. Co (mg/ml) is the initial FD-70/-40/-4 concentration on the donor side, and A (cm2) is the effective surface area of the insert. 8. Calculations Read RFU for time points 0–120 min. Convert RFU to concentration (from standard curve). Convert to mmol. ●●
70,000 g FD-70, 4,000 g FD-4 = 1 mol.
●●
X g FD-70/4 = ((1 mol) × X)/70,000 mol or 4,000 mol.
●●
Y mol = (Y mol) × 1,000 mmol.
Calculate the cumulative amount Q. ●●
●●
Qt = (CtVr + (sum of all previous concentrations)Vs) where Qt = cumulative amount at time t, C t = concentration (mmol) at time t, Vr = receiver side volume (ml), Vs = sampling volume (ml).
9. Plot Q versus time (seconds) The slope (m from y = mx + c) is dQ/dt, the rate of appearance of compound in the receiver side. 10. Calculate Papp = (dQ/dt)/(A×Co) ●●
where A = area, Co = initial concentration.
11. With regard to the interpretation of this data with corresponding Western analyses, please refer to Note 3. 3.4. Western Blot Analysis of Occludin Expression in RMEC’s 3.4.1. Preparation of SDS-PAGE Gels
1. Electrophoresis plates are washed thoroughly before use, with water and ethanol. 2. The resolving gel is poured into the plates, leaving a space, the length of the Teflon comb, for the stacking gel, plus 1 cm. Distilled H2O is gently pipetted on top of the resolving gel, to prevent bubbles forming. The gel is allowed to set for approximately 30 min. 3. The stacking gel is poured on top of the resolving gel, and the Teflon comb is placed into the gel and allowed to set for approximately 20 min. Once set, the plates are placed into electrophoretic apparatus. The electrophoretic apparatus was
24 Size-Selective and In Vitro Assessment of Inner Blood Retina Barrier Permeability
361
then filled with 1× running buffer prepared from the 10× running buffer stock. 3.4.2. Loading of Protein Samples into Wells
1. Following protein quantitation by BCA-assay, 30 ml of protein sample was added to 10 ml of 4× sample buffer. 2. Protein samples are then boiled for 2 min, to allow for de-naturing. 3. Samples are then loaded into separate lanes, with a pre-stained protein marker on the outermost lane. 4. The electrophoretic equipment is run at ~15 mA per gel for approximately 2 h.
3.4.3. Semi-dry Transfer
1. Once the protein front is approximately 1 cm from the bottom of the gel, the plates are removed from the electrophoretic apparatus, and the stacking gel is removed. 2. A BIO-RAD semi-dry electroblot apparatus is prepared by cutting eight pieces of blotting paper to the exact size of the gel. 3. Four pieces of blotting paper are soaked in transfer buffer, and placed on the anode of the electroblot apparatus. 4. Nitrocellulose membrane is then placed on top of the soaked blotting paper, and the resolving gel is gently placed over this, followed by the remaining four pieces of blotting paper, which are also soaked in transfer buffer. 5. The cathode is placed over this, and the electroblot apparatus is run at 0.9 mA per cm2 of gel for approximately 2–3 h. 6. The proteins should move from the gel to the nitrocellulose membrane, due to the negative charge induced by SDS, and will be attracted to the positive charge of the anode on which the nitrocellulose membrane is placed.
3.4.4. Processing of Blots
1. Following electroblotting, the membrane is briefly washed in TBS in order to remove trace amounts of methanol, which may increase background. 2. The membrane is then incubated with 5% blocking solution. Blocking can be carried out for 1 h at room temperature on a shaker, or overnight at 4°C. This will prevent non-specific binding of the 1o antibody, as it will bind to all non-specific protein binding sites on the membrane. 3. Following blocking, the membrane is briefly washed with TBS. 4. A primary antibody, specific for the protein of interest, i.e. rabbit anti-occludin (1:1,000 dilution) is added to the membrane in 5% blocking solution.
362
M. Campbell and P. Humphries
5. The membrane is incubated with 1° antibody at room temperature for approximately 3 h, or overnight at 4°C. 6. The antibody/blocker solution is pipetted off the membrane to be re-used, and the membrane is washed for 15 min with TBS/Tween. Following this washing, the membrane was washed for 3 × 15 min with TBS without Tween. 7. A secondary antibody is then incubated with the membrane at a dilution of 1:2,500 for approximately 1 h at room temperature. Secondary antibodies are species specific with respect to the IgG, and can be conjugated to enzymes for use in chemiluminescent detection of proteins. Conjugates used in this are the enzyme horseradish-peroxidase. These conjugates can cleave the chemiluminescent substrate luminol to an activated intermediate which decays to the ground state by emitting light. 8. Following incubation with the secondary antibody, the membrane is washed 4 × 15 min with TBS/Tween (0.1%), ensuring the shaker is moving at a high speed. 3.4.5. Chemiluminescent Detection of Proteins
1. Preparation of chemiluminescent reagent is done in the dark. Using a PIERCE reagent kit, equal volumes of substrate solution and enhancer solution are added to a SARSTEDT tube, to be re-used for up to 5 days. 2. Exposure of the blot to the detection reagents must be done in the dark, and excess TBS is drained off the membrane before exposure to the reagents. Detection solution is left on the membrane for 1 min, after which time it is drained off the membrane using tissue or blotting paper. 3. The membrane is placed on a transparent sheet, (i.e., saran wrap/acetate film), and a sheet of X-Ray film is cut out to cover the blot. 4. The film is placed over the blot in a film cassette, and exposed to the membrane for between 1 min and 20 min, depending on the intensity of the signal. Typically occludin signal will need 15 min to develop sufficiently. 5. The film is developed using a film developing machine. 6. For normalisation, the membrane is stripped using commercially available stripping solution (Pierce) and without reblocking, a rabbit anti-b-actin antibody is applied to the membrane overnight (1:2,000 dilution) at 4°C. The protocol for subsequent development of the b-actin blot is essentially identical to that described above for occludin. 7. Blots can be re-probed for numerous other tight junction associated proteins (see Note 1).
24 Size-Selective and In Vitro Assessment of Inner Blood Retina Barrier Permeability
3.5. Indirect Immunostaining of Tight Junction Proteins for Confocal or Epifluorescent Microscopy
363
1. RMECs on Transwell filters are washed twice with ice-cold PBS and subsequently fixed for 10 min at room temperature with pre-cooled methanol and subsequently washed twice with PBS. 2. Cells are soaked in PBS for 20 min at room temperature and subsequently permeabilised with permeabilisation buffer for 30 min at room temperature. 3. Non-specific binding is blocked by incubation with blocking solution for 30 min at room temperature. 4. 1o anti-rabbit IgG antibody specific for either occludin (1:100), ZO-1 (1:100), or claudin-5 (1:100) is diluted in 1% NGS in permeabilisation buffer, and incubated with cells overnight at 4°C. 5. Cells are washed three times with PBS, and blocked again as described above with blocking solution for 30 min at room temperature. The 5% blocking solution is replaced by 1% NGS with a 2o anti-rabbit IgG:Cy3 conjugated secondary antibody (1:200). This is incubated with cells for 1 h at 37°C. Cells are washed three times with PBS and nuclei are stained with DAPI for 30 s. 6. Transwell membranes are cut from filters using a sterile scalpel and are mounted on a microscope slide using vectashield. Primary isolated RMECs should be seeded at a high cell density on Transwell inserts so that a robust monolayer is formed. The determination of TEER for these cells is necessary in ensuring that all wells have approximately equal monolayers and that all monolayers are intact as a disruption in cell integrity on the membrane will have consequences for the FITC-dextran permeability assay. In general however, TEER values for these cells will be too low to accurately assess subtle changes in barrier physiology. Once seeded on Transwells, TEER can be determined each day post-seeding of cells until a plateau has been reached. Typically, primary bovine RMECs will generate TEER values of between 50 and 60 W/cm2 and the addition of 0.1% SDS to the apical chamber will act as a positive control (Fig. 1).
3.6. Treatment with Angiogenic Factors and Analysis of Changes in the Apparent Permeability Coefficient (Papp )
Treatment of RMECs with 10 ng/ml VEGF165 added to the basolateral side of a trans-well insert elicited a significant increase (P £ 0.05) in the permeability of RMECs after 24 h (FD-70 flux across the monolayer). Endostatin was shown to reverse this effect. Simultaneous treatment of cells with 10 ng/ml VEGF165 and 20 ng/ml Endostatin showed a marked decrease in permeability, as did initial treatment for 2 h with 10 ng/ml VEGF165 followed by treatment with 20 ng/ml Endostatin and vice versa (P £ 0.05). Apical treatment of cells with VEGF165 elicited no
364
M. Campbell and P. Humphries
Fig. 1. Treatment of RMEC’s with 0.1% SDS acts as a positive control and causes a decrease in trans-endothelial electrical resistance (TEER) across the RMEC monolayer.
Fig. 2. Treatment of RMECs on the basolateral side of a Transwell filter elicited a significant increase in the permeability coefficient across the monolayer P £ 0.05. This VEGF-mediated permeability was shown to decrease significantly upon treatment of cells for 24 h with 20 ng/ml Endostatin (P £ 0.05). VEGF-mediated permeability was also shown to decrease upon prior and subsequent treatment of cells with 20 ng/ml Endostatin (P £ 0.05).
change in permeability, whereas both apical and basolateral treatment with Endostatin showed a marked decrease in VEGF165mediated permeability (Fig. 2). Similar to FD-70 flux across the RMEC monolayer, it is observed that FD-4 flux is also increased following the treatment of cells with 10 ng/ml VEGF and that
24 Size-Selective and In Vitro Assessment of Inner Blood Retina Barrier Permeability
365
Fig. 3. Treatment of RMECs on the basolateral side of a Transwell filter elicited a significant increase in the permeability of FD-4 across the monolayer P £ 0.05. This VEGF-mediated permeability was shown to decrease significantly upon treatment of cells for 24 h with 20 ng/ml Endostatin (P £ 0.05).
Fig. 4. Western blot analysis of occludin expression post-treatment of RMECs with VEGF showed a distinct change in the pattern of migration of occludin with extra bands evident below the main 60 and 62 kDa immunoreactive forms of occludin. This resolving pattern was prevented by following addition of endostatin to the growth medium.
this increase in permeability can be attenuated with exposure of cells to the anti-angiogenic factor endostatin (20 ng/ml) for 24 h prior to permeability assay (Fig. 3). Western blot analysis of occludin expression in RMEC cells will generally correlate with any changes in permeability that are observed in the assays described above as can be seen in Fig. 4, where treatment of cells with VEGF causes a distinct change in
366
M. Campbell and P. Humphries
the pattern of migration of occludin in the resolving gel. Occludin is highly susceptible to post-translational modifications (9) and these modifications have previously been shown to cause increases in the permeability of both endothelial cells and epithelial cells, due, in part to post-translational modifications of this tight junction protein (10). Therefore, these methods can be applied in all cell types that form tight junctions.
4. Notes 1. While this protocol only describes Western blot analysis of occludin as described in Subheading 3.4, it can also be applied to the analysis of claudins-1, -3, -5, and -12 which have previously been shown to mediate the paracellular integrity of endothelial cells and which can all be analysed using the exact same protocol as described above (7). 2. In general, when assessing changes in paracellular permeability of endothelial cells or indeed epithelial cells (Subheadings 3.2 and 3.3), it is good scientific practice to use two tracer molecules of varying molecular weight so that a size-selectivity of barrier modulation can be assessed. In addition, Western blots can be stripped and re-probed accurately up to three times and therefore, more than one tight junction protein should be analysed per experiment (8). 3. When correlating changes in permeability to transcriptional regulation of tight junction proteins, it is important to bear in mind that in general, it is occludin and claudin expression that will represent a clear molecular event, with changes in ZO-1, -2, and -3 expression playing a lesser and more indirect role in paracellular permeability changes (Subheadings 3.2–3.4).
Acknowledgement The authors thank the foundation Fighting Blindness Ireland who supported the laboratory during the development of these protocols. References 1. Sakakibara A, Furuse M, Saitou M, AndoAkatsuka Y, Tsukita S. (1997) Possible involvement of phosphorylation of occludin in tight junction formation. J Cell Biol. Jun 16;137(6):1393–401.
2. Fanning AS, Little BP, Rahner C, Utepbergenov D, Walther Z, Anderson JM. (2007). The unique-5 and -6 motifs of ZO-1 regulate tight junction strand localization and scaffolding properties. Mol Biol Cell. Mar;18(3):721–31.
24 Size-Selective and In Vitro Assessment of Inner Blood Retina Barrier Permeability 3. Riesen FK, Rothen-Rutishauser B, WunderliAllenspach H. (2002) A ZO1-GFP fusion protein to study the dynamics of tight junctions in living cells. Histochem Cell Biol. Apr;117(4):307–15. 4. Zahraoui A. (2004) Tight junctions, a platform regulating cell proliferation and polarity. Med Sci (Paris). May;20(5):580–5. Review. 5. Huber D, Balda MS, Matter K. (2000) Occludin modulates transepithelial migration of neutrophils. J Biol Chem. Feb 25;275(8):5773–8. 6. Gardner, T.W., Antonetti, D.A., Barber, A.J., Lieth, E., Tarbell, J.A. & the Penn State Retina Research Group. (1999) The molecular structure and function of the inner blood-retinal barrier. 97:229–237. 7. Campbell M, Collery R, McEvoy A, Gardiner TA, Stitt AW, Brankin B. (2006) Involvement of MAPKs in endostatin-mediated regulation
367
of blood-retinal barrier function. Curr Eye Res. Dec;31(12):1033–45. 8. Brankin B, Campbell M, Canning P, Gardiner TA, Stitt AW. (2005) Endostatin modulates VEGF-mediated barrier dysfunction in the retinal microvascular endothelium. Exp Eye Res. Jul;81(1):22–31. 9. Murakami T, Felinski EA, Antonetti DA. (2009) Occludin phosphorylation and ubiquitination regulate tight junction trafficking and vascular endothelial growth factor-induced permeability. J Biol Chem. Jul 31;284(31):21036–46. 10. Elias BC, Suzuki T, Seth A, Giorgianni F, Kale G, Shen L, Turner JR, Naren A, Desiderio DM, Rao R. Phosphorylation of Tyr-398 and Tyr-402 in occludin prevents its interaction with ZO-1 and destabilizes its assembly at the tight junctions. (2009) J Biol Chem. Jan 16;284(3):1559–69.
wwwwwwwwwwwww
Chapter 25 Assessment of Permeability in Barrier Type of Endothelium in Brain Using Tracers: Evans Blue, Sodium Fluorescein, and Horseradish Peroxidase Mehmet Kaya and Bulent Ahishali Abstract Blood–brain barrier (BBB) constituted primarily by the capillary endothelial cells functions to maintain a constant environment for the brain, by preventing or slowing down the passage of a variety of bloodborne substances, such as serum proteins, chemical compounds, ions, and hormones from the circulation into the brain parenchyma. Various diseases such as brain tumors, epilepsy, and sepsis disturb the BBB integrity leading to enhanced permeability of brain microvessels. In animal models, a variety of experimental insults targeted to the BBB integrity have been shown to increase BBB permeability causing enhanced passage of molecules into the brain paranchyma by transcellular and/or paracellular pathways. This alteration can be demonstrated by intravascular infusion of exogenous tracers and subsequent detection of extravasated molecules in the brain tissue. A number of exogenous BBB tracers are available, and they can be used for functional and structural analysis of BBB permeability. In this chapter, we aimed to highlight the basic knowledge on the use of three most commonly performed tracers, namely Evans blue dye, sodium fluorescein, and horseradish peroxidase. The experimental methodologies that we use in our laboratory for the detection of these tracers by macroscopy, spectrophotometry, spectrophotofluorometry, and electron microscopy are also discussed. While tracing studies at the morphological level are mainly aimed at the identification and characterization of the tracers both in the barrier related cells and brain parenchyma, spectrophotometric and spectrophotofluorometric assays enable quantification of BBB permeability. The results of our studies that we performed using the mentioned tracers indicate that barrier type of endothelial cells in brain play an important role in paracellular and/or transcytoplasmic trafficking of macromolecules across BBB under various experimental settings, which may provide new insights in both designing approaches for the management of diseases with BBB breakdown and developing novel trans-BBB drug delivery strategies. Key words: Blood–brain barrier, Permeability, Endothelium, Evans blue, Sodium fluorescein, Horseradish peroxidase
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_25, © Springer Science+Business Media, LLC 2011
369
370
M. Kaya and B. Ahishali
1. Introduction The concept of the blood–brain barrier (BBB) was initiated in the late nineteenth century, and anatomical evidence of BBB was confirmed using electron microscopy in the 1960s. The BBB which is primarily constituted by the brain capillary endothelial cells interacts with neighboring cells such as pericytes, astrocytes, and neurons to maintain the barrier integrity (1–5). Physiologically, BBB protects the brain from the compositional fluctuations of substances that occur in the circulation and plays a major role in maintaining the constant environment required for normal brain function which is called neuronal homeostasis. Under normal physiological conditions, the presence of continuous strands of tight junctions between adjacent endothelial cells of brain capillaries significantly prevents transport of polar solutes and macromolecules from circulation into the brain through the paracellular pathway (6–8). Transcellular pathway in barrier type of endothelial cells is highly selective and the rate of transcytosis of solutes is also very limited (9). Despite these limitations, the uptake of essential molecules into the brain is mediated through specific transport or carrier molecules, which renders transcytosis the main mechanism of trafficking of molecules across the BBB (10, 11). As a result, barrier type of brain capillaries governs a fairly limited capacity of transporting of substances from circulation to brain and vice versa. Under pathological conditions, the leakage of intravascular substances through the disrupted BBB to brain parenchyma occurs mainly by increased function of transcellular pathway (i.e., vesicular trancytosis and formation of transendothelial channels) and/or paracellular pathway (the opening of the intercellular tight junctions). There is a variety of low molecular weight vascular permeability markers such as radiolabeled alpha aminoisobutyric acid (MW 103), radiolabeled sucrose (MW 342), sodium fluorescein (MW 376), and radiolabeled inulin (MW 5,000); and high molecular weight vascular permeability markers such as horseradish peroxidase (MW 40,000), dextran (MW 70,000), and Evans blue (MW 961) which binds to albumin (Evans blue-Albumin, MW 69,000). This chapter focuses on the general characteristics of the three most commonly performed exogenous tracers, Evans blue, sodium fluorescein, and horseradish peroxidase, their use in assessment of BBB damage and their detection in brain parenchyma by macroscopy, spectrophotometry, spectrophotofluorometry, and electron microscopy after administration into the vascular system. Determination of extravasation of Evans blue, sodium fluorescein, and horseradish peroxidase tracers into brain parenchyma provides both qualitative and quantitative information about microvascular permeability of small and large
25 Assessment of Permeability in Barrier Type of Endothelium in Brain Using Tracers…
371
molecules. Under physiological conditions, systemically injected tracers, such as Evans blue, sodium fluorescein, and horseradish peroxidase cannot diffuse into brain parenchyma owing to the presence of tight junctions between endothelial cells and restricted transcellular pathway. However, an impairment in BBB integrity has been described in a number of experimental insults such as brain ischemia (12, 13), hypertension (14, 15), traumatic brain injury (16, 17), epileptic seizures (18, 19), and irradiation (20, 21) which have been shown to cause the extravasation of the above-mentioned tracers into the brain parenchyma as a result of disruption of BBB in animal models. 1.1. Evans Blue
Evans blue (T-1824) dye introduced as high-molecular weight permeability marker is widely used to study capillary and cellular membrane permeability as it is nontoxic in vivo. Evans blue binds to serum albumin as soon as it is presented to vascular system (22). The dye then leaks through the BBB in case of impairment and stains the brain lesion where the BBB is disrupted (Figs. 1–3). The advantage of using Evans blue tracer is its accurate representation of the zone of altered permeability that is especially useful in animal studies in which lesions with BBB breakdown in specific brain regions are investigated. Clasen et al. have shown that the
Fig. 1. Photograph of a rat brain exposed to left-sided carotid infusion of mannitol. Extensive Evans blue-albumin extravasation (dark areas) is prominent in the left hemisphere of the brain depicting unilateral BBB disruption (reproduced from ref. 28 with permission from Elsevier Science).
372
M. Kaya and B. Ahishali
Fig. 2. Representative picture of a brain taken after Evans blue administration in an Nw-nitro-L-arginine methyl ester + angiotensin II treated rat. Evans blue albumin extravasation (dark areas) is prominent in subcortical regions (reproduced from ref. 30 with permission from Elsevier Science).
Evans blue method may be used with confidence to assess vascular protein leakage macroscopically (23). The extravasated Evans blue can be extracted by different solvents and the content is determined by colorimetry at the absorbance maximum of 600–620 nm.
25 Assessment of Permeability in Barrier Type of Endothelium in Brain Using Tracers…
373
Fig. 3. Representative coronal sections of a rat brain after Evans blue administration during severe insulin-induced hypoglycemia. Note the extensive Evans blue extravasation (dark areas) in both cortical and subcortical regions (reproduced from ref. 26 with permission from NRC Research Press).
Fluorescence of the extravasated dye can also be determined by spectrophotofluorometry (excitation at 620 nm and emission at 680 nm). The tissue content of Evans blue is quantified from a linear standard curve derived from known amounts of the dye
374
M. Kaya and B. Ahishali
and expressed as mg per mg of brain tissue. Evans blue has been used as a tracer in BBB studies for years by many researchers. In a variety of experimental models, the extravasated Evans blue dye has been identified macroscopically within brain tissue (24–30), observed by red autofluorescence in brain tissue sections by fluorescence microscopy (31–33), or quantified by spectrophotometry or spectrophotofluorometry in homogenized brain tissue samples (26, 27, 34–37). In conclusion, Evans blue provides a rapid, reliable, and highly sensitive assessment of global BBB permeability. Furthermore, it also enables microscopic visualization as well as gross macroscopic observation of discrete areas of BBB disruption. 1.2. Sodium Fluorescein
Sodium fluorescein, which was introduced as a small molecular weight tracer has been widely used in a variety of model systems for the evaluation of BBB permeability (18, 38–43). Sodium fluorescein is of very small size, freely diffusible, nontoxic, and detectable at very low concentrations. The tissue distribution of sodium fluorescein is highly dependent on plasma concentration due to its low affinity for nonspecific protein binding (22). Owing to its small molecular weight, sodium fluorescein may cross the BBB much more readily than larger molecular weight tracers. Despite being inexpensive and nonradioactive, an important disadvantage of this tracer is that it cannot be detected by visual examination well. Spectrophotofluorimetric sodium fluorescein uptake measurements (excitation at 440 nm and emission at 525 nm) may enable detection of more subtle alterations in BBB permeability when compared to the use of radioactive tracers. It is suggested that changes in BBB permeability to sodium fluorescein may be the earliest and the most sensitive indicator of BBB disruption and therefore in many instances, it can be the tracer of choice or at least a useful complement to existing tracers.
1.3. Horseradish Peroxidase
Horseradish peroxidase as a tracer has been used for years in morphological studies of vascular permeability since 1960s (44). Several types of horseradish peroxidase (such as types II, IV, and VI) are commercially available for BBB permeability studies. Horseradish peroxidase reacts with the 3-3¢-diaminobenzidine to produce a brown reaction product which can be easily visualized macroscopically (Fig. 4) and by electron microscopy (Fig. 5). For years, the nature and location of the BBB were in debate, and the true significance of the capillary endothelium was brought to light in 1967 through the use of intravenously injected horseradish peroxidase (6, 44). Restriction of this tracer from the brain tissue was attributed to two unique characteristics of endothelial cells of cerebral microvessels; the presence of tight junctions, and the paucity of vesicular transport. Under a number of experimental conditions, BBB becomes permeable to horseradish peroxidase which provides morphological evidence for a barrier opening (45–53).
25 Assessment of Permeability in Barrier Type of Endothelium in Brain Using Tracers…
375
Fig. 4. Macrophotography of a coronal section of brain from an animal which was subjected to hyperbaric oxygen therapy and received a single intravascular injection of horseradish peroxidase 20 min prior to sacrification. The reaction product of horseradish peroxidase is seen to be localized to cortical and subcortical regions.
Fig. 5. Electron micrograph of a capillary from cerebral cortex of an animal subjected to cecal ligation and puncture and injected with a single dose of horseradish peroxidase 20 min prior to perfusion-fixation. Pinocytotic vesicles in the cytoplasm of endothelial cells contain horseradish peroxidase reaction product (arrows).
Owing to the electron-dense nature of the reaction product of the tracer, horseradish peroxidase not only provides ultrastructural evidence for a barrier, but also may offer an indication of the route of tracer extravasation which can be detected in brain sections under electron microscopy.
376
M. Kaya and B. Ahishali
2. Materials 2.1. Evans Blue
1. Evans blue (2%; Sigma Chem. Co., St Louis, MO) is dissolved in physiological saline solution, filtered through a filter paper, and kept at room temperature. 2. Sodium pentobarbital. 3. 60% trichloroacetic acid. 4. 0.1 M phosphate-buffered saline (PBS).
2.2. Sodium Fluorescein
1. Sodium fluorescein (2%; Sigma Chem. Co., St Louis, MO) is dissolved in physiological saline solution and kept at room temperature. 2. Sodium pentobarbital. 3. 60% trichloroacetic acid. 4. 0.1 M PBS.
2.3. Horseradish Peroxidase
1. Horseradish peroxidase (Type II; Sigma Chem. Co., St Louis, MO) is dissolved in physiological saline solution and kept at room temperature. 2. Sodium pentobarbital. 3. 3,3¢-Diaminobenzidine (DAB; Sigma Chem. Co., St Louis, MO) is dissolved in 250 mg/ml of distilled water and stored in single use aliquots of 100 ml at −80ºC. 4. 35% hydrogen peroxide. 5. 0.05 M Tris-buffer (pH 7.6). 6. Fixative solution: Paraformaldehyde (2%) is dissolved in 0.1 M phosphate buffer (pH 7.4) by stirring on a heating plate at 56ºC until the solution becomes colorless. After cooling to 37ºC, glutaraldehyde is added to a final concentration of 2.5%. The fixative is then filtered through a filter paper. 7. Epoxy Embedding Medium Kit (Sigma Chem. Co., St Louis, MO). 8. Osmium tetroxide.
3. Methods 3.1. Evans Blue
1. Catheterization: A PE-10 or 50 polyethylene catheter is inserted into the right femoral vein of anesthetized animal (we prefer anesthesia with sodium pentobarbital; 35 mg/ kg, i.p).
25 Assessment of Permeability in Barrier Type of Endothelium in Brain Using Tracers…
377
2. Evans Blue Delivery: Evans blue is administered at a dose of 4 ml per kg body weight of animal through the femoral vein (see Note 1). In our studies, Evans blue is allowed to circulate for 30 min before rats are perfused. 3. Perfusion: A lethal dose of sodium pentobarbital (over 100 mg/kg, i.p) is administered and the thoracic cavity is opened. The rats are perfused transcardially through the left ventricle with 100–150 ml of physiological saline solution at a pressure of about 110 mmHg (see Note 2). A concomitant small cut is made in the right atrium immediately to remove intravascular blood and tracer and the perfusion is continued for about 15 min until the fluid from the right atrium becomes colorless. 4. Following the perfusion, the animals are decapitated and brains are removed. 5. Meninges and choroid plexuses are removed and tissue samples are dissected from brain regions of interest. 6. Each brain region is weighed for quantitative measurement of Evans blue extravasation. 7. Brain regions are homogenized in 2 ml of PBS and then mixed with a vortex for 2 min after the addition of 2 ml of 60% trichloroacetic acid at 4°C to precipitate proteins. 8. Homogenized samples are kept in cold room (4°C) for 30 min and centrifuged at 18,000 g at 4°C for 10 min. 9. 250 ml of supernatants are put into microplate containers and the concentration of Evans blue in the supernatant is analyzed at a wavelength of 620 nm spectrophotometrically using a microplate reader (DTX 880, Multimode Detector). 10. The tissue content of Evans blue is quantified from a linear standard curve derived from the dye and expressed in mg per mg of brain tissue by a software application (Multimode Analysis Software) of microplate reader. 3.2. Sodium Fluorescein
1. Catheterization: A PE-10 or 50 polyethylene catheter is inserted into the right femoral vein of anesthetized animal (we prefer anesthesia with sodium pentobarbital; 35 mg/kg, i.p). 2. Sodium Fluorescein Delivery: Sodium fluorescein is administered at a dose of 5 ml per kg body weight of animal through the femoral vein (see Note 1). In our studies, sodium fluorescein is allowed to circulate for 30 min before rats are perfused. 3. Perfusion: A lethal dose of sodium pentobarbital (over 100 mg/ kg, i.p) is administered and the thoracic cavity is opened. The rats are perfused transcardially through the left ventricle with 100–150 ml of physiological saline solution at a pressure of
378
M. Kaya and B. Ahishali
about 110 mmHg (see Note 2). A concomitant small cut is made in the right atrium immediately to remove intravascular blood and tracer, and the perfusion is continued for about 15 min until the fluid from the right atrium becomes colorless. 4. Following the perfusion, the animals are decapitated and brains are removed. 5. Meninges and choroid plexuses are removed and tissue samples are dissected from brain regions of interest. 6. Each brain region is weighed for quantitative measurement of sodium fluorescein extravasation. 7. Brain regions are homogenized in 2 ml of PBS and then mixed with a vortex for 2 min after the addition of 2 ml of 60% trichloroacetic acid at 4°C to precipitate proteins. 8. Homogenized samples are kept in cold room (4°C) for 30 min and centrifuged at 18,000 g at 4°C for 10 min. 9. 250 ml of supernatants are put into microplate containers and the concentration of sodium fluorescein in the supernatant is analyzed at an excitation wavelength of 440 nm and emission wavelength of 525 nm spectrophotofluorometrically using a microplate reader (DTX 880, Multimode Detector). 10. The tissue content of sodium fluorescein is quantified from a linear standard curve derived from the dye and expressed in ng per mg of brain tissue by a software application (Multimode Analysis Software) of microplate reader. 3.3. Horseradish Peroxidase
1. Catheterization: A PE-10 or 50 polyethylene catheter is inserted into the right femoral vein of anesthetized animal (we prefer anesthesia with sodium pentobarbital; 35 mg/kg, i.p). 2. Horseradish Peroxidase Delivery: Horseradish peroxidase is administered at a dose of 0.2 mg/g body weight of animal through the femoral vein (see Notes 1 and 3). In our studies, horseradish peroxidase is allowed to circulate for 20 min before rats are perfused. 3. Perfusion-fixation: A lethal dose of sodium pentobarbital (over 100 mg/kg, i.p) is administered and the thoracic cavity is opened. The rats are perfused with physiological saline solution (approximately 75 ml) through the left ventricle to flush out intravascular blood by means of a small concomitant cut in the right atrium. The transcardial perfusion is continued until the fluid from the right atrium becomes colorless and followed with 200 ml of fixative solution (2% paraformaldehyde and 2.5% glutaraldehyde in 0.1 M phosphate buffer) for about 15 min (see Note 4). 4. Following the perfusion, the animals are decapitated and brains are removed.
25 Assessment of Permeability in Barrier Type of Endothelium in Brain Using Tracers…
379
5. Vibratome sectioning (Vibratome 3000 plus, St Louis, MO, USA): A coronal cut is made to the rostral part of the brain at a level close to cerebellum. The brain is then mounted by sticking the rostral face on one of the surfaces of the specimen-mounting block using an adhesive preferably with a cyanoacrylate base and the block is clamped into the vise jaws. The bath of the Vibratome is filled up with approximately 500 ml of 0.05 M Tris-buffer. After the initial gross trimming, sectioning is initiated preferably by the serial cut option at a thickness of 50 mm. 6. Section Processing for Horseradish Peroxidase Visualization: The brain sections are carried into a Petri dish containing 45 ml of 0.05 M Tris-buffer, and 100 ml DAB solution and 5 ml of freshly prepared 0.1% hydrogen peroxide are added (see Note 5). Following incubation for about 5–30 min at room temperature, the sections are washed for 30 min in two changes of 0.05 M Tris-buffer. On tissue sections, brown colored areas containing horseradish peroxidase-reaction products depict regions of BBB damage. Small specimens of approximately 1 mm2 are cut from these areas of interest for electron microscopic processing. 7. Processing for Transmission Electron Microscopy: The specimens are postfixed in 1% osmium tetroxide in 0.1 M phosphate buffer for 1 h at 4°C (see Note 4). After washing with distilled water thrice (5 min each), the specimens are dehydrated through a series of graded ethanol from 70 to 100%. After dehydration, the specimens are immersed in two changes of propylene oxide (10 min each) and infiltrated with 1:1 followed by 1:3 mixtures of propylene oxide and epoxy embedding medium (1 h each). The specimens are then transferred to a fresh change of 100% epoxy embedding medium for at least 1 h. The specimens are placed into gelatin capsules filled with resin and kept in an oven at 60°C for polymerization for at least 18 h. Ultrathin sections are cut at 500 Å thickness using an ultramicrotome. The sections are carried on a clean copper grid and evaluated under a transmission electron microscope (see Note 6).
4. Notes 1. Injection of Evans blue, sodium fluorescein, and horseradish peroxidase tracers should be given slowly over 1 min to avoid a sudden increase in vascular blood pressure. Care should be taken to avoid plugging of the catheter or sending air bubbles.
380
M. Kaya and B. Ahishali
2. During perfusion adequate pressure of the physiological saline solution is necessary to ensure thorough perfusion of the brain parenchyma. 3. The enzymatic activity of horseradish peroxidase decays over time (check the expiration date). 4. Paraformaldehyde, glutaraldehyde, osmium tetroxide, and propylene oxide are toxic and the solutions are extremely volatile. Always wear gloves and work in the chemical fume hood. 5. DAB is a potentially carcinogenic substance and requires caution in handling. At the end of the experiment DAB solution should be destroyed by 5% sodium hypochlorite. 6. During the electron microscopic procedures, en bloc staining of the specimens with uranyl acetate and staining of the ultrathin sections on grids with uranyl acetate and lead citrate are omitted in an attempt to avoid misinterpretation of the artifactual precipitates as HRP reaction products. References 1. Neuwelt, E. A., Abbott, N. J., Drewes, L., Smith, Q.R., Couraud, P. O., Chiocca, E. A., Audus, K. L., Greig, N. H., Doolittle, N. D. (1999) Cerebrovascular Biology and the various neural barriers: challenges and future directions. Neurosurgery 44, 604–608. 2. Gloor, S. M., Wachtel, M., Bolliger, M.F., Ishihara, H., Landmann, R., Frei. K. (2001) Molecular and cellular permeability control at the blood-brain barrier. Brain Res. Brain Res. Rev. 36, 258–264. 3. Abbott, N. J., Rönnbäck, L., Hansson, E. (2006) Astrocyte-endothelial interactions at the blood-brain barrier. Nat. Rev. Neurosci. 7, 1–53. 4. Correale, J., Villa, A. (2009) Cellular elements of the blood-brain barrier. Neurochem. Res. 34, 2067–2077. 5. Weiss, N., Miller, F., Cazaubon, S., Couraud, P. O. (2009) The blood-brain barrier in brain homeostasis and neurological diseases. Biochim. Biophys. Acta. 1788, 842–857. 6. Brightman, M. W., Reese, T. S. (1969) Junctions between intimately apposed cell membranes in the vertebrate brain. J. Cell Biol. 40, 648–677. 7. Begley, D. J., Brightman, M. W. (2003) Structural and functional aspects of the blood– brain barrier. Progr. Drug Res. 61, 39–78. 8. Abbott, N. J., Patabendige, A. A., Dolman, D. E., Yusof, S. R., Begley, D. J. (2010) Structure and function of the blood-brain barrier. Neurobiol. Dis. 37, 13–25.
9. Rubin, L. L., Staddon, J. M. (1999) The cell biology of the blood-brain barrier. Annul. Rev. Neurosci. 22, 11–28. 10. Ek, C. J., Habgood, M. D., Dziegielewska, K. M., Potter, A., Saunders, N. R. (2001) Permeability and route of entry for lipid-insoluble molecules across brain barriers in developing Monodelphis domestica. J. Physiol. 536, 841–853. 11. Tuma, P. L., Hubbard, A. L. (2003) Transcytosis: crossing cellular barriers. Physiol. Rev. 83, 871–932. 12. Rosenberg, G. A., Yang, Y. (2007) Vasogenic edema due to tight junction disruption by matrix metalloproteinases in cerebral ischemia. Neurosurg. Focus 22, E4. 13. Strbian, D., Durukan, A., Pitkonen, M., Marinkovic, I., Tatlisumak, E., Pedrono, E., Abo-Ramadan, U., Tatlisumak, T. (2008) The blood-brain barrier is continuously open for several weeks following transient focal cerebral ischemia. Neuroscience 153, 175–181. 14. Hardebo, J. E., Johansson, B. B. (1980) Effect of an anion transport inhibitor on blood-brain barrier lesions during acute hypertension. Possible prevention of transendothelial vesicular transport. Acta Neuropathol. 51, 33–38. 15. Kaya, M., Kalayci, R., Küçük, M., Arican, N., Elmas, I., Kudat, H., Korkut, F. (2003) Effect of losartan on the blood-brain barrier permeability in diabetic hypertensive rats. Life Sci. 73, 3235–3244.
25 Assessment of Permeability in Barrier Type of Endothelium in Brain Using Tracers… 16. Dietrich, W.D., Alonso, O., Halley, M. (1994) Early microvascular and neuronal consequences of traumatic brain injury: a light and electron microscopic study in rats. J. Neurotrauma 11, 289–301. 17. Beaumont, A., Marmarou, A., Hayasaki, K., Barzo, P., Fatouros, P., Corwin, F., Marmarou, C., Dunbar, J. (2000) The permissive nature of blood brain barrier (BBB) opening in edema formation following traumatic brain injury. Acta Neurochir. Suppl. 76, 125–129. 18. Gurses, C., Ekizoglu, O., Orhan, N., Ustek, D., Arican, N., Ahishali, B., Elmas, I., Kucuk, M., Bilgic, B., Kemikler, G., Kalayci, R., Karadeniz, A., Kaya, M. (2009) Levetiracetam decreases the seizure activity and blood-brain barrier permeability in pentylenetetrazolekindled rats with cortical dysplasia. Brain Res. 1281, 71–83. 19. Ndode-Ekane, X. E., Hayward, N., Gröhn, O., Pitkänen, A. (2010) Vascular changes in epilepsy: Functional consequences and association with network plasticity in pilocarpineinduced experimental epilepsy. Neuroscience 166, 312–332. 20. d’Avella, D., Cicciarello, R., Angileri, F. F., Lucerna, S., La Torre, D., Tomasello, F. (1998) Radiation-induced blood-brain barrier changes: pathophysiological mechanisms and clinical implications. Acta Neurochir. Suppl. 71, 282–294. 21. Kaya, M., Palanduz, A., Kalayci, R., Kemikler, G., Simsek, G., Bilgic, B., Ahishali, B., Arican, N., Kocyildiz, Z. C., Elmas, I., Kucuk, M., Karadeniz, A. (2004) Effects of Lipopolysaccharide on the Radiation-Induced Changes in the Blood-Brain Barrier and the Astrocytes. Brain Res. 1019, 105–112. 22. Wolman, M., Klatzo, I. and Chui, E. (1981) Evaluation of the dye-protein tracer in pathophysiology of the blood-brain barrier. Acta Neuropathol. (Berl.) 54, 55–61. 23. Clasen, R. A., Pandolfi, S., Hass, G. M. (1970) Vital staining, serum albumin and the bloodbrain barrier. J. Neuropathol. Exp. Neurol. 29, 266–284. 24. Zuccarello, M., Anderson, D. K. (1989) Protective effect of a 21-aminosteroid on the blood-brain barrier following subarachnoid hemorrhage in rats. Stroke 20, 367–371. 25. Kaya, M., Küçük, M., Bulut, R. K., Palandüz, Ş. (1999) Acute hyperglycemia augments blood-brain barrier damage in experimental status epilepticus. Neurosci. Res. Commun. 25, 111–119. 26. Kaya, M., Küçük, M., Kalaycı, R., Çimen, V., Gürses, C., Elmas, İ., Arıcan, N. (2001) Magnesium sullfate attenuates increased blood-brain barrier permeability during insu-
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
381
lin-induced hypoglycemia in rats. Can. J. Physiol. and Pharmacol. 79, 793–798. Kaya, M., Kalaycı, R., Arıcan, N., Küçük, M., Elmas, I. (2003-a) Effect of aluminum on the blood-brain barrier permeability during nitric oxide-blockade-induced chronic hypertension in rats. Biol. Trace Element Res. 92, 221–230. Kaya, M., Gulturk, S., Elmas, I., Kalayci, R., Arican, N., Kocyildiz, Z.C., Kucuk, M., Yorulmaz, H., Sivas, A. (2004-a) The effects of magnesium sulfate on blood-brain barrier disruption caused by intracarotid injection of hyperosmolar mannitol in rats. Life Sci. 76, 201–212. Esen, F., Erdem, T., Aktan, D., Orhan, M., Kaya, M., Eraksoy, H., Cakar, N., Telci, L. (2005) Effect of magnesium sulfate administration on blood–brain barrier in a rat model of intraperitoneal sepsis: a randomized controlled experimental study. Crit. Care 9, R18–23. Kalayci R, Kaya M, Ahishali B, Arican N, Elmas I, Kucuk M. Long-term L-NAME treatment potentiates the blood-brain barrier disruption during pentylenetetrazole-induced seizures in rats. Life Sci. 79(1):16–20, 2006. Lo, W. D., McNeely, D. L., Boesel, C. W. (1991) Blood-brain barrier permeability in an experimental model of bacterial cerebritis. Neurosurgery 29, 888–892. Murakami, K., Kawase, M., Kondo, T., Chan, P. H. (1998) Cellular accumulation of extravasated serum protein and DNA fragmentation following vasogenic edema. J. Neurotrauma 15, 825–835. Akiyama, H., Kondoh, T., Kokunai, T., Nagashima, T., Saito, N., Tamaki, N. (2000) Blood-brain barrier formation of grafted human umbilical vein endothelial cells in athymic mouse brain. Brain Res. 858, 172–176. Beach, V. L., Steinetz, B. G. 1961. Quantitative measurement of Evans blue space in the tissues of the rat: influence of 5-hydroxytryptamine antagonists and phenelzine on experimental inflammation. J. Pharmacol. Exp. Ther. 131, 400–406. Kajiwara, K., Ito, H., Fukumoto, T. (1990) Lymphocyte infiltration into normal rat brain following hyperosmotic blood-brain barrier opening. J. Neuroimmunol. 27, 133–140. Mikawa, S., Kinouchi, H., Kamii, H., Gobbel, G. T., Chen, S. F., Carlson, E., Epstein, C. J., Chan, P. H. (1996) Attenuation of acute and chronic damage following traumatic brain injury in copper, zinc-superoxide dismutase transgenic mice. J. Neurosurg. 85, 885–891. Saria, A., Lundberg, J. M. (1983) Evans blue fluorescence: quantitative and morphological
382
38. 39.
40.
41.
42.
43.
44.
45.
M. Kaya and B. Ahishali evaluation of vascular permeability in animal tissues. J. Neurosci. Methods 8, 41–49. Hatashita, S., Hoff, J. T. (1990) Brain edema and cerebrovascular permeability during cerebral ischemia in rats. Stroke 21, 582–288. Gupta, A., Agarwal, R., Shukla, G. S. (1999) Functional impairment of blood-brain barrier following pesticide exposure during early development in rats. Hum. Exp. Toxicol. 18, 174–179. Hooper, D. C., Kean R. B., Scott, G. S., Spitsin, S. V., Mikheeva, T., Morimoto, K., Bette, M., Röhrenbeck, A. M., Dietzschold, B., Weihe, E. (2001) The Central Nervous System Inflammatory Response to Neurotropic Virus Infection Is Peroxynitrite Dependent. The Journal of Immunol. 167, 3470–3477. Erdlenbruch, B., Alipour, M., Fricker, G., Miller, D. S., Kugler, W., Eibl, H., Lakomek, M. (2003) Alkylglycerol opening of the blood-brain barrier to small and large fluorescence markers in normal and C6 glioma-bearing rats and isolated rat brain capillaries. Br. J. Pharmacol. 140, 1201–1210. Natah, S. S., Mouihate, A., Pittman, Q. J., Sharkey, K. A. (2005) Disruption of the blood-brain barrier during TNBS colitis. Neurogastroenterol. Motil. 17, 433–46. Kaya, M., Gurses, C., Kalayci, R., Ekizoglu, O., Ahishali, B., Orhan, N., Oku, B., Arican, N., Ustek, D., Bilgic, B., Elmas, I., Kucuk, M., Kemikler, G. (2008) Morphological and functional changes of blood-brain barrier in kindled rats with cortical dysplasia. Brain Res. 1208, 181–191. Reese, T. S., Karnovsky, M. J. (1967) Fine structural localization of a blood-brain barrier to exogenous peroxidase. J. Cell Biol. 34, 207–217. Bolwig, T. G., Hertz, M. M., Westergaard, E. (1977) Acute hypertension causing bloodbrain barrier breakdown during epileptic seizures. Acta Neurol. Scand. 56, 335–342.
46. Sheikov, N., McDannold, N., Sharma, S., Hynynen, K. (2008) Effect of focused ultrasound applied with an ultrasound contrast agent on the tight junctional integrity of the brain microvascular endothelium. Ultrasound Med. Biol. 34, 1093–1104. 47. Westergaard, E., Brightman, M. W. (1973) Transport of protein across normal cerebral arteriols. J. Comp. Neurol. 152, 17–44. 48. Reyners, H., de Reyners, E. G., Jadin, J. M., Maisin, J. R. (1975) An ultrastructural quantitative method for the evaluation of the permeability to horseradish peroxidase of cerebral cortex endothelial cells of the rat. Cell Tissue Res. 157, 93–99. 49. Broadwell, R. D., Charlton, H. M., Balin, B. J., Salcman, M. (1987) Angioarchitecture of the CNS, pituitary gland, and intracerebral grafts revealed with peroxidase cytochemistry. J. Comp. Neurol. 260, 47–62. 50. Villegas, J. C., Broadwell, R.D. (1993) Transcytosis of protein through the mammalian cerebral epithelium and endothelium II. Adsorptive trancytosis of WGA-HRP and the blood-brain and brain-blood barriers. J. Neurocytol. 22, 67–80. 51. Nag, S. (1998) Blood-brain barrier permeability measured with histochemisty. Ed. Pardridge WM. Introduction to the bloodbrain barrier-Methodology, biology and pathology. pp. 113–121, Cambridge University Press. 52. Nag, S. (2003) Pathophysiology of bloodbrain barrier breakdown. Ed: Nag, S. The Blood-brain barrier-Biology and research protocols – Method in molecular medicine. Chapter 6, pp: 97–119, Humana press. 53. Nag, S. (2003a) Blood-brain barrier permeability using tracers and immunohistochemistry. Ed: Nag, S. The Blood-brain barrier-Biology and research protocols – Method in molecular medicine. Chapter 8, pp: 133–144, Humana press.
Chapter 26 In Vitro and In Vivo Methods for Assessing FcRn-Mediated Reverse Transcytosis Across the Blood–Brain Barrier Nadia Caram-Salas, Eve Boileau, Graham K. Farrington, Ellen Garber, Eric Brunette, Abedelnasser Abulrob, and Danica Stanimirovic Abstract The neonatal Fc receptor, FcRn, mediates endocytic recycling pathway that prevents degradation of IgG and is expressed in most endothelial cells. The blood–brain barrier (BBB), formed by brain endothelial cells sealed with tight junctions, restricts transport of IgG from the blood to the brain. In contrast, it has been suggested that IgG undergoes efflux from the brain parenchyma via reverse transcytosis across the BBB mediated by FcRn. The fast elimination of therapeutic antibodies from the brain via this route may limit their therapeutic potency. In vitro and in vivo methods described in this chapter were developed to facilitate research into mechanisms and dynamics of brain efflux of compounds, including FcRn-mediated reverse transcytosis across the BBB. The in vitro model uses immortalized adult rat brain endothelial cells which express high levels of FcRn. In vivo models use Prospective optical imaging to measure the clearance rate of intracerebrally injected FcRn-transported molecules tagged with near-infrared fluorescent probes. Key words: FcRn, Blood–brain barrier, Reverse transcytosis, Optical imaging
1. Introduction The blood–brain barrier (BBB) is formed by specialized endothelial cells lining the extensive brain capillary network. These endothelial cells function as a restrictive gate to control the composition of extracellular fluid in the central nervous system (CNS) selectively restricting and/or controlling the access of bloodborne molecules to the brain (1). The brain capillary endothelium exhibits unique anatomical and biochemical features, including elaborate tight junctions that form a physical barrier to the majority of hydrophilic molecules larger than 500 Da, and the
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_26, © Springer Science+Business Media, LLC 2011
383
384
N. Caram-Salas et al.
polarized expression of transporters that control transport, both influx and efflux, of molecular substrates across the BBB (2). While blood-borne antibodies are almost completely restricted from the brain by the BBB, antibodies targeted against specific receptors expressed by brain endothelium, including transferrin receptor and insulin growth factor receptor, have been shown to penetrate the BBB into the brain via a receptor-mediated transcytosis, and could be exploited as carriers to systemically deliver therapeutics including therapeutic antibodies into the brain (2, 3). To exploit the potential of therapeutic antibodies being developed for diseases of the brain, a better understanding of their pharmacodynamics in the brain has become of critical importance. The work by Pardridge and Zhang (2001) (4) suggested that radioactively labeled antibodies (IgG) injected into the brain parenchyma in rats have residence half-life of approximately 45 min. A predicted normal width of the extracellular spaces ranges between 38 and 64 nm (5), whereas the width of perivascular and periaxonal spaces is estimated at 80–150 nm (5, 6), suggesting that these spaces are permissive for diffusion of large molecules, including antibodies. The body of prior work, summarized in a recent review (7) suggests that intracerebrally injected radiolabeled tracers of different molecular weights spanning a fivefold range in diffusion coefficients, clear at similar rates (with half residence times of 6–16 h), consistent with bulk (convective) flow of the interstitial fluid (ISF) along perivascular spaces and axonal tracts, rather than diffusion. The short half-life of brain injected antibodies was therefore attributed to active efflux (reverse transcytosis) across the BBB, presumably mediated by the neonatal Fc receptor (FcRn), which, similar to the non-BBB capillaries, shows high expression in brain endothelial cells (8). FcRn is responsible for the transfer of passive humoral immunity from mother to fetus in humans (9). In adult rodents and humans FcRn is expressed in varied cell types, including polarized epithelia and endothelia, and is responsible for bidirectional IgG translocation, from either the lumen or tissue spaces to the opposite pole of the cell (10, 11). Endothelial cells lining the capillaries express FcRn and are involved in IgG recycling (12); this mechanism serves to protect circulating IgG from degradation extending the beta-phase half-life of antibodies by several fold. The recycling path involves passive endosomal pinocytotic capture of sera, followed by acidification to pH 6.0–6.5 in the sorting endosome where FcRn binds IgG in a pH-dependent manner; FcRn bound IgG is then directed to the cellular surface for release thereby preventing IgG degradation (10, 13, 14). FcRn is expressed at low levels in the plasma membrane because of its rapid recycling after incomplete fusion with the plasma membrane, known as “kiss and run” (13). However, the potential role for FcRn in brain efflux of antibodies across the BBB remains
26 In Vitro and In Vivo Methods for Assessing FcRn-Mediated Reverse…
385
obscure. A recent study (15) comparing plasma to brain ratio of systemically injected radiolabeled antibodies in FcRn knockout and wild-type animals concluded that lack of FcRn-mediated antibody recycling did not affect their brain accumulation. Other studies demonstrated slower brain clearance of amyloid-betabinding antibodies in FcRn knockout animals (16) but suggested that yet unknown accessory IgG receptor at the abluminal BBB membrane likely facilitates IgG endocytosis before its FcRnmediated migration to the exocytotic site at the luminal BBB membrane. To enable better understanding of the role FcRn plays in antibody trafficking across the BBB, we have developed in vitro and in vivo models suitable for studying antibody efflux from the brain. The in vitro model uses immortalized adult rat brain endothelial cells (SV-ARBEC) developed in our laboratory (17) and shown to highly express FcRn. The in vivo model is based on monitoring the disappearance of brain-injected fluorescently labeled compounds using prospective noninvasive optical imaging.
2. Materials 2.1. In Vitro BBB Model (See Note 1)
1. SV-ARBEC: immortalized adult rat brain microvascular endothelial cells (see Note 2). 2. SV-RAS: immortalized neonatal rat astrocytes (see Note 3). 3. SV-ARBEC growth media (per 500 ml of media). (a) M199 media, phenol red free (Gibco/BRL Canada, Burlington, ON), 370 ml. (b) Peptone (Sigma Aldrich Canada, Oakville, ON); 250 mg dissolved in 50 ml of M199. (c) d-Glucose 45% (Sigma Aldrich Canada, Oakville, ON); 10 ml. (d) BME amino acids 50× (Sigma Aldrich Canada, Oakville, ON); 10 ml. (e) BME vitamins 100× (Sigma Aldrich Canada, Oakville, ON); 5 ml. (f ) Fetal bovine serum (FBS) heat inactivated at 56°C for 35 min (Multicell Premium Wisent Inc. Canada, St-Bruno, QC); 50 ml. (g) Antibiotic/antimytotic (100×) (Sigma Aldrich Canada, Oakville, ON); 5 ml. 4. SV-RAS growth media. (h) DMEM-high glucose, no phenol red (Gibco/BRL Canada, Burlington, ON).
386
N. Caram-Salas et al.
(i) FBS 10%, heat inactivated at 56°C for 35 min (Multicell Premium Wisent Inc. Canada, St-Bruno, QC). (j) Antibiotic/antimytotic (100×) (Sigma Aldrich Canada, Oakville, ON); 5 ml. 5. SV-RAS-conditioned media. (k) DMEM-high glucose, no phenol red (Gibco/BRL Canada, Burlington, ON). (l) FBS 1%, heat inactivated at 56°C for 35 min (Multicell Premium Wisent Inc. Canada, St-Bruno, QC). (m) Incubate SV-RAS in above media for 72 h. (n) Collect media and filter through 45 mm filter to remove cellular debris. (o) Store in aliquots at −20°C. 6. Cell culture T75 flasks (BD Falcon Canada, Mississauga, ON). 7. Rat tail collagen type 1 (Collaborative Biomedical products Canada, Mississauga, ON). 8. Falcon tubes, 15 ml (BD Falcon Canada, Mississauga, ON). 9. 0.83 cm2 Falcon cell inserts 1 mm pore size (BD Falcon Canada, Mississauga, ON). 10. 12-Well tissue culture plates (BD Falcon Canada, Mississauga, ON). 11. BCA kit (Thermo Fisher Scientific Canada, Ottawa, ON) for protein content determination. 12. 0.25% Trypsin (Gibco/BRL Canada, Burlington, ON). 13. Hank’s buffered saline solution phenol red free (Multicell Premium Wisent Inc. Canada, St-Bruno, QC). 14. Bovine serum albumin (BSA), cell culture tested (Sigma Aldrich Canada, Oakville, ON). 2.2. Quantification of IgG Using EnzymeLinked Immunosorbent Assay
1. PBS 10 mM 10× (NaCl 137 mM, KCl 2.7 mM, Na2HPO4 100 mM, KH2PO4 2 mM; Sigma Aldrich Canada, Oakville, ON). Dilute to 1× with dH2O before use. 2. Wash buffer (0.05% Tween-20 in 10 mM PBS, pH 7.0). 3. Blocking buffer (1% casein acid hydrolysate in 10 mM PBS pH 7.0; Sigma Aldrich Canada, Oakville, ON). 4. Sample diluent (10 mM PBS, 0.05% Tween-20, 0.1% casein, 5% FBS, 0.1% BSA, pH 7.0). 5. MaxiSorp Immunoplate (NuncTM, VWR International, Ltd. Canada, Mississauga, ON). 6. Coating antibody: Immunoglobulin G (Fc)-specific monoclonal antibody, clone M6102721 (Fitzgerald Industries International USA, Acton, MA).
26 In Vitro and In Vivo Methods for Assessing FcRn-Mediated Reverse…
387
7. HRP-conjugated donkey anti-human Fc fragment antibody (Jackson ImmunoResearch, Medicorp Inc. Canada, Montreal, QC). 8. 3,3¢,5,5¢-Tetramethylbenzidine substrate reagent (TMB substrate; R&D systems USA, Minneapolis, MN). 9. Sulfuric acid 1 N (JT-Backer Canada, Mississauga, ON). 10. Plate reader SpectraMax340 (Molecular Devices, Sunnyvale, CA, USA). 11. Software SoftMaxPro installed on the plate reader. 2.3. Animals
1. Male Wistar rats, 4–6 weeks old (Charles River Laboratories, Wilmington, MA) (see Note 4).
2.4. Stereotactic and Surgery Equipment
1. Lab standard stereotaxic frame (Harvard Apparatus Canada, Saint-Laurent QC). 2. Rat adaptor for stereotactic apparatus (Harvard Apparatus Canada, Saint-Laurent QC). 3. Gas anesthesia mask adapter for rats (Harvard Apparatus Canada, Saint-Laurent QC). 4. Harvard Apparatus model 11Plus microsyringe digital pump (Harvard Apparatus, Canada, Saint-Laurent QC) with 10 ml Hamilton microsyringe connected by means of PE10 tubing with 31 gauge stainless steel injection cannulae. 5. Surgical instruments: Fine Science Tools (North Vancouver, BC, Canada) microdrill with dental drill bits, sterile scalpel blade, stitches 4.0, carbide jaws, scissors (Harvard Apparatus Canada, Saint-Laurent QC). 6. Microsyringes, 0.5 and 1 ml, 32 gauge (Stoelting USA, Kiel, WI). 7. 1 cm3 insulin syringe, 30 gauge (BD Canada, Mississauga, ON).
2.5. Imaging Equipment and Software
1. Time-domain eXplore Optix MX2 preclinical imager with 680 nm and/or 780 nm pulsed laser diode (Advanced Research Technologies, Montreal, QC) (see Note 5). 2. Acquisition Software (Optix MX2, Advance Research Technologies, ART, Montreal, QC, Canada). 3. Analysis Software (OptiView, Advance Research Technologies, ART, Montreal, QC, Canada).
2.6. Anesthesia and Analgesia
1. Isoflurane (Baxter Canada, Mississauga, ON). 2. O2/N2 balance gas tank (Praxair Canada, Ottawa, ON). 3. Local analgesic (0.5% Marcaine; Novocol Pharmaceutical of Canada, Inc. Cambridge, ON, Canada).
388
N. Caram-Salas et al.
2.7. Compound Labeling with NIRF Dyes
1. IgG or human Fc antibody fragment (Biogenidec, Cambridge, MA). 2. BSA (Sigma Aldrich Canada, Oakville, ON). 3. IR-800 dye (IRDye 800CW NH Ester Infrared Dye; LI-COR Biosciences USA, Nebraska). Absorption/emission: 778/789 nm. 4. Cy5.5-Dextran 10 kDa (Nanocs, New York, USA). 5. Methanol (Sigma Aldrich Canada, Oakville, ON). 6. Carbonate buffer pH 9.3 (0.6 mM NaHCO3 in dH2O, adjusted to pH 9.3 with 6N NaOH). 7. DMSO (Dimethyl sulfoxide) (Sigma Aldrich Canada, Oakville, ON). 8. 0.22 ml filters (Millipore, MA, USA). 9. Purification columns, 3–100 kDa cut off (Amicon Centricon Billerica, MA, USA).
3. Methods The methods in this chapter will describe how to (1) measure the rate of reverse transcytosis using in vitro BBB model and (2) of brain efflux of intracerebrally injected molecules using in vivo optical imaging. 3.1. Determining the Rate of Reverse Transcytosis Using In Vitro BBB Model 3.1.1. Setting Up In Vitro BBB Model
SV-ARBEC were derived according to a previously described method (18). 1. Coat T75 cell culture flasks with rat-tail collagen type 1 at a final concentration of 50 mg/ml in 10% acetic acid. 2. Wash culture flasks three times in Hanks Buffered Saline Solution (HBSS) and leave the last wash in flasks until cell seeding. 3. Seed and grow SV-ARBEC cells in rat-tail collagen type 1-coated T-75 cell culture flasks in 10 ml of SV-ARBEC growth medium (see Subheading 2.1, item 3). Divide cells 1:20 when they reach confluence (every 5–6 days). 4. To divide cells, wash cells twice with HBSS. Aspirate all HBSS, then add 0.25% trypsin (2 ml for each T-75 flask), incubate at 37°C for 2–3 min. 5. Add 8 ml of SV-ARBEC culture media to the flask; to divide 1:20, add 0.5 ml of cells from this flask to 9.5 ml of fresh SV-ARBEC culture media and seed into new flasks. 6. Coat separate cell culture flasks with poly-l-lysine (25 mg/ml) dissolved in distillated water (dH2O), and wash twice with HBSS prior to plating SV-RAS cell.
26 In Vitro and In Vivo Methods for Assessing FcRn-Mediated Reverse…
389
7. Grow SV-RAS cells in SV-RAS growth medium (see Subheading 2.1, item 4) and split 1:20 every week as described in step 4. Seed and grow cells as in step 5. Collect the astrocyte conditioned media (ACM) as described in Subheading 2.1, item 5. 3.1.2. Transport Experiments
1. Seed 80,000 SV-ARBEC cells in each rat-tail collagen-coated Falcon cell insert (pore size 1 mm; surface area 0.83 cm2) in 1 ml of phenol red free SV-ARBEC growth media (see Subheading 2.1, item 3). Immerse SV-ARBEC-seeded tissue culture inserts into wells of a 12-well plate, each containing 2 ml of 50:50 (v/v) ratio of SV-ARBEC growth media without phenol red and SV-RAS conditioned media (Subheading 2.1, item 5) (see Note 6). 2. Incubate inserts at 37°C in 5% of CO2 for 4–6 days without feeding. 3. At day 4, evaluate the integrity and the tightness of the cell monolayer by measuring transendothelial electrical resistance (TEER) or permeability for [14C sucrose] as described previously (18, 19) (see Note 7). 4. To wash cells, gently immerse inserts to be tested in three consecutive wells of a 12-well plate containing HBSS, leaving them in place for 10 s in each well. 5. Place inserts in a new 12-well plate containing 2 ml/well of prewarmed transport buffer (HBSS containing 0.05% BSA and filtered). 6. For luminal (apical) to abluminal (basolateral) transport (i.e., top to bottom), replace transport buffer from the top chamber (due to evaporation the volume is smaller than 500 ml) with 500 ml of transport buffer containing 30 mg/ml antibody (IgG) or hFc antibody fusion protein drop by drop into the top chamber. Apply the same protocol to tissue culture inserts without seeded cells, which are used as controls. 7. For abluminal (basolateral) to luminal (apical) transport (bottom to top), add 500 ml of transport buffer in the top chamber and 2 ml of transport buffer containing 30 mg/ml antibody (IgG) or hFc antibody fusion protein drop by drop into the bottom chamber. Apply the same protocol to tissue culture inserts without seeded cells, which are used as controls (see Note 8). 8. Transport studies in both directions are conducted at 37°C with plates positioned onto a rotating platform, stirring at 30–40 rpm. 9. Collect aliquots of 100 ml from the chamber opposite to the chamber containing antibody/antibody fusion protein at successive time points: 15 min, 1, 2, 3, and 4 h. After collecting
390
N. Caram-Salas et al.
each aliquot, replace the volume with 100 ml of the transport buffer. 10. Quantify the amount of the IgG or hFc antibody fragment in each collected aliquot using sandwich enzyme-linked immunosorbent assay (ELISA) (see Subheading 3.1.3). 3.1.3. Quantification of IgG Using Sandwich EnzymeLinked Immunosorbent Assay
1. Coat a 96-well plate (MaxiSorp) with 50 ml of monoclonal Immunoglobulin G (Fc) Specific, 0.5 mg/ml (Fitzgerald, MA, USA). 2. Incubate overnight at 4°C. 3. Flick the plate to remove the coating buffer and tap upside down on paper towel to empty all wells from unbound material by gentle tapping. Fill all wells with washing buffer, flick, and tap the plate face down on paper towel to empty wells of wash buffer; repeat three times for a total of four washes. 4. Block the plate with 250 ml of blocking buffer for 90 min at room temperature. 5. Prepare dilutions of IgG or hFc from 0 to 3,200 ng/ml in sample diluent for a standard curve. 6. Dilute the samples appropriately so that they “fit” onto linear portion of the standard curve. 7. Turn the plate upside down on a paper towel to remove blocking buffer from anti-Fc antibody-coated plates. 8. Transfer in duplicate 100 ml of standards and samples in each well and incubate for 120 min at room temperature; wash plate four times with washing buffer as described in step 3. 9. Add 100 ml of HRP-conjugated donkey anti-human-Fc antibody at 1:10,000 dilution for 90 min at room temperature; wash plate four times with washing buffer as described in step 3 (see Note 9). 10. Add 100 ml/well of tetramethylbenzidine (3,3¢,5,5¢tetramethylbenzidine, TMB; see Note 10) and develop color for 2–4 min. 11. Stop the reaction by adding 100 ml/well of 1N sulfuric acid. 12. Optical density of developed color is measured at 450 nm in a microtiter plate reader. 13. Plot the standard curve and from that curve calculate IgG (or hFc) concentrations in aliquots using SoftmaxPro software. 14. Calculate the clearance and the endothelial permeability coefficient (Pe) for IgG or hFc in either direction (top-to-bottom and bottom-to-top) using published methods/formulas (see Note 11).
26 In Vitro and In Vivo Methods for Assessing FcRn-Mediated Reverse…
3.2. Brain Efflux of Intracerebrally Injected Molecules Using In Vivo Optical Imaging
3.2.1. Labeling of Compounds with the NIRF Dye
391
Optical imaging using eXplore Optix time domain small animal imaging system allows a real-time, prospective, noninvasive in vivo tracking of the fluorescently labeled molecules injected systemically or locally (i.e., intracerebrally). This section describes methods used for assessing fluorescence intensity (FI) decay of intracerebrally injected compounds labeled with the NIRF probe(s) using eXplore Optix. Although this method can be used for assessing efflux/elimination of any passively or actively transported compound, the examples provided in this chapter focus on the molecules known to be recycled via FcRn, human Fc antibody fragment (MW 50 kDa), and BSA (MW 66 kDa) (see Note 12). 1. Reconstitute IR800 dye by dissolving 1 mg of the dye into 50 ml of DMSO (stock solution), pipet up and down to mix, flush air with nitrogen for 10 s and vortex for 20 s (see Note 13). 2. Aliquot the stock solution in 10 ml in Eppendorf tubes; flush the tubes with air/nitrogen for 20 s. 3. Add 250 ml of 1 mg/ml IgG or hFc to the Eppendorf tube [250 mg of IgG or hFc per 10 ml of IR800]. 4. Add carbonate buffer pH 9.3 to make 10% of total volume (26 ml), vortex immediately and flush air with nitrogen. 5. Wrap the tube in aluminum foil and leave at room temperature for 120 min, and then at 4°C overnight. 6. Purify IR800-IgG/hFc by transferring the reaction to the Amicon Centricon column with 3 kDa cut-off size (to eliminate unbound dye). Top up the volume to 4 ml with PBS and centrifuge at 4,000 × g in a swinging bucket rotor at 4°C between 25 and 30 min. Repeat this washing step two to four times until the flow-through is completely clear of dye. 7. Collect the conjugated molecule. 8. Measure protein levels in collected sample using BCA kit. Follow instructions provided by the manufacturer, but use 2.5 ml of protein (instead of 10 ml). Calculate dye/protein ratio by measuring the absorbance at 280 nm (protein) and 778 nm (IR800) (see Note 14). 9. Aliquots could be stored at −80°C for 3–4 months.
3.2.2. Animal Preparation and Stereotaxic Injection of Labeled Compounds
1. Deeply anesthetize a male Wistar rat under isoflurane anesthesia (apparatus dispenses from 2.5 to 4.5 L/min O2/N2 balance gas with 2–3% isoflurane) until the respiratory pace slows down (usually few minutes).
392
N. Caram-Salas et al.
2. Reduce anesthesia to maintenance level from 1 to 1.5 L/min of O2/N2 balance and from 1.5 to 2% isoflurane and shave the head of the rat. 3. Prescan the head region where the injection will be made in eXplore Optix as described in Subheading 3.2.3 to obtain background fluorescence reading (this will be subtracted from all subsequent experimental readings). 4. Under isoflurane anesthesia, Place the rat in a stereotaxic frame and hold the nose with a nosebar set at 2.4 mm for head angle. 5. Immobilize the head symmetrically with ear bars. 6. Apply two drops of artificial tears in each eye to protect them from light. 7. Make an incision in the right area of the scalp with a scalpel blade No. 15. 8. Retract the scalp and clean the scull from remaining tissue and blood. 9. Make a hole in the scull using a drill screw 0.7 mm anterior to Bregma; 3.0 lateral to midline (caudate-putamen area, according to the rat atlas of Paxinos and Watson) (20). 10. Clean the skull of blood with a cotton swab and inject 5.15 ml of NIRF-labeled compound with a microsyringe connected to a pump at the 5.0-mm depth from the skull using an injection rate of 1 ml/min (see Note 15). 11. After the needle is removed from the injection site, bone wax to block the hole and prevent the back flow of the compound from the brain, then apply local analgesic and inject pain killer intramuscularly. 12. Clean the area with a cotton swab, and close the skin with three to four stitches. 13. Immediately after closing the skin, deeply anesthetize the rat under isoflurane anesthesia (2.5–4.5 L/min of O2/N2 balance gas, 2–3% isoflurane) and place it into the eXplore Optix imager. 3.2.3. Scanning Parameters for eXplore Optix Acquisition Software
1. Place the animal on the animal bed, ventral side facing down, in the eXplore Optix imager. Set the bed temperature at 36°C (see Note 16). 2. While in eXplore Optix, maintain the animal under anesthesia with 1.5–2.5 l/min of O2/N2 balance gas with 1.5–2.0% of isoflurane delivered through the fitted anesthesia mask (available with the imaging equipment). 3. Using the side viewing digital camera inside the machine, set the elevation of the bed such that the animal head is just below the green line limit.
26 In Vitro and In Vivo Methods for Assessing FcRn-Mediated Reverse…
393
4. Draw a region of interest (ROI) around the area of the head where the injection has been made using the polygon draw tool in the eXplore Optix acquisition software version 2.0 (see Note 5). 5. Set the step-size of the scan to 0.5 mm. 6. Select the appropriate laser to be used (IR800-780 nm excitation) or (Cy5.5 – 670 nm excitation) (see Note 13). 7. The following step, power automation, is run by the imaging software to optimize the power of the excitation laser and the integration time based on the signal emitted by the fluorophore from the selected region of interest (see Note 17). 8. Image ROI using program sequence set up in the acquisition software. After completion of data acquisition, all information is automatically saved. Turn off anesthesia. 9. Remove the animal from the scanner and return it into the cage to recover. 10. Repeat the imaging protocol described in steps 1–8 at desired time points after the injection (e.g., 0, 1, 2, 4, 8, 24, 48 h) (see Note 18). 3.2.4. Data Analysis
1. Reconstruct fluorescence intensity, fluorescence lifetime, and/or fluorescence chi-square image map, concentrationdepth maps and 3D volumetric representation (optical tomography) using OptiView analysis software version 2.0 from temporal point-spread functions (TPSF) data (21) (see Note 19). 2. Quantify total fluorescence intensity (FI) in the selected ROI at each time point after the injection using selection tools provided by the OptiView analysis software. Express data as a percent FI remaining (relative to time 0) at each subsequent time point (see Fig. 2). Calculate efflux rates and (intracerebral) residence half lives of injected molecules using appropriate fitting models in statistical analyses software.
3.2.5. Examples of Results and Their Interpretation
Models, methods, and protocols described in this chapter enable combined in vitro and in vivo assessment of brain efflux rates of compounds across the BBB. In the example described here, hFc antibody fusion protein was used as a model compound known to be recycled via endothelial FcRn receptor. It has been suggested that antibodies are actively transported from the brain via FcRnmediated reverse transcytosis (i.e., transcellular transport) across the BBB (4, 13, 16). In vitro compartmentalized BBB model uses SV-ARBEC cells shown to express FcRn (Fig. 1a). Transendothelial transport of hFc determined in this model using described protocols shows
N. Caram-Salas et al.
a
b
sv-ARBEC ACM 450 400 350 [ng/ml]
394
300 250 200 150 100 50 0
0
1
2 Time [hours]
3
4
Fig. 1. Polarized abluminal-to-luminal transport of hFc antibody fragment across in vitro blood–brain barrier (BBB) model using SV-ARBEC cells. (a) Immunofluorescence photo micrograph demonstrating the expression of FcRn [anti-human FcRn Ab from Santa Cruz, CA; green] in SV-ARBEC cells. Cell nuclei stained with Hoechst dye are shown in blue. Cell membrane stained with WGA (Vector Laboratories, Burlington, ON, Canada) is shown in red. (b) Schematic of the in vitro BBB model used to determine hFc clearance (line graph) across SV-ARBEC in luminal-to-abluminal (blue) and abluminal-to-luminal (red ) direction using protocols described in the chapter.
abluminal to luminal polarization; faster clearance rates were observed from bottom to top chamber (i.e., brain to blood). Since FcRn is principal receptor for hFc recycling, it is likely that this directional transport across the BBB in vitro is mediated by FcRn. This further suggests that, in contrast to endothelia from
395
26 In Vitro and In Vivo Methods for Assessing FcRn-Mediated Reverse…
hFc-IR800
8
1
24
2
48
120 Percent FI decay
0
80 60 40 20 0
BSA-IR800
8
1
24
2
48
8
24
48
20 30 Time [h]
40
50
40
50
40
50
80 60 40 20 0
10
120
2
Percent FI decay
Dextran10 kD -Cy5.5
1
10
T1/2 = 1.5 h
100
0 0
0
120 Percent FI decay
0
T1/2 = 2.36 h
100
100
20 30 Time [h]
T1/2 = 4.1 h
80 60 40 20 0
0
10
20 30 Time [h]
Fig. 2. Clearance of intracerebrally injected compounds labeled with the NIRF probes determined by prospective in vivo optical imaging. Fluorescence intensity (FI) images of the head region of interest (ROI) (left panels) in rats injected with IR800-hFc, IR800-BSA, or Cy5.5-Dextran 10 kDa into caudate-putamen, acquired at indicated time points after injection. Brain clearance rates (right panels) of IR800-hFc, IR800-BSA, or Cy5.5-Dextran 10 kDa determined from FI imaging data analyses in three animals.
peripheral vascular beds where FcRn has been shown to facilitate IgG transport in bi-directional manner BBB endothelia may be functionally polarized in FcRn-mediated antibody transport. This observation is consistent with the proposed directional transport/ elimination of antibodies from the brain across the BBB. The prospective in vivo imaging of the fluorescence intensity (FI) decay rates in the ROI regions corresponding to the brain injection sites of the NIRF-labeled compounds recycled by FcRn, hFc, and BSA, as well as inert, nontransported Dextrans (10 kDa) indicated that hFc and BSA have similar elimination/decay rates (residence half lives of 2.2 and 1.5 h, respectively) (Fig. 2), significantly shorter than that of Dextran 10 kDa (4.1 h), since previously measured residence half lives of compounds eliminated from the brain by bulk flow of ISF range between 6 and 16 h (7), shorter brain half lives of hFc and BSA measured in this study could be attributed to their active reverse transport across the BBB. The described methods could be applied to examine elimination rates
396
N. Caram-Salas et al.
of brain-injected compounds by both active (carrier-mediated) and passive (bulk-flow of the ISF and lymph drainage) routes. To specifically discern FcRn-mediated component of the active efflux of antibodies, similar studies using this method could be performed in FcRn-knockout animals.
4. Notes 1. Several in vitro models of the BBB similar to the one described here have been developed to analyze drug delivery and efflux from the brain (17, 22). The model used here consists of tissue culture inserts with semipermeable membranes seeded with a monolayer of brain endothelial cells which divides two media compartments. Various types of endothelial cells as well as co-culture models using brain cells (astrocytes, neurons, pericytes, etc.) or tissues (brain slices) that induce BBB phenotype of endothelial cells have been described and characterized in the literature (for review see ref. 22), and could be used as alternative to the set-up described here. For the purpose of evaluating active transport of antibodies, it is important to confirm that the selected model (i.e., endothelial cells) expresses FcRn receptor. 2. Immortalized rat brain endothelial cells (SV-ARBEC) were established by SV-40 transfection of primary rat brain microvascular endothelial cells, isolated from 24 to 30 days old Sprague–Dawley rats and were characterized as described previously (17). For the purpose of described studies, the expression of FcRn protein was confirmed in SV-ARBEC by immunochemistry (Fig. 1a). 3. Immortalized neonatal rat astrocyte cells, SV-RAS, were established by SV40 transfection of rat astrocyte cultures established from cortex of 4–8-day-old neonatal Sprague– Dawley rats. In this study, the conditioned media was used to induce BBB phenotype of SV-ARBEC (23). Cells can also be cultured at the bottom of wells in which SV-ARBECcontaining inserts are immersed, or on the opposite sides (to that of SV-ARBEC) of tissue culture inserts. 4. Rats are chosen for intracerebral injection studies because the size of their brain enables injection of higher volumes (5 ml), in contrast to the mouse brain (1 ml). Other strains of rats that can be used for this study include Sprague–Dawley or Long-Evans. 5. eXplore Optix is an in vivo small animal optical imaging equipment that uses Time Domain (TD) imaging technology.
26 In Vitro and In Vivo Methods for Assessing FcRn-Mediated Reverse…
397
For in vivo optical imaging, the observation of photon behavior in the near-infrared (NIR) region is favored because of tissue’s low absorption properties in this spectral band (between 650 and 1,100 nm), thus allowing light to penetrate several centimeters of tissue (24). In TD optical imaging, short pulses of light are sent to illuminate the specimen under study. The system then detects the photons according to their time-of-flight within the tissue; this time-of-flight distribution (generally called a TPSF or temporal point-spread function) is used to recover the optical characteristics of the specimen, discriminating absorption from scattering properties. Due to the temporal dimension in TD measurements, the signal already contains volumetric information about the tissue and enables tomographic (3D) image reconstruction. The fluorescence emission is collected by a highly sensitive time correlated single-photon counting system and detected through a fast photomultiplier tube (PMT). The images are reconstructed as fluorescence intensity (FI), fluorescence lifetime (t), and fluorescence concentration (Conc) using the OptiView software from ART Inc. (21). Refer to the eXplore Optix MX2 Operator’s manual for more details and options. Other fluorescence imaging equipments can be used to measure fluorescence intensity (e.g., In vivo multispectral FX® imager from Care-stream Imaging, IVIS® imaging system from Caliper Lifesciences, and Maestro® from CRI), but not fluorescence lifetime. 6. The assay is usually performed in triplicate membranes seeded with cells and in triplicate control membranes without seeded cells (for each direction). Evaluation of the tightness of cell monolayer using radioactive tracers (see below Note 7) is performed in a separate set of membranes. 7. The integrity of the brain endothelial cell layer forming an in vitro BBB can be monitored by measuring TEER or by determining its permeability to a paracellular tracer, such as 14 C-sucrose. For measurements of TEER, we routinely use an Endohm-12 Apparatus (World Precision Instruments Inc., Sarasota, FL, USA). Cell membranes showing TEER values higher than 40 W cm2 are accepted for subsequent studies. Although these TEER values are lower than those typically measured in similar in vitro BBB models using primary brain endothelial cells (bovine) (17), they were found acceptable for evaluating transcellular transport of larger molecules, such are peptides and antibodies, which show minimal paracellular diffusion at these TEER values. Alternatively, endothelial permeability coefficient for 14C-sucrose could be determined across separate endothelial monolayers using described protocols (17). Pesucrose values of well-formed SV-ARBEC
398
N. Caram-Salas et al.
monolayers are typically less than 8.5 × 10−6 cm/s. Cell membranes with Pesucrose values > 9 × 10−6 cm/s should be eliminated from further studies. 8. To protect cell monolayer integrity, especially when longer collection times are needed (over 2 h), instead of completely replacing plating (growth) media with transport buffer, only half the volume of plating media can be replaced with transport buffer. 9. The capture antibody from Fitzgerald is a monoclonal antibody, which will bind to only one epitope on the human Fc. The hFc is captured by this antibody only at this epitope, which leaves the majority of the hFc exposed. The second or detecting antibody (goat anti-huIgG Fc HRP) is a polyclonal antibody which contains multiple antibodies that bind different epitopes on the hFc. The exposed portion of the hFc will have some of these epitopes. 10. TMB is a chromogen that yields a blue color when oxidized with hydrogen peroxide (reaction catalyzed by HRP). 11. Calculate the permeability coefficient (Pe) and the clearance for each membrane (each condition is usually repeated in triplicate membranes) as: CI (ml) = ([C]A × VA)/[C]L, where CI is the initial drug concentration in the donor chamber, CA is the drug concentration in the acceptor chamber, and VA is the volume of the acceptor chamber. The average volume cleared is plotted versus time, and the slope is estimated by linear regression analysis. The slope of the clearance curves is denoted PSt, where PS is the permeability surface area product (in ml/ min). The slope of the clearance curve with the control filter is denoted PSf. The PS value for the endothelial cell monolayer (PSe) is calculated from: 1/PSe = 1/PSt − 1/PSf·PSe values is divided with the surface area of the filter (0.83 cm2) on which the cells are cultured to generate the “endothelial” cell permeability coefficient (Pe, in centimeters per second). The apparent permeability, Papp (cm/s), is calculated as: Papp = dQ/ dt·1/(A × C0), where dQ/dt is the transport rate of the compound (mol/s), A is the area of the cell monolayers (cm2), and C0 is the initial donor concentration (mol/L). 12. Since the primary purpose of described methods was to evaluate rates of reverse transcytosis of FcRn recycled compounds across the BBB, the “model” compounds used for demonstration are Fc containing molecules (13, 15, 16) and BSA, shown to be recycled by FcRn (25, 26). Molecule with similar size that is not substrate for FcRn recycling should be used as control; this molecule may be F(ab)2 fragment, dextran or engineered antibody molecules in which binding to FcRn has been neutralized by appropriate mutations. When choosing dextran or other branched polymers, take into account
26 In Vitro and In Vivo Methods for Assessing FcRn-Mediated Reverse…
399
that their diffusion is governed by the Stokes–Einstein hydrodynamic diameter rather than their molecular weight. The hydrodynamic radius of Dextran 10 kDa (5 nm) is similar to that of the antibody fragments used in this study. 13. Absorption of light in the far red spectrum (between 650 and 1,100 nm) by tissues is low, permitting light to penetrate several centimeters of tissue (23). Providing that the scanner is equipped with appropriate laser(s), dyes emitting in the far red spectrum, such as CyDyes monoreactive NHEster (Cy5.5emission 694 nm, GE Healthcare, Buckinghamshire, UK) and RDye800CW (IR800emission 789 nm) are dyes of choice for protein/antibody labeling applications requiring high labeling density. The NHS ester reactive group provides the functionality for labeling primary and secondary amino groups of test compounds. 14. The dye/protein ratio is calculated by taking into consideration the molar extinction coefficient (IR800 is 240,000 M−1 cm−1 and IgG is 150,000 M−1 cm−1) and the % of absorption of the dye at 280 nm. [IR800] = (O.D. at 778 nm × dilution)/240,000. [IgG] = [(O.D. at 280 nm × dilution) − 0.03 (O.D. at 778 nm × dilution)]/150,000. The correction factor 0.03 is used because the absorbance of the dye at 280 nm is approximately 3% of that at 778 nm. Dye/protein ratio equals [IR800]/ [IgG]. The fluorescence intensity is not always directly proportional to the amount (number) of dye molecules per labeled protein due to a possibility of self quenching. For an IgG, GE Healthcare suggest optimized labeling ratio with CyDye ranging from 4 to 12. The ratio may be optimized by varying the amount of dye per mg of protein in labeling protocol. For the studies presented here, dye/protein ratio ranged from 4 to 6. 15. The injection needle should be left in place for an additional 1 min after the injection to minimize the back flow of the injected compounds. 16. It is very important to turn on the machine for at least 30 min before placing animal into the scanner to heat the animal bed and to stabilize the laser. 17. Laser power and integration time per pixel are automatically optimized by the acquisition software. The animal bed is fitted with a monitoring device that detects the presence of the animal during the scan. If the animal wakes up and moves from the bed, eXpore Optix software will interrupt the scan. 18. The last in vivo scan could be followed by animal sacrifice by transcardial perfusion. Brain could be removed from the skull and imaged ex vivo in eXplore Optix, to obtain additional visual information on where the conjugate had been injected. To obtain more precise anatomical verification of the site of
400
N. Caram-Salas et al.
injection, brain can be fixed (4% formalin for 5 days) or flash frozen and then sectioned using vibratome or cryostat, respectively. The needle track and fluorescence distribution of the injected compounds could be visualized using appropriate filters on the fluorescence microscope. Additional immunofluorescence staining for cells (e.g., endothelial, neuronal, glial) or molecules of interest (e.g., FcRn) could be performed on these sections to investigate cellular co-localization or molecular interactions of injected compounds and correlate this information with results obtained by in vivo imaging. 19. The images obtained can be reconstructed and quantified as fluorescence intensity (FI), fluorescence lifetime (t), and fluorescence concentration (Conc) using the OptiView software from ART Inc. (21). Fluorescence lifetime is a characteristic of the conjugate in its environment. It should not significantly vary overtime or due to concentration differences. Fluorescence lifetime of the free fluorescent probe is different from the fluorescence lifetime of the probe conjugated to other molecules. Therefore, fluorescence life time gating of acquired images should be used to assure the integrity of the protein-fluorescent probe conjugate during the experimental follow-up time. Depth-concentration analysis is computed according to the lifetime of the conjugate. This feature allows quantification of the fluorophore concentration at different tissue depths and can be computed/displayed in 3D profile of the animal. While these analyses are complex and require proficiency with the equipment, analyses of fluorescence intensity described in this chapter will provide sufficient information to calculate basic parameters of fluorescence decay/efflux from the brain such as the rate of efflux and brain half life. References 1. Ballabh P, Braun A, Nedergaard M (2004) The blood-brain barrier: an overview: structure, regulation, and clinical implications. Neurobiol Dis 16:1–13. 2. Gaillard PJ, Visser CC, de Boer AG (2005) Targeted delivery across the blood-brain barrier. Expert Opin Drug Deliv 2:299–309. 3. Boado RJ, Hui EK, Lu JZ, Zhou QH, Pardridge WM (2010) Selective targeting of a TNFR decoy receptor pharmaceutical to the primate brain as a receptor-specific IgG fusion protein. J Biotechnol 146:84–91. 4. Zhang Y, Pardridge WM (2001) Mediated efflux of IgG molecules from brain to blood across the blood-brain barrier. J Neuroimmunol 114:168–172. 5. Thorne RG, Nicholson C (2006) In vivo diffusion analysis with quantum dots and dex-
6.
7.
8.
9.
trans predicts the width of brain extracellular space. PNAS 103:5567–5572. Weller RO, Galea I, Carare RO, Minagar A (2009) Pathophysiology of the lymphatic drainage of the central nervous system: Implications for pathogenesis and therapy of multiple sclerosis. J Pathophysiology (doi: 10.1015/j.pathophysiol.2009.10.007). Abbot JN (2004) Evidence for bulk flow of brain interstitial fluid: significance for physiology and pathology. Neurochemistry International 45:545–552 Schlachetzki F, Zhu C, Pardridge WM (2002) Expression of the neonatal Fc receptor (FcRn) at the blood-brain barrier. J Neurochem 81:203–206. Leach JL, Sedmak DD, Osborne JM, Rahill B, Lairmore MD, Anderson CL (1996) Isolation
26 In Vitro and In Vivo Methods for Assessing FcRn-Mediated Reverse… from human placenta of the IgG transporter, FcRn, and localization to the syncytiotrophoblast: implications for maternal-fetal antibody transport. J Immunol 157:3317–3322. 10. Roopenian DC, Christianson GJ, Sproule TJ, Brown AC, Akilesh S, Jung N, Petkova S, Avanessian L, Choi EY, Shaffer DJ, Eden PA, Anderson CL (2003) The MHC class I-like IgG receptor controls perinatal IgG transport, IgG homeostasis, and fate of IgG-Fc-coupled drugs. J Immunol 170: 3528–3533. 11. Yoshida M, Claypool SM, Wagner JS, Mizoguchi E, Mizoguchi A, Roopenian DC, Lencer WI, Blumberg RS (2004) Human neonatal Fc receptor mediates transport of IgG into luminal secretions for delivery of antigens to mucosal dendritic cells. Immunity 20:769–783. 12. Ober RJ, Martinez C, Lai X, Zhou J, Ward ES (2004) Exocytosis of IgG as mediated by the receptor, FcRn: an analysis at the single-molecule level. PNAS 101:11076–11081. 13. Roopenian DC, Akilesh S (2007) FcRn: the neonatal Fc receptor comes of age. Nat Rev Immunol 7:715–725. 14. He W, Ladinsky MS, Huey-Tubman KE, Jensen GJ, McIntosh JR, Björkman PJ (2008) FcRn-mediated antibody transport across epithelial cells revealed by electron tomography. Nature 455:542–546. 15. Garg A, Balthasar JP (2009) Investigation of the influence of FcRn on the distribution of IgG to the brain. AAPS J 11:553–557. 16. Deane R, Sagare A, Hamm K, Parisi M, LaRue B, Guo H, Wu Z, Holtzman DM, Zlokovic BV (2005) IgG-assisted age-dependent clearance of Alzheimer’s amyloid beta peptide by the blood-brain barrier neonatal Fc receptor. J Neurosci 25:11495–11503. 17. Garberg P, Ball M, Borg N, Cecchelli R, Fenart L, Hurst RD, Lindmark T, Mabondzo A, Nilsson JE, Raub TJ, Stanimirovic D, Terasaki T, Oberg JO, Osterberg T (2005) In vitro models for the blood-brain barrier. Toxicol In Vitro 19:299–334.
401
18. Muruganandam A, Herx LM, Monette R, Durkin JP, Stanimirovic DB (1997) Development of immortalized human cerebromicrovascular endothelial cell line as an in vitro model of the human blood-brain barrier. FASEB J 11:1187–1197. 19. Cecchelli R, Dehouck B, Descamps L, Fenart L, Buée-Scherrer VV, Duhem C, Lundquist S, Rentfel M, Torpier G, Dehouck MP (1999) In vitro model for evaluating drug transport across the blood-brain barrier. Adv Drug Deliv Rev 36:165–178. 20. Paxinos G, Watson C (2005) The rat brain in stereotaxic coordinates. 5th Edition, Elsevier Academic Press, Amsterdam Boston. 21. Abulrob A, Brunette E, Slinn J, Baumann E, Stanimirovic D (2007) In vivo time domain optical imaging of renal ischemia-reperfusion injury: discrimination based on fluorescence lifetime. Mol Imaging 6:304–314. 22. Gumbleton M, Audus KL (2001) Progress and limitations in the use of in vitro cell cultures to serve as a permeability screen for the blood-brain barrier. J Pharm Sci 90: 1681–1698. 23. Colgan OC, Collins NT, Ferguson G, Murphy RP, Birney YA, Cahill PA, Cummins PM (2008) Influence of basolateral condition on the regulation of brain microvascular endothelial tight junction properties and barrier function. Brain Res 1193:84–92. 24. Frangioni JV (2003) In vivo near-infrared fluorescence imaging Curr Opin Chem Biol 7:626–34. 25. Andersen JT, Sandlie I (2009) the versatile MHC class-I-related FcRn protects Igg and albumin from degradation: implication for development of new diagnostics and therapeutics. Drug Metab Pharmacokinet 24:318–332. 26. Chaudhury C, Mehnaz, S, Robinson JM, Hayton WL, Pearl DK, Roopenian DC, Anderson CL (2003) The major histocompatibility complex-related Fc receptor for IgG (FcRn) binds albumin and prolongs its lifespan. J Exp Med 197:315–322.
wwwwwwwwwwwww
Chapter 27 Evaluation of VEGF-Induced Vascular Permeability in Mice Sara M. Weis Abstract Vascular endothelial growth factor (VEGF) is a potent inducer of angiogenesis and vascular leak involved in development, wound healing, tumor growth, macular degeneration, and ischemia. Studying the effects of VEGF in vitro is not always sufficient to approximate the complex in vivo response that involves multiple cell types within functioning tissues. Treating mice with an intravenous injection of recombinant VEGF produces a rapid and transient biochemical response that is accompanied by a series of ultrastructural changes. Similar events are induced by hypoxia-induced VEGF in the heart following myocardial infarction or by tumor cell-released VEGF during metastasis. Studying how intact blood vessels respond to VEGF will augment the further development of antipermeability strategies to improve disease progression in a number of pathologies. Key words: Vascular permeability, Vascular leak, Vascular endothelial growth factor, Endothelial cells, Transmission electron microscopy, Mouse models
1. Introduction Vascular endothelial growth factor (VEGF) was initially described as a vascular permeability factor produced by some tumors, since it was a potent inducer of vascular leak and edema (1). In the nearly 30 years since, understanding the many aspects of VEGF signaling has led to a number of therapies such as Avastin (an anti-VEGF monoclonal antibody (2)) or Macugen (a pegylated anti-VEGF aptamer (3)) to block the ability of VEGF to induce vascular permeability and angiogenesis. The VEGF family of growth factors (VEGF-A, -B, -C, -D, and placenta growth factor (PlGF)) functions as soluble or matrix-bound ligands that exert activity by binding to tyrosine kinase receptors on the surface of target cells. Activation of receptor signaling leads to a number of
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_27, © Springer Science+Business Media, LLC 2011
403
404
S.M. Weis
downstream signaling pathways that induce changes in cell activity to induce vascular leak and/or to support vascular proliferation. In addition to targeting VEGF directly, blocking the function of the VEGF receptors or other critical downstream intermediates with antibodies or kinase inhibitors is a popular strategy to suppress VEGF signaling. A range of approaches have been employed to better understand how blood vessels respond to VEGF stimulation or VEGF pathway inhibition. Signaling pathways are often studied using cultured endothelial cells, while the ability of these cells to form tubes in 3D collagen gels serves as a model of angiogenesis and Transwell or Boyden chamber assays are used to investigate VEGF-induced permeability. To complement these in vitro models, we have developed a system to stimulate intact blood vessels with VEGF in order to closely examine how the vascular permeability response is regulated in vivo. To do this, mice are injected intravenously with a bolus or repeated injections of recombinant VEGF. At discrete time points, tissues are harvested for biochemical or ultrastructural analysis. This approach can be used to study how the permeability response varies in the presence of pharmacological inhibitors or in mice with gene-targeted deletions of critical downstream signaling proteins. Using these methods, we have shown that VEGF production and release by tumor cells (4) or ischemic tissues (5) induces the same signaling pathways and ultrastructural changes as a simple intravenous injection of VEGF. We discovered that Src family kinases are required downstream of VEGF in order to produce a vascular leak response, since mice lacking Src or Yes expression are resistant to the permeability-inducing effects of VEGF (6). Accordingly, Src-knockout mice are protected from the extravasation of VEGF-expressing tumor cells during metastasis (4), as well as VEGF-induced edema in the heart following myocardial infarction (5). Both of these VEGF-mediated permeability events could be blocked by simple pretreatment with inhibitors of VEGF, VEGF receptors, or Src family kinases (4, 5). Biochemically, we have implicated the VE-cadherin/b-catenin cell–cell adhesion pathway as a major effector of VEGF/Src required for the induction of vascular leak in vivo. At the ultrastructural level, VEGF and Src activity induce changes in endothelial cell–cell adhesion that result in vascular leak, platelet recruitment and activation, and start a series of events that lead to tissue damage when VEGF stimulation is prolonged (which is the case during the progression of ischemia or cancer). The methods described here can easily be adapted to study how other permeability factors exert activity in vivo, as well as to further test the mechanism of new therapeutic approaches targeting the VEGF signaling pathway.
27 Evaluation of VEGF-Induced Vascular Permeability in Mice
405
2. Materials 2.1. Intravenous Injection of VEGF to Induce a Systemic Vascular Response
1. Examples of agents that can block the permeability-inducing effects of VEGF: the Src family kinase inhibitor SKI-606 ((7, 8), 10 mg/kg i.p., 236 nM), the VEGFR2 inhibitor SU1498 (Calbiochem, 20 mg/kg i.p., 385 nM), and the VEGF inhibitor Cyclo-VEGI (CBO-P11) (Calbiochem, 2 mg/kg i.p., 13 nM). The same volume of appropriate vehicle controls must be used, such as sterile water, sterile saline, DMSO (dimethyl sulfoxide, 276855, Sigma-Aldrich), or PEG-400 (polyethylene glycol, 91893, Sigma-Aldrich). 2. For intravenous injection, mice are gently placed into a restrainer (available from Braintree Scientific), the tail submerged in warm tap water for a minute to dilate the tail blood vessels, and then the tail vein is injected using an insulin syringe (29 G × 1/2 cc, Terumo). 3. Recombinant human VEGFA-165 is from Peprotech, Inc. (100-20). Injection of 2 mg per mouse is costly, so it is prudent to purchase 1 mg in bulk ($4/mg) rather than smaller amounts which can cost as much as $40/mg. A 100 mg/ml stock solution is stored in aliquots at −80°C and diluted in sterile saline for an injection volume of 50–100 ml. Recombinant murine VEGFA is also available from Peprotech and produces an identical response in our experience.
2.2. Analysis of VEGF-Induced Biochemical Signaling Pathways In Vivo
1. RIPA lysis buffer: 6.05 g Tris base pH 7.5, 4.385 g NaCl, 5 g deoxycholic acid, 5 ml Triton X-100, 0.5 g SDS, and bring up to 500 ml with water. Store at 4°C. 2. Immediately before use, add protease and phosphatase inhibitors to 25 ml lysis buffer: 1.25 ml of 1 M sodium fluoride (NaF, protein phosphatase inhibitor), 500 ml of 100 mM phenylmethanesulfonyl fluoride (PMSF, serine protease inhibitor), 100 ml of 500 mM sodium orthovanadate (Na3VO4, tyrosine phosphatase inhibitor), 1 Complete Mini protease inhibitor tablet (Roche, protease inhibitor). 3. 6× Laemmli sample buffer: 28 ml 0.5 M Tris–HCl pH 6.8, 10 ml 100% glycerol, 4 g SDS, 3.72 g dithiothreitol (DTT), 100 ml 1% bromophenol blue, and water to 40 ml. Aliquot and store at −80°C. 6× buffer stock is diluted to 2× with water and added 1:1 to lysates for SDS-PAGE. 4. Although most immunoblotting antibodies will work, we use the following: VEGFR2 (SC-315, Santa Cruz), VE-cadherin (SC-6458, Santa Cruz), Phospho-p42/44 ERK antibody (9101, Cell Signaling), Phospho-Tyrosine (SC-7020 or SC-508). Typical dilution is 1:500 to 1:1,000 in TBS-T
406
S.M. Weis
containing 1% dry milk (Sanalac, Saco Foods) for 1 h at room temperature. 2.3. Analysis of Endothelial Barrier Integrity In Vivo Using Electron Microscopy
1. First fixative: 4% paraformaldehyde + 1.5% glutaraldehyde in 0.1 M cacodylate buffer: 17.5 ml of 16% paraformaldehyde, 4.2 ml of 25% glutaraldehyde, 17.5 ml of 0.4 M Na cacodylate, 30.8 ml of d2H2O. Make fresh and keep on ice. 2. Second fixative: 5% glutaraldehyde in 0.1 M cacodylate buffer: 28 ml of 25% glutaraldehyde, 35 ml of 0.4 M Na cacodylate, 77 ml of d2H2O. Make fresh and keep on ice. 3. Epon/Araldite resin: 24.70 g Epon 812, 33.25 g dodecenyl succinic anhydride, 31.05 g araldite, 2.3 ml dibutylphthalate, 2.5 ml DMP-30. Mix well by stirring and avoid bubbles. Store in 10 ml syringes at −20°C.
3. Methods 3.1. Intravenous Injection of VEGF to Induce a Systemic Vascular Response
Close examination of the vascular response to VEGF in the live mouse provides a novel means to investigate how different genetic alterations or pharmacological inhibitors might impact VEGFinduced vascular signaling pathways, permeability, and angiogenesis. To do this, we have developed a number of complimentary techniques to examine the vascular response to VEGF. 1. To test the impact of genetic manipulations, age- and gendermatched littermates should be used. The vascular response to VEGF can vary between male versus female subjects, and thus both genders should be examined if possible. Care should be taken to analyze results separated by gender and age to determine whether significant differences are observed (see Note 1). 2. When normal mice are being used, the choice of background strain can significantly impact the responsiveness to VEGF and other growth factors (9, 10) (see Note 2). 3. (Optional) Pretreatment with pharmacological inhibitors. Mice can be injected i.v. with test agents 5–30 min prior to the injection of VEGF. For example, we have found that pretreatment with inhibitors of Src family kinases (e.g., SKI-606) 5 min prior to VEGF injection can reduce the permeability-inducing ability of VEGF (5). Injection volume of test agents should not exceed 50–100 ml since the mouse will subsequently receive a 100-ml injection of VEGF (described below). Injection of appropriate vehicle (e.g., DMSO or PEG400) at equivalent volume should be performed in separate mice as a control. 4. Intravenous injection of VEGF. Adult mice are injected via the tail vein with 2 mg of either human or mouse recombinant
27 Evaluation of VEGF-Induced Vascular Permeability in Mice
407
VEGFA-165 (Peprotech) diluted in 100 ml sterile saline. Injection of 100 ml sterile saline serves as a control. 5. In our experience, a single i.v. injection of VEGF produces a very rapid and transient response in terms of biochemical signaling and ultrastructural response. For some experiments, it is therefore useful to subject mice to a series of VEGF injections to mimic exposure to VEGF produced by a tumor or ischemic tissue. To model this, 2 mg VEGF is injected i.v. every 30 min for 2 h. 6. At defined time points following injection (e.g., 1–60 min to study the biochemical response to VEGF), the mouse is euthanized using a carbon dioxide chamber followed by cervical dislocation, and the organs of interest are rapidly removed (time management is critical, see Note 3). 3.2. Analysis of VEGF-Induced Biochemical Signaling Pathways In Vivo
It is important to emphasize that in vitro analysis of the response to VEGF using cultured endothelial cells often does not yield similar results. There are inherent differences between endothelial cells isolated from large arteries, large veins, or capillary beds. Furthermore, the consequence of disrupted barrier function within a monolayer of cultured cells is quite different from the in vivo situation where endothelial cells act as gatekeepers between the bloodstream and extravascular space. The in vitro models cannot recapitulate the presence and physiologic impact of the basement membrane, perivascular cells, circulating myeloid cells, and pressure gradients that govern vascular permeability in vivo. By closely monitoring the consequences of VEGF stimulation in vivo, we have developed methods to show that exposure of blood vessels to VEGF induces a rapid and transient phosphorylation of VE-cadherin, resulting in the dissociation of VE-cadherin and b-catenin that contributes to breakdown of the physical link between the adherens junctions and the actin cytoskeleton (4). We believe that this in vivo signaling pathway is highly sensitive to VEGF since VE-cadherin is constitutively associated with VEGFR2 in quiescent vessels (5). Upon VEGF stimulation, Src is recruited to the VEGFR2/VE-cadherin signaling complex (5), resulting in phosphorylation of VE-cadherin at Y658 (which disrupts its association with p120-catenin) and Y731 (which disrupts its association with b-catenin) (11). This section describes preparation of samples for analysis by immunoprecipitation and/or immunoblotting. Mice are first injected intravenously with VEGF as described in Subheading 3.1 above. 1. Immediately upon removal, each organ is minced on a glass plate using razor blades and scraped into a 5-ml round bottom tube containing ice-cold RIPA lysis buffer with freshly prepared protease and phosphatase inhibitors. 3 ml lysis buffer is typically used for lung, kidney, or hearts each weighing
408
S.M. Weis
approximately 0.1–0.2 g, whereas a section of aorta should be homogenized in 200 ml so that the resulting lysates are not too dilute for standard immunoblotting. As a general rule, 10–20 ml RIPA per gram tissue is sufficient for lysis. 2. This mixture is homogenized for 20 s using a tissue grinder (e.g., a PRO-200 from PRO Scientific, Inc. fitted with a 7-mm saw tooth generator tip) while keeping the tube on ice. 3. The sample is transferred to 2 ml Eppendorf tubes and rotated end-over-end at 4°C for 20 min, and then centrifuged at 14,000 × g at 4°C for 10 min. 4. The supernatant is transferred to new tubes for storage at −80°C, and the pellet is typically discarded. 5. Protein concentration of the supernatant is determined, for example, using a BCA protein assay kit from Pierce. This method typically yields several milligrams of protein per organ. Samples must often be diluted with RIPA lysis buffer (e.g., 1:5) in order to stay within range of the 0–2 mg/ml albumin standards. 6. To confirm effective VEGF stimulation of various tissues, immunoblotting for activated phospho-Erk or phosphoVEGFR2 is performed as shown in Fig. 1. It might be necessary to “pull-down” proteins of interest in order to detect changes in phosphorylation (peak stimulation times can vary between organ beds, see Note 4). 7. Since the mice are treated with intravenous VEGF, the luminal surface of the endothelial cells is the first to respond to this stimulation. Immunoprecipitation and/or blotting for endothelial cell-specific proteins such as VE-cadherin or VEGFR2 produce the cleanest signaling response. In contrast, analysis of a protein expressed by all cell types within an organ (e.g., focal adhesion kinase) might poorly showcase the effects of VEGF since many cell types such as fibroblasts or smooth muscle cells represent a significant portion of the total organ lysate, but are not directly stimulated by the intravenous injection of VEGF. 8. Standard SDS-PAGE techniques are used. Samples are mixed 1:1 with a reducing Laemmli sample buffer, boiled at 100°C for 4 min, run on Novex 4–20% Tris–glycine gels (Invitrogen), and transferred to nitrocellulose at 20 V overnight for standard immunoblotting. 3.3. Analysis of Endothelial Barrier Integrity In Vivo Using Electron Microscopy
The preceding section provides a method to analyze the biochemical response of an entire organ to systemic treatment with VEGF. Whereas the measured response represents the consensus of all blood vessels, some subtle vascular remodeling events may remain unnoticed. Here, an alternative approach at the opposite end of the spectrum is used to examine the response of individual
27 Evaluation of VEGF-Induced Vascular Permeability in Mice
409
Fig. 1. Effect of mouse background strain on VEGF-induced angiogenesis and leak. The Matrigel plug assay was used to assess the ability of VEGF to induce vascular proliferation and vascular leak in mice of different background strain. Here, Matrigel containing 50 mg of human recombinant VEGFA-165 (Peprotech) was injected subcutaneously to the flanks of 8–11-week-old male mice. After 6 days, the plugs were removed and photographed to compare the extent of angiogenesis and vascular leak between mice of different background strain. Whereas mice on a 129 background showed a robust VEGF-induced response, FVB or C57BL6 mice showed little effect.
endothelial cells to VEGF using transmission electron microscopy. By examining the ultrastructural response to VEGF stimulation in vivo, we have been able to visualize how vascular permeability manifests within intact tissues. For example, we learned that intravenous VEGF injection rapidly induces changes in the appearance of endothelial cells lining capillaries in the heart. Within several minutes following VEGF stimulation, capillaries begin to exhibit irregular endothelial thickness, endothelial protrusions into the lumen, and expanded cytoplasmic vacuoles (12). VEGF treatment also induces gaps between adjacent endothelial cells (5), illustrating the permeability-inducing effects of VEGF. We have observed a similar vascular response to VEGF released by tumor cells at sites of extravasation in the lung (4). Many of the VEGF-induced endothelial gaps contain activated platelets adhering to exposed basement membrane (5). While platelet adhesion to exposed basal lamina may prevent further leak at that site, platelet activation begins to recruit and aggregate additional platelets, which reduces vessel patency and produces sites of microthrombosis. Acutely, activated platelets release VEGF from
410
S.M. Weis
their a-granules, which further potentiates the permeability response in the already affected vessels. In the longer term, microthrombi induces sites of focal ischemia which activate hypoxiaresponsive gene expression pathways leading to the increased production and release of VEGF, creating a series of secondary events leading to prolonged permeability and tissue damage. Therefore, examining the ultrastructural response to VEGF stimulation in vivo provides a unique method to study the pathological consequences of VEGF-induced leak (13). This section describes preparation of tissue for transmission electron microscopy. Mice are first injected intravenously with VEGF as described in Subheading 3.1 above. 1. At specified time points, mice are euthanized by carbon dioxide overdose followed by cervical dislocation. The ultrastructural changes induced by VEGF stimulation should be examined at a range of time points, and may vary between organs. We have closely examined the ultrastructural response to VEGF in the heart, brain, and lungs (4, 5, 12, 13). 2. For some experiments, mice are perfused with fixative to flush out circulating cells and produce open lumens. Extreme care must be taken to perfuse at a low physiological pressure so as not to damage the fragile capillary beds. Alternatively, the interaction of platelets or red blood cells with the endothelial surface may be of interest, in which case perfusion is not performed. 3. Organs are removed with extreme care not to poke, deform, or damage the morphology. Blunt forceps or scoops are helpful. Once the tissue has been fixed for an hour, tissues are less fragile and can be further sliced with a scalpel into smaller pieces to promote thorough fixation. 4. Tissue is fixed in 0.1 M sodium cacodylate buffer (pH 7.3) containing 4% paraformaldehyde and 1.5% glutaraldehyde for 2 h, transferred to 5% glutaraldehyde overnight, then to 1% osmium tetroxide for 1 h. 5. Blocks are washed, dehydrated in a graded ethanol series, and embedded in Epon/Araldite resin. Ultrathin sections are stained with uranyl acetate and lead citrate. 6. Samples are viewed using a transmission electron microscope (e.g., Philips CM-100). 7. The vascular response to VEGF can be evaluated by assessing the following: (a) Endothelial barrier function (initial response). • Evidence of gaps between adjacent endothelial cells. • Fenestrations within individual endothelial cells. • Presence of blood cells which have extravasated to the extravascular space.
27 Evaluation of VEGF-Induced Vascular Permeability in Mice
411
(b) Platelet activation and adhesion (secondary response). • Presence of platelets within vessel lumen. • Presence of degranulated platelets (indicative of platelet activation, a-granules release VEGF). • Platelet adhesion to basement membrane exposed by endothelial cell gaps. (c) Endothelial cell injury (cause or effect?). • Electron-lucent or swollen endothelial cells. • Endothelial cells with large vacuoles or many caveolae. • Vessel lumen is collapsed or occluded by abnormal endothelial cells. 3.4. VEGF Produced in Response to Hypoxia or Released by Tumor Cells
As opposed to direct injection of VEGF, the methods described above can also be applied to situations in which VEGF is released by other means. For example, we have used both the biochemical (Subheading 3.2) and transmission electron microscopy (Subheading 3.3) approaches to examine how hypoxia-induced VEGF expression during myocardial infarction induces edema and tissue damage in the heart (5). Mice were subject to ligation of the left anterior descending coronary artery and tissue was harvested at a range of timepoints (3–24 h). Care was taken to dissect and prepare tissues samples from the infarct, peri-infarct, and remote zones. Using electron microscopy, we observed signs of impaired endothelial barrier function and adhesion, as well as significant platelet activation, by 3 h following ligation. After 24 h, we observed a prevalence of endothelial cell injury and cardiac damage (evidenced as mitochondrial swelling, disordered cristae, and myofilament disintegration within the myocytes). VEGF was a contributor to the progression of injury in the heart, since mice that were insensitive to VEGF-induced permeability (those lacking Src or Yes expression, or normal mice pretreated with Src family kinase inhibitors) showed a dramatic reduction in these ultrastructural signs of vascular leak, platelet activation, and the resulting tissue damage. The benefits of blocking VEGF-induced leak observed using electron microscopy were associated with improved survival, decreased infarct size, improved cardiac function (measured by echocardiography), and decreased edema (measured as wet-to-dry weight or by T2-weighted MRI). We have also examined how VEGF-expressing tumor cells extravasate from the circulation into the lung by inducing a local vascular leak response (4). Using electron microscopy, we confirmed that VEGF-expressing tumor cells lodge in the lung capillaries immediately following intravenous injection through the tail vein. Within 1–3 h, we observed some tumor cells extending finger-like processes toward endothelial cell–cell junctions and
412
S.M. Weis
others with these processes already contacting the basal lamina. After 3 h, we began to see extravasated tumor cells in the extravascular space between the endothelial and pneumocyte layers. After several days, the lung was filled with established metastatic tumor foci. Importantly, VEGF-expressing tumor cells showed a significant advantage in their ability to extravasate and colonize the lung. Furthermore, the extravasation of VEGF-expressing tumor cells was dramatically reduced in mice that were insensitive to VEGF-induced permeability (mice lacking Src or Yes kinases) as well as normal mice treated with pharmacological inhibitors of VEGF, VEGFR2, or Src family kinases during the period of tumor cell extravasation.
4. Notes 1. We found an interesting vascular phenotype in male mice lacking integrin b3, but not female littermates (12). In the developing myocardium, VEGF-dependent neovascularization occurs by division of existing vessels, a process that persists for several weeks following birth. In male mice lacking b3, coronary capillaries fail to mature and continue to exhibit irregular endothelial thickness, endothelial protrusions into the lumen, and expanded cytoplasmic vacuoles. We were able to attribute this phenotype to enhanced VEGF signaling, since these hyperactivated vessels in male b3-knockout mice could be normalized by inhibitors of VEGF or VEGFR2. Moreover, we found that intravenous injection of VEGF induced a similar angiogenic phenotype in hearts of adult wild-type mice. It is interesting to consider why this phenotype was not seen in female b3-null mice. It has long been appreciated that estrogen modulates sensitivity to growth factors or cytokines, and that females show significant cardioprotective mechanisms (14). In fact, we measured lower VEGF gene expression in female b3-knockout mice, suggesting that females might have a more efficient negative feedback system capable of counteracting the hypersensitivity to VEGF that results from loss of b3 integrin expression. This example is discussed here to emphasize the need to closely examine the vascular response to VEGF in mice of similar age and gender so as to produce less variability within a given experiment. Furthermore, the comparison of VEGF-induced responses between genders should be incorporated into the experimental design when testing the impact of genetic deletions or pharmacological inhibitors. 2. In our experience, mice on the 129 mouse background strain consistently show the most robust response to VEGF (Fig. 1).
27 Evaluation of VEGF-Induced Vascular Permeability in Mice
413
This can be problematic when studying the effects of genetargeted deletion in mice on other strains such as FVB or C57BL6, since these strains consistently show a very poor response to VEGF (Fig. 1). When possible, experiments to test different permeability factors and/or pharmacological inhibitors can be performed using 129 mice to achieve the optimal permeability response with the best signal-to-noise ratio. Gene-targeted mice should always be compared to wild-type littermates to avoid any strain-dependent responses. Furthermore, comparing the VEGF-induced permeability response between mice with different genetic mutations may be complicated if mice are on different background strains. 3. Circulating VEGF produces an immediate and transient stimulation in vivo, and it is critical that the duration of stimulation be tightly measured and controlled between experiments. Activation of VEGF receptors can be observed within 1–2 min, and thus all tools, tubes, and reagents should be prepared in advance so that the removal and processing of tissues is efficient and consistent. Furthermore, if more than one organ will be harvested for a given experiment, the order of removal and processing should be standardized to reduce variability between animals. 4. The timeline of stimulation is transient and can vary by several minutes between different vascular beds (Fig. 2). The investigator must take into account which organ system is of primary interest for a given project, and then optimize the choice of timepoints. In addition, the optimal harvest time to produce activation of discrete downstream pathways can vary. For example, VEGFR2 activation tends to precede phosphorylation of Erk or Src kinases. If mice and reagents are not
Fig. 2. VEGF injection produces different activation kinetics in various organs. Mice were subject to an intravenous injection of recombinant VEGFA-165 as described (or sterile saline as a control), and organs of interest were harvested after 1–10 min. Tissue lysates were prepared and immunoblotting was performed for phospho-Erk to confirm VEGF stimulation. VEGF-induced Erk activation peaks by 1–2 min in the kidney or heart, whereas activation appears prolonged in the spleen. Thus, it is necessary to establish the kinetics of the pathways of interest in the organ of choice before proceeding with a more complicated analysis of response to pharmacological inhibitors or gene-targeted deletions.
414
S.M. Weis
limiting, it is generally useful to inject multiple mice with VEGF and take timepoints at 2, 5, 10, 20, 30, and 60 min. One mouse per timepoint is sufficient, although the experiment should be repeated several times.
Acknowledgments The methods described here were developed in the lab of David Cheresh formerly at The Scripps Research Institute in La Jolla and now at the University of California San Diego. The author would like to thank Dr. Cheresh as well as past and present members of the Cheresh lab for their contribution to these projects. The ultrastructural work would not have been possible without the technical expertise of Dr. Malcolm Wood at the TSRI Electron Microscopy Core. References 1. Senger, D. R., Galli, S. J., Dvorak, A. M., Perruzzi, C. A., Harvey, V. S., and Dvorak, H. F. (1983) Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites fluid, Science 219, 983–985. 2. Ferrara, N., Hillan, K. J., and Novotny, W. (2005) Bevacizumab (Avastin), a humanized anti-VEGF monoclonal antibody for cancer therapy, Biochem Biophys Res Commun 333, 328–335. 3. Rosina, C., Bottoni, F., and Staurenghi, G. (2008) Clinical experience with pegaptanib sodium, Clin Ophthalmol 2, 485–488. 4. Weis, S., Cui, J., Barnes, L., and Cheresh, D. (2004) Endothelial barrier disruption by VEGF-mediated Src activity potentiates tumor cell extravasation and metastasis, J Cell Biol 167, 223–229. 5. Weis, S., Shintani, S., Weber, A., Kirchmair, R., Wood, M., Cravens, A., McSharry, H., Iwakura, A., Yoon, Y. S., Himes, N., Burstein, D., Doukas, J., Soll, R., Losordo, D., and Cheresh, D. (2004) Src blockade stabilizes a Flk/cadherin complex, reducing edema and tissue injury following myocardial infarction, J Clin Invest 113, 885–894. 6. Eliceiri, B. P., Paul, R., Schwartzberg, P. L., Hood, J. D., Leng, J., and Cheresh, D. A. (1999) Selective requirement for Src kinases during VEGF-induced angiogenesis and vascular permeability, Mol Cell 4, 915–924. 7. Boschelli, D. H., Ye, F., Wang, Y. D., Dutia, M., Johnson, S. L., Wu, B., Miller, K., Powell,
D. W., Yaczko, D., Young, M., Tischler, M., Arndt, K., Discafani, C., Etienne, C., Gibbons, J., Grod, J., Lucas, J., Weber, J. M., and Boschelli, F. (2001) Optimization of 4-phenylamino-3-quinolinecarbonitriles as potent inhibitors of Src kinase activity, J Med Chem 44, 3965–3977. 8. Golas, J. M., Arndt, K., Etienne, C., Lucas, J., Nardin, D., Gibbons, J., Frost, P., Ye, F., Boschelli, D. H., and Boschelli, F. (2003) SKI606, a 4-anilino-3-quinolinecarbonitrile dual inhibitor of Src and Abl kinases, is a potent antiproliferative agent against chronic myelogenous leukemia cells in culture and causes regression of K562 xenografts in nude mice, Cancer Res 63, 375–381. 9. Rogers, M. S., Rohan, R. M., Birsner, A. E., and D’Amato, R. J. (2003) Genetic loci that control vascular endothelial growth factorinduced angiogenesis, Faseb J 17, 2112–2114. 10. Chan, C. K., Pham, L. N., Zhou, J., Spee, C., Ryan, S. J., and Hinton, D. R. (2005) Differential expression of pro- and antiangiogenic factors in mouse strain-dependent hypoxia-induced retinal neovascularization, Lab Invest 85, 721. 11. Potter, M. D., Barbero, S., and Cheresh, D. A. (2005) Tyrosine phosphorylation of VE-cadherin prevents binding of p120- and beta-catenin and maintains the cellular mesenchymal state, J Biol Chem 280, 31906–31912. 12. Weis, S. M., Lindquist, J. N., Barnes, L. A., Lutu-Fuga, K. M., Cui, J., Wood, M. R., and
27 Evaluation of VEGF-Induced Vascular Permeability in Mice Cheresh, D. A. (2007) Cooperation between VEGF and beta3 integrin during cardiac vascular development, Blood 109, 1962–1970. 13. Weis, S. M., and Cheresh, D. A. (2005) Pathophysiological consequences of VEGF-
415
induced vascular permeability, Nature 437, 497–504. 14. Mendelsohn, M. E., and Karas, R. H. (2005) Molecular and Cellular Basis of Cardiovascular Gender Differences, Science 308, 1583–1587.
wwwwwwwwwwwww
Chapter 28 In Vivo Measurement of Glioma-Induced Vascular Permeability Jisook Lee, Andrew Baird, and Brian P. Eliceiri Abstract The normal blood–brain barrier (BBB) consists of tight interendothelial cell junctions and adjacent astrocyte end feet separated by a basal lamina surrounding the endothelium. The interactions between the different cell types of BBB are disrupted in distinct patterns in the microenvironment of glioma. Malignant gliomas infiltrate the surrounding normal brain parenchyma; a process associated with vascular permeability (VP) and breakdown of the BBB. Herein, we describe methods to quantitatively measure gliomainduced vascular permeability, utilizing an orthotopic xenograft model of glioma. Key words: Vascular permeability, Glioma, Bioluminescent imaging, Blood–brain barrier
1. Introduction In normal brain, endothelial cells held together by tight junctions form a tight barrier between blood and brain parenchyma (blood– brain barrier, BBB). The integrity of BBB is regulated by endothelial cells, basal lamina surrounding the capillary walls, and adjacent astrocyte end feet that form a structure collectively known as the neurovascular unit, NVU (1, 2). The interaction between these different cell types of NVU maintains a tight barrier to protect the brain from various insults and regulate the fluid balance in the brain. However, the balance is disrupted in the microenvironment of various pathologies including glioma (3). These pathologies are often accompanied with vascular permeability (VP) and leakage of fluid into the brain parenchyma which results in high pressure and brain edema.
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_28, © Springer Science+Business Media, LLC 2011
417
418
J. Lee et al.
Glioma growth induces different morphological and functional alterations to each component of the NVU (4). Especially, glioma vasculature demonstrate dilated, tortuous, and leaky vessels with increased VP and breakdown of the BBB and this compromise is associated with increased glioma tumor growth and infiltration in human and rodents (5, 6). To examine glioma-induced vascular permeability in vivo, an orthotopic xenograft model is well suited to distinguish the role of host/tumor in tumor growth and progression. Xenograft models can be engineered such that tumor cells tagged with luciferase can be used to monitor tumor growth noninvasively or tagged with fluorescent markers for studies that need higher resolution. In fact, in vivo experimental models of glioma have been used to demonstrate that tagged tumor cells can be used to monitor the infiltration of tumor cells in the parenchyma and/or perivascular spaces even at a single cell level (5, 7–9). We describe methods to tag human glioma cells, implant them in mice by stereotactic injection, and monitor its growth in vivo. Following tumor implantation, tumor-induced vascular permeability and BBB breakdown can be assessed by measuring extravasation of labeled tracers in a quantitative manner.
2. Materials 2.1. Cell Culture and Labeling Tumor Cells
1. Dulbecco’s modified Eagle’s minimum essential medium (DMEM). 2. Fetal bovine serum (FBS), penicillin, streptomycin. 3. 37°C CO2 incubator. 4. DBTRG Human glioma cells (ATCC).
2.2. Intracranial Stereotactic Injection
1. Stereotactic frame (Kopf Instruments, Tujunga, CA). 2. Microsyringe (Hamilton, Reno, NV). 3. Isoflurane. 4. Buprenorphine. 5. Sterile suture material. 6. Lidocaine.
2.3. Noninvasive Imaging of Tumor Growth
1. d-Luciferin (Caliper Life Science): make 15 mg/ml stock in PBS, aliquots stored at −80°C. Thaw at room temperature before use. 2. Clipper/trimmer (WAHL Peanut clipper). 3. Nair lotion hair remover.
28 In Vivo Measurement of Glioma-Induced Vascular Permeability
2.4. Vascular Permeability Assay
419
1. FITC-dextran 70 kDa (Sigma), reconstituted in PBS at 50 mg/ml, aliquots stored at −80°C. Thaw at room temperature before use. Avoid exposure to light. 2. Heating pad. 3. 0.3-ml Syringe. 4. Heparin/Saline: 1 USP unit/ml of heparin dissolved in saline, make fresh before use. 5. Avertin (2,2,2-tribromoethanol, Aldrich): for 50× stock, make a 1:1 (w/v) solution of tribromoethanol in tert-amyl alcohol, store in the dark at 4°C. For 1× working solution, dilute the stock solution 1:50 in PBS in a glass vial and store in the dark at 4°C. 6. Surgery tools: forceps, chest retractor, scissors, rodent brain matrix for coronal sections.
3. Methods 3.1. Glioma Cells
1. Clone in open reading frame of firefly luciferase (pGL4.14, Promega) or red fluorescent protein (RFP) (Clontech) into a lentiviral expression vector (pLVX-puro, Clontech). 2. Prepare infectious lentiviral particles in 293 T packaging cells (ATCC) (Lenti-X Lentiviral Expression Systems, Clontech), store at −80°C. 3. Transduce DBTRG cells with lentivirus with polybrene added to regular growth medium (DMEM + 10% FBS with antibiotics) at a final concentration of 4 mg/ml. 4. Change to regular growth media 8–24 h after transduction (longer incubation of cells with polybrene may be toxic). 5. Select transduced cells by either flow cytometry (DBTRGRFP) or by adding 2 mg/ml puromycin in the culture medium for at least 2 weeks (DBTRG-luc-puro).
3.2. Intracranial Stereotactic Injections
1. Trypsinize tumor cells and resuspend cells in PBS at 1–2 E8/ ml, keep cells on ice. 2. Shave hair of mice in brain area prior to surgery (use Nair for complete hair removal). 3. Anesthetize animal before surgery with 2–3% isoflurane. 4. Immobilize 10-week-old immuno-deficient male mouse (Rag2 null) in a rodent stereotactic frame, make an incision in the skin, and make a burr hole in the skull. 5. Mix tumor cells and inject 5–10 ml (one million cells) at a rate of 1–2 ml/min using a microsyringe (Hamilton, Reno, NV) mounted on a stereotactic frame (Kopf Instruments,
420
J. Lee et al.
Tujunga, CA) using coordinates of 1 mm lateral and 1 mm anterior to the bregma and 1 mm below the dura. 6. Incisions are sutured with sterile suture material. 7. Apply lidocaine to the incision area. 8. Inject 0.05 mg/kg buprenorphine subcutaneously. 9. 5–10 ml of PBS is used as a control. 3.3. In Vivo Bioluminescent Imaging
Tumor growth is monitored noninvasively immediately after surgery as well as 4, 7, 14, 21, and 28 days after tumor implantation to generate a growth curve. 1. Anesthetize mice with 2–3% isoflurane. 2. Trim hair with clipper and remove remaining hair at the imaging site with Nair before each imaging as necessary (black hair quenches bioluminescent signal). 3. Inject mice intraperitonealy with 15 mg/ml stock).
d-luciferin
(150 ml of
4. Ten minutes after luciferin injection assess bioluminescent signal over an integration time of 10 and 60 s using a cooled charge-coupled device (CCD) camera (Spectrum; Caliper Life Sciences, Hopkinton, MA). 5. Quantitate luciferase signal in the brain using Live Image software. 6. Plot bioluminescent signal (corresponding to tumor growth) over time. 3.4. Measuring Vascular Permeability
Vascular permeability can be measured by measuring extravasation of fluorescently labeled dextran tracers after perfusion of blood vessels. 1. Immobilize mice and dilate tail-vein with heating pad. 2. Inject 100 ml of 50 mg/ml FITC-dextran 70 kDa into the tail vein. 3. Wait 15–20 min to allow circulation of dextran in the body. 4. For luciferase expressing cells, inject luciferin for bioluminescent imaging. 5. Inject 0.3–0.5 ml avertin i.p. to make sure animal is deeply anesthetized. 6. Open chest, perfuse animal with 5 ml of heparin/saline. 7. Place brain in rodent brain matrix to make 1-mm coronal sections. 8. Image tumor burden and FITC-dextran using a CCD camera (Fig. 1) (Spectrum, Caliper Life Sciences) or on a confocal microscope for higher resolution (Fig. 2).
28 In Vivo Measurement of Glioma-Induced Vascular Permeability
421
Fig. 1. Rag2 null mice were implanted with DBTRG-RFP cells by stereotaxic injection. Twenty-eight days after tumor implantation, mice were injected with FITC-dextran, systemically perfused and 1-mm brain sections were made. Brain sections were imaged to measure vascular permeability (FITC) and tumor (RFP) with a deep cooled CCD imaging system equipped with appropriate fluorescence filter cubes, background subtraction, and image integration software. Note the co-registry of FITC and RFP signal.
Fig. 2. Rag2 null mice were implanted with DBTRG-RFP cells by stereotaxic injection. Twenty-eight days after tumor implantation, mice were injected with FITC-dextran, systemically perfused and 1-mm brain sections were made. Extravastion of dextran (FITC) and tumor burden (labeled with RFP) in fresh brain sections were imaged with an Olympus Fluoview 1000 (ASW 1.7b) laser scanning confocal microscope (Olympus, Melville, NY).
4. Notes 1. Bicistronic lentiviral vectors expressing both luciferase and fluorescent proteins (i.e., green fluorescent protein, GFP) can be used to both perform noninvasive bioluminescent imaging on the basis of firefly luciferase as well as confocal fluorescent imaging for GFP localization. 2. Optimization of the peak bioluminescence is essential in initial studies to determine the ideal substrate incubation time. This can be most easily performed by imaging animal at short intervals over 1–15 min to determine peak bioluminescence.
422
J. Lee et al.
References 1. Iadecola, C. (2004), Neurovascular regulation in the normal brain and in Alzheimer’s disease. Nat Rev Neurosci,5(5): p. 347–60. 2. Simard, M. and M. Nedergaard (2004), The neurobiology of glia in the context of water and ion homeostasis. Neuroscience, 129(4): p. 877–96. 3. Abbott, N.J., L. Ronnback, and E. Hansson (2006), Astrocyte-endothelial interactions at the blood-brain barrier. Nat Rev Neurosci, 7(1): p. 41–53. 4. Lee, J., et al. (2009), Glioma-induced remodeling of the neurovascular unit. Brain Res, 1288: p. 125–34. 5. Lund, C.V., et al. (2006), Reduced glioma infiltration in Src-deficient mice. J Neurooncol, 78(1): p. 19–29.
6. Neuwelt, E.A. and S.I. Rapoport (1984), Modification of the blood-brain barrier in the chemotherapy of malignant brain tumors. Fed Proc, 43(2): p. 214–9. 7. Zhang, X., et al. (2002), Experiment and observation on invasion of brain glioma in vivo. J Clin Neurosci, 9(6): p. 668–71. 8. Winkler, F., et al. (2009), Imaging glioma cell invasion in vivo reveals mechanisms of dissemination and peritumoral angiogenesis. Glia, 57(12): p.1306-15 9. Lampson, L.A., M.A. Lampson, and A.D. Dunne (1993), Exploiting the lacZ reporter gene for quantitative analysis of disseminated tumor growth within the brain: use of the lacZ gene product as a tumor antigen, for evaluation of antigenic modulation, and to facilitate image analysis of tumor growth in situ. Cancer Res, 53(1): p. 176–82.
Chapter 29 In Vivo Optical Imaging of Ischemic Blood–Brain Barrier Disruption Abedelnasser Abulrob, Eric Brunette, Jacqueline Slinn, Ewa Baumann, and Danica Stanimirovic Abstract The blood–brain barrier (BBB) disruption following cerebral ischemia (stroke) contributes to the development of life-threatening brain edema. Recent studies suggested that the ischemic BBB disruption is not uniform throughout the affected brain region. The aim of this study was to establish in vivo optical imaging methods to assess the size selectivity and spatial distribution of the BBB disruption after a focal cerebral ischemia. The BBB permeability was assessed in mice subjected to a 60-min middle cerebral artery occlusion and 24 h of reperfusion using in vivo time domain near-infrared optical imaging after contrast enhancement with two tracers of different molecular size, Cy5.5 (1 kDa) and Cy5.5 conjugated with bovine serum albumin (BSA) (67 kDa). Volumetric reconstruction of contrast-enhanced brain areas in vivo and ex vivo indicated that the BSA-Cy5.5-enhancement is identical to the volume of infarct determined by TTC staining, whereas the volume of enhancement with Cy5.5 was 40% greater. The volume differential between areas of BBB disruption for small and large-size molecules could be useful for determining the size of peri-infarct tissues (penumbra) that can respond to neuroprotective therapies. Key words: Optical imaging, Ischemia, Near-infrared fluorescence, Blood–brain barrier
1. Introduction The blood–brain barrier (BBB), formed by endothelial cells lining cerebral microvessels, plays a pivotal role in protecting the neuronal microenvironment from blood-borne agents. While this is essential for physiological function, the BBB hinders the entry of drugs into the central nervous system (CNS). The endothelial cells forming the BBB are characterized by tight intercellular junctions, minimal pinocytosis and polarized expression of various transporters (see review (1)). It is now recognized that BBB permeability to solutes is determined by both passive and active Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8_29, © Springer Science+Business Media, LLC 2011
423
424
A. Abulrob et al.
processes and exists in a spectrum of permeability states rather than either open or closed. The BBB is of great clinical importance because many brain pathologies, including stroke, are associated with an increase in permeability or complete disruption of barrier properties resulting in extravasation of serum proteins and consequent vasogenic edema. The extent of the BBB damage during and after an ischemic event is a positive indicator of cerebral edema, hemorrhagic transformation, and worsened outcomes in stroke patients (2, 3). The BBB disruption occurs early after stroke and evolves with time, depending on the severity of the initial insult; it is variable in both the degree of disruption for different molecules and in its spatial relation to the infarct area. The recent study (4) demonstrated spatially more distributed areas of MRI enhancement due to the BBB damage for Gd-DTPA (550 Da) compared to Gd-DTPA linked to bovine serum albumin (BSA) (68 kDa) 24 h after a 3-h middle cerebral artery occlusion (MCAO) in rats. Measuring spatial spread of the BBB “breakdown” to molecules of different sizes is important for two reasons: while barrier permeability is increased, its protective function is reduced; however, the period when the barrier is more permeable can represent a therapeutic opportunity to deliver neuroprotective agents that would not normally enter the CNS. This chapter describes new in vivo and ex vivo optical imaging methods developed to evaluate size selectivity and spatial spread of the BBB disruption after focal cerebral ischemia. The method uses time-domain optical imaging in near-infrared spectrum, which allows deeper tissue penetration of several centimeters (5–7), analyses of the time domain characteristics of the fluorescence (8, 9), and optical tissue sectioning to reconstruct 3D images and calculate volume(s) of enhancement. The imaging method is then compared with a method commonly used to measure infarct volume in experimental stroke models in brain slices ex vivo.
2. Materials 2.1. Middle Cerebral Artery Occlusion
1. Male CD-1 mice 23–25 g (Charles River Laboratories, Wilmington, MA) (see Note 1). 2. Dissecting microscope (10–40×). 3. Digital thermometer with rectal probe. 4. Hot water-circulating pad. 5. Heat lamp (Variable Transformer, Variac, Cleveland, OH, USA). 6. Sundt AVM Micro Aneurysm Clip, curved 0.8 × 6 mm and Applier (Harvard Apparatus, Holliston, MA, USA). 7. #4 and #6 silk suture.
29 In Vivo Optical Imaging of Ischemic Blood–Brain Barrier Disruption
425
8. #8 black nylon monofilament suture. 9. Cutter cat. # 35166 hardener. 10. Cutter cat. # 35514 light silicon. 11. Surgical tools (Fine Science Tools Inc., North Vancouver, BC, Canada): (a) Curved extra fine Graefe forceps (10 cm 0.5-mm tip). (b) Two Dumont #5 microsurgery forceps with micro blunted, traumatic tips, one straight, one curved. (c) Olsen-Hegar needle holder with T-C jaws (12 cm). (d) Two fine Iris Scissors, 9 cm, one straight, one curved. (e) Small bulldog type hemostatic clamp, straight, 28 mm, gentle serrations. 2.2. Anesthesia
1. Isoflurane (Baxter Canada, Mississauga, ON). 2. 30% O2/N2 balance gas tank (Praxair Canada, Ottawa, ON). 3. Small animal anesthetic machine-tabletop (Surgivet Anesco, Waukesha, WI, USA).
2.3. Hair Removal
1. Small animal shaver (Peanut series) (Wahl, Shelton, CT, USA) 2. Hair removal cream (Veet, Reckitt Benckiser, Parsisppany, NJ, USA).
2.4. Preparation of Optical Contrast Tracers
1. Cy5.5 mono-reactive NHS ester dye (GE Healthcare, Amersham Place Little Chalfont, Buckinghamshire, UK). 2. Dimethyl sulfoxide (Sigma Aldrich Canada, Oakville, ON). 3. BSA (Sigma Aldrich Canada, Oakville, ON). 4. PBS (10× phosphate-buffered saline) (Wisent, ST-BRUNO, Quebec, Canada). 5. Carbonate buffer pH 9.3: Sodium Bicarbonate (NaHCO3, ACP, Montreal, QC, Canada), Sodium Carbonate (Na2CO3, BDH Chemicals, Toronto, ON, Canada). 6. Amicon Ultracel-4, 10-K centrifugal filter device (Millipore, Billerica, MA, USA). 7. Nanodrop Spectrophotometer (ND-1000) Technologies, Wilmington, DE, USA).
2.5. In Vivo Time Domain Near-Infrared Optical Imaging
(Nanodrop
1. Pre-clinical time-domain optical imager, eXplore Optix MX2 equipped with 670 nm laser and 700 nm filter (Advance Research Technologies, ART, Montreal, QC, Canada) (see Note 2). 2. Acquisition software (Optix MX2, Advance Research Technologies, ART, Montreal, QC, Canada). 3. Analysis software (Optiview, Advance Research Technologies, ART, Montreal, QC, Canada).
426
A. Abulrob et al.
2.6. Transcardiac Perfusion
1. Heparin, Hepalean (Organon, Toronto, ON, Canada). 2. Saline, 0.9% sodium chloride solution (Baxter, Toronto, ON, Canada). 3. Surgical tape, Transpore (3M, St Paul, MN, USA). 4. Peristaltic perfusion pump (Harvard Apparatus, Holliston, MA, USA).
2.7. TTC Staining and Infarct Determination
1. Stainless steel coronal brain matrices (Harvard Apparatus Canada, St. Laurent, QC). 2. 2% (2,3,5)-Triphenyltetrazolium chloride (TTC, Sigma, Oakville, ON). 3. AxioVision version 4.6 software (Carl Zeiss Canada Ltd., Toronto, ON). 4. 10% Buffered formalin phosphate solution (Fisher, Ottawa, ON).
3. Methods Protocols and methods described below consist of the following steps: (a) inducing a transient focal cerebral ischemia (stroke) by occluding (for 60 min) and then reperfusing (for 24 h) middle cerebral artery (MCA) in the experimental animal; (b) injecting one group of stroke-induced animals with the optical contrast agent of large molecular size (BSA-Cy5.5, 67 kDa; similar in size to serum proteins) and assessing contrast enhancement area in the brain due to the BBB disruption using in vivo time-domain optical imaging; (c) injecting another group of stroke-induced animals with the optical contrast agent of small molecular size (Cy5.5, 1 kDa; similar in size to nutrients and some drugs) and assessing contrast enhancement area in the brain using in vivo time-domain optical imaging (see Note 3); (d) sacrificing animals by transcardial perfusion and determining volume of contrast enhancement by BSA-Cy5.5 and Cy5.5 using optical tomography/sectioning ex vivo; (e) determining volume of infarct (i.e., irreversibly damaged/dead tissue) by TTC staining of serial thick brain sections, and (f) comparing infarct volume with volumes of contrast enhancement achieved by small- and large-molecular weight contrast tracers. This analysis provides the information on the degree and spatial spread of the BBB disruption for small- and large-molecular weight contrast tracers in the zone of the infarct and in peri-infarct areas. 3.1. MCAO in Mouse 3.1.1. Preparation of Threads
1. Using small measuring scale, measure and cut 11-mm long pieces of 8-0 (0.4 metric) Ethilon black monofilament under dissecting microscope. They have to be straight.
29 In Vivo Optical Imaging of Ischemic Blood–Brain Barrier Disruption
427
2. Squeeze a small amount of silicon (approximately 0.5 ml) onto a piece of paper. Stir gently with student quality forceps, e.g., Dumont # 3 to ensure mixture is homogeneous. Create a well in center and allow one drop of hardener from a 1-ml syringe to fall in well. Mix well with forceps, approximately 5 s. 3. Immediately pick up thread by one end with clean forceps, e.g., Dumont #5, and draw 7–8 mm of it though the silicone. Without letting go of the thread, place it on a clean section of the paper and gently pull excess silicone off. It should leave most of the silicon on the pad but remain well coated (120–180 mm thickness). 4. Threads should be placed standing on the gauze to dry overnight (see Note 4). 3.1.2. Anesthesia
1. Place a mouse into an anesthetic box. Induce anesthesia using a vaporizer with a 3 ml/min flow of 4% isoflurane in 30% O2/70% N2 gas mixture. 2. Anesthesia is maintained at 1 ml/min flow of 1.7% isoflurane in 30% O2/70% N2 gas mixture during surgery (the same anesthesia protocol is applied for head shaving and imaging procedures described below in Subheadings 3.3.1 and 3.3.2).
3.1.3. MCAO Procedure
1. After anesthetic induction the mouse is transferred to the surgery table, the nose and mouth are placed into a nose cone, taped in place with forelegs; tail is also taped with small strips of 3M Transpore Surgical Tape. The neck area should be slightly taut. Insert rectal temperature probe and adjust heating lamp to provide additional warmth if necessary. Mouse should be maintained at 37°C (see Note 5). 2. Clip fur with scissors, use masking tape to remove clippings and clean skin with appropriate disinfectant. A midline neck incision around 1-cm long is made with surgical scissors. 3. The midline submandibular glands are bluntly divided leaving right one in situ and left one retracted to the side with 28-mm hemostatic clamp and secured to the table by a 2-0 silk ligature and tape. 4. Under magnification the omohyoid muscle is carefully divided with Dumont microsurgery forceps and the common carotid artery (CCA) and carotid bifurcation are carefully isolated from the surrounding tissues. 5. Two 5-cm lengths of 6.0 silk suture are placed under the left external carotid artery (ECA). One is placed as far as possible cranially and used to permanently occlude ECA. Second one is tied in loose knot. 6. The left CCA is temporary occluded by a butterfly knot.
428
A. Abulrob et al.
7. A small (0.8 × 6 mm) MicroAneurysm Clip is placed on internal carotid artery (ICA). 8. Using a 30-gauge needle, a small hole is made in ECA between the two knots. 9. A silicon covered nylon thread is inserted into the hole and pushed until it reaches the butterfly knot in the CCA. The loose knot on the ECA is tightened gently and the microaneurysm clip is removed. The thread is maneuvered in such a way to introduce it into the ICA and advanced into ICA until approximately 1 mm is left remaining outside the ECA. The tightened knot may need to be loosened slightly to allow smooth movement of the thread. 10. The loose knot is again tightened holding thread in place; timing of occlusion begins. 11. The skin incision is closed with 4.0 silk suture. 12. For the time of occlusion the mice are placed in a cage standing on the warming pad (see Note 6). 3.1.4. Reperfusion
1. 5 min prior to the end of the ischemic period the mouse is re-anesthetized as described in Subheading 3.1.2 and taped to the surgery table. Rectal temperature probe is put in place and temperature is recorded. 2. The stitches are removed from the neck incision site, and the left submandibular gland is retracted to expose the filament tip. 3. The silk ligature is loosened, the nylon thread is withdrawn with the loose knot immediately tightened and tied permanently. 4. The butterfly knot on the CCA is removed and blood circulation restored. 5. The neck incision is restitched and mouse is placed for recovery in a clean cage.
3.2. Preparation of Contrast Tracers (Cy5.5 and BSA-Cy5.5) for Imaging
1. Cy5.5 NHS ester (N-hydroxysuccinimide ester of Cy5.5) (see Note 7) is supplied in a powder form, each vial containing 1 mg. Prepare a (10 mg/ml) stock solution by dissolving 1 mg of dye in 100 ml of DMSO (see Note 8).
3.2.1. Preparation of Cy5.5 NHS Ester Stocks and Inactivated Cy5.5
2. Aliquot the dye into 200-ml microcentrifuge tubes in 10 ml volume to avoid repeat freeze thaw cycles. Blow N2 gas in the tubes and cap tubes. Cy5.5 stocks can be stored at −80°C for 1 month or at −20°C for no longer than 2 weeks. 3. To inactivate the NHS ester on the Cy5.5, add 6 ml of (10 mg/ml) stock to 360 ml of milliQ H2O in a 1.5-ml microcentrifuge tube, vortex and place the tube on a rotator overnight at 4°C. The NHS ester will be hydrolyzed hence inactivating the
29 In Vivo Optical Imaging of Ischemic Blood–Brain Barrier Disruption
molecule. Final stock Cy5.5 = 0.1667 mg/ml.
concentration
of
429
inactivated
4. Once the NHS ester is hydrolyzed, the dye can be stored temporarily at 4°C or at −20°C for 3 months. 3.2.2. Preparation of Bovine Serum Albumin Conjugated to Cy5.5 Dye
1. Weigh out 1 mg of BSA directly into a 1.5-ml microcentrifuge tube and add 100 ml PBS to make a (10 mg/ml) solution (see Note 9). 2. Add 13 ml (~10% of final volume) of 1 M carbonate buffer pH 9.3 and vortex (see Note 10). Add 15 ml of (10 mg/ml) stock solution of Cy5.5 NHS ester to the mixture (see Note 11). Blow N2 gas in the tube to remove oxygen and humidity. A tenfold molar ratio of dye to protein will give a labeling ratio dye/protein in the range of 4–6 (see Note 12). 3. Incubate at RT for 1 h on tube rotator. 4. Remove unreacted dye from the BSA-Cy5.5 using Ultracel-4 10-K column. These columns have a molecular weight cutoff of 10 kDa; the Cy5.5 has a molecular weight of 1,128 g/mol and the BSA-Cy5.5 has a molecular weight of ~71,500 Da [(BSA = 67,000 Da) + (4× cy5.5 molecules = 4,512) = 71,512]. Add the BSA-Cy5.5 sample to the top chamber of the column, complete volume to 4 ml with PBS. Spin at 4,000 × g for 5 min at room temperature using a swinging bucket rotor centrifuge. Volume will be reduced to ~500 ml, discard the flow through, which will have a faint blue color. Reconstitute volume in the top chamber to 4 ml with PBS and spin sample as previously described. Repeat until flow through is clear, free of dye. Usually, three to five washes are required. 5. Measure the absorbance at 280 and 678 nm using nanodrop spectrophotometer. Use 2 ml PBS to blank the instrument. Use protein setting for the 280 nm reading and use the Cy5.5 fluorescence setting for the 678 nm. 6. Determine the molar dye/protein ratio from the readings using the following formulas:
[Cy5.5] = (A678nm ´
extinction coefficient (250, 000)
dilution factor) /
[BSA ] = (A280 nm ´
dilution factor) - (0.18 ´ A678nm
´ dilution factor) / extinction coefficient D / P ratio = [Cy5.5] / [BSA ] 7. BSA-Cy5.5 can be stored at −20°C for 3 months.
430
A. Abulrob et al.
3.3. Imaging Protocol(s)
1. The mouse is anesthetized as described in Subheading 3.1.2.
3.3.1. Hair Removal (See Note 13)
3. An electric small animal clipper is used to shave the fur off the ROI (see Note 14).
3.3.2. In Vivo Time-Domain Near-Infrared Fluorescence Imaging Using eXplore Optix Pre-clinical Scanner
1. Divide animals subjected to MCAO occlusion (described in Subheading 3.1) into two groups.
2. Tape can be used to immobilize the mouse.
2. Shave the head of animals as described in Subheading 3.3.1. 3. Image each mouse head as described below in steps 6–14. This prescan will be used to remove background from the experimental (contrast-enhanced) scans during the subsequent analysis. 4. In one group of animals, administer 50 mg BSA-Cy5.5 by intravenous injection (in total volume of 100 ml) through tail vein 15 min after inducing reperfusion (see Note 15). Allow animals to recover for 24 h and proceed with imaging (Fig. 1a, protocol schematic). 5. In the second group of animals, allow reperfusion for 24 h, then administer 100 nmol Cy5.5 by intravenous injection (in total volume of 50 ml) through tail vein (see Note 15); wait 15 min, then proceed with imaging (Fig. 1b, protocol schematic) (see Note 16).
a
b
Ischemia
Ischemia
Reperfusion
0 Background scan
1
Fluorescence scans BSA-Cy5.5 Injection
Cy5.5 Injection Reperfusion
25
0
Time (hours)
Background scan
R
R
L
L
1
25
Time (hours)
Fluorescence scans
Fig. 1. Schematic drawings of the experimental imaging protocol used for imaging the blood–brain barrier (BBB) disruption after an 1-h middle-cerebral artery occlusion (MCAO) followed by 24 h of reperfusion in mice using (a) large molecular weight tracer, BSA-Cy5.5 (65 kDa) and (b) small molecular weight tracer, Cy5.5 (1 kDa) and representative in vivo images of the head ROI (acquired using eXplore Optix) using these protocols. The images are reconstructed as head concentration volumetric planes [showing coronal (x ), sagital (y ), and axial (z ) planes] superimposed on the threedimensional image of the animal head profile.
29 In Vivo Optical Imaging of Ischemic Blood–Brain Barrier Disruption
431
6. For imaging, anesthetize the mouse as described in Subheading 3.1.2. 7. Once the induction phase is done, transfer the mouse from the anesthetic box onto the imager’s bed. In order to help maintain the mouse body temperature during the scans, the bed is heated to 36°C. 8. Supply anesthetic gas to imager to maintain anesthesia during scanning. Anesthesia is maintained at 1 ml/min flow of 1.5% isoflurane in 30% O2/70% N2 gas mixture. 9. Place the mouse in a prone position with its nose in the nose cone through which the anesthetic gas is supplied. Apply ophthalmic ointment or sterile eye drop to prevent the eyes from drying. 10. Secure the mouse position by taping down its front and rear paws to the bed. An overhead camera provides a live image which can be seen on the monitor; use this live image to position the mouse in a consistent matter. 11. Adjust the bed height using the live image from the side camera; align the top of the region of interest (ROI) to be scanned with the digitally created green line on the monitor. The green line is the focal plane; it is used to calculate time of flight of the photons which is then used to reconstruct depth and concentration. 12. Follow the instructions of the software and enter parameters in prompt windows. Parameters include selecting laser wavelength, selecting ROI, selecting steps (pixel) size, etc. Draw an ROI over the head being careful not to include the eyes in the ROI. 13. Once all parameters are entered, the imager will perform power automation in which the optimal laser power and integration time will be defined. The laser power and integration time will be set in accordance with the brightest pixel in the ROI so that the maximum number of photons will be received without damaging the PMT. 14. After the power automation is done, image ROI using program sequence setup in the acquisition software. After completion of data acquisition, all information are automatically saved. Turn off anesthesia. 15. Once the scan is completed, proceed with transcardiac perfusion (see Note 17). 16. Analyze the data using Optiview software from ART (10, 11). 3.4. Transcardiac Perfusion
1. Anesthetize the mouse as described in Subheading 3.1.2. 2. Prepare a saline–heparin solution by dissolving 200 ml heparin (1,000 U/ml) in 10 ml of saline.
432
A. Abulrob et al.
3. Place the mouse in a supine position on the imaging bed, maintain anesthesia with a nose cone. 4. Secure the mouse position by taping down its two front paws and its tail. 5. Remove the skin from upper abdomen and chest area. 6. Grab sternum with forceps and pull upward, with curved baby lexer scissors, cut along each side of the ribs cage to expose the diaphragm. 7. Using curved baby lexer scissors, cut through the diaphragm from one side to the other being careful not to cut any of the organs surrounding that area. 8. Expose the heart by cutting the rib cage with straight baby lexer scissors. Create a flap by making two lateral incisions on either side of the sternum up pass the heart level. Use a hemostat to keep the flap opened, exposing the heart. 9. Insert the perfusion needle (25G) in the left ventricle of the heart and secure it in place with a hemostat if desired (see Note 18). 10. Perform an incision in the right atrium to let the blood flow out. 11. Start the perfusion pump and perfuse at 2 ml/ml with heparinized saline. Make sure that the blood and perfusion solution is coming out of the atrium. 20 ml should be enough to clear the organs. 12. Excise the brain and perform ex vivo organ imaging using the eXplore Optix preclinical scanner. 3.5. Imaging Brains Ex Vivo
1. Place the excised brain on a black piece of plastic and on the imaging bad (platform). 2. Follow imaging procedures described in Subheading 3.3.2, steps 11–16. 3. The Optiview software can provide you with topographic three-dimensional representation of the axial (x-axis), sagittal (y-axis), and coronal (z-axis) volume planes (sections) of the head and optical tomography in 1-mm thickness increments. 4. Calculate the volume of contrast enhancement by multiplying the enhancement area in each slice with the thickness of the slice (11).
3.6. Triphenyltetrazolium Chloride Staining of the Brain Macrosections
1. After ex vivo brain imaging, mouse brain is sectioned into a four 2-mm coronal slices. 2. Expose sections to 2% (2,3,5)-triphenyltetrazolium chloride (TTC) solution for 15 min at 37°C and observe formation of red color. Intact brain regions stain dark red while infarcted regions
29 In Vivo Optical Imaging of Ischemic Blood–Brain Barrier Disruption
433
remain white (see Note 19) which allows easy demarcation of ischemic lesions from the intact brain areas. 3. Photograph stained sections using digital camera and then place them in 10% buffered formalin solution. 4. The infarct volume is determined from captured images using AxioVision version 4.6 software. 5. First, the scale bar is created by using advance window in the scaling tab of the software. The line is drawn against the ruler in the image of interest and measuring unit (in micrometers) is assigned. 6. Apply the scale bar to each image. 7. Using outline function from the “measure” tab of the software, outline the area of interest (i.e., white area representing infracted area) in each brain slice. The area of infarct in each slice is then multiplied by 2 mm (thickness) to determine the volume of dead tissue in each slice. The total infarct volume is derived by adding volumes from four thick brain slices (see Note 20). Examples of results and their interpretation The representative in vivo images [reconstructed as head concentration volumetric planes showing coronal (x), sagital (y), and axial (z) planes] of animals subjected to a 60-min MCAO and 24 h reperfusion using contrast-enhancing protocols with BSA-Cy5.5 and Cy5.5 are shown in Fig. 1a, b, respectively. Images show that BSA-Cy5.5 concentration signal is spatially confined to the hemisphere ipsilateral to MCAO, whereas Cy5.5 concentration signal is also detected in the contralateral frontal cortical region adjacent to brain midline. These observations suggest that the BBB disruption for a small tracer (Cy5.5) has wider spatial distribution than that for a large tracer (BSA-Cy5.5). Detailed analyses of these differences can be found in Abulrob et al. (11). These in vivo observations were also corroborated by demonstrated extravasation of Cy5.5 in brain sections using fluorescence microscopy (11). Volumetric analyses of the infarct volume based on TTC staining (Fig. 2c) and of contrast enhancement by BSA-Cy5.5 (Fig. 2a) and Cy5.5 (Fig. 2b) using optical tomography in brains of ex vivo corroborated in vivo imaging data. The volume of BSA-Cy5.5 contrast enhancement was similar to (Fig. 2d), whereas the volume of Cy.5.5 enhancement was 40% bigger than the infarct volume determined by TTC staining. Infarct volume determined by TTC staining in this example is consistent with values reported in literature in the same model (12). With some constraints, discussed in detail in Abulrob et al. (11), these observations could be interpreted as follows: in the ischemic core, the BBB is sufficiently damaged to allow free tissue accumulation of albumin and serum
434
A. Abulrob et al.
a
b
1
2 L
3
R
4
c L
R * +
x
Infarct Volume (mm3)
d 100 80 60 40 20
.5 y5 C
BS
A-
C
y5
TT
.5
C
0
Fig. 2. Volumetric analyses of contrast enhancement and infarct size in ex vivo brains of mice subjected to a transient 60-min middle cerebral artery occlusion followed by a 24-h reperfusion processed for optical imaging using (a) large molecular weight tracer, BSA-Cy5.5 (65 kDa) and (b) small molecular weight tracer, Cy5.5 (1 kDa). (a, b) Ex vivo brain images are reconstructed as (z) axis sections (ten 1-mm sections) through the thickness of the brain (optical tomography) showing Cy5.5 signal distribution in volumetric planes. (c) TTC-staining. (d) The volumes of TTC-staining (infarct size) and image contrast enhancement with small and large MW tracers are calculated by multiplying the enhancement area in each slice with the thickness of the slice. The results are means ± SD of four animals. The difference between volume of contrast enhancement between BSA-Cy5.5 and Cy5.5 is indicated in red lines and represents the volume of peri-infarct area in which the BBB is disrupted.
proteins; hence, contrast enhancement observed with BSA-Cy.5.5 roughly demarcates a region of severely damaged tissue. In contrast, the BBB permeability for smaller tracer is enhanced in both the ischemic core and in peri-infarct areas in which the BBB remains restrictive for serum proteins. Hence, the volumetric difference between BSA-Cy5.5 and Cy5.5 enhancement (see arrows in Fig. 2d) may represent the volume of brain “tissue at risk” which is not irreversibly damaged but is susceptible to secondary damage due to selective “breakdown” of the BBB. This tissue at risk is potentially “salvageable” by therapeutic treatment. The information that the BBB in this region is partially permissive for molecules of certain sizes could guide the selection of appropriate
29 In Vivo Optical Imaging of Ischemic Blood–Brain Barrier Disruption
435
neuroprotective drugs that could access their brain targets. Determining volume differential between the BBB disruption for small and large molecules could further be useful as a surrogate imaging marker for assessing prognosis and efficacy of therapeutic interventions in stroke patients.
4. Notes 1. The most important clinical equivalent of experimental focal brain ischemia is ischemic stroke. The most commonly used experimental model of focal brain ischemia is MCAO produced by various techniques (e.g., transorbital MCAO, transcranial MCAO, filament occlusion of MCA, clot embolism of MCA, etc.) in mice and rats. Detailed review of animal models of cerebral ischemia is provided in (13). A filament occlusion of MCA in the mouse was described here because of laboratory’s long experience with this model. Other focal ischemia models in mice and rats can be used in described experiments. 2. eXplore Optix is an in vivo small animal optical imaging equipment that uses time domain (TD) imaging technology. For in vivo optical imaging, the observation of photon behavior in the near-infrared (NIR) region is favored because of tissue’s low absorption properties in this spectral band (between 650 and 1,100 nm), thus allowing light to penetrate several centimeters of tissue (7). In TD optical imaging, short pulses of light are sent to illuminate the specimen under study. The system then detects the photons according to their time-of-flight within the tissue; this time-of-flight distribution (generally called a TPSF or Temporal Point Spread Function) is used to recover the optical characteristics of the specimen, discriminating absorption from scattering properties. Due to the temporal dimension in TD measurements, the signal already contains volumetric information about the tissue and enables tomographic (3D) image reconstruction. The eXplore Optix MX2 from ART uses a 670-nm pulse laser diode at 80 MHz with a time resolution of 12 ps light pulse. The fluorescence emission is collected at 700 nm by a highly sensitive time correlated single-photon counting system and detected through a fast photomultiplier tube (PMT). The images are reconstructed as fluorescence intensity (FI), fluorescence lifetime (t) and fluorescence concentration (Conc) using the OptiView software from ART Inc. (10). Refer to the eXplore Optix MX2 Operator’s manual for more details and different laser options. Other fluorescence imaging equipments can be used to measure fluorescence intensity
436
A. Abulrob et al.
(e.g., In vivo multispectral FX® imager from Care-stream Imaging, IVIS® imaging system from Caliper Lifesciences, and Maestro® from CRI), but not fluorescence lifetime. 3. 99.5% of hydrophilic compounds with molecular weight >500 kDa does not cross the BBB by paracellular diffusion (14). Neither Cy5.5 nor BSA-Cy5.5 crosses the intact BBB. Therefore, contrast enhancement in the brain after systemic injection of either optical tracer will indicate the BBB disruption. 4. Threads are best prepared at least 1 day before surgery. 5. Temperature measurement and control are essential in most ischemia experiments. Ischemic hypothermia lessens brain injury whereas hyperthermia exacerbates it (15). Furthermore, both delayed hypothermia and hyperthermia can significantly affect outcome in animal models of ischemia (16). Therefore, monitoring and maintenance of core and brain temperature during and after cerebral ischemia are important for producing reproducible infarct size. An additional problem is the sometimes discordant relationship between core and brain temperature measurements in rodent models (16). Brain temperature monitoring requires implantation of the brain probe and often complex telemetry equipment which is not readily available in nonspecialized labs. Therefore, the minimal requirement for described focal cerebral ischemia experiments is monitoring and maintenance of the core (rectal) temperature during both the occlusion and reperfusion at 37–38°C. 6. Total procedure time from cage to recovery cage should not last longer than 20–25 min per mouse. The length of MCAO will determine the extent of brain injury, infarct size, and animal survival (13). For 60-min MCAO described in this study, 24-h survival rates should be close to 100%. Mice with occlusion times of 60 min or less generally survive, recover, and can be kept indefinitely. 7. Dyes emitting in the far red spectrum, such as CyDyes monoreactive NHEster (Cy5.5 lemission 694 nm, GE Healthcare, Buckinghamshire, UK) are dyes of choice for protein/antibody labeling applications requiring high labeling density. The NHS ester reactive group provides the functionality for labeling primary and secondary amino groups of test compounds. It is important to inactivate Cy5.5 prior to injection to avoid random binding of monoreactive Cy5.5 to endogenous proteins. 8. NHS ester hydrolyzes in the presence of water. Therefore, the vial should be warmed up to RT before it is opened to avoid condensation of humidity from ambient air. Once the DMSO
29 In Vivo Optical Imaging of Ischemic Blood–Brain Barrier Disruption
437
is added, humidity should be removed by blowing N2 gas in the vial. Vortex until the dye is fully dissolved. 9. The buffer used to dilute protein to be labeled must not contain any primary or secondary amines as these will compete against the amine groups on the protein for labeling. 10. The NHS reaction is optimal between pH 7 and 9. The higher the pH, the faster the acylation reaction; however the hydrolysis is faster as well. 11. The most convenient and widely used functional group for labeling of peptides and proteins is the primary amino group provided by the e-amino group of lysine or the N-terminal amino group. Lysine is a relatively common amino acid and most proteins will have at least one. 12. If working with a less concentrated protein solution, i.e. (1 mg/ml), hydrolysis will be more predominant, hence increasing the dye to protein molar ratio to 20-fold will increase the acylation reaction so that the final dye/protein ratio will be similar to the described example using a (10 mg/ml) protein solution. 13. The fur can scatter or reflect the excitation light; removing it will improve light penetration into the tissue and reduce noise or artifacts from reflected light. The fur helps regulate the body temperature; so only shave the required areas for a given experiment. 14. Care must be taken not to cut the skin, especially when trying to shave a difficult area such as the head or the arm pits. Cuts can be painful to the animal and can also create artifacts. Use your finger to keep the skin stretched, this will help get a closer shave and prevent pinching and cutting of the skin. Cosmetic hair removal or depilatory agent can also be used to remove the fur. If using such products, clean thoroughly after use to avoid skin irritation. 15. To dilate the tail vein, the mouse cage should be placed under a heat lamp for roughly 1 min without the lid. The mouse tail has two parallel veins on each side that run from the tip to the base. Hold the tail near the tip with the major and the thumb. Locate one of the two tail veins; bend the tail at 90° over the index finger, this will allow you to insert the needle bevel up at an angle that follows the vein. The veins are very superficial and small, so the angle used to insert the needle is very important. Verify that the needle is well in the vein by drawing some blood back in the needle before injecting. If the needle moves out of the vein while injecting, the plunger will become hard to push and a subcutaneous lump will appear.
438
A. Abulrob et al.
16. The differences in imaging protocols (i.e., times of injections) between Cy5.5 and BSA-Cy5.5 are due to different molecular weights and circulation half lives of two tracers. Cy5.5 is rapidly cleared from the circulation, mainly through kidney excretion. Therefore, sufficient contrast to noise ratio for in vivo imaging of extravasated dye is 15–20 min. In contrast, the circulation half-life of BSA-Cy5.5 is much longer (i.e., several hours) and sufficient circulation clearance for in vivo imaging is achieved only 6 h after injection. Based on our previously published studies (11) one injection of BSA-Cy5.5 is sufficient for longitudinal in vivo imaging of up to 24 h. 17. The protocol described here terminates 24 h after MCAO. However, if of interest, the same animals could be imaged prospectively after this time to assess the resolution of BBB permeability changes or infarct size over longer periods of time. Potential prospective imaging protocols for BBB permeability are described in (11). 18. It is easy to puncture the heart wall during this procedure. The needle should be kept stable during the perfusion. 19. Incubation of tissue with TTC solution generates a red precipitate in tissue with active mitochondria and thus provides good staining contrast between living and dead tissue. 20. Typically, acute rodent stroke studies utilize TTC staining of the brain at the end of the study to determine infarct volume. The TTC staining and computer-based image analysis protocol can be automated to increase the throughput (17). Alternatively, more informative sequential tissue section imaging techniques ranging from H&E staining to multiparametric imaging and quantitative autoradiography (13) could be used to determine infarct size/volume and surrounding penumbra. References 1. Pardridge W.M. (2002) Drug and gene delivery to the brain: the vascular route. Neuron 36,555–8. 2. Wahlgren N.G., Ahmed N. (2004) Neuroprotection in cerebral ischaemia: facts and fancies-the need for new approaches. Cerebrovasc Dis. 17, 153–66. 3. Hofmeijer J., Veldhuis W.B., Schepers J., Nicolay K., Kappelle L.J., Bär P.R., van der Worp H.B. (2004) The time course of ischemic damage and cerebral perfusion in a rat model of space-occupying cerebral infarction. Brain Res. 1013, 74–82. 4. Nagaraja T.N., Karki K., Ewing J.R., Croxen R.L., Knight R.A. (2008) Identification of
Variations in Blood-Brain Barrier Opening After Cerebral Ischemia by Dual ContrastEnhanced Magnetic Resonance Imaging and T1sat Measurements. Stroke. 39, 427–432. 5. Wilkinson J.M., Kuok M.H., and Adamson G. (2004) Biomedical applications of optical imaging. Med Device Technol. 15, 22–24. 6. Licha K., Olbrich C. (2005) Optical imaging in drug discovery and diagnostic applications. Adv Drug Deliv Rev. 57,1087–108. 7. Ntziachristos V., Ripoll J., Wang L.V., and Weissleder R. (2005) Looking and listening to light: the evolution of whole-body photonic imaging. Nat Biotechnol. 23,313–320.
29 In Vivo Optical Imaging of Ischemic Blood–Brain Barrier Disruption 8. Elson D., Requejo-Isidro J., Munro I., Reavell F., Siegel J., Suhling K., Tadrous P., Benninger R., Lever J., Neil M., Phillips D., Stamp G., French P. (2004) Time-domain fluorescence lifetime imaging applied to biological tissue. Photochem Photobiol Sci. 3,795–801. 9. Bloch S., Xu B., Ye Y., Liang K., Nikiforovich G.V., Achilefu S. (2006) Targeting Beta-3 integrin using a linear hexapeptide labeled with a near-infrared fluorescent molecular probe. Mol Pharm. 3,539–549. 10. Abulrob A., Brunette E., Slinn J., Baumann E., Stanimirovic D. (2007). In vivo time domain optical imaging of renal ischemia- reperfusion injury: discrimination based on fluorescence lifetime. Mol Imaging 6,304–14. 11. Abulrob A., Brunette E., Slinn J., Baumann E., Stanimirovic D. (2008). Dynamic analysis of the blood-brain barrier disruption in experimental stroke using time domain in vivo fluorescence imaging. Mol Imaging 7,248–62. 12. Türeyen K., Vemuganti R., Sailor K.A., and Dempsey R.J. (2004). Infarct volume
439
uantification in mouse focal cerebral ischemia: q a comparison of triphenyltetrazolium chloride and cresyl violet staining techniques. J Neurosci. Methods 139,203–207. 13. Hossmann K.A. (2008) Cerebral ischemia: models, methods and outcomes. Neuropharmacology. 55, 257–70. 14. Pardridge W.M. (2007) Blood-brain barrier delivery. Drug Discov Today. 12, 54–61. 15. Colbourne F., Sutherland G., Corbett D. (1997). Postischemic hypothermia: A critical appraisal with implications for clinical treatment. Mol Neurobiol. 97, 171–201. 16. DeBow S., Colbourne F. (2003) Brain temperature measurement and regulation in awake and freely moving rodents. Methods. 30,167–71. 17. Regan H.K., Detwiler T.J., Huang J.C., Lynch J.J., Regan C.P. (2007) An improved automated method to quantitate infarct volume in triphenyltetrazolium stained rat brain sections. J Pharmacol Toxicol Methods. 56, 339–43.
wwwwwwwwwwwww
Index A Adherens junction................................................... 271–272 Albumin diffusion assay BLMVEC monolayer................................................298 bovine lung cell.................................................. 297–298 permeability coefficient.............................................299 Albumin flux............................................340–342, 344, 351 Albumin permeability materials....................................................................306 methods............................................................. 313–315 Analysis of variance (ANOVA).................................. 57–58 Angiogenic factor occludin expression............................................ 365–366 post-translational modification..................................366 VEGF-mediated permeability........................... 364, 365 ANOVA. See Analysis of variance Apparent permeability coefficient. See Angiogenic factor Avidin-biotin complex (ABC)........................................ 229
B Barrier function. See Endothelial barrier function Barrier regulation. See Mechano-transduction and barrier regulation Basal ectoplasmic specializations.....................................238 BBB. See Blood–brain barrier Bioluminescent imaging..................................................420 Blood–brain barrier (BBB). See also Glioma-induced vascular permeability; Ischemic blood–brain barrier anesthesia and analgesia.....................387 animal preparation materials..............................................................387 methods....................................................... 391–392 brain parenchyma......................................................371 cerebral microvasculature...........................................269 co-culture model........................................................396 data analysis...............................................................393 dye/protein ratio........................................................399 endothelial cells................................................. 370, 383 enzyme-linked immunosorbent assay materials...................................................... 386–387 methods...............................................................390 Evans blue carotid infusion....................................................371
cortical and subcortical region..................... 371, 373 materials..............................................................376 methods....................................................... 376–377 spectrophotofluorometry............................. 373–374 subcortical region......................................... 371, 372 vascular protein leakage.......................................372 Fitzgerald..................................................................398 fluorescence lifetime..................................................400 horseradish peroxidase electron microscopy..................................... 374, 375 macrophotography....................................... 374, 375 materials..............................................................376 methods....................................................... 378–379 morphological study............................................374 humoral immunity.....................................................384 imaging equipment and software...............................387 in vitro model materials...................................................... 385–386 methods....................................................... 388–390 in vivo imaging..........................................................395 NIRF dye materials..............................................................388 methods...............................................................391 optical imaging...................................385, 388, 391, 396 optix acquisition software.................................. 392–393 permeability coefficient.............................................398 permeability markers.................................................380 sodium fluorescein materials..............................................................376 methods....................................................... 377–378 molecular weight..................................................374 stereotactic and surgery equipment............................387 SV-ARBEC cells............................................... 393, 394 TEER 397 therapeutic antibodies................................................384 time-of-flight.............................................................397 transcardial perfusion.................................................399 uranyl acetate.............................................................380 Blood retina barrier angiogenic factor............................................... 363–366 cell isolation....................................................... 358–359 epifluorescent microscopy.................................. 363, 364 FD-70/FD-4 permeability................................ 359–360 immunohistochemistry..............................................358
Kursad Turksen (ed.), Permeability Barrier: Methods and Protocols, Methods in Molecular Biology, vol. 763, DOI 10.1007/978-1-61779-191-8, © Springer Science+Business Media, LLC 2011
441
Permeability Barrier 442 Index
Blood retina barrier (Continued ) occludin.....................................................................356 paracellular permeability............................................366 permeability assay......................................................357 primary cell culture............................................ 356–357 SDS-PAGE...............................................................357 TEER .......................................................................359 tight junction..................................................... 355–356 transcriptional regulation...........................................366 western blotting materials..............................................................358 methods....................................................... 360–362 Blood–retinal barrier (BRB)............................................180 Blood-testis barrier (BTB), 224. See also Sertoli cell Bovine serum albumin (BSA)..........................................143 Breast cancer resistance protein (BCRP).........................147 Bruch’s membrane...........................................................179
C Caco-2 permeability assay assay controls.............................................................146 BSA.... . ............................................................. 149–150 cassette dosing...........................................................152 cell culture and monolayer maintenance....................141 digoxin permeability..................................................148 energy-dependent mechanisms.................................140 enterocytes.................................................................140 HPLC and mass spectrometry materials..............................................................142 methods...............................................................150 intestinal absorption..................................................140 linear calibration curves.............................................150 lipophilic drugs..........................................................153 MRM transitions.......................................................150 oral delivery...............................................................139 P-gp, BCRP and MRP-2 efflux pumps............ 147–148 protein-sink conditions..............................................152 reagents......................................................................141 reference control compound and inhibitor stocks............................... 141–142 seeding monolayers............................................ 143–144 standard assay design......................................... 144–146 Capillary filtration coefficient materials....................................................................336 methods lung perfusion.............................................. 345, 346 lung preparation........................................... 345–347 measurement................................................ 348–349 Starling equation......................................... 345, 347 surgical procedure........................................ 347–348 Catenins... .......................................................................276 Catheterization Evans blue.................................................................376 horseradish peroxidase...............................................378 sodium fluorescein.....................................................377
Caveolin-1. See Pulmonary endothelial permeability Cell–cell contact...................................................... 272, 276 Cell column chromatography materials....................................................................255 methods............................................................. 260–261 Cell culture and in vivo model capillary filtration coefficient (see Capillary filtration coefficient)endothelial monolayer permeability materials..............................................................335 methods....................................................... 340–342 Evans blue-labeled albumin materials..............................................................336 methods....................................................... 344–345 hydraulic conductance materials..............................................................335 methods....................................................... 342–343 intraperitoneal injection.............................................353 lung injury and edema....................................... 334, 335 osmotic reflection coefficient (s) materials..............................................................337 methods....................................................... 350–352 pathological condition...............................................333 permeability determination........................................334 permeability × surface area product materials..............................................................337 methods....................................................... 349–350 TER materials..............................................................335 methods....................................................... 337–339 wet-dry weight ratio materials..............................................................336 methods....................................................... 343–344 Cell index (CI), 287 Cell junction. See Endothelial cell–cell junction Charge-coupled device (CCD).......................................420 Chemokines....................................................................330 Chopstick electrodes............................................... 219–220 Chromatin immuno-precipitation (ChIP) assays materials................................................................ 54–55 methods................................................................. 62–66 Claudin 11 protein antigen markers................................................. 231–233 dehydratation and mounting.....................................225 histological stain materials...................................................... 225–226 methods....................................................... 227–228 immunohistochemistry materials...................................................... 225–226 methods....................................................... 228–229 PAS stain........................................................... 230, 231 pathophysiology.........................................................229 sample preparation............................................. 226–227 signaling mechanism......................................... 229–230 spermatogenesis.........................................................224 tissue fixation..................................................... 233–235
tissue processing and sectioning................................225 Tuchmann-Duplessis trichrome stain............... 230, 231 Clonetics® .......................................................................320 Common carotid artery (CCA), 427 Costar®, 357 Cryosectioning....................................................... 36, 37, 41 Cultureware.....................................................................210 Cyber-T.... ................................................................... 56–57
D Dewaxing 225 Dextran permeability materials....................................................................321 methods............................................................. 324–326 Diabetic macular edema (DME)............................. 180–181 Diabetic retinopathy (DR)...................................... 180, 181 Diaminobenzidine substrate (DAB)................................229 Differentiated human urothelium in vitro barrier function..........................................................207 “biomimetic” urothelium........................... 208, 212–213 cell culture reagents...................................................209 cell harvesting agents......................................... 209–210 cellular differentiation................................................207 dextran permeability assessment materials..............................................................210 methods....................................................... 215–216 human urothelial cell cultures............................ 211–212 Krebs solution............................................................220 noxious urine components.........................................207 Snapwell™ membranes..................................... 219, 221 TER assessment.................................................... 213–215 measurements...................................... 210, 219, 220 monitoring, portable EVOM...............................215 Ussing chamber vs. chopstick electrodes..... 219–220 tissue engineering and transplantation strategies.......209 transcellular and paracellular urinary barrier properties............................................208 water and solute transport.........................................208 water and urea permeability assessment materials...................................................... 210–211 methods....................................................... 216–218 Dulbecco’s phosphate-buffered saline (dPBS).................209
E Electric cell-substrate impedance sensing system (ECIS™)............................................337 Electromobility shift assays (EMSA).......................... 60–62 Electronic volt-ohmmeter (EVOM)....................... 210, 215 Electron microscopy materials....................................................................406 methods biochemical response...........................................408 focal ischemia......................................................410
Permeability Barrier 443 Index platelet activation.................................................409 vascular response.......................................... 410–411 EMSA. See Electromobility shift assays Endothelial barrier. See Cell culture and in vivo model Endothelial barrier function cell column chromatography materials..............................................................255 methods....................................................... 260–261 electric resistance measurement.................................254 horseradish permeability assay materials..............................................................255 methods....................................................... 258–259 immunofluorescent staining materials..............................................................254 methods....................................................... 256–257 Marquardt iteration method......................................263 TEER measurement.......................................... 257–258 tracer elutes................................................................264 trypan blue albumin permeability materials..............................................................255 methods...............................................................262 Endothelial cell–cell junction adhesion molecules....................................................266 antibodies..................................................................268 biochemical analysis materials..............................................................268 methods...............................................................274 cell culture materials..............................................................268 methods...............................................................271 cell imaging adherens junction......................................... 271–272 tight junction.......................................................272 co immunoprecipitation..................................... 272–273 definition........................................................... 265–266 immortalization.........................................................275 immunostaining.........................................................268 molecular mechanism........................................ 267–268 phenotype and morphology.......................................275 phosphorylation.........................................................277 physiology and pathology.................................. 266–267 primary cell and cell lines..........................................270 vascular system................................................... 269–270 VE-cadherin...................................................... 276, 277 Endothelial cell permeability barrier function..........................................................281 cell–cell junction........................................................282 cell culture materials..............................................................282 methods...............................................................283 FITC-dextran............................................................289 paracellular pathway.......................................... 281–282 real-time cell electric impedance materials..............................................................283 methods....................................................... 287–288
Permeability Barrier 444 Index
Endothelial cell permeability (Continued ) TER materials..............................................................283 methods....................................................... 283–284 tracer flux assay materials..............................................................283 methods....................................................... 284–287 Endothelial electric resistance measurement...................254 Endothelial permeability. See Endothelial cell–cell junction; Human endothelial cell monolayer; Pulmonary endothelial permeability Enzyme-linked immunosorbent assay materials............................................................ 386–387 methods.....................................................................390 EPB. See Epidermal permeability barrier Epidermal barrier formation, transcription factor regulation ANOVA................................................................ 57–58 ChIP assays materials.......................................................... 54–55 methods........................................................... 62–66 Cyber-T................................................................. 56–57 EMSA 60–62 gel shift assays..............................................................54 gene program, experiments.................................... 53–54 keratinocytes.......................................................... 51–52 K-means clustering................................................ 58–59 luciferase reporter assays........................................ 66–70 Pearson’s product moment correlation coefficient............................................ 59–60, 68 potential gene targets evaluation..................................52 reporter assays........................................................ 55–56 TFBS.. . .......................................................................60 time course microarray experiment..............................53 Epidermal permeability barrier (EPB), mammalian skin dye diffusion assays......................................................75 embryos.......................................................................76 epidermal barrier.........................................................74 keratinocytes.......................................................... 73–74 Lucifer yellow........................................................ 77–78 materials......................................................................75 TEWL.................................................................. 78–79 X-gal diffusion....................................................... 76, 77 Epifluorescent microscopy....................................... 363, 364 Epithelial barrier integrity barrier disruption, reagents........................................197 cell culture materials...................................................... 196–197 methods....................................................... 198–199 data analysis materials..............................................................198 methods....................................................... 202–203 electron microscopy........................................... 196, 206 FITC-dextran solutions.............................................205 flux determination.....................................................206 human airways...........................................................195
influenza A virus........................................................205 lung epithelium..........................................................198 paracellular pathway..................................................206 permeability assay materials..............................................................197 methods....................................................... 200–201 permeability changing reagents......................... 199–200 polarized human lung epithelium..............................198 pulmonary epithelium...............................................195 TEER measurement materials..............................................................197 methods...............................................................199 Transwell® inserts.......................................................201 Epithelial dome formation adherens junction.......................................................170 basolateral transepithelial transport process...............170 cell culture materials...................................................... 170–171 methods...............................................................172 cellular necrosis..........................................................176 FLICA apoptosis detection kits.................................176 live-cell imaging................................................ 171–172 permanent cell culture end-point assays...................................................175 hemocytometer....................................................172 incubation media.................................................175 inverted microscope.............................................172 Translight™ wells................................................172 renal tubular cells.......................................................170 tight junction..................................................... 169–170 Evans blue dye method.................................... 84, 85, 87–88 External carotid artery (ECA).........................................427
F FcRn-mediated reverse transcytosis. See Blood–brain barrier FD-70/FD-4 permeability assay............................. 359–360 Filtration coefficient........................................................295 Fitzgerald. .......................................................................398 Flow cytometry...............................................................330 Fluorescein isothiocyanate (FITC)..................................197 Fluorescence intensity (FI)..............................................391 Fluorescence lifetime.......................................................400 Fluorescence microscopy......................................... 245, 246 Focal adhesion kinase (FAK)...........................................238 Focal cerebral ischemia.................................... 424, 426, 435
G Gas chromatography................................................. 99–100 Glioma-induced vascular permeability bicistronic lentiviral vectors.......................................421 bioluminescent imaging.............................................420 cell culture materials..............................................................418 methods...............................................................419
Permeability Barrier 445 Index
luciferase....................................................................418 noninvasive imaging..................................................418 stereotactic injection materials..............................................................418 methods....................................................... 419–420 tight junction.............................................................417 vascular permeability assay materials..............................................................419 methods....................................................... 420–421
H Horseradish permeability assay materials....................................................................255 methods............................................................. 258–259 Human endothelial cell monolayer chemokines................................................................330 dextran permeability materials..............................................................321 methods....................................................... 324–326 impedance measurement materials...................................................... 320–321 methods....................................................... 322–324 inflammation.............................................................319 insulation property.....................................................320 leukocyte transmigration materials..............................................................321 methods....................................................... 326–329 mechanical stress.......................................................329 neutrophil transmigration..........................................329 Human umbilical vein endothelial cells (HUVECs).......275 Humoral immunity.........................................................384 Hydraulic conductance materials....................................................................335 methods............................................................. 342–343 Hypotonic treatment............................................... 243–244 Hypoxia.... . ............................................................. 411–412
I I-albumin.....................................................................307 IFNg cytokine model.......................................................120 Immunofluorescent staining method materials....................................................................254 methods............................................................. 256–257 Immunohistochemistry...................................................358 Impedance materials............................................................ 320–321 methods............................................................. 322–324 Infrared densitometry (IR-D)...........................................38 Internal carotid artery (ICA)...........................................428 Internalization.................................................................274 Interstitial fluid (ISF)......................................................384 Intestinal epithelium barrier function..........................................................105 cell shedding and gap identification
125
acriflavine-stained villous epithelial cells..... 111–113 BCECF-AM.......................................................106 dynamic process...................................................106 goblet cells................................................... 110, 111 image analysis......................................................108 shedding nuclei....................................................106 stained cell nuclei, Hoechst 33258............... 110, 112 TNF-a................................................................. 106 confocal endoscopy, human materials..............................................................108 methods...............................................................110 confocal imaging, mice materials...................................................... 107–108 methods....................................................... 109, 110 Pasteur pipette bent, ‘L’ shape............................ 107, 113 rigid pen confocal probe microscopy materials..............................................................108 methods...............................................................110 small intestine............................................................105 surgical preparation materials..............................................................107 methods...............................................................109 Intestinal permeability, premature neonates NEC............................................................................96 sugar absorption test calculations.................................................. 102–103 derivatisation, pre-purification, and analysis.............................................. 99–101 lactulose and mannitol quantification.......... 101–102 materials.......................................................... 97–98 in neonates....................................................... 98–99 procedure...............................................................96 sugar solution.........................................................98 urinary pools preparation.......................................99 tight junctions..............................................................95 In vitro barrier function. See Endothelial barrier function IR-D. See Infrared densitometry Ischemic blood–brain barrier anesthesia...................................................................425 brain imaging.............................................................432 clinical uses................................................................424 depilatory agent.........................................................437 endothelial cells.........................................................423 hair removal materials..............................................................425 methods...............................................................430 hypothermia..............................................................436 ischemic stroke..........................................................435 lysine..........................................................................437 MCAO (see Middle cerebral artery occlusion) multiparametric imaging................................438 near-infrared optical imaging materials..............................................................425 methods....................................................... 430–431 NHS ester..................................................................436
Permeability Barrier 446 Index
Ischemic blood–brain barrier (Continued ) quantitative autoradiography.....................................438 TD optical imaging...................................................435 tracer preparation materials..............................................................425 methods....................................................... 428–429 transcardiac perfusion materials..............................................................426 methods....................................................... 431–432 TTC staining materials..............................................................426 methods....................................................... 432–435 Ischemic stroke................................................................435
reperfusion...........................................................428 thread preparation........................................ 426–427 Mouse model. See VEGF-induced vascular permeability Multiparametric imaging.................................................438 Multiple reaction monitoring (MRM) transitions..........150
N
Kedem–Katchalsky equation...........................................340 K-means clustering...................................................... 58–59
Near-infrared fluorescence...................................... 430–431 Necrotising enterocolitis (NEC).......................................96 Neurovascular unit (NVU)..............................................417 Neutrophils. See Human endothelial cell monolayer NHU cells. See Normal human urothelial cells Normal human bronchial epithelial (NHBE) cells..........196 Normal human urothelial (NHU) cells cultureware................................................................210 differentiated “biomimetic” urothelium.....................208 finite cell lines.................................................... 211, 212 growth medium.........................................................209 permeable Snapwell™...............................................217 Normalized cell index......................................................287
L
O
Leukocyte transmigration. See Human endothelial cell monolayer Lucifer yellow diffusion............................................... 77–78 Lungs. See Mechano-transduction and barrier regulation
Occludin... . .....................................................................120 Optical imaging. See Ischemic blood–brain barrier Optix acquisition software....................................... 392–393 Osmotic reflection coefficient (s) materials....................................................................337 methods............................................................. 350–352 Ouabain, cell contacts calcium phosphate precipitation materials..............................................................160 methods....................................................... 164–165 cell adhesion...................................................... 156–158 cell–cell connection....................................................156 cell culture materials...................................................... 158–159 methods....................................................... 161–162 claudin 2 mRNA.......................................................158 colon mucosa.............................................................157 connexin (cnx) 32......................................................157 differentiation............................................................166 epithelial growth factor..............................................158 FITC dextran materials..............................................................160 methods...............................................................164 MDCK cells..............................................................156 mRNA isolation................................................ 165–166 polyacrylamide gel electrophoresis..................... 162–163 RNA isolation...........................................................161 RT-PCR materials..............................................................161 methods...............................................................166 SDS polyacrylamide gel electrophoresis............ 159–160 tight junction (TJ).....................................................157
J Junction adhesion molecule ( JAM).................................356
K
M Marquardt iteration method............................................263 Matrigel™ . .....................................................................244 Matrigel plug assay.......................................... 409, 412–413 Mechano-transduction and barrier regulation albumin diffusion assay...................................... 297–300 BLMVEC monolayer................................................296 cell culture.................................................................292 edemagenic condition................................................300 inflammatory condition.............................................301 in vitro hydraulic system materials...................................................... 293–294 methods....................................................... 296–297 lung perfusion............................................................292 lung preparation................................................ 294–296 lung weight calculation..............................................296 mechanical force........................................................291 pharmacological agents..............................................300 rat isolation................................................................293 Microcarrier beads...........................................................260 Microvascular endothelial cell. See Mechano-transduction and barrier regulation Middle cerebral artery occlusion (MCAO) materials............................................................ 424–425 methods anesthesia.............................................................427 procedure..................................................... 427–428
Permeability Barrier 447 Index
transepithelial electrical resistance materials..............................................................159 methods...............................................................162 trypsin solution..........................................................167 western blotting materials..............................................................160 methods....................................................... 163–164
P Paracellular permeability barrier function.................................................. 311–312 immunofluorescent staining.............................. 312–314 regulation...................................................................304 PDR. See Proliferative diabetic retinopathy Permeability. See Blood–brain barrier; Blood retina barrier; Cell culture and in vivo model Permeability coefficient...................................................299 Permeability index (PI)....................................................326 Permeability inducing factors (PIF)................................276 Permeability × surface area product (PS) materials....................................................................337 methods............................................................. 349–350 Photomultiplier tube (PMT)...........................................397 Photoreceptors................................................................179 Placenta growth factor (PlGF)........................................403 Platelet activating factor (PAF).......................................259 Polyacrylamide gel electrophoresis (PAGE)............ 183, 188 Premature neonates, intestinal permeability NEC... .........................................................................96 sugar absorption test (see Sugar absorption test, premature neonates) tight junctions.................95 Primary decidual zone (PDZ) characteristics........................................................ 84, 85 macromolecules, visualization materials.................................................... 87, 90–91 methods........................................................... 90–91 Proliferative diabetic retinopathy (PDR)................. 180, 181 Pulmonary edema............................................................343 Pulmonary endothelial permeability, caveolin-1 albumin permeability materials..............................................................306 methods....................................................... 313–315 albumin uptake.................................................. 305–306 downregulation..........................................................304 endocytosis........................................................ 304–305 125 I handling...............................................................315 measurement.............................................................316 paracellular permeability.................................... 311–313 plasma protein...........................................................304 TER... .......................................................................306 transcellular permeability (see Transcellular permeability)
Q Quantitative autoradiography..........................................438
R Rat lung microvascular endothelial cells (RLMVECs).................................................304 Real-time cell electric impedance materials....................................................................283 methods............................................................. 287–288 Region of interest (ROI).................................................393 Rehydratation..................................................................225 Retinal microvascular endothelial cells (RMECs)...........356 Retinal pigment epithelium (RPE) Alexa secondary antibodies........................................193 apical and basolateral membrane...............................179 ARPE-19 culture......................................................185 bifidus .......................................................................192 BRB.... . .....................................................................180 confocal immunofluorescence, tight junctions materials..............................................................184 methods....................................................... 190–191 definition...................................................................179 dextran permeability measurement materials..............................................................183 methods....................................................... 186–187 diabetic retinopathy........................................... 180, 181 DME 180–181 hematoxilin–eosin......................................................180 human cell culture............................................. 181–182 paracellular epithelial electrical resistance measurement materials..............................................................182 methods....................................................... 187–188 PDR.................................................................. 180, 181 PMSF .......................................................................192 real-time PCR materials..............................................................184 methods...............................................................190 SDS PAGE materials..............................................................183 methods...............................................................188 sealing function..........................................................180 western blotting, tight junctions materials...................................................... 183–184 methods....................................................... 189–190 Reverse transcytosis. See Blood–brain barrier Rodent endometrium animals.........................................................................86 artificial stimuli............................................................84 decidualization.............................................................85 ECFV, quantitative assessment materials................................................................86 methods........................................................... 89–90 embryo implantation...................................................84 mammal uterus............................................................84 PDZ characteristics.................................................. 84, 85 macromolecules, visualization.................... 87, 90–91
Permeability Barrier 448 Index
Rodent endometrium (Continued ) vascular permeability Evans blue dye method.................................... 84, 85 qualitative assessment...................................... 86–88 quantitative assessment.............................. 86, 88–89 RPE. See Retinal pigment epithelium
S Scintillation.....................................................................211 SDS-PAGE.....................................................................357 Sertoli cell assessment......................................................... 249–250 culture preparation............................................. 248–249 function.....................................................................237 hypotonic treatment.......................................... 243–244 isolation and culturing materials...................................................... 240–241 methods....................................................... 241–243 occludin dissociation..................................................238 pre-isolation...............................................................241 seminiferous tubules..................................................224 spermatogenesis.........................................................224 spermiogenesis...........................................................239 TER (see Transepithelial electrical resistance) TJ-permeability materials..............................................................241 methods...............................................................244 Single-tracer method.......................................................350 Size-selective and in vitro assessment. See Blood retina barrier Skin barrier geometry models...........................................................2 keratinocyte cells...........................................................2 lag time, approximation......................................... 25–27 layers... . .....................................................................1–2 relative permeability, approximation cell volume....................................................... 23–24 2D and 3D cuboid model................................ 21, 22 horizontal cell overlap...................................... 22, 23 layer N............................................................. 24–25 membrane, effectiveness........................................21 TKD model..........................................................3D 21, 23 SC modeling (see Stratum corneum) virtual diffusion experiment................................................. 28–30 Skin segmentation and drug concentration cryosectioning materials.......................................................... 36, 37 methods.................................................................41 data treatment........................................................ 42–44 drug extraction materials................................................................36 methods.................................................................41
in vitro method............................................................34 IR-D measurements....................................................38 SC...... .........................................................................37 skin depth calculation........................................................ 44–48 profiles, parameters................................................37 skin thickness, measurement materials................................................................35 methods........................................................... 38–39 tape stripping materials.......................................................... 35–36 methods.................................................................40 Small airway epithelial cells (SAEC)..............................196 Smooth muscle cells (SMCs)..........................................269 Starling equation.............................................................342 Stratum corneum (SC) concentration density................................................. 7, 9 constant coefficients model membrane...................................................11 reformulation.....................................................9–10 cuboid model 2D, 2–3 3D ....................................................................... 3, 4 drug diffusion..............................................................13 flux, permeability, and lag time.............................. 10–11 homogenization...........................................................13 investigated model geometries...................................6–8 IR-D... . .......................................................................38 literature, dimensions.................................................5–6 mathematical models............................................. 13–21 numerical methods................................................ 11–13 tape stripping...............................................................37 TKD model 3D.........................................................3–5 transmission conditions.................................................9 Sugar absorption test, premature neonates calculations........................................................ 102–103 derivatisation, pre-purification, and analysis.............................................. 99–101 lactulose and mannitol quantification................ 101–102 materials................................................................ 97–98 in neonates............................................................. 98–99 procedure.....................................................................96 sugar solution...............................................................98 urinary pools preparation.............................................99
T Tape stripping....................................................... 35–36, 40 Temporal point-spread functions (TPSF).......................393 TER/TEER. See Transendothelial electrical resistance; Transepithelial electrical resistance Testis deficiency. See Claudin 11 protein Tetrakaidecahedra (TKD)...............................................3–5 TEWL. See Transepidermal water loss
Permeability Barrier 449 Index
TFBS. See Transcription factor binding sites Thrombin .......................................................................319 Tight junction (TJ) claudin 11 protein......................................................223 endothelial cell...........................................................272 Sertoli cell..................................................................237 Time domain (TD) imaging...........................................396 T84 intestinal epithelial cells vs. Caco-2 cells..........................................................116 cell culture materials..............................................................117 monolayers establishment method............... 122, 123 cell lysis and sample preparation materials..............................................................118 methods....................................................... 126, 127 gene expression..........................................................121 IFNg cytokine model.................................................120 immuno-cytochemical detection materials..............................................................120 methods....................................................... 131–133 immunocytochemistry...............................................136 IRF-1 detection materials..............................................................119 western blotting methods............................ 128–130 mycoplasma infection................................................121 paracellular permeability measurement materials..............................................................117 methods....................................................... 123, 124 SDS-polyacrylamide gel electrophoresis materials...................................................... 118, 119 methods...............................................................128 T84 cells transfection materials...................................................... 117, 118 stable cell lines establishment method......... 124–126 TER and FITC-dextran flux..................... 116, 120–121 b-tubulin materials...................................................... 119, 120 methods....................................................... 130, 131 Tracer flux assay materials....................................................................283 methods............................................................. 284–287 Transcellular permeability endocytosis........................................................ 307–308 fluorescent albumin uptake................................ 310–311 transendothelial permeability............................ 308–310 Transcription factor binding sites (TFBS)................... 53, 60 Transendothelial electrical resistance (TER/TEER) blood–brain barrier....................................................397 blood retina barrier....................................................359 cell culture and in vivo model materials..............................................................283 methods....................................................... 283–284 endothelial barrier function............................... 257–258
endothelial cell permeability materials..............................................................335 methods....................................................... 337–339 Transepidermal water loss (TEWL)............................ 78–79 Transepithelial electrical resistance data analysis....................................................... 202–203 differentiated human urothelium assessment.................................................... 213–215 measurements...................................... 210, 219, 220 monitoring, portable EVOM...............................215 Ussing chamber vs. chopstick electrodes............................................... 219–220 and FITC-dextran flux..............................................116 functional characteristics................................... 245, 246 measurement..................................................... 247–248 monolayer integrity....................................................144 Sertoli cells........................................................ 244, 245 TJ-permeability.........................................................244 Transmigration materials....................................................................321 methods............................................................. 326–329 Transmission electron microscopy................... 379, 409–411 Transwell® ............................................................... 321, 357 Transwells ............................................................... 117, 118 Triphenyltetrazolium chloride (TTC) stain materials....................................................................426 methods contrast-enhancing protocol........................ 430, 433 neuroprotective drugs..........................................435 volumetric analyses...................................... 433, 434 Trypsine assay..................................................................274 Trypsin inhibitor.............................................................210 Trypsin Versene (TV).....................................................210
U Uroplakin proteins...........................................................207 Ussing chamber....................................................... 219–220
V Vascular endothelial growth factor (VEGF)................................................. 266, 403 Vascular permeability. See Glioma-induced vascular permeability Vascular permeability (Vp) Evans blue dye method.......................................... 84, 85 qualitative assessment materials................................................................86 methods........................................................... 87–88 quantitative assessment materials................................................................86 methods........................................................... 88–89 VE-cadherin endothelial cell–cell junction............................. 276, 277
Permeability Barrier 450 Index
VE-cadherin (Continued ) materials....................................................................254 methods............................................................. 256–257 VEGF-induced vascular permeability activation kinetics.............................................. 413–414 electron microscopy (see Electron microscopy) hypoxia................................................... 411–412 integrin b3.................................................................412 intravenous injection materials..............................................................405 methods....................................................... 406–407 matrigel plug assay..................................... 409, 412–413 neovascularization......................................................412 potent inducer............................................................403 signaling pathway materials...................................................... 405–406 methods....................................................... 407–408 Vesicular permeability......................................................304 Vp. See Vascular permeability
W Western blotting blood retina barrier materials..............................................................358 methods....................................................... 360–362 cell lysis and sample preparation..................................... 118, 126, 127 IRF-1 detection......................................... 119, 128–130 ouabain materials..............................................................160 methods....................................................... 163–164 RPE materials...................................................... 183–184 methods....................................................... 189–190 semi-quantitative assessment.....................................121
X X-gal diffusion............................................................. 76, 77