Lymphangiogenesis

Methods in Molecular Biology 1846

Guillermo Oliver Mark L. Kahn Editors

Lymphangiogenesis Methods and Protocols

Methods

in

M o l e c u l a r B i o lo g y

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

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

Lymphangiogenesis Methods and Protocols

Edited by

Guillermo Oliver Center for Vascular & Developmental Biology, Feinberg Cardiovascular Research Institute, Northwestern University, Chicago, IL, USA

Mark L. Kahn Department of Medicine and Cardiovascular Institute, University of Pennsylvania, Philadelphia, PA, USA

Editors Guillermo Oliver Center for Vascular & Developmental Biology Feinberg Cardiovascular Research Institute Northwestern University Chicago, IL, USA

Mark L. Kahn Department of Medicine and Cardiovascular Institute University of Pennsylvania Philadelphia, PA, USA

ISSN 1064-3745     ISSN 1940-6029 (electronic) Methods in Molecular Biology ISBN 978-1-4939-8711-5    ISBN 978-1-4939-8712-2 (eBook) https://doi.org/10.1007/978-1-4939-8712-2 Library of Congress Control Number: 2018953044 © Springer Science+Business Media, LLC, part of Springer Nature 2018 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Humana Press imprint is published by the registered company Springer Science+Business Media, LLC part of Springer Nature. The registered company address is: 233 Spring Street, New York, NY 10013, U.S.A.

Preface All animals with a blood vascular system also have a lymphatic vascular system. The lymphatic vasculature is essential for maintaining fluid homeostasis, immune surveillance, and fat uptake from the intestinal tract. The lymphatic vasculature forms a complex, hierarchical network that contains two morphologically different types of vessels with distinct functions. Blind-ended initial lymphatics with discontinuous button-like junctions enable the rapid uptake of cells, proteins, and fluid. Larger collecting lymphatics transport lymph to lymph nodes, where immune responses are coordinated, and return lymph to the venous blood vascular network. Congenital malfunction of the lymphatic vasculature leads to primary lymphedema, and recent studies have revealed a myriad of novel functional roles for lymphatic vessels in diseases such as cancer, obesity, atherosclerosis, and hypertension. The last few years has seen an explosion in our understanding of the genetic and molecular mechanisms by which lymphatic vessels are specified, grow, and function. These discoveries have been made possible by the creation of new tools, techniques, and animal models designed specifically to dissect the processes of lymphatic vascular formation and lymphatic function in health and disease. In this book we provide a detailed description of cutting-edge protocols and reagents to help researchers utilize these new approaches to address their own questions related to the lymphatic vasculature. Chicago, IL, USA Philadelphia, PA, USA 

Guillermo Oliver Mark L. Kahn

v

Contents Preface�������������������������������������������������������������������������������������������������������������������������������    v Contributors ����������������������������������������������������������������������������������������������������������������������   ix 1 Three-Dimensional Visualization of the Lymphatic Vasculature�������������������������    1 Cathrin Dierkes, Aaron Scherzinger, and Friedemann Kiefer 2 Histological and Morphological Characterization of Developing Dermal Lymphatic Vessels���������������������������������������������������������������������������������������������  19 Kelly L. Betterman and Natasha L. Harvey 3 Genetic Lineage Tracing of Lymphatic Endothelial Cells in Mice�����������������������  37 Ines Martinez-Corral and Taija Makinen 4 Visualization and Tools for Analysis of Zebrafish Lymphatic Development���������  55 Kazuhide S. Okuda, Sungmin Baek, and Benjamin M. Hogan 5 Characterization of Zebrafish Facial Lymphatics�������������������������������������������������  71 Tiffany C. Y. Eng and Jonathan W. Astin 6 Correlative Fluorescence and Scanning Electron Microscopy to Study Lymphovenous Valve Development�������������������������������������������������������������������  85 Xin Geng and R. Sathish Srinivasan 7 Characterization of Mouse Mesenteric Lymphatic Valve Structure and Function�����������������������������������������������������������������������������������������������������  97 Amélie Sabine, Michael J. Davis, Esther Bovay, and Tatiana V. Petrova 8 Morphological Analysis of Lacteal Structure in the Small Intestine of Adult Mice ��������������������������������������������������������������������������������������������������� 131 Sang Heon Suh, Seon Pyo Hong, Intae Park, Joo-Hye Song, and Gou Young Koh 9 Morphological and Functional Analysis of CNS-Associated Lymphatics������������� 141 Jasmin Herz, Antoine Louveau, Sandro Da Mesquita, and Jonathan Kipnis 10 Morphological Analysis of Schlemm’s Canal in Mice ����������������������������������������� 153 Benjamin R. Thomson and Susan E. Quaggin 11 Imaging Lymphatics in Mouse Lungs ��������������������������������������������������������������� 161 Peter Baluk and Donald M. McDonald 12 Imaging of Endothelial Cell Dynamic Behavior in Zebrafish������������������������������� 181 Baptiste Coxam and Holger Gerhardt 13 Visualization and Measurement of Lymphatic Function In Vivo������������������������� 197 Samia B. Bachmann, Michael Detmar, and Steven T. Proulx 14 Investigating Effects of Fluid Shear Stress on Lymphatic Endothelial Cells��������� 213 Daniel T. Sweet, Joshua D. Hall, John Welsh, Mark L. Kahn, and Juan M. Jiménez

vii

viii

Contents

15 Methods for Assessing the Contractile Function of Mouse Lymphatic Vessels Ex Vivo ������������������������������������������������������������������������������������������������� 229 Jorge A. Castorena-Gonzalez, Joshua P. Scallan, and Michael J. Davis 16 Evaluation and Characterization of Endothelial Cell Invasion and Sprouting Behavior������������������������������������������������������������������������������������� 249 Jocelynda Salvador and George E. Davis 17 Dorsal Ear Skin Window for Intravital Imaging and Functional Analysis of Lymphangiogenesis��������������������������������������������������������������������������������������� 261 Witold W. Kilarski, Esra Güç, and Melody A. Swartz 18 Isolation of Human Skin Lymphatic Endothelial Cells and 3D Reconstruction of the Lymphatic Vasculature In Vitro��������������������������������������������������������������� 279 Anita Rogic, Francois Auger, and Mihaela Skobe 19 Stimulation and Inhibition of Lymphangiogenesis Via Adeno-Associated Viral Gene Delivery������������������������������������������������������������������������������������������� 291 Sinem Karaman, Harri Nurmi, Salli Antila, and Kari Alitalo 20 Isolation and Characterization of Mouse Organ-Specific Endothelial Transcriptomes������������������������������������������������������������������������������� 301 Shahin Rafii and Brisa Palikuqi 21 Single-Cell mRNA Sequencing of the Mouse Brain Vasculature������������������������� 309 Michael Vanlandewijck and Christer Betsholtz 22 Metabolic Analysis of Lymphatic Endothelial Cells��������������������������������������������� 325 Pengchun Yu, Tiago C. Alves, Richard G. Kibbey, and Michael Simons 23 Characterizing Epigenetic Changes in Endothelial Cells������������������������������������� 335 Matthew T. Menendez and Courtney T. Griffin Index���������������������������������������������������������������������������������������������������������������������������������  345

Contributors Kari Alitalo  •  Wihuri Research Institute and Translational Cancer Biology Program, Biomedicum Helsinki, University of Helsinki, Helsinki, Finland Tiago C. Alves  •  Section of Endocrinology, Department of Internal Medicine, Yale University School of Medicine, New Haven, CT, USA Salli Antila  •  Wihuri Research Institute and Translational Cancer Biology Program, Biomedicum Helsinki, University of Helsinki, Helsinki, Finland Jonathan W. Astin  •  Department of Molecular Medicine and Pathology, School of Medical Sciences, The University of Auckland, Auckland, New Zealand Francois Auger  •  Centre LOEX de l’Université Laval, Regenerative Medicine Section of the FRQS Research Center of the CHU de Québec, Quebec, QC, Canada Samia B. Bachmann  •  Institute of Pharmaceutical Sciences, Swiss Federal Institute of Technology, ETH Zurich, Zurich, Switzerland Sungmin Baek  •  Stowers Institute for Medical Research, Kansas city, MO, USA; Division of Genomics of Development and Disease, Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD, Australia Peter Baluk  •  Department of Anatomy, Helen Diller Family Comprehensive Cancer Center, Cardiovascular Research Institute, University of California, San Francisco, CA, USA Christer Betsholtz  •  Karolinska Institutet/AstraZeneca Integrated Cardio Metabolic Centre (KI/AZ ICMC), Huddinge, Sweden; Department of Immunology, Genetics and Pathology, Rudbeck Laboratory, Uppsala University, Uppsala, Sweden Kelly L. Betterman  •  Centre for Cancer Biology, University of South Australia and SA Pathology, Adelaide, SA, Australia Esther Bovay  •  Division of Experimental Pathology, Department of Oncology, CHUV, Ludwig Institute for Cancer Research, University of Lausanne, Epalinges, Switzerland Jorge A. Castorena-Gonzalez  •  Department of Medical Pharmacology and Physiology, University of Missouri, Columbia, MO, USA Baptiste Coxam  •  Integrative Vascular Biology Laboratory, Max-Delbrück Center for Molecular Medicine (MDC), Berlin, Germany; DZHK (German Center for Cardiovascular Research), Berlin, Germany; Berlin Institute of Health (BIH), Berlin, Germany Sandro Da Mesquita  •  Department of Neuroscience, Center for Brain Immunology and Glia, School of Medicine, University of Virginia, Charlottesville, VA, USA George E. Davis  •  Department of Medical Pharmacology and Physiology, Dalton Cardiovascular Research Center, University of Missouri School of Medicine, Columbia, MO, USA Michael J. Davis  •  Department of Medical Pharmacology and Physiology, University of Missouri, Columbia, MO, USA Michael Detmar  •  Institute of Pharmaceutical Sciences, Swiss Federal Institute of Technology, ETH Zurich, Zurich, Switzerland

ix

x

Contributors

Cathrin Dierkes  •  Max Planck Institute for Molecular Biomedicine, Münster, Germany Tiffany C. Y. Eng  •  Department of Molecular Medicine and Pathology, School of Medical Sciences, The University of Auckland, Auckland, New Zealand Xin Geng  •  Cardiovascular Biology Research Program, Oklahoma Medical Research Foundation, Oklahoma, OK, USA Holger Gerhardt  •  Integrative Vascular Biology Laboratory, Max-Delbrück Center for Molecular Medicine (MDC), Berlin, Germany; DZHK (German Center for Cardiovascular Research), Berlin, Germany; Berlin Institute of Health (BIH), Berlin, Germany; Vascular Patterning Laboratory, Vesalius Research Center, Leuven, Belgium; Department of Oncology, KU Leuven, Leuven, Belgium Courtney T. Griffin  •  Cardiovascular Biology Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA; Department of Cell Biology, University of Oklahoma Health Sciences Center, Oklahoma City, OK, USA Esra Güç  •  MRC Centre for Reproductive Health, Queen’s Medical Research Institute, The University of Edinburgh, Edinburgh Scotland, IL, USA Joshua D. Hall  •  Department of Mechanical and Industrial Engineering, University of Massachusetts, Amherst, MA, USA Natasha L. Harvey  •  Centre for Cancer Biology, University of South Australia and SA Pathology, Adelaide, SA, Australia Jasmin Herz  •  Department of Neuroscience, School of Medicine, Center for Brain Immunology and Glia, University of Virginia, Charlottesville, VA, USA Benjamin M. Hogan  •  Division of Genomics of Development and Disease, Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD, Australia Seon Pyo Hong  •  Center for Vascular Research, Institute for Basic Science, Daejeon, Republic of Korea Juan M. Jiménez  •  Department of Mechanical and Industrial Engineering, University of Massachusetts, Amherst, MA, USA Mark L. Kahn  •  Department of Medicine and Cardiovascular Institute, University of Pennsylvania, Philadelphia, PA, USA Sinem Karaman  •  Wihuri Research Institute and Translational Cancer Biology Program, Biomedicum Helsinki, University of Helsinki, Helsinki, Finland Richard G. Kibbey  •  Section of Endocrinology, Department of Internal Medicine, Yale University School of Medicine, New Haven, CT, USA Friedemann Kiefer  •  Max Planck Institute for Molecular Biomedicine, Münster, Germany; European Institute for Molecular Imaging (EIMI), University of Münster, Münster, Germany Witold W. Kilarski  •  Institute for Molecular Engineering, The University of Chicago, Chicago, IL, USA Jonathan Kipnis  •  Center for Brain Immunology and Glia, Department of Neuroscience, School of Medicine, University of Virginia, Charlottesville, VA, USA Gou Young Koh  •  Graduate School of Medical Science and Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea; Center for Vascular Research, Institute for Basic Science, Daejeon, Republic of Korea Antoine Louveau  •  Department of Neuroscience, Center for Brain Immunology and Glia, School of Medicine, University of Virginia, Charlottesville, VA, USA Taija Makinen  •  Department of Immunology, Genetics and Pathology, Uppsala University, Uppsala, Sweden

Contributors

xi

Ines Martinez-Corral  •  Department of Immunology, Genetics and Pathology, Uppsala University, Uppsala, Sweden Donald M. McDonald  •  Department of Anatomy, Helen Diller Family Comprehensive Cancer Center, Cardiovascular Research Institute, University of California, San Francisco, CA, USA Matthew T. Menendez  •  Cardiovascular Biology Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA Harri Nurmi  •  Wihuri Research Institute and Translational Cancer Biology Program, Biomedicum Helsinki, University of Helsinki, Helsinki, Finland Kazuhide S. Okuda  •  Division of Genomics of Development and Disease, Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD, Australia Brisa Palikuqi  •  Department of Regenerative Medicine, Weill Cornell Medical College, New York, NY, USA Intae Park  •  Graduate School of Medical Science and Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea Tatiana V. Petrova  •  Division of Experimental Pathology, Department of Oncology, CHUV, Ludwig Institute for Cancer Research, University of Lausanne, Epalinges, Switzerland; Swiss Institute for Cancer Research, EPFL, Lausanne, Switzerland Steven T. Proulx  •  Institute of Pharmaceutical Sciences, Swiss Federal Institute of Technology, ETH Zurich, Zurich, Switzerland Susan E. Quaggin  •  Feinberg Cardiovascular and Renal Research Institute, Northwestern University, Chicago, IL, USA; The Division of Nephrology/Hypertension, Northwestern University Feinberg School of Medicine, Chicago, IL, USA Shahin Rafii  •  Department of Regenerative Medicine, Weill Cornell Medical College, New York, NY, USA Anita Rogic  •  Department of Oncological Sciences and Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New York, NY, USA Amélie Sabine  •  Division of Experimental Pathology, Department of Oncology, CHUV, Ludwig Institute for Cancer Research, University of Lausanne, Epalinges, Switzerland Jocelynda Salvador  •  Department of Medical Pharmacology and Physiology, Dalton Cardiovascular Research Center, University of Missouri School of Medicine, Columbia, MO, USA Joshua P. Scallan  •  Department of Molecular Pharmacology and Physiology, University of South Florida, Tampa, FL, USA Aaron Scherzinger  •  Department of Computer Science, University of Münster, Münster, Germany Michael Simons  •  Department of Internal Medicine, Yale Cardiovascular Research Center, Yale University School of Medicine, New Haven, CT, USA; Department of Cell Biology, Yale University School of Medicine, New Haven, CT, USA Mihaela Skobe  •  Department of Oncological Sciences and Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New York, NY, USA Joo-Hye Song  •  Center for Vascular Research, Institute for Basic Science, Daejeon, Republic of Korea R. Sathish Srinivasan  •  Cardiovascular Biology Research Program, Oklahoma Medical Research Foundation, Oklahoma, OK, USA; Department of Cell Biology, University of Oklahoma Health Sciences Center, Oklahoma, OK, USA

xii

Contributors

Sang Heon Suh  •  Graduate School of Medical Science and Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea Melody A. Swartz  •  Institute for Molecular Engineering, The University of Chicago, Chicago, IL, USA; Ben May Department for Cancer Research, The University of Chicago, Chicago, IL, USA Daniel T. Sweet  •  GlaxoSmithKline, Target Incubator, Exploratory Discovery, Collegeville, PA, USA Benjamin R. Thomson  •  Feinberg Cardiovascular and Renal Research Institute, Northwestern University, Chicago, IL, USA; The Division of Nephrology/Hypertension, Northwestern University Feinberg School of Medicine, Chicago, IL, USA Michael Vanlandewijck  •  Karolinska Institutet/AstraZeneca Integrated Cardio Metabolic Centre (KI/AZ ICMC), Huddinge, Sweden; Department of Immunology, Genetics and Pathology, Rudbeck Laboratory, Uppsala University, Uppsala, Sweden John Welsh  •  Department of Medicine and Division of Cardiology, University of Pennsylvania, Philadelphia, PA, USA Pengchun Yu  •  Department of Internal Medicine, Yale Cardiovascular Research Center, Yale University School of Medicine, New Haven, CT, USA

Chapter 1 Three-Dimensional Visualization of the Lymphatic Vasculature Cathrin Dierkes, Aaron Scherzinger, and Friedemann Kiefer Abstract Like the circulatory blood vessel system, the dendriform lymphatic vascular system forms a disseminated organ that is virtually indispensible for the function of most other organs. Formation and maintenance of the correct topology are essential for lymph vessel physiology and hence analysis of its three-dimensional architecture provides crucial functional information. Here we describe protocols for whole-mount immunostaining of the vessel systems in various mouse tissues, mouse fetuses, and human skin biopsies. The resulting samples are suited after flat mounting for confocal microscopy or after optical tissue clearing for light sheet microscopy. Both microscopic modalities use optical sectioning to generate image stacks from which the three-dimensional vessel structure can be digitally reconstructed. We introduce the open software package Voreen, developed at the University of Münster. Voreen has been adapted and extended for the interactive visualization of large multichannel image stacks on commodity hardware, allowing for a faithful digital representation of the spatial structure of the vessel systems in whole-mount stained tissue samples. Key words Whole-mount staining, Vascular systems, Blood vessels, Lymph vessels, Light sheet microscopy, Interactive 3D rendering, Multiresolution octree

1  Introduction Lymphatic vessels form an organ of considerable size, which is central to the function of virtually all other organ systems. Therefore, understanding the development, biology and pathobiology of lymph vessels is of paramount importance [1]. Being a systemically disseminated organ, the spatial, three-­ dimensional structure of lymph vessels provides vital information about their functional state, however, due to its relative invisibility the lymph vessel system has so far eluted extensive structural investigation. This lack of investigatory and diagnostic possibilities has also been a major contributing factor to the failure to mechanistically understand lymphatic disorders and to develop effective drug and surgical therapies [2]. In research and clinical settings, lymphoscintigraphy

Guillermo Oliver and Mark L. Kahn (eds.), Lymphangiogenesis: Methods and Protocols, Methods in Molecular Biology, vol. 1846, https://doi.org/10.1007/978-1-4939-8712-2_1, © Springer Science+Business Media, LLC, part of Springer Nature 2018

1

2

Cathrin Dierkes et al.

and ICG lymphangiography, which are based on the injection, uptake, and transport of radioactive or fluorescent probes by the lymphatic vasculature, provide valuable information on lymphatic function and gross architecture, but do not inform on microanatomical and spatial details. A central difficulty in understanding the spatial organization of the lymphatic vessel system has arisen from the use of standard 2D tissue sections. Besides being labor intensive due to the requirement for staining, handling, and microscopy of large numbers of sections, three-­dimensional reconstruction of the lymphatic architecture from serial sections is computationally demanding due to tissue loss and distortion associated with the sectioning process [3]. Over the last decade planar illumination or light sheet microscopy experienced a brilliant renaissance and underwent rapid technological advancement [4, 5]. This technology is based on sample illumination with a thin sheet of light that is generated from a laser beam. Fluorescence emitted from this entire illuminated sample plane is detected orthogonally using a high resolution area detector, e.g., a sCMOS camera. By stepwise scanning of the sample, i.e., movement of the illumination plane through the sample, a light sheet microscope generates a contiguous series of optical sections from the specimen, which has to be either translucent or requires optical clearing. From these image stacks a nearly lossless representation of the spatial structure of the specimen can be generated by digital reconstruction, provided step size, axial and z-resolution are carefully matched. Therefore, light microscopy has been successfully used in experimental 3D reconstructions and analysis of developmental processes and the vascular systems [5–8]. We have developed a work flow that combines whole-mount staining of developmental or tissue specimen and subsequent clearing with the subsequent computer-assisted visualization of light sheet microscope-generated image stacks [9, 10]. Central to this approach is the identification of suitable antibodies and the development of whole-mount staining protocols using these antibodies that provide reliable identification of lymphatic vessels or the particular structural features of lymph vessels under investigation. Whole-mount preparations of thin specimen, like the dermis of the mouse ear or the mouse mesentery may be directly analyzed using a confocal microscope. Larger specimen need to be optically cleared for light sheet microscopy. Optical clearing is based on the extraction of lipids and a homogenization of the refractive index of the remaining tissue and the solvent of the sample. Numerous protocols have been developed over the last years for that purpose that are partly organic solvent-­based or partly water-based (for a direct comparison see [11]). For antibodystained specimen, tissue clearing using a mixture of benzyl alcohol and benzyl benzoate (BABB) following methanol dehydration/ delipidation described here is a widely applicable procedure.

Lymph Vessels in 3D

3

This BABB-based clearing protocol provides uniformly satisfying clearing results and stable samples that are resistant to microbial decomposition and can be reimaged for more than a year. The last decade has seen the appearance of a range of light sheet microscope designs dedicated to fluorescence microscopy of biological samples. This includes fully commercial “push button” instruments as well as detailed construction plans and kits for building custom microscopes. Imaging mouse fetal development or adult tissues entails the analysis of large sample volumes, which is facilitated by a large viewfield (i.e., small magnification objectives ideally with a high numerical aperture). For many years, we have successfully used a commercially available single lens light sheet microscope with bidirectional illumination (Ultramicroscope) by LaVision BioTec, (Bielefeld, Germany). This instrument allows the acquisition of sample volumes of close to a cubic centimeter with still cellular resolution. Because sample mounting and the actual imaging procedure are highly specific for the brand of microscope used and the proprietary software of the manufacturer, we have not included details on this part. Due to the substantial sample volumes imaged with cellular or in the latest setups even subcellular resolution, light sheet microscopy image stacks tend to become exceptionally large and consequently can be very slow in interactive rendering applications. To address this issue, we provide information on an open source software package for visualization, which is called Voreen (Volume Rendering Engine) [12]. Voreen is developed and maintained by the Department of Computer Science at the University of Münster, Germany, and has been especially adapted and extended to support interactive visualization of large multichannel image stacks on commodity hardware by using an octree data structure in combination with an OpenCL (Open Computing Language)-based raycaster [13]. In the octree data structure each node or point in space it represents corresponds to a data volume that is subdivided into eight octants. The Voreen framework has already been used in different contexts for visualizing and analyzing light sheet fluorescence microscopy data [9, 14]. Image stacks are imported into Voreen using the OME-TIFF format, which is directly provided by the image acquisition software of most, if not all manufacturers. Voreen then generates a multiresolution octree data structure based on the original images. The octree representation of the data is cached on the hard drive so that the conversion only needs to be computed once as a preprocessing step. The software provides an interactive 3D rendering as well as three axis-aligned slice views (one in the direction of each main axis) to interactively explore the data. Moreover, the volume dataset can be clipped using both axisaligned and arbitrarily oriented clipping planes to cut away parts of the image stack. Color maps for all channels can be edited individually. Special tools allow the efficient compensation of color

4

Cathrin Dierkes et al.

channel offsets and the localization of selected positions in the three axis-aligned slice views in the 3D visualization. Moreover, Voreen allows to generate movies by animating the camera, clipping planes, and other parameters of the visualization in multiple ways.

2  Materials 1. Ice bucket. 2. Petri dish. 3. 70% ethanol. 4. PBS (ice cold). 5. Two pairs of fine forceps (Dumont #5). 6. Blunt forceps. 7. Scissors. 8. Ophthalmic scissors. 9. Fine perforated spoon. 10. Petri dish with 5–8 mm layer of Sylguard 184 (Dow corning). 11. Insect pins. 12. Scalpel blade. 13. Mounting medium based on Mowiol 4.88. 14. Nail polish. 15. Workstation for image visualization: minimally recommended system configuration, quad core processor, two PCIe SSDs (500GB) (Peripheral Component Interconnect Express Solid State Drives), 128GB RAM, current generation graphics card (e.g., NVIDIA GeForce GTX 1080Ti), 20 TB hard drive space. 2.1  Buffers and Solutions

1. PBS (Phosphate buffered saline): 8 g/L NaCl, 0.2 g/L KCl, 1.44 g/L Na2HPO4, 0.24 g/L KH2PO4, pH 7.4. 2. 4% PFA: 4% paraformaldehyde dissolved in PBS, pH 7.4. 3. PBST: 0.1%(v/v) Tween®20 dissolved in PBS). 4. Permeabilization buffer: 0.5% Triton X-100 dissolved in PBS. 5. Permblock Solution: 1%(w/v) bovine serum albumin (BSA), 0.5%(v/v) Tween®20 dissolved in PBS. 6. Methanol (anhydrous). 7. Benzyl benzoate (anhydrous). 8. Benzyl alcohol (anhydrous).

Lymph Vessels in 3D

5

9. low melting agarose: Add 1 g of low melting agarose slowly into 100 mL of ultrapure water, while stirring. Heat the suspension in a microwave oven until the agarose is completely dissolved. Before use, allow the solution to cool, however, not below 40 °C. 10. BABB: To prepare 900 mL of BABB mix 300 mL of benzyl alcohol and 600 mL of benzyl benzoate in a glass vial.

3  Methods 3.1  Preparation of Tissues and Whole-­ Mount Staining

3.1.1  Ear (Dermis)

Before starting, prepare at your workspace an ice bucket with petri dishes containing PBS and a multiwell plate containing 4% PFA. Dilutions or working concentrations for the different primary and secondary antibodies are listed in Table 1. 1. Sacrifice mouse. 2. Cut the ears above the cartilage of the auricle. 3. Under inspection with a stereomicroscope, use two fine forceps to tear apart the dorsal and ventral half of the ear skin. Dispose the ventral sheet and keep the dorsal half. 4. Fix the dermis in 4% PFA for 30 min at room temperature. Replace the fixative and cover the ear with Permeabilization Buffer, incubate for 24 h at 4 °C with gentle shaking. 5. Wash the sample 3× with PBS at room temperature, allow 20 min for washing each step, gently shaking. 6. Cover the sample with Permblock solution and incubate for 24 h at 4 °C, gently shaking. 7. Incubate the sample in primary antibody diluted in Permblock solution at 4 °C for 3 days. 8. Wash 3× with PBST, allow 1 h for each step. Incubate the sample in secondary antibody diluted in Permblock solution at 4 °C for 3 days. 9. Wash 3× with PBST, allow 1 h for each step. 10. Mount the split ear sample in Mowiol on a glass slide with the dermis facing upward, be careful not to enclose air bubbles when covering with a coverslip. 11. Let the slides dry in the dark for at least 1 day, then seal the coverslip with nail polish.

3.1.2  Trachea

1. Sacrifice mouse and spray fur with 70% ethanol. 2. Open the abdomen and cut thorax up to the throat. 3. Remove the ribcage and cut the end of the trachea above the bronchial bifurcation.

6

Cathrin Dierkes et al.

Table 1 Antibodies and dilutions: To visualize the blood and lymphatic vasculature in tissue whole-mount preparations, the following antibodies and final dilutions in Permblock solution were used

Antigen

Dilution/final concentration

Species

Company

hLyve1

pAb goat IgG

R&D Systems, Minneapolis,USA (#AF2089)

1:200

mLyve1

pAb rabbit IgG

ReliaTech, Wolfenbüttel (#103-PA50)

1:250

mAb mouse IgG

Dako

1:200

hProx-1

pAb goat IgG

R&D Systems, Minneapolis,USA (#AF2727)

1:200

hProx-1

pAb rabbit IgG

ReliaTech, Wolfenbüttel (#102-PA32)

1:200

hPDPN

Clone

D2-­40

hPecam1

89C2

mAb mouse IgG

Cell Signaling Technologies (#3528)

1:200

mPecam-1

5D2.6

mAb rat IgG2a

S.Butz, MPI Münster

20 μg/mL

mPecam-1

1G5.2

mAb rat IgG2b

S.Butz, MPI Münster

20 μg/mL

pAb goat IgG

R&D systems, Minneapolis,USA (#AF743)

1:125

mVEGFR-3

Antigen

Conjugation

Species

Company

Dilution

Goat IgG (H + L)

Alexa Fluor 488

Donkey IgG

Invitrogen, Karlsruhe

2 μg/mL

Goat IgG (H + L)

Alexa Fluor 568

Donkey IgG

Invitrogen, Karlsruhe

2 μg/mL

Goat IgG (H + L)

Alexa Fluor 647

Donkey IgG

Invitrogen, Karlsruhe

2 μg/mL

Hamster IgG (H + L)

Alexa Fluor 488

Goat IgG

Invitrogen, Karlsruhe

2 μg/mL

Rabbit IgG (H + L)

Alexa Fluor 488

Donkey IgG

Invitrogen, Karlsruhe

2 μg/mL

Rabbit IgG (H + L)

Alexa Fluor 568

Donkey IgG

Invitrogen, Karlsruhe

2 μg/mL

Rabbit IgG (H + L)

Alexa Fluor 647

Donkey IgG

Invitrogen, Karlsruhe

2 μg/mL

Rat IgG (H + L)

Alexa Fluor 488

Donkey IgG

Invitrogen, Karlsruhe

2 μg/mL

Rat IgG (H + L)

Alexa Fluor 568

Donkey IgG

Invitrogen, Karlsruhe

2 μg/mL

Rat IgG (H + L)

Alexa Fluor 647

Chicken IgG

Invitrogen, Karlsruhe

2 μg/mL

Sheep IgG (H + L)

Alexa Fluor 488

Donkey IgG

Invitrogen, Karlsruhe

2 μg/mL

Sheep IgG (H + L)

Alexa Fluor 568

Donkey IgG

Invitrogen, Karlsruhe

2 μg/mL

Lymph Vessels in 3D

7

4. Using ophthalmic scissors and fine forceps, carefully lift the trachea upward and separate the esophagus. 5. Cut the trachea at the larynx and place it into a petri dish filled with PBS. 6. Using a stereomicroscope, carefully dissect away the connective tissue around the trachea and remaining cartilage of the larynx. 7. Pin the stretched trachea to the bottom of the dish using the insect pins. 8. For observation using a confocal microscope, cut the trachea alongside and pin the flattened trachea to the dish with the inside facing the bottom of the dish. For light sheet microscopy leave trachea intact. 9. Fix the trachea with 4% PFA for 30 min at room temperature. 10. Wash the trachea 3× in PBS, allow 10 min for each washing step. 11. Cover the trachea with Permeabilization Buffer and incubate for 12 h at 4 °C. 12. Discard the Permeabilization Buffer and wash the trachea 3× with PBS, wash for 10 min each step. 13. Transfer the trachea into Permblock solution and incubate for 12 h at 4 °C. 14. For staining, incubate the trachea in primary antibody diluted in Permblock solution at 4 °C for 2 days. 15. Wash 3× with PBST, allow 1 h for each wash step. 16. Incubate the trachea in fluorophore-coupled secondary antibody, diluted in Permblock solution at 4 °C in the dark for 2 days. 17. Wash 3× with PBST, allow 1 h for each wash step. 18. For confocal microscopy place the flattened trachea on a slide into a drop of MOWIOL and cover with a coverslip. 19. For light sheet microscopic observation, place the immunostained trachea into a dish with 1% low melting agarose. 20. Let the agarose cure, then use a scalpel to cut the agarose around forming a block containing the trachea. 21. Pour PBS on top of the agarose before taking out the rectangular block. 22. Put the block into a glass vial and continue to clear the tissue (see Subheading 3.5).

8

Cathrin Dierkes et al.

3.1.3  Mesentery

1. Sacrifice mouse and spray fur with 70% ethanol. 2. Open the abdomen. 3. Cut the small intestine, avoiding the pancreas and remove the colon. 4. Put the intestine into petri dish filled with a silicon layer and covered with PBS. 5. Pin the entire length of the intestine with insect pins and thereby unravel the mesentery. 6. Discard the PBS and replace it with 4% PFA to fix the intestine for 2 h at room temperature. 7. Wash the intestine 3× in PBS, allow 20 min for each step. 8. Use a sharp scalpel blade to separate the mesentery from the gut. 9. Dispose the gut and put the mesentery into a clean dish. 10. Cover the mesentery with Permeabilization Buffer and incubate for 24 h at 4 °C, gently shaking on a tumbling shaker. 11. Wash the mesentery 3× with PBS, 20 min each step, with gentle shaking. 12. Cover the mesentery with Permblock solution and incubate for 24 h at 4 °C, gently shaking. 13. Incubate the mesentery in primary antibody diluted in Permblock solution at 4 °C for 3 days. 14. Wash 3× with PBST, 1 h each step. 15. Incubate the mesentery in secondary antibody diluted in Permblock solution at 4 °C for 3 days. 16. Wash 3× with PBST, 1 h each step. 17. Mount the mesentery in Mowiol on a slide, be careful not to enclose air bubbles when covering with a coverslip.

3.1.4  Intestinal Villi

1. Sacrifice mouse and spray fur with 70% ethanol. 2. Open the abdomen. 3. Dissect the small intestine, avoiding to damage the pancreas and subsequently remove the colon. 4. Put the intestine into a petri dish with PBS. 5. Cut a small section of the small intestine. 6. Remove any remaining parts of the mesentery. 7. Open the section longitudinally and pin it stretched and flattened out to a silicon coated dish using insect pins. 8. Fix the tissue with 4% PFA for 1 h at room temperature. 9. Discard the PFA and wash 3× with PBS, 10 min each step.

Lymph Vessels in 3D

9

10. Using a stereomicroscope, use fine forceps and ophthalmic scissors to cut individual rows of villi. 11. Incubate these rows in single wells of a 96-well-dish with Permeabilization Buffer for 6 h at room temperature. 12. Remove the Permeabilization Buffer with a pipette and wash the villi with PBS. Do so by very gently aspirating and expelling the buffer using a pipette with 200 μL disposable tip. Take care not to aspirate the tissue. 13. Fill the wells with Permblock solution and incubate the villi overnight at 4 °C. 14. Gently remove the Permblock Solution and refill the well with primary antibodies diluted in Permblock solution. Allow the staining to proceed for 3 h. 15. Wash the villi with PBST 3×, 20 min for each step. 16. Carefully pipet up and down the buffer to rinse the sample when changing the washing buffer. 17. Refill the wells with secondary antibodies diluted in Permblock solution. Stain for 3 h in the dark. 18. Again wash the villi with PBST 3×, 20 min each step. 19. Again carefully rinse the sample by pipetting up and down the wash buffer. 20. For confocal observation, use fine forceps to gently immerse rows of villi in a drop of Mowiol and transfer to a microscope slide, with the cut surface of the intestinal wall facing up and put a coverslip on top. 23. For light sheet microscopic observation, place the immunostained intestinal sample into a dish with 1% low melting agarose. 24. Let the agarose cure, then use a scalpel to cut the agarose around forming a block containing the intestinal sample. 25. Pour PBS on top of the agarose before taking out the rectangular block. 26. Put the block into a glass vial and continue to clear the tissue (see Subheading 3.5). 3.2  Preparation of Mouse Developmental Stages

Before starting, prepare at your workspace a full ice bucket with petri dishes containing PBS and a multiwell plate containing 4% PFA. 1. Sacrifice pregnant female mouse from a timed mating. 2. Spray fur with 70% ethanol and open up the abdomen. 3. Hold the uterus with blunt forceps and cut the oviduct on one side below the ovary.

10

Cathrin Dierkes et al.

Table 2 Incubation times for fixation and permeabilization of various developmental stages of the mouse Age of embryo/fetus (days post fertilization)

Fixation time

Permeabilization and blocking time

9.5

20 min

24 h

10.5

30 min

24 h

11.5

45 min

24 h

12.5

1 h

48 h

13.5

1.5 h

48 h

14.5

2 h

48 h

4. Pull the uterus toward you and slice it open alongside, avoid damaging the embryo and the membranes ensheathing them. Carefully scoop out the embryo proper with a fine perforated spoon and place it into ice-cold PBS. 5. Use a stereomicroscope to dissect the yolk sac, liberate the embryo and cut the umbilical cord to remove the placenta. 6. Clip the tip of the tail or a limp to obtain a sample for genotyping. 7. Place each embryo into a single well with 4% PFA for fixation. 8. Fix embryo at room temperature for 20 min up to 2 h, times depend on size (see Table 2). 9. Wash embryo 3× with PBS, 10 min each wash with gentle shaking. 10. Place the embryo in Permeabilization Buffer for 24–48 h at 4 °C (see Table 2), gently shaking. 11. Wash embryo 3× with PBS, 10 min each wash with gentle shaking. 12. Place the embryos in Permblock solution for 24–48 h at 4 °C, gently shaking. 13. Put the embryo into small wells (see Note 1), filled with primary antibody diluted in Permblock solution and incubate for 7 days at 4 °C (see Note 2). 14. Wash the embryos in larger wells (e.g., for E9.5 to E12.5 use a 24-wells-dish, for larger embryos use a 12-well-dish) with PBST, gently shaking. 15. Wash over 2 days, change the PBST every 2 h.

Lymph Vessels in 3D

11

16. Put the embryos into small wells (see Note 1) filled with secondary antibody diluted in Permblock solution and incubate for 7 days at 4 °C (see Note 2). 17. Wash the embryos in larger wells with PBST, gently shaking. 18. Wash for 2 days, change the PBST every 2 h. 19. For observation in an ultramicroscope, place the immunostained embryos into a dish with 1% low melting agarose. 20. Let the agarose cure, then use a scalpel to cut a rectangular block around the embryo (see Note 3). 21. Put the block into a glass vial and continue to clear the tissue (see Subheading 3.5). 3.3  Preparation of Fetal Skin

1. Sacrifice pregnant mice from timed matings as described in Subheading 3.2. For the preparation of skin, use fetuses between E13.5 and E16.5. 2. Fix fetus in 4% PFA for 4 h at room temperature. 3. Wash fetus 3× with PBS, 10 min each, gently shaking. 4. Put the fetus into a new dish filled with PBS. 5. Clip the limbs and tail. 6. Hold the fetus with blunt forceps to avoid damage to the skin. 7. Use ophthalmic scissors to create a deep incision in the fetus from tail to head on the abdominal side. The incision must be deep enough to cut through the ribcage and the facial structures. Open up the fetus and hold it open with blunt scissors. 8. Take out the brain and the inner organs. 9. Using ophthalmic scissors, cut underneath the spine and take out the whole backbone. 10. Carefully using ophthalmic scissors cave out the complete fetus. 11. Transfer the skin into a new well with Permeabilization Buffer and incubate at 4 °C overnight. 12. Wash the skin 3× with PBS, 10 min each step while gently shaking. 13. Discard the PBS and incubate the skin in Permblock solution at 4 °C overnight. 14. Cover the skin with primary antibody diluted in Permblock solution and incubate for 4 days at 4 °C. 15. Wash the skin in a 10 cm petri dish with PBST, with gentle agitation. 16. Wash for 1 day and change the PBST every 2 h. 17. Cover the skin with fluorophore coupled secondary antibody diluted in Permblock Solution and incubate for 4 days at 4 °C.

12

Cathrin Dierkes et al.

18. Wash the skin in a large dish with PBST, with gentle agitation. 19. Wash for 1 day, change the PBST every 2 h. 20. For observation using a confocal microscope, prepare a flatmount of the skin. To do so, add a large drop of Mowiol on a microscope slide. 21. Use blunt forceps to transfer the stained skin onto the slide. 22. Stretch out the skin, the epidermis is facing the bottom of the slide. 23. Carefully cover the skin with a coverslip and let the mounting medium settle at room temperature in the dark. 24. For observation using a light sheet microscope, transfer the antibody-stained skin into a dish with 1% low melting agarose. 25. Once submerged, the skin will adopt its original shape. 26. Let the agarose cure, then use a scalpel to cut a rectangular block around the skin (see Note 3). 27. Put the block into a glass vial and continue to clear the tissue (see Subheading 3.5). 3.4  Preparation of Human Skin Biopsies

1. Fix fresh biopsies of human skin in 4% PFA for 4 h at room temperature. 2. Wash samples 3× with PBS, 10 min each, with gentle agitation on a tumbling shaker. 3. Transfer the skin into a new well with Permeabilization Buffer and incubate at 4 °C for 2 days. 4. Wash the skin 3× with PBS, allow 10 min for each step with gentle agitation. 5. Discard the PBS and incubate the skin in Permblock solution at 4 °C for 2 days. 6. Incubate the skin samples in primary antibody diluted in Permblock Solution at 4 °C for 1 week. 7. Wash the biopsies on a tumbling shaker in PBST 1 day, change the washing buffer every hour. 8. Incubate the skin in secondary antibody diluted in Permblock solution at 4 °C for 1 week in the dark. 9. Wash the biopsies on a tumbling shaker in PBST 1 day, change the washing buffer every hour. 10. Upon completion of the incubation, transfer the antibody-­ stained skin into a dish with 1% low melting agarose. 11. Let the agarose cure, then use a scalpel to cut a cuboidal block around the skin (see Note 3). 12. Put the block into a glass vial and continue to clear the tissue (see Subheading 3.5).

Lymph Vessels in 3D

3.5  Clearing of Tissues with BABB to Prepare them for Light Sheet Microscopy

13

1. Perform all incubation steps in the dark. 2. For the clearing steps only use glass vials or polypropylene tubes, BABB dissolves polystyrene. 3. Wash the agarose blocks twice with PBS. 4. Dehydrate the samples by incubation in an ascending methanol series (50%, 70%, 95%, 2×>99.5% (v/v)) while gently shaking. 5. Be sure to carefully remove all fluid of the previous step before adding the next solution in the series (see Note 4). 6. Incubation times until each dehydration step is complete depend on the size of the sample-containing blocks, at least 30 min should be allowed for each step. Average-sized blocks take 1 h; confirm that there is no longer “Schlieren” formation detectable before you consider a dehydration step complete (Schlieren here describe the typical streaks appearing at the interface of solutions of different density or refractory index). 7. After dehydration, place the sample into a 1:1 mixture of 100% methanol and BABB for at least 5 h. 8. Finally, incubate the agarose block in 100% BABB overnight. Store the blocks in fresh BABB solution at 4 °C in the dark. Keep the samples in BABB during microscopic observation. Be sure not to spill the BABB, as it is corrosive and might damage surface coatings, adhesive joints and the electronic components of the microscope.

3.6  Visualization of Light Sheet Microscope-Generated Image Stacks in Voreen 3.6.1  Software Installation and Preliminary Steps

1. Download the Voreen source code and follow the build instructions at “voreen.uni-muenster.de”. Voreen currently supports Linux (recommended) and Windows platforms, Mac OS is presently not supported. When working in a Windows environment, you can also download a prebuilt version of Voreen. 2. Start the application and select a workspace, i.e., a Voreen project. For the purpose of visualizing light sheet data, you should open the “ultramicroscopy.vws” file that is included with Voreen. 3. When starting Voreen for the first time, it is advisable you configure the settings for the octree cache (Settings -> General Settings). Here you can choose the “Cache Directory”, which is the location where Voreen stores the octree data structures, and the “Octree Cache Size”, which is the amount of memory on the hard drive that Voreen may use for the octree caching (see Note 5). 4. Select Voreen’s application mode to switch into a more streamlined interface of the workspace (View -> Application Mode or press F4). You only have to do this when first using Voreen— the next time you start the application, Voreen will already select the application mode by default.

14

Cathrin Dierkes et al.

5. Select your image data by using the “Data Input” menu on the right-hand side of the Voreen application. Each channel of your image stack needs to be loaded individually (hence the four file dialog properties, as the octree raycaster supports up to four channels, see Note 6). 6. When loading your data for the first time, click the “Generate Octree” button. If you have already generated an octree representation for this data previously, the button changes to “Load Octree” and directly allows loading the cached octree data structure. 7. You will now see four rendering views in the central quad view. The upper left view shows a 3D rendering and the other three views show axis-aligned slice views along each of the coordinate axes. You can enlarge each of the views by double-clicking on it. By double-clicking again, you can go back to the standard quad view. See Fig. 1 for a workspace with one color channel loaded. 3.6.2  Interactive Visualization Routine

You can now interact with the image stack in various ways. The following steps do not have to be performed in any particular order, but can be applied as required to interactively explore the data.

Fig. 1 Voreen workspace for the visualization of light sheet microscope-generated image stacks with a single color channel loaded. The top left panel represents the 3D visualization, while the bottom left and two right panels show the corresponding 2D slice views with the respective axis indicated by the colored arrows at the right bottom. The right hand panel shows the property groups that allow control of the parameters for the digital rendering

Lymph Vessels in 3D

15

1. To adjust the rendering settings and color maps for each channel use the property groups on the right-hand side of the application. Select “3D Rendering” to change the 3D visualization and select “2D Slice Views” for settings related to the axis-aligned slice views. 2. In the 3D view, hold the left mouse button and move the mouse in order to rotate the view of the data. Hold the right mouse button and move the mouse or use the mouse wheel to change the distance of the camera to the data (i.e., “zoom in or out”). When holding down the Shift button while holding the left mouse button, you can Shift the camera. By holding down the Control key, you can initiate a continuous rotation. To reset the camera shift, click on the “Camera” button in the “3D Rendering” property group on the right side, then click on “Reset Camera Focus to Trackball Center”. 3. In the 2D slice views, use the mouse wheel or hold down the middle mouse button (or mouse wheel) and move the mouse in order to traverse the image stack. Hold down the Control button while holding down the middle mouse button and move the mouse in order to zoom in or out. Hold down the Control button while holding down the left mouse button and move the mouse to shift the view. To reset the shift and zoom, click on the “Reset Shift and Zoom” button in the “2D Slice Views” property group on the right-hand side of the application. 4. For both the 3D rendering and the 2D slice views, you can select to show the orientation overlays (i.e., coordinate systems) by selecting the corresponding check boxes in the “3D Rendering” and “2D Slice Views” property groups. Here, you can also find some more options such as showing the bounding box of the image stack in the 3D rendering, or visualizing the positions of the 2D slices in the 3D view. 5. In order to crop the image stack in the 3D rendering to explore specific structures of interest, select the “Enable” checkbox in the “Axis-aligned Clipping” property group. You can either change the size of the region of interest by using the sliders provided in the property group or by directly interacting with the on-screen handles in the 3D rendering view. 6. In addition to the axis-aligned cropping, you can use three clipping planes with arbitrary orientation using the “Clipping Plane 1-3” property groups. To enable a clipping plane, select the “Enable” checkbox in the corresponding property group. You can then interact with the clipping plane via the on-screen handle in the 3D rendering view. To change the orientation of the plane, hold the left mouse button after clicking on the spherical end of the handle and move the mouse. To move the plane along the specified axis, hold the left mouse button after

16

Cathrin Dierkes et al.

clicking on the shaft of the handle and move the mouse. To change the position of the handle on the clipping plane, hold the right mouse button after clicking on the handle and move the mouse. You can also deactivate the rendering of the handle and plane indicator by using the “Show Plane” checkbox in the property group. 7. Depending on the mode of acquisition (single track versus multi track) and through chromatic aberration, color channels may be recorded with a significant offset in the image stacks. You can use the “Channel Shift Correction” property group on the right-hand side of the application to correct the offset. Change the channel shift for each channel by using the sliders provided in the property group and inspect the results in the three axis-aligned slice views until the channels match in all dimensions. 8. You can export your data by using the “Save Clipped Volume or Octree” property group. You can either select the “VVD” format, which stands for “Voreen Volume Data” and is a Voreen-specific data format. Using this format, each channel will be exported separately to two files, a “.raw” file containing the actual image data, and a “.vvd” file containing meta data. You can also choose to export the image stack to the HDF5 format, which is compatible with tools such as Fiji or Ilastik. In both cases, you can choose to directly apply the axis-aligned clipping (i.e., cropping) and the channel shift correction to the exported version. Note that the arbitrarily oriented clipping planes will not be considered during export. 9. There is also the option to export the image stack in an octree format. This can be helpful to store an octree for later use when it might be removed from the octree cache due to the intermediate octree generation for other image stacks, although the octree export has some limitations (see Note 7).

4  Notes 1. Choose the multiwell dish depending on the size of the embryo to expend minimal amounts of antibody solution. For E9.5 and E10.5 embryos a 96-well dish is suitable, E11.5 and E12.5 fetuses require a 48-well-dish, E13.5 and E14.5 a 24-well-dish. 2. Be careful to not let the specimen dry out during the staining! If you use small volumes of fluid, e.g., in a 96-well-dish, seal the dish with Parafilm. 3. Make sure the agarose is covered with PBS before easing out the rectangular block containing the specimen, to prevent the tissue from slipping out of the agarose.

Lymph Vessels in 3D

17

4. Avoid handling of the sample as it gets brittle during the process. 5. The octree data structure for a 16-bit image stack is about 50% larger than the original stack in uncompressed form due to the multiresolution representation. Voreen will delete the least recently used cached octree from the hard drive once the set cache size is exceeded to stay within the maximum cache size. Once an octree is removed from cache, it needs to be computed from scratch when working with the dataset at a later point. Depending on the size of the image stack, this process might take considerable time. It is advisable to use a PCIe SSD (Peripheral Component Interconnect Express Solid State Drive) for the octree cache (octree storage site) for better rendering performance when visualizing large image stacks. 6. Voreen supports different file formats such as HDF5 (e.g., for interaction with Fiji or similar software). For microscopy image stacks, the OME-TIFF format is a good data format. Note that when storing your data in this format, it is advisable to select the option to write the header (containing meta data such as referencing the list of image files) into each of the individual files. 7. The octree representation stores details about your system’s hardware such as the amount of main memory that is available for the octree visualization, which may cause problems when trying to visualize the exported octree on a different machine, which has less main memory. Moreover, exporting the octree representation for an image stack does not allow to incorporate the cropping or channel shift correction into the data export. References 1. Tammela T, Alitalo K (2010) Lymphangiogenesis: molecular mechanisms and future promise. Cell 140(4):460–476. https://doi.org/10.1016/j.cell.2010.01.045 2. Mortimer PS, Rockson SG (2014) New developments in clinical aspects of lymphatic disease. J Clin Invest 124(3):915–921. https:// doi.org/10.1172/JCI71608 3. Jones AS, Milthorpe BK, Howlett CR (1994) Measurement of microtomy-induced section distortion and its correction for 3-dimensional histological reconstructions. Cytometry 15(2):95–105. https://doi.org/10.1002/ cyto.990150203 4. Mertz J (2011) Optical sectioning microscopy with planar or structured illumination. Nat

Methods 8(10):811–819. https://doi. org/10.1038/nmeth.1709 5. Keller PJ, Dodt HU (2012) Light sheet microscopy of living or cleared specimens. Curr Opin Neurobiol 22(1):138–143. https://doi. org/10.1016/j.conb.2011.08.003 6. Lavina B, Gaengel K (2015) New imaging methods and tools to study vascular biology. Curr Opin Hematol 22(3):258–266. https:// doi.org/10.1097/MOH.000000000 0000141 7. Pollmann C, Hagerling R, Kiefer F (2014) Visualization of lymphatic vessel development, growth, and function. Adv Anat Embryol Cell Biol 214:167–186. https://doi.org/10.1007/ 978-3-7091-1646-3_13

18

Cathrin Dierkes et al.

8. de Medeiros G, Balazs B, Hufnagel L (2016) Light-sheet imaging of mammalian development. Semin Cell Dev Biol 55:148–155. https:// doi.org/10.1016/j.semcdb.2015.11.001 9. Hagerling R, Drees D, Scherzinger A, Dierkes C, Martin-Almedina S, Butz S, Gordon K, Schafers M, Hinrichs K, Ostergaard P, Vestweber D, Goerge T, Mansour S, Jiang X, Mortimer PS, Kiefer F (2017) VIPAR, a quantitative approach to 3D histopathology applied to lymphatic malformations. JCI Insight 2:16. https://doi.org/10.1172/jci. insight.93424 10. Hagerling R, Pollmann C, Andreas M, Schmidt C, Nurmi H, Adams RH, Alitalo K, Andresen V, Schulte-Merker S, Kiefer F (2013) A novel multistep mechanism for initial lymphangiogenesis in mouse embryos based on ultramicroscopy. EMBO J 32(5):629–644. https:// doi.org/10.1038/emboj.2012.340

11. Orlich M, Kiefer F (2017) A qualitative comparison of ten tissue clearing techniques. Histol Histopathol 33(2):181–199. https://doi. org/10.14670/HH-11-903 12. Meyer-Spradow J, Ropinski T, Mensmann J, Hinrichs K (2009) Voreen: a rapid-prototyping environment for ray-casting-based volume visualizations. IEEE Comput Graph Appl 29(6):6–13. https://doi.org/10.1109/ MCG.2009.130 13. Brix T, Praßni J-S, Hinrichs KH (2014) Visualization of large volumetric multi-channel microscopy data streams on standard PCs. CoRR abs/1407.2074 14. Scherzinger A, Kleene F, Dierkes C, Kiefer F, Hinrichs KH, Jiang X (2016) Automated segmentation of immunostained cell nuclei in 3D ultramicroscopy images. Paper presented at the Pattern Recognition GCPR 2016, Lecture Notes in Computer Science

Chapter 2 Histological and Morphological Characterization of Developing Dermal Lymphatic Vessels Kelly L. Betterman and Natasha L. Harvey Abstract The capacity to visualize the lymphatic vasculature in three-dimensions has revolutionized our understanding of the morphogenetic mechanisms important for constructing the lymphatic vascular network during development. Two complementary approaches are commonly employed to assess the function of genes and signaling pathways important for development of the dermal lymphatic vasculature in the mouse embryo. The first of these is whole-mount immunostaining of embryonic skin to analyze dermal lymphatic vessel network patterning and morphology in two and three dimensions. The second is immunostaining of thin tissue sections to examine lymphatic vessel identity, lumen formation and protein localization within discrete lymphatic endothelial cells in a two-dimensional setting. Here we present detailed protocols for multicolor immunofluorescent immunostaining of embryonic dorsal skin and thin tissue cryosections. Each of these methods generates high-resolution images of the dermal lymphatic vasculature, yielding information integral to in-depth characterization of lymphatic vessel phenotypes in the developing mouse embryo. Key words Lymphangiogenesis, Skin, Dermal lymphatic vessels, Embryonic skin, Immunostaining, Whole-mount immunostaining, Immunofluorescence, Confocal microscopy

1  Introduction Advances in high-resolution microscopy have provided unprecedented insight to the cellular and morphogenetic events important for building the vasculature during development. Mouse tissues including the postnatal retina [1], embryonic mesentery [2], small intestine [3], mammary gland [4], and skin [5] are highly amenable to flat-mounting and imaging using high-resolution confocal microscopy and, as such, are routinely employed for the assessment of gene function in developmental angiogenesis and lymphangiogenesis. A key tool for the analysis of vascular phenotypes in genetically modified mice in our laboratory and others is whole-mount immunostaining of embryonic dorsal skin. The dermal lymphatic vasculature in the mouse is a highly organized, arborized network that

Guillermo Oliver and Mark L. Kahn (eds.), Lymphangiogenesis: Methods and Protocols, Methods in Molecular Biology, vol. 1846, https://doi.org/10.1007/978-1-4939-8712-2_2, © Springer Science+Business Media, LLC, part of Springer Nature 2018

19

20

Kelly L. Betterman and Natasha L. Harvey

develops in a stepwise fashion [6]. At embryonic day (E) 12.5, generally the earliest time point we analyze the skin using this technique, lymphatic endothelial cells begin to invade the anterior dorsal skin from either side of the embryo. Two days later (E14.5), the lymphatic vascular network has rapidly expanded by sprouting lymphangiogenesis and covers the majority of the dorsal skin, but in general, has not yet reached the embryonic midline. Between E15.5 and E16.5, lymphatic vessels reach the dorsal midline and anastomose to form an integrated network. At this point in time, valve forming regions become apparent. By imaging the lymphatic vasculature using high-resolution confocal microscopy, we can visualize this stereotypical process and dissect gene function in cell processes including sprouting, migration, proliferation, lumen formation, and junctional integrity. The measurement of parameters including vessel caliber, vessel length, branch points, filopodial number and length, tip cell morphology, endothelial cell proliferation, endothelial cell identity, and valve formation provides detailed insight to the mechanisms by which genes and signaling pathways of interest direct lymphangiogenesis. Here we present a step-by-step protocol for multicolor immunofluorescent whole-mount immunolabeling of embryonic dorsal skin to enable the visualization of blood and lymphatic vessel networks in the mouse embryo between E12.5 and E18.5. We also provide a complementary protocol for multicolor immunofluorescent immunostaining of thin tissue cryosections, enabling thorough characterization of lymphatic vessel lumen formation and protein localization within discrete lymphatic endothelial cells; detail that is occasionally masked in three-dimensions. In our experience some antibodies yield far superior staining; we highlight here the sources of antibodies that have proven most robust in our hands for both purposes. The techniques described in these two simple protocols can be performed by anyone who is comfortable with handling mice and has skills in mouse dissection and confocal microscopy.

2  Materials Prepare all solutions using ultrapure water (prepared by purifying deionized water to attain a sensitivity of 18 MΩ cm at 25 °C) and analytical grade reagents. Unless specified, graduated cylinders are not necessary to measure solution volumes. Prepare and store all reagents at room temperature (unless indicated otherwise). Use a rocking platform mixer for all steps requiring gentle agitation at room temperature or 4 °C. 2.1  Reagents

1. Phosphate-buffered saline (PBS; 10×): 2.0 g/L KCl, 2.0 g/L KH2PO4, 80.0 g/L NaCl, 11.5 g/L Na2HPO4 (see Note 1).

Tools to Characterise the Dermal Lymphatic Vasculature

21

2. Paraformaldehyde (PFA): 4% (w/v) solution in 1× PBS (see Notes 2–5). 3. Thimerosal: 1% (w/v) solution in 1× PBS (see Note 6). 4. Triton X-100: 10% (v/v) solution in 1× PBS (see Note 7). 5. Blocking solution: 1× PBS containing 1% (w/v) bovine serum albumin (BSA) and 0.3% (v/v) Triton X-100 (see Note 8). 6. Wash solution: 1× PBS containing 0.1% (v/v) Triton X-100 (see Note 9). 7. Sucrose: 30% (w/v) solution in 1× PBS (see Note 10). 8. Tris-buffered saline (TBS; 20×): 3 M NaCl, 0.4 M Tris (pH 7.4). 9. Tween-20: 0.1% (v/v) solution in 1× TBS (TBS-T) (see Note 11). 10. Primary antibodies (see Table 1). 11. Alexa Fluor® conjugated secondary antibodies. 12. DAPI Fluoromount G™ mounting medium. 13. O.C.T.  Compound. 2.2  Equipment

1. Dissection equipment (e.g., forceps, scissors, spatula). 2. Petri dishes. 3. Dissecting microscope. 4. Rocking platform mixer. 5. Multiwell, flat-bottom tissue culture plates (e.g., 24-well format). 6. Parafilm. 7. Aluminum foil. 8. Glass slides. 9. Kimwipes. 10. Coverslips: 22 × 22 mm and 24 × 50 mm. 11. Clear nail polish. 12. Cardboard slide folders. 13. Confocal microscope. 14. Cryomolds. 15. Cryostat. 16. Slide storage boxes. 17. Wax pen. 18. Coplin staining jars. 19. Humidified chamber.

22

Kelly L. Betterman and Natasha L. Harvey

Table 1 Primary antibodies routinely used to analyze dermal lymphatic vessels in whole-mount tissues and sections Name

Antibody against

Company

Species Dilution

CD31

Blood and lymphatic vessels

BioLegend; #102502 (MEC13.3)

Rat

1:500

Endomucin

Blood vessels

Santa Cruz Biotechnology; sc-65495 (V.7C7)

Rat

1:500

F4/80

Macrophages

ThermoFisher Scientific; MF48000 (BM8)

Rat

1:500

LYVE-1

Lymphatic vessels and macrophages

AngioBio; #11-034

Rabbit

1:1000

Neuropilin 2 (NRP2)

Lymphatic vessels (highlights filopodia), nerves and hair follicles

R&D Systems; AF2215 Cell Signaling; #3366 (D39A5)

Goat Rabbit

1:500 1:500

Phospho-histone H3 (PH3)

Proliferating cells

Millipore; #06-570 (Ser10)

Rabbit

1:500

PROX1

Lymphatic vessels

R&D Systems; AF2727 Abcam; ab101851

Goat Rabbit

1:500 1:1000

VEGFR-3

Lymphatic vessels

R&D Systems; AF743

Goat

1:500

Sigma; C6198 (Cy3 conjugate)

Mouse

1:1000

α-Smooth muscle Arteries actin

3  Methods Carry out all procedures at room temperature, unless otherwise specified. 3.1  Embryo Harvest and Fixation

1. Following the humane killing of a pregnant female mouse (see Note 12), make an incision in the abdominal cavity to expose the uterine horns containing embryos. 2. Carefully dissect out the two uterine horns and place immediately in a petri dish containing ice-cold 1× PBS. 3. Under a dissecting microscope dissect embryos free from the uterine horn and surrounding extraembryonic tissue in cold 1× PBS. 4. Using a spatula transfer dissected embryos to a petri dish containing fresh 1× PBS on ice (see Note 13). 5. Once all embryos are dissected, remove 1× PBS and add fresh 1× PBS. Wash embryos with gentle agitation for 10 min on ice. Repeat this step two more times.

Tools to Characterise the Dermal Lymphatic Vasculature

23

6. Transfer embryos to a 50 mL conical centrifuge tube containing around 40 mL of 4% PFA (see Notes 13–15). Fix embryos overnight at 4° with gentle agitation. 7. Remove 4% PFA and replace with 1× PBS. Wash embryos with gentle agitation for 10 min on ice (see Note 16). Repeat this step two more times. 8. After final wash replace 1× PBS with 1× PBS containing 0.01% (w/v) thimerosal (see Note 17). 9. Store embryos at 4 °C until ready for use (see Note 18). 3.2  Whole-Mount Immunostaining of Embryonic Dorsal Skin

1. Using a dissecting microscope to facilitate visualization, carefully dissect thoracic region of dorsal skin tissue (see Notes 19–21) away from the embryo proper (see Fig. 1, red shaded region) in 1× PBS.

3.2.1  Embryonic Skin Dissection

2. Using forceps carefully remove extra layers of muscle and underlying connective tissue from the dissected piece of skin, ensuring the superficial vascular networks remain largely intact (see Note 22). 3. Transfer each piece of dissected skin to an individual well of a 24-well, flat-bottom tissue culture plate containing 300– 500 μL blocking solution (see Note 23).

3.2.2  Whole Mount Immunostaining

1. Incubate skin in blocking solution with gentle agitation for 1 h at room temperate or at 4 °C overnight (see Note 24).

Fig. 1 Anatomical regions of the dorsal skin used for whole-mount immunostaining from an E14.5 mouse embryo. Side and dorsal view of an E14.5 mouse embryo illustrating three anatomical regions of the dorsal skin, cervical (orange), thoracic (red), and lumbar (green), used to visualize the dermal vasculature using whole-mount immunostaining. Grey dashed line represents the dorsal midline of the embryo

24

Kelly L. Betterman and Natasha L. Harvey

2. Prepare appropriate primary antibody combinations diluted in blocking solution (see Notes 25 and 26). Table 1 outlines primary antibodies that we routinely use for visualization of the dermal vasculature (see Note 27). A negative control can be added by omitting primary antibodies or replacing them with isotype control antibodies. 3. Remove the blocking solution from the samples using a P200 or P1000 micropipette and add primary antibody solution. 4. Incubate skin at 4 °C overnight with gentle agitation. 5. Remove the primary antibody solution and add around 500 μL of wash solution. Wash samples extensively at 4 °C for 1 h with gentle agitation. Repeat this step five more times (see Note 28). 6. Dilute appropriate Alexa Fluor® conjugated secondary antibodies 1:500 in blocking solution (see Note 29). Prepare 500 μL per well. 7. Replace wash solution with secondary antibody solution. Incubate the skin samples overnight at 4 °C in the dark (see Note 30) with gentle agitation. This step and all subsequent steps must be carried out in the dark to avoid photobleaching. 8. Remove secondary antibody solution and add 500 μL of wash solution. Wash samples to remove unbound secondary antibodies at 4 °C for 1 h with gentle agitation. Ensure plate remains wrapped in foil. Repeat this step five more times. 9. Under a dissecting microscope place each skin sample, one at a time (epidermis facing down and orientated horizontally across the width of the slide, see Note 31), with a drop of wash solution (to prevent the sample drying out under the light of the dissecting microscope) on a glass slide and carefully flatten out using forceps. 10. Using paper towel or a Kimwipe, absorb excess wash solution by placing it near the skin sample. Be careful not to touch the sample. 11. Carefully add one drop of DAPI Fluoromount G™ mounting medium (see Note 32) on top of the piece of skin. 12. Gently overlay the skin sample with a 22 × 22 mm glass coverslip. 13. To permanently adhere the coverslip in place, add a drop of clear nail polish to each corner of the coverslip. 14. Store slides in a cardboard slide folder at 4 °C in the dark (see Note 33). 15. Acquire z-stack images using a confocal microscope (see Note 34 and Fig. 2). 16. Convert images into maximum-intensity projections to view the complete z-stack image.

Tools to Characterise the Dermal Lymphatic Vasculature

25

Fig. 2 Representative confocal tile scan of an E14.5 dorsal skin sample. (a) Confocal tiled z-stack image of dissected cervicothoracic dorsal skin sample from an E14.5 wild-type embryo immunolabeled with NRP2, highlighting the dermal lymphatic vascular network, along with developing hair follicles (discrete spots). The pink shaded region stretching from flank to flank defines a typical area employed for downstream lymphatic vessel network analyses. (b) E14.5 dorsal skin sample stained with antibodies against PROX1 (red) and NRP2 (cyan) to visualize lymphatic vessels, in combination with CD31 (green) to detect the blood vasculature. Lymphatic vessels closest to the midline (pink arrows) are defined as the sprouting front and represent a region of active sprouting lymphangiogenesis. Grey dashed line represents the dorsal midline of the embryo. Scale bar represents 500 μm

17. Open images in ImageJ software [7], set scale depending on size of captured image in μm (Analyze > Set Scale) and overlay image with a grid (Plugins > GridNoOffset) to enable randomization of vessel width measurements (see Fig. 3). 18. Use Lymphatic Vessel Analysis Protocol (LVAP) [8] (Plugins > Lymphatic Vessel Analysis) to quantify lymphatic vessel parameters such as vessel width and branch points from acquired images of whole-mount immunostained embryonic skin samples (see Fig. 3 for example quantification). 19. Cut and paste the data into a spreadsheet for further analysis. 3.3  Tissue Section Immunostaining (See Note 35)

1. Following collection and fixation of embryos (see Subheading 3.1), transfer embryos to equilibrate overnight in 1× PBS containing 30% (w/v) sucrose at 4 °C with gentle agitation (see Note 36).

26

Kelly L. Betterman and Natasha L. Harvey

Fig. 3 Quantification of dermal lymphatic vessel width and branch points. E14.5 wild-type embryonic dorsal skin immunolabeled with NRP2* (from Fig. 2b) to show lymphatic vessel network quantification parameters, including vessel width (red lines) and branch points (green dots). Measure vessel width (μm) at any point where the grid lines intersect over part of a vessel (red lines). The average of these measurements will yield an average lymphatic vessel width for the sample. Complete these analyses on images from both sides of the dorsal midline. Branch points (green dots) can be counted on the same images. *Image converted to greyscale to enable easier visualization of marks used for measurement

2. Embed embryos in a cryomold in O.C.T. Compound, noting their orientation in the cryomold and allow embedding medium to completely set on dry ice. 3. Transfer cryomolds to −70 °C freezer where they can be stored indefinitely. 4. Section embryo cryoblocks transversely (i.e., in a head-to-­ rump direction), sagittally, or coronally, depending on required information (see Note 37), at 10 μm on a cryostat (see Note 38). Slides can be stored in a slide storage box at −20 °C until ready to stain. 5. Select appropriate slides to stain and allow to air-dry for 5 min. 6. Frame sections with a wax pen (see Note 39) and allow to air-­ dry for 5 min. 7. Place slides in a Coplin staining jar(s) filled with TBS-T for 15 min with gentle agitation to remove surrounding O.C.T. Compound from tissue sections.

Tools to Characterise the Dermal Lymphatic Vasculature

27

8. Remove slides from Coplin jar and dab excess TBS-T from slides by blotting with a Kimwipe. Lay slides flat in a humidified chamber (see Note 40). 9. Add 50–100 μL of blocking solution per row of sections and incubate in humidified chamber for 30 min to 3 h. From this point on, do not let slides dry out at any stage of the protocol. 10. Prepare appropriate primary antibody combinations diluted in blocking solution (see Note 26). Table 1 outlines primary antibodies that we routinely use for visualization of the dermal vasculature (see Note 27). A negative control can be added by omitting primary antibodies or replacing them with isotype control antibodies. 11. Tip off blocking solution and replace with primary antibody solution (see Note 41). Incubate overnight at room temperature in humidified chamber. 12. Tip off primary antibody solution and place slides in a Coplin jar(s) filled with TBS-T for 10 min with gentle agitation. Tip off TBS-T and replace with fresh TBS-T. Repeat this step twice more. 13. Whilst slides are washing, dilute appropriate Alexa Fluor® conjugated secondary antibodies 1:500 in blocking solution (see Note 29). Prepare 50–100 μL per row of tissue sections. 14. Remove slides from Coplin jar(s) and dab excess TBS-T from slides by blotting with a Kimwipe. Lay slides flat in humidified chamber. 15. Add secondary antibody solutions to sections (see Note 42) and incubate for 2–3 h at room temperature in humidified chamber. Ensure slides are kept in the dark from this point forward to prevent photobleaching. 16. Tip off secondary antibody solution and place slides in a Coplin jar(s) filled with TBS-T to wash for 10 min with gentle agitation (see Note 43). Tip off TBS-T and replace with fresh TBS-T. Repeat this step twice more. 17. Following the final wash in TBS-T, fill Coplin jar(s) with water and wash for 5 min with gentle agitation to remove excess salts. 18. Remove slides from Coplin jar(s), lay slides flat on paper towel and allow to air-dry for approximately 15 min in the dark. 19. Add one drop of DAPI Fluoromount G™ mounting medium (see Note 32) to each row of sections on the slide and carefully overlay with an appropriately sized glass coverslip. 20. To permanently adhere the coverslip in place add a drop of clear nail polish to each corner of the coverslip.

28

Kelly L. Betterman and Natasha L. Harvey

Fig. 4 Immunostaining of the dermal blood and lymphatic vasculature in thin embryonic tissue sections. Representative confocal z-stack image through the dermis of a cryosectioned E14.5 wild-type mouse embryo immunolabeled with antibodies to the pan-endothelial maker CD31 (green), in combination with PROX1 (red) and NRP2 (cyan) to detect the dermal lymphatic (arrows) and blood vessels (green). DAPI (grey) marks all nuclei. Scale bar represents 50 μm

21. Store slides in a cardboard slide folder at 4 °C in the dark (see Note 33). 22. Acquire z-stack images using a confocal microscope (see Fig. 4 for an example).

4  Notes 1. We purchase 10× Dulbecco’s Phosphate Buffered Saline (without calcium chloride and magnesium chloride). 2. Due to the hazardous nature of the chemicals involved, PFA should be prepared in a fume hood or with adequate ventilation. Add approximately 300 mL of water to a 1 L glass beaker and heat (see Note 3). Weigh 40 g PFA (see Note 4) and transfer to a 1 L Schott bottle. Carefully add heated water and swirl to mix. Add 1 mL of 10 M NaOH and place closed bottle on a roller mixer to facilitate PFA going into solution. Using a 1 L graduated cylinder add water to a final volume of 900 mL. Transfer solution back into the same Schott bottle and add 100 mL of 10× PBS. Mix and adjust pH to 7.4 with concentrated HCl (see Note 5). Filter through a 0.45 μm bottle top filter, aliquot (10–30 mL volumes) and store at −20 °C for up to 12 months.

Tools to Characterise the Dermal Lymphatic Vasculature

29

3. Microwave for around 2 min, or until water temperature reaches approximately 70 °C. Heating the water allows the PFA powder to dissolve more rapidly. 4. Wear a mask when weighing out and handling PFA in its powdered state. Should transport of powdered PFA be required ensure it is contained appropriately. Avoid breathing in PFA powder and contact with skin. 5. We found that approximately 820 μL of concentrated HCl is sufficient to effectively adjust the pH to around 7.4. We use pH strips as a guide to determine the pH. 6. Wear a mask when weighing out thimerosal. Weigh 1 g of thimerosal into a 100 mL Schott bottle. Make volume to approximately 70 mL with water. Add 10 mL of 10× PBS stock solution. Make up to 100 mL with water and mix. Can be stored indefinitely at room temperature. 7. Add 5 mL of Triton X-100 (cut end of blue tip to aspirate Triton X-100 more easily or use a 1ml plastic transfer pipette) to a 50 mL conical centrifuge tube. Make up volume to approximately 40 mL with water. Add 5 mL of 10× PBS stock solution. Top up to 50 mL with water. Agitate tube on roller mixer to allow Triton X-100 to completely dissolve. Shake a few times to facilitate this process. Can be stored indefinitely at room temperature. 8. Add 0.5 g of BSA to a 50 mL conical centrifuge tube containing approximately 30 mL of water. Having water at the bottom of the tube helps to dissolve the BSA more easily. Add 5 mL of 10× PBS stock solution and agitate on roller mixer until BSA is completely dissolved. Add 1.5 mL of 10% Triton X-100 stock and mix. Filter solution through a 0.45 μm syringe filter. Can be stored indefinitely at 4 °C, providing the solution is free of microbial contamination and precipitate. 9. Add 1:10 dilution of 10× PBS and 1:100 dilution of 10% Triton X-100 stock solutions to required volume of water and gently shake to mix. Can be stored indefinitely at room temperature. 10. Add 150 g of sucrose to a Schott bottle containing approximately 300 mL of water. Add 50 mL of 10× PBS stock solution and agitate on roller mixer until sucrose has completely dissolved. Make up to 500 mL with water. This solution can be stored for months at 4 °C, providing the solution is free of any microbial contamination. 11. TBS-T is prepared by adding 50 mL of 20× TBS to 950 mL of water in a Schott bottle, along with 1 mL of Tween-20 (cut end of blue tip to aspirate Tween-20 more easily or use a 1 mL plastic transfer pipette). Shake to mix. Can be stored indefi-

30

Kelly L. Betterman and Natasha L. Harvey

nitely at room temperature, providing the solution is free of microbial contamination. 12. All procedures involving animal experiments should follow approved institutional and governmental animal protocols and comply with the relevant guidelines and regulations of local animal ethics committees. 13. If embryos are genetically modified and need to be genotyped, it is at this point that tissue samples (usually from tail or yolk sac) should be collected for genotyping and embryos numbered. From this point forward, embryos should be kept individually (i.e., washed and fixed) in numbered wells of an appropriate sized multiwell tissue culture plate. We find a 24-well plate the best for E9.5 to E13.5 embryos, a 12-well plate for E14.5 to E15.5, 6-well plate for E16.5 and 50 mL conical centrifuge tubes for E18.5 embryos. If, however, embryos are all of the same genotype (i.e., wild-type), then embryos can be washed and fixed together in petri dishes and 15 mL or 50 mL conical centrifuge tubes, respectively, dependent on age of embryos. 14. We find that thawing PFA aliquots fresh on the day of use in a 37 °C water bath until PFA is completely in solution best practice. Discard any unused thawed PFA, ensuring that PFA waste disposal regulations are followed. 15. If fixing embryos in multiwell plates (described above in Note 13) then simply fill each well with 4% PFA. For larger embryos fixed in tubes, a 10:1 ratio of fixative to tissue/embryo volume is best practice. 16. Washes can be carried out on ice or in a cold room/fridge at 4 °C. 17. Add 1:10 dilution of 10× PBS and 1:100 dilution of 1% thimerosal (antimicrobial) stock solutions to required volume of water and gently shake to mix. Can be stored indefinitely at room temperature. 18. Embryos can be stored in PBS at 4 °C for weeks to months in 15 mL or 50 mL conical centrifuge tubes, or multiwell tissue culture plates wrapped in Parafilm, if unable to stain immediately. Ensure that multiwell plates do not dry out. 19. For experiments involving the analysis of genetically modified embryos, littermate (wild-type) controls should be used where possible. 20. We routinely stain skin for dermal lymphatic vessel analyses from E14.5 embryos, as this stage is when the lymphatic vessel network is actively sprouting and has almost joined at the midline (from the flanks) to form an interconnected network across the embryo. Earlier at E13.5 is ideal for investigating lymphatic

Tools to Characterise the Dermal Lymphatic Vasculature

31

cluster formation in the skin and E16.5 to E18.5 are the best stages for examining lymphatic vessel valve formation in the skin. This protocol can be used successfully for any embryonic stage ranging from E12.5 to E18.5. 21. We routinely dissect the thoracic region of the dorsal skin for dermal lymphatic vessel analyses as this region generally is the most developed in terms of blood vessels converging at the midline and lymphatic vessel network formation. Depending on your area of interest, similar analyses can be performed in the cervical (see Fig. 1, orange shaded region) and/or lumbar (see Fig. 1, green shaded region) areas. 22. The older the embryo (i.e., E16.5 onward), the more underlying connective tissue and muscle there is to remove. 23. The volume of blocking solution used is not critical at this stage. We find for E13.5 and E14.5 skin pieces 300 μL is sufficient, but for larger pieces of skin from E15.5 to E18.5 embryos a larger volume of 500 μL is better to ensure the samples remain submerged in solution. 24. Blocking time is flexible and can be done at room temperate for a few hours or overnight at 4 °C, depending on experimental window. If the blocking step is performed for a few hours, the protocol will be 3 days in duration (with two overnight incubations), whilst if blocking is performed overnight the protocol will be 4 days in length (with three overnight incubations). 25. The volume of primary antibody solution is dependent on the age of embryo and the size of dissected skin pieces being stained. Prepare 250–300 μL for E13.5 and E14.5 samples and 500 μL for larger pieces of skin from E15.5 to E18.5 embryos. 26. The number of primary antibodies added to each sample will depend on the secondary antibodies available and the capability of the confocal microscope used for downstream imaging. 27. Ensure primary antibodies used for each combination are all raised in different species to prevent cross-reactivity. We routinely use a combination of goat anti-PROX1, rabbit anti-­ NRP2, and rat anti-CD31 for clear visualization of dermal lymphatic vessels and blood vessels, respectively. See Table 1 for further details. 28. Wash volume is not critical and can be added using a 1 mL plastic transfer pipette. To ensure skin samples are not aspirated it is best to use a P200 or P1000 micropipette to carefully remove the wash solution. This step can be carried out on ice if a rocking platform mixer is not easily accessible in a cold room/fridge.

32

Kelly L. Betterman and Natasha L. Harvey

29. Secondary antibodies employed are dependent on the species of primary antibodies added and on the lasers available on your confocal microscope. We routinely use Alexa Fluor® 488 (green), Alexa Fluor® 555 (red), and Alexa Fluor® 647 (far red) fluorophore-conjugated secondary antibodies. The 555 fluorophore offers the best signal-to-noise ratio, followed by 647 and then 488. Therefore it is best practice to detect your antibody of interest with a 555-conjugated secondary antibody. 30. Wrap multiwell plate with aluminum foil to minimize photobleaching. 31. Orientation of the skin sample is more critical for late stage embryos (E16.5 onward) when the skin is thicker and the epidermis influences depth of resolution. 32. Any aqueous-based mounting medium that preserves tissue fluorescence and contains DAPI (for nuclei visualization) can be used. 33. Slides can be stored for up to 12–18 months in the dark at 4 °C without loss of sample fluorescence, providing mounting medium adequately covers samples. 34. For imaging of the overall dermal vascular network, use a 10× objective. For more detailed analyses at the sprouting front (see Fig. 2, pink arrows), such as filopodial number and length, tip cell morphology and endothelial cell proliferation analyses (dual PROX1-positive, PH3-positve cells) use either a 20×, 40×, or 63× objective. It is important to use the same laser settings for all analyses from one experiment. 35. This protocol can be performed on any age embryo and/or tissue sample, but for the purpose of dermal lymphatic vessel examination we use E14.5 mouse embryos as our developmental time point of choice, as these smaller sized embryos fix and cryosection efficiently. 36. If embryos need to be kept separate this step can be performed in an appropriate sized multiwell tissue culture plate or conical centrifuge tubes (similar to Note 13). 37. Embryos can be sectioned in any plane, including transverse (i.e., in a head to rump direction), frontal/coronal (i.e., ­dorsal/back to ventral/front) and sagittal (i.e., left to right) depending on anatomical structures to be examined. 38. We generally aim to fit three rows and three sections per row on a slide depending on the age of embryo (see Fig. 5). This means that upon immunostaining three different antibody combinations can be applied per slide.

Tools to Characterise the Dermal Lymphatic Vasculature

33

Fig. 5 Slide layout of embryonic tissue sections. Schematic depicting ideal placement of three rows of three embryonic cryosections per row on a glass slide. Wax pen (blue lines) is applied around each row of sections to facilitate the use of different antibody combinations

39. Apply wax pen carefully around the three distinct rows of sections on each slide (see Fig. 5, blue lines) to enable complete separation of different antibody combinations and prevent cross-contamination between rows. 40. For a humidified chamber we use an opaque 100 place slide storage box containing a small amount of water in the base of the box (see Fig. 6). Lay the slides across the longitudinal ridge and close the lid to prevent evaporation of small volumes of solution. 41. Ensure surface tension is retained when adding the primary antibody solution to prevent cross-contamination between rows (i.e., do not add excessive volumes of primary antibody solution). It is best to determine ideal volume using blocking solution prior to adding antibodies to slides. 42. Check integrity of wax pen demarcations around sections following washing in TBS-T and redraw if required. It is important that secondary antibodies do not mix between rows. Retain surface tension when adding secondary antibody solution. 43. Cover Coplin jar with a small cardboard box or wrap in aluminum foil to prevent photobleaching of samples during washes.

34

Kelly L. Betterman and Natasha L. Harvey

Fig. 6 Humidified chamber set-up. Humidified chamber set-up consisting of an opaque 100 place slide storage box containing a small amount of water in the base of the box. Slides are placed across the ridge of the slide rack

Acknowledgments This work was supported by grants from the NHMRC (APP1061365) and ARC (DP150103110). NLH is supported by an ARC Future Fellowship (FT130101254). References 1. Gerhardt H, Golding M, Fruttiger M, Ruhrberg C, Lundkvist A, Abramsson A, Jeltsch M, Mitchell C, Alitalo K, Shima D, Betsholtz C (2003) VEGF guides angiogenic sprouting utilizing endothelial tip cell filopodia. J Cell Biol 161(6):1163–1177. https://doi. org/10.1083/jcb.200302047 2. Norrmen C, Ivanov KI, Cheng J, Zangger N, Delorenzi M, Jaquet M, Miura N, Puolakkainen P, Horsley V, Hu J, Augustin HG, Yla-Herttuala S, Alitalo K, Petrova TV (2009) FOXC2 controls formation and maturation of lymphatic collecting vessels through cooperation with NFATc1. J Cell Biol 185(3):439–457. https:// doi.org/10.1083/jcb.200901104

3. Bernier-Latmani J, Petrova TV (2016) High-­ resolution 3D analysis of mouse small-intestinal stroma. Nat Protoc 11(9):1617–1629. https:// doi.org/10.1038/nprot.2016.092 4. Betterman KL, Paquet-Fifield S, Asselin-Labat ML, Visvader JE, Butler LM, Stacker SA, Achen MG, Harvey NL (2012) Remodeling of the lymphatic vasculature during mouse mammary gland morphogenesis is mediated via epithelialderived lymphangiogenic stimuli. Am J Pathol 181(6):2225–2238. https://doi. org/10.1016/j.ajpath.2012.08.035 5. Gordon EJ, Rao S, Pollard JW, Nutt SL, Lang RA, Harvey NL (2010) Macrophages define dermal lymphatic vessel calibre during develop-

Tools to Characterise the Dermal Lymphatic Vasculature ment by regulating lymphatic endothelial cell proliferation. Development 137(22):3899– 3910. https://doi.org/10.1242/dev.050021 6. James JM, Nalbandian A, Mukouyama YS (2013) TGFbeta signaling is required for sprouting lymphangiogenesis during lymphatic network development in the skin. Development 140(18):3903–3914. https://doi. org/10.1242/dev.095026

35

7. Abràmoff MD, Magalhães PJ, Ram SJ (2004) Image processing with ImageJ. Biophoton Int 11(7):36–42 8. Shayan R, Karnezis T, Tsantikos E, Williams SP, Runting AS, Ashton MW, Achen MG, Hibbs ML, Stacker SA (2007) A system for quantifying the patterning of the lymphatic vasculature. Growth Factors 25(6):417–425. https://doi. org/10.1080/08977190801932550

Chapter 3 Genetic Lineage Tracing of Lymphatic Endothelial Cells in Mice Ines Martinez-Corral and Taija Makinen Abstract Lineage tracing allows for identification of all progeny produced by a single cell or groups of cells and can thus be used to assess developmental fate of cells. Here we focus on one of the most widely used lineage tracing approaches that utilize the Cre/loxP system for site-specific genetic recombination in studying the developmental origins of lymphatic endothelial cells (LECs) in the mouse embryo. We discuss general considerations for a successful Cre/loxP based lineage tracing experiment and provide information about strains that are available for genetic lineage tracing of LECs. A protocol for lineage tracing analysis of the lymphatic vasculature by whole-mount immunofluorescence in two embryonic tissues, the skin and the mesentery, is also provided. Key words Lineage tracing, Cre Reporter, Lymphatic vessel, Endothelial cell, Cre/loxP

1  Introduction Lineage tracing is a powerful method that allows for identification of all progeny produced by a single cell or groups of cells [1]. It is commonly used in developmental biology to assess developmental fate of cells, but is increasingly applied to defining and studying the properties of stem cells also in adult tissues and disease contexts, such as cancer [2–4]. Different experimental strategies used for lineage tracing have been discussed in recent reviews [1, 5, 6]. One of the well-known uses of these approaches utilizes the Cre/loxP system for site-specific genetic recombination. For this approach, a transgenic mouse expressing the Cre recombinase under the control of a cell-type or tissue-specific promoter is crossed with a Cre reporter strain. The latter carries a transgene encoding for example a fluorescent reporter protein, which is usually under a ubiquitously expressed promoter, but expression is prevented by a loxP-­ stop-­ loxP cassette. Cre excises the stop cassette leading to constitutive expression of the reporter gene in Cre-expressing cells

Guillermo Oliver and Mark L. Kahn (eds.), Lymphangiogenesis: Methods and Protocols, Methods in Molecular Biology, vol. 1846, https://doi.org/10.1007/978-1-4939-8712-2_3, © Springer Science+Business Media, LLC, part of Springer Nature 2018

37

38

Ines Martinez-Corral and Taija Makinen

a

b

4-OHT LEC progeny

Inducible Cre 4OH-Tamoxifen HSP90 CreERT C

Constitutive Cre Cre

Nucleus

Cyto pla sm

CreERT loxP

Target

Tie2+ Prox1-

HSP90

CreERT

Endothelial cell

Venous EC Tie2+ Prox1-

Venousderived LEC

LEC

Tie2+ Prox1+

Tie2+ Prox1+

Tie2+ Prox1+

Tie2+ Prox1-

LEC progeny

Arterial EC Non-venousderived LEC Tie2Prox1+/-

LEC

Tie2+ Prox1+

Tie2+ Prox1+

loxP

Lineage tracing using a constitutiveTie2-Cre Induction of Tie2 expression and Cre recombination Lineage tracing using a LEC specific Prox1-CreERT2

Fig. 1 General principle of the Cre/loxP based genetic lineage tracing. (a) Schematic of the Cre/LoxP recombination system. The constitutively active Cre recombinase (grey) recognizes and excises target sequences in DNA that are flanked by loxP sites. The inducible CreERT2 recombinase (green) is retained in the cytoplasm until activated by 4OH-Tamoxifen. For genetic labeling of cells, Cre-mediated removal of a loxP-stop-loxP cassette leads to constitutive expression of a reporter gene. (b) Schematic representation of lineage tracing of endothelial cells using a constitutive EC-specific (Tie2-Cre) and an inducible LEC-specific (Prox1-CreERT2) Cre line. Constitutive Tie2-Cre labels all Tie2-expressing cells and their progeny, and does not thus allow distinguishing between current and historical expression. Inducible Prox1-CreERT2 allows labeling of Prox1-expressing cells at a specific developmental time point, and following the progeny of those cells

and their progeny (Fig. 1a). The fundamental basis of genetic lineage tracing is that once the reporter expression has been ­ switched on by irreversible Cre-mediated recombination, it is expressed constitutively in these cells and all their descendants independent of later recombinase expression, thus enabling the detection and tracking of fluorescent founder cells and their lineage. Here, we focus on Cre/loxP based lineage tracing approach in studying the developmental origins of lymphatic endothelial cells (LECs) in the mouse embryo. We first discuss general considerations for a successful Cre/loxP lineage tracing experiment and provide information about strains that are available for genetic lineage tracing of LECs. A detailed protocol for lineage tracing analysis of the lymphatic vasculature by whole-mount

Genetic Lineage Tracing of Lymphatic Endothelium

39

immunofluorescence in two embryonic tissues, the skin and the mesentery, where different embryonic origins of LECs have been described [7, 8] is also provided. 1.1  Key Considerations for a Successful Cre/ loxP Based Lineage Tracing Experiment 1.1.1  Choice Between a Constitutive and an Inducible Cre Line

Constitutive Cre allows for irreversible genetic labeling of all Cre-­ expressing cells and their progeny. Positive labeling may thus indicate either current expression in the cells of interest or historical expression in their ancestors at an earlier developmental time point (Fig. 1b). Since gene expression is dynamically regulated, constitutive Cre lines do not allow determining temporal lineage contributions. However, they can be powerful when excluding the contribution of a particular cell lineage to a tissue of interest. To capture dynamic changes in gene expression, inducible approaches where Cre can be activated transiently at a specific developmental time point have been developed. Tamoxifen-­ inducible form of Cre utilizes a mutated hormone binding domain of the estrogen receptor (ER) fused to Cre. The CreER protein is insensitive to the natural ligand 17β-estradiol at physiological concentrations, but is activated and translocates to the nucleus upon binding of the synthetic ligand, the Tamoxifen metabolite 4-hydroxytamoxifen (4-OHT) (Fig. 1a). A modified CreERT2 is a widely used form of CreER, which has low background recombinase activity and robust activation following Tamoxifen treatment [9]. A critical consideration when planning and interpreting data from inducible Cre/loxP lineage tracing experiments is to determine the cell type(s) targeted at the time of induction (see the next paragraph). The efficiency of recombination in the target cell population(s), i.e., to what extent the initial population(s) to be traced are labeled, and the absence of Tamoxifen independent recombination should also be assessed. The latter is an important but often underappreciated consideration, since CreER lines can present low level of Cre activity in the absence of Tamoxifen ­induction [10]. Low level of activity is unlikely of critical importance when Cre is used to delete a gene of interest, which usually requires recombination of two floxed alleles in the same cell. However, it can become important in lineage tracing studies where a single recombination event in a stem/progenitor cell, or any proliferative cell can result in labeling of a large progeny of (different types of) cells observed at the stage of analysis. Transgenic Cre and CreER lines are usually phenotypically normal. However, it is important to note that some lines may exhibit a phenotype due to random transgene integration that can influence nearby genes, or Cre toxicity that has been associated with DNA damage and apoptosis [11, 12]. Cre toxicity has been reported in several lines and appears to depend on the level of

40

Ines Martinez-Corral and Taija Makinen

expression of the recombinase. In the case of inducible lines, it may only appear after Tamoxifen administration [11, 12]. Different cell types may be differentially sensitive to Cre, which can potentially affect results obtained from lineage tracing experiments. An important principle of temporally controlled genetic lineage tracing is that the inductive agent is administered transiently at the start of the experiment. Both Tamoxifen and 4-OHT have been widely used in the literature for lineage tracing studies. However, there are important differences related to their activity that may affect the outcome of the tracing experiment and should thus be taken into consideration. Tamoxifen is metabolized in the liver into active metabolites, which include 4-OHT. This conversion of Tamoxifen takes approximately 6–12 h in vivo [5], and thus results in a delay in the onset of Cre-mediated recombination when compared to administration of 4-OHT. Apart from differences in the initiation of Cre-mediated recombination, the duration of the recombination window is different depending on whether Tamoxifen or 4-OHT was administered. Measurement of 4-OHT in the blood plasma of C57BL/6J mice administered with different doses of Tamoxifen or 4-OHT showed different kinetics, translating into different windows of Cre activity (Fig. 2) [7, 13]. While serum levels of 4-OHT peaked after intraperitoneal 4-OHT administration rapidly at 6 h and declined within 24 h, Tamoxifen administration led to slower

Serum levels of 4-OHT 2000

4-OHT (ng/ml)

1.1.2  Choice Between 4-OHT and Tamoxifen

1500

1 mg 4-OHT 2 mg 4-OHT

1000

5 mg Tamoxifen

500 0

0

12

24

36

48

60

72

Time (hours)

Fig. 2 Kinetics of 4-OHT and Tamoxifen induced Cre-activity. Analysis of 4-OHT metabolite levels in serum after administration of a single dose of Tamoxifen or 4-OHT. Doses used in this study: 1 mg of 4-OHT, 2 mg of 4-OHT, and 5 mg of Tamoxifen. A single dose of 4-OHT showed a 24-h time window for Cre activity, while Tamoxifen administration led to up to a 3-day window of activity. Reproduced from [7]

Genetic Lineage Tracing of Lymphatic Endothelium

41

kinetics with levels peaking at 12 h and sustained for up to 48–72 h. Duration of the recombination window as well as the efficiency of recombination will also depend on the dose and the number of administered doses of Tamoxifen/4-OHT. Repeated administration of Tamoxifen at commonly used doses was reported to induce Cre-mediated recombination for up to weeks after the last t­ reatment [14, 15]. Route of Tamoxifen administration, the accessibility of loxP sites in the reporter construct (see the next paragraph), the tissue where it is expressed, as well as the developmental stage may also influence the efficiency of Cre recombination. Tamoxifen/4OHT dose should be therefore optimized for each experiment to achieve the desired recombination efficiency and window of Cre activity. Tamoxifen is an established endocrine disruptor acting via endogenous estrogen receptors. Its use can therefore lead to side effects. At high doses, Tamoxifen can be lethal to the developing embryo [16]. In addition, administration of Tamoxifen to pregnant females impacts their ability to give natural birth [17]. For analysis of embryonically labeled cells and their contribution to postnatal tissues, pups should be delivered by cesarean section and fostered with another female [17]. In postnatal mice, Tamoxifen treatment has been reported to lead to long-term adverse effects on the reproductive system in juvenile males [12] and dramatic rearrangement of the gastric mucosa [18]. 1.1.3  Choice of a Cre Reporter Line

A number of Cre reporter lines are available that utilize different promoter/enhancer elements and express different fluorescent proteins localized to various subcellular compartments (e.g., nuclear, membrane, and cytoplasmic) [19]. Typically, Cre reporter lines utilize a ubiquitous promoter to ensure that any cell at any developmental or adult stage has the potential to express the reporter transgene. Targeting the reporter gene into the endogenous Rosa26 locus is one of the most successfully used approaches, yet even Rosa26 locus is suppressed in some tissues such as the adult brain [20]. This is an important consideration, since lack of reporter expression could lead to a wrong interpretation of negative tracing results. In addition to reporter lines expressing a single reporter protein, dual and multicolor reporters have been developed that provide important advantages in certain experimental settings. Double reporters, such as the R26-mT/mG line, express one fluorescent protein (in this case Tomato) which is replaced upon Cre recombination by another fluorescent protein (in this case GFP) [21]. Since all cells can be visualized, dual reporters can verify that lack of recombination in the cell of interest is not due to lack of reporter expression. In addition, double reporters may help distinguishing between recent and historical recombination

42

Ines Martinez-Corral and Taija Makinen

events due to perdurance of the Tomato protein, with simultaneous presence of the two fluorescent proteins indicating recent recombination [21]. This can be useful when used in combination with constitutive Cre lines to distinguish current expression in the cell type of interest as opposed to historical expression in their progenitors. However, the decay of Tomato protein (in case of the R26-mT/mG reporter) is likely to vary depending on the developmental stage and cell type and cannot accurately determine the timing of the recombination event. Multicolor reporters, such as the R26R-Confetti [22] and ifgMosaic [23], express multiple fluorescent proteins from a single genomic locus. In these mice Cre recombination allows labeling of individual cells stochastically with different fluorescent proteins, which is ideal for studying clonal expansion and dynamics. It is generally recognized that each Cre line should be separately assessed for their efficiency and specificity for driving recombination. However, the efficiency of Cre-mediated excision should be validated also for each floxed allele, including the reporter alleles. It has not been systemically evaluated to which extend the available reporters differ in their efficiency to report Cre recombination, but it is likely that the choice of promoter, accessibility of loxP sites, and length in between loxP sites affect the recombination efficiency. Thus, different results may be obtained using different reporter lines. 1.2  Genetic Tools for Lineage Tracing of Lymphatic Endothelial Cells

Genetic lineage tracing studies using combinations of different Cre lines and fluorescent reporters have provided important insight into the developmental mechanisms of the lymphatic vasculature. Such studies demonstrated the venous origin of lymphatic vessels [24], and uncovered the contribution of alternative nonvenous derived progenitors to lymphatic vessel development in the skin, mesentery, and heart [7, 8, 25]. Cre lines available for targeting the lymphatic vasculature, and used to delineate cellular origins of LECs are summarized in Tables 1 and 2, respectively. It should be noted that none of these lines is specific to one cell type, and therefore the use of a combination of Cre lines is often needed.

2  Materials 2.1  Induction of Cre Recombination

1. Tamoxifen or 4-OH-Tamoxifen (4-OHT). 2. Peanut oil. 3. Sonicator. 4. 25 G needles. 5. 1 mL insulin syringes.

Genetic Lineage Tracing of Lymphatic Endothelium

43

Table 1 Cre lines for targeting of lymphatic endothelial cells Cre line

Genetic model

Target cells/organs

References

Constitutive pan-EC targeting Cre lines Tie2-Cre

Transgenic

All types of ECs, some hematopoietic cells

[26]

Tie2-Cre

Transgenic

All types of ECs, some hematopoietic cells

[27]

Nfatc1-Cre

Knockin

ECs of coronary arteries, lymphatic vessels and valves

[28, 29]

Inducible pan-EC targeting Cre lines Cdh5-CreERT2

Transgenic

Blood ECs (incomplete), in embryonic LECs [30] only if Tamoxifen administered before E11.5

Cdh5-CreERT2

PAC transgenic

All types of ECs, including LECs

[31]

All types of EC, including LECs. No/limited recombination in the large vessels and liver (neonate) and CNS vasculature (adult)

[32]

Pdgfb-CreERT2-­ PAC transgenic iresGFP Constitutive LEC targeting Cre lines Lyve1-Cre

Knockin (GFPCre)

LECs, a proportion of BECs and hematopoietic [33, 34] cells, ECs of yolk sac and vitelline vessels

Pdpn-Cre

BAC transgenic

LECs, subset of FRCs (lymph node). [35] Information about other organs not available

Pdpn-Cre

Transgenic (1.3kb promoter; GFPCre)

LECs (after exiting from the cardinal vein), other Pdpn-expressing tissues

[36]

Nrp2-Cre

BAC transgenic (GFPCre)

Primitive blood cells, venous and lymphatic EC, peripheral sensory ganglia, lung bud

[37]

Prox1-Cre

Knockin (GFPCre)

LECs

[38]

[24]

Inducible LEC targeting Cre lines Prox1-CreERT2

Knockin

LECs, other Prox1-expressing tissues

Prox1-CreERT2

BAC transgenic

LECs, ECs of large veins (iliac, mesenteric) and [39] venous valves, nonvascular cells in the liver, heart, CNS, and muscle

Vegfr3-CreERT2 Knockin Lyve1-CreERT2

BAC transgenic

LECs, a proportion of BECs in embryos and knockin homozygous mice

[40]

LECs

[41]

PAC P1 phage artificial chromosome, BAC bacterial artificial chromosome, FRC fibroblastic reticular cells (lymph node), CNS central nervous system

Cre reporter

E10 or E11: 1 mg 4-OHT

Vav-Cre

–

E14 or E15: 1 mg 4-OHT

E12 or E13: 1 mg 4-OHT

Nearly complete absence of GFP+ LECs in dermal LV (E17.5) and jugular LS (E13.5) GFP+ LECs in jugular LS (E13.5) but high proportion of GFP− LECs in the sprouting front of dermal vasculature (E17.5) Efficient labeling of dermal LECs except for isolated GFP− LEC clusters at the dorsal midline (E17.5) No GFP+ LECs (E17.5)

–

Nonvenous origin—skin [40] Tie2-Cre R26-­mTmG

Prox1-­ CreERT2

~30% of dermal LECs GFP− (E13)

E9.5: 5 mg Tam/40 g body weight

lacZ+ LECs within or near CV (24 h post tam); in jugular LS and peripheral LV (E13.5) lacZ+ LECs within (E12.5) and sprouting (E13.5) from the lymph sac, in the lung and mesenteric LV (E16.5) lacZ+ CV and budding LECs (E11.5) and lymph sac LECs (E13.5) lacZ+ cells (descendants of HSCs) not positive for LEC markers

No lacZ+ LECs; scattered hepatocytes (E13.5) No labeled cells

Runx1-­MER-­ Cre-MER

E10.5: 5 mg Tam/40 g body weight

E8.5: 3 mg Tam/40 g body weight E9: 3 mg Tam/40 g body weight E9.5: 3 mg Tam/40 g body weight

Targeted cells

–

R26R-­lacZ

Tam/4-OHT treatment

Tie2-Cre

Prox1-­ CreERT2

Venous origin [24]

Cre line

Table 2 Cre lines used for lineage tracing analysis of LEC origins

Low labeling selectively at vessel tips and LEC clusters suggests incorporation of newly differentiated cells into the growing vessels at the vascular front Excludes contribution from definitive HCs

Evidence for non-Tie2-­lineage (BEC) origin

Evidence for Tie2-­lineage (CV BEC) origin

Exclusive labeling of the earliest Prox1+ LECs in the CV, evidence for venous origin

Note

[43]

[39]

[27]

[42]

[26]

[24]

References

R26-­mTmG R26-­tdTom

Pdgfrb-Crea Csf1r-­CreER

– –

tdTom+ cardiac LECs and LYVE1+ macrophages (E17.5) ~30% of cardiac LECs GFP+ (E17.5) ~5% of cardiac LECs tdTom+ (E17.5)

–

CV cardinal vein, HC hematopoietic cell, HSC hematopoietic stem cell, LV lymphatic vessel, LS lymph sac a Subsequently shown to target LECs of both venous and nonvenous origins, as well as BECs in multiple tissues [52]

R26-­tdTom

Vav-Cre

No YFP+ cardiac LECs

~20% of LECs in vessels proximal to the outflow tract YFP− (E17.5) Incomplete labeling of cardiac LECs (E17.5)

~40% LECs in mesenteric clusters GFP+

ECs of major arteries (E9.5), HCs and mesenteric LECs (E13.5) HCs and mesenteric BECs and LECs (E13.5) Progressively increased labeling of mesenteric BECs, decreased labeling of HCs and LECs GFP+ LECs in mesenteric clusters (E13.5; in 30% of the embryos with overall low labeling efficiency) No GFP+ LECs (E14.5)

E9.5: 4-OHT – – –

E9: 4-OHT

–

Nonvenous origin—heart [25] Tie2-Cre R26-YFP

R26-­tdTom or mTmG R26-YFP

–

Pdgfrb-Crea

Pdgfb-­ CreERT2 Wt1-­CreERT2 Mesp1-Cre Nkx2.5-Cre Wnt1-Cre

–

Vav-Cre

cKit-­CreERT2

E10: 2 mg 4-OHT E10 + 11 or E11 + 12 or E13 or E14: 1 mg 4-OHT/dose E10 and/or E11: 2 mg 4-OHT

Nonvenous origin—mesentery [8] Pdgfb-­ R26-­mTmG E8: 2 mg 4-OHT CreERT2 E9: 2 mg 4-OHT

Excludes contribution of proepicardial organ (Wt1+), lateral plate mesoderm-derived progenitors (Mesp1+, Nkx2.5+) and cardiac neural crest (Wnt1+) Evidence for hematopoietic origin

Evidence for non-Tie2-­lineage (BEC) origin

[43]

Excludes contribution from definitive HCs

[45] [51]

[50]

[46] [47] [48] [49]

[32]

[27]

[45]

[44]

[32]

Evidence for contribution from cKit+ progenitors

Selective targeting of major arteries (E9.5) and mesenteric LEC progenitors (E13.5) upon early induction suggests contribution from (hemogenic) arterial EC

46

Ines Martinez-Corral and Taija Makinen

2.2  Tissue Dissection and Preparation

1. Dissection plates: Prepared using the SYLGARD® 184 SILICONE ELASTOMER KIT. Mix 1:10 curing agent to base solution and stir well for 5 min at room temperature (RT). Pour 15–20 mL per 10 cm dish and let it rest for 15 min to allow the trapped air to escape. Bake at 60 °C for 30 min. The plates are ready to use as soon as they cool down. 2. Insert 0.1 mm pins for dissection of mesenteries and of 0.2 mm for dissection of the embryonic skin. 3. Forceps (such as Dumont #5, biologie tips, 11 cm and standard tips, 11 cm). 4. Scissors (such as Iris Scissors, Super Cut, straight, 9 cm; Vannas Delicate, Spring Scissors, straight, 8 cm). 5. 4% (w/v) Paraformaldehyde. 6. Phosphate buffered saline (PBS): 135 mM NaCl, 3 mM KCl, 8 mM Na2HPO4, 2 mM KH2PO4, pH 7.4.

2.3  Immuno-­ fluorescence

1. Primary and secondary antibodies, determined by the experimental question. 2. PBST: 1× phosphate buffered saline, 0.3% Triton X-100 (v/v). 3. Milk powder. 4. Mounting medium (e.g., Mowiol).

3  Methods 3.1  Induction of Cre Recombination in Mouse Embryos

Systemic administration of Tamoxifen (or 4-OHT) by oral gavage or intraperitoneal injection is commonly used to induce recombination at different developmental time-points. Below we describe our standard protocol for administration of 4-OHT/Tamoxifen by intraperitoneal injection to pregnant females in order to induce Cre recombination in embryos. 1. Prepare 4-OHT/Tamoxifen stock solution (10 mg/mL) by dissolving the product directly into peanut oil (see Note 1). Protect the tube from light, place it on ice and sonicate in cycles of 40 s ON +30 s OFF to avoid overheating. Stop sonication as soon as 4-OHT/Tamoxifen is dissolved (see Note 2). 2. Once dissolved, 4-OHT/Tamoxifen can be used immediately or stored at 4 °C until needed (see Note 3). 3. Administer 4-OHT/Tamoxifen to pregnant females by intraperitoneal injection. The dose and dosage regimen should be optimized for each Cre line and experiment, to achieve the

Genetic Lineage Tracing of Lymphatic Endothelium

47

desired recombination efficiency and window of Cre activity (see above and Note 4). 4. Analysis of recombination in the skin and the mesentery is typically done 1 day after treatment, but it is possible to visualize genetically labeled cells already 8–12 h after treatment. 3.2  Analysis of Genetically Labeled Cells

Genetically labeled cells expressing a fluorescent reporter protein can be analyzed in the tissue of interest by immunofluorescence or flow cytometry. Here we provide a protocol for the analysis of embryonic skin and mesentery by whole-mount immunofluorescence.

3.2.1  Dissection and Preparation of Embryonic Back Skin (E13.5–E18.5) for Whole-­ Mount Immunofluorescence

1. Remove the uterus of the pregnant female and place it on a 10 mm plate containing PBS. With the help of dissection scissors and fine forceps release each embryo from the uterus leaving the yolk sac and the placenta attached to facilitate transferring of the embryo to the dissection plate (see Note 5). 2. Fill the dissection plate with PBS. Transfer the embryo and pin it down as shown in Fig. 3 (see Note 6). 3. With the embryo placed on the side, cut the skin as shown in Fig.  3, and carefully separate it from the embryo using fine forceps (see Note 7).

Fig. 3 Dissection and whole-mount analysis of embryonic back skin. (a) E15 embryo is immobilized to the dissection plate using insect pins. Dashed line indicates the outline of the skin sample to be dissected. (b) Back skin of an E15 embryo immobilized to the dissection plate using seven insect pins (white dots). Dorsal midline and different areas of the skin are indicated. Red boxes correspond to the areas imaged in (c). (c) Whole-­mount immunofluorescence of E14 wild type thoracic (upper image) and lumbar (lower image) back skin for Nrp2. Note the different developmental status of the different regions in the skin; the thoracic lymphatic vessels have a venous origin and form by sprouting while the vessels in the lumbar region assemble from nonvenous derived LEC progenitors (asterisks)

48

Ines Martinez-Corral and Taija Makinen

4. Once the skin is released from one side of the embryo, remove the pins and turn the embryo to the other side. Cut and separate the skin from this side and pull gently to completely detach the skin from the embryo. Take a small piece of tail for genotyping if needed. 5. Pin down the skin to the dissection plate as shown in Fig. 3, using 0.2 mm insect pins (see Note 8). 6. Remove PBS and fix with 4% PFA at RT for 2 h (see Note 9). 7. Wash with PBS. 8. Gently remove the muscle from the skin to prepare it for immunostaining. Remove the pins and continue to the staining ­protocol. Skins can be stored in PBS at 4 °C until analysis (see Note 10). 3.2.2  Dissection and Preparation of Embryonic Mesenteries (E13.5–E18.5) for Whole-­ Mount Immunofluorescence

1. Follow steps 1 and 2 of the skin dissection protocol to remove embryos and immobilize them to the dissection plate. 2. Cut out the gut tube and the associated mesentery (see Fig. 4) (see Note 11). 3. Pin down the gut tube and the associated mesentery to the dissection plate as shown in Fig. 4. Use 0.1 mm insect pins. 4. Remove PBS and fix with 4% PFA at RT for 1 h (see Note 9). 5. Wash with PBS.

Fig. 4 Dissection and whole-mount analysis of embryonic mesenteries. (a) E15 embryo on a dissection plate. At this stage the gut tube and the associated mesentery is localized outside the embryo (arrow). (b) Mesentery of an E15 embryo pinned down to the dissection plate for fixation. Intestine is indicated. Red box corresponds to the area imaged in (c). (c) Whole-mount immunofluorescence of E15 wild-type mesentery for PECAM1 (green) and Nrp2 (red). Single channel image for Nrp2 staining for lymphatic vessels is shown in the lower panel

Genetic Lineage Tracing of Lymphatic Endothelium

49

6. To proceed with immunofluorescence, unpin the mesentery from the plate and continue directly to the staining protocol. Alternatively, mesenteries can be stored in Eppendorf tubes in PBS at 4 °C until analysis (see Note 10). 3.3  Whole-Mount Immunofluorescence

1. Wash and permeabilize the tissues (skin or mesenteries) in PBST at RT for 10 min (see Note 12). 2. Block in PBST +3% (w/v) milk at RT for 1.5–2 h. 3. Incubate with the primary antibody in PBST +3% milk o/n at 4 °C (see Note 13). 4. Wash in PBST at RT for 2 h. Change the solution every 15–30 min. 5. Incubate with the secondary antibody in PBST +3% milk at RT for 2 h. 6. Wash in PBST at RT for 2 h. Change the solution every 15–30 min. 7. Mount the tissue on a microscope slide (see Note 14). 8. Analyze the samples by confocal microscopy (see Note 15).

4  Notes 1. Alternative protocols exist for dissolving Tamoxifen first in 100% Ethanol followed by dilution in oil. Different types of oil can be used, including corn oil, sesame oil and peanut oil. 2. The length of sonication time may vary depending on the model and parameters of the sonicator. Using out set up (Soniprep 150 (MSE) with sonication amplitude of 18 μm), the length of sonication time for 4-OHT stock is 30–40 min. Preparation should be checked every 15 min. Tamoxifen is more soluble in oil than 4-OHT and the length of sonication time is around 10 min. 3. If the 4-OHT stock solution has been stored at 4 °C it should be warmed briefly at 37 °C before injection. 4-OHT can be stored at 4 °C for several weeks when protected from light, however, if the color of the solution changes or precipitation forms it should be discarded. 4. We normally use 7–10 weeks old females on a C57BL/6 background. If females of a large range of ages are used, the dose of 4-OHT/Tamoxifen can be standardized per weight so that females within the same weight range receive the same dose.

50

Ines Martinez-Corral and Taija Makinen

5. Leaving the placenta attached to the embryo facilitates transferring of the embryos without risking any damage to the skin and the mesentery. If the placenta gets detached during ­dissection, a curved forceps or a spoon can be used to transfer them. To avoid inducing tissue damage, never use the forceps directly on to hold the embryo. 6. Embryos are sacrificed by decapitation [Note: The regulations governing the killing of laboratory animals may vary between different countries and the compliance of the selected method with the regulations should be confirmed]. For dissection of skin, leave the neck area intact to allow immobilization of the embryo to the dissection plate and dissection of skin covering this region. 7. It can be difficult to separate the skin from the underlying muscle. Removal of the muscle tissue should be done with care to avoid damaging the dermis with forceps. 8. A sufficient number of pins [7–8] should be used to allow unfolding and flattening of the skin. Over-stretching should be avoided as it may cause tearing of the skin and affect vessel morphology. 9. Fixation time and fixative may need to be optimized for different antibodies. 10. Fixed tissues that are stored in PBS at 4 °C can be used for several months if they are stored in optimal conditions (good fixation and sterile PBS). For long-term storage, sodium azide 4% can be added to the PBS to prevent bacterial growth. 11. Until approximately E15, the gut tube is herniated (Fig. 3). 12. Permeabilization, blocking, and washing steps are done in 2 mL Eppendorf tubes on a rocker. 13. Both primary and secondary antibody solutions are prepared in a volume of 200 μL, Incubation is done in 0.5 mL Eppendorf tubes on a rocker. 14. Different mounting mediums exist and may be used depending on the experimental requirements. For standard purposes we use home-made Mowiol. 15. Three-dimensional projections can be digitally reconstructed from confocal z-stacks to allow visualization of the entire vascular network. For imaging of the skin, a tile scan image covering an area of approximately 3400 μm × 1700 μm and centered in the midline of the back skin is recommended to provide sufficient information. It is important that the images of different embryos are taken from the same area of the skin, taking into account that the vasculature shows rostral to caudal progression and develops via different mechanisms in different

Genetic Lineage Tracing of Lymphatic Endothelium

51

regions of the skin (Fig. 3). For analysis of mesenteric lymphatic vessels, the different segments of the mesentery can be imaged separately (Fig. 4). Within the same mesentery, the developmental progression of vessel formation in the different segments can be different, thus it is recommended to image several segments per mesentery. References 1. Kretzschmar K, Watt FM (2012) Lineage tracing. Cell 148:33–45 2. Driessens G, Beck B, Caauwe A et al (2012) Defining the mode of tumour growth by clonal analysis. Nature 488:527–530 3. Chen J, Li Y, Yu T-S et al (2012) A restricted cell population propagates glioblastoma growth after chemotherapy. Nature 488:522–526 4. Schepers AG, Snippert HJ, Stange DE et al (2012) Lineage tracing reveals Lgr5+ stem cell activity in mouse intestinal adenomas. Science 337(6095):730–735 5. Jensen P, Dymecki SM (2014) Essentials of recombinase-based genetic fate mapping in mice. Methods Mol Biol 1092:437–454 6. Hsu Y-C (2015) The theory and practice of lineage tracing. Stem Cells 33:3197–3204 7. Martinez-Corral I, Ulvmar MH, Stanczuk L et al (2015) Nonvenous origin of dermal lymphatic vasculature. Circ Res 116:1649–1654 8. Stanczuk L, Martinez-Corral I, Ulvmar MH et al (2015) cKit lineage Hemogenic endothelium-­derived cells contribute to mesenteric lymphatic vessels. Cell Rep. https:// doi.org/10.1016/j.celrep.2015.02.026 9. Feil R, Wagner J, Metzger D et al (1997) Regulation of Cre recombinase activity by mutated estrogen receptor ligand-binding domains. Biochem Biophys Res Commun 237:752–757 10. Liu Y, Suckale J, Masjkur J et al (2010) Tamoxifen-independent recombination in the RIP-CreER mouse. PLoS One 5:e13533 11. Naiche LA, Papaioannou VE (2007) Cre activity causes widespread apoptosis and lethal anemia during embryonic development. Genesis 45:768–775 12. Smith L (2011) Good planning and serendipity: exploiting the Cre/lox system in the testis. Reproduction 141:151–161 13. Zovein AC, Hofmann JJ, Lynch M et al (2008) Fate tracing reveals the endothelial origin of hematopoietic stem cells. Cell Stem Cell 3:625–636

14. Reinert RB, Kantz J, Misfeldt AA et al (2012) Tamoxifen-induced Cre-loxP recombination is prolonged in pancreatic islets of adult mice. PLoS One 7:e33529 15. Ye R, Wang QA, Tao C et al (2015) Impact of tamoxifen on adipocyte lineage tracing: inducer of adipogenesis and prolonged nuclear translocation of Cre recombinase. Mol Metab 4:771–778 16. Hayashi S, McMahon AP (2002) Efficient recombination in diverse tissues by a tamoxifen-­ inducible form of Cre: a tool for temporally regulated gene activation/inactivation in the mouse. Dev Biol 244:305–318 17. Lizen B, Claus M, Jeannotte L et al (2015) Perinatal induction of Cre recombination with tamoxifen. Transgenic Res 24:1065–1077 18. Huh WJ, Khurana SS, Geahlen JH et al (2012) Tamoxifen induces rapid, reversible atrophy, and metaplasia in mouse stomach. Gastroenterology 142:21–24.e7 19. Abe T, Fujimori T (2013) Reporter mouse lines for fluorescence imaging. Develop Growth Differ 55:390–405 20. Madisen L, Zwingman TA, Sunkin SM et al (2010) A robust and high-throughput Cre reporting and characterization system for the whole mouse brain. Nat Neurosci 13:133–140 21. Muzumdar MD, Tasic B, Miyamichi K et al (2007) A global double-fluorescent Cre reporter mouse. Genesis 45:593–605 22. Snippert HJ, van der Flier LG, Sato T et al (2010) Intestinal crypt homeostasis results from neutral competition between symmetrically dividing Lgr5 stem cells. Cell 143:134–144 23. Pontes-Quero S, Heredia L, Casquero-Garcia V, Fernandez-Chacon M et al (2017) Dual ifgMosaic: A versatile method for multispectral and combinatorial mosaic gene-function analysis. Cell 170:800-814. https://doi.org/ 10.1016/j.cell.2017.07.031 24. Srinivasan RS, Dillard ME, Lagutin OV et al (2007) Lineage tracing demonstrates the venous origin of the mammalian lymphatic vasculature. Genes Dev 21:2422–2432

52

Ines Martinez-Corral and Taija Makinen

25. Klotz L, Norman S, Vieira JM et al (2015) 38. Srinivasan RS, Geng X, Yang Y et al (2010) The nuclear hormone receptor coup-TFII is Cardiac lymphatics are heterogeneous in origin required for the initiation and early mainteand respond to injury. Nature 522:62–67 nance of Prox1 expression in lymphatic endo 26. Kisanuki YY, Hammer RE, Miyazaki J et al thelial cells. Genes Dev 24:696–707 (2001) Tie2-Cre transgenic mice: a new model for endothelial cell-lineage analysis in vivo. Dev 39. Bazigou E, Lyons OTA, Smith A et al (2011) Genes regulating lymphangiogenesis control Biol 230:230–242 venous valve formation and maintenance in 27. Koni PA, Joshi SK, Temann U-A et al (2001) mice. J Clin Invest 121:2984–2992 Conditional vascular cell adhesion molecule 1 40. Martinez-Corral I, Stanczuk L, Frye M, deletion in mice. J Exp Med 193:741–754 Ulvmar MH, Diéguez-Hurtado R, Olmeda D, 28. Qu X, Zhou B, Baldwin HS (2015) Tie1 is Makinen T, Ortega S (2016) Vegfr3-CreERT2 required for lymphatic valve and collecting vesmouse, a new genetic tool for targeting the sel development. Dev Biol 399:117–128 lymphatic system. Angiogenesis 19:433–445. 29. Wu B, Zhang Z, Lui W et al (2012) Endocardial https://doi.org/10.1007/s10456-016cells form the coronary arteries by angiogenesis 9505-x through myocardial-endocardial VEGF signal 4 1. Connor AL, Kelley PM, Tempero RM (2016) ing. Cell 151:1083–1096 Lymphatic endothelial lineage assemblage dur 30. Monvoisin A, Alva JA, Hofmann JJ et al (2006) ing corneal lymphangiogenesis. Lab Investig VE-cadherin-CreERT2 transgenic mouse: a J Tech Methods Pathol 96:270–282 model for inducible recombination in the 42. Samokhvalov IM, Samokhvalova NI, Nishikawa endothelium. Dev Dyn 235:3413–3422 S (2007) Cell tracing shows the contribution of 31. Wang Y, Nakayama M, Pitulescu ME et al the yolk sac to adult haematopoiesis. Nature (2010) Ephrin-B2 controls VEGF-induced 446:1056–1061 angiogenesis and lymphangiogenesis. Nature 4 3. de Boer J, Williams A, Skavdis G et al (2003) 465:483–486 Transgenic mice with hematopoietic and lym 32. Claxton S, Kostourou V, Jadeja S et al (2008) phoid specific expression of Cre. Eur J Immunol Efficient, inducible Cre-recombinase activation 33:314–325 in vascular endothelium. Genesis 46:74–80 4 4. Klein S, Seidler B, Kettenberger A et al (2013) 33. Pham THM, Baluk P, Xu Y et al (2010) Interstitial cells of Cajal integrate excitatory Lymphatic endothelial cell sphingosine kinase and inhibitory neurotransmission with intestiactivity is required for lymphocyte egress and nal slow-wave activity. Nat Commun 4:1630 lymphatic patterning. J Exp Med 207:17–27 4 5. Foo SS, Turner CJ, Adams S et al (2006) 34. Lee LK, Ghorbanian Y, Wang W et al (2016) Ephrin-B2 controls cell motility and adhesion LYVE1 marks the divergence of yolk sac definiduring blood-Vessel-Wall assembly. Cell tive Hemogenic endothelium from the primi124:161–173 tive Erythroid lineage. Cell Rep 46. Zhou B, Ma Q, Rajagopal S et al (2008) 17:2286–2298 Epicardial progenitors contribute to the car 35. Onder L, Scandella E, Chai Q et al (2011) A diomyocyte lineage in the developing heart. novel bacterial artificial chromosome-­ Nature 454:109–113 transgenic podoplanin–Cre mouse targets lym 4 7. Saga Y, Miyagawa-Tomita S, Takagi A et al phoid organ stromal cells in vivo. Front (1999) MesP1 is expressed in the heart precurImmunol 2:50 sor cells and required for the formation of a 36. Gil HJ, Ma W, Oliver G (2018) A novel podosingle heart tube. Development planin-GFPCre mouse strain for gene deletion 126:3437–3447 in lymphatic endothelial cells. Genesis 56:e23102. https://doi.org/10.1002/ 48. Moses KA, DeMayo F, Braun RM et al (2001) Embryonic expression of an Nkx2-5/Cre gene dvg.23102 using ROSA26 reporter mice. Genesis 37. Wiszniak S, Scherer M, Ramshaw H et al 31:176–180 (2015) Neuropilin-2 genomic elements drive cre recombinase expression in primitive blood, 49. Jiang X, Rowitch DH, Soriano P et al (2000) Fate of the mammalian cardiac neural crest. vascular and neuronal lineages. Genesis Development 127:1607–1616 53:709–717

Genetic Lineage Tracing of Lymphatic Endothelium 50. Georgiades P, Ogilvy S, Duval H et al (2002) vavCre transgenic mice: a tool for mutagenesis in hematopoietic and endothelial lineages. Genesis 34:251–256 51. Gomez Perdiguero E, Klapproth K, Schulz C et al (2015) Tissue-resident macrophages orig-

53

inate from yolk-sac-derived erythro-myeloid progenitors. Nature 518:547–551 52. Ulvmar MH, Martinez-Corral I, Stanczuk L et al (2016) Pdgfrb-Cre targets lymphatic endothelial cells of both venous and non-­ venous origins. Genesis 54:350–358

Chapter 4 Visualization and Tools for Analysis of Zebrafish Lymphatic Development Kazuhide S. Okuda, Sungmin Baek, and Benjamin M. Hogan Abstract The accessibility and optical transparency of the zebrafish embryo offers a unique platform for live-imaging of developmental lymphangiogenesis. Transgenic lines labelling lymphatic progenitors and vessels enable researchers to visualize cellular processes and ask how they contribute to lymphatic development in genetic models. Furthermore, validated immunofluorescence staining for key signaling and cell fate markers (phosphorylated Erk and Prox1) allow single cell resolution studies of lymphatic differentiation. Here, we describe in detail how zebrafish embryos and larvae can be mounted for high resolution, staged imaging of lymphatic networks, how lymphangiogenesis can be reliably quantified and how immunofluorescence can reveal lymphatic signaling and differentiation. These methods offer researchers the opportunity to experimentally dissect developmental lymphangiogenesis with outstanding resolution. Key words Zebrafish, Lymphatic, Lymphangiogenesis, Live imaging, Immunofluorescence, Erk, Prox1

1  Introduction The initial description of lymphangiogenesis in zebrafish was made a little over a decade ago with the description of zebrafish thoracic duct development [1, 2]. Since then, the zebrafish has emerged as an excellent complementary vertebrate model for lymphatic research, allowing the study of complex lymphatic vessel networks and lymphangiogenesis at single cell resolution [3–5]. Similar to human lymphatics, zebrafish lymphatics lack blood flow, are able to take up and transport subcutaneously injected dye, and express orthologs of human lymphatic markers such as LYVE1 and PROX1 [1–3, 5–8]. Importantly, the molecular mechanisms controlling lymphatic development are largely conserved between humans and zebrafish, with zebrafish homologues of mammalian lymphatic genes such as vegfc [9–11], vegfd [12], vegfr3 [13], ccbe1 [14, 15], and prox1a [3] being essential for lymphatic development. The conservation of some lymphatic specification gene functions such Guillermo Oliver and Mark L. Kahn (eds.), Lymphangiogenesis: Methods and Protocols, Methods in Molecular Biology, vol. 1846, https://doi.org/10.1007/978-1-4939-8712-2_4, © Springer Science+Business Media, LLC, part of Springer Nature 2018

55

56

Kazuhide S. Okuda et al.

as for sox18 and nrf2f have been questioned [7]. However, most display conserved expression patterns and redundant functions are likely to play a role in zebrafish, as previously described for SoxF transcription factors [16–19]. Given the amenability of zebrafish to both forward and reverse genetic studies, it provides a powerful platform for identifying novel genes in lymphangiogenesis and for dissecting molecular mechanisms [4, 8, 15, 20–23]. The optical transparency of zebrafish embryos enables researchers to conduct real-time live imaging of lymphatic development using various transgenic lines that fluorescently label lymphatic vessels (see Table 1). The spatiotemporal resolution achieved from this can be used to dissect cellular mechanisms that drive early lymphatic development, answering questions that are hard to elucidate using the mouse model. For example, Yaniv and colleagues provided the first conclusive in vivo evidence that lymphatic endothelial cells (LECs) originate from veins using time-­lapse imaging in zebrafish. They showed that LECs in the thoracic duct originate in the parachordal vessel (also known as parachordal lymphangioblasts (PL)), which had been earlier shown to derive from the posterior cardinal vein (PCV) [1, 24]. Further characterization of this process using time-lapse imaging revealed that precursor LECs in the PCV are prox1a-positive and are bipotential, giving rise to LECs which progressively upregulate prox1a and venous endothelial cells (VECs) which eventually turn off prox1a expression [3]. While live-imaging offers a unique strength in zebrafish, immunofluorescence staining (IF) and analysis can also be conducted on whole zebrafish embryos. Recent studies have characterized expression of phosphorylated Erk and Prox1 using IF and have labelled key functional steps in lymphatic sprouting and specification [3, 4, 23, 32]. This technique is used in zebrafish mutant or transgenic lines to investigate whether modification of gene function results in altered Erk signaling or Prox1 expression in LECs. Of note, the Vegfc/Vegfr3 pathway, acting via downstream Erk, was shown to be both necessary and sufficient for Prox1 induction in the PCV [3]. Zebrafish with overexpression of vegfc show vastly increased nuclear Prox1-positive endothelial cells in the PCV in an Erk-dependent manner. In agreement, Shin and colleagues showed that flt4C562Δ mutants lack Prox1-positive LECs and that EC-specific activation of Erk results in increased nuclear Prox1-positive PCV ECs [32]. These early stages of lymphatic differentiation are particularly accessible for experimental observation using these simple readouts in zebrafish. In this chapter, we describe in detail how zebrafish embryos and larvae can be carefully mounted for high resolution confocal live imaging of lymphatic vasculature. We cover how zebrafish lymphatic vessels can be accurately quantified as a readout for perturbed developmental lymphangiogenesis. Instead of quantifying thoracic duct tissue fragments relative to body length, we describe

Visualization and Analysis of Zebrafish Lymphatics

57

Table 1 Commonly used transgenic lines for zebrafish lymphatic research Transgenic line/s

Utility

Tg(fli1a:EGFP)y1

Labels all endothelial Lymphatics can be difficult to cells including blood distinguish from blood vessels and lymphatic vessels without combining a second lymphatic or blood vascular restricted marker. Also labels macrophages and craniofacial neural crest cells

[1, 2, 25]

Tg(fli1a:nEGFP)y7

Can be used to quantify Lymphatics can be difficult to distinguish from blood vessels endothelial cell without combining a second number in blood and marker. Also, weak expression lymphatic vessels observed in the epidermis

[1, 3, 25]

TgBAC(flt4:Citrine)hu7135

Labels the trunk, facial lymphatic networks and major veins

Tg(kdrl:nlsmcherry)nz10

Can be used to quantify kdrl expression diminishes in lymphatic vessels over time. A endothelial cell second lymphatic marker is number in blood required to distinguish vessels and early lymphatics from blood vessels LECs (e.g., PLs)

Tg(-5.2lyve1b:DsRed2)nz101 Labels all zebrafish lymphatic networks Tg(-5.2lyve1b:EGFP)nz150 and major veins Tg(-5.2lyve1b:BFPCaax)uq18bh

Pitfalls

References

An additional blood vessel marker [7] is required to distinguish lymphatics from veins [5, 26, 27]

Weak expression in the fins. An additional blood vessel marker is required to distinguish lymphatics from veins

[5, 28]

Labels other tissues mosaically and blood vasculature weakly. An additional blood vessel marker is required to identify lymphatics

[29, 30]

Tg(map3k20nksagff27cGt; UAS:GFP)

Labels the trunk lymphatic network when in a heterozygous state

Tg(mrc1a:EGFP)y251

Labels the trunk, facial An additional blood vessel marker is required to and the superficial distinguish lymphatics from lymphatic network as veins well as major veins and macrophages

TgBAC(prox1a:KalTA44xUAS-­ADV. E1b:TagRFP)nim5

Labels the trunk and facial lymphatic networks. The only transgenic line that does not label blood vessels as well

[3, 7, 8] This transgenic line is highly expressed in nonvascular tissues such as muscle, eye and neural tube. Hence, a second marker is best used to help unequivocally identify lymphatics

TgBAC(stab1:YFP)hu4453

Labels the trunk lymphatic network and major veins.

Very weak at early stages (before 5 [15] dpf). An additional blood vessel marker is required to distinguish lymphatics from veins

[31]

58

Kazuhide S. Okuda et al.

how LECs in each trunk lymphatic vessel can be quantified to obtain an accurate readout of lymphatic development [3, 12]. Finally, we describe how IF can be conducted on zebrafish embryos to investigate two key drivers of lymphangiogenesis: phosphorylated Erk and Prox1 levels.

2  Materials 2.1  Zebrafish

2.2  Instrumentation

For live-imaging several transgenic zebrafish strains are suitable for examination of lymphatics (see Table 1) but the nuclear localized strain Tg(fli1a:nEGFP)y7 or equivalent should be used to allow quantification of LEC numbers. 1. Suitable stereomicroscope: We use the Nikon SMZ800 or the Leica MZ6 stereomicroscopes. 2. Suitable fluorescent stereo-dissecting microscope: We use the Leica M165 FC or the Nikon SMZ1500 fluorescent stereo-­ dissecting microscopes. 3. Suitable confocal microscope: We use the Zeiss LSM 710 inverted confocal microscope. 4. Objectives: For quantifying total LEC number in the trunk, we use either 10× (for imaging nine somites as shown in Fig. 2a, b) or 20× objectives (for imaging 4.5 somites). We use the Zeiss HC PL APO 10×/0.45 CS2 or the Zeiss HC PL APO 20×/0.8 CS objectives. We find that a 20× objective is best used for immunofluorescence analysis. Objectives with higher magnification can also be used depending on the image resolution required.

2.3  Small Equipment

1. Suitable tube heating block. 2. MatTek glass bottom dish (MatTek Corporation). 3. No. 1 superfine eyelash (ProSciTech).

2.4  Solutions and Reagents

1. E3 medium: To prepare a 60× stock of E3 medium, add 17.2 g of NaCl, 0.76 g of KCL, 2.9 g of CaCl2 2H2O, and 4.9 g of MgSO4 7H2O in 1 L of reverse osmosis (RO) water. The 60× stock can be diluted (1/60 dilution) to a working 1× E3 medium with RO water. 2. 1% (w/v) low melting point (LMP) agarose solution: Add 0.5 g of LMP agarose in 50 mL of E3 medium. Microwave for approximately 1 min until LMP agarose is dissolved and cool down at room temperature. Once sufficiently cooled, aliquot 1 mL of 1% LMP agarose solution to 1.5 mL Eppendorf tubes and store at 4 °C.

Visualization and Analysis of Zebrafish Lymphatics

59

3. Hank’s solution: To make Hank’s stock #1, add 8 g of NaCl and 0.4 g of KCl in 100 mL in 100 mL of RO water. To make Hank’s stock #2, add 0.358 g of Na2HPO4 anhydrous and 0.6 g of KH2PO4 in 100 mL of RO water. To make Hank’s stock #4, add 0.72 g of CaCl2 in 50 mL of RO water. To make Hank’s stock #5, add 0.601 g of MgSO4 in 50 mL of RO water. To make Hank’s stock #6, add 0.35 g of NAHCO3 in 10 mL of RO water. To make Hank’s premix, add 10 mL of Hank’s stock #1, 1 mL of Hank’s stock #2, 1 mL of Hank’s stock #4, 86 mL of RO water and 1 mL Hank’s stock #5 in order. To make Hank’s solution, mix 9.9 mL of Hank’s premix and 0.1 mL of Hank’s stock #6. Store Hank’s premix and Hank’s solution in 4 °C. 4. 0.3% (w/v) 1-phenyl-2-thiourea (PTU) stock solution: Add 3 g of PTU (Merck) into 1 L of Hank’s solution. Stir at 65 °C overnight to completely dissolve the PTU and keep at room temperature. The final concentration for PTU in E3 medium should be 0.003% (1/100 dilution of 0.3% PTU stock solution). 5. 50× tricaine stock solution: Add 4 g of tricaine methanesulfonate (Merck) and 10 g of Na2HPO4 to 1 L of E3 medium. Aliquot desired volumes and store at −20 °C. 6. Phosphate-buffered saline (PBS): Add 8 g of NaCl, 0.2 g of KCl, 1.44 g of Na2HPO4, 0.24 g of KH2PO4 to 1 L of sterile water and adjust pH to 7.4 with HCl. 7. 4% paraformaldehyde (PFA) in PBS: Add 40 g of paraformaldehyde to 1 L of PBS. Heat at 65 °C overnight to dissolve. Aliquot and store at −20 °C. 8. 3% (v/v) H2O2 in methanol: Add 1 mL of 30% (v/v) H2O2 in 9 mL of 100% methanol. Prepare fresh for each experiment and store in dark. 9. PBS with Tween 20 (PBST): Add 500 μL of Tween 20 in 500 mL of PBS. 10. 30% sucrose in PBST: Add 3 g of sucrose in 10 mL of PBST. 11. 10 mg/mL proteinase K stock solution: To prepare the reconstitution buffer, add 500 μL of 1 M Tris–HCl, pH 7.5, 1 mL of 1 M CaCl2 and 25 mL of glycerol to 23.5 mL of milli-Q water. To make the 10 mg/mL proteinase K stock solution, dissolve 100 mg of proteinase K (Thermo Fisher) in 10 mL of reconstitution buffer. 12. Tris-buffered saline with Tween 20 (TBST): To prepare a 10× TBS stock solution, add 12 g of Tris base, 44 g of NaCl into 500 mL of sterile water and adjust to pH 7.6 with HCl.

60

Kazuhide S. Okuda et al.

To make TBST, add 5 mL of 10× TBS stock solution, 500 μL of 10% Triton X (v/v) in sterile water, and 44.5 mL of sterile water. 13. Blocking buffer 1: Add 0.1 g of bovine serum albumin, 1 mL of deactivated horse serum in 9 mL of TBST. Prepare fresh for each experiment. 14. Anti-phospho-p44/42 MAPK (ERK1/2) (Thr202/Thr204) rabbit monoclonal antibody (Cell Signaling Technology, #4370). 15. Anti-PROX1 rabbit polyclonal antibody (Angiobio, 11-002). 16. Anti-GFP chicken polyclonal antibody (Abcam, ab13970). 17. Maleic acid buffer: To make 1 M maleic acid, add 17.41 g of maleic acid (Merck) into 1 L of sterile water. Adjust to pH 7.4–7.5 with 10 M NaOH. To make maleic acid buffer, add 7.5 mL of 1 M maleic acid, 1 mL of 5 M NaCl, and 5 μL of 10% Tween 20 (v/v) in sterile water, into 41.495 μL of sterile water. 18. Blocking buffer 2: Add 0.2 g of blocking reagent (Merck) to 10 mL of maleic acid buffer. 19. Goat anti-rabbit IgG, HRP-linked antibody (Cell Signaling, #7074). 20. Goat anti-chicken IgY secondary antibody, Alexa Fluor 488 (ThermoFisher, A-11039). 21. TSA™ Plus Cyanine 3 system (PerkinElmer).

3  Methods 3.1  Mounting Live Zebrafish Embryos/ Larvae (Herein Referred To as Specimens) for Imaging

Before starting, specimens with fluorescent expression (transgene-­ positive) should be selected for mounting using a fluorescent stereo-dissecting microscope. 1. Melt 1 mL of 1% LMP agarose solution (see Note 1) at 95 °C on a tube heating block for 20 min. Then, transfer the melted 1% LMP agarose solution to a tube heating block set at 42 °C. Leave the tube in this heating block for at least 10 min before starting step 4. 2. Transfer approximately 10–50 specimens to a petri dish with 25 mL E3 with 250 μL 0.3% (w/v) PTU stock solution (see Note 2). Add 500 μL 50× tricaine stock solution to anesthetize the specimens. 3. Transfer 8 anesthetized specimens (see Notes 3 and 4) to a 35 mm MatTek glass bottom dish. Remove as much E3 medium as possible but leave some as shown in Fig. 1a. 4. Add 10 μL 0.3% PTU stock solution and 20 μL 50× tricaine stock solution into the melted 1% LMP agarose solution from

Visualization and Analysis of Zebrafish Lymphatics

61

Fig. 1 Specimen mounting on MatTek glass bottom dish for imaging (a) Schematic representation of how 1% LMP agarose solution should be added to the 35 mm MatTek glass bottom dish containing anesthetized larvae. (b) Photo of a 35 mm MatTek glass bottom dish with 8 5 dpf larvae mounted in a lateral position

step 1 and mix by pipetting up and down several times. Carefully transfer this solution to the MatTek glass bottom dish from step 3 as shown in Fig. 1a (see Note 5). 5. Using a superfine eyelash or gently using fine forceps, adjust specimens laterally under a stereo microscope as shown in Fig.  1b (see Notes 6 and 7). Once specimens are immobile, leave the dish at room temperature for at least 5 min to properly solidify the agarose. 6. Specimens can now be imaged using your microscope of choice. For time-lapse imaging, E3 medium with 0.003% PTU (add 1/100 of 0.3% PTU medium) and 1× tricaine (add 1/50 of 50× tricaine) can be added to the dish to avoid drying (see Note 8). 3.2  Quantifying LEC Number in Developing Zebrafish Vasculature

Quantification of endothelial cell number in different lymphatic vascular beds has been shown to provide an accurate measure of the extent of successful zebrafish developmental lymphangiogenesis [3, 12]. Early studies typically used the gross tissue measurement of the extent of the thoracic duct in the trunk and expressed these measurements as number of fragments per body segment [1, 7, 15]. However, recent studies have shown that a near complete thoracic duct can form in the trunk, despite having vastly reduced numbers of LECs in the thoracic duct, PLs, intersegmental lymphatics, and dorsal longitudinal lymphatics [3, 12]. For example, in the thoracic duct, many fragments are still present in hypomorphic vegfchu5055 mutants; however, LEC numbers in the thoracic duct and other trunk lymphatic vessels are vastly reduced (see Fig. 2) [12]. This method for live imaging and quantification can

62

Kazuhide S. Okuda et al.

Fig. 2 Quantification of endothelial cell number in different lymphatic vascular beds provides an accurate measure of zebrafish lymphatic development (a, b) Confocal projection of 5 dpf Tg(lyve1:DsRed2)nz101;Tg(fli1a:n EGFP)y7 wild type larvae (a, a’ = lyve1:DsRed2 only) and vegfchu5055 mutant embryos (b, b’ = lyve1:DsRed2 only). (c–f) Quantification of thoracic duct fragments (c, n = 7 embryos), thoracic duct (TD) nuclei number (d, n = 7 embryos), intersegmental lymphatic vessel (ISLV) nuclei number (e, n = 7 embryos), or dorsal longitudinal lymphatic vessel (DLLV) nuclei number (f, n = 7 embryos) across nine somites. Although TD fragments in vegfchu5055 mutants were reduced by only ~10%, endothelial cell number in TD, ISLV, and DLLV were reduced by ~45%, ~53%, and ~62%, respectively. Error bars: SEM, ***P30 s), as this can lead to photobleaching and/or cell death. 4. Photoconverted Kaede will eventually degrade or be bleached by subsequent imaging. In our experience, using the lyve1b:Kaede transgenic, it is not feasible to track photoconverted cells beyond 2 days if embryos are photoconverted between 24 and 36 hpf, or beyond 3 days if embryos are photoconverted ≥72 hpf. If longer cell tracking is required, steps 2 and 3 may be repeated to reconvert structures with diminishing photoconverted Kaede signals. 5. If counting facial lymphatic cell number use a lymphatic reporter with a nuclear localization signal (i.e. lyve1b:nlsmCherry). 6. 2 mL microcentrifuge tubes that contain a tapered bottom are not suitable, as embryos are more likely to become curved/ bent, stick together and not be washed properly. 7. 10–20 embryos per tube is the recommended sample density per 2 mL microcentrifuge tube. 8. When removing solutions in between washes or incubations, use a 3 mL transfer pipette, taking care not to suck up embryos in the process. It is recommended to leave a small amount of liquid in the tubes rather than attempting to completely remove the solution, as this may cause damage to samples. 9. Wrapping tube racks with aluminum foil is an easy method for keeping samples in the dark once the secondary antibody/DAPI solution is applied. 10. If DAPI is required to visualize nuclei add 12.5 μg/mL DAPI to the secondary antibody solution.

82

Tiffany C. Y. Eng and Jonathan W. Astin

11. Mounting and imaging of immunostained samples is similar to Subheading 3.1, however embedding medium is 1% (w/v) Ultrapure low melting pointing agarose in PBS, and no tricaine or PTU is necessary. Once set, agarose is overlaid with PBS for imaging. Images are taken with a 20× or 60× water immersion lens with optical sections taken at 5 μm or 4 μm intervals, respectively.

Acknowledgments This work was supported by grants awarded to JWA (Marsden Fund—Royal Society of New Zealand, Health Research Council of New Zealand, and the Auckland Medical Research Foundation). References 1. Kuchler AM, Gjini E, Peterson-Maduro J, Cancilla B, Wolburg H, Schulte-Merker S (2006) Development of the zebrafish lymphatic system requires VEGFC signaling. Curr Biol 16(12):1244–1248. https://doi. org/10.1016/j.cub.2006.05.026 2. Yaniv K, Isogai S, Castranova D, Dye L, Hitomi J, Weinstein BM (2006) Live imaging of lymphatic development in the zebrafish. Nat Med 12(6):711–716. https://doi.org/10.1038/ nm1427 3. Hogan BM, Bos FL, Bussmann J, Witte M, Chi NC, Duckers HJ, Schulte-Merker S (2009) Ccbe1 is required for embryonic lymphangiogenesis and venous sprouting. Nat Genet 41(4):396–398. https://doi.org/10.1038/ ng.321 4. Hogan BM, Herpers R, Witte M, Helotera H, Alitalo K, Duckers HJ, Schulte-Merker S (2009) Vegfc/Flt4 signalling is suppressed by Dll4 in developing zebrafish intersegmental arteries. Development 136(23):4001–4009. https://doi. org/10.1242/dev.039990 5. Koltowska K, Lagendijk AK, Pichol-Thievend C, Fischer JC, Francois M, Ober EA, Yap AS, Hogan BM (2015) Vegfc regulates bipotential precursor division and Prox1 expression to promote lymphatic identity in zebrafish. Cell Rep 13(9):1828–1841. https://doi. org/10.1016/j.celrep.2015.10.055 6. Koltowska K, Betterman KL, Harvey NL, Hogan BM (2013) Getting out and about: the emergence and morphogenesis of the vertebrate lymphatic vasculature. Development

140(9):1857–1870. https://doi. org/10.1242/dev.089565 7. Okuda KS, Astin JW, Misa JP, Flores MV, Crosier KE, Crosier PS (2012) Lyve1 expression reveals novel lymphatic vessels and new mechanisms for lymphatic vessel development in zebrafish. Development 139(13):2381– 2391. https://doi.org/10.1242/dev.077701 8. Jung HM, Castranova D, Swift MR, Pham VN, Venero Galanternik M, Isogai S, Butler MG, Mulligan TS, Weinstein BM (2017) Development of the larval lymphatic system in zebrafish. Development 144(11):2070–2081. https://doi.org/10.1242/dev.145755 9. Kampmeier OT (1969) Evolution and comparative morphology of the lymphatic system. Thomas, London 10. Lawson ND, Weinstein BM (2002) In vivo imaging of embryonic vascular development using transgenic zebrafish. Dev Biol. 248(2):307–318 Epub 2002/08/09 11. Bussmann J, Bos FL, Urasaki A, Kawakami K, Duckers HJ, Schulte-Merker S (2010) Arteries provide essential guidance cues for lymphatic endothelial cells in the zebrafish trunk. Development. 137(16):2653–2657. Epub 2010/07/09. https://doi. org/10.1242/dev.048207 12. van Impel A, Zhao Z, Hermkens DM, Roukens MG, Fischer JC, Peterson-Maduro J et al (2014) Divergence of zebrafish and mouse lymphatic cell fate specification pathways. Development. 141(6):1228–1238. Epub 2014/02/14. https://doi.org/10.1242/ dev.105031

Zebrafish Facial Lymphatics 13. Dunworth WP, Cardona-Costa J, Bozkulak EC, Kim JD, Meadows S, Fischer JC et al (2014) Bone morphogenetic protein 2 signaling negatively modulates lymphatic development in vertebrate embryos. Circ Res. 114(1):56–66. Epub 2013/10/15. https://doi.org/10.1161/ CIRCRESAHA.114.302452

83

14. Astin JW, Haggerty MJ, Okuda KS, Le Guen L, Misa JP, Tromp A, Hogan BM, Crosier KE, Crosier PS (2014) Vegfd can compensate for loss of Vegfc in zebrafish facial lymphatic sprouting. Development 141(13):2680–2690. https://doi.org/10.1242/dev.106591

Chapter 6 Correlative Fluorescence and Scanning Electron Microscopy to Study Lymphovenous Valve Development Xin Geng and R. Sathish Srinivasan Abstract Lymph collected from throughout the body is exclusively returned to blood circulation via two pairs of bilaterally located lymphovenous valves. Lymphovenous valves share numerous similarities with lymphatic and venous valves and are defective in multiple mouse models of lymphedema or lymphatic dysfunction. Here we describe a protocol that combines the strengths of fluorescence microscopy and scanning electron microscopy to precisely locate and analyze the topography of developing lymphovenous valves at high resolution. Key words Lymphatic vasculature, Lymphovenous valves, Scanning electron microscopy, Confocal microscopy, Correlative microscopy, Venous valves

1  Introduction The lymphatic vasculature collects interstitial fluid and digested lipids and returns it to blood circulation bilaterally at the junction of jugular and subclavian veins [1]. A pair of LVVs is present at each of the junctions [2, 3]. LVVs are the gatekeepers that regulate lymph return to blood while preventing blood from entering the lymphatic vessels [4]. Furthermore, LVVs are defective in multiple mouse models of lymphatic dysfunction [2, 5, 6]. Therefore, characterization of LVVs (along with that of lymphatic vessels and lymphatic valves) must be an integral part of any phenotypic analysis. LVVs develop in a stereotypic manner at the jugulosubclavian junction, which is one of the largest (if not the largest) vessels in the entire body. The size of this vessel and the predictability of LVV development provide a unique opportunity for imaging and characterization. We recently developed an approach to study the developing LVVs by synergizing the strengths of optical-microscopy and electron-microcopy techniques  [2]. Fluorescent immunohistochemistry followed by light microscopy permits the identification of specific cell types such as LVV-forming endothelial cells Guillermo Oliver and Mark L. Kahn (eds.), Lymphangiogenesis: Methods and Protocols, Methods in Molecular Biology, vol. 1846, https://doi.org/10.1007/978-1-4939-8712-2_6, © Springer Science+Business Media, LLC, part of Springer Nature 2018

85

86

Xin Geng and R. Sathish Srinivasan

(LVV-ECs). SEM provides high resolution, high magnification and three-dimensional images of a sample. We first performed fluorescence microscopy using LVV-EC specific markers to locate LVVs. Subsequently the same samples were processed and imaged by SEM to obtain topographical information about LVVs at cellular level. This correlative microscopy protocol detects phenotypic abnormalities in LVVs with great accuracy and clarity. LVVs share numerous structural and molecular similarities with lymphatic and venous valves [7]. In fact, three venous valves develop very close to LVVs [2, 3]. Therefore development of LVVs could be compared side-by-side with that of venous valves.

2  Materials Prepare all solutions using ultrapure water. Prepare and store all solutions at room temperature, unless specified. 2.1  Vibratome Section of Tg(Prox1-­ tdTomato) Embryos

1. Tg(Prox1-tdTomato) mice (MMRRC, Cat# 036531-UCD) [8]. 2. 20× PBS: Weigh 160 g NaCl, 4 g KCl, 28.8 g Na2HPO4, and 4.8 g KH2PO4. Transfer to a 1 L glass beaker. Add water to a volume of 900 mL. Mix and adjust pH to 7.4 with NaOH. Make up to 1 L with water. Autoclave. 3. 20% paraformaldehyde (PFA): Weigh 200 g PFA and transfer to a 1 L glass beaker (see Note 1). Add 500 mL water and stir with a magnetic stirrer on a hot plate in a fume hood. Heat the PFA suspension to 55–60 °C (never over 60 °C). Add 5 N NaOH in a drop wise manner until the suspension turn clear. Cool down the solution to room temperature. Add 50 mL of 20× PBS and adjust pH to 7.4 by adding 5 N NaOH in a dropwise manner. Make up to 1 L with water. Aliquot the solution with 50 mL falcon tubes and store at −20 °C. 4. 20% NaN3: Weigh 20 g NaN3 and dissolve in 100 mL water. 5. 7% low melting point agarose: Weigh 3.5 g low melting point agarose in 250 mL media bottle. Add 50 mL 1× PBS. In a 900w microwave, set cooking time to 30 min and power level to 1. Most of the agarose will melt after 30 min. If clumps of agarose remain then heat at full power in short pulses and check frequently. Keep melted agarose at 60 °C. 6. Tissue-Tek Cryomold (10 mm × 10 mm × 5 mm). 7. Double edged blade. 8. Peel-A-Way disposable tissue (22 mm × 30 mm) (see Note 2).

embedding

9. Leica VT1000 S vibration blade microtome. 10. 24-well plate.

molds

SEM Analysis of Lymphovenous Valves

87

11. Instant glue. 12. No. 10 flat paintbrush. 13. Fluorescent stereomicroscope and camera. 2.2  Immunostaining of Vibratome Sections

1. 10% Triton X-100: Mix 10 mL Triton X-100 with 90 mL water. Filter through 0.45 μm filter. 2. 1× PBS + 0.3% Triton X-100 (1× PBST): Mix 50 mL 20× PBS and 30 mL 10% Triton X-100 in a 1 L cylinder. Make up to 1 L with water. 3. 5% BSA: Dissolve 5 g BSA with 100 mL water by stirring. Filter through 0.45 μm filter. Aliquot and store at −20 °C. 4. Donkey serum. 5. Blocking buffer: Add 400 μL 5% BSA and 500 μL donkey serum to a 15 mL centrifuge tube. Make up to 10 mL with 1× PBST. 6. For Fig. 5a we used goat anti-mouse LYVE1 and rat anti-­mouse CD31 as primary antibodies. We used Alexa 488-conjugated donkey anti-goat and Cy5-conjugated donkey anti-rat as secondary antibodies. Other markers such as VEGFR3, NRP2, endomucin, and VE-Cadherin can also be used. However, the sample fixation condition will have to be optimized individually. 7. Aluminum foil. 8. 1-in. white vinyl tape. 9. Scalpel. 10. Microscope cover glass. 11. Mounting media. 12. Confocal microscope.

2.3  Scanning Electron Microscopy Analysis of Immunostained Vibratome Sections

1. 10 cm petri dish. 2. 0.1 M sodium cacodylate buffer: Dilute 0.4 M sodium cacodylate buffer with water. 3. 2% glutaraldehyde: Dilute 25% glutaraldehyde solution with 0.1 M sodium cacodylate buffer. 4. 1% osmium tetroxide (OsO4): Dilute 4% osmium tetroxide solution with 0.1 M sodium cacodylate buffer. 5. Ethanol: Dilute 100% ethanol with 0.1 M sodium cacodylate buffer to prepare graded ethanol (50%, 70%, 90%, and 95%). 6. Hexamethyldisilazane (HMDS). 7. 12 mm diameter carbon adhesive tabs. 8. Aluminum SEM mount slotted head. 9. #5 forceps.

88

Xin Geng and R. Sathish Srinivasan

10. Sputter coater. 11. Scanning electron microscope. 12. Specimen mount holder box. 13. Desiccator.

3  Methods Carry out all procedures at room temperature, unless otherwise specified. 3.1  Vibratome Section of Tg(Prox1-­ tdTomato) Embryos

1. Dissect E12.0-E18.5 Tg(Prox1-tdTomato) embryos from the uterus and remove extra embryonic membranes in 1× PBS (see Note 3). 2. Fix embryos in 4% PFA at 4 °C (see Note 4 for fixation time). 3. Wash embryos three times with 1× PBS, 30 min each time. Embryos can be stored in 1× PBS with 0.01% sodium azide at 4 °C for up to 1 month. 4. Prepare 7% low melting point agarose in 1× PBS. 5. Embed embryos in 7% low melting point agarose (see Note 5) (Fig. 1). 6. Modify the vibratome setting according to the developmental stage of the sample (see Note 6). 7. Remove agarose block from the mold, trim the block and glue it to the sample holder with instant glue (see Fig. 1). 8. Fill a 24-well plate with 1× PBS and place it on ice (see Note 7). 9. Section the samples according to the instructions of the vibratome manual. 10. Pick up sections using the paintbrush, transfer to the 24-well plate and visualize under the fluorescent stereomicroscope. 11. Sections containing the LVVs are chosen based on tdTomato expression (Figs. 2 and 3). 12. Sections can be stored in 1× PBS with 0.01% sodium azide at 4 °C for up to 1 month (see Note 8).

3.2  Immunostaining of Vibratome Sections

Immunostaining protocol is modified from Yang et al. [9]. Steps 1–7 are carried out using a shaker with agitation. During these steps the samples are kept immersed in the appropriate solutions in individual wells of a 24-well plate. 1. Permeabilize the samples by adding 1× PBST to the sections. Incubate at 4 °C for 1 h. 2. Incubate the sections in blocking buffer at 4 °C for 2 h.

SEM Analysis of Lymphovenous Valves

89

Fig. 1 Sample embedding in agarose block. (a) E12.0 mouse embryo embedded in agarose in a disposable mold. (b) Agarose block is trimmed and ready to be glued to the sample holder. (c) The black dotted line indicates the side that will be glued to the sample holder. Midline of the E12.0 embryo (red dotted line) is parallel to the bottom of the block (black dotted line). (d) The dotted black line indicates the position where an E16.5 embryo must be trimmed. (e) Two trimmed E16.5 embryos are embedded in agarose in a disposable mold. (f) A trimmed block with an E16.5 embryo that is ready to be glued to the sample holder. (g) The black dotted line indicates the side that will be glued to the sample holder. The midline of E16.5 embryo (red dotted line) is vertical to the bottom of the block (black dotted line)

Fig. 2 Anterior to posterior serial cross sections of E14.5 embryos and P0 pups. (a–d) Bright field images of serial 300 μm thick cross sections of an E14.5 Tg(Prox1-tdTomato) embryo. (e–h) Bright field images of serial 500 μm thick cross sections of a P0 Tg(Prox1-tdTomato) pup. Black arrows in panels (c) and (g) indicate the location of LVVs and the inset in panels (c) and (g) are overlay images of bright field and tdTomato fluorescent images of a LVV complex

90

Xin Geng and R. Sathish Srinivasan

Fig. 3 Lateral to medial serial sagittal sections of E12.0 embryos. (a, b) Bright field images of serial 800 μm thick sagittal sections of an E12.0 Tg(Prox1-tdTomato) embryo. (c) tdTomato fluorescent image of panel (b). Black arrow in panel (b) and white arrow in panel (c) indicate the LVV complex

3. Add CD31 and LYVE1 antibodies diluted with blocking buffer and incubate the sections at 4 °C overnight. 4. Wash the sections 3 times with 1× PBST, 30 min each time at 4 °C. 5. Add Cy5 anti-rat and Alex 488 anti-goat antibodies diluted with blocking buffer, cover the plate with foil and incubate for 2 h. 6. Wash the sections 3 times with 1× PBST, 30 min each time at 4 °C. 7. Rinse the sections 2 times with 1× PBS. 8. Prepare the imaging slides using cover glass (see Note 9 and Fig. 4). 9. Fill the chambers in the imaging slide with mounting media. Transfer one section to the chamber using the paintbrush (Fig. 4d). 10. Use another cover glass to cover the chamber. The sample identification number can be written on the corner of the cover glass (Fig. 4d). 11. Image the samples using a confocal microscope (Fig. 5a). 3.3  Scanning Electron Microscopy Analysis of Immunostained Vibratome Sections

Steps 5–11 should be performed in a fume hood. 1. After taking confocal images put the cover glass–sample–cover glass sandwich into a 10 cm petri dish filled with 1× PBS and incubate at 37 °C for 30 min. 2. Mounting media will reliquefy releasing the cover glasses and the sample.

SEM Analysis of Lymphovenous Valves

91

Fig. 4 Preparing spacers and mounting samples for imaging. (a) Five layers of white 1″ vinyl tape are adhered to the cover of a culture plate. The midline (arrowhead), edges (arrows) and the chambers (red arrows) are outlined. (b) The edges of the imaging spacer are trimmed and the chambers are cut out using a scalpel. (c) An imaging spacer is peeled out from the plate and adhered to a cover glass. (d) 500 μm E16.5 sections are placed inside the chambers filled with mounting media. A second cover glass is placed over the spacer and the setup is ready for imaging

Fig. 5 Correlative fluorescence and electron microscopy images of E14.5 LVV complex. (a) Confocal image of an LVV complex of E14.5 Tg(Prox1-tdTomato) embryo. White arrows indicate the LVVs. (b) SEM image of the section in panel a. The two LVVs are pseudo colored in magenta. Abbreviations: T thymus, IJV internal jugular vein, EJV external jugular vein. Scale bars: 50 μm. Image adopted with permission from Geng et al. [2]

3. By using the paintbrush transfer the sample to a well of 24-well plate filled with 1× PBS. 4. Wash the samples 2 times with 1× PBS and once with 0.1 M cacodylate buffer. 5. Fix the samples with 2% glutaraldehyde in 0.1 M sodium cacodylate buffer for 2 h. 6. Wash the samples 3 times with 0.1 M sodium cacodylate buffer, 15 min each time (see Note 10). 7. Remove the agarose around the sections by using forceps. 8. Fix the samples with 1% OsO4 in 0.1 M sodium cacodylate buffer for 2 h.

92

Xin Geng and R. Sathish Srinivasan

Fig. 6 Mounting dehydrated samples to SEM mount slotted head. (a) Write sample ID on the back of the slotted head (arrowhead). Arrow points to the top of the head. (b) Paste the carbon adhesive tab (arrowhead) on the top of the slotted head (arrow) and place it on the specimen mount holder (red arrow). (c) Samples are mounted to the adhesive tabs, sputter-coated with Au-Pd particles, and ready to be imaged by SEM

9. Wash the samples 3 times with 0.1 M sodium cacodylate buffer, 15 min each time (see Note 10). 10. Dehydrate the samples in steps by incubating the samples in increasing concentrations of ethanol: 50%, 70%, 90%, 95% and 3 times in 100%, 15 min for each step. 11. Incubate the samples 2 times in HMDS for further dehydration, 10 min each time. 12. Remove HMDS, uncover the plate and dry the samples overnight. 13. Mount the samples onto SEM mount (see Note 11 and Fig. 6). 14. Using a sputter coater coat the samples with gold–palladium (Au-Pd) particles. 15. Image the samples using scanning electron microscope (Fig. 5b). 16. For long-term storage, the samples can be placed in specimen mount holder boxes and stored inside a desiccator.

4  Notes 1. Wear mask to avoid breathing paraformaldehyde dust. 2. We normally put two embryos in the same mold. Smaller diameter-mold can be used if only one embryo is embedded. 3. Start with Tg(Prox1-tdTomato) embryos to locate the LVVs. Embryos without reporters can be used once you are familiar with the anatomical landmarks around LVVs. Embryonic day (E) 12.0-to-postnatal day (P) 5 mice have been analyzed using this protocol. It becomes progressively difficult to section pups due to bone mineralization.

SEM Analysis of Lymphovenous Valves

93

4. The fixation time depends on the age of the embryos. In general, E12.0–E15.5 embryos are fixed overnight. For embryos older than E15.5 and for postnatal pups the abdominal cavity and the skin on the chest are opened before fixation for 48 h. Additionally, for E18.5 embryos and postnatal pups the samples are washed 3 times with 1× PBS after the first day of fixation. Then the skin is removed before fixing for another 24 h. 5. The position of LVVs rotates clockwise from being parallel to the midline at E12.0 to vertically above the superior vena cava (SVC) at E14.5 [2]. Therefore, sagittal section of the cardinal vein followed by imaging of the inner surface of the cardinal vein is the best way to visualize LVVs at E12.0 and E12.5. From E14.5 onward, the best way to visualize the LVVs is by transverse sectioning the samples, which exposes the branch point of internal jugular vein (IJV), external jugular vein (EJV) and subclavian vein (SCV). Protocol for embedding E12.0 and E12.5 embryos for sagittal section: pour agarose into a 10 mm × 10 mm × 5 mm mold. Transfer the embryo from PBS to a tissue paper and pat dry the embryo gently using another tissue paper. Transfer the embryo into the mold filled with agarose, position the embryo using a 200 μL pipette tip to let it lie on its side and make sure that the midline of the embryo is parallel to the bottom of the mold. Place the mold on a cold surface to accelerate the agarose solidification and keep checking the position of the embryo until the agarose is solidified (Fig. 1a). To embed embryos from E14.5 onward and postnatal pups: pour agarose into a 22 mm × 30 mm mold, transfer the embryo from PBS to a tissue paper and pat dry the embryo gently using another tissue paper. Cut the embryo transversely below the ribcage using a razor blade (Fig. 1d), transfer the embryo into the mold filled with agarose, position the embryo using a 200 μL tip to let it stand on the cut surface and make sure that the midline of the embryo is vertical to the bottom of the mold. Transfer another embryo to the mold following the steps described previously (optional). Place the mold on a cold surface to accelerate the agarose solidification and keep checking the position of the embryo until the agarose is solidified (Fig. 1e). If agarose gets solidified during embedding process, it can be reliquefied by heating in a microwave oven and cooled down before use. If the position of the embryo is not optimal, wait until the agarose is fully solidified, cut the solidified agarose using razor blade, remove all the agarose from the embryo and reembed. 6. The setting of the vibratome is based on the hardness of the sample. The older the embryos and pups, the harder they

94

Xin Geng and R. Sathish Srinivasan

Table 1 Vibratome setting for sectioning mouse embryos and pups Stages

Amplitude

Frequency

Speed

E12.0-E12.5

0.8 mm

6

8

E14.5-E16.5

0.8 mm

6

7

E18.5-P0

0.8 mm

8

6

P5

1 mm

10

3

become. As the samples get harder increase the frequency and the amplitude, and reduce the speed. The setting of Leica VT1000 S vibratome for samples at various stages is listed in Table 1. The thickness of sections must be optimized to capture the entire LVVs while also avoiding sectioning through the structures. Additionally, LVVs must be as close to the sectioning plane as possible. LVVs that are located too deep inside the SVC may not be visualized by SEM. Normally, 300 μm thickness is used to section E14.5 embryos (Fig. 2a–d) and 500 μm thickness is used to section E16.5 or older samples (Fig. 2e– h). Based on our experience, the distance between the LVVs to the surface of E12.0 and E12.5 embryos is approximately 800  μm. To section E12.0 to E12.5 embryos, 300 μm to 500  μm thickness can be used first. Once the limbs are sectioned off from the embryo, increase the thickness to 800 μm (Fig. 3a–c). The thickness can be adjusted based on the distance between the last section plane to the surface of the embryo and the size of the embryo. 7. 12-well plate can be used while sectioning postnatal samples. 8. If immunohistochemistry is not required for the study, Subheading 3.2 can be skipped and continue to step 4 in Subheading 3.3. 9. Adhere 1″ white vinyl tape onto the cover of a 12- or 24-well plate. The length of the plates is exactly equal to that of two cover slips. Make sure that the tape is parallel to the edge of the plate (Fig. 4a). The thickness of tape is approximately 100 μm. Stick multiple layers of tape according to the thickness of the section. For example, if the section is 500 μm, five layers of tape will be used. Draw lines on the tape along the edges and the center (Fig. 4a, arrows and arrowhead respectively). Outline the chambers on the tape based on the size of the section (Fig. 4a, red arrows). Cut along the lines using a scalpel (Fig. 4b). Remove one stack of tapes from the plate and adhere it to a cover glass (Fig. 4c).

SEM Analysis of Lymphovenous Valves

95

10. After washing, if a stopping point is preferred, store the samples at 4 °C. 11. The samples are very fragile after dehydration. Therefore extreme care is necessary during sample manipulation. Write the sample identification number at the bottom of the SEM mount slotted head (Fig. 6a, arrowhead). Paste the carbon adhesive tab (Fig. 6b, arrowhead) on the top of the SEM mount slotted head and place the mount on the mount holder (Fig.  6b, arrow). Transfer the samples from 24-well plate to a 10 cm petri dish using forceps by holding the edge of the sample. Check the sample under stereoscope and determine the surface that needs to be imaged. With that side up, transfer the sample from the petri dish onto the adhesive tab on the mount (Fig. 6c). The samples might have bent after dehydration. Therefore, the samples need to be positioned with an angle to make sure the imaging surface is parallel to the mount head and will be fully exposed to the electron beam. As the tab is very sticky, the sample cannot be repositioned after adhering to the tab.

Acknowledgments This work is supported by NIH/NHLBI (R01HL131652 and R01HL133216 to RSS), Oklahoma Center for Adult Stem Cell Research (4340) to RSS, NIH/NIGMS COBRE (P20 GM103441 to XG; PI: Dr. Rodger McEver) and American Heart Association (15BGIA25710032 for RSS). References 1. Tammela T, Alitalo K (2010) Lymphangiogenesis: molecular mechanisms and future promise. Cell 140(4):460–476. https://doi.org/10.1016/j.cell.2010.01.045 2. Geng X, Cha B, Mahamud MR, Lim KC, SilasiMansat R, Uddin MK, Miura N, Xia L, Simon AM, Engel JD, Chen H, Lupu F, Srinivasan RS (2016) Multiple mouse models of primary lymphedema exhibit distinct defects in lymphovenous valve development. Dev Biol 409(1):218–233. https://doi.org/10.1016/j. ydbio.2015.10.022 3. Srinivasan RS, Oliver G (2011) Prox1 dosage controls the number of lymphatic endothelial cell progenitors and the formation of the lymphovenous valves. Genes Dev 25(20): 2187–2197. https://doi.org/10.1101/gad. 16974811

4. Hess PR, Rawnsley DR, Jakus Z, Yang Y, Sweet DT, Fu J, Herzog B, Lu M, Nieswandt B, Oliver G, Makinen T, Xia L, Kahn ML (2014) Platelets mediate lymphovenous hemostasis to maintain blood-lymphatic separation throughout life. J Clin Invest 124(1):273–284. https:// doi.org/10.1172/JCI70422 5. Kazenwadel J, Betterman KL, Chong CE, Stokes PH, Lee YK, Secker GA, Agalarov Y, Demir CS, Lawrence DM, Sutton DL, Tabruyn SP, Miura N, Salminen M, Petrova TV, Matthews JM, Hahn CN, Scott HS, Harvey NL (2015) GATA2 is required for lymphatic vessel valve development and maintenance. J Clin Invest 125(8):2979–2994. https://doi. org/10.1172/JCI78888 6. Martin-Almedina S, Martinez-Corral I, Holdhus R, Vicente A, Fotiou E, Lin S,

96

Xin Geng and R. Sathish Srinivasan

Petersen K, Simpson MA, Hoischen A, Gilissen C, Jeffery H, Atton G, Karapouliou C, Brice G, Gordon K, Wiseman JW, Wedin M, Rockson SG, Jeffery S, Mortimer PS, Snyder MP, Berland S, Mansour S, Makinen T, Ostergaard P (2016) EPHB4 kinase-inactivating mutations cause autosomal dominant lymphatic-related hydrops fetalis. J Clin Invest 126(8):3080– 3088. https://doi.org/10.1172/JCI85794 7. Geng X, Cha B, Mahamud MR, Srinivasan RS (2017) Intraluminal valves: development, function and disease. Dis Model Mech 10(11):1273–1287. https://doi.org/10.1242/ dmm.030825

8. Gong S, Zheng C, Doughty ML, Losos K, Didkovsky N, Schambra UB, Nowak NJ, Joyner A, Leblanc G, Hatten ME, Heintz N (2003) A gene expression atlas of the central nervous system based on bacterial artificial chromosomes. Nature 425(6961):917–925. https://doi.org/10.1038/nature02033 9. Yang Y, Garcia-Verdugo JM, Soriano-Navarro M, Srinivasan RS, Scallan JP, Singh MK, Epstein JA, Oliver G (2012) Lymphatic endothelial progenitors bud from the cardinal vein and intersomitic vessels in mammalian embryos. Blood 120(11):2340–2348. https://doi. org/10.1182/blood-2012-05-428607

Chapter 7 Characterization of Mouse Mesenteric Lymphatic Valve Structure and Function Amélie Sabine, Michael J. Davis, Esther Bovay, and Tatiana V. Petrova Abstract Intraluminal valves of collecting lymphatic vessels ensure unidirectional lymph transport against hydrostatic pressure gradient. Mouse mesentery harbors up to 800 valves and represents a convenient model for lymphatic valve quantification, high resolution imaging of different stages of valve development as well as for analysis of valve function. The protocol describes embryonic and postnatal mesenteric lymphatic vessel preparation for whole-mount immunofluorescent staining and visualization of valve organization, quantification of main morphological parameters such as valve size and leaflet length, and the quantitative assessment of functional properties of adult valves using back-leak and closure tests. Key words Mesentery, Collecting vessel, Lymphatic valve

1  Introduction Lymphatic valves are semilunar bileaflet structures abundantly present in mature lymphatic collecting vessels (see Fig. 1). Each side of the leaflet is covered by a monolayer of lymphatic endothelial cells apposed on a thin sheet of specialized extracellular matrix. Lymphatic valves usually have two leaflets of the same size. The convex side of the leaflet is attached to the vessel wall, forming the valve annulus that is reinforced at the commissure by the valve buttress, while the semilunar part remains free and is oriented in the direction of lymph flow (see Fig. 1). Movement of lymph in the opposite direction pushes the valve leaflets together, and thus prevents retrograde flow. The main role of lymphatic valves is to direct, in concert with the pumping activity of lymphatic smooth muscle cells, the flow of lymph toward the draining lymph nodes in peripheral lymphatic vascular networks or toward the connection of central lymphatic vessels such as the thoracic duct or the right lymphatic duct with the blood circulation (reviewed in [4, 5]). Proper lymphatic valve organization and function is therefore indispensable for the functionality of the entire lymphatic system and failure of Guillermo Oliver and Mark L. Kahn (eds.), Lymphangiogenesis: Methods and Protocols, Methods in Molecular Biology, vol. 1846, https://doi.org/10.1007/978-1-4939-8712-2_7, © Springer Science+Business Media, LLC, part of Springer Nature 2018

97

98

Amélie Sabine et al.

Fig. 1 Lymphatic valve organization and the associated lymph flow patterns. The leaflet is covered on both sides by Prox1HIGH LECs, which are more compact on the sinusal side subjected to recirculating flow and more elongated on the luminal side subjected to high laminar shear stress [1, 2]. A core of specialized ECM is sandwiched between the two layers of LECs. Leaflets are inserted within the vessel wall at the annulus, a fibrous circumferential fold rich in collagens and elastin, which changes shape during the valve cycle [3].

lymphatic valve development underlies human hereditary diseases of lymphatic vessels, such as lymphedema-distichiasis, caused by mutations in the transcription factor FOXC2 [6, 7]. Collecting lymphatic vessels and valves are formed during late stages of embyogenesis in response to increased lymph flow [8]; this process continues during postnatal growth period and is reactivated during tissue regeneration [7, 9]. Previous studies established that valve LECs are molecularly distinct from lymphangion LECs [5, 7, 8, 10–13]. In particular valve LECs produce high levels of the transcription factors Prox1, Foxc2, Gata2, and Nfatc1 [8, 14], and the ECM components and integrin receptors laminin alpha5, FN-EIIIA, and Itga9 [11], which, together with other pan-endothelial markers such as PECAM1, can be used for the identification and characterization of lymphatic valve structures. Given the complexity of valve organization, visualization in 3D is mandatory for the assessment of eventual lymphatic valve phenotype, which is otherwise difficult if not impossible to characterize using conventional staining approaches on thin sections. Due to its flat structure, ease of preparation, and relatively stereotypic organization, the mouse mesentery became a tissue of choice for the analyses of lymphatic collecting vessel and valve organization [7, 8, 10, 11, 14, 15] (see Figs. 2 and 3). Here, we provide a protocol optimized for whole-mount analysis of the mesenteric lymphatic vessels of mouse pups at postnatal day 4–14 (see Fig. 4). A shorter

Mouse Mesenteric Lymphatic Valve Analysis

99

Fig. 2 Scheme of a dissected anterior mesentery, unfolded clockwise from duodenum (Duo) to ileum highlighting its main components: intestine, mesentery, lymph node, and five vascular branches. The black dashed box indicates the vascular branch area selected for lymphatic vessel length and valve quantifications. The orange and pink dashed lines indicate the vascular branch segment that is selected for analysis of valves from collecting vessels and precollectors, respectively.

Fig. 3 Bright-field pictures of a P6 neonate mesentery taken with a dissecting stereomicroscope. (a) Mesenteric loop. Mes mesentery, LN lymph node, In intestine. The red dashed box is magnified in (b). Scale bar: 5 mm. (b) Mesenteric vascular branch. Red arrowheads: lymphatic valves. The red dashed box is magnified in (c). Scale bar: 1 mm. (c) Lymphatic collecting vessel filled with milky chyle. Red arrows: lymphatic valves

version of this protocol is also provided for analysis of embryonic lymphatic vessels (see Note 1). Furthermore, as even morphologically normal lymphatic valves may be dysfunctional, we include a protocol for quantitative assessment of the mechanical properties of adult mesenteric valves using back-leak and closure tests.

100

Amélie Sabine et al.

Fig. 4 Flowchart of the method for whole-mount staining and imaging of postnatal mouse mesentery

2  Materials 2.1  Mice

The procedure described can be used for mesenteries of mice up to P14. If mice are treated prior to mesentery collection, see Note 2. All procedures involving animal experiments should follow approved institutional and governmental animal protocols and comply with the relevant guidelines and regulations of the local animal ethics committee.

2.2  Mesentery Dissection from Mouse Neonates

1. Dissecting microscope with reflected light illumination and a zoom magnification of at least 10×. 2. Dissecting board and needles (e.g., 26 G1/2, 0.45 × 12 mm). 3. Dissection tools: sharp straight surgical scissors (41 mm sharp edge, or more), microscissors (e.g., cutting edge: 5 mm), serrated forceps (e.g., tip: 1.3 × 1 mm), fine forceps (e.g., tip: 0.05 × 0.02 mm). 4. Stainless steel insect pins (e.g., Austerlitz Insect Pins, diameter: 0.2 mm; tip: 0.02 mm). 5. 70% ethanol spray. 6. 12-Well elastomer-coated cell culture plate. Coat wells with a thick (3–5 mm) layer of liquid elastomer (Sylgard® 184 Silicone elastomer kit, Dow Corning) and dry overnight under a fume hood.

Mouse Mesenteric Lymphatic Valve Analysis

101

7. PBS: dilute from a sterile stock of 10× PBS using autoclaved distilled water, store at 4 °C. 8. PBS-A: 0.1% sodium azide (w/v) dissolved in PBS (see Note 3), store at 4 °C. Sodium azide is a hazardous reagent; use a chemical fume hood and wear protective gloves and mask when preparing the solution. 9. 1 ml micropipette and corresponding tips. 10. Hazardous liquid waste container. 2.3  Mesentery Preparation for Whole-­ Mount Staining

1. Horizontal shaker at 4 °C. 2. 4% PFA: dissolve 4% paraformaldehyde (w/v) in PBS at 50 °C while stirring, filter using 0.45 μm filter and store aliquots at −20 °C. After thawing store at 4 °C and use within 7 days. Paraformaldehyde is a hazardous reagent; use a chemical fume hood and wear protective gloves and mask when preparing the fixative. 3. 10% sucrose solution: dissolve 10% d(+)-sucrose (w/v) in PBS-A with stirring, filter sterilize and store at 4 °C. 4. 20% sucrose solution: dissolve 20% d(+)-sucrose (w/v) and 10% glycerol (v/v) in PBS-A with stirring, filter sterilize and store at 4 °C.

2.4  Immunostaining

1. Blocking buffer: 0.5% bovine serum albumin (w/v), 5% donkey serum (v/v) (see Note 4) and 0.5% Triton X-100 (v/v) in PBS-­A. Filter sterilize and store at 4 °C for maximum 3 months. 2. Washing buffer: 0.5% Triton X-100 (v/v) in PBS, store at 4 °C. 3. Primary antibodies (see Table 1). 4. Secondary antibodies (see Note 4). 5. P20 and P10 micropipettes, and corresponding tips. 6. Aluminium foil. 7. Stereomicroscope with epifluorescence illumination

2.5  Sample Mounting on Slides

1. Glass microscope slides (e.g., Thermo Scientific, 76 × 26 mm). 2. Glass coverslips (e.g., Menzel-Gläser, 24 × 24 mm, #1). 3. Adhesive spacers (e.g., diameter: 20 mm; depth: 0.12 mm deep). 4. Mounting medium with DAPI (e.g., ProLong Gold antifade reagent from Life Technologies). 5. P200 micropipette and corresponding tips.

2.6  Imaging and Image Analysis

1. Microscope with epifluorescence illumination (e.g., Zeiss Axiovision equipped with a Standard HBO arc lamp and Dapi, FITC, DsRED and Alexa 660 filter cubes).

LEC membrane (enriched in cell-cell junctions) All ECs

Pecam-1

All ECs

Claudin-5

Collecting vessels, arteries and veins

αSMA

Basement membrane of lymphangions and blood vessels Basement membrane of blood vessels

Laminin α5 Leaflet core ECM

ECM at the free edge of valve leaflet

Valve LEC membrane anchored to ECM: Valve ≫ Lymphangion

LEC membrane (endocytosed upon ligand binding): Valve ≫ Lymphangion

FN-EIIIA

Integrin α9

Vegfr-3

All LECs

Arteries and veins, low in collecting vessels

Basement membrane of all vessels

Collagen IV Valve annulus ECM and valve basement membrane: Valve  Lymphangion

All LECs

LEC nuclei: Valve leaflet cells ≫ Lymphangion cells

Prox1

Other mesenteric components

Lymphatic valve compartment

Antigen

Staining

Table 1 Examples of lymphatic valve immunostaining and the list of antibodies used

Reference

–

–

–

Fig. 9b

Fig. 9c

Fig. 9a

Fig. 10b

Fig. 10c

–

0.5 μg/ml

1 μg/ml

1 μg/ml

1 μg/ml

1 μg/ml

Concentration

R&D, #AF743

R&D, #AF3827

Clone FN-3E2; Sigma-­ Aldrich, F6140

See Refs. 8, 11

Millipore, #AB756P

2 μg/ml

0.5 μg/ml

1 μg/ml

2 μg/ml

Clone 1A4; Sigma-Aldrich, 1 μg/ml #C6198

Clone 192,703; R&D, #MAB1556

Invitrogen, #34-1600

R&D, #AF1002

Figs. 9a Clone MEC13.3; BD and 10a Pharmingen, #553370

Fig. 10b–d R&D, #AF2727

Figure

Antibodies

102 Amélie Sabine et al.

Mouse Mesenteric Lymphatic Valve Analysis

103

2. Confocal microscope (e.g., Leica SP5 Tandem equipped with 3 lasers and confocal detection channels: 488, 543 and 633). 3. Image analysis software for confocal images (z-stack projection, 3D reconstruction, etc) (e.g., Imaris, Bitplane). 4. Image processing software (area, length, event and staining intensity measurement) (e.g., ImageJ, https://imagej.nih. gov/ij/). 2.7  Valve Function Analysis

See Chapter 15 for additional details. 1. Krebs solution supplemented with 0.5% BSA. 2. Sharpened forceps and microscissors. 3. Dissection chamber, recessed into table. 4. Fine wire for pinning. 5. Two glass micropipettes with tip diameters 30–50 μm and lightly fire-polished tips. 6. Two pipette holders with micromanipulator mounting system. 7. 12-0 monofilament suture or teased strands of 2-0 monofilament suture. 8. Two 10 ml syringes + 0.8 μm filters + two 3-way stopcocks + two 1 ft lengths of PE-190 tubing. 9. Dissection microscope, magnification range: 8–64×. 10. Dual fiber optic illuminator. 11. Heated perfusion chamber for ex vivo experimental studies. 12. Inverted microscope. 13. Bath perfusion pump. 14. Three low-pressure transducers and amplifiers. 15. Pressure control system. 16. Data acquisition software.

interface,

computer

and

controlling

17. Servo-nulling pressure measurement system. 18. Camera and diameter measurement system.

3  Methods 3.1  Lymphatic Valve Imaging 3.1.1  Mesentery Dissection

1. Euthanize the mouse pup by decapitation. Lay the body on its back on a dissecting board and stretch it flat by applying a needle in each limb (see Fig. 5b). Wipe the belly with 70% ethanol. 2. Open the abdominal cavity with a small incision in the skin under the sternum. Insert one tip of the scissors between the

104

Amélie Sabine et al.

Fig. 5 Mesentery collection procedure. (a) Tamoxifen is injected SC in the neck of a P6 mouse neonate. (b) The euthanized mouse is pinned on a dissecting board. (c) The abdominal skin is detached. (d) The peritoneum is removed to free the intestine. (e) The colon is cut above rectum. (f) The duodenum is cut under stomach. The intestine and mesentery are pulled out from the abdominal cavity (g) and transferred to PBS-A in a 12-well elastomere-coated plate (h)

skin and the abdominal wall. Open the skin by making a mid-­ line incision down to the pelvis. Make transverse incisions on both sides under the diaphragm and along the two hind limbs. Pin skin flaps to the dissecting board (see Fig. 5c). 3. Grab the abdominal wall with forceps and cut it open over the entire abdominal cavity (see Fig. 5d). Pay attention not to damage the internal organs during the procedure. 4. Push the intestines upward to get free access to the colon and cut it above the rectum (see Fig. 5e). Then, push the intestine downward to get free access to the duodenum and cut it below the stomach (see Fig. 5f). It is important that the intestine is completely severed at both extremities. 5. Insert serrated forceps in an open position under the intestine, from the stomach toward the pelvis, grab it from below (see Fig. 5g) and pull it out firmly to detach the mesenteric lymph node from underlying tissues. 6. Transfer the intestine to PBS-A in an elastomere-coated 12-well plate on ice (see Fig. 5h). It should be entirely covered with PBS-A to prevent drying. 3.1.2  Mesentery Pinning

Work under a dissection microscope for precise manipulation of the tissue. Mesentery should be pinned immediately after collection from the animal (see Note 5). Avoid directly grasping the mesentery with forceps to prevent tissue damage.

Mouse Mesenteric Lymphatic Valve Analysis

105

Fig. 6 Mesentery dissection procedure. (a) The duodenum is pinned on elastomere to allow further clockwise unfolding of the intestine. (b) Clockwise unfolding gives access to the colon, which is excised above the caecum. (c) The pancreas (white diffuse tissue spread in the mesentery at the top of the image) is excised. (d) The mesentery is further unfolded to make a complete wheel. (e) The remaining part of the intestine is cut off. (f) The upper half of the lymph node is excised. (g) Small holes are made at the connection of each vascular branch with the intestine. (h) Blood and chyle are flushed out from mesenteric vessels

1. Identify the duodenum and orientate the intestine to allow clockwise unfolding of the gut (see Note 6). 2. Pin down the duodenal extremity to the elastomere (see Fig. 6a). 3. Progressively unfold the mesentery in clockwise direction by applying more pins, approximately at each bend of the i­ ntestine (see Fig. 6b–e). After inserting a pin half through the intestine at a bend, gently transport it away from the previous pin and from the mesenteric lymph node to sufficiently stretch the mesentery while avoiding rupture (see Note 7). 4. Identify the colon extremity, grasp it with forceps and cut the colon mesentery to detach it from the small intestine and mesenteric lymph node. Discard colon and caecum by cutting the intestine above the caecum (see Fig. 6b). 5. When the first wheel is half-completed, identify the pancreas located at the duodenal extremity as a soft and diffuse white tissue spread throughout the duodenal mesentery, partly attached to the mesenteric lymph node (see Fig. 6c). Using small scissors remove the pancreas by cutting the mesentery after the first vascular branch that follows the pancreas and discard it (see Note 8). 6. Continue unfolding the intestine and the mesentery to complete the wheel (see Fig. 6d).

106

Amélie Sabine et al.

Fig. 7 Mesenteric vessels before (left) and after (right) flushing with PBS-A. Red arrow: blood-filled vessel; white arrowhead: chyle-filled lymphatic vessel

7. Cut the mesenteric lymh node in two halves with microscissors; the anterior half will remain as the central “hub” of this first wheel, while the posterior half will constitute the “hub” of a second mesenteric wheel. Further cut the mesentery between two vascular branches, and eventually the intestine (see Fig. 6e). The posterior mesentery should detach entirely from the anterior one (see Note 9). 8. Once a wheel is complete, dissect out excess of tissues around the mesenteric lymph node and remove upper half of the lymph node using microscissors (see Fig. 6f) (see Note 10). 9. With microscissors, make small incisions in the mesentery at the end of each large vessel at the transition between the mesentery and the intestine (see Fig. 6g). 10. Replace PBS-A. 11. Using a 1 ml pipette, flush each vascular branch with PBS-A in lymph flow direction, i.e., from the intestine toward the lymph node (see Fig. 6h). Repeat until most blood and chyle are washed out from vessel lumens (see Fig. 7). 12. Rinse twice with cold PBS-A and keep in PBS-A. 13. Between each mesentery, invert the plate over a liquid waste container and refill it with cold PBS-A. 14. When all mesenteries in a plate are prepared, rinse twice with cold PBS, and then replace PBS with cold 4% PFA under a ventilated hood. Plates with pinned mesenteries are kept on ice during the following procedure. All further incubations are done at 4 °C on a rotating platform. Attention should be paid not to allow tissues to dry up (see Note 11).

Mouse Mesenteric Lymphatic Valve Analysis 3.1.3  Mesentery Preparation for Immunostaining

107

1. Fix mesenteries for 6 h. 2. Under a ventilated hood, discard PFA and replace it with cold PBS (see Note 12). 3. Rinse three times with PBS, then wash 3 × 20 min with PBS-A. 4. Incubate for 2 h in 10% sucrose solution (see Note 13). 5. Incubate for 2 h in 20% sucrose solution (see Note 13). 6. Rinse three times with PBS, then wash 3 × 1 h with PBS-A. 7. Tissues can be stored in PBS-A at 4 °C up to 3 months.

3.1.4  Immunostaining

1. Rinse tissues twice with PBS-A (see Note 14). 2. Incubate for at least 8 h with blocking buffer. 3. Prepare the mix of primary antibodies in blocking buffer (see Table 1) and store it on ice. Enough mix should be prepared to allow complete immersion of the samples (e.g., 1 ml per well in a 12-well plate). 4. Rinse tissues once with blocking buffer to remove any floating tissue pieces. 5. Add primary antibody mix, making sure that the tissue is completely immersed. 6. Incubate overnight. 7. Discard primary antibody mix one well after another to prevent mixing antibodies (see Note 15), and wash 5 × 1 h with washing buffer. 8. Prepare secondary antibody mix in blocking buffer (see Note 16). Keep away from light. 9. Add secondary antibody mix, filling one well after another to prevent drying of tissues. Wrap the plate with aluminium foil to protect from light, from this step on. 10. Incubate overnight. 11. Discard secondary antibody mix (see Note 17) and wash 3–10 times for 30 min with washing buffer. The number of washes is antigen- and antibody-dependent. Check the staining after three washes using a fluorescent stereomicroscope to decide whether additional washes are necessary. If background staining of the surrounding tissues is high, up to ten washes can be done. If the specific staining is low, stop after five washes. 12. Rinse twice with PBS and refix for 2 days with 4% PFA (see Note 18). 13. Rinse 3 times with PBS, then 3 × 1 h with PBS-A. Tissues can be kept for up to 1 month in PBS-A at 4 °C away from light before mounting without loss of quality (see Note 19).

108

Amélie Sabine et al.

Fig. 8 Procedure of mesentery mounting on microscopy slide. (a) The intestine is cut off the mesentery at the junction between the two. Arrowhead: detachment of the mesentery (b) Microscopy slide is prepared with a central spacer. Mesentery is grasped at the two external extremities (arrowheads) (c) and transferred to the slide within the spacer well, when the opaque liner is still present (d). Asterisk: empty well. (e) Opaque liner is removed and mesentery is covered with mounting medium to fill in the spacer well. Arrowheads: well filled with mounting medium. A coverslip (arrowhead) is applied to a sticky border of the spacer (f), let to fall on the mesentery and tightly attached to the spacer (g). (h) Mounting medium is added to remove air bubbles under the coverslip

3.1.5  Tissue Mounting on Slides

1. Rinse tissues once with PBS. 2. Under a dissection microscope remove the intestine using microscissors, cutting as close as possible to the gut wall (see Fig. 8a). Discard the intestine and the pins. 3. Rinse mesenteries three times with PBS to remove traces of sodium azide. 4. Prepare slides with spacers. To place a spacer (see Note 20 and Fig. 8b) remove the transparent liner from the spacer without touching the adhesive, attach the spacer at the center of the slide as flat as possible, eliminate air bubbles first by applying a firm pressure with your finger on the thick opaque liner on each side, then using a dissection tool (e.g., scissors) as a roller to ensure complete adhesion to the slide. Mark the slides. 5. Seize the mesentery at the two extremities using two forceps (see Fig. 8c) and transfer it on the slide in the same orientation (see Note 21). 6. Under the stereomicroscope unfold the mesentery using closed forceps. Touch tissue as little as possible to avoid damaging the vessels (see Note 22).

Mouse Mesenteric Lymphatic Valve Analysis

109

7. When the mesenteric wheel is unfolded flat on the slide (see Fig. 8d), drain the excess of PBS with a tissue paper. Remove the thick opaque liner. 8. Quickly add 3 drops of mounting medium containing DAPI on top of the mesentery (see Note 23). 9. Gently tilt and rotate the slide to fill in entirely the room delimited by the spacer (see Fig. 8e). Remove all air bubbles using a 200 μl pipette. 10. Under the microscope, raise each of the two mesentery extremities to allow mounting medium to flow beneath. Carefully replace the mesentery in the wheel position. Add a new drop of mounting medium on top of the mesentery. 11. Apply a coverslip by placing one of its borders to the spacer and letting coverslip fall on the mesentery (see Fig. 8f). Quickly apply pressure with fingers on the coverslip borders to ensure adhesion to the spacer (see Fig. 8g). 12. Gently release the pressure on the coverslip while, if necessary, adding more mounting medium using a 200 μl pipet to prevent the formation of bubbles (see Fig. 8h). 13. Release completely the pressure. Check that all four borders of the coverslip are attached to the spacer. 14. Store the slide in a horizontal position at 4 °C for 24 h to allow polymerization of the mounting medium. 15. Control the absence of air bubbles under the coverslip and refill with mounting medium if necessary (see Note 24). Slides can be kept at 4 °C for up to 4 weeks without loss of fluorescence (see Note 25). 3.1.6  Valve Imaging and Analysis

The number and size of valves varies depending on the collecting vessel location in the mesenteric vasculature; therefore, the anterior and posterior mesenteries are analyzed separately. Careful mapping within the mesentery at low magnification (see Notes 26 and 27) and examination at high magnification are necessary to allow optimal selection of valves for quantification and unbiased comparison (see Table 2). 1. Quantification of lymphatic valve number. This analysis is done with mesenteries stained for Prox1. (a) Imaging. Using epifluorescence microscope take a tiled picture of the entire mesenteric wheel comprising 3–5 vascular branches (see Note 28). If tiling is not an option, take a picture of three entire vascular branches per mesentery (see Fig. 2). (b) Lymphatic vessel length. Using ImageJ software, adjust the image levels to visualize all Prox1+ LECs. Determine the

110

Amélie Sabine et al.

Table 2 Examples of applications and criteria of fluorescent microscopy imaging of mesenteric lymphatic valves. Illustrations are shown in Figs. 9 and 10. n.a. not applicable (High-magnification imaging of valves necessitates confocal microscopy, due to their multilayered organization.)

Imaging techniques and AIMS Magnification (objective)

Microscope Epifluorescence

Confocal fluorescence

10–20×

• Number of valves per vessel length • Valve localization within vessel

• Number of valves per vessel length • Valve localization within vessel

20–40×

• Valve size • Valve coverage (mural cells, basement membrane)

• Leaflet length • Leaflet core matrix • Cellular organization within the valve (cell shape, cell neighbors, intercellular spreading)

>40×

n.a.

• Valve cell number • Valve cell phenotype (identity, shape, junctions, cytoskeleton) • Valve buttress •  Valve cell event (e.g., proliferation)

lymphatic vessel length of each branch by using the segmented line tool and measure action. Calculate the total lymphatic vessel length by adding up the data for all branches analyzed. (c) Number of valves. Using ImageJ software, adjust the image levels to visualize Prox1HIGH valves and Prox1LOW lymphangions. Quantify manually the number of valves (Prox1HIGH regions) for each branch. Calculate the total number of valves by adding up data for all branches analyzed. (d) Graph. Plot the total number of valves per total lymphatic vessel length (mm) (see Note 29). 2. Quantification of valve size and leaflet length. This analysis is done with mesenteries costained for Prox1 and a leaflet marker such as an ECM component laminin alpha5 (Lama5) (see Fig. 9b), an adhesion receptor to the core matrix such as integrin alpha9 (Itga9) or a constituent of the glycocalyx that accumulates in high-shear regions such as podocalyxin (see Fig. 10b). (a) Valve selection. Given that Lama5, Itga9, and podocalyxin antibodies also stain blood vessels, select valves that do not overlap with the blood vasculature (see Fig. 2). (b) Imaging. Using a confocal microscope take a 40×-picture of the valves in an orientation that allows clear distinction

Mouse Mesenteric Lymphatic Valve Analysis

111

Fig. 9 Examples of mesenteric lymphatic vessels and valve ECM stainings. (a) Collecting vessel bifurcation stained to analyze its coverage by mural cells. Pecam1 in green, alphaSMA in red. Scale bar: 100 μm. (b) Valve leaflet core matrix stained for laminin alpha5. Scale bar: 30 μm. (c) Valve matrix stained for collagen IV. Scale bar: 50 μm. Arrowhead, lymphatic valve devoid of mural cells; arrow: valve buttress; asterisk: luminal free edge of the valve leaflets; (a) valve length; (b) leaflet length

of the leaflets (see Fig. 11 and Note 30). Acquire 3–5 pictures per mesentery. (c) Reconstruct each valve as a 2D z-stack picture using ImageJ or Imaris softwares and export the image, e.g., as a TIFF file. (d) Exclusion of precollectors. Using ImageJ software, discard the outer one-third of the mesentery, as it comprises most precollectors (see Fig. 2 and Note 31). (e) Valve length. Using ImageJ segmented line tool and measure action measure the valve length, defined as the vessel portion associated with high levels of Prox1 and Lama5/ Itga9/podocalyxin (indicated as “a” in Fig. 11). This quantitative measure can be useful when there is almost complete degeneration of leaflets. (f) Leaflet length. Select one z-slice layer from the z-stack (e.g., using the slice view tool of Imaris) that shows at the same time the most external sinus margin of the valve annulus and the free edge of the corresponding leaflet. Export the corresponding image to ImageJ and measure the leaflet length, (indicated as “b” on Fig. 11). (g) Graph. Calculate the averaged valve and leaflet lengths to obtain an average value for one animal (n = 1). 3. Analysis of lymphatic valve endothelial cells.

112

Amélie Sabine et al.

Fig. 10 Imaging of valve LECs. (a) Valve stained for Pecam1 to show valve cell shape. Arrow: elongated cell; arrowhead: rounder valve cell. Scale bar: 20 μm. (b) Valve cells stained for Prox1 and podocalyxin to highlight the sinusal free edge of the valve leaflets. Arrowhead: Prox1HIGH cell at the free edge; arrows: increased deposition of podocalyxin at the free edge. Scale bar: 20 μm. (c) Staining for Prox1 and claudin5 to show intercellular junctions (arrowhead) between valve cells. Scale bar: 25 μm. (d) Valve cell proliferation showed by staining for Prox1 and Ki67. White staining: Lama5. Arrowhead: proliferating Ki67+/Prox1HIGH valve cell; arrow: proliferating Ki67+/Prox1LOW lymphangion cell; asterisk: Ki67+/Prox1neg proliferating non-LEC. Scale bar: 20 μm

This analysis is done with mesenteries costained for Prox1 and other markers of your choice (see Note 32). Analysis of valve cell proliferation using Ki67 as a marker of cell proliferation is described below (see Fig. 10d). (a) Imaging. Under a confocal microscope, take a 40×-picture of the well-defined valves that are located away from the

Mouse Mesenteric Lymphatic Valve Analysis

113

Fig. 11 Overview of different possible valve appearances, under the microscope and on 2D pictures, depending on their orientation and opening/closure status. Blue arrows show lymph flow direction. The main valve component, which can help in their orientation (sinus, annulus, buttress and leaflet) are depicted. (a) Valve length; (b) leaflet length

blood vessels. Collect 3–5 pictures per mesentery. Lymphatic valves should be selected as described above (see above steps 1 and 2). (b) Reconstruct each valve as a 3D picture using Imaris software. Export the 2D z-stack image as an illustration (e.g., TIFF file). (c) Proliferating valve cells. Using the 3D view tool of Imaris, by rotating and magnifying the 3D image, identify and count the double-positive Prox1HIGH/Ki67+ cells (see Fig. 10d and Note 33). (d) Data analysis. The number of proliferating valve cells per valve can be averaged for all valves to obtain a mean value per animal. Alternatively, given the low number of proliferating valve LECs, normalize the total number of Prox1HIGH/Ki67+ cells to the number of valves that have been analyzed. For both measurements, a single value per animal should be used for statistical analysis. 3.2  Lymphatic Valve Function Analysis 3.2.1  Mesenteric Vessel Dissection

Dissect and cannulate a mouse lymphatic vessel using the methods described in Chap. 15; for mesenteric lymphatics use the protocol described for popliteal afferent lymphatics with the following variations. 1. After the mouse is anesthetized and has reached a surgical plane of anesthesia, make a midline incision in the abdomen, about 2 cm long.

114

Amélie Sabine et al.

2. Turn the mouse on its side and exteriorize a 5–6 cm loop of intestine, preferably duodenum. 3. Pin it to a Sylgard platform using 00 insect pins. The dissection and cannulation steps will be much easier if the vessel is filled with chyle, which is more likely to be present if the animal has eaten or been gavaged with olive oil 30–60 min prior to surgery. 4. Locate an appropriate segment containing 2–4 valves (see Fig. 3b); the larger diameter vessels in the central arcades near the root of the mesenteric node will be easiest to cannulate. 5. Remove the surface fat with the coarse microscissors, being careful not to cut too close to the vessel wall. 6. Cut the thin layer of mesothelium along each side of the vessel and then cut each vessel end, starting with the proximal end first to increase the likelihood of retaining chyle in the lumen. 7. Transfer the vessel to a Sylgard dissection dish filled with Krebs-­BSA solution (see Chap. 15) and pin it down using short pieces of 40 μm stainless steel wire. 8. Under pseudo-oblique illumination (see Chap. 15) carefully remove most of the fat, especially at the two ends. It is usually best not to remove all the fat until the vessel is cannulated and pressurized. 9. Transfer the vessel to the cannulation chamber, filled with Krebs-BSA solution, with prefilled cannulation pipettes, all mounted to a portable stage (see Chap. 15). Keep the vessel from floating during cannulation by weighing one end down with a 1–1.5 cm long piece of 40 μm wire. 3.2.2  Cannulation and Cleaning

1. Cannulate the input end as described in Chapter 15. 2. Tease open the output end using the cannulation forceps and temporarily raise the inflow pressure from 3 to 10 cm H2O for 30 s to flush out the lumenal contents. 3. Return the inflow pressure to 3 cm H2O and cannulate the outflow end of the vessel. 4. Once both ends are cannulated and pressurized to 3 cm H2O, increase the axial length until there is no lateral bowing and use the 45° angled cannulation forceps and fine microscissors to remove the remaining fat. Be particularly careful near the fragile valve sinuses. 5. Once the fat is removed the vessel should be shortened to one valve for valve tests. Starting with a segment that contains 2–3 valves provides flexibility in the event that one part of the vessel wall is accidently nicked or damaged during cleaning. 6. If all valves are intact and the wall is undamaged the easiest way to shorten to one valve is to loosen the tie on the inflow

Mouse Mesenteric Lymphatic Valve Analysis

115

pipette and advance it through the appropriate number of valves, leaving a single valve near the outflow pipette. The inflow pipette can only be advanced ∼500  μm at a time (depending on the taper angle of the pipette) and will retract unless it is retied. After retying, cut the vessel off around the shank of the inflow pipette before advancing it further. Repeat this procedure of advancing and cutting as necessary to obtain a one-valve segment. Be sure to leave a reasonable length of vessel (200–300 μm) for micropuncture on the input side of the remaining valve. 3.2.3  Valve Orientation

1. Rotate the vessel so that both valve leaflets are visible, with the valve buttresses [16] on the top and bottom surfaces (“V-shape” in Fig. 11). To rotate it, loosen the suture at one end and turn the vessel by pulling gently on its two sides behind the loose tie. Retie it and repeat this step at the other end. (If only one end is rotated in an attempt to orient the valve the vessel likely will twist during subsequent protocols as outflow pressure is raised, so rotate the vessel only ½ way at each end). 2. Carefully remove any remaining mesothelium, fat or connective tissue along the inflow end of the vessel, paying particular attention to the top surface of the vessel near the pipette where the wall will be micropunctured. 3. Secure the ties and transfer the vessel, pipette assembly, and cannulation chamber to the stage of an inverted microscope (see Chapter 15).

3.2.4  Vessel Perfusion

1. Disconnect the syringe lines from the back ends of the cannulation pipettes/holders and connect them to a water-filled, adjustable reservoir system (see Chapter 15) set to 3 cm H2O. Make sure that one port of the 3-way stopcock is open to atmosphere when disconnecting or connecting each line. 2. Connect the water bath to the water jacket of the cannulation chamber. 3. Begin perfusing the chamber with Krebs at 0.5 ml/min and allow the vessel to equilibrate to 37 °C for 30–60 min. Mesenteric vessels will develop small amplitude vasomotion but not spontaneous contractions [17]. Mesenteric valve tests can therefore be conducted in Krebs solution, but the contractions of lymphatics from other regions must be stopped for valve tests by perfusion with Ca2+-free Krebs solution. 4. Calibrate the pressure transducers by connecting them through a T-connector to a single, adjustable reservoir to ensure that both the inflow and outflow pressure transducers (and the servo-nulling pressure transducer) are calibrated to within 0.1– 0.2 cm H2O of each other (see Note 34).

116

Amélie Sabine et al.

5. Switch the 3-way valve so the pressure controller is connected to the cannulating pipettes (see Note 35). 3.2.5  Vessel Micropuncture

To measure pressure on the inflow side of the valve, use a servo-­ nulling micropressure system with a sharp micropipette broken back to a 3 μm tip (see Note 36). The pump, line, and holder should be prefilled with 2 M NaCl and all air bubbles removed. 1. Raise both pressures to 10 cm H2O, remove any axial slack, if present (a taut vessel is easier to micropuncture). 2. Using the servo-nulling pipette to make the micropuncture hole will usually plug up the micropipette, so first use a pilot pipette—an unfilled micropipette made of the same glass with a sharper taper and fine point (unbroken)—to make a small hole in the vessel wall. Puncturing the wall near the inflow cannulation pipette will be easier than in the middle of the vessel because the cannulation pipette will anchor it and help prevent bowing during micropuncture. During this procedure temporary use of a higher magnification objective (e.g., 16×) allows better visualization of the micropuncture hole than a lower magnification objective (e.g., 4 or 6.3×) routinely used during the valve test protocols. 3. Mount the pilot pipette in a holder at a 45° angle to the vessel (see Fig. 12a and Note 36). 4. Make a small indentation in the vessel wall (top surface) by lowering the pipette in the Z-axis. 5. Rapidly thrust the pipette forward for ∼100 μm using the 45° axis control; it should push easily through the wall. 6. Pull back slightly on the control for that axis and also lift the micropipette slightly in the Z-axis so that the tip is centered in the vessel lumen; slide it in and out using the 45° axis control, pushing it a bit further into the lumen each time to widen the pilot hole to ∼5–10 μm. 7. After memorizing the location of the hole (see Note 37), withdraw the pilot pipette completely, raise it out of the bath solution, and replace it with the 2 M NaCl servo-nulling pipette. 8. Lower the servo-nulling pipette next to (but outside of) the vessel, confirm that the tip is clear (see Note 38), turn up the servo null gain, and adjust the offset to read zero (see Note 39). 9. Use the manipulator controls to carefully guide the servo-­ nulling pipette back into the hole made by the pilot pipette: once the tip is centered in the hole, lower it in the Z-axis and then advance it at a 45° angle. As soon as the servo-nulling micropipette enters the lumen the readout on the servo-­ nulling pressure transducer (Psn) should be 10 cm H2O. If it is

Mouse Mesenteric Lymphatic Valve Analysis

117

Fig. 12 (a) Rear view of converted x-axis (moved to 45° angle) of Narishige M102 micromanipulator head (pipette holder is on back side). (b) Bottom view of inflow, outflow and servo-nulling pipettes, showing their positions relative to the cannulated lymphatic vessel

slightly high or low, adjust the balance control of the servo-null system so that the Psn readout is 10 cm H2O (see Note 39). 10. Advance the micropipette into the lumen until the shank of the micropipette wedges into the pilot hole, sealing the hole (see Fig. 12b). 11. Raise the tip in the Z-axis as necessary to keep the tip clear of the endothelium. 12. Adjust the set pressures on the controller for both inflow (Pin) and outflow (Pout) pressures to 0.5 cm H2O; the servo-null reading should drop immediately to 0.5 cm H2O. If the reading is high or low adjust the offset control of the Psn amplifier; then again raise pressure to 10 cm H2O and if the servo-null reading is not exactly 10 cm H2O, adjust the gain slightly. 13. Set the pressure to 0 and then 10 cm H2O, check the values and repeat if necessary until you are confident of the calibration.

118

Amélie Sabine et al.

3.2.6  Back-Leak Test

This test measures the degree of pressure back-leak through a closed valve when outflow pressure is raised [7]. 1. Set both Pin and Pout to 0.5 cm H2O. 2. Using an appropriate computer program to drive the pressure controller, initiate a ramp-wise increase in Pout from 0.5 to 10 cm H2O (rate ∼8 cm H2O/min) while Pin is held at 0.5 cm H2O. A normal valve should close before Pout reaches ∼2 cm H2O; if it does not, then tap gently on the Pout line to encourage closure. 3. When Pout reaches 10 cm H2O, return it to 0.5 cm H2O and repeat the test twice. 4. For analysis, take the average of the three final servo-null pressures recorded when Pout reaches 10 cm H2O. A value of 0.5 cm H2O is used for Pin because it is low enough that a normal valve will easily close, but not so low that the vessel partially collapses and touches the tip of the servo-nulling micropipette. See examples of the behavior of normal, abnormal, and intermediate valves in Fig. 13. For a normal valve the average Psn value when Pout reaches 10 cm H2O should be 0.5 cm H2O (i.e., no back-leak); for a completely incompetent valve the servo-null reading should be ∼5.2 cm H2O (approximately ½ between 0.5 and 10 cm H2O, but will depend both on the position of the valve, i.e., if the valve is exactly midway between the two cannulating pipette tips, and the position of the servo-null pipette relative to the valve, i.e., if the Psn

Fig. 13 Valve back-leak test. Experimental traces showing Psn recordings during back-leak tests on normal (a), intermediate (b), and abnormally leaky valves (c). Pin: inflow pipette pressure; Pout: outflow pipette pressure; Psn; servo-nulling micropipette pressure

Mouse Mesenteric Lymphatic Valve Analysis

119

tip is near the valve or near the inflow pipette). If the valve is leaky, there will be backflow through it and therefore a continuous pressure drop along the length of the vessel toward the inflow pipette. It is important to be sure that no debris has accumulated on the cannulating pipette tips, particularly the inflow pipette tip during this procedure, or the measurements will be invalid. Be sure the servo-nulling micropipette tip does not contact the vessel wall during the Pout ramp (some vessel bowing may occur); doing so will produce an artifactual rise in the Psn reading that can be corrected by adjusting the micropipette tip position. Check the calibration of the Psn pipette by stepping both Pin and Pout from 0.5 to 10 cm H2O and back again (see Note 39 ). If the pipette calibration is off, then correct it by adjusting the offset/gain controls of the Psn amplifier, or change the pipette and repeat the test(s). 3.2.7  Valve Closure Test

This test measures the adverse pressure gradient required to close a valve; the value will vary as a nonlinear function of the passive vessel diameter, which in turn is determined by the baseline pressure in the passive vessel [3]. 1. Set Pin and Pout to 0.1 cm H2O. 2. Note the value of the vessel diameter (see Note 40). 3. Increase Pout ramp-wise to ∼30 cm H2O at a rate of ∼8 cm H2O/ min while watching the valve leaflet positions and/or Psn reading. Psn will rise with Pout, but when the valve snaps shut Psn will fall back rapidly to 0.5 cm H2O if there is no back-leak through the valve. (If there is some back-leak then the Psn reading will be commensurately higher than 0.5 cm H2O.) A normal valve will close at Pout 50 cm H2O, depending on the amount of gain set on the servo-nulling system). If this happens adjust its position slightly with the micromanipulator controls (e.g., raise it vertically) so that it is not contacting the inside of the vessel wall. If the tip is still plugged, withdraw it, turn off the servo-null system gain, switch the stopcock on the pump so that the tip can be flushed clear by application of positive pressure to the syringe filled with 2 M NaCl; this syringe should be a 10 ml glass syringe, allowing easy movement of the barrel so that increased resistance associated with a plugged tip can be felt when applying positive pressure. 39. Operation of the servo-nulling system. We use a model 4A servo-­ nulling micropipette pressure system from Instrumentation for Physiology & Medicine (San Diego, CA); the company is no longer in business but used systems can be occasionally found on ebay. The same system is/was also made by Vista Electronics (Ramona, CA), although current availability is uncertain. An alternative air-filled system is made by WPI Instruments (Sarasota, FL). The general operating principle is to measure the resistance across the tip of a micropipette filled with 2 M NaCl continually using an AC bridge circuit. When the tip is inserted into a pressurized blood or lymphatic vessel containing 0.15 M blood or lymph, the interface between the two solutions moves from the tip into the shank of the micropipette, thereby increasing the electrical resistance across the tip. This is sensed by a headstage/amplifier and appropriate internal circuitry generates a counter pressure that is applied to the rear of the pipette using a hydraulic pump (model V203, Ling Instruments, Hertz, UK) filled with silicone oil (or with air for the WPI system) until the pipette resistance is returned (“servoed”) to its initial set point. A pressure transducer (CyQ) on a side port of the pump measures the counter pressure, Psn, which is equal to the pressure in the lumen of the vessel being micropunctured. The step-by-step operation of the servo-nulling system is described in the manual from the manufacturer; we recommend connecting the pump in such a way as to fill the 2 M NaCl line and holder by suction rather than positive

Mouse Mesenteric Lymphatic Valve Analysis

127

pressure as the latter tends to leave air bubbles. Once the pipette is filled, the basic procedure is to apply a slight negative pressure to the micropipette (using the pump offset control on the servo-null system) to draw the high/low saline interface slightly up into the tip. The balance point (the “null resistance set point”) is then specified by another control on the front panel of the instrument and the system gain control is turned up to a point sufficiently high to counter the highest pressure that will occur inside the vessel but not so high as to induce oscillations. The servo-nulling zero level is typically set with the tip positioned outside the vessel just prior to micropuncture. The Psn transducer should be calibrated before the experiment, matching the Pin and Pout transducers, using the reservoir system. One advantage of using the system in conjunction with a cannulated vessel is that the calibration and zero point can be checked at any time simply by raising Pin and Pout simultaneously by the same amount and verifying that Psn = Pin = Pout. 40. Vessel diameter measurement. A video port on the inverted microscope should be connected to a digital monochrome camera and the camera interfaced to a computer with the appropriate image acquisition hardware and software (see Chap. 16). We use custom LabVIEW (National Instruments, Austin TX) programs for video/data acquisition and diameter tracking. Automated diameter tracking systems are commercially available, e.g., from Danish Myo (Aarhus, Denmark), IonOptix (Westfield, MA). These are accurate for outside edge detection but less so for internal diameter measurement. We use our own method for inner diameter measurement [22]. Continuous diameter measurements are not necessary for the valve tests described here, so an inexpensive alternative is simply to measure the internal vessel diameter with a ruler on the computer screen and calibrate the measurement against a video image of a stage micrometer at the same magnification. 41. Calibration range of Psn for the valve closure test. For the valve closure test, the Psn reading does not need to be accurate; relative changes in Psn are used in that test simply as an indicator of valve closure. If the Psn transducer is calibrated only to 10 cm H2O it is likely to be inaccurate during closure tests of defective valves if higher pressures are reached (>30 cm H2O, outside of the calibration range). The critical value to note at the point of valve closure is the Pout reading. However, a rapid fall in Psn during the test often is additional verification that the valve has closed and, if valve is not oriented exactly right to see both leaflets (which is common), the Psn reading may be the best indication of when the valve closes (see Fig. 14a). Also note that Pout measurements recorded during times of backflow through the valve are not corrected for the pressure drop

128

Amélie Sabine et al.

across the outflow-cannulating pipette (which can be significant) [23]. For this reason the same pairs of cannulating pipettes should be used when comparing different, e.g., normal and abnormal valves.

Acknowledgments This work was supported by the Swiss National Science Foundation (31003A-156266 and CR32I3_166326), MEDIC, the Emma Muschamp Foundation, Fondation Leenaards, the TheraLymph ERA-NET E-Rare Research Program (FNS 31ER30_160674), the Commission for Technology and Innovation, and the Swiss Cancer League (KLS 3406-02-2016) (to T.V.P), Theodor and Gabriela Kummer funds from UNIL-FBM and Société Académique Vaudoise fellowships (to E.B.), Fondation Pierre Mercier pour la Science and Novartis Foundation for medical-biological research (to A.S.), and grants from the National Institutes of Health R01 HL-120867, R01 HL-122608, R01 HL-122578 (to M.J.D.). References 1. Zawieja DC (2010) Contractile physiology of Lymphatics. Lymphat Res Biol 7:87–96 2. Bazigou E, Wilson JT, Moore JE (2014) Primary and secondary lymphatic valve development: molecular, functional and mechanical insights. Microvasc Res 96:38–45. https:// doi.org/10.1016/j.mvr.2014.07.008 3. Davis MJ, Rahbar E, Gashev AA, Zawieja DC, Moore JE (2011) Determinants of valve gating in collecting lymphatic vessels from rat mesentery. Am J Physiol Heart Circ Physiol 301:H48– H60. https://doi.org/10.1152/ ajpheart.00133.2011 4. Schulte-Merker S, Sabine A, Petrova TV (2011) Lymphatic vascular morphogenesis in development, physiology, and disease. J Cell Biol 193:607–618. https://doi.org/10.1083/ jcb.201012094 5. Sabine A, Saygili Demir C, Petrova TV (2016) Endothelial cell responses to biomechanical forces in lymphatic vessels. Antioxid Redox Signal 25:451–465. https://doi. org/10.1089/ars.2016.6685 6. Petrova TV, Karpanen T, Norrmén C, Mellor RH, Tamakoshi T, Finegold DN, Ferrell RE, Kerjaschki D, Mostoslavsky G, Ylä-Herttuala S, Miura N, Alitalo K (2004) Defective valves and abnormal mural cell recruitment underlie lymphatic vascular failure in lymphedema distichia-

sis. Nat Med 10:974–981. https://doi. org/10.1038/nm1094 7. Sabine A, Bovay E, Saygili Demir C, Kimura W, Jaquet M, Agalarov Y, Zangger N, Scallan JP, Graber W, Gulpinar E, Kwak BR, Mäkinen T, Martinez-Corral I, Ortega S, Delorenzi M, Kiefer F, Davis MJ, Djonov V, Miura N, Petrova TV (2015) FOXC2 and fluid shear stress stabilize postnatal lymphatic vasculature. J Clin Invest 125:3861–3877. https://doi. org/10.1172/JCI80454 8. Sabine A, Agalarov Y, Maby-El Hajjami H, Jaquet M, Hägerling R, Pollmann C, Bebber D, Pfenniger A, Miura N, Dormond O, Calmes J-M, Adams RH, Mäkinen T, Kiefer F, Kwak BR, Petrova TV (2012) Mechanotransduction, PROX1, and FOXC2 cooperate to control connexin37 and calcineurin during lymphatic-valve formation. Dev Cell 22:430–445. https://doi. org/10.1016/j.devcel.2011.12.020 9. Tammela T, Saaristo A, Holopainen T, Lyytikkä J, Kotronen A, Pitkonen M, Abo-Ramadan U, Ylä-Herttuala S, Petrova TV, Alitalo K (2007) Therapeutic differentiation and maturation of lymphatic vessels after lymph node dissection and transplantation. Nat Med 13:1458–1466. https://doi.org/10.1038/nm1689 10. Norrmén C, Ivanov KI, Cheng J, Zangger N, Delorenzi M, Jaquet M, Miura N, Puolakkainen

Mouse Mesenteric Lymphatic Valve Analysis P, Horsley V, Hu J, Augustin HG, Ylä-­ Herttuala S, Alitalo K, Petrova TV (2009) FOXC2 controls formation and maturation of lymphatic collecting vessels through cooperation with NFATc1. J Cell Biol 185:439–457. https://doi.org/10.1083/jcb.200901104 11. Bazigou E, Xie S, Chen C, Weston A, Miura N, Sorokin LM, Adams R, Muro AF, Sheppard D, Mäkinen T (2009) Integrin-alpha9 is required for fibronectin matrix assembly during lymphatic valve morphogenesis. Dev Cell 17:175– 186. https://doi.org/10.1016/j.devcel.2009. 06.017 12. Bazigou E, Mäkinen T (2013) Flow control in our vessels: vascular valves make sure there is no way back. Cell Mol Life Sci 70:1055–1066. https://doi.org/10.1007/s00018012-1110-6 13. Geng X, Cha B, Mahamud MR, Srinivasan RS (2017) Intraluminal valves: development, function and disease. Dis Model Mech 10:1273– 1287. https://doi.org/10.1242/dmm. 030825 14. Kazenwadel J, Betterman KL, Chong C-E, Stokes PH, Lee YK, Secker GA, Agalarov Y, Saygili Demir C, Lawrence DM, Sutton DL, Tabruyn SP, Miura N, Salminen M, Petrova TV, Matthews JM, Hahn CN, Scott HS, Harvey NL (2015) GATA2 is required for lymphatic vessel valve development and maintenance. J Clin Invest 125:2979–2994. https:// doi.org/10.1172/JCI78888 15. Sweet DT, Jiménez JM, Chang J, Hess PR, Mericko-Ishizuka P, Fu J, Xia L, Davies PF, Kahn ML (2015) Lymph flow regulates collecting lymphatic vessel maturation in vivo. J Clin Invest 125:2995–3007. https://doi. org/10.1172/JCI79386 16. Schmid-Schonbein GW (1990) Microlymphatics and lymph flow. Physiol Rev 70:987–1028. https://doi.org/10.1152/ physrev.1990.70.4.987

129

17. Zawieja SD, Castorena-Gonzalez JA, Scallan J, Davis MJ (2018) Differences in L-type calcium channel activity partially underlie the regional dichotomy in pumping behavior by murine peripheral and visceral lymphatic vessels. Am J Physiol Heart Circ Physiol 5:e9863. https:// doi.org/10.1152/ajpheart.00499.2017 18. Wiederhielm CA, WOODBURY JW, KIRK S, RUSHMER RF (1964) Pulsatile pressures in the microcirculation of Frog's mesentery. Am J Phys 207:173–176. https://doi. org/10.1152/ajplegacy.1964.207.1.173 19. Intaglietta M, Tompkins WR (1971) Micropressure measurement with 1 micron and smaller cannulae. Microvasc Res 3:211–214 20. Fox JR, Wiederhielm CA (1973) Characteristics of the servo-controlled micropipet pressure system. Microvasc Res 5:324–335 21. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ (1981) Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflugers Arch 391:85–100 22. Davis MJ (2005) An improved, computer-­ based method to automatically track internal and external diameter of isolated microvessels. Microcirculation 12:361–372. https://doi. org/10.1080/10739680590934772 23. Bertram CD, Macaskill C, Davis MJ, Moore JE (2014) Development of a model of a multilymphangion lymphatic vessel incorporating realistic and measured parameter values. Biomech Model Mechanobiol 13:401–416. https://doi. org/10.1007/s10237-013-0505-0 24. Jamalian S, Jafarnejad M, Zawieja SD, Bertram CD, Gashev AA, Zawieja DC, Davis MJ, Moore JE (2017) Demonstration and analysis of the suction effect for pumping lymph from tissue beds at subatmospheric pressure. Sci Rep 7:12080. ­https://doi.org/10.1038/s41598017-11599-x

Chapter 8 Morphological Analysis of Lacteal Structure in the Small Intestine of Adult Mice Sang Heon Suh, Seon Pyo Hong, Intae Park, Joo-Hye Song, and Gou Young Koh Abstract The lacteal is a blunt-ended lymphatic capillary located at the center of a villus in the small intestine that plays multifaceted roles under both physiologic and pathologic conditions. However, studies of its biology are limited by the lack of a feasible method to visualize all the relevant components for its regulation. Here, we describe an efficient whole-mount protocol to visualize the intact structure of lacteals and surrounding cells in villi of the small intestine of adult mouse. Key words Whole-mount, Lacteal, Small intestine, Immunohistochemistry, Confocal imaging

1  Introduction The lacteal is a blunt-ended lymphatic capillary located at the center of a villus, which not only performs the common lymphatic function of interstitial fluid and immune cell drainage but is also specialized in absorbing lipids and lipid-soluble vitamins and drugs [1]. Recently, studies have shown that lacteal not only functions as a mere drainage conduit but also plays significant roles in pathologic conditions such as inflammatory bowel disease [2–5]. With the increasing awareness of the importance of lacteal, many studies presented evidence that the maintenance of lacteal structure and function is dependent on neighboring populations. For instance, the periodic contraction of villus smooth muscle cells squeeze lacteals to facilitate efficient drainage [6], and VEGF-C secreted by neighboring smooth muscle cells maintains the lacteal integrity under steady-state conditions [7]. Indeed, en bloc visualization of lacteal with surrounding environment is critical to understand lacteal biology. Sectional staining images have intrinsic limitations in visualizing longitudinally projecting structures and provide only a few Guillermo Oliver and Mark L. Kahn (eds.), Lymphangiogenesis: Methods and Protocols, Methods in Molecular Biology, vol. 1846, https://doi.org/10.1007/978-1-4939-8712-2_8, © Springer Science+Business Media, LLC, part of Springer Nature 2018

131

132

Sang Heon Suh et al.

details; in terms of lacteal imaging, the tips cannot be definitely recognized. On the other hand, whole-mount technique presents the entire structure, providing mechanistic insight into the lacteal and neighboring tissue components [8]. Here, we present a protocol suitable for imaging lacteal and stromal cells in the intestinal villi, which is modified from a previous protocol to shorten the duration by 4 to 7 days and eliminate the need of a potentially explosive reagent (i.e., picric acid). Using this protocol, we were able to obtain clear images of not only the lacteal but also the various surrounding stromal cellular components in the intestinal villi.

2  Materials 2.1  Tissue Preparation

1. 1× PBS: Store at room temperature (RT). 2. Fixatives: 3.7% (wt/vol) formaldehyde in PBS. To be made fresh shortly before use. 3. 10% (vol/vol) sucrose solution: Dissolve 50 g of sucrose in 400 mL of 1× PBS. When the sucrose is dissolved, add 1× PBS up to 500 mL. Sterilize with a filter and store at 4 °C. 4. 20% (vol/vol) sucrose solution with 10% (vol/vol) glycerol: Dissolve 100 g of sucrose and 50 mL of glycerol in 400 mL of 1× PBS. When the sucrose and glycerol are dissolved, add 1× PBS up to 500 mL. Sterilize with a filter and store at 4 °C. 5. Anesthetic agents: Mixture of Zoletil (20 mg/kg) and xylazine (11 mg/kg). 6. Silicone plates: Mix the curing agent with the silicone vigorously according to the manufacturer’s instructions. Pour the mixture into 10-cm culture dishes. Let them cure in a hood for 3–4 days at RT. Silicone plates can be reused after washing. 7. Equipment: 20-mL syringes fitted with a 26-guage needle, scissors, forceps, artery scissors, 15-cm culture dish, micro forceps, and insect pins.

2.2  Whole-Mount Immunostaining

1. Preserving solution: 0.1% (wt/vol) NaN3 in 1× PBS. Dissolve 1 g of NaN3 in 1 L of 1× PBS. Store at 4 °C. 2. Washing buffer: 0.3% (vol/vol) Triton X-100 in 1× PBS. Add 30 mL of 10% Triton X-100 in 1 L of 1× PBS. Store at 4 °C. 3. Blocking buffer: 5% (vol/vol) goat or donkey serum dissolved in washing buffer. Sterilize with a filter and store at 4 °C. 4. Primary antibodies. 5. Secondary antibodies. 6. DAPI solution. 7. VECTASHIELD Antifade Mounting Medium.

Morphological Analysis of Lacteal Structure in the Small Intestine of Adult Mice

133

8. Silicone plates: Prepare silicone plates in 12-well culture dishes as described in Subheading 2.1. 9. Equipment: Dissection microscope, scissors, microscissors, microforceps, insect pins, 6-cm culture plate, microscope slides, and coverslips.

3  Methods 3.1  Harvest and Preparation of Small Intestine

1. 1× PBS and 3.7% formaldehyde solution for cardiac perfusion should be adapted to RT. All the other reagents should be ice-­ cold at use. 2. Anesthetize the mouse by intraperitoneal injection of anesthetic agent. 3. Fill 20-mL syringes with PBS and 3.7% formaldehyde solution. 4. Open the thoracic cage and expose the heart. Insert the syringe filled with PBS in the left ventricle of the mouse via cardiac apex. Immediately after insertion, make a small cut in the right atrium to allow perfusion outlet. 5. Perfuse 20-mL of PBS. After PBS perfusion, remove the syringe, and insert the fixative-filled syringe. Resume perfusion with fixatives. 6. Cut the pyloric ring at the proximal end and the cecum at the distal end. Pull the intestine out from mesentery with a pair of forceps. To avoid drying of the tissue, keep the small intestine in 15-cm culture plate filled with ice-cold PBS (see Note 1). 7. Cut the small intestine into segments less than 5 cm in length to separate duodenum, jejunum, and ileum (see Note 2). With scissors, cut the intestine segment longitudinally along the mesenteric border to expose the lumen. Flush out the feces and other luminal contents with a 20-mL syringe fitted with a 26-gauge needle filled with ice-cold PBS (see Note 3). 8. Transfer the washed intestine segments to a 6-cm silicone plate filled with ice-cold PBS. Using the insect pins, anchor the intestine segments with the luminal side facing up (see Fig. 1 and Note 4). 9. Discard PBS. For postfixation, fill the silicone plates with ice-­cold fixative. Make sure that the samples are completely submerged. Incubate the samples on an orbital shaker for 2 h at 4 °C. 10. After postfixation, briefly wash the samples with ice-cold PBS 3 times. 11. Discard PBS. Fill the silicone plates with 10% (vol/vol) sucrose solution. Incubate the samples on an orbital shaker for 3 h at 4 °C.

134

Sang Heon Suh et al.

Fig. 1 Procedure details. (a) Intestine segments anchored with insect pins. Note the location of pins. The intestine segments are flattened and positioned so that the luminal side is facing up. (b) Small pieces (0.5–1.0 cm) of intestine anchored in a well of a 12-well silicone plate. (c) Dissection microscopic view demonstrating how to cut a strip of intestinal villi. Note that the width of a strip is even (1 or 2 villi). (d) An example of successfully mounted strips. Note that the direction of villi projection of each strip is uniform. Scale bars, 500 μm

12. Discard 10% (vol/vol) sucrose solution. Fill the silicone plate with 20% (vol/vol) sucrose with 10% (vol/vol) glycerol solution. Incubate the samples on an orbital shaker overnight (O/N) at 4 °C. 3.2  Whole-Mount Immunostaining

1. Briefly wash the samples with ice-cold PBS and replace it with preserving solution (see Note 5). 2. Cut the intestine segment into small pieces (0.5–1.0 cm). Transfer the intestine pieces into the well of a 12-well silicone plate filled with blocking buffer. Pin the four corners of the samples with insect pins. Incubate the samples on an orbital shaker for 1 h at 4 °C (see Note 6). 3. Prepare primary antibody mixture in blocking buffer. Fully submerge the samples to avoid drying during the prolonged incubation.

Morphological Analysis of Lacteal Structure in the Small Intestine of Adult Mice

135

4. After incubation with blocking buffer, replace it with appropriate primary antibody mixture. Incubate the samples on an orbital shaker O/N at 4 °C. 5. Discard the primary antibody mixture. 6. Replace with 1–2 mL of ice-cold washing buffer. Using a P1000 micropipette with volume set to 600 μL, wash the samples by pipetting the buffer intensely more than 10 times. Change the pipette tip with a new one to wash the next sample. Repeat this step at least 5 times. Be careful not to touch the intestine samples directly with pipette tip. 7. After washing, replace the washing buffer with ice-cold 1× PBS. Wash the samples by pipetting PBS intensely. Discard PBS, and keep the samples in fresh, ice-cold PBS for a while. 8. Prepare secondary antibody mixture in blocking buffer. 9. Discard PBS, and add secondary antibody mixture to the well. Incubate the samples on an orbital shaker O/N at 4 °C. Incubation time for secondary antibodies should not exceed 16 h (see Note 7). 10. Discard the secondary antibody mixture. 11. Replace with 1–2 mL of ice-cold washing buffer. Using a P1000 micropipette with volume set to 600 μL, wash the samples by pipetting the buffer intensely more than 10 times. Change the pipette tip with a new one to wash the next sample. Repeat this step at least 5 times. Be careful not to touch the intestine samples directly with pipette tip. 12. At the end of washing, replace the washing buffer with ice-­cold 1× PBS. Wash the samples by pipetting the buffer intensely. 13. Discard PBS, and incubate the samples with DAPI solution on an orbital shaker for 3–4 h at 4 °C (see Note 8). 14. Discard DAPI solution and replace with ice-cold PBS. Keep the plate on ice and protected from lights. 15. Pin off the samples from the silicone plate, and transfer it to a 6-cm culture plate on a dissection microscope. To avoid drying of the sample, add a small amount of PBS (less than 100 μL) (see Note 9). 16. Look at the sample at the highest magnification under retroillumination (see Note 10). Using microforceps and microscissors cut a strip of intestine with a width of one or two villi. A strip of even width should be obtained to enable flat lateral mounting. Obtain a sufficient number of strips (usually 6–8 stripes/sample) (see Note 11). 17. Transfer the strips on a microscope glass. Using microforceps and dissection microscope, place the strips on the microscope glass so that the lateral side of villi is well-exposed (see Note 12).

136

Sang Heon Suh et al.

18. Add 2–3 drops of VECTASHIELD Antifade Mounting Medium to the strips. After spreading the mounding medium evenly, place a cover slip on top of the samples (see Fig. 2 and Note 13).

4  Notes 1. Try to handle the intestine gently. Physical forces applied with forceps will result in traces in the villi, which hinders observation under a confocal microscope. Holding the border of intestinal tube, rather than the center, would help avoid undesirable artifact. 2. To compare the villus structure between individual mice, the same location of the intestine should be examined, since the villus structure differs significantly depending on the region of small intestine. Therefore, the same, corresponding segment should be harvested and compared. The operator should also keep in mind which side is the proximal and distal end of intestine segments throughout the entire processes of tissue harvest and immunostaining. 3. Thorough washing of luminal contents is a critical step for successful immunostaining. Remnant mucus or other luminal contents could be a cause of inadequate postfixation and nonspecific antibody staining. They can also accelerate the degradation of stained sample. Flushing with PBS would not damage adequately perfusion-fixed villi, unless the needle touches the samples directly. 4. When inserting the pins, make sure that the intestine is completely flat. If the tissue is not pulled tensely enough due to inappropriate anchoring, intestine segments will be wrinkled after postfixation. 5. Samples kept in a preserving solution can be stored at 4 °C for several days unless the solution is grossly contaminated. To minimize the contamination of samples, use autoclaved instruments (i.e., scissors and microforceps) when cutting the samples into small pieces for immunostaining.

Fig. 2 (continued) duodenum, jejunum, and ileum. Scale bars, 100 μm. (c) Representative images of VE-cadherin junctions in lacteal. The area marked by a dotted box is further magnified in the right panel for a detailed view of junctional pattern in a lacteal. (d) Filopodia at lacteal tip. The area marked by a dotted box is further magnified in the right panel for a detailed view of lacteal filopodia (arrowheads). Scale bars, 20 μm. (e) Representative images of LYVE-1+ lacteal and surrounding αSMA+ smooth muscle fibers. The area marked by a dotted box is further magnified in the right panel for a detailed view of the close proximity of smooth muscle fibers to the lacteal. Scale bars, 50 μm

Morphological Analysis of Lacteal Structure in the Small Intestine of Adult Mice

137

Fig. 2 Visualization of lacteal structure. (a) Measurement of lengths of CD31+/LYVE-1+ lacteal, CD31+/LYVE-1− capillary plexus, and DAPI+ villus. (b) Representative images of lacteals and blood capillary plexi in the villi of

138

Sang Heon Suh et al.

6. A maximum of 4–6 pieces (0.5–1.0 cm sized) can be incubated simultaneously in a well of a 12-well plate. Make sure that the tissue is completely flat. Wrinkling of samples during immunostaining prevents optimal results during mounting and confocal imaging. 7. Prolonged incubation of samples with secondary antibodies results in nonspecific binding of antibodies. Keep to the recommended incubation time. In case of primary antibodies, it is acceptable to incubate for several days unless the samples have dried. 8. Short incubation time may cause partial staining only in the tip portion of villi. Spend enough time to stain the entire villi including the crypt. This is also true for primary and secondary antibody incubation. 9. It is important to protect the sample from drying during the mounting process. Add ice-cold PBS repeatedly if strip dissection procedure is prolonged. At the same time, be careful not to add too much PBS as it will cause samples to float and make handling of samples difficult in the next step. 10. Use dim lights to minimize bleaching of stained samples. 11. Even width of strips is critical for imaging quality. Uneven width of strips leads to variations in distances from light source to samples, which could be an obstacle for accurate measurement of signal intensities. 12. The villi should be completely straight and projected unidirectionally in a strip. If some villi are folded or the direction of villi projection is inverted, gently ‘brush’ them to align correctly using microforceps. 13. Use an adequate volume of mounting medium. Insufficient mounting medium might result in air bubbles under coverslips. Conversely, excess mounting media may cause strips to float in the medium and distract the alignment of the samples. References 1. Petrova TV, Koh GY (2018) Organ-specific lymphatic vasculature: From development to pathophysiology. J Exp Med 215(1):35–49. https://doi.org/10.1084/jem.20171868 2. Jang JY, Koh YJ, Lee SH, Lee J, Kim KH, Kim D, Koh GY, Yoo OJ (2013) Conditional ablation of LYVE-1+ cells unveils defensive roles of lymphatic vessels in intestine and lymph nodes. Blood 122(13):2151–2161. https://doi. org/10.1182/blood-2013-01-478941 3. Vetrano S, Borroni EM, Sarukhan A, Savino B, Bonecchi R, Correale C, Arena V, Fantini M, Roncalli M, Malesci A, Mantovani A, Locati M, Danese S (2010) The lymphatic system

controls intestinal inflammation and inflammation-­associated colon cancer through the chemokine decoy receptor D6. Gut 59(2):197–206. https://doi.org/10.1136/ gut.2009.183772 4. Van Kruiningen HJ, Colombel JF (2008) The forgotten role of lymphangitis in Crohn’s disease. Gut 57(1):1–4. https://doi. org/10.1136/gut.2007.123166 5. von der Weid PY, Rehal S, Ferraz JG (2011) Role of the lymphatic system in the pathogenesis of Crohn’s disease. Curr Opin Gastroenterol 27(4):335–341. https://doi.org/10.1097/ MOG.0b013e3283476e8f

Morphological Analysis of Lacteal Structure in the Small Intestine of Adult Mice 6. Choe K, Jang JY, Park I, Kim Y, Ahn S, Park DY, Hong YK, Alitalo K, Koh GY, Kim P (2015) Intravital imaging of intestinal lacteals unveils lipid drainage through contractility. J Clin Invest 125(11):4042–4052. https:// doi.org/10.1172/JCI76509 7. Nurmi H, Saharinen P, Zarkada G, Zheng W, Robciuc MR, Alitalo K (2015) VEGF-C is required

139

for intestinal lymphatic vessel maintenance and lipid absorption. EMBO Mol Med 7(11): 1418–1425. https://doi.org/10.15252/emmm. 201505731 8. Bernier-Latmani J, Petrova TV (2016) High-­ resolution 3D analysis of mouse small-intestinal stroma. Nat Protoc 11(9):1617–1629. https:// doi.org/10.1038/nprot.2016.092

Chapter 9 Morphological and Functional Analysis of CNS-Associated Lymphatics Jasmin Herz, Antoine Louveau, Sandro Da Mesquita, and Jonathan Kipnis Abstract The study of meningeal lymphatic vessels of the central nervous system (CNS) has recently gathered momentum, with several papers dissecting their role in draining solutes from cerebrospinal fluid and brain (Louveau et al., Nature 523(7560):337–341, 2015; Antila et al., J Exp Med 214(12):3645– 3667, 2017; Aspelund et al., J Exp Med 212(7):991–999, 2015). Methodological capabilities, however, have been limited to few laboratories due to difficulties reproducibly visualizing these rare cell subsets in the meninges. To explore meningeal lymphatics fundamental role during homeostasis and how they may contribute to human pathology, the field has begun to require purification and characterization of lymphatic endothelial cells. Here, modern cell biological techniques involving a combination of histological, flow-­cytometric, and functional drainage assays are applied to brain and spinal cord meninges and detailed stepwise procedures used for successful in vivo and ex vivo characterization of meningeal lymphatic vessels. Key words Brain meninges, Spinal cord meninges, Neuroimmunology, CSF drainage

1  Introduction The meningeal microenvironment is composed of many cell types, including innate and adaptive immune cells. Lymphatic vessels within the meninges have been shown to support the drainage of macromolecules and solutes [1–3]. Studying their functions that may govern inflammatory processes and immune surveillance of the meninges and CNS poses a valuable contribution to our understanding of neurological disease, and even resistance to therapies of the CNS [4]. Recent studies from our laboratory indicated that the meningeal environment is changed following performance of cognitive tasks [5]. T cells are recruited to the meningeal spaces, supporting neuronal function through production of IL-4 and regulating meningeal myeloid cell phenotypes and subsequent brain-derived neurotrophic factor expression. Thus, investigating Guillermo Oliver and Mark L. Kahn (eds.), Lymphangiogenesis: Methods and Protocols, Methods in Molecular Biology, vol. 1846, https://doi.org/10.1007/978-1-4939-8712-2_9, © Springer Science+Business Media, LLC, part of Springer Nature 2018

141

142

Jasmin Herz et al.

the migration in and out of the meninges of different cell types through lymphatic vessels is critical for further understanding the crosstalk between immune system and nervous system. The use of single cell analysis such as flow cytometry or RNA-seq allows phenotypic and even functional characterization of the lymphatic cells, aiming at deciphering their role in guiding immune sentinels. Such fundamental features as morphology, surface phenotype, and function are unreported in neurological diseases. Here we provide simple methods to extract and analyze lymphatic endothelial cells from the meninges to effectively identify their contribution to physiological processes in health and disease. These techniques could also be used to address this and many other outstanding questions pertaining human meningeal lymphatics biology.

2  Materials 2.1  Animal Preparation

1. Mice: C57BL/6 (Jackson Laboratories, stock#000664) and Prox1creERT2xROSA(tdTomato) (Jackson Laboratories, stock# 022075, 007909). 2. Surgical instruments: blunt scissors and forceps. 3. 10 or 30 ml syringes with 25-G needle. 4. Phosphate-buffered saline (PBS), supplemented with 5 U/ml heparin. 5. Paraformaldehyde fixative (PFA): Dissolve 40 g of paraformaldehyde powder in 900 ml ultrapure water in a fume hood. Heat the solution to 60 °C with constant stirring. Add 100  ml 10× PBS to make 1 L. Store at 4 °C for up to 1 month or freeze at −20 °C for long-term storage. 6. Anesthesia: Lethal dose of pentobarbital or other approved anesthetic agent.

2.2  Meninges Dissection

1. Surgical instruments: Spring scissors, microscissors, fine forceps such as Dumont#5 or 7-Inox-H forceps. 2. Dissecting microscope. 3. 10-cm2 petri dish. 4. Phosphate-buffered saline (PBS), pH 7.4.

2.3  Immunohisto-­ chemistry Staining

1. 24-well plate. 2. Orbital shaker. 3. Permeabilizing and wash buffer (PBST): Add 1 ml of Triton X-100 to 1 L of PBS. If sterile filtered or supplemented with 0.01% sodium azide, the solution can be stored up to 1 year at room temperature.

Morphological and Functional Analysis of CNS-Associated Lymphatics

143

4. Blocking solution: Add 1 g bovine serum albumin to 100 ml of PBST. Sterile filter and store at 4 °C. The day of use, prepare 10% chicken, goat, or donkey whole IgG in blocking solution (see Note 1). 5. Primary and secondary immune-staining antibodies. 6. 4,6-Diamidino-2-phenylindole (DAPI) stock solution: Dilute 1 mg in 1 ml of ultrapure water. Store at 4 °C. 7. Mounting medium. 8. Glass slides and coverslips. 2.4  Single Cell Suspension from Dural Meninges

1. Meninges: Animal should be perfused using PBS with 5 U/ml heparin and meninges dissected as described below (see Subheading 3.1.1). 2. Phosphate-buffered saline (PBS), pH 7.4. 3. Cell culture medium: RPMI-1640. 4. 2 ml Eppendorf tubes. 5. Enzymatic cocktail: For 1 ml enzymatic cocktail prepare the following solution fresh at the day of use: Add 20 μl of 140 U/ ml Collagenase 8 (in PBS) and 20 μl of 500 U/ml DNase I (in PBS) to 0.96 ml of RPMI-1640. Keep on ice for no longer than 1 h before adding it to the tissue. 6. 37 °C incubator. 7. Rotating mixer or orbital shaker. 8. 50 ml centrifuge tubes. 9. 70 μm cell strainer. 10. Reagents and equipment for counting cells (see Note 2). 11. Sorting buffer: Add 2 ml of a 0.5 M EDTA solution to 1 L of PBS. Dissolve 10 g bovine serum albumin, filter-sterilize and store at 4 °C.

2.5  Immunostaining of Lymphatic Endothelial Cells for Cytometry

1. Antibodies: Alexa Fluor 647 conjugated anti-CD31, PE conjugated podoplanin, FITC anti-CD45 purchased from BD Biosciences or eBiosciences. 2. Flow cytometry microcentrifuge tubes. 3. Hoechst 33342 or DAPI solution (see item 6 in Subheading 2.3).

2.6  Cisterna Magna Injection

1. Mice: C57Bl/6, Jackson Laboratories, stock# 000664. 2. Anesthesia: Ketamine and Xylazine or other institutionally approved drug. 3. Surgical instruments: blunt scissors, forceps, and retractors. 4. Stereotaxic instrument.

144

Jasmin Herz et al.

5. Hamilton syringe: 600 or 700 series with 33-G needle. 6. Artificial cerebrospinal fluid (aCSF) (see Note 3). 7. Fluorescent tracer: Dilute 10 μl of a 1 mg/ml fluorescent ovalbumin stock (e.g., Alexa Fluor 647 conjugated ovalbumin reconstituted in aCSF) in 10 μl aCSF (see Note 4). 8. 5-0 sutures. 2.7  Lymph Node Preparation and Staining

1. Paraformaldehyde fixative (see item 5 in Subheading 2.1). 2. Sucrose solution: Stir 30 g of sucrose in 50 ml ultrapure water until it dissolved. Add 10 ml of 10× PBS. Adjust total volume with ultrapure water to make 100 ml of a 30% (w/v) solution. If sterile filtered, sucrose can be stored at 4 °C for several months. 3. Gelatin-coated glass slides: Heat 500 ml of ultrapure water at 60 °C. Dissolve 1.5 g gelatin type A (220 or 275 Bloom) on a magnetic stirrer. Stir in 0.25 g chromium potassium sulfate while maintaining 60 °C. Dip clean glass slides in 40–50 °C gelatin solution. Air-dry or in 37 °C incubator overnight. 4. Hydrophobic pen. 5. Optimal Cutting Temperature medium (OCT). 6. Freezing molds. 7. Blocking solution: Add 10 μl anti-CD16/32 antibody and 50 μl of 20% BSA in 0.94 ml PBS. 8. Primary and secondary antibodies. 9. Mounting medium.

3  Methods 3.1  Immunohisto-­ chemical Detection of CNS Associated Lymphatics 3.1.1  Meninges Dissection

1. Anesthetize mouse by injection of a suitable dosage of anesthesia [6] in accordance with the regulations of your institution’s animal care and use committee. 2. Make an incision through the abdominal wall of the chest, cut both sides of the rib cage to expose the heart and clip the right auricle with a small scissor. 3. Perfuse the mouse through the left ventricle with ~15 ml of ice-­cold PBS with 5 U/ml heparin at a rate of ~10 ml per minute until the liver is clear of blood. 4. Replace the PBS-containing syringe with the PFA-containing syringe and perfuse manually with ~15 ml of 4% PFA at a rate of ~10 ml/min (see Note 5). 5. Remove the skin from the back and head of the dorsal site of the mouse with a pair of blunt scissors.

Morphological and Functional Analysis of CNS-Associated Lymphatics

145

6. Decapitate mouse close to the skull, approximately at C1 level. Make a transverse cut through the L5 vertebrae, just above the hips. 7. Continue with a longitudinal incision from the anterior end, down the right side of the spinal cord until reaching the cervical region. Repeat longitudinal cut along the left side of the spinal cord. Lift the column and separate it from muscles, organs and connective tissue underneath. 8. Keep the skull and spinal cord in 4% PFA at 4 °C for 24 h (see Note 6). 3.1.2  Spinal Cord Meninges Whole-Mount Preparation

1. Fill a petri dish with PBS and place the spinal cord with the ventral site up. Take angled spring scissors and clip the neural arch bilaterally one to two segments at a time. Then lift the end of the lose bone and cut across the neural spine to remove the bone and expose the ventral surface of the spinal cord. Continue this process with the next vertebrae until the whole ventral surface of the spinal cord is exposed. 2. To widen the spinal column for dissection, hold the edges of the spinal column on either side with two straight forceps and crack open carefully. 3. Remove spinal cord and adherent meninges in one piece and place in petri dish with fresh PBS (see Note 7). 4. Use spring microscissors to make a midline incision of the meninges along the ventral surface of the spinal cord. 5. Remove meninges and place in 24-well plate with 1 ml PBS.

3.1.3  Dural Brain Meninges Whole Mount Preparation

1. Remove muscles caudal of the skull and use fine angled scissors to make two incisions: (1) cut the skull starting at the brainstem and go counterclockwise toward the olfactory bulb. (2) Make a second incision in clockwise direction on the left side of the head until the rostral part is reached on both sides. Stay above the mandible muscles. 2. Use forceps to pick up the bone at the inferior end and bend it over, taking care not to damage the attached dura mater. 3. Place the skullcap in 24-well plate with 1 ml of 4% PFA and incubate at least 2 h at room temperature or overnight at 4 °C. 4. Wash once with 1 ml PBS. Carefully remove meninges (dura and some arachnoid) from interior part of the skull cap with fine forceps under a dissecting scope. It is easiest to start at the top of the skull and pull the meninges down in the middle part, and then remove the sides. Move along the edge and peal the meninges from each site and collect it by the pineal gland. If done correctly, the whole meninges can be scoped up in on piece (see Note 8). 5. Place dissected meninges in 1 ml of PBS in a 24-well plate.

146

Jasmin Herz et al.

Fig. 1 Immunohistochemical detection of CNS-associated lymphatics. (a) Meningeal whole mount from brain and spinal cord of a 10-week-old Prox1creERT2xROSA (tdTomato) mouse, 3 weeks after tamoxifen chow. (b) Immunostaining of a Prox-1 tdTomato reporter animal with 0.5 μg anti-Lyve1 eFluor660 3.1.4  Immunohisto-­ chemistry and Imaging Meninges

1. Wash tissue once in 0.5 ml PBS and incubate in 0.3 ml blocking solution for 1 h at room temperature (see Note 9). 2. Immunostain the meninges with 0.3 ml primary antibody, such as VEGFR-3, Lyve-1 and podoplanin (at a concentration of 1:100, 1:400, and 1:100) in PBST overnight at 4 °C (see Note 10). 3. Decant the solution and wash with 0.5 ml PBS for 5 min at room temperature. Repeat two times with fresh PBS. 4. Incubate the tissue with 0.3 ml secondary antibody conjugated with fluorescent protein (at a concentration of 1–2 μg/ml) in PBST with 1% BSA for 1 h at room temperature in the dark. 5. Decant the secondary antibody solution and wash once with PBS and incubate in 0.3 ml Hoechst or DAPI (at a concentration of 0.1–1 μg/ml in PBS) for 3 min at room temperature in the dark. 6. Rinse twice with 0.5 ml PBS. 7. Transfer meninges with paintbrushes to a glass slide with a few drops of PBS. Unfold them carefully and aspirate PBS to flatten meninges and let them dry. Cover with mounting medium to preserve fluorescence and seal with a nail polish to prevent movement (see Note 11). 8. The tissue is now ready for imaging by using a conventional widefield microscope, confocal microscope, or an ultramicroscope (see Note 12 and Fig. 1).

3.2  Sorting of Meningeal Lymphatic Endothelial Cells

1. Transfer meninges from brain (or spinal cord) into 2 ml tube with 0.5 ml RPMI (see Note 13).

3.2.1  Meningeal Single Cell Suspension

3. Filter the resulting single-cell suspension through a 70 μm cell strainer into a 50 ml tube. Add 20 ml of ice-cold PBS to wash the strainer.

2. Add 0.5 ml 2× enzymatic cocktail and place tube in 37 °C incubator on a rotating mixer for 15 min.

4. Centrifuge cells for 10 min at 320 × g at 4 °C and discard the supernatant

Morphological and Functional Analysis of CNS-Associated Lymphatics

147

5. Resuspend each pellet in 200 μl ice-cold sorting buffer and count total number using an automated cell counter or hemacytometer (see Note 2). 3.2.2  Staining of Meningeal Lymphatic Endothelial Cells

1. Place cells in 96-V well plate and centrifuge for 5 min at 320 × g and 4 °C (see Note 14). 2. Add 100  μl primary antibody cocktail and resuspend with pipette. Incubate for 30 min on ice. 3. Add 10 μl of DAPI (10 μg/ml) and incubate for 5 min on ice. 4. Centrifuge cells for 5 min at 4 °C, discard supernatant and wash with 100 μl sorting buffer. 5. Repeat step 4 and keep cell samples on ice until they are ready for sorting (or analyzing) on a flow cytometer (see Note 15 and Fig. 2).

3.3  Quantification of CSF Drainage to Cervical Lymph Nodes 3.3.1  Cisterna Magna Injection of Tracers

1. Mice are anesthetized by intraperitoneal injection of 100 mg/ kg Ketamine and 10 mg/kg Xylazine (see Note 16). 2. The neck is shaved, the mouse placed on a stereotaxic frame and the head secured with tooth and ear bars. 3. The surgical site is swabbed with 10% iodine, followed by 70% ethanol, and a 1 cm sagittal incision of the skin over the occipital bone and cervical spinal cord is made. 4. Under the dissection microscope, the covering muscle layers are separated by blunt dissection with forceps and a pair of retractors used to hold the muscles apart.

gated on live singlets

gated on CD45104

60 K

LEC

103

0.32

40 K 102

101

Podoplanin

SSC

20 K

100

0 0

10

CD45

101

102

103

104

100

101

102

103

104

CD31

Fig. 2 Flow cytometry analysis of lymphatic endothelial cells (LEC) within the brain meninges. Cells obtained were immunostained with antibodies against CD31, podoplanin (for endothelial cells), CD45 (hematopoietic), and DAPI (DNA stain). Typical flow-cytometric profile of LEC (CD45− podoplanin+ CD31+) freshly isolated from the meninges of an individual mouse

148

Jasmin Herz et al.

5. The mouse’s body is lowered by ~1 cm so that the head forms a 140° angle (see Note 17). 6. The posterior atlantooccipital membrane is exposed and penetrated with a Hamilton syringe equipped with a bevel ended 33-G needle (see Note 18). 7. A volume of 2 μl of tracer, such as fluorescent ovalbumin, is diluted in artificial cerebrospinal fluid (see Note 4) and slowly infused intra cisternally. The needle is held in place for ~2 min to prevent leakage of the tracer and slowly removed while observing if any backflow occurs. 8. After injection, the retractors are removed and the skin sutured. The mouse is monitored until its recovery. 3.3.2  Preparation and Staining of Cervical Lymph Nodes

1. Two hours after tracer injection, mice are perfused with cold saline, followed by 4% PFA. CSF-draining lymph nodes (deep and superficial cervical) present in both sides of the neck and nondraining lymph nodes (peritoneal or inguinal) are harvested into 4% PFA (see Note 19). 2. Tissue is washed with PBS and lymph nodes placed in 30% sucrose overnight at 4 °C or until sunk to the bottom of the tube. 3. Lymph nodes are placed in a mold with OCT and frozen on dry ice. 4. 30 μm sections are cut on a cryostat and collected onto gelatin-­ coated glass slides. In order to get a representation of the entire lymph node at least ten sections of each lymph node should be collected (see Note 20). 5. Draw a border around the sectioned tissue with a hydrophobic pen and let air-dry for a few minutes at room temperature. 6. Rinse in PBS and then wash with PBST for 3 min (in a glass staining jar) at room temperature to remove residues of OCT. 7. Incubate with 200–250 μl (depending on the slide area containing tissue sections) of blocking solution in PBST for 20 min at room temperature in a humidifying chamber, then wash once with PBST for 3 min at room temperature. 8. Add 200–250 μl antibody cocktail in PBST with 0.5% BSA and incubate for 1 h at room temperature or overnight at 4 °C in a humidifying chamber. 9. Wash with PBS for 5 min at room temperature. Repeat wash step twice. 10. Add 200–250 μl of secondary antibody mix in PBST with 1% BSA for 1 h at room temperature in a humidifying chamber. 11. Wash once with PBS for 5 min at room temperature.

Morphological and Functional Analysis of CNS-Associated Lymphatics

149

Fig. 3 Quantification of drainage from the CSF to cervical lymph nodes. (a) Shown is a confocal image of a 30 μm section of a deep cervical lymph node. Two hours post injection of fluorescently labeled ovalbumin into the cisterna magna, all nodes were collected, sliced and immunostained for imaging. (b) Quantifications of five individual animals are depicted. Each dot represents the area of all lymph node (outlined by DAPI) covered with fluorescent protein

12. Add 200–250  μl DAPI or Hoechst solution and incubate for 5 min at room temperature in a humidifying chamber. 13. Wash twice with PBS for 5 min at room temperature. 14. Let slides air-dry, then add mounting medium plus glass coverslip and seal with nail polish. 15. Tissue is now ready to be imaged with a microscope. Tracer are being quantified by total counts (micrometer-sized FluoSpheres) and/or area covered (by peptides or other smaller tracers) in draining versus nondraining lymph nodes (Fig. 3).

4  Notes 1. The choice of whole animal serum depends on the secondary antibody used. For example, if the secondary reagent is an antimouse that was made in chicken, 10% chicken serum is used for the first blocking step. 2. We mix 10 μl of the cell suspension with 10 μl of Acridine orange/Propidium iodine solution and count with an automated Cellometer Auto 2000 cell counter. Obtaining cell viability and total counts can also be quickly performed with a hemocytometer and 0.4% Trypan Blue Solution. 3. Artificial cerebrospinal fluid (aCSF) can be made but is usually stable for 3–4 weeks and precipitation apparent. We recommend

150

Jasmin Herz et al.

obtaining aCSF fluid from Harvard Apparatus to make stock solutions of tracers that are injected in vivo. They are micro filtered, match the electrolyte concentration of CSF and stable for months. 4. Other proteins or particulate tracers can be used to test lymphatic function in vivo. To study differences in size-dependent drainage for example, we have used fluorescent microspheres such as FluoSpheres beads from Thermo Fisher (Carboxylate-­ modified in yellow-green or red) in different sizes such as 0.5, 1, or 2 μm. 5. Twitching and stiffing of the muscles indicate successful perfusion of the fixative. 6. Over fixation for 2–3 days may cause some antigens to be masked and not be recognized by specific antibodies. The tissue should be stored in PBS supplemented with 0.01% sodium azide or 0.02% thimerosal at 4 °C for up to a year. If fluorescent proteins are used, tissue should be protected from light to avoid photobleaching. 7. Optional: dorsal root ganglions can be carefully pulled out of bone segments before the spinal cord is resected. 8. Be careful to not tear the meninges and only touch the edges to collect the meninges in the center of the skullcap. 9. The volume used should be just enough to cover the entire tissue. In general, we recommend all steps to be performed on an orbital shaker (with slow agitation). The serum used for blocking depends on the species the secondary antibody was raised in. If the target antigen is intracellular, it is important to permeabilize with Triton X-100 to improve the penetration of antibodies. 10. Longer incubation periods do not enhance the staining signal significantly. We routinely use directly conjugated antibodies for multicolor staining if clones were generated in the same animal species and have corresponding secondary antibodies. 11. Avoid any bubbles to not affect the preservation of the tissue and quality of the image. 12. Depending on the mounting medium, slides can be stored at 4 °C or −20 °C in the dark for weeks or month. 13. In the case that the extracted cells are to be cultured, all procedures should be performed in a laminar flow hood and proper aseptic techniques applied. 14. The total number of meningeal cells per animal will be  0.95. If multiple injection sites are used in one mouse (e.g., both ears, or both paws), then the K rate and half-life values can be averaged together to yield one value per mouse. In some regions of the body (e.g., back skin) or under specific conditions where lymphatic clearance may be slower than expected, there may be a diffusion of the tracer from the injection site through the interstitial tissue that complicates the analysis. In these cases, it may be more appropriate to calculate via an area under the curve approach, as the data will not closely fit an exponential decay model. Alternatively, one could calculate clearance rates starting at a later time point after which it is assumed that diffusion does not occur to a significant degree [12, 19]. 8. For noninvasive contractility assessments, we use the combination of ketamine 80 mg/kg and medetomidine 0.2 mg/ kg. Mice (20 g) are usually asleep for 60–90 min with this

In Vivo Imaging of Lymphatic Function

209

dosage. For invasive contractility assessments, we recommend a mixture of xylazine 20 mg/kg, ketamine 100 mg/kg, and acepromazine 3 mg/kg. However, be careful during induction as the latter anesthesia mixture may overdose the mouse if the full dosage is applied at once. Therefore, we usually inject 3/4 of the dose, wait for 15 min and inject the remaining 1/4. The mouse should be ready after about 20 min of the initial injection. As an alternative to injection anesthesia, isoflurane could be used. However, be aware that in a systematic study of the effects of different anesthesia regimens on lymphatic vessel contractility, we found that isoflurane inhibits the pumping activity of lymphatic collectors (Bachmann et al., in preparation). For euthanasia we inject mice i.p. with ketamine (at least 160 mg/kg) and medetomidine (at least 0.4 mg/kg). 9. A well-performed intradermal injection in normal mice should result in an immediate perfusion of the downstream collecting vessels and popliteal lymph node. Be careful not to overload the vessels as this may interfere with the contractility [20]. In mice with severe lymphatic dysfunction, you may observe retrograde flow to the dermal lymphatic capillaries and/or rerouting of lymphatic flow to alternate lymph nodes [2, 21, 22]. If no collecting lymphatic vessels are visualized or the perfusion is very weak, the injection was likely performed into the subcutaneous tissue. In this case, gentle massage of the injection site may improve perfusion (see Note 10). 10. For example, repeated depression of the injection site (1 per second for 10 s) with a cotton swab (Q-tip) will increase the fluid load on the collecting vessels, leading to a temporary increase in contractility. Pressure cuffs around the limb could also be employed to test the lymphatic pumping or perfusion responses to increases in downstream pressure [23]. 11. Use containers that can be thrown away afterward; the silicone mixture is very sticky. You need a 10 cm petri dish and a silicone elastomer base and curing agent (e.g., from Sylgard). Add ten parts of the silicone base into the mixing container (e.g., PET bottle) and add one part of the curing agent (e.g., using a pipetboy with one-way pipette). Stir the mixture with the pipette. Now the mix is ready to be poured into the petri dish. There will be many small air bubbles; just let the dish dry over 2 days, during which time the air bubbles will mostly disappear. Now your dish is ready to be used. Side note: cover your working area with thick paper, since silicone spills will stay on your bench forever. 12. When working with a silicone-filled petri dish, keep in mind that it takes quite some time to be heated up and keep the mouse warm. Best is to place the dish on the heating pad for

210

Samia B. Bachmann et al.

30–40  min and then do all the preparation and dissection work on the silicone dish still placed on the heating pad. Also set the heating pad 1–2  °C higher than usual (39 °C), so that the silicone dish has a temperature of about 37 °C. 13. The inguinal lymph node is located as depicted in Fig.  3. Three blood vessels (indicated with asterisk in the figure) merge on top of the lymph node and help finding the node. When placing the catheter, take care not to puncture any of them in order to avoid bleeding.

Acknowledgments We thank Prof. Jean-Christophe Leroux, Dr. Paola Luciani, Dr. Diana Andina, and Dr. Anna Polomska for the development and continued provision of pegylated tracers. We also would like to acknowledge Dr. Sinem Karaman and Chloé Chong for assistance during the development of these assays and Dr. Felix Scholkmann for development of the Matlab algorithm to quantify lymphatic contractility. References 1. Proulx ST, Detmar M (2013) Molecular mechanisms and imaging of lymphatic metastasis. Exp Cell Res. https://doi.org/10.1016/j. yexcr.2013.03.009 2. Proulx ST, Luciani P, Christiansen A, Karaman S, Blum KS, Rinderknecht M, Leroux JC, Detmar M (2013) Use of a PEG-conjugated bright near-infrared dye for functional imaging of rerouting of tumor lymphatic drainage after sentinel lymph node metastasis. Biomaterials 34(21):5128–5137. https://doi.org/10.1016/j.biomaterials. 2013.03.034 3. Karaman S, Buschle D, Luciani P, Leroux JC, Detmar M, Proulx ST (2015) Decline of lymphatic vessel density and function in murine skin during aging. Angiogenesis 18(4):489–498. https://doi.org/10.1007/ s10456-015-9479-0 4. Chong C, Scholkmann F, Bachmann SB, Luciani P, Leroux JC, Detmar M, Proulx ST (2016) In vivo visualization and quantification of collecting lymphatic vessel contractility using near-infrared imaging. Sci Rep 6:22930. https://doi.org/10.1038/srep22930 5. Choi I, Chung HK, Ramu S, Lee HN, Kim KE, Lee S, Yoo J, Choi D, Lee YS, Aguilar B, Hong YK (2011) Visualization of lymphatic vessels by

Prox1-promoter directed GFP reporter in a bacterial artificial chromosome-based transgenic mouse. Blood 117(1):362–365. https://doi. org/10.1182/blood-2010-07-298562 6. Proulx ST, Luciani P, Alitalo A, Mumprecht V, Christiansen AJ, Huggenberger R, Leroux JC, Detmar M (2013) Non-invasive dynamic near-infrared imaging and quantification of vascular leakage in  vivo. Angiogenesis 16(3):525–540. https://doi.org/10.1007/ s10456-013-9332-2 7. Teijeira A, Hunter MC, Russo E, Proulx ST, Frei T, Debes GF, Coles M, Melero I, Detmar M, Rouzaut A, Halin C (2017) T cell migration from inflamed skin to draining lymph nodes requires intralymphatic crawling supported by ICAM-1/LFA-1 interactions. Cell Rep 18(4):857–865. https://doi.org/10.1016/j. celrep.2016.12.078 8. Kwon S, Sevick-Muraca EM (2007) Noninvasive quantitative imaging of lymph function in mice. Lymphat Res Biol 5(4):219–231. https://doi. org/10.1089/lrb.2007.1013 9. Sevick-Muraca EM, Kwon S, Rasmussen JC (2014) Emerging lymphatic imaging technologies for mouse and man. J  Clin Invest 124(3):905–914. https://doi.org/10.1172/ JCI71612

In Vivo Imaging of Lymphatic Function 10. Brambilla D, Proulx ST, Marschalkova P, Detmar M, Leroux JC (2016) Microneedles for the noninvasive structural and functional assessment of dermal lymphatic vessels. Small 12(8):1053–1061. https://doi. org/10.1002/smll.201503093 11. Proulx ST, Luciani P, Dieterich LC, Karaman S, Leroux JC, Detmar M (2013) Expansion of the lymphatic vasculature in cancer and inflammation: new opportunities for in  vivo imaging and drug delivery. J  Control Release 172(2):550–557. https://doi.org/10.1016/j. jconrel.2013.04.027 12. Karlsen TV, McCormack E, Mujic M, Tenstad O, Wiig H (2012) Minimally invasive quantification of lymph flow in mice and rats by imaging depot clearance of near-­ infrared albumin. Am J  Physiol Heart Circ Physiol 302(2):H391–H401. https://doi. org/10.1152/ajpheart.00842.2011 13. Levick JR, Michel CC (2010) Microvascular fluid exchange and the revised starling principle. Cardiovasc Res 87(2):198–210. https:// doi.org/10.1093/cvr/cvq062 14. Liao S, Cheng G, Conner DA, Huang Y, Kucherlapati RS, Munn LL, Ruddle NH, Jain RK, Fukumura D, Padera TP (2011) Impaired lymphatic contraction associated with immunosuppression. Proc Natl Acad Sci U S A 108(46):18784–18789. https://doi. org/10.1073/pnas.1116152108 15. Scallan JP, Davis MJ (2013) Genetic removal of basal nitric oxide enhances contractile activity in isolated murine collecting lymphatic vessels. J Physiol 591(Pt 8):2139–2156. https:// doi.org/10.1113/jphysiol.2012.250662 16. Gousopoulos E, Proulx ST, Scholl J, Uecker M, Detmar M (2016) Prominent lymphatic vessel hyperplasia with progressive dysfunction and distinct immune cell infiltration in lymphedema. Am J Pathol 186(8):2193–2203. https://doi. org/10.1016/j.ajpath.2016.04.006

211

17. Gogineni A, Caunt M, Crow A, Lee CV, Fuh G, van Bruggen N, Ye W, Weimer RM (2013) Inhibition of VEGF-C modulates distal lymphatic remodeling and secondary metastasis. PLoS One 8(7):e68755. https://doi. org/10.1371/journal.pone.0068755 18. Ma Q, Ineichen BV, Detmar M, Proulx ST (2017) Outflow of cerebrospinal fluid is predominantly through lymphatic vessels and is reduced in aged mice. Nat Commun 8(1):1434. https:// doi.org/10.1038/s41467-017-01484-6 19. Modi S, Stanton AW, Mortimer PS, Levick JR (2007) Clinical assessment of human lymph flow using removal rate constants of interstitial macromolecules: a critical review of lymphoscintigraphy. Lymph Res Biol 5(3):183–202. https://doi.org/10.1089/lrb.2007.5306 20. Bouta EM, Blatter C, Ruggieri TA, Meijer EF, Munn LL, Vakoc BJ, Padera TP (2018) Lymphatic function measurements influenced by contrast agent volume and body position. JCI Insight 3(2). https://doi.org/10.1172/ jci.insight.96591 21. Blum KS, Proulx ST, Luciani P, Leroux JC, Detmar M (2013) Dynamics of lymphatic regeneration and flow patterns after lymph node dissection. Breast Cancer Res Treat 139(1):81–86. https://doi.org/10.1007/ s10549-013-2537-7 22. Escobedo N, Proulx ST, Karaman S, Dillard ME, Johnson N, Detmar M, Oliver G (2016) Restoration of lymphatic function rescues obesity in Prox1-haploinsufficient mice. JCI Insight 1(2). https://doi.org/10.1172/jci. insight.85096 23. Nelson TS, Akin RE, Weiler MJ, Kassis T, Kornuta JA, Dixon JB (2014) Minimally invasive method for determining the effective lymphatic pumping pressure in rats using near-­ infrared imaging. Am J  Physiol Regul Integr Comp Physiol 306(5):R281–R290. https:// doi.org/10.1152/ajpregu.00369.2013

Chapter 14 Investigating Effects of Fluid Shear Stress on Lymphatic Endothelial Cells Daniel T. Sweet, Joshua D. Hall, John Welsh, Mark L. Kahn, and Juan M. Jiménez Abstract Recent studies using in vivo models have characterized lymph flow and demonstrated that lymph flow plays a key role in the later stages of lymphatic vascular development, including vascular remodeling, to create a hierarchical collecting vessel network and lymphatic valves (Sweet et al., J Clin Invest 125, 2995–3007, 2015). However, mechanistic insights into the response of lymphatic endothelial cells to fluid flow are difficult to obtain from in vivo studies because of the small size of lymphatic vessels and the technical challenge of lymphatic endothelial cell isolation. On the other hand, in vitro experiments can be tailored to isolate and test specific mechanotransduction pathways more cleanly than conditions in vivo. To measure in vitro the cellular response to flow, cultured primary lymphatic endothelial cells can be exposed to highly specific fluid forces like those believed to exist in vivo. Such in vitro studies have recently helped identify FOXC2 and GATA2 as important transcriptional regulators of lymphatic function during valve formation that are regulated by lymph flow dynamics. This chapter discusses the methods used to expose primary lymphatic endothelial cells (LECs) to lymph fluid dynamics and the relationship of these in vitro studies to in vivo lymphatic biology. Key words Lymphatic endothelial cells, Shear stress, In vivo lymphatic imaging

1  Introduction Endothelial cells, which line blood and lymphatic vessels, are constantly exposed to frictional, pulsatile forces induced by flowing blood or lymph. Endothelial cells are able to recognize and adapt to different wall shear stresses, amplitudes, and patterns via mechanotransduction [2]. Mechanosensors located on the cell membrane, such as receptor tyrosine kinases, G protein-coupled receptors, ion channels, integrins, and glycocalyx, transduce this stimulus into intracellular signals and changes in gene and protein expression. Transcriptional and translational changes are not the only response to fluid flow; endothelial cells also undergo cytoskeletal remodeling and align parallel to the direction of flow [3]. Guillermo Oliver and Mark L. Kahn (eds.), Lymphangiogenesis: Methods and Protocols, Methods in Molecular Biology, vol. 1846, https://doi.org/10.1007/978-1-4939-8712-2_14, © Springer Science+Business Media, LLC, part of Springer Nature 2018

213

214

Daniel T. Sweet et al.

Endothelial cells in the body can experience disturbed or undisturbed flow. Mechanobiological responses by blood endothelial cells to disturbed flow have been associated with cardiovascular disease [4], while undisturbed flow is important in vascular patterning and remodeling, and atheroprotection [2]. Lymphatic flow activates signaling within lymphatic endothelial cells through currently unknown mechanisms to drive lymphatic valve development via activation of FOXC2 and GATA2 transcription factors [1, 5–7]. Studies of endothelial mechanobiology often utilize parallel-­ plate flow chambers [8], cone-plate chambers [9], and microfluidic devices [10] to recreate wall shear stress environments in vitro. In vitro flow studies can focus on a range of cell phenomena, such as morphology [11], arrangement [12], and concentration of reactive oxygen species [10]. They also permit the use of siRNA and other reagents to alter gene and/or protein expression during the experimental process. Parallel plate flow chambers (PPFCs) are relatively simple to manufacture and operate, and the cells can be monitored in real time under a microscope. The PPFC is commonly perfused by peristaltic pumps that do not come in direct contact with the cell media, ensuring sterility. Furthermore, PPFCs allow for the recreation of various physiological profiles of pressure, flow, and wall shear stress in vitro [13]. Although very important in endothelial mechanobiology, wall shear stress is very difficult to measure inside PPFCs. For this reason, it is not uncommon to indirectly determine the in vitro flow field via computational fluid dynamics (CFD) simulations [14]. An advantage of using PPFCs for endothelial mechanobiology is the ability to expose large numbers of cells to the same stimuli in a controlled environment and then quantify transcriptional and translational changes, as well as other cellular responses such as intracellular signaling responses. The controlled environment of the PPFC also allows for exposure to a variety of stimuli including chemical, temperature, and stiffness of the underlying matrix, in the presence of a fluid flow field. The PPFCs can be adapted for evaluation of mechanical flow stimuli of both lymphatic and blood endothelial cells, as long as the pertinent waveform is known. Therefore, in vivo measurements of the flow field are paramount to accurately recreate the flow field in vitro. Herein, we demonstrate steps to measure the flow field in a lymphatic vessel in a mouse model. The resultant waveform can be employed in vitro to expose lymphatic endothelial cells to the in vivo mechanical stimuli, and reproduce in vivo flow conditions.

215

Testing Lymphatic Endothelial Responses to Fluid Shear

2  Human Lymphatic Endothelial In Vitro Cell Culture, Transfection and Lymphatic Flow Protocol 2.1  Materials

1. Cell medium: EGM-2MV bullet kit (Lonza) (see Note 1).

2.1.1  Cell Culture

2. T75 tissue culture flask with filtered cap. 3. 15 mL conical tube. 4. Sterile square Petri dishes (100 × 100 mm2). 5. 75 × 38 mm2 glass microscope slide (sterilized) (see Note 2). 6. Phosphate buffered saline (sterile), pH = 7.4. 7. Human dermal lymphatic microvascular endothelial cells. 8. Human plasma fibronectin. 9. Trypsin–EDTA (0.25%). 10. Dextran sulfate (see Note 3). 11. High-vacuum silicone grease.

2.1.2  In Vitro Flow System

1. Peristaltic pump (see Note 4). 2. Parallel plate flow chamber—PPFC (see Note 5). 3. Tubing and bottles for flow system (see Notes 6 and 7). 4. Flow meter (see Note 8). 5. Forceps, stainless steel dissection scissors, Petri dish with high-­ vacuum grease and cotton tip applicators inside. 6. Instant sealing (61 × 30 cm2).

2.1.3  Cell Isolation

sterilization

pouches

autoclave

bags

1. Cell scraper. 2. Lysis buffer (see Note 9).

2.2  Methods 2.2.1  Cell Culture

Carry out all procedures at room temperature, unless otherwise specified. 1. Thaw lymphatic endothelial cells following the manufacturer’s instructions. Seed cell content from vial into one T75 flask in 37 °C EGM2-MV media in the tissue culture hood (see Note 9). 2. The next day, change media to fresh EGM2-MV. Change media every 2–3 days until the flask is completely confluent (see Note 11). 3. Place six 38 × 75 mm2 sterilized glass slides into three square Petri dishes (two slides per dish, side by side) (see Note 2). 4. With a serological pipette thinly coat slides with 10 μg/mL fibronectin diluted in sterile PBS+. Incubate fibronectin on slides for about 1 h at room temperature. Aspirate fibronectin

216

Daniel T. Sweet et al.

and allow slides to dry for approximately 5 min in the tissue culture hood. 5. Split cells by gently washing one time with warm sterile PBS-, then remove cells from the flask surface by adding trypsin– EDTA and incubating at 37 °C for approximately 5 min until cells round up and detach. Suspend cells with PBS+ to neutralize the EDTA (see Notes 12 and 13). 6. Collect cell suspension in 15 mL conical tube, spin down for 3 min at 800 × g at room temperature to pellet the cells. Aspirate supernatant and resuspend cell pellet in 6 mL of warm EGM2-MV media. 7. Add 1 mL cell suspension to each of the six slides and distribute uniformly with a serological pipette tip. Let it sit for about 20 min in the hood to allow cells to begin attaching to fibronectin surface. Then, very gently, add 9 mL EGM2-MV media per square Petri dish and place in CO2 incubator (see Note 14). 2.2.2  siRNA Transfection (Optional Step)

1. Incubate cells on glass slides for 4 h or more in the CO2 incubator until slides are approximately 75% confluent (see Notes 15 and 16). 2. Mix transfection reagents according to manufacturer’s protocol. For each square petri dish (containing two slides), dilute a total of 150 pmol siRNA in 500 μL Opti-MEM media (serum free) and separately, 45 μL Lipofectamine RNAiMAX in 500  μL Opti-MEM. Mix diluted siRNA and diluted Lipofectamine, incubate 5 min at room temperature, and then add the entire 1 mL mixture to the EGM2-MV media in the square dish. siRNA sequences are shown in Table 1 (see Note 17) 3. Incubate overnight (about 18 h), then change media the next morning to fresh warm EGM2-MV. 4. Incubate cells for 1–2 days depending on the kinetics of the siRNA knockdown, then subject to lymphatic fluid flow waveform in full EGM-2MV media +5% dextran (see Note 17).

2.2.3  Flow Experiment

1. Warm up cell media to 37 °C. 2. Place two forceps and stainless steel dissecting scissors in self-­ sealing autoclave pouch and seal. 3. In a 61 × 30 cm2 autoclave bag, place the top plate of the parallel plate flow chamber (PPFC). In another 61 × 30 cm2 autoclave bag, place the bottom plate metal half of the PPFC. In a third 61 × 30 cm2 autoclave bag, place tubing assembly with caps and 0.2 μm filter attached, Petri dish with grease and cotton tip applicator, petri dish with bolts, nuts and washers, and two 250 mL Pyrex bottles. Seal all bags with packing tape.

Direction

GAPDH F

GAPDH R

PROX1 F1

PROX1 R1

FoxC2 F1

FoxC2 R1

Cx37 F

Cx37 R

VEGFR3 F

VEGFR3 R

LYVE1 F

LYVE1 R

Nrp-1 F

Nrp-1 R

ephrinB2 F

ephrinB2 R

Itga9 F

Itga9 R

Gata2 F

Gata2 R

Gata2 F2

Gata2 R2

Gene

GAPDH

GAPDH

Prox1

Prox1

Foxc2

Foxc2

Cx37

Cx37

VEGFR3

VEGFR3

Lyve1

Lyve1

Nrp-1

Nrp-1

ephrinB2

ephrinB2

Itga9

Itga9

Gata2

Gata2

Gata2–2

Gata2–2

Table 1 siRNA sequences

CACAGGCATTGCACAGGTAGT

CCTCCAGCTTCACCCCTAA

CATGGTCAGTGGCCTGTTAACATTGTGCAG

CTGCACAATGTTAACAGGCCACTGACCATG

ATCCACTCATCATCGCGGTC

GAGATGCACCGAACTGGACA

CCCTCTCCTCAACTGTGCCAAACC

TCCAGGCCCTCCAAAGACCCA

CAGAGCGCTCCCGCCTGAAC

AAATGGCGCCCTGTGTCCCG

AGCCTACAGGCCTCCTTAGC

ACTTCCATCTGGACCACGAG

CTCTGCCTGGGACTCCTG

GGTGTCGATGACGTGTGACT

CATCGTCCCCACCTCCAC

GGTCCAGCAACTTCTCCAGG

GAGCCGTCTCGGAAGCAG

CGGCTTCACCAGGTCCTTAG

ACAGGGCTCTGAACATGCAC

GGCATTGAAAAACTCCCGTA

GGCATGGACTGTGGTCATGAG

TGCACCACCAACTGCTTAGC

Sequence

human

human

human

human

human

human

human

human

human

human

human

human

human

human

human

human

human

human

human

human

human

human

Species

[21]

[21]

[20]

[20]

[19]

[19]

[19]

[19]

[19]

[19]

[18]

[18]

[18]

[18]

[19]

[19]

[19]

[19]

[18]

[18]

[18]

[18]

References

Testing Lymphatic Endothelial Responses to Fluid Shear 217

218

Daniel T. Sweet et al.

4. Autoclave items in wet cycle for 30 min at 105 °C and after sterilization allow for items to cool down. 5. In a sterilized tissue culture hood, introduce pouch with scissors and forceps, autoclave bag with tubing, and the bag containing the top half of flow chamber (see Note 18). 6. Open pouch containing the scissors, remove sterilized scissors and use scissors to cut open autoclave bags with flow chamber accessories. Place top half of PPFC on a shallow test tube tray and connect tubing to fittings. From the 500 mL of cell media, add about 240 mL to each 250 mL Pyrex bottle. Push in retrofitted stopper into bottle closest to PPFC inlet and screw retrofitted caps on both bottles. Gently push assembled half toward back of tissue culture hood and place bottom half of PPFC on a low profile test tube tray closer to user. 7. Add a small dab of grease to each slide well, and then add about 1 mL of cell media to each slide well (Fig. 1). With the sterile forceps, gently place a glass slide with cells (cells on top side) on slide well and gently, but firmly press downward (see Note 19). 8. Once all slides have been placed and secured, another 1 mL of cell media should be added to each glass slide ensuring hydration during assembly of PPFC. 9. Top half of PPFC should be flipped over and placed on top of bottom half. Nuts and bolts should be installed (see Note 20).

Fig. 1 Metal half of parallel plate flow chamber denoting the slide wells for slide, side channels that enable quick removal of glass slides and vinyl screw that fastens glass slides in place

Testing Lymphatic Endothelial Responses to Fluid Shear

219

10. Once assembled, carry PPFC and tubing-bottle assembly to incubator and insert tubing into pump head (see Note 21). 11. PPFC is initially perfused at low flow rates and held at a 45° angle with PPFC outlet raised to ensure exit of any bubbles, which can take about 30 s. Once bubbles have been cleared, the desired waveform can be selected. For these experiments, a lymphatic flow waveform from the literature was chosen [1]. 2.2.4  Ending Flow Experiment and Isolating Cells

1. After 48 h, the experiment is stopped. PPFC is removed from incubator and placed on ice on benchtop to decrease biomolecular kinetics (except if cells will be used for immunofluorescence, see Note 22). 2. Glass slides with cells are washed with cold PBS+ and immediately incubated in lysis buffer (RNA/Protein), cells are detached from surface with cell scraper and aspirated with micropipette for further experiments. Lysate can now be processed to isolate RNA and/or protein by standard protocols and subsequently used for qPCR or western blot assays (see Note 8).

2.2.5  Notes

1. Add the supplements provided with the kit to the cell medium bottle. 2. 38 × 75 mm2 glass slides should be sterilized by autoclaving before use at a minimum temperature of 120 °C for 30 min. 3. In order to recapitulate the physiological wall shear stress while maintaining the low flow rate present in lymphatic vessels in vivo, add dextran to increase the dynamic viscosity of the cell culture media. For our experiments adding 5% dextran yielded a 2.74 cP dynamic viscosity cell culture media (see Table 2 for additional values). The optimal concentration of dextran can be determined empirically with a viscometer similar to the Cannon-Ubbelohde type 0C viscometer used in these experiments. The viscosity of cell media with different dextran Table 2 Dynamic viscosity of EGM-2MV cell media as function of dextran concentration % Dextran

Dynamic viscosity (cP)

0

0.72

1

1.00

2

1.29

5

2.74

10

6.98

220

Daniel T. Sweet et al.

Fig. 2 Example of (a) flow waveform and (b) input signals necessary to generate a flow waveform. (c) The control input signals for the Watson Marlow 520U pump provided via DB25 port in the back of the pump. Signals are unique for each system and need to be modified as necessary

c­ oncentrations is determined with the viscometer as it stands upright inside a CO2 incubator per manufacturer’s instructions. 4. Peristaltic pumps are preferred because the pump parts do not come in direct contact with the cell media, lessening probability of contamination. For these experiments, a Watson-­Marlow 520U peristaltic pump was used. The pump was controlled with a computer along with the LabVIEW software and data acquisition (DAQ) card from National Instruments. Although the LabVIEW software and DAQ card were used to control the pump, similar objectives can be accomplished with other brand controllers. Figure 2a shows a sample waveform that is generated by the pump speeds and pump direction signals (Fig. 2b). Figure 2c shows the DB25 port on the rear of the pump and the signals required to control the Watson-Marlow 520U peristaltic pump. 5. Many examples can be found of parallel plate flow chambers (PPFC) in the literature. PPFCs can be manufactured relatively easily by a machine shop. Some of the basic design variables that are to be considered are discussed in the next few sentences. PPFC materials selected should be easily machined and capable of withstanding the high temperature and high pressure environment of an autoclave. The thermal coefficient of the different materials that together make up the PPFC should be considered, since materials with very different thermal coefficients will expand by different amounts potentially causing failure. The width of the domain where fluid travels should be much wider than taller. It is suggested that the width of the channel is at least 50 times wider than taller. This ensures that the cells on the glass slides experience a homogeneous flow

Testing Lymphatic Endothelial Responses to Fluid Shear

221

field. A glass slide with cells should never rest on or be in close proximity to the side walls, since a boundary layer develops add the wall yielding a nonuniform flow field over the glass slide. The bottom half of chamber is manufactured with 318 stainless steel. We use top halves made out of either PTFE or PEEK. 6. Tubing similar to PharMed BPT is recommended. In our experience, this type of tubing can withstand repeated peristaltic pump cycles and autoclaving. When new tubing is used, it is suggested to conduct a 2 day run with water, since when new it is common to observe small particles in the solution due to a break-in period of the tubing. 7. Figures 3 and 4 detail the different parts that make up the flow system [15]. 8. A Transonic T402 flow meter was used to measure the flow rate at the inlet of the PPFC. Throughout the experiment, the inlet PPFC flow rate is measured continuously with the flow meter, while sampled and saved for 3 s every 5 min, allowing monitoring of the experiment. In addition, the flow meter output signal is displayed continuously on an oscilloscope. It is highly recommended that a flow meter be used, since the input signal given to the pump is meaningless unless the flow system has been calibrated to determine the relationship between input voltage to the pump and fluid flow rate in the PPFC. Calibration of the system can be conducted by providing the pump an input voltage and measuring the flow rate at the inlet of the PPFC with the flow meter. This process is

Fig. 3 The parallel plate flow chamber (PPFC) has an inlet and outlet that are connected to Pyrex bottles. Tubings of different sizes are used to accommodate for spatial constraints of the PPFC and pump heads

222

Daniel T. Sweet et al.

Fig. 4 Flow system running inside a 5% CO2 incubator. Pump is placed inside incubator and no water is added to incubator to protect electronics. Flow meter sensor is installed at inlet. Two bottles serve as reservoirs, where the hermetically sealed bottle, on the back of the top shelf, also serves as a flow dampener by removing high frequency fluctuations, while the second bottle, on the bottom shelf, allows gas exchange through the 0.2 μm filter. The outlet of the flow chamber is raised about 15° to allow air bubbles to exit flow chamber, since air bubbles can affect cells negatively

repeated at discrete voltage intervals throughout the 0–10 V range for both the clockwise and counterclockwise rotational directions of the pump head. This is information is used to determine the voltage to flow rate relationship, which is unique for each system. 9. A standard RIPA buffer with phosphatase and protease inhibitors was used for protein isolation by standard protocol. RNA was isolated using Qiagen RNEasy kit following manufacturer’s instructions. Characterization of western Blot or qPCR conditions is beyond the scope of this chapter. These conditions should be optimized by the operator for the individual experiments. 10. All cell culture work should be performed in a sterile biosafety cabinet following strict aseptic technique to avoid cell contamination. 11. Best practice is to expand early passage cells and freeze stocks of approximately one million cells per vial in EGM2-MV supplemented with 20% (final concentration) FBS and 10% DMSO. 12. Human dermal lymphatic microvascular endothelial cells are primary cells, so passage number should be recorded and cell morphology should be carefully monitored often. Do not use

Testing Lymphatic Endothelial Responses to Fluid Shear

223

cells after passage 9 or if growth becomes noticeably slower or morphology changes from the typical cobblestone shape. 13. If cells will not detach after 5 min in Trypsin-EDTA, it is safe to tap the side of the flask gently to mechanically dislodge them. 14. Add cell medium to petri dish while avoiding glass slide surface in order to not dislodge the cells. 15. It is important for cells to be subconfluent during siRNA transfection. If slides are under 75 percent confluent wait a few hours for some cell division to occur. If slides are nearly confluent, transfection efficiency will decrease. 16. If gene knockdown is not required, skip to Subheading 2.2.2. 17. This protocol was optimized using ThermoFisher Silencer Select siRNA oligos, results with siRNA from other suppliers may vary. It is important to use Lipofectamine RNAiMAX, which produced far superior gene knockdown than other transient transfection reagents in the human LECs. 18. When placing halves of PPFC on tissue culture hood surface, it helps to place them on top of a shallow test tube tray for ease of manipulating parts. 19. Minimize manipulation of glass slides with forceps. Every contact will remove cells from the surface. 20. Bolts and nuts should be installed and gently tightened from center of the PPFC outward, alternating from side to side and top to bottom following a crisscross pattern. Then, tighten each nut in the clockwise direction. Ensure not to overtighten, otherwise cell media leakage may occur. If media leakage occurs, slightly loosen bolts and nuts in affected area. 21. Tubing should be marked so that placement within pump head is constant for all experiments. This will ensure repeatability of waveform. It is advised to draw a line on tubing as a reference point. 22. If cells on a slide are to be probed via immunofluorescence, do not place on ice, but wash with PBS+ at room temperature three times and then incubate in 10% formalin for 20 min.

3  Mouse In Vivo Tracking of Lymphatic Flow to Determine Local Flow Patterns 3.1  Materials 3.2.1  In Vivo Fluid Flow Measurement

1. FluoSpheres—1  μm diameter carboxylate-coated microspheres—fluorescent tag of your choice. 2. 32 gauge needles. 3. Glass syringe (10 μL). 4. N-hydroxysuccinimide.

224

Daniel T. Sweet et al.

5. N-ethyl-N′-(3-dimethylaminopropyl) chloride.

carbodiimide

hydro-­

6. Polyethylene Glycol (PEG). 7. Sterile PBS+. 8. 3-aminopropyltriethoxysilane (APTES solution). 9. Mice (approx. 4–8 weeks old) (see Note 23). 10. Live animal imaging suite with sterile surgical tools, isoflurane anesthesia, heating pad, fluorescent dissecting microscope and high frame rate video camera. 3.2  Methods 3.2.1  Cell Culture

1. PEG-ylation of FluoSpheres fluorescent microspheres (see Note 24) (adapted from [16]):

(a) Make a solution containing 10 mM N-­hydroxysuccinimide and 20 mM N-ethyl-N′-(3-dimethylaminopropyl) carbodiimide hydrochloride. (b) To 950  μL of this solution, add 50 μL of microspheres (≈1.82 × 109 microspheres) and incubate for 1 h at room temperature.





(c) Centrifuge briefly to pellet beads at 1000 × g, remove supernatant and resuspend beads in 1 mL of 95% ethanol. Spin again to pellet beads and remove as much supernatant as possible. Air dry.



(d)  Incubate PEG with APTES solution (2% by volume 3-­aminopropyltriethoxysilane in 100% ethanol) for 2 h at room temperature.



(e) Centrifuge PEG solution at 1000 × g, remove supernatant and wash pellet with 1 mL of 95% ethanol one time. Gently dry with filtered compressed air. Install a 0.2 μm on compressed air line.



(f) Combine the chemically activated microspheres from step c with the amino-terminated PEG from step d for 30 min at room temperature. Spin down microspheres, remove supernatant and resuspend in 180 μL sterile PBS+ (≈1010 particles/mL).



(g) Take 5  μL from microsphere solution in step (f) and add 495 μL of sterile PBS+ for an approximate concentration of 1 × 108 microspheres per mL. Approximately 10 μL beads are required per lymph node to be injected.

(h) Anesthetize mouse using an isoflurane vaporizer at a flow rate of 200 mL/min with 3% isoflurane, keep on warm heating pad (37 °C) and place in the supine position in an isoflurane nosecone to maintain proper surgical plane in which mouse is deeply anesthetized and does not react to toe pinch.

Testing Lymphatic Endothelial Responses to Fluid Shear

3.2.2  Particle Imaging

225



(i) Cut the ventral abdominal skin up the midline (from pelvis to sternum), then cut skin laterally toward front and hind limbs. Gently peel skin flap away from abdomen and pin down to board.



(j) In the posterior end of the skin flap, identify the inguinal lymph node in the skin overlying the groin and the efferent collecting vessel coming out of the lymph node and going toward the forelimb (see Note 25).

1. Place mouse and heating pad in dissecting microscope region of interest. Rotate mouse-heating pad so that vessel of interest aligns with horizontal axis of camera (see Note 26). 2. Using a Hamilton syringe and a 32 gauge needle, inject 5 μL or less of bead solution directly into the lumen of the inguinal lymph node, applying minimal pressure (see Note 27). 3. Image efferent lymphatic vessels several millimeters downstream from injection site (see Notes 28–30).

3.2.3  Image Analysis

1. Analyze the movement of lymphatic vessel walls and particles in the vessels to reconstruct waveforms that can be used in in vitro experiments to stimulate lymphatic endothelial cells (see Notes 31–33).

3.3  Notes

1. Young mice are easier to visualize because as mice age, adipose tissue is deposited around lymph nodes and lymphatic collecting vessels, making imaging and accurate injection of the inguinal lymph node difficult. Additionally, transgenic mice with lymphatic endothelial reporter genes improve visualization of lymphatic walls. We prefer the Prox1-GFP BAC-­ transgenic [17] mouse; however, several different cre-activated and constitutive lines are available. This is especially important when measuring flow patterns because lymphatic vessels can contract and relax at high frequencies and contractions can decrease the lumen diameter significantly. 2. Microspheres may be injected without any modification; however, in our experience, microspheres rapidly adhere to lymphatic vessel walls and each other, making tracking difficult. Coating microspheres with PEG decreases the interaction of particles with each other and other cells in the vessel. 3. Identification of lymph node and lymphatic vessels is straightforward in transgenic reporter animals, as both will light up intensely. If no reporter is present, the lymph node is a small white pouch, and the efferent lymphatic is transparent or slightly white vessel running alongside a large vein. 4. It is recommended that the dissecting microscope is equipped with a dual-wavelength fluorescent filter to enable s­ imultaneous

226

Daniel T. Sweet et al.

visualization of both the transgenic reporter molecule and the fluorescent microspheres. Ensure that an IR filter is installed to control thermal effects from the light source. A color camera is preferable to separate the lymphatic vessel wall and microsphere information. 5. This can be performed using a single bolus injection of a small volume, or by slow infusion of microspheres directly into the lymph node. 6. Animal breathing and local movements can result in some out of focus images. Record the flow for several minutes to ensure acquisition of in focus periods of time. 7. A camera with a minimum ISO = 10,000 sensitivity is required to capture the rapidly moving particles at short exposure times on the order of milliseconds. 8. Only capture images about 10 s after the first particles appear in the region of interest. Stop image acquisition when the volume of particles decreases, since the only particles present will be near the vessel walls. Hence, the flow rate will be interpreted as lower than it actually is. 9. Before particle and wall motion can be analyzed, the images need to be stabilized computationally to subtract breathing and local motions. 10. The authors tried multiple software packages, including freeware and commercial programs, to track particles in lymphatic vessels. However, the highly complex nature of lymphatic flow could only be accurately tracked using an interactive plugin with the ImageJ software, especially because of tissue movement. 11. The vessel wall motion can be extracted with different software packages, including multiple plugins in the ImageJ software. References 1. Sweet DT et al (2015) Lymph flow regulates collecting lymphatic vessel maturation in vivo. J Clin Invest 125:2995–3007 2. Davies PF (1995) Flow-mediated endothelial mechanotransduction. Physiol Rev 75:519–560 3. DePaola N et al (1999) Spatial and temporal regulation of gap junction connexin43 in vascular endothelial cells exposed to controlled disturbed flows in vitro. Proc Natl Acad Sci 96:3154–3159 4. Passerini AG et al (2005) Regional determinants of arterial endothelial phenotype dominate the impact of gender or short-term

exposure to a high-fat diet. Biochem Biophys Res Commun 332:142–148 5. Sabine A et al (2015) FOXC2 and fluid shear stress stabilize postnatal lymphatic vasculature. J Clin Invest 125:3861–3877 6. Kazenwadel J et al (2015) GATA2 is required for lymphatic vessel valve development and maintenance. J Clin Invest 125:2879–2994 7. Sabine A et al (2012) Mechanotransduction, PROX1, and FOXC2 Cooperate to Control Connexin37 and Calcineurin during Lymphatic-Valve Formation. Dev Cell 22:430–445 8. Levesque MJ, Nerem RM (1985) The elongation and orientation of cultured endothelial

Testing Lymphatic Endothelial Responses to Fluid Shear cells in response to shear stress. J Biomech Eng 107:341–347 9. Davies PF, Remuzzi A, Gordon EJ, Dewey CF, Gimbrone MA (1986) Turbulent fluid shear stress induces vascular endothelial cell turnover in vitro. Proc Natl Acad Sci U S A 83:2114–2117 10. Chin LK et al (2011) Production of reactive oxygen species in endothelial cells under different pulsatile shear stresses and glucose concentrations. Lab Chip 11:1856 11. Helmke BP, Davies PF (2002) The cytoskeleton under external fluid mechanical forces: Hemodynamic forces acting on the endothelium. Ann Biomed Eng 30:284–296 12. Kuo YC et al (2015) Oscillatory shear stress mediates directional reorganization of actin cytoskeleton and alters differentiation propensity of mesenchymal stem cells. Stem Cells 33:429–442 13. Estrada R et al (2011) Endothelial cell culture model for replication of physiological profiles of pressure, flow, stretch, and shear stress in vitro. Anal Chem 83:3170–3177 14. Vogel M, Franke J, Frank W, Schroten H (2007) Flow in the well: computational fluid dynamics is essential in flow chamber construction. Cytotechnology 55:41–54

227

15. Jiménez JM et al (2014) Macro- and microscale variables regulate stent haemodynamics, fibrin deposition and thrombomodulin expression. J R Soc Interface 11:20131079 16. Zhang ZL, Crozatier C, Le Berre M, Chen Y (2005) In situ bio-functionalization and cell adhesion in microfluidic devices. Microelectron Eng 78–79:556–562 17. Choi I et al (2011) Visualization of lymphatic vessels by Prox1-promoter directed GFP reporter in a bacterial artificial chromosome-­ based transgenic mouse. Blood 117:362–365 18. Deng Y, Atri D, Eichmann A, Simons M (2013) Endothelial ERK signaling controls lymphatic fate specification. J Clin Invest 123:1202–1215 19. Levet S et al (2013) Bone morphogenetic protein 9 (BMP9) controls lymphatic vessel maturation and valve formation. Blood 122:598–607 20. Shimahara A, Yamakawa N, Nishikata I, Morishita K (2010) Acetylation of lysine 564 adjacent to the C-terminal binding protein-­ binding motif in EVI1 is crucial for transcriptional activation of GATA2. J Biol Chem 285:16967–16977 21. Ahn EE et al (2013) SON protein regulates GATA-2 through transcriptional control of the MicroRNA 23a 27a 24-2 cluster. J Biol Chem 288:5381–5388

Chapter 15 Methods for Assessing the Contractile Function of Mouse Lymphatic Vessels Ex Vivo Jorge A. Castorena-Gonzalez, Joshua P. Scallan, and Michael J. Davis Abstract Lymphatic contractile dysfunction has been identified in several diseases, including lymphedema, yet a detailed molecular understanding of lymphatic muscle physiology has remained elusive. With the advent of genetic methods to manipulate gene expression in mice, a set of new tools became available for the investigation and visualization of the lymphatic vasculature. To gain insight into the molecular regulators of lymphatic contractile function, regulated primarily by the muscle cell layer encircling lymphatic collecting vessels, ex vivo approaches to allow control of hydrostatic and oncotic pressures and flow have been invaluable, complementing in vivo methods. While the original ex vivo techniques were developed for lymphatic vessels from large animals, and later adapted to rat vessels, here we describe modifications that enable the study of isolated, pressurized murine lymphatic collecting vessels. These methods, used in combination with transgenic mice, can be a powerful tool to investigate the molecular and cellular mechanisms of lymphatic function. Key words Cannulation, Pressure, Diameter, Contraction, Perfusion, Collecting lymphatic vessel

1  Introduction Ex vivo methods to study pressurized large blood vessels have been around for decades [1] but were scaled down in the 1980s for studies of small arteries and arterioles [2, 3]. The scaling process borrowed heavily from perfusion techniques for renal tubules developed by Burg and colleagues at the National Institutes of Health [4]. In the 1990s, the small vessel methods were adapted further for the ex vivo study of rat collecting lymphatic vessels [5, 6] and subsequently by us and others for mouse collecting lymphatic vessels [7–9]. The following chapter describes the protocols used in our laboratories for the isolation, cannulation, and ex vivo study of mouse collecting lymphatic vessels. Although the detailed descriptions given here are specific for popliteal afferent lymphatics, they are readily adaptable to lymphatic vessels from several different anatomic regions of the mouse as we have reported recently Guillermo Oliver and Mark L. Kahn (eds.), Lymphangiogenesis: Methods and Protocols, Methods in Molecular Biology, vol. 1846, https://doi.org/10.1007/978-1-4939-8712-2_15, © Springer Science+Business Media, LLC, part of Springer Nature 2018

229

230

Jorge A. Castorena-Gonzalez et al.

[10]. These methods are in large part modifications of techniques described for arterioles [11], with adaptations to control inflow and outflow pressures independently over a much lower pressure range, 0–20 cm H2O, corresponding to the intraluminal pressure range measured in mesenteric lymphatic networks of the rat [12, 13]. Unfortunately, corresponding pressure measurements are not available for the mouse, but it is assumed that the pressure levels are comparable to, or perhaps slightly lower than, those in the rat. In vivo protocols for the study of mouse popliteal lymphatics are given in other chapters, and comparisons of in vivo and ex vivo methods are the subject of a recent review [14].

2  Materials Prepare all solutions using ultrapure water (purified to 18 MΩ cm at 25 °C) and analytical grade reagents. 2.1  Krebs Solution Supplemented with 0.5% BSA

1. Krebs buffer (in mM): 146.9 NaCl, 4.7 KCl, 2 CaCl2∙2H2O, 1.2 MgSO4, 1.2 NaH2PO4∙H2O, 3 NaHCO3, 1.5 Na-HEPES, and 5 d-glucose. Add BSA and adjust the pH to 7.4 at 37 °C (see Note 1).

2.2  Sharpened Forceps and Microscissors

1. Use Moria forceps and Vannas scissors (Fine Science Tools #11399-87, #15369-00, respectively) for initial dissection. The forceps are sharpened using successive grades of Aluminum grit (Thomas Scientific #6775E38) adhered to a 1/8″ thick piece of Lucite. Sharpening procedures are described in ref. 2 (see Note 2). 2. Dumont 45° angled fine forceps (Fine Science Tools #11251-­ 35), sharpened to 10-μm tips, are used for cannulation.

2.3  Dissection Chamber, Recessed into Table (See Note 3) 2.4  Fine Steel Wire for Pinning 2.5  Two Glass Micropipettes with Tip Diameters 30–50 μm

1. The dissection dish should be recessed into a table top, allowing the operator’s elbows, forearms, and hands to rest on a stable base during the pinning and initial cleaning of the lymphatic vessel. See Fig. 2. See Note 4. 1. Pull glass pipette tubing on a pipette puller or microforge (see Note 5). 2. Lightly fire-polish the tips (see Note 6). 3. Use a heated filament on the microforge to bend the tip 30° so that it matches the angle of the V-track system; once mounted, the end of the pipette will be parallel with the bottom of the cannulation chamber.

Ex Vivo Methods to Study Mouse Lymphatic Contractile Function

231

2.6  Two Pipette Holders with Micromanipulator Mounting System

1. Clamp the prefilled pipettes to the manipulators. The manipulators must be mounted on a stable base platform to which the perfusion chamber is also mounted (see Note 7).

2.7  12-0 Monofilament Suture or Teased Strands of 4-0 Silk Suture (See Note 8)

1. Cut the suture to 2–3 mm lengths and pretie into single-tie loops using a loose surgical knot. This is best performed on a wetted Kimwipes (to help hold the suture in place and reduce static electricity), using the dissection microscope in advance of the experiment.

2. Use the y-axis control to lower the pipette tip to about 2 mm above the surface of the glass slide that comprises the bottom of the cannulation/bath chamber. Use the x-axis control(s) to separate the pipette tips by about 1 cm in preparation for vessel cannulation.

2. A loop of suture should be loosely secured to the shank of each glass micropipette tip before cannulating the vessel. 2.8  Syringe with Filtered Krebs-­ BSA Solution to Fill Pipettes

1. Fill two 10-mL syringes with Krebs + BSA. 2. Attach a 0.8 μm filter to each syringe. 3. Attach a 3-way stopcock and 16-G needle hub to each filter. 4. Attach a 1-ft length of PE-190 tubing to each needle hub. 5. With the stopcock turned appropriately, pressurize each syringe to fill the entire line with Krebs-BSA, making sure there are no remaining air bubbles. 6. Attach each piece of tubing to one port of the pipette holder (or directly to the back of the cannulating pipette if not using a pipette holder) (see Note 7).

2.9  Dissection Microscope, Magnification Range 8–64×

See Note 9.

2.10  Dual Fiber-­ Optic Light Source

See Note 10.

2.11  Heated Perfusion Chamber for the Ex Vivo Experimental Studies

See Note 11.

2.12  Inverted Microscope

See Note 12.

232

Jorge A. Castorena-Gonzalez et al.

2.13  Temperature-­ Controlled Water Circulator and Bath Superfusion Pump 2.14  Pressure Control System (See Note 14)

See Note 13.

1. The simplest pressure control system is a single (or pair of) moveable reservoir(s) mounted on a post or wall (e.g., with pulley system) near the microscope. Inverted glass syringe (e.g., 30 ml) barrels can serve as reservoirs, connected through needle hubs to appropriate lengths of Silastic (Cole-Palmer) and/or polyethylene (Intramedic) tubing to the back of the glass cannulation pipettes. The zero point of the reservoir should be level with the vertical position of the vessel in the perfusion chamber after it is mounted on the microscope (e.g., 1–2 mm above the bottom glass surface of the chamber). The transmural pressure is the important variable determining vessel responses and may need to be corrected for the depth of fluid covering the vessel. 2. For most experiments the tubing from the reservoirs to the pipette holder can be filled with distilled H2O because there will not be sufficient cumulative flow through the cannulated mouse vessel, even over several hours, to displace the 0.3– 0.5 mL of Krebs-BSA in the connecting tubing/pipette holder/cannulating pipette. However, protocols involving extended periods of unidirectional flow may require filling the entire inflow line with Krebs-BSA, which should then be thoroughly cleaned after each experiment. 3. If a more sophisticated (e.g., computer-controlled) pressure system is available, connect both it and the reservoir system to a high-quality, three-way stopcock (e.g., HPLC grade; Hamilton, Reno NV) and connect the outflow port(s) to the back(s) of the cannulation pipette(s) or pipette holder(s).

2.15  Video-Camera, Computer, and Diameter Measurement Software

See Note 15.

3  Methods Carry out all procedures at room temperature, unless otherwise specified. 3.1  Surgical Procedures to Expose the Popliteal Afferent Lymphatic Vessels

1. A 6- to 8-week-old mouse (optimal age) is anesthetized with an injection of sodium pentobarbital or ketamine–xylazine, i.p. Once a surgical plane of anesthesia is reached, shave the dorsal surface of one hind leg and place the mouse face down on a

Ex Vivo Methods to Study Mouse Lymphatic Contractile Function

233

heating pad. Extend one hind leg and hold it in place with a piece of tape. 2. Using a dissection scope at low power (8×), locate the saphenous vein on the medial-dorsal surface of the hind leg and make a 1 cm-long incision over and parallel to the vein while gently pulling upward on the skin to avoid nicking the vein. 3. As soon as the skin is opened, moisten the exposed tissue with room-temperature Krebs-BSA. Pull the medial edge of the cut skin toward the animal’s midline and hold it open using 2–3 small clamps (#18038-45 or 18054-28, Fine Science Tools, Foster City, CA), spaced a few mm apart. Slide the wide end of a cotton-tipped applicator (Puritan, #806-WCL) under the clamps so they are raised slightly and form a well that holds Krebs-BSA. Flush any loose hair or debris from the incision by dripping Krebs-BSA as needed from a syringe at the proximal end of the incision while using a suction tube at the distal end, formed from PE-190 tubing (Intramedic, Fisher Scientific) and connected to a vacuum line (with water trap). 3.2  Excise the Vessel

1. A pair of popliteal afferent lymphatic vessels run alongside and parallel to the saphenous vein from the ankle to the knee, deviating from the saphenous vein as they approach the popliteal node deeper behind the knee (see Fig. 1a; a short segment of each vessel is demarcated with dotted lines). With practice these vessels can be observed readily in younger mice (due to less perivascular adipose tissue) without the distal injection of any contrast agents [14]. Use coarse microscissors to cut the loose connective tissue on the lateral side of the more superficial popliteal lymphatic (away from the saphenous vein) for a distance of 5–8 mm. The vessel will retract toward the saphenous vein. 2. Carefully grab the remaining loose connective tissue near the distal end of the vessel with one of the coarse forceps, being careful not to grab the vein, and pull it slightly upward. While still holding the loose connective tissue, use the coarse microscissors to cut between the popliteal lymphatic and the vein. Once they separate, cut the distal end of the vessel and, continuing to hold the loose connective tissue and severed end of the popliteal lymphatic and pulling it up slightly from the rest of the tissue (but still covered by Krebs-BSA), extend the cut between the popliteal lymphatic and the saphenous vein proximally. (It is also possible to first cut the proximal end, retaining lymph and pressure in the vessel and then work distally, but that method is more difficult for a beginner.) The lymphatic vessel, covered by connective tissue and fat will separate from the vein and cutting should become easier (without risk of nicking the saphenous vein) as the cut is extended. Stop when

234

Jorge A. Castorena-Gonzalez et al.

Fig. 1 (a) Photo of surgically exposed region of the mouse hind limb showing the two popliteal afferent lymphatic vessels (diameter ~60 μm) coursing in parallel with the superficial saphenous vein. The skin overlying the region has been retracted using several small clamps (one clamp is shown). (b) A photomontage of a cannulated popliteal afferent lymphatic; the diameter of the outflow pipette is 50 μm

the vessel turns and runs deeper toward the popliteal node. The vessel can be dissected more proximally but at this point there should be at least 0.5 cm length of vessel free. 3. Cut the proximal end and, continuing to grasp the distal end, lift the vessel out of the animal and place it in a shallow 35-mm petri dish filled with Krebs-BSA. The BSA helps prevent the vessel from sticking to the forceps and other surfaces. If needed, the medial popliteal lymphatic can be removed using a somewhat similar procedure; the medial vessel tends to be more branched than the lateral vessel. If the vein is nicked, it will bleed profusely. In that case keep flushing the area with KrebsBSA for 2–3 min until the bleeding stops; the vein will constrict, making both popliteal vessels easier to see and remove. If the area is flushed well to prevent blood from clotting around the popliteal vessels there will be no permanent damage to the lymphatics from the blood. (A similar procedure can be used to dissect lymphatic vessels from other regions of the mouse, with only minor modifications, as we have recently described [10]; the dissection of mesenteric lymphatics is described in Chapter 7.)

Ex Vivo Methods to Study Mouse Lymphatic Contractile Function

235

4. Upon completion of this step, the animal can be removed from the heating pad and euthanized. 3.3  Set Up Pipettes and Chamber

1. Both cannulation pipettes will be connected at their back ends to a 10-mL disposable syringe filled with Krebs-BSA. After filling the syringe, connect it to the luer port of a 0.8-μm disposable filter. Connect this filter to a 3-way stopcock that has a 16-G needle hub on the opposite end. Affix ~6–12″ of PE-190 tubing to the 16 G needle hub. Pressurize the syringe to fill all of the components, taking care to avoid introducing small pockets of air. 2. Fill the perfusion chamber with filtered Krebs-BSA and then attach the syringe to the pipette holder (or pipette if not using a holder) and pressurize further to fill the holder, making sure not to leave any bubbles. Insert the pipette into the holder and pressurize the syringe to fill the pipette with buffer. Mount the pipette/holder to the micromanipulator. 3. Repeat this procedure for the other cannulation pipette.

3.4  Prepare the Vessel for Cannulation

1. Using a siliconized Pasteur pipette with fire-polished tip, transfer the excised popliteal lymphatic from the 35-mm petri dish to the recessed dissection dish, which should be prefilled 0.5 cm deep with room temperature, filtered Krebs-BSA (see previous step, Fig. 2). Dissection in room temperature Krebs-­ BSA is preferred for rat and mouse lymphatic vessels, in contrast to arterioles, which require dissection at 4 °C, necessitating the use of a cooling jacket around the dissection chamber. 2. Using the coarse forceps, grab a piece of 40-μm stainless steel wire and pin one end of the vessel to the Sylgard-coated bottom of the dissection chamber. Then grab the connective tissue covering the other end of the vessel, lengthen the vessel until slack is removed, and pin the other end. The pins should be driven through the connective tissue or fat rather than the vessel wall itself. 3. Use both fine forceps to gently tease and loosen the connective tissue/fat from the vessel by grabbing on both sides and pulling gently and equally in opposite directions. Be careful never to touch the wall of the vessel during this and subsequent procedures. As the connective tissue loosens, pull it away from the vessel with one pair of fine forceps and use the fine microscissors to cut close to the vessel wall. Repeat while moving along the length of the vessel. Take care to avoid overstretching the vessel, as this could damage the smooth muscle cells and prevent contractions during the experiment. It is not necessary to clean the vessel completely at this point and in fact it is not advisable because the lumen will be collapsed, difficult to see, and easy to nick inadvertently. Only the two ends of the vessel need to

236

Jorge A. Castorena-Gonzalez et al.

Fig. 2 Photo of dissection chamber. The chamber opening is 5.3 cm in diameter

be cleaned well and they will be damaged anyway during cannulation. 4. Once the vessel is cleaned, it will float to the surface if a sufficient number of fat cells remain attached; otherwise it will simply settle to the bottom of the chamber. Either way, draw it up into a Pasteur pipette and transfer it to the cannulation/bath chamber (Fig. 3) with mounted cannulation pipettes (Krebs-­ BSA in all). 5. A piece of 40-μm stainless steel wire, 5–8 mm in length can be used to weight the vessel down and keep it from floating; weighting is also helpful during the next step even if no fat cells remain attached. 3.5  Adjust the Stage and Illumination

1. Position the stage assembly with pipettes/holders at a 45° angle relative to the operator (Fig. 4). Position the two fiber optic light guides as close as possible to the vessel/chamber to provide maximum illumination without interference from finger tips and surgical instruments; adjust the lenses appropriately for the distance used (typically 1–2 in.). Position the moveable arm rests so that the operator can reach each side of the vessel while keeping one cannulating pipette/holder/ manipulator pointed at one of the operator’s shoulders (i.e., the large dissecting microscope objective will not allow the vessel to be oriented perfectly perpendicular to the operator). 2. During cannulation and cleaning, magnifications of 50–64× provide good resolution of the vessel ends while retaining

Ex Vivo Methods to Study Mouse Lymphatic Contractile Function

237

Fig. 3 Photo of cannulation/bath chamber with water jacket and connecting tubing. Outer chamber edge is 6.5 cm on each side. A glass slide (0.17 mm thickness) is glued to the chamber bottom

Fig. 4 Photo of stage plate with pair of coarse micromanipulators (Narishige model MN-189) attached underneath, each supporting a V-tracks with a pipette holder for a cannulation pipette. Stage is raised onto a moveable platform so that the manipulators clear the surface of the table. Arm rests are positioned for right-­handed operator to cannulate a lymphatic vessel in the cannulation chamber

238

Jorge A. Castorena-Gonzalez et al.

some depth of field. Good lighting will aid in minimizing eye strain. Adjust the height of operator’s chair relative to the microscope eyepieces and chamber/stage; it is important to maintain good posture (e.g., a straight back) during the next few steps as they may require 1 h or more of intense concentration. After extensive practice cannulation should take less than 10 min. Cleaning will require an additional 30–40 min. Vessel viability is in general inversely proportional to the amount of time required in Subheadings 3.6–3.8. 3.6  Cannulate the First Vessel End

1. With one end of the vessel weighted down to the bottom of the perfusion chamber, lower one cannulation pipette close to, but not touching, the chamber bottom; the cut ends of the suture tied around the pipette tip will likely touch the bottom. The other pipette can be lowered to an intermediate depth. It is essential that the stopcock-syringe be opened to atmosphere prior to cannulation. The syringe should be turned and/or elevated so that the open port is 2–3 cm H2O above the level of the chamber, creating a slight positive pressure head prior to cannulation. 2. Grab the free end of the vessel (preferably the input end if orientation has been retained from Subheading 3.2) using the two 45° angled forceps and carefully pull in opposite directions on the crimped vessel end until the lumen appears to open partially. Slide that end onto the cannulating pipette, being careful not to force it; the slight positive pressure in the pipette will suddenly fill and expand the vessel if the pipette tip cleanly enters the lumen. Mouse vessels are covered in sticky, dense connective tissue that tightens like a Chinese finger trap [15] when pulled with too much force, so pull gently to open the vessel end while simultaneously attempting to slide the open end onto the pipette. If unsuccessful, remove the vessel from the pipette tip and rotate it, pulling from a different angle; repeat if necessary to partially open the lumen. Grabbing the vessel wall rather than the connective tissue around the wall, usually is required for a successful cannulation. 3. Expansion of the vessel for only part of its length upon approach of the open end to the cannulation pipette indicates that the outflow end is being cannulated, as any valves that are present will prevent retrograde flow from filling up the entire vessel. Either end can be cannulated first but cannulating the inflow end will result in the filling of the entire vessel and make subsequent cannulation of the outflow end easier. 4. Once the lumen is open and the vessel is filled or partly filled, loosen the suture tie, slide it over the end of the vessel and retighten it. Readjusting the tie can usually be performed with one forceps, while maintaining a hold on the vessel end with

Ex Vivo Methods to Study Mouse Lymphatic Contractile Function

239

the other forceps to prevent the internal pressure from pushing it off of the cannulation pipette. 3.7  Cannulate the Second Vessel End

1. Using the z-axis control of the manipulator, raise the vessel off the chamber bottom 1–2 mm, adjust the z-axis control of the other manipulator so that the both pipette tips are in focus, and repeat Subheading 3.6 with the other vessel end and other cannulation pipette.

3.8  Clean the Vessel

1. When both ends are cannulated, the lumen should be filled and distended over the entire length of the vessel, causing it to bow laterally between the cannulae. Use one manipulator (x-axis control) to move the pipettes apart in the axial direction and thereby eliminate slack from the vessel. A stretched vessel is easier to clean. 2. [This description is for a right-handed operator; reverse it for left-handed operator]. Starting at one end, grab the loose connective tissue on both sides of the vessel with the two 45° angled forceps and gently pull in equal and opposite directions, working around the circumference of the vessel. The tissue will slowly retract after it is released, so work in 200-μm segments. After the tissue and/or fat are loosened, pull the tissue away from the vessel (causing the vessel to bow slightly) with the left-­hand forceps while using the fine microscissors to cut along the left edge of the vessel wall. The scissors will cross over the top of the vessel in order to cut along the left edge. The scissors must be very sharp and only the very tips should be used to cut. If necessary, anchor the scissors on the oval edge of the chamber (1–2 cm away) so that the scissor tips remain stationary and use the forceps to lift the connective tissue and vessel up to the scissors. Do not touch the wall with either the forceps or scissors. The vessel may occasionally contract during the cannulation procedure even with the vessel at room temperature. Contractions during the cleaning step can be used to one’s advantage if the connective tissue is held to one side during the contraction and a scissor cut is made close to the wall during the peak of the contraction. 3. Repeat this cleaning routine on the next 200-μm length segment and then move slowly toward the other cannulating pipette. If tissue remains on the right edge of the vessel, turn the entire stage assembly with cannulating pipettes 180° and repeat the procedure on the opposite vessel edge. 4. Repeat steps 1–3 two or three times, as necessary, to thoroughly clean the vessel. Depending on the design of the experiment, a clearly visible outer wall may be needed for diameter tracking only along a short section of the vessel, but that option also dictates that there should be no damaged areas of the wall and that

240

Jorge A. Castorena-Gonzalez et al.

the contraction amplitude of the vessel (when subsequently warmed) be consistent along the entire vessel length. Cleaning the entire length of the vessel provides options later for diameter tracking at various sites and/or assessment or exclusion of damaged areas. For difficulties with vessels twisting (see Note 16). For dealing with side branches (see Note 17). 5. Make sure that both pipette tips are still clear after the cleaning procedure by jiggling each syringe slightly and observing the vessel; the smallest movement of either syringe should cause the vessel to move as the luminal pressure changes slightly. For help with detecting partial cannulations (see Note 18). 6. Finally, tighten both suture knots as needed. If the knots loosen later during the experiment (this will depend on the taper of the pipettes and the type of suture used), two sets of ties may be needed on each vessel end. A photomontage of a cannulated popliteal lymphatic from an SVF129 mouse is shown in Fig. 1b. 3.9  Mount the Stage to the Inverted Microscope

1. Transfer the pipette/perfusion chamber/micromanipulator assembly to the inverted microscope and secure it to the stage. 2. Disconnect the PE or Silastic tubing from the 16-G blunt needle hub and reconnect it to the tubing from the pressure reservoir set at 3 cm H2O. It is imperative that the stopcocks be opened to the atmosphere when disconnecting and connecting the tubing; otherwise damaging levels of negative and positive intraluminal pressure, respectively, will be imposed on the vessel. 3. Connect the perfusion tube and suction tube to the chamber and begin bath perfusion with temperature control (see Note 13). A typical vessel will begin to show slow, large-amplitude (>50% passive diameter) contractions within 10 min. These will accelerate in frequency and decline somewhat in amplitude as the bath temperature stabilizes at 37 °C. 4. Allow 30–60 min for full equilibration and stabilization of contractions. Contraction frequency is very sensitive to temperature. Vessels will contract between 28 °C and 40 °C but sustained temperatures >38 °C often lead to declining function.

3.10  Conduct the Experimental Protocol

1. Numerous types of protocols are possible [14], including testing the effects of (a) changes in intraluminal pressures, either inflow, outflow or both [16], (b) changes in imposed flow by raising inflow pressure while simultaneously lowering outflow pressure [5] within certain ranges (negative outflow pressures will close the vessel), (c) valve function tests [17] as described in Chapter 7, and (d) bath application of endothelial-dependent [9] or independent [18] agonists/inhibitors.

Ex Vivo Methods to Study Mouse Lymphatic Contractile Function

3.11  Disassembly and Cleaning

241

1. Pipettes can be reused if they are cleaned well after each experiment: remove the pipette from its holder, wash the outside with distilled water, connect the back end to a vacuum tube (with water trap), dip the tip in filtered distilled H2O for 20 s, then HPLC-grade acetone for 20 s and allow it to air-dry for 20 s, all while connected to suction. 2. Store in a covered dish. 3. Clean the pipette holders (disassemble if necessary) and the perfusion chamber after each experiment with 10% neutral detergent followed by a distilled H2O rinse, a rinse with 30% pure ETOH in distilled H2O, and another distilled H2O rinse. 4. The dissection chamber should be drained and rinsed several times with boiling-hot distilled H2O. 5. Hot distilled H2O should also be flushed liberally through the tubing in the bath perfusion pump and then allowed to air-dry by running the pump for a few minutes. 6. Wash all surgical instruments carefully with distilled water and then 70% ETOH and store them in a protected case.

4  Notes 1. If unpurified BSA (e.g., Sigma #A2058) is used it should be extensively dialyzed and lyophilized or normal vessel reactivity may be affected. As an alternative we use a more expensive BSA from US Biochemicals (#10856, Cleveland, OH), which is purified and lyophilized. Krebs + 0.5% BSA is used to fill the dissection dish, the cannulating pipettes and holders and (initially) the perfusion chamber. Albumin-free Krebs is used to perfuse the chamber during the actual experiment. 2. The highest quality, hardened stainless steel instruments are preferable because they can withstand daily use and maintain fine points. We recommend two pairs of forceps and two pairs of scissors for each investigator. One “coarse” pair of each is sharpened to ~50 μm tips and used for gross dissection (holding skin, removing the vessel, manipulating pins); the other “fine” pair of forceps and scissors is sharpened to