Ultrastructure of Mast Cells and Basophils

U U U U U U U U U U U U U U U U U U U U U U U U U U U Chemical Immunology and Allergy Vol. 85 Series Editors Johanne...

Author: Dvorak A.M.

58 downloads 639 Views 9MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Chemical Immunology and Allergy Vol. 85

Series Editors

Johannes Ring Munich Luciano Adorini Milan Claudia Berek Berlin Kurt Blaser Davos Monique Capron Lille Judah A. Denburg Hamilton Stephen T. Holgate Southampton Gianni Marone Napoli Hirohisa Saito Tokyo

Ann M. Dvorak U U U U U U U U U U U U U U U U U U U U U U U U U U U

Ultrastructure of Mast Cells and Basophils

185 ﬁgures, 4 in color, 2005

Basel • Freiburg • Paris • London • New York • Bangalore • Bangkok • Singapore • Tokyo • Sydney

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Chemical Immunology and Allergy Formerly published as ‘Progress in Allergy’ (Founded 1939) continued 1990-2002 as ‘Chemical Immunology’ Edited by Paul Kallos 1939-1988, Byron H. Waksman 1962–2002 U U U U U U U U U U U U U U U U U U U U U U U U U U U

Ann M. Dvorak, MD Department of Pathology - East Campus Beth Israel Deaconess Medical Center 330 Brookline Avenue Boston, MA 02215 (USA) Tel. +1 617 667 3692, Fax +1 617 667 2943, E-Mail [email protected]

Bibliographic Indices. This publication is listed in bibliographic services, including Current Contents® and Index Medicus. Drug Dosage. The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant ﬂow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher. © Copyright 2005 by S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) www.karger.com Printed in Switzerland on acid-free paper by Reinhardt Druck, Basel ISSN 1660–2242 ISBN 3–8055–7864–4

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 1 Introduction

IX X

..............................................................................

1

Chapter 2 Ultrastructural Analysis Is Necessary and Sufficient for Identification of Basophils and Mast Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Development of Human Mast Cells and Basophils de novo in vitro . . . . . . . . . . . 2.2.1. Human Basophils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2. Human Mast Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Circulating Mast Cell Precursor in Embryonic Mice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Mouse Basophils Are Not Lymphocytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5. Mouse Basophils Are Not Neutrophils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6. Ultrastructural Identiﬁcation of Monkey Basophils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7. Ultrastructural Veriﬁcation of the Identity of Mast Cells or Basophils in Cell Samples Obtained from Human Organs and Fluids . . . . . . . . . . . . . . . . . . . . . . . .

10 10 11 11 17 39 42 55 56

Chapter 3 Cyclooxygenase, a Key Enzyme Family for Production of Prostaglandins, Is Present in Human Mast Cell Lipid Bodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Lipid Bodies Are Distinct Non-Membrane-Bound Organelles . . . . . . . . . . . . . . . . . . 3.3. Ultrastructural Autoradiographs Localize Arachidonic Acid to Human Mast Cell Lipid Bodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Immunogold Ultrastructural Localization of Prostaglandin Endoperoxide Synthase (Cyclooxygenase) in Human Mast Cell Lipid Bodies . . . . . . . . . . . . . . . . . .

62

68 68 68 69 70

Chapter 4 Subcellular Localization of the Cytokines, Basic Fibroblast Growth Factor and Tumor Necrosis Factor-␣ in Mast Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Cytokines in Lipid Bodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1. Tumor Necrosis Factor-␣ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.2. Basic Fibroblast Growth Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Cytokines in Granules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.1. Basic Fibroblast Growth Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.2. Tumor Necrosis Factor-␣ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

72 72 73 73 77 77 78 82

Chapter 5 Immunogold Ultrastructural Techniques Identify Subcellular Sites of Chymase, Charcot-Leyden Crystal Protein, and Histamine in Basophils and Mast Cells . . . . . . . 5.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Chymase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. Charcot-Leyden Crystal Protein . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4. Histamine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

89 89 89 91 95

Chapter 6 Ultrastructural Enzyme-Affinity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells . . . . . . . . . . . . 6.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2. Histamine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2.1. Diamine Oxidase-Gold (DAO-G) Method to Image Histamine . . . . . . 6.2.2. DAO-G Labels Human Mast Cell Secretory Granules . . . . . . . . . . . . . . . . . . . 6.2.3. Enzyme-Afﬁnity Cytochemical Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2.4. Comment regarding a New Ultrastructural Method to Detect Histamine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3. Heparin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.1. Ribonuclease-Gold (R-G) Method to Image Heparin . . . . . . . . . . . . . . 6.3.2. Inhibitor-Gold Cytochemical Controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.3. RNase-Gold-Labeled Sites in Human Mast Cells . . . . . . . . . . . . . . . . . . . . . . . . 6.3.4. Experiments to Determine the Basis of R-G Staining of HMC Granules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.5. Effect of General Inhibitors, Blockers, and Enzyme Digestions on R-G Labeling of HMC Granules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.6. Effect of Speciﬁc Blocking and Enzyme Digestions on R-G Labeling of HMC Granules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.7. R-G Staining of Agar Blocks Containing Heparin, RNA or Histamine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.8. R-G Labels Chondroitin Sulfate (CS) in Guinea Pig Basophil Granules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.9. R-G Labels CS in Human Basophil Granules . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.10. R-G Labels Proteoglycans in Rabbit Basophil Granules . . . . . . . . . . . . . . . . . 6.3.11. R-G Labels Heparin in Rat Peritoneal Mast Cell Granules . . . . . . . . . . . . . 6.3.12. R-G Labels CS in Cultured, Bone-Marrow-Derived Mouse Mast Cell Granules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Contents

98 98 99 99 100 101 103 114 115 116 118 118 119 120 124 125 126 128 128 130

VI

Chapter 7 Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2. Cutaneous Basophil Hypersensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3. Mast Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.1. Human Mast Cells in vivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.2. Human Mast Cells ex vivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.3. Mouse Mast Cells in vivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3.4. Rat Mast Cells ex vivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4. Basophils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4.1. Piecemeal Degranulation of Human Basophils ex vivo Is Stimulated by Bacterial Products, Tumor-Promoting Agents, and Cytokines . . . . . . . 7.4.2. Ultrastructural Morphology of Secretion by FMLP- and TPA-Stimulated Human Basophils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4.3. Stimulated Human Basophils Show Differences in Ultrastructural Phenotypes, Granules and Vesicles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4.4. Vesicular Transport of Charcot-Leyden Crystal Protein in Piecemeal Degranulation of Human Basophils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4.5. Vesicular Transport of Histamine in Piecemeal Degranulation of Human Basophils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4.6. Comparative Analysis of Vesicle Transport of CLC Protein and Histamine in Stimulated Human Basophils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4.7. Comment regarding Basophil Secretion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 8 Mast Cell-Derived Mediators of Enhanced Microvascular Permeability, Vascular Permeability Factor/Vascular Endothelial Growth Factor, Histamine, and Serotonin, Cause Leakage of Macromolecules through a New Endothelial Cell Permeability Organelle, the Vesiculo-Vacuolar Organelle . . . . . . . . . . 8.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2. What Are Vesiculo-Vacuolar Organelles (VVOs)? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3. Serial Section Analysis Demonstrates the Transendothelial Cell Pathway Provided by VVOs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.4. Electron-Dense Ultrastructural Tracer Analysis Demonstrates the Transendothelial Cell Pathway Provided by VVOs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.5. VVOs in Tumor Vessels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.6. VVOs in Inﬂammatory Eye Disease Vessels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.7. Permeability through VVOs in Normal Vessels Is Induced by VPF/VEGF, Histamine, and Serotonin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.8. Stomatal Diameters in Venule VVOs and Capillary Caveolae Respond Differently when Exposed to Permeabilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.9. Ultrastructural Localization of Receptors and Ligands to VVOs . . . . . . . . . . . . . . . . 8.9.1. Histamine, a Potent Permeability-Producing Ligand, Binds to VVOs . . . 8.9.2. VPF/VEGF, a More Potent Permeability-Producing Ligand, Binds to VVOs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.9.3. VPF/VEGFR-2, a High-Afﬁnity Receptor for VPF/VEGF, Is Present in VVOs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Contents

135 135 138 145 145 153 153 154 156 156 162 166 173 175 175 179

185 185 186 190 190 192 196 198 198 198 200 200 202

VII

Chapter 9 Degranulation and Recovery from Degranulation of Basophils and Mast Cells . . . . . 9.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2. Murine Mast Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2.1. Spontaneous Degranulation of Cultured Bone-Marrow-Derived Immature Mast Cells from X-Linked-Immunodeﬁcient Mice . . . . . . . . . . . 9.2.2. Anaphylactic Degranulation of Beige Rat Mast Cells . . . . . . . . . . . . . . . . . . . . 9.2.3. Rat Peritoneal Mast Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3. Human Basophils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3.1. Anaphylactic Degranulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3.2. Recovery from Anaphylactic Degranulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4. Human Mast Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4.1. Anaphylactic Degranulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4.2. Recovery from Degranulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 10 Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.2. Regressive Ethylene Diamine Tetraacetic Acid Staining of HMCs Reveals Granule- and Lipid Body-Associated Ribonuclear Proteins . . . . . . . . . . . . . . . . . . . . . 10.3. Ribonuclease-Gold Stains Heparin in HMC Granules and Labels RNA in HMC Lipid Bodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.4. Ultrastructural Autoradiography of Uridine Incorporation Labels RNA in Human Mast Cell Granules and Lipid Bodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.5. Ultrastructural Immunogold Imaging of Uridine in Human Mast Cell Granule and Lipid Body RNA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10.6. Ultrastructural Cytochemical, Immunocytochemical and in situ Hybridization Methods Detect mRNA in Human Mast Cell Granules and Lipid Bodies . . . . . . 10.7. Ultrastructural Immunogold Cytochemistry of HMC with Speciﬁc Autoimmune Human Sera Detects RNA-Associated Protein Antigens of Different RNA Species in and Associated with Lipid Bodies and Granules . . . . 10.8. Comment regarding Human Mast Cell Granules and Lipid Bodies Containing RNA Machinery Important for Protein Synthesis . . . . . . . . . . . . . . . . . . 10.9. RNA: Site-Speciﬁc Synthesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 11 Concluding Statement

...................................................................

205 205 205 205 206 206 206 206 209 219 219 235

252 252 257 262 262 275 279

298 303 309

316

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319 Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345

Contents

VIII

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Preface Basophils and mast cells are similar but unique secretory cells with a welldocumented role in immediate hypersensitivity reactions. The presence of these cells in various cell-mediated hypersensitivity reactions, in tissues of multiple diseases, and as a component of the host reaction to injury and repair in numerous circumstances is also well documented. Release of stored and newly generated mediators of inﬂammation from basophils and mast cells contributes to the cascade of pathogenetic events in circumstances under which these release reactions occur. In this book, I review the ultrastructural studies of basophils and mast cells that we have done since our earlier monograph appeared in 1991 [7]. These studies include the use of ultrastructural rules that allow identiﬁcation of basophils and mast cells in new circumstances. We establish the presence of secretory granules and lipid bodies in these cells as different, important organelles and use, or develop, a number of ultrastructural imaging methods, to deﬁne the subcellular locations of chymase, Charcot-Leyden crystal protein, histamine, and heparin as well as materials important to arachidonate and RNA metabolism, and to cytokine biology in basophils and mast cells. These studies have allowed us to advance knowledge regarding the secretory mechanics of basophil and mast cells, and their recovery possibilities as well as to establish the transendothelial route of passage of macromolecules stimulated by permeability mediators released from basophils and mast cells. I acknowledge the continuous encouragement and helpful advice of my husband and collaborator Dr. Harold F. Dvorak. Also, many of these studies could not have been done without the excellent assistance of my colleagues Drs. Stephen Galli, Teruko Ishizaka and Lawrence Lictenstein, and the numerous members of their laboratories. A large number of colleagues in many other distinguished laboratories have been a ready source of material for some studies. Manuscript processing was provided by Tracey Sciuto and expert technical assistance by Ellen Morgan, Rita Monahan-Earley, Kathryn Pyne, Patricia Fox and Linda Letourneau. All of these studies have been made possible by the National Institutes of Health grants CA 15136, CA 19141, CA/AI 28834, AI 33372, and AI 44066, representing continuous support of the research efforts in basophil and mast cell biology in my laboratory. The publication of this monograph was partially supported by Lillian Gong. Ann M. Dvorak, Boston, Mass., USA

Preface

IX

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Abbreviations

AND AND-I/II bFGF BSA CBC(s) CBH CD CDB(s) CF CG CLC(s) CS DAO-G DH DNA DNase EDTA FE FMET FMLP GAGs HB(s) HCl HGMCs HLMC(s) HMC(s) hnRNA HRP HSMC(s) HSYMCs IBD IL ISH LC LFA LFB MA

anaphylactic degranulation anaphylactic degranulating basophil I/II basic ﬁbroblast growth factor bovine serum albumin cord blood cell(s) cutaneous basophil hypersensitivity Crohn’s disease completely degranulated basophil(s) cationized/cationic ferritin cationized gold Charcot-Leyden crystal(s) chondroitin sulfate diamine oxidase-gold delayed hypersensitivity deoxyribonucleic acid deoxyribonuclease ethylene diamine tetraacetic acid anionic ferritin formyl-methionyl formyl-methionyl-leucyl-phenylalanine glycosaminoglycans human basophil(s) hydrochloric acid human gut/gastrointestinal mast cells human lung mast cell(s) human mast cell(s) heterogeneous nuclear RNA horseradish peroxidase human skin mast cell(s) human synovial mast cells inﬂammatory bowel disease interleukin in situ hybridization lead citrate lymphocyte factor A lymphocyte factor B macrophage

Abbreviations

X

MCGF MCP-1 MFF MMC(s) NM PBS PLL PMD PMD-I/II RBs RB-I/II RER R-G rhIL RNP(s) rHRF rhSCF RMC(s) rmMCGF RNA RNase SCF snRNP TBS-BSA TPA TNF-␣ TV(s) UA UC VEGF VG/TV VPF VPF/VEGF VVO(s) Xid

mast cell growth factor monocyte chemotactic protein-1 murine ﬁbroblast factor(s) mouse mast cell(s) neutrophilic myelocyte phosphate-buffered saline poly-L-lysine piecemeal degranulation piecemeal degranulating basophil I/II recovering basophils recovering basophil I/II rough endoplasmic reticulum ribonuclease-gold recombinant human interleukin ribonuclear protein(s) recombinant histamine-releasing factor recombinant human stem cell factor rat mast cell(s) recombinant murine MCGF ribonucleic acid ribonuclease stem cell factor small nuclear ribonuclear protein Tris(hydroxymethyl)aminomethane-bovine serum albumin tetradecanoyl phorbol acetate tumor necrosis factor-␣ total vesicle(s) uranyl acetate ulcerative colitis vascular endothelial growth factor gold-loaded vesicles/total vesicles vascular permeability factor vascular permeability factor/vascular endothelial growth factor vesiculo-vacuolar organelle(s) X-linked immunodeﬁcient

Abbreviations

XI

Chapter 1

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Introduction

Basophils (ﬁg. 1) and mast cells (ﬁg. 2) are two unique but related cells which were both originally described by Ehrlich [1, 2]. Our interest in these fascinating cells began in the late 1960s when we found large numbers of basophils in skin lesions of guinea pigs sensitized to antigen emulsiﬁed in incomplete Freund’s adjuvant and studied by electron microscopy within 1–2 weeks after skin testing – reactions we called cutaneous basophil hypersensitivity (CBH) [3, 4]. Pursuant to these studies, our interest in basophil and mast cell biology has continued unabated for 30 years. In early studies (1970–1990) we used ultrastructural analyses to identify the metachromatic granule-containing (ﬁg. 3A) cell populations as either mast cells (ﬁg. 2) or basophils (ﬁg. 1) in a number of species and circumstances, identiﬁed an important new organelle system, lipid bodies (ﬁg. 3B), in human mast cells [5], deﬁned a new mode of secretion from basophils and mast cells which we called piecemeal degranulation (PMD) (ﬁg. 4, 5) [6], and described ultrastructural features of anaphylactic degranulation (AND) (ﬁg. 6, 7) and, recovery therefrom, of basophils and mast cells. These studies have all been reviewed in an earlier monograph [7]. In this review, I present ultrastructural studies primarily covering work done in our laboratory from 1990 to the present. In this interval we used ultrastructural rules for basophil and mast cell identiﬁcation to facilitate the ﬁrst descriptions of the de novo development of human mast cells and basophils in vitro [8, 9] and the ﬁrst description of a circulating embryonic mouse mast

cell precursor [10], to clarify the identiﬁcation of mouse basophils in unusual circumstances [11], to verify the presence of basophils or mast cells in a wide variety of puriﬁed human cell populations, and to provide the ﬁrst description of basophils in monkeys [12]. Additionally, we concentrated on subcellular localization of products in basophils and mast cells in order to provide proof of principle to our hypothesis that PMD is effected by vesicular transport of granule products [6]. The model cells for these studies include human mast cells (HMCs), human basophils (HBs), rat mast cells (RMCs), and mouse mast cells (MMCs) studied either in vivo or ex vivo. The products localized include chymase, Charcot-Leyden crystal (CLC) protein, histamine, heparin, tumor necrosis factor-␣ (TNF-␣) and basic ﬁbroblast growth factor (bFGF). Other areas, which we review here, are as follows: (1) subcellular localization of cytokines, and of enzymes important to eicosanoid production, to lipid bodies in HMCs; (2) the functional and anatomic description of a newly recognized endothelial cell organelle, which we termed the vesiculo-vacuolar organelle (VVO) [13–15], which is responsible for the enhanced vascular permeability induced by the stimulated release of mast cell and basophil mediators stored in cytoplasmic granules, and (3) the implication of HMC granules and lipid bodies in ribonucleic acid (RNA) metabolism [16, 17], which may facilitate renewal of the contents of granules and lipid bodies in the recovery phase following PMD or AND.

Ultrastructure of Mast Cells and Basophils

2

Fig. 1. Human basophil in the peripheral blood shows a polylobed nucleus (N), short irregular surface processes, monoparticulate electron-dense glycogen particles and large particle-ﬁlled secretory granules. Cationized ferritin stains the plasma membrane but not the granule membranes. Osmium potassium ferrocyanide, cationized ferritin. !13,000. [From 53, with permission.]

Introduction

3

Fig. 2. Human mast cell (isolated from a lung specimen and maintained in culture for 1 month) shows numerous electron-dense secretory granules with variable numbers of scrolls within them. A monolobed nucleus and Golgi area (G) are seen. The surface is adorned with narrow surface folds. Osmium collidine uranyl en bloc. !14,600. [From 28, with permission.]

Ultrastructure of Mast Cells and Basophils

4

Fig. 3. Membrane-bound, homogeneous, moderately electron-dense secretory granule (A) and non-membrane-bound lipid body (B) in the cytoplasm of a human mucosal mast cell from a rectal biopsy specimen. Osmium collidine uranyl en bloc. Bar = 0.1 ␮m. [From 7, with permission.]

Introduction

5

Fig. 4. Human peripheral blood basophil shows piecemeal degranulation characterized by empty, electron-lucent, membrane-bound secretory granules. Several granules contain particles. The cytoplasm also contains many empty vesicles and one small granule (arrow). N = Nucleus. Osmium potassium ferrocyanide. Bar = 0.8 ␮m. [From 293, with permission.]

Ultrastructure of Mast Cells and Basophils

6

Fig. 5. Human gut mast cell in situ from continent ileal pouch biopsy of a patient with Crohn’s disease shows PMD. Note that non-fused granule chambers are partly empty. Osmium collidine uranyl en bloc. !14,000. [From 55, with permission.]

Introduction

7

Fig. 6. Human peripheral blood basophil shows AND 30 min after exposure to antigen. Secretory granules are being extruded to the exterior at separate membrane openings. Arrows indicate a narrow cationized ferritin-ﬁlled neck connecting to one granule (bottom). Other arrows indicate an interconnected granule chain that is also being extruded (top). N = Nucleus. Bar = 1.2 ␮m. [From 405, with permission.]

Ultrastructure of Mast Cells and Basophils

8

Fig. 7. Human skin mast cell ex vivo, 3 min after stimulation with anti-IgE, shows AND characterized by extrusion of swollen, membrane-free granules through the cell membrane at four separate points. A few unaltered granules remain in the cytoplasm. Osmium collidine uranyl en bloc. !17,000. [From 35, with permission.]

Introduction

9

Chapter 2

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Ultrastructural Analysis Is Necessary and Sufficient for Identification of Basophils and Mast Cells

2.1. Overview Ultrastructural analysis is necessary and sufﬁcient for correct identiﬁcation of basophils and mast cells. Although these cells are closely related they are sufﬁciently different to be assigned to one of these two lineages in most species. We have previously reviewed the ultrastructural criteria for doing so in three species: guinea pig, mouse and human [7]. These criteria have been useful in a number of new circumstances. These include the identiﬁcation of HBs and HMCs which developed de novo in vitro [8, 9, 18–29], the identiﬁcation of circulating, granule-containing, MMC precursors in embryonic mice [10], the identiﬁcation of mouse basophils comprising the cells in a non-B-, non-Tcell population derived from spleen or bone marrow [11, 30, 31], and those interpreted as neutrophils in the published literature [32, 33]. Basophils were identiﬁed for the ﬁrst time in a new species, the monkey [12], and the ultrastructural criteria were of use to verify that mast cells or basophils were present in cellular preparations obtained from human organs and ﬂuids [34–39]. Knowledge of various pitfalls in the identiﬁcation of basophils and mast cells, reviewed earlier [7], were also of use when ultrastructural rules were applied in the studies reviewed here. These include recognition that immature basophils, or basophil myelocytes, can be confused with mast cells [7]. Other cells that have also been confused with mast cells include immature eosinophils, or eosinophilic myleocytes [19, 40, 41], and macrophages [42].

2.2. Development of Human Mast Cells and Basophils de novo in vitro Sophisticated culture systems and newly identiﬁed speciﬁc growth factors have led to important new studies of the development of HBs and HMCs in vitro [8, 9, 18–25, 27, 43–46]. In 1983, a program was initiated that made use of conditioned media and human umbilical cord blood agranular mononuclear cell cultures [8, 43, 44] in an attempt to identify the necessary conditions for de novo HMC development in vitro. These studies were based on knowledge acquired from murine systems in which mast cells developed in quantity in conditioned media now known to contain interleukin (IL)-3 [7, 47–50]. Selective growth of cells containing metachromatic granules occurred in suspension cultures of cord blood cells (CBCs) that were supplemented with a fraction of the culture supernatant of phytohemagglutinin-stimulated, IL-2depleted human T cells [43]. These cells were identiﬁed ultrastructurally to be mature HBs [8]. Mature HMCs were ﬁrst identiﬁed, by electron microscopy, to arise from CBCs in cocultures with ﬁbroblasts [9, 20]. A soluble, ﬁbroblastderived factor, ultimately identiﬁed as the c-kit ligand [stem cell factor (SCF), steel factor, mast cell growth factor (MCGF)], was reported to support the development of large numbers of immature HMCs (deﬁned by ultrastructural analysis) from CBCs in suspension cultures, in 1993 [23, 27] – 10 years after the initiation of this culture program. Thus, the initial goal, to identify the necessary conditions for the de novo development of HMCs in vitro, was reached in 10 years. The studies for this purpose provided new information regarding both basophil and mast cell lineages in man. Ultrastructural aspects of these studies [8, 9, 18–29, 40–42, 51] are reviewed here. 2.2.1. Human Basophils

The metachromatic granule-containing cells that arose in 2- to 3-week cultures of CBCs, supplemented with a fraction of the culture supernatant of phytohemagglutinin-stimulated, IL-2-depleted human T cells [lymphocyte factor A (LFA)] [43], were mature cells resembling circulating peripheral blood basophils by ultrastructural criteria [8, 18]. Immunological studies showed that these cultured basophils had high-afﬁnity IgE receptors, contained histamine, and released both histamine and arachidonate products by an IgE-dependent mechanism [8]. Concomitant with stimulation, the mature, cultured HBs released entire granules by exocytosis [8], typical of AND – a form of regulated secretion expressed by HBs and HMCs [7, 52]. The ultrastructural features of the mature HBs developing in 2 weeks in this system were identical to those of peripheral blood basophils [53], with the exception that lipid bodies were more numerous [8] than they generally are in peripheral blood basophils. Thus,

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

11

mature HBs with typical polylobed nuclei and basophil particle-ﬁlled granules characterized these cells arising from mononuclear, agranular precursor cells that are present in CBCs [53]. No cells with the ultrastructural criteria necessary for the identiﬁcation of HMCs [7, 54, 55] were present. The studies of HBs arising from suspension cultures of CBCs were extended with detailed ultrastructural and cytochemical analyses [18]. These studies provided the ﬁrst opportunity to examine human basophilopoiesis in a non-leukemic setting [18, 53]. In these studies [18, 53], CBCs were cultured in LFA [18], in supernatants of cloned mouse lymphocytes [lymphocyte factor B (LFB)] [18, 25, 53] known to contain IL-3 [47, 56–58], or in LFB followed by LFA [18]. Cultures supplemented with LFA contained numerous mature basophils, which underwent a sequence of maturation similar in ultrastructure to that previously described in guinea pigs and mice [7, 59]. Substantial numbers of mature eosinophils were also present [18]. Eosinophils and their immature precursors exhibited peroxidatic activity in secretory structures (Golgi, rough endoplasmic reticular, and perinuclear cisternae) and cytoplasmic granules. Eosinophils undergoing necrosis and apoptosis released membrane-bound, peroxidase-positive granules into the medium. By contrast, mature basophils and immature basophilic myelocytes did not exhibit peroxidase in Golgi structures or in the cisterns of rough endoplasmic reticulum (RER), or in the cistern which surrounds the nucleus. Mature basophils frequently contained variable numbers of peroxidase-positive, particle-ﬁlled granules and small cytoplasmic vesicles. Taken together, these ﬁndings suggest that basophils do not synthesize peroxidase but acquire this enzyme by vesicular uptake of peroxidase, which is released by eosinophils [18]. In cultures supplemented with LFB, cells of the basophil lineage developed, but their maturation generally did not progress beyond the basophilic myelocyte stage [18]. These immature cells of the basophil lineage are morphologically distinctive [53] and, thus, are readily distinguished from mast cells by ultrastructural criteria, although they are generally similar to mast cells in size. Eosinophilic myelocytes are also large, distinctive cells that are ﬁlled with large, electron-dense, immature secretory granules. These immature eosinophils are readily distinguished from mast cells by ultrastructural criteria [41, 60]. Cytochemical studies revealed some eosinophilic myelocytes to be undergoing vesicular transport of the eosinophil peroxidase from the matrix compartment of their secondary granules – a process termed piecemeal degranulation (PMD) [51]. These observations were made in uninjured cells and indicate, for the ﬁrst time, that immature cells (eosinophilic myelocytes) have the capacity to undergo this type of secretion [51]. Eosinophilic myelocytes that arose in LFB-supplemented cultures also generally did not progress beyond this stage of development [18].

Ultrastructure of Mast Cells and Basophils

12

When cord blood cell (CBC) cultures were initiated with LFB and later changed to LFA, basophils (and eosinophils) developed fully into small, mature polylobed granulocytes [18]. This was accomplished by size reduction, nuclear segmentation and chromatin condensation, cytoplasmic granule maturation, and elimination of much of the cytoplasmic synthetic machinery. Thus, the two IL-3-containing media used here were permissive of the selective development of mature basophils (and eosinophils) when human lymphocyte-derived, IL-3-containing medium was used, but maturation of both lineages did not proceed beyond the myelocyte stage when murine lymphocyte-derived, IL-3containing medium was used [18]. The maturational arrest imposed by murine-derived LFB in each granulocyte lineage was reversed by human-derived LFA. In no instance did cultures supplemented with LFA or LFB, or with the sequential combination of LFB and LFA, give rise to mast cells. Nor did we ﬁnd cells with ultrastructural features intermediate between those of basophils and mast cells [61]. The effects of recombinant human IL (rhIL)-3 [62] or rhIL-5 [63] on hematopoiesis were explored using CBC suspension cultures [19, 40, 64, 65]. The results showed that CBCs cultured for 2 weeks in rhIL-3 developed into a mixture of basophilic, eosinophilic, and neutrophilic myelocytes and macrophages but that, by 3 weeks, most cells were eosinophilic myelocytes [40]. Similarly, CBCs cultured with rhIL-5 showed selective differentiation and proliferation of eosinophils; essentially all cells present by 3 weeks in rhIL-5-supplemented cultures were eosinophilic myelocytes [40]. The identiﬁcation of basophilic myelocytes in rhIL-3-supplemented CBC cultures and eosinophilic myelocytes in rhIL-3- or rhIL-5-supplemented CBC cultures was established by routine and cytochemical ultrastructural studies that detect eosinophil peroxidase in immature granules and synthetic organelles of eosinophilic myelocytes, an enzyme that is lacking in these structures in basophilic myelocytes [40]. In contrast to the effect of murine IL-3 on the development of MMCs from their agranular precursors [7, 31, 50], rhIL-3 did not induce the differentiation of mast cells in suspension cultures of human CBCs [40]. Conversely, neither recombinant murine IL-3 (rmIL-3) nor recombinant murine SCF was permissive for the survival (or development) of cells in the mouse basophil lineage in cultures of bone marrow cells [11, 30, 31, 66]. We extended the culture intervals of human CBCs in suspensions supplemented with rhIL-3, rhIL-4 or rhIL-5 to 5 weeks and prepared routine and cytochemical samples at several intervals for electron microscopy [19]. These studies showed that rhIL-5 primarily supported the eosinophil lineage (with fewer basophils), that rhIL-3 initially supported all granulocyte lineages but eventually supported the eosinophil lineage (with fewer basophils), and that rhIL-4 did not support the granulocyte lineages in these cultures. Mast cells

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

13

were absent in all cultures. By 5 weeks in cultures containing either rhIL-3 or rhIL-5, variable numbers of mature basophils and eosinophils developed [19]. Mature HBs displayed extensive evidence of PMD in rhIL-3-containing cultures. In addition, we noted extensive formation of intragranular CharcotLeyden crystals (CLCs) in mature basophils [19]. The next cultures within which we detected the development of HBs were suspension cultures of CBCs that were supplemented with cultured murine ﬁbroblast supernatants, with partially puriﬁed murine ﬁbroblast factor(s) (MFFs), or with recombinant human stem cell factor (rhSCF) [22–25]. (These systems primarily stimulated the development of HMCs.) The suspension cultures were examined with routine, peroxidase cytochemical and immunogold (to detect CLC protein) ultrastructural techniques. A number of time intervals were used for the sampling of multiple cord blood specimens, spanning 3–17 weeks of culture. Mature basophils, present in quantity in 3-week cultures, decreased in number and released their granule contents by PMD (ﬁg. 4) at later culture times [24]. Basophil counts as high as 52.5% at 3 weeks of culture in rhSCF dropped to 7% in 14-week cultures with rhSCF [24]. No basophilic myelocytes remained in cultures obtained over this time interval. Some basophils at earlier culture times (3 weeks) were noted to have decreased numbers of particle granules, others had increased numbers of homogeneously dense granules. Cytochemical preparations revealed some peroxidase in mature basophil particle-ﬁlled granules (but not in other granules or sparse synthetic structures), analogous to the uptake of eosinophil peroxidase by HBs that we detected in IL-3-supplemented suspension cultures of CBCs [18]. By 6 weeks of culture in rhSCF, images indicative of PMD had diminished in many mature HBs [24]. In addition to enhanced particulate contents in secretory granules, HBs displayed more CLCs in their granules (ﬁg. 8) [24]. Hexagonal and bipyramidal CLCs were also noted in the cytoplasm (ﬁg. 9) and nuclei of some of these cells [24]. A homogeneously dense subset of membrane-bound granules also was more evident (ﬁg. 10) [24]. We applied post-embedding ultrastructural immunogold technology to identify cells and subcellular sites that express the CLC protein in CBC suspension cultures supplemented with various sources of soluble c-kit ligand and sampled from 3 to 14 weeks [25]. Basophils (and eosinophils, which were present in smaller numbers) contained CLC protein, but mast cells did not. The subcellular sites in basophils that contained CLC protein included formed intragranular (ﬁg. 8), cytoplasmic (ﬁg. 9) and nuclear CLCs, cytoplasmic particle-ﬁlled and homogeneously dense granules (ﬁg. 10), cytoplasm, nucleus (ﬁg. 8, 10), plasma membrane, and Golgi area vesicles (ﬁg. 10). Individual basophil ultrastructural phenotypes, similar to those associated with stimulated release and recovery reactions [67], showed similar variations in gold-labeled

Ultrastructure of Mast Cells and Basophils

14

Fig. 8. A mature basophil (7-week suspension culture of CBCs in murine 3T3 ﬁbroblast culture supernatant), prepared with immunogold to detect CLC protein, shows gold label, which is concentrated over an intragranular CLC (closed arrowhead) and granule particles (open arrowhead) but not over intragranular, dense concentric membranes. Sparse label is present in a nearly empty granule, which shows granule membrane label. Focal areas of diffuse cytoplasmic and nuclear label are also evident. Background extracellular label is absent. !43,500. [From 25, with permission.]

subcellular compartments [25]. Macrophages also were labeled for CLC protein within endocytotic-lysosomal structures; neutrophilic myelocytes did not contain CLC protein. Thus, the combined ultrastructural morphology and immunogold phenotyping for CLC protein in c-kit ligand-supplemented cultures allows accurate lineage assignment of the developing cells [25]. Suspension cultures of human umbilical cord blood mononuclear cells supplemented with c-kit ligand-containing additives, in addition to basophils, gave rise to a mixture of cells belonging to several lineages [22–26]. Among those cells that developed in quantity were immature mast cells, neutrophilic myelocytes (ﬁg. 11, 12) and macrophages (ﬁg. 11). Variable numbers of eosinophils [41] and small numbers of megakaryocytes and endothelial cells were also present. Speciﬁc identiﬁcation of endothelial cells in these cultures was made possible by the presence of endothelial cell-speciﬁc, elongated, tubule-ﬁlled,

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

15

Fig. 9. Another mature basophil obtained and prepared similarly to ﬁgure 8, shows partially empty, particle-containing granules that are not labeled, and numerous heavily labeled intragranular CLCs in ﬁve granules. A large, elongated, heavily labeled CLC is free in the cytoplasm, traversing nearly the entire width of this cell. Note the reduced diffuse nuclear and cytoplasmic label in this cell with large numbers of CLCs. Background label is minimal. !25,000. [From 25, with permission.]

Ultrastructure of Mast Cells and Basophils

16

membrane-bound Weibel-Palade bodies in their cytoplasm [22]. Neutrophilic myelocytes were identiﬁed by their ultrastructural features in routine preparations (ﬁg. 11, 12A) as well as by the presence of myeloperoxidase-positive azurophilic granules and synthetic structures (ﬁg. 12B). Unlike basophils and eosinophils, which became completely mature in these cultures, neutrophils did not develop peroxidase-negative secondary (speciﬁc) granules or polylobed nuclei. Rather, they persisted as immature myelocytes or underwent apoptosis [26]. 2.2.2. Human Mast Cells

Cultures of isolated, partially pure, mature human lung mast cells (HLMCs) were established with mouse 3T3 embryonic skin ﬁbroblasts, or alone, to determine the effect of 3T3 ﬁbroblasts (or lung stromal ﬁbroblasts) on the maintenance of ultrastructural morphology of mature HMLCs for 1 month in vitro (ﬁg. 2) [28]. In both cases, mast cell viability was maintained, and mast cells retained the ultrastructural features of their in vivo counterparts in the lung, i.e., scroll granules predominated and numerous lipid bodies were present in the cytoplasm [28]. This system conﬁrms and extends those reported by LeviSchaffer et al. [68]. We particularly noted that the mature, fully granulated HLMCs, cocultured for 1 month with 3T3 ﬁbroblasts (or with lung stromal ﬁbroblasts), showed no evidence of PMD (or AND), and cytoplasmic vesicles, although present, were not prominent features of these cells. Thus, the electron microscopic evidence suggests that unstimulated HLMCs do not secrete mediators from their granule stores in this culture system [28] – a ﬁnding that is supported by the full complement of preformed mediators in cells and the lack of spontaneous release of preformed mediators up to 13 days in culture [68]. Human umbilical CBCs were cultured with mouse embryonic skin 3T3 ﬁbroblasts to search for a culture system that would allow HMCs to develop in vitro [9, 20, 21, 42]. Between 7 and 8 weeks, metachromatic granule-containing mononuclear cells were seen in these cultures by light microscopy, and their numbers increased up to 3 months [9]. We examined this coculture system at 3 months by electron microscopy and deﬁnitively identiﬁed the presence of mature, fully granulated HMCs in them (ﬁg. 13) [9, 20]. These cells ranged in size from 7 to 12 ␮m, and variable numbers of full, mature secretory granules were present. Other cytoplasmic organelles included elongated, tubular mitochondria and free ribosomes. The secretory granules contained regular crystalline arrays of several periodicities identical to the prevalent granule pattern of human skin mast cells (HSMCs) in vivo. Also, granules contained spirals, scrolls, particles, homogeneously dense areas, and mixtures of these patterns. Lipid bodies, as in HSMCs in vivo, were rarely present in mature cells. Cell surfaces displayed narrow surface folds [9, 20]. Light microscopic immunomorpho-

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

17

logical methods showed that all of these newly developed mast cells contained tryptase, and many contained chymase as well [9]. The cells contained 1.8– 2 ␮g of histamine per 106 cells and had receptors for IgE [9]. Thus, a reproducible culture system was established that permitted the reliable development of HMCs in vitro from agranular CBCs [9]. A detailed ultrastructural analysis of this established culture system for the development of HMCs was performed [20]. In concert with other ultrastructural features of HMCs, differing from those of HBs, the presence of crystal granules (ﬁg. 14) provided the means for the deﬁnitive identiﬁcation of HMCs, since this speciﬁc mast cell granule pattern does not occur in HBs [7, 53, 54]. Not all of the mast cells present after 3 months of coculture with ﬁbroblasts were mature by ultrastructural morphological criteria [20]. A mixture of mature (approx. 75%) (ﬁg. 13) and immature (approx. 25%) (ﬁg. 15) mast cells were identiﬁed [20]. Most of the non-adherent cells present in long-term cultures of CBCs and ﬁbroblasts were not mast cells. A few 3T3 ﬁbroblasts contaminated the non-adherent fraction of cells prepared for electron microscopy, but most of the non-mast cells present were either macrophages or endothelial cells [20]. Immature mast cells (ﬁg. 15) that were present in the 3-month coculture system were larger than their mature counterparts. They displayed ultrastructural features of immaturity, including large, lobular nuclei with dispersed chromatin and large nucleoli. Numerous large, immature, partially ﬁlled granules contained mixtures of membranous, vesicular, and dense materials. Immature granules had foci of dense condensation products superimposed on less dense matrix materials (ﬁg. 16). Small numbers of typical, mature crystal granules were present in these maturing mast cells. Large numbers of elongated mitochondria, free ribosomes, and expanded Golgi structures were present, but progranule production was minimal [20]. While most mature granules were small (similar in size to in vivo cells), giant crystal granules were also present, but rarely so [20]. Ultrastructural criteria were used to identify the endothelial cells that were also present in this coculture system [20]. Most endothelial cells were larger than the mast cells. Mast cells were often closely associated with the outer

Fig. 10. The cytoplasm and Golgi (G) area of a mature basophil obtained and prepared similarly to ﬁgure 8 are shown. Gold particles are associated with small, smooth membrane-bound vesicles in the Golgi area. The nucleus and cytoplasm are also labeled; extracellular background is not labeled. The plasma membrane is not labeled, but subplasma membrane cytoplasmic sites are. Non-particle-ﬁlled, homogeneously dense granules are heavily labeled (closed arrowhead). One particle-ﬁlled granule is not labeled (open arrowhead). !31,000. [From 25, with permission.]

Ultrastructure of Mast Cells and Basophils

18

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

19

surfaces of endothelial cells; some were completely enclosed within them – a process which has been termed emperipolesis. Macrophages (ﬁg. 17) were often similar in size to mast cells [20, 42]. They were ﬁlled with lysosomes and phagolysosomes (ﬁg. 18). These membrane-bound structures contained variable mixtures of membranous, dense amorphous material, and vesicular contents – a circumstance which might pose difﬁculties to observers in terms of distinguishing these typical phagocytes from immature cells of the mast cell lineage (compare ﬁg. 17 and 18 with ﬁg. 15 and 16) [42]. The ultrastructural criteria necessary for discriminating between these two possibilities have been reviewed [20, 42]. The ultrastructural ontogeny of HMCs was investigated in sequential cocultures of CBCs and 3T3 mouse ﬁbroblasts [21]. Clearly, agranular precursors in CBCs gave rise to mature HMCs after long-term culture with ﬁbroblasts [9, 20]. Therefore, we compared light microscopic identiﬁcation of metachromatic cells, immunoﬂuorescent determination of tryptase-positive cells, and incidence of mast cell progenitors and basophils in sequential cocultures of CBCs and ﬁbroblasts [21]. The samples were prepared at 2 and 3 weeks, 25 days, 4 and 5 weeks, and 3 and 3.5 months of culture. A remarkable concordance was obtained using these morphological methods. For example, by 5 weeks of culture, light microscopy revealed 0.7% HBs, and electron microscopy revealed 1% HBs [21]. A similar evaluation for mast cells revealed 3.8% by light microscopy, 3.5% by electron microscopy, and 4.4% tryptase-positive cells [21]. Thus, the identiﬁcation and quantiﬁcation of mast cells and their progenitors evolving from agranular precursors in CBCs were remarkably consistent using three morphological methods with three observers from two separate laboratories [21]. The ultrastructural analysis of HMC development was of interest as well [21]. For example, particulate materials were the earliest visible granule contents that appeared in granule chambers (ﬁg. 19, 20); later, scrolls that were superimposed on particulate materials became evident. HMC progenitors did

Fig. 11. CBCs cultured in suspension for 6 weeks in rhSCF give rise to a mixture of cells. These include neutrophilic myelocytes (NM), phagosome-laden macrophages (MA), and immature mast cells (MC). NM show large, lobular nuclei with dispersed chromatin. Large numbers of primary granules with loosely ﬂocculent contents (arrows) ﬁll the cytoplasm. Distended cisterns of rough endoplasmic reticulum and mitochondria are also present. Note the mixture of full and partially full granules in the MC with narrow, short surface folds. The mast cell granule contents are electron-dense compared with the primary granules (azurophilic granules) which ﬁll the NM cytoplasm. Also note the different surface contour of MC and NM. MA are ﬁlled with phagolysosomes, which vary in size, shape and contents. Bar = 2.2 ␮m. [From 26, with permission.]

Ultrastructure of Mast Cells and Basophils

20

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

21

not contain crystal granules in early cultures, but fully mature HMCs present in the long-term cocultures did. Particle granules are rarely evident in HMCs in some in vivo locations (i.e., skin), but they are prominent in several circumstances; two of these are (1) when mast cells occupy intraepithelial sites [55] in chronic inﬂammatory conditions [69–71] and (2) in developing mast cells studied at early gestational times in human fetuses [72]. The developmental sequence of the evolution of the HMC lineage in vitro, made available by this new culture system, was identical to that described for the development of HMCs in vivo in sequentially examined samples of human fetal tissues [72]. Mature HBs were present in small numbers at early times in these cocultures, but their immature, myelocytic precursors were absent [21]. (Note that we showed previously in suspension cultures that mature HBs developed rapidly from their myelocyte precursors, a process essentially complete by 2–3 weeks [8, 18] – times corresponding to the earliest cocultures that we examined here.) Mature HBs were characteristically secretory, many having only residual empty granule containers – a process termed PMD – whereas HMC progenitors were characteristically synthetic, as evidenced by active granule building. In aggregate, these studies of ﬁbroblast/CBC cocultures have provided documentation of the differentiation and early maturation of HMCs in a system known to give rise to morphologically mature HMCs with crystal granules – cells which contain both tryptase and chymase [9]. All HBs present were small, mature cells which were releasing (or had released) their granule contents. The HMC progenitors were morphologically distinctive [21] from previously described HB progenitors [18]. Fibroblasts have been implicated as culture-competent cells for the development of the mast cell lineage in several species [9, 20, 22, 28, 42, 68, 73–76]. In man, ﬁbroblast monolayers sustain the ultrastructural phenotype and function of isolated HLMCs [28, 68] and permit the differentiation and full maturation of HMCs from their agranular precursors in CBCs [9, 20, 21, 42]. We next investigated whether ﬁbroblasts, or their soluble products, were responsible for the observed development and maturation of the mast cell lineage in man [22, 27]. For these experiments, CBCs were cultured in suspenFig. 12. NM (6-week suspension cultures of CBCs in MFF) prepared either routinely (A) or with a cytochemical procedure (B) to detect myeloperoxidase. A The NM cytoplasm is ﬁlled with primary (azurophilic) granules. These are lightly electron dense. They are membrane bound and contain ﬂocculent and vesicular materials. They are similar in size to the darker mitochondria nearby. Dilated cisterns of RER (arrows) extend throughout the cytoplasm. B Peroxidase is present in the cisterns both around the nucleus and bounded by RER; it is also present in several azurophilic granules in this NM. Mitochondria and nucleus are not peroxidase positive. A !22,000. B !25,000. [From 26, with permission.]

Ultrastructure of Mast Cells and Basophils

22

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

23

Fig. 13. Mature human mast cell (3-month coculture of CBCs and 3T3 ﬁbroblasts) shows a monolobed nucleus with partially condensed peripheral chromatin, narrow surface folds, and numerous membrane-bound granules ﬁlled with electron-dense material. !16,000. [From 9, with permission.]

sion, either with or without the addition of mouse ﬁbroblast culture supernatants, and sampled for light and electron microscopy at early (6–8 weeks) and late (4 months) culture intervals [22]. We found that mast cells developed in quantity in cultures that had been supplemented with the supernatants from mouse ﬁbroblasts (ﬁg. 21) but not at all in unsupplemented culture media [22]. By ultrastructural criteria, the newly developed mast cells did not achieve full maturity, were actively synthesizing new granules, and were undergoing intragranular maturational events in long-term suspension cultures of comparable times to the cocultures in which fully mature HMCs had developed [9, 20]. Small numbers of mature basophils persisted in suspension cultures, and many

Ultrastructure of Mast Cells and Basophils

24

Fig. 14. High magniﬁcation electron micrographs of MC granules (similar source and preparation as in ﬁg. 13) show crystals that are seen in longitudinal (A) and cross section (B). These structures in granules provide deﬁnitive proof that the metachromatic cells containing them are of the MC lineage in humans. A !97,000. B !116,500. [A From 55, with permission. B From 20, with permission.]

were undergoing PMD. Other cell lineages also present included eosinophils, neutrophils, macrophages, and endothelial cells (ﬁg. 22, 23). Ultrastructural inspection of the HMC secretory granules which developed in these suspension cultures showed that the number of granules exceeded those present in the earliest, hypogranular mast cell progenitors which had developed in the coculture system after 3 weeks [21]. The cytoplasmic granules in mast cells developing in the suspension cultures were immature. They were not completely ﬁlled but contained particulate, membranous, vesicular, homogeneously dense, and ﬁnely granular materials. Individual granules pre-

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

25

Fig. 15. Immature HMC, obtained and prepared similarly as in ﬁgure 13, shows a large, lobular nucleus with a deep nuclear cleft, elongated, narrow surface folds, and a mixture of immature and mature granules. The cytoplasm contains large numbers of nonmembrane-attached ribosomes; some ribosomes are attached to endoplasmic reticulum with non-dilated cisterns. Immature granules are large, membrane-bound vacuoles that contain centrally distributed condensations of dense nucleoids, often superimposed on less dense matrix material (arrowhead). !15,000. [From 20, with permission.]

Ultrastructure of Mast Cells and Basophils

26

Fig. 16. Higher magniﬁcation of MC cytoplasm (same source and preparation as in ﬁg. 13) shows a mixture of mature and partially mature granules. Granule contents are heterogeneous, including crystalline arrays (black arrow) and eccentric, concentric, dense lamellar membranes (white arrow). Very dense, homogeneous condensations (nucleoids) are centrally placed in several granules (arrowhead) with peripherally located partial scrolls surrounding them. !63,000. [From 20, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

27

17

Fig. 17. This macrophage (6-week suspension culture of CBCs in 3T3 ﬁbroblast supernatant) illustrates typical macrophage morphology. The monolobed nucleus has an irregular shape, and the chromatin is dispersed. The surface has focal, narrow folds. The cytoplasm is ﬁlled with a mixture of electron-lucent vacuoles and membrane-bound dense bodies of variable size, shape and contents. These are typical macrophage lysosomes and phagolysosomes. Other cytoplasmic organelles include Golgi structures, vesicles, short, narrow strands of RER, and several mitochondria. !11,500. [From 42, with permission.] Fig. 18. Higher magniﬁcation of a macrophage cytoplasm (culture similar to ﬁg. 17) shows the heterogeneous size, shape, and contents of cytoplasmic phagolysosomes. !25,000. [From 42, with permission.]

Ultrastructure of Mast Cells and Basophils

28

18

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

29

19

Ultrastructure of Mast Cells and Basophils

30

20

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

31

dominantly containing particles, scrolls, or ﬁngerprint-like whorls were present among the larger numbers of granules with mixed contents (ﬁg. 24) [22]. Regular parallel and hexagonal arrays of crystalline material – a characteristic routinely seen in granules of mature HSMCs [35, 54, 55, 77, 78] and in mature HMCs arising in cocultures of 3T3 mouse ﬁbroblasts and CBCs (ﬁg. 14) [9, 20] – were absent from the immature mast cells that arose in suspension cultures of CBCs supplemented with mouse ﬁbroblast supernatants, even when the cultures were extended to times comparable to those of the cocultures [22]. We also noted that the immature mast cells developing in the suspension cultures of the CBCs and ﬁbroblast supernatant had quantities of lipid bodies in excess of those present in the coculture system. These ﬁndings show that ﬁbroblasts produce a soluble factor(s) which permits HMCs to develop from agranular precursors in CBCs [22, 27]. Regardless of the length of time in culture, however, these mast cells did not mature completely. Thus, direct contact of CBCs with ﬁbroblasts is not essential for the differentiation and partial maturation of HMCs from their agranular precursors; however, direct contact with ﬁbroblasts or alternatively with ﬁbroblast membrane-bound factor(s) or ﬁbroblast-generated matrix-bound factor(s) is essential for the production of morphologically mature HMCs in vitro [22]. In 1988, it was determined that a proto-oncogene c-kit was encoded by the dominant-white spotting (W) locus of the mouse [79, 80]. c-kit is the cellular homologue of the oncogene v-kit of the H24 feline sarcoma virus [81] and is a member of the tyrosine kinase receptor family [82]. Mice with mutations at the steel (Sl) locus have deﬁcits in the mast cell lineage, as do mice with mutations at the W locus [50, 83–87]. A novel growth factor for murine mast cells has been puriﬁed and identiﬁed as the ligand for the c-kit tyrosine kinase receptor and has been genetically mapped to the Sl locus in mice by several groups [88–92]. This growth factor has been diversely termed by several groups, as follows: (1) SCF [88, 89], (2) kit ligand [90, 91], and (3) MCGF [92]. Recombinant murine and human growth factors for the c-kit ligand are now available [92, 93], and recent studies clearly show that the deﬁcit in mast cells in steel mice [85] can be repaired by injections of the c-kit ligand [89]. More recently, injections of recombinant SCF have been shown to increase mast cell numbers in vivo in normal mice and rats [94–96]. The presence of recombinant SCF in suspension cultures of immature MMCs induced biochemical and morphological maturation [96], and the addition of recombinant SCF to bone marrow cells devoid of visibly granulated MMCs stimulated the development and maturation of granule-ﬁlled mast cells from their agranular bone marrow precursors [31]. The studies initially reported in 1989 [9] and thereafter [20, 21], involving cocultures of murine ﬁbroblasts and human CBCs, clearly showed that the system reliably gave rise to mature, crystal granule-ﬁlled, and tryptase- and

Ultrastructure of Mast Cells and Basophils

32

21

Fig. 19. MC progenitors in 3-week cocultures of CBCs and 3T3 ﬁbroblasts are distinguished by variable numbers of granule-sized, membrane-bound containers that cluster around the Golgi (G) area (A) and occupy the peripheral cytoplasm (B). Some of these immature granules have accumulated dense particulate materials (arrows). The cytoplasm is ﬁlled with free ribosomes, small amounts of RER, which do not have dilated cisternae, mitochondria, and lipid bodies (arrowhead). The large, lobular nucleus often appears separated into several lobes in thin sections (A). Narrow surface folds (B) are variably present. A !13,000. B !10,500. [From 21, with permission.] Fig. 20. MC progenitor in a 5-week coculture of CBCs and 3T3 ﬁbroblasts shows an active Golgi area and peripheral granules containing mixtures of vesicles, membranes, particles, and dense materials. !31,000. [From 21, with permission.] Fig. 21. Immature MCs (7-week suspension culture of CBCs in 3T3 ﬁbroblast culture supernatant) show large nucleoli in monolobed nuclei and large numbers of variably ﬁlled cytoplasmic granules, which are dispersed throughout the cytoplasm. Many of the granules are electron-lucent (A). The cell surface has several long, narrow folds in A but is smoothly contoured in B. A !10,000. B !13,000. [From Dvorak and Ishizaka, Int J Clin Lab Res 1995;25: 7–24, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

33

chymase-containing HMCs. Since the murine c-kit ligand is a product of murine ﬁbroblasts [27, 90], and in order to determine whether the c-kit ligand was a growth factor for HMCs, replicate suspension cultures of CBCs were supplemented with rhSCF, recombinant murine MCGF (rmMCGF), or without growth factor and were examined by electron microscopic, immunological and light microscopic methods [23–25, 27]. We found that each source of the c-kit ligand favored the development of large, immature mast cells which did not complete their granule maturational program and did not contain crystal granules (ﬁg. 25) [23, 27]. These cells expressed Fc⑀RI and could be sensitized with human IgE for anti-IgE-induced release of histamine, prostaglandin D2, and leukotriene C4 [27]. Light microscopic immunohistochemistry showed that the majority of mast cells contained tryptase only [27], whereas the mast cells developing in the ﬁbroblast cocultures (ﬁg. 13) contained both tryptase and chymase [9]. Mast cells did not develop in CBC suspension cultures without the addition of growth factor [23, 24]. The immature mast cells that did develop in cultures supplemented with either source of c-kit ligand – e.g., rhSCF (ﬁg. 25) or rmMCGF (ﬁg. 26) [23] – were ultrastructurally identical to the immature mast cells that developed in suspension CBC cultures supplemented with either full mouse ﬁbroblast cultures supernatants (ﬁg. 21) [22] or with a partially puriﬁed fraction of these culture supernatants [23–25, 27]. Thus, each source of the c-kit ligand that was used to supplement suspension cultures of CBCs, whether recombinant proteins of human (rhSCF) or mouse (rmMGF) origin [23, 27] or naturally occurring c-kit ligand of mouse origin [22, 24, 25, 27], exerted identical effects on the ontogeny of HMCs as they developed from their agranular precursors in cord blood. Full morphological maturity of factor-supported mast cells did not occur [23, 24, 27]. Routine ultrastructural [24] and ultrastructural cytochemical [25] studies to detect CLC protein were performed on multiple suspension cultures of CBCs that were designed so that the effects on the developing mast cell lineage of growth factor source, individual cord sample, and culture time (3–17 weeks) could be individually evaluated [24, 25]. Thus, replicate suspension cultures were set up with cells from individual cord blood samples. These cultures were supplemented with partially puriﬁed murine ﬁbroblast culture supernatants or with rhSCF and were recovered for morphological studies at 3, 6, 8, or 14 weeks [24, 25]. These matched samples showed that a mixture of cellular lineages was present, regardless of the supplement source. That is, either the rh-kit ligand or the naturally occurring mouse kit ligand stimulated HMCs, basophils, neutrophils, eosinophils, macrophages, megakaryocytes, or endothelial cells to develop. The numbers of mast cells and their granules increased with culture time; mature basophils, present in quantity in 3-week cultures, decreased thereafter and released their granule contents (ﬁg. 27). The

Ultrastructure of Mast Cells and Basophils

34

Fig. 22. An endothelial cell is present in a 7-week suspension culture of CBCs in 3T3 ﬁbroblast culture supernatant. Note the large, irregularly contoured nucleus with a dispersed chromatin pattern. The cell surface is relatively smooth. Elongated, endothelial cell-speciﬁc Weibel-Palade bodies (arrows) in the cytoplasm positively identify this cell as an endothelial cell. Other cytoplasmic structures include serpentine strands of RER with non-dilated cisterns, elongated mitochondria, and vesiculo-vacuolar organelles (arrowheads) in the peripheral cytoplasm. !11,500. [From 22, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

35

Fig. 23. Higher magniﬁcation of the endothelial cell in ﬁgure 22 illustrates the endothelial cell-speciﬁc Weibel-Palade bodies in longitudinal and cross sections. These secretory granules contain an array of elongated tubules, which are visible in both views (arrows). !42,000. [From 22, with permission.]

Ultrastructure of Mast Cells and Basophils

36

Fig. 24. Immature MCs (7-week suspension culture of CBCs in 3T3 ﬁbroblast supernatant) with immature cytoplasmic granules. They contain mixtures of particulate, vesicular, homogeneously dense, membranous, and ﬁnely granular materials (A, B). Some granules are nearly electron-lucent (A). Individual granules containing particles (A) or scrolls (B) are apparent, but crystals are absent. A large, round, osmiophilic lipid body is also seen (B). Note the large amounts of intermediate ﬁlaments and free ribosomes in the cytoplasm (B). A !39,000. B !45,000. [From 22, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

37

Fig. 25. This immature MC (14-week suspension culture of CBCs in rhSCF) has a large, immature nucleus with partially dispersed chromatin and a large, round central nucleolus. The cell surface has a large number of short, narrow folds. Many membrane-bound cytoplasmic secretory granules with heterogeneous contents are present. Some granules contain scrolls; most contain mixtures of membranous and dense materials. Nearly half of the secretory granules are immature, with incomplete deposition and condensation of dense granule materials. Other cytoplasmic organelles include elongated mitochondria, free ribosomes, and a cluster of small, dense lipid bodies near the nucleus. !14,000. [From 27, with permission.]

Ultrastructure of Mast Cells and Basophils

38

mast cell lineage developed similarly, regardless of which factor preparation was added to cultures, but considerable variability existed among individual donors from whom cord blood samples had been obtained. Unlike the mature, crystal granule-containing mast cells that regularly developed in ﬁbroblast cord blood cocultures [9, 20] and similar to CBC suspension cultures supplemented with ﬁbroblast culture supernatants [22], HMCs failed to attain full maturity in the suspension cultures, regardless of the individual cord sample, added growth factor, or culture time [24]. Morphological evidence of granule secretion from mast cells was absent. Rather, HMCs were actively undergoing granule building with prominent granule condensation patterns (ﬁg. 28), numerous formation of progranules in Golgi areas, increased number of granules, and ﬁlling of granule containers over time in culture. Crystal-ﬁlled granules never developed. Thus, it was concluded that ﬁbroblasts were necessary and sufﬁcient for the complete maturation of HMCs in vitro from their agranular precursors [9, 20] but, although soluble c-kit ligand was necessary, it was not sufﬁcient [24]. Ultrastructural immunogold studies were performed to detect cellular CLC protein in the suspension cultures described above [25]. These studies showed labeled cells and organelles in the basophil lineage (ﬁg. 8–10), but no cells in the mast cell lineage were labeled (ﬁg. 29) [25]. Thus, the combined ultrastructural morphology and immunogold phenotyping for the presence of CLC protein in cells differentiating in c-kit ligand-supplemented suspension cultures of human CBCs allows accurate assignment of developing basophils and mast cells to their respective lineages [25].

2.3. Circulating Mast Cell Precursor in Embryonic Mice A cell population of murine fetal blood that fulﬁlls the criteria of progenitor mastocytes was identiﬁed [10]. It is deﬁned by the phenotype Thy-1loc-kithi, contains cytoplasmic granules, and expresses RNAs encoding mast cell-associated proteases but lacks expression of the high-afﬁnity immunoglobulin E receptor. Thy-1loc-kithi cells generated functionally competent mast cells at high frequencies in vitro but lacked developmental potential for other hematopoietic lineages. When transferred intraperitoneally, this population reconstituted the peritoneal mast cell compartment of genetically mast cell-deﬁcient W/Wv mice to wild-type levels [10]. Mast cells originate from hematopoietic stem cells [97], and in vitro assays of colony formation indicate that mast cell precursor activity occurs at low frequency in the bone marrow, peripheral blood, and mesenteric lymph nodes of murine rodents [98, 99]. It has been proposed that mast cell precursors leave the bone marrow, migrate in the peripheral blood, and invade mucosal and

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

39

26

Ultrastructure of Mast Cells and Basophils

40

27 Fig. 26. Immature HMCs develop in 8-week suspension cultures of CBCs in recombinant murine c-kit ligand. An expanded Golgi (G) area, with numerous vesicles and vacuoles, is present (A). The cytoplasm is ﬁlled with a mixture of mature, partially mature, condensing, and nearly empty granules. Numerous osmiophilic lipid bodies are also noted (arrowhead). At higher magniﬁcation (B), mature granules contain ﬁnely granular material and scrolls. Large collections of irregular membranes are present in one otherwise empty granule. Several progranules (arrow) are noted. A !13,300. B !44,000. [From Dvorak and Ishizaka, Int J Clin Lab Res 1995;25:7–24, with permission.] Fig. 27. Fourteen-week suspension culture of CBCs in rhSCF shows a mature basophil undergoing PMD. The mature cell is ﬁlled with expanded, partially empty, and nearly empty granule containers. Some retain dense membranes and particles; one contains a large, bipyramidal CLC, which is pushing the granule membrane outward at each of its pointed ends. !10,000. [From 24, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

41

connective tissues where they undergo differentiation into morphologically characteristic mature mast cells. The identiﬁcation of a cell population puriﬁed from murine fetal blood satisﬁes the criteria of a mast cell-committed precursor at a stage before tissue invasion. The leukocyte fraction of fetal blood from day 15.5 of gestation was sorted into Thy-1loc-kithi cells and examined by electron microscopy [10]. These cells exhibited criteria characteristic of very immature mast cells (ﬁg. 30) [48] and differed from immature basophils according to the ultrastructural characteristics of their nuclei, immature cytoplasmic granules, Golgi apparatus, and endoplasmic reticulum [11, 30, 31, 66]. When fetal blood Thy-1loc-kithi cells were cultured for 13 days with IL-3 and SCF, ultrastructural examination revealed a more abundant cytoplasm and more numerous granules, which are indicative of further maturation. Ultrastructural analysis of sorted fetal blood Thy-1–c-kit+ and fetal blood Thy-1+c-kitlo cells, representing multi-potent and pro-thymocyte populations [100], respectively, revealed no evidence for cytoplasmic granule development.

2.4. Mouse Basophils Are Not Lymphocytes We established ultrastructural criteria for the identiﬁcation of mouse basophils and their myelocytic precursors in 1982 [66]. Since then, ultrastructural examination of unknown cellular populations that were otherwise well characterized has led to the recognition of this rare granulocyte lineage in improbable samples [11]. By identifying these cells as non-B, non-T cells they were assigned to the non-lymphoid compartment of spleen and bone marrow and found to contain all of the IL-4-producing capacity of these compartments. Further characterization showed that they were Fc⑀R+, and contained histamine and metachromatic granules. Electron microscopy of these samples obtained from mice and sorted for the presence of the high-afﬁnity Fc⑀R+ established their identity as basophils (ﬁg. 31) [11].

Fig. 28. Eight-week suspension cultures of CBCs in MFF (A) or rhSCF (B) show condensation foci in immature granules of immature MCs. Granule condensation is a late event in the maturation of HMC secretory granules. A The central, dense condensation foci (open arrowhead) are large; one completely ﬁlled, homogeneously dense mature granule is present in this MC (closed arrowhead). B In addition to single, central dense condensation nucleoids, immature granules have multiple small, dense condensation foci superimposed on a less dense matrix (open arrowheads). Note the large lipid bodies (L) surrounded by an outer shell, which is more electron-dense than the lipid body interior. A !47,000. B !38,000. [From 24, with permission.]

Ultrastructure of Mast Cells and Basophils

42

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

43

29

Ultrastructure of Mast Cells and Basophils

44

30 Fig. 29. This nearly mature MC, with many ﬁlled granules and narrow surface folds, developed in a 6-week suspension culture of CBCs supplemented with rhSCF. The section was stained with immunogold to detect CLC protein, and it shows no gold label in the MC. Electron-dense osmium has been removed from two cytoplasmic lipid bodies by the sodium metaperiodate step in the immunogold procedure. These lipid bodies are represented by two perfectly round, electron-lucent cytoplasmic spaces in the Epon embedment (arrowheads). !16,000. [From 25, with permission.] Fig. 30. Puriﬁed mouse fetal blood Thy-1lo c-kithi cell isolated ex vivo and examined by transmission electron microscopy. A representative ex vivo isolated mast cell precursor has a lobular nucleus and several immature cytoplasmic granules containing mixtures of vesicles, electron-dense progranules, lightly dense matrix material, and electron-lucent areas. !10,000. [From 10, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

45

Fig. 31. Basophil in the Fc⑀R+ cell population from spleens of anti-IgD-injected mice. This basophil has a polylobed nucleus with a condensed chromatin pattern, as well as irregular, broad surface processes and a small number of homogeneously electron-dense, membrane-bound cytoplasmic granules. !17,600. [From 11, with permission.]

A detailed ultrastructural analysis of Fc⑀R+ bone marrow cells and splenic non-B, non-T cells sorted from mouse samples was done [30]. These studies showed that the ultrastructural features of the mature basophils (ﬁg. 32) conformed to previous descriptions of this lineage in mice [50, 66, 101, 102]. That

Ultrastructure of Mast Cells and Basophils

46

is, they had polylobed nuclei with condensed chromatin, irregular, broad surface processes and a small number of homogeneously dense, membrane-bound granules in their cytoplasm. Other cytoplasmic organelles included mitochondria, free ribosomes and large numbers of perigranular and cytoplasmic smooth membrane-bound vesicles (ﬁg. 33). The Golgi area was poorly developed. Whereas cells of the basophil lineage in the Fc⑀R+ cells sorted from spleen cells were nearly all (195%) mature basophils, immature basophils were plentiful (~30% of all basophils) in the Fc⑀R+ cells sorted from bone marrow. The ultrastructural features of these immature cells of the basophil lineage (basophilic myelocytes) (ﬁg. 34) also conformed to previous descriptions of these immature cells in mice [66, 102] and other mammalian species [101], that is, they were larger than mature basophils, and their nuclei were lobular and had less chromatin condensation than did mature cells. Broad surface processes and cytoplasmic granules similar to those of mature cells prevailed. Additional features of the basophilic myelocytes included RER with dilated cisternal spaces (ﬁg. 35A), large Golgi areas, and immature cytoplasmic granules. The immature granules (ﬁg. 34A) were larger than mature granules and contained less electron-dense homogeneous material, often bounded by peripheral subgranular membrane vesicles. Cells with cytoplasmic granules, other than basophils, represented !5% of the Fc⑀R+ cells derived from the bone marrow cells of goat anti-IgD-injected mice, and ~2–3% of the Fc⑀R+, non-B, non-T cells sorted from the spleens of such mice. These cells represented a mixture of neutrophils, eosinophils, and monocytes. Mature mast cells were absent from the Fc⑀R+ cell preparations sorted from either the spleen or bone marrow of mice [30]. Very immature poorly to non-granulated mast cell precursors were, however, present in small numbers (ﬁg. 35B). We assessed the ultrastructure and the cell-surface expression of the receptor for immunoglobulin E (Fc⑀R), and c-kit, the receptor for SCF, in mouse basophils (ﬁg. 36) and mast cells present in short-term cultures of mouse bone marrow cells in IL-3 with or without SCF [31]. Basophils did not develop increased numbers of cytoplasmic granules and underwent apoptosis (ﬁg. 37) in cultures containing IL-3 with or without SCF, whereas mast cells thrived and developed increased numbers of granules (ﬁg. 38). Basophils were nearly all Fc⑀R+ c-kit– when sorted after culture in IL-3 and SCF; most mast cells were Fc⑀R+ c-kit+. However, a second population of Fc⑀R+ c-kit– mast cells was present after culture in IL-3 and SCF. These c-kit– mast cells were less mature than c-kit+ mast cells and contained signiﬁcantly fewer cytoplasmic granules than the c-kit+ mast cells present in the same cultures (p ! 0.001). Thus, mouse basophils express little or no c-kit receptor on their surface, nor can they survive for long periods in SCF-supplemented cultures. By contrast, MMCs seem to

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

47

32

Ultrastructure of Mast Cells and Basophils

48

33 Fig. 32. Mature basophils in Fc⑀R+ cells sorted from the bone marrow (A) or spleen (B) cells of mice injected with goat anti-IgD. Note the bilobed nucleus with extensive condensation of peripheral nuclear chromatin, irregular, blunt surface processes, and small numbers of homogeneous, electron-dense, membrane-bound cytoplasmic granules. A !15,500. B !16,000. [From 30, with permission.] Fig. 33. Higher magniﬁcation views of basophils in Fc⑀R+ cells sorted from the bone marrow (A) or spleen (B) cells of mice injected with goat anti-IgD. Note large number of smooth membrane vesicles in the cytoplasm. Many of these vesicles cluster around granules (arrowheads). One perigranular vesicle is ﬁlled with dense content (arrow). A !31,000. B !36,000. [From 30, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

49

Ultrastructure of Mast Cells and Basophils

50

express the Fc⑀R early in their development, even before they express detectable c-kit receptors on their surface. IL-3 promotes cytoplasmic granule formation in immature mast cells, but even more granules are formed when c-kit+ immature mast cells are cultured in both SCF and IL-3. Basophilic myelocytes (ﬁg. 39) that were present in short-term cultures of mouse bone marrow cells in rmIL-3 were identical in ultrastructure to those present either in unfractionated bone marrow cells [66] or in Fc⑀R+ cells that had been sorted directly from freshly isolated bone marrow cell preparations [30]. These cells differ substantially in ultrastructure from the most immature mast cell precursors also present in short-term cultures (ﬁg. 38, 40). For example, basophilic myelocytes are distinctive for their extensive contents of dilated cisterns of RER – structures that generally are absent from cells in the MMC lineage. In addition to their large immature granules, which contain ﬁnely granular, moderately dense material, basophilic myelocytes also usually contain several smaller, homogeneously dense mature basophil granules that aid in their identiﬁcation. Their nuclei may or may not be segmented. By contrast, at 4 days of culture in rmIL-3, the most immature mast cells have a single nucleus and an ample cytoplasm packed with free ribosomes and contain several long strands of RER (ﬁg. 38A). These structures, in contrast to those present in basophilic myelocytes, do not have dilated cisterns. Mast cell precursors also typically have many mitochondria and several granule-sized vacuoles containing small, electron-lucent vesicles (ﬁg. 38A). By 7 days of culture in rmIL-3, mononuclear mast cells exhibit expanded Golgi structures, as well as dense material within some granules (ﬁg. 38B). When cytoplasmic granules with dense content were enumerated in thin sections of these cells, an average of 6.6 granules/mast cell was present in the mast cells that developed after unsorted bone marrow cells from normal mice had been cultured in rmIL-3 for 7 days. This value is signiﬁcantly greater than in the immature mast cells before culture in IL-3. Even greater numbers of cytoplasmic granules were present in mast cells after 11 days of culture (14.9 granules/mast cell, p ! 0.025 versus day 7 value). Moreover, after 11 days in culture, immature mast cells had developed numerous elongated, narrow Fig. 34. Basophilic myeloytes in Fc⑀R+ cells sorted from the bone marrow of goat anti-IgD-injected mice. A The basophilic myelocyte exhibits increased vesicles in the Golgi area (G), dilated cisterns of rough endoplasmic reticulum (closed arrowhead), and immature granules (open arrowhead). These large granules contain central, moderately dense contents surrounded by vesicles beneath their membranes. B The basophilic myelocyte is more mature. It also has extensive dilated cisterns of rough endoplasmic reticulum (closed arrowhead), elongated mitochondria, an expanded Golgi area (G), and small numbers of homogeneously dense mature granules (open arrowhead). A !17,600. B !14,300. [From 30, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

51

35

Ultrastructure of Mast Cells and Basophils

52

36 Fig. 35. Fc⑀R+ cells sorted from the bone marrow of goat anti-IgD-injected mice contain basophilic myelocytes (A) and cells that represent very immature, poorly to nongranulated mast cell precursors (B). A The basophilic myelocyte exhibits nuclear lobation with condensed nuclear chromatin, large amounts of dilated rough endoplasmic reticulum (open arrowhead), and small numbers of homogeneously dense, immature (closed arrowhead) or mature (arrow) cytoplasmic granules. B The immature mast cell, by contrast, has a large, lobular nucleus with dispersed chromatin, primarily free ribosomes with small amounts of non-dilated rough endoplasmic reticulum, numerous mitochondria, and small numbers of large, lucent, membrane-bound vacuoles (closed arrowhead) containing vesicles and focal, small aggregates of dense progranule material. A !11,000. B !10,000. [From 30, with permission.] Fig. 36. Mature basophil in Fc⑀R+ population sorted from normal mouse bone marrow cells. This cell exhibits features typical of a mature mouse basophil: a polylobed nucleus with heavily condensed chromatin, short, blunt surface processes, mitochondria, vesicles, and homogeneously dense granules in the cytoplasm. !17,700. [From 31, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

53

Fig. 37. A Mouse basophil exhibiting apoptosis in Fc⑀R+ population sorted from bone marrow cells after 1 week of culture with rmIL-3 and SCF. Note completely pyknotic, rounded nuclear masses (arrowhead). B Basophilic myelocyte exhibiting apoptosis in Fc⑀R+ population sorted from bone marrow cells after 11 days of culture with rmIL-3 and SCF. Note the condensed pyknotic nucleus (N). A !8,500. B !10,000. [From 31, with permission.]

surface folds, as well as large numbers of granules (ﬁg. 38C). After 21 days of culture with rmIL-3, some immature mast cells contained predominantly mature granules with completely dense matrix, while still displaying evidence of nuclear immaturity, e.g., large nucleoli (ﬁg. 38D), whereas other immature mast cells still exhibited a mixture of immature, incompletely ﬁlled granules and completely dense mature granules.

Ultrastructure of Mast Cells and Basophils

54

We also analyzed the Fc⑀R+ c-kit+ cells that were sorted from cultures of normal mouse bone marrow cells after 7 or 11 days in a mixture of rmIL-3 plus SCF. We found that 96% of these cells were immature mast cells, whereas only 2% could be classiﬁed as basophils. Thus, substantially more mast cells comprise the Fc⑀R+ c-kit+ cells (96%) than are present in the Fc⑀R+ c-kit– cells (34%). The number of granules present in the Fc⑀R+ c-kit+ mast cells sorted from bone marrow cells cultured for 7 or 11 days (ﬁg. 41) was signiﬁcantly greater than that in all of the other mast cell populations that we quantiﬁed. In summary, mast cells underwent maturation, as reﬂected in increased numbers of secretory granules, in all cultures containing IL-3 and/or SCF. By contrast, under similar conditions of culture, basophils exhibited a diminution in the numbers of their secretory granules and, in some cases, underwent apoptosis.

2.5. Mouse Basophils Are Not Neutrophils Ultrastructural data were used to identify defects in terminal differentiation of neutrophils in the CCAAT enhancer binding protein family of transcriptional factor knockout mice (C/EBP⑀–/–) [32]. Peripheral blood of C/EBP⑀–/– and control mice was obtained from the retro-ocular venous plexus and prepared for electron microscopy, whereas the remainder of the molecular, functional, and biochemical data in this report were obtained using bone marrow cells, peritoneal cells elicited with thioglycollate injections, or the cell line U937 stably transfected with a zinc inducible C/EBP⑀ (p32) expression vector, or empty vector. The electron micrographs illustrated in ﬁgure 1A and B of Verbeek et al. [32] illustrate a classic mature mouse basophil (see ﬁg. 31, 36), not an immature neutrophil as stated, in a C/EBP⑀–/– mouse [33]. The granules shown are typical for histamine-containing mouse basophil granules [11, 30, 31, 66], which are less numerous and larger than mature granules in mature mouse eosinophils and neutrophils [66]. Basophils do not have secondary and tertiary granules, as do neutrophils, and the electron-lucent structures referred to as glutaraldehyde-extracted secondary granules [32] are typical cytoplasmic vesicles and vacuoles, which are present in basophils from multiple species (ﬁg. 33) [30, 31, 103–106]. Figure 1C of Verbeek et al. [32] shows a classic neutrophil from a control mouse, as indicated in the legend. Whether or not the basophil illustrated in Verbeek et al.’s ﬁgure 1A and B represents a real increase in this rare granulocyte class in C/EBP⑀–/– mice cannot be determined from the single cell present here. Because much of the molecular, functional and biochemical data for defective terminal neutrophil differentiation was based on studies of bone marrow and elicited peritoneal exudate cells in C/EBP⑀–/– mice,

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

55

it would be important to do ultrastructural studies of these cell populations in these knockout mice and their wild-type controls. The mouse basophil lineage was deﬁned by ultrastructural analysis, contrasted with developing mast cell, eosinophil, and neutrophil lineages, and reported in 1982 [66]. These images clearly established the identiﬁcation and evolution of each lineage in this species and provided the necessary background for the identiﬁcation of mouse basophils as IL-4-producing, high-afﬁnity Fc⑀R+, histamine-containing cells sorted from mouse bone marrow and spleen (ﬁg. 31, 36) [11], previously referred to as non-B, non-T cells [107]. The ultrastructural characteristics of these basophils are described and contrasted with those of Fc⑀R+ mast cells and with Fc⑀R– neutrophils and eosinophils in detail [30]. Further ultrastructural studies with this model revealed that mouse basophils are Fc⑀R+ and c-kit–, whereas mast cells are Fc⑀R+ and c-kit+. Also, mouse basophils did not thrive in cultures supplemented with IL-3 or SCF, on the one hand, or with a combination of these two MMC growth factors, on the other [31]. In all of these references [11, 30, 31, 66], ultrastructural images of mouse basophils identical to the basophil in Verbeek et al.’s ﬁgure 1A and B are included.

2.6. Ultrastructural Identification of Monkey Basophils Ultrastructural and cytochemical studies of peripheral blood samples from a monkey continuously infused with rhIL-3 were performed [12]. rhIL-3 stimulated a delayed granulocytosis primarily characterized by numerous mature basophils and fewer eosinophils and neutrophils. Basophilic leukocytes were identiﬁed by ultrastructural analysis (ﬁg. 42). They were found to be typical granulocytes with polylobed nuclei containing condensed chromatin and numerous cytoplasmic granules. Basophil secretory granules were ﬁlled with homogeneous dense contents and were larger than eosinophil and neutrophil secretory granules. Evidence of increased basophil production was accompanied by IL-3-associated activation morphologies. These included an increased number of cytoplasmic and granule-associated vesicles, as are routinely present in a non-IgE-mediated basophil release reaction, termed piecemeal degranulation (PMD), and focal perigranular matrix swelling and granule membrane fusion which accompanies AND of basophils in other species. Monkey basophils were shown to have a different ultrastructural morphology than that published for monkey mast cells, but exhibited general morphologic criteria for the identiﬁcation of circulating mature basophils in a number of species. Like human and guinea pig basophils, monkey basophils did not display endogenous peroxidase or peroxidatic activity in a cytochemical

Ultrastructure of Mast Cells and Basophils

56

Fig. 38. Mouse bone marrow cells cultured for 4 (A), 7 (B), 11 (C), or 21 (D) days in rmIL-3, illustrating the developing mast cell lineage. A The most immature mast cell precursors do not have dense content in their granules. They characteristically contain 2–6 apparently empty or vesicle-containing vacuoles (immature granules, arrowhead) in an abundant free ribosome-packed cytoplasm. Non-dilated strands of rough endoplasmic reticulum and mitochondria are also present. The cell surface lacks typical mast cell folds, and a large nucleolus is present in a single-lobed nucleus. B This immature mast cell shows an expanded Golgi (G) area with active granulogenesis. Note that immature granules now contain some focal deposits of dense material (arrowhead). C An immature mast cell after 11 days in culture with rmIL-3 contains more granules than the cell in B, and many of these are more completely ﬁlled with dense material than at earlier intervals of culture. The single nucleus is irregularly contoured; many narrow folds are now evident on the cell surface. D After 21 days of culture in rmIL-3, an immature mast cell contains many granules that are completely electron-dense and a few vesicle-ﬁlled immature granules (arrowhead). The surface displays the numerous narrow folds typical of mast cells. A large nucleolus persists in the large lobular nucleus. A !9,350. B, C !7,725. D !6,375. [From 31, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

57

Fig. 39. Typical basophilic myelocyte, which differs substantially in ultrastructure from immature mast cells (compare with ﬁg. 40 from same cell culture), in Fc⑀R+ population sorted from the bone marrow cells of a goat anti-mouse IgD-injected mouse and then cultured in rmIL-3 for 11 days. Note that the cytoplasm of the basophilic myelocyte, unlike that of the immature mast cell (see ﬁg. 40), contains many dilated cisterns of rough endoplasmic reticulum and many elongated mitochondria. The immature basophil granules contain centrally located, homogeneously dense material surrounded by small vesicles beneath the granule membrane. !11,000. [From 31, with permission.]

Ultrastructure of Mast Cells and Basophils

58

Fig. 40. Immature mast cell in Fc⑀R+ population sorted from the bone marrow cells of a goat anti-mouse IgD-injected mouse and then cultured in rmIL-3 for 11 days. The immature mast cell granules contain a mixture of dense progranules and vesicles; progranule formation is apparent in the Golgi (G) area. Cytoplasmic ribosomes are plentiful; most are free of membranes. !7,500. [From 31, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

59

Fig. 41. Mouse bone marrow cells cultured for 1 week in rmIL-3 plus SCF and then sorted for Fc⑀R+ c-kit+ cells contained virtually all mast cells (96%, average of three experiments). These cells still displayed some ultrastructural criteria of immaturity, but the average number of granules (22.3/cell at 7 days and 22.6/cell at 11 days) exceeded that for all other immature mast cell populations analyzed in this study. !10,000. [From 31, with permission.]

Ultrastructure of Mast Cells and Basophils

60

Fig. 42. Peripheral blood cells from a monkey given rhIL-3 processed with the osmium potassium ferrocyanide method shows a typical monkey basophil. The cytoplasm is ﬁlled with secretory granules. Also, note the large number of dark glycogen particles in the cytoplasm. !23,300. [From 12, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

61

assay which simultaneously identiﬁed peroxidase-positive granules in neutrophils and eosinophils as well as in synthetic structures in eosinophils (ﬁg. 43). Thus, electron microscopy and knowledge of the ultrastructural features of basophils in several species facilitated the identiﬁcation of basophils in a new species [12].

2.7. Ultrastructural Verification of the Identity of Mast Cells or Basophils in Cell Samples Obtained from Human Organs and Fluids Ultrastructural studies of biopsies from a wide variety of human organs in health and disease contributed to the interpretative knowledge base necessary to verify the identity of cells in samples prepared from various organ sites for functional studies [reviewed in 55]. These studies showed that basophils comprised the histamine-containing metachromatic cell population in human peritoneal ﬂuid preparations from patients on peritoneal dialysis [39]. Initially, we characterized the ultrastructural morphology of isolated HLMCs (ﬁg. 2) [36] and isolated human gut mast cells (HGMCs) from the colon (ﬁg. 44) [38]. The morphological diversity of HLMCs noted in situ [reviewed in 55] extends to these cells when they are isolated from excised human lungs containing malignant tumors. Despite the described variability, HMCs prepared from this source express mostly scroll granules and numerous lipid bodies (ﬁg. 2) [36]. HGMCs obtained from colons resected for carcinoma resembled their in situ counterparts [38, 55]. Mast cells from this location also contained numerous lipid bodies; granules often were ﬁlled with particles but scroll-ﬁlled granules were also evident (ﬁg. 44). Human synovia, obtained at arthroplasty from patients with rheumatoid arthritis and osteoarthritis, were the source for preparation of isolated human synovial mast cells (HSYMCs) [34]. These preparations were assessed functionally and by light and electron microscopy. Mast cell numbers were elevated in synovial tissues obtained from either arthritic disorder, and by functional and morphological criteria, HSYMCs did not differ from HLMCs or HGMCs [34]. By electron microscopy, HSYMCs were rounded cells ﬁlled with cytoplasmic granules (ﬁg. 45) which most commonly displayed scrolls. Crystal-containing granules, like those in HSMCs, were less frequent and particle-ﬁlled granules, like those in HGMCs, were rare. Lipid bodies were evident and these displayed focal lucencies within them (ﬁg. 45). Human uterine mast cells were also characterized by electron microscopy (ﬁg. 46) [37]. Mast cells in this location primarily had granules ﬁlled with scrolls and/or crystals; particles were conspicuously absent. Lipid bodies were

Ultrastructure of Mast Cells and Basophils

62

Fig. 43. Peripheral blood cells from a monkey given rhIL-3 processed to demonstrate peroxidase activity shows peroxidase-positive azurophilic (primary) granules in a neutrophil in A, peroxidase-positive eosinophil granule matrix and portions of endoplasmic reticulum in B, and peroxidase-negative basophil in C. DAB. !7,500. [From 12, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

63

Fig. 44. HGMC ex vivo is ﬁlled with particle granules (short arrow). Several granules contain beaded threads in a reticular array (long arrow). Lipid bodies are moderately dense and contain central vesicles and densities (white arrow). !21,500. [From 38, with permission.]

present and these had focal lucencies much like those seen in HGMCs and HSYMCs [37]. The ultrastructural morphology of isolated HSMCs (ﬁg. 47) differed somewhat from those obtained from lung, gut, uterus, and synovium [35]. These differences included crystal granules as the predominant granule pattern and exceedingly rare lipid bodies.

Ultrastructure of Mast Cells and Basophils

64

Fig. 45. HSYMC ex vivo shows scroll granules, focal lucencies in lipid bodies (arrowhead), and PMD. !12,000. [From 55, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

65

Fig. 46. Human uterine mast cell ex vivo is a mononuclear cell with narrow surface folds that is ﬁlled with electron-dense granules. Several partially dense lipid bodies are seen. !14,500. [From 55, with permission.]

Ultrastructure of Mast Cells and Basophils

66

Fig. 47. HSMC ex vivo shows well-preserved cell with full complement of dense granules. Lipid bodies are not present. !11,000. [From 55, with permission.]

Ultrastructural Analysis Is Necessary and Sufﬁcient for Identiﬁcation of Basophils and Mast Cells

67

Chapter 3

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Cyclooxygenase, a Key Enzyme Family for Production of Prostaglandins, Is Present in Human Mast Cell Lipid Bodies

3.1. Overview Lipid bodies are a major cytoplasmic organelle in HMCs [5, 7, 36, 108, 109]. Our early studies [reviewed in 7] of these important organelles deﬁned their widespread presence in mammalian cells, their ﬂuctuations in number and size and substructural detail and their distinction from cytoplasmic secretory granules in HMCs. These studies led us to consider the possibility for the ﬁrst time that these lipid-rich domains might be important in arachidonic acid metabolism, which is essential for the generation of potent mediators of inﬂammation, such as prostaglandins and leukotrienes. Subsequently, we showed with ultrastructural autoradiography that [3H]-arachidonic acid ([3H]-AA) was present in HMC lipid bodies [5, 109]. These ﬁndings fueled our search for cyclooxygenase in lipid bodies using ultrastructural immunogold technology [110].

3.2. Lipid Bodies Are Distinct Non-Membrane-Bound Organelles Lipid bodies are rounded, non-membrane-bound structures of variable size and density that are present in the cytoplasm of cells of diverse lineages. They are distinct cytoplasmic organelles that have generally been regarded to contain neutral lipids [111]. Emerging data, however, indicate that lipid

bodies may have a role in the oxidative formation of eicosanoids [112–114]. Eicosanoids are synthesized from substrate arachidonic acid that is esteriﬁed principally within phospholipids of cells. These arachidonyl-phospholipids are generally believed to reside solely within cell membranes, and membranes are considered to serve both as repositories of arachidonate storage and as sites of synthesis of arachidonate-derived eicosanoids. Our interest in lipid bodies as organelles involved in eicosanoid formation is based on a series of morphological ﬁndings. First, lipid bodies are widely distributed in mammalian cells, many of which are known to participate in the generation of eicosanoids [115]. Second, lipid bodies characteristically increase in cells as they participate in a wide variety of acute, subacute, and chronic inﬂammatory processes in tissues [5]. Third, cells that demonstrate increased lipid bodies are not damaged, using ultrastructural morphological criteria [115]. These morphological observations indicate that lipid body formation in vivo represents a morphological correlate for cell participation in inﬂammation. Additional insights into the formation of lipid bodies were derived from studies with human neutrophils [116, 117]. Lipid bodies, morphologically identical to those present in reactive tissue neutrophils in vivo, could be induced to form de novo within minutes by exposure of cells to cis-unsaturated fatty acids [116, 117]. In addition to cis fatty acids, other activators of protein kinase C, including cell-permeant diglycerides and speciﬁc phobol esters, induced lipid body formation, whereas inhibitors of protein kinase C (i.e., H-7 staurosporine, and 1-O-hexadecyl-2-O-methyl-rac-glycerol) inhibited lipid body formation in neutrophils induced by cis fatty acids and other protein kinase C activators [117]. Thus, lipid body formation within cells represents a concerted cell response, mediated by protein kinase C activation, which develops in diverse cells associated with varied inﬂammatory responses.

3.3. Ultrastructural Autoradiographs Localize Arachidonic Acid to Human Mast Cell Lipid Bodies A role for lipid bodies as repositories of substrate arachidonic acid available for eicosanoid formation has been suggested from ultrastructural and biochemical studies with a number of cell types [reviewed in 112–114]. Early observations indicated that a variety of [3H]-labeled lipid synthetic precursors, including palmitic acid, methylcholine, and myoinositol, could be localized to lipid bodies in ultrastructural autoradiographic studies of cultured guinea pig peritoneal macrophages [118]. Direct demonstration that the speciﬁc arachidonate fatty acid precursor for eicosanoid synthesis by either the lipoxygenase or the cyclooxygenase pathway was localized in lipid bodies was ob-

Cyclooxygenase, a Key Enzyme Family for Production of Prostaglandins, Is Present in Human Mast Cell Lipid Bodies

69

tained from ultrastructural autoradiographic localization of cell-incorporated [3H]-AA. [3H]-AA incorporation within lipid bodies was found in guinea pig, mouse and human macrophages [5], human ciliated respiratory epithelial cells [113], guinea pig line 10 hepatoma cells [113], mouse peritoneal lymphoblasts [113], human eosinophils [119], and HMCs (ﬁg. 48A) [5, 109]. Biochemical analysis of total cell lipids indicated that the [3H]-AA in lipid bodies was not free fatty acid, but rather was esteriﬁed in glycerolipids. In HLMCs, most [3H]-AA was in neutral lipids [120], whereas in mouse peritoneal macrophages [5, 121], human neutrophils [116], and eosinophils [119] most [3H]-AA was esteriﬁed in phospholipid classes. Lipid bodies isolated by subcellular fractionation from guinea pig peritoneal macrophages and line 10 hepatocarcinoma cells [122] and human eosinophils [123] contained incorporated [3H]-AA, demonstrable by autoradiography. Most directly, with lipid bodies isolated free of cell membranes by subcellular fractionation of lipid body-rich human eosinophils, most [3H]-AA incorporated within lipid bodies was shown to reside in phospholipid classes, including the same proportions of phosphatidylcholine, phosphatidylinositol, phosphatidylethanolamine and phosphatidylserine as [3H]-AA labeled in cell membranes [123]. Therefore, lipid bodies are major intracellular sites of incorporation of arachidonate, which is largely present in the form of arachidonyl-phospholipids.

3.4. Immunogold Ultrastructural Localization of Prostaglandin Endoperoxide Synthase (Cyclooxygenase) in Human Mast Cell Lipid Bodies If arachidonyl-phospholipids present in lipid bodies are to provide substrate for the formation of eicosanoids, such lipids must either be translocated to cell membranes for eicosanoid formation or be acted on locally at lipid body sites. That enzymatic activities reside in lipid bodies has been indicated by the regular demonstration of endogenous peroxidase and/or non-speciﬁc esterase in lipid bodies of HLMCs [112], guinea pig tumor cells [113], human eosinophils [60, 65, 124, 125], human monocytes [124], guinea pig basophils [112, 126], cultured cloned guinea pig endothelial cells [127], HBs [18, 53, 112, 124], and human skin histiocytes [125]. Whether this esterase activity might reﬂect lipolytic enzymes and whether some of the peroxidase activity might be mediated by the peroxidatic function of eicosanoid-forming enzymes (e.g., prostaglandin endoperoxide synthase [128]) has not been ascertained. Prostaglandin synthase, also referred to as cyclooxygenase, is the initial rate-limiting enzyme in the cyclooxygenase pathway of synthesis of prostacyclin, thromboxanes, and prostaglandins. Because the direct localization of this initial enzyme of the cy-

Ultrastructure of Mast Cells and Basophils

70

Fig. 48. Isolated human lung mast cells, prepared for electron microscopic autoradiography after an overnight pulse of [3H]-AA (A) or prepared by an immunogold method to demonstrate prostaglandin H (PGH) synthase (B), are illustrated. A The numerous osmiophilic lipid bodies contain silver grains, indicating the presence of [3H]-AA (open arrowhead). Granules (closed arrowhead) and plasma membrane (arrow) are not labeled. B PGH synthase is present in lipid bodies (L), as indicated by the presence of gold particles, but not in granules (G) or plasma membrane (arrow). A ⫻20,000. B ⫻36,000. [From 122, with permission.]

clooxygenase pathway (often believed to reside solely in plasma membranes) to lipid bodies would indicate that lipid bodies are both stores of arachidonate and possible sites of eicosanoid synthesis, we utilized ultrastructural immunogold localization to document that prostaglandin synthase resides in lipid bodies of HLMCs (ﬁg. 48B).

Cyclooxygenase, a Key Enzyme Family for Production of Prostaglandins, Is Present in Human Mast Cell Lipid Bodies

71

Chapter 4

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Subcellular Localization of the Cytokines, Basic Fibroblast Growth Factor and Tumor Necrosis Factor-␣ in Mast Cells

4.1. Overview The realization that the mast cell repertoire includes the production, storage and release of cytokines considerably enhanced understanding of mast cell biology [reviewed in 129–131]. Prior to this, mast cells were primarily known for their stored products (heparin, histamine, chymase and tryptase, for example) and for the rapid generation of eicosanoids from the precursor, phospholipid-derived arachidonic acid [reviewed in 129, 132]. Release of mast cell-derived mediators of inﬂammation impacts tissues adjacent to them upon their release. Consequently, mast cell-derived cytokines must now be included in the assessment of functional and pathological processes in mast cell-rich human tissues in health and disease. Subcellular locations of cytokines were sought to establish possible secretory mechanisms for their release from mast cells in health and disease. For example, cytokines in secretory granules could be released by traditional regulated exocytotic secretion of entire granules, as seen in IgE-mediated AND or in small vesicular packets which bud from granules and are the effectors of PMD [see Chapter 7]. Ultrastructural post-embedding immunogold procedures and specialized ﬁxation technologies were used to establish subcellular sites of two model cytokines in two mast cell populations [133–137].

4.2. Cytokines in Lipid Bodies Lipid bodies are prominent subcellular organelles in HMCs present in the lung and gut [see Chapters 2, 3]. We examined isolated HLMCs in vitro for the presence of bFGF and HGMCs in situ, which were present in colonic biopsies of patients with Crohn’s disease (CD), for the presence of TNF-, with ultrastructural methods designed to preserve and image these two cytokines. 4.2.1. Tumor Necrosis Factor-

The localization of cytokines and investigations of the mechanisms of actions in the pathophysiology of intestinal disorders has included studies of a large number of cytokines and, speciﬁcally, studies of TNF-. Recent progress regarding TNF- has been facilitated by new studies using high magniﬁcation imaging to localize and quantitate cellular, subcellular and extracellular TNF- in intestinal biopsies [133]. These localizations provide an anatomic basis for a role for TNF- in intestinal inﬂammation that is supported by data acquired by hemolytic plaque assays to assess single cell contributions to TNF- production [138] and steady-state perfusions with biochemical determinations of TNF- in perfusates [139] of patients with inﬂammatory bowel disease (IBD). Important mechanistic studies have been based on functional analyses following (1) the addition of cytokine (using puriﬁed TNF- or TNF--producing disease models) or (2) the ablation of cytokine (using appropriate antibodies, receptor blockade, or knockout mouse models). Altogether, these studies provide new insights into the role of the proinﬂammatory cytokine, TNF-, in human intestinal diseases. TNF- was originally identiﬁed as an endotoxin-induced serum factor causing necrosis of tumors, hence the name tumor necrosis factor [140]. Because of its proinﬂammatory properties, a role(s) for TNF- in IBD has been actively sought, with controversial results [reviewed in 133]. The modalities with which this issue has been examined may have inﬂuenced and even directed this controversy. For example, in reports on the pathophysiology of CD which examine a role for TNF- (or a lack thereof), the results may reﬂect issues of test sensitivity and resolving power. The issue was again addressed as a result of the increased sensitivity, quantitation and resolving power made possible by ultrastructural immunogold imaging of intestinal biopsies from patients with CD [133]. In these studies, TNF- was speciﬁcally localized to subcellular sites in six cell lineages, with a rank order for labeling density (gold particles per cellular square micrometer) as follows (in descending order): ﬁbroblast, eosinophil, mast cell, macrophage, absorptive epithelial cell, and Paneth cell. Neutrophils also contained label for TNF- but were not quantitated, since their numbers in CD biopsies were small.

Subcellular Localization of the Cytokines, Basic Fibroblast Growth Factor and Tumor Necrosis Factor- in Mast Cells

73

The TNF--containing subcellular organelles included lipid bodies (ﬁg. 49A), which were present in each TNF--positive cellular type (except Paneth cells), the secretory granules of eosinophils and Paneth cells, phagolysosomes of macrophages and epithelial absorptive cells, and Golgi structures and cytoplasmic vesicular membranes in neutrophils [133]. A gradient of extracellular TNF- immunoreactivity surrounded eosinophils, mast cells and macrophages. Probability values of gold counts per square micrometer were signiﬁcant for all cells, subcellular organelles, and extravascular pericellular spaces, and all TNF--positive structures signiﬁcantly exceeded the background labeling density per square micrometer. Speciﬁcity controls (normal rabbit serum, TNF--absorbed primary antibody) either failed to label these sites or yielded markedly reduced speciﬁc TNF- labeling, respectively (ﬁg. 49B) [133]. The relative TNF- labeling density in individual subcellular organelles for TNF--positive cells in CD biopsy specimens was also determined in the ultrastructural immunogold analysis. This approach revealed that lipid bodies were intensely labeled (ﬁg. 49A). Indeed, these organelles, recently implicated in intracellular eicosanoid cascades [141, 142], either were the sole intracellular site of TNF- (ﬁbroblasts, mast cells) (ﬁg. 49), exceeded labeled phagolysosomes (an additional subcellular site in macrophages), or were equivalent to the secretory granule stores of TNF- in eosinophils [133]. Paneth cells had no lipid bodies whatsoever; TNF- resided in the immature and mature secretory granules in these epithelial secretory cells [133]. Altogether, the added sensitivity and resolution of this ultrastructural immunogold modality has provided deﬁnitive proof that multiple cells in the chronic inﬂammatory process of CD contain immunoreactivity for TNF-. Moreover, TNF--labeled extracellular gradients around three secretory cells provide evidence for the secretion of TNF- into the bowel wall in CD [133]. The cells containing TNF- in CD biopsy specimens are of interest. Eosinophils are regularly present in these biopsy specimens, and many are activated, as determined by ultrastructural criteria [141]. TNF- was ﬁrst detected ultrastructurally in human peripheral blood eosinophils from patients with the hypereosinophilic syndrome [143]; TNF- mRNA was also detected in hypereosinophilic syndrome eosinophils by in situ hybridization [144, 145], as well as in tissue eosinophils in necrotizing enterocolitis tissues [146] and nasal polyps [145]. Mast cells are increased in CD tissues, and TNF- has been detected Fig. 49. Human mast cells demonstrate TNF- label in numerous large, non-membrane-bound lipid bodies (A, open arrowhead). The smaller secretory granules do not label above background (A, ﬁlled arrowhead and inset). Lipid bodies (B, open arrowhead) and granules (B, ﬁlled arrowheads) were negative in a TNF- absorption control. A and inset !33,300. B !38,200. [From 133, with permission.]

Ultrastructure of Mast Cells and Basophils

74

Subcellular Localization of the Cytokines, Basic Fibroblast Growth Factor and Tumor Necrosis Factor- in Mast Cells

75

in mast cells of several species [reviewed in 133, 134]. However, TNF- was ﬁrst localized to mast cell lipid bodies in CD biopsy specimens [133]. Macrophages are both phagocytic and secretory cells, and the subcellular localization of TNF- of macrophages present in the CD biopsy specimens [133] suggests that lipid bodies may be associated with synthetic processes, and phagolysosomes with the endocytosis of exogenous TNF-. Previously, TNF- mRNA and protein were localized to macrophages (using light microscopy) in IBD [147] and in necrotizing enterocolitis [146]. Fibroblasts, capable of producing TNF- [148], contained high concentrations of TNF- in numerous lipid bodies in CD biopsies [133]. Fibroblast lipid bodies are sites of eicosanoid metabolism [142]. Absorptive epithelial cells had TNF- in their lipid bodies in CD tissues – a novel ﬁnding which implicates TNF--containing epithelial cells in the barrier, secretory and resorptive functions of these epithelia [133]. It was conﬁrmed that Paneth cells contained TNF- [146] (in their secretory granules) in CD biopsy specimens but that Paneth cells were just one of the many cell types to do so; the relative density of Paneth cell label ranked last of the six cell lineages thus quantitated [133]. A reverse hemolytic plaque assay, identifying TNF--secreting cells isolated from mucosal biopsy specimens of children with IBD, provided a sensitive quantitative estimate of this cytokine in CD, ulcerative colitis (UC), mild, non-speciﬁcally inﬂamed intestine, and in histologically normal intestine [138]. These studies revealed a signiﬁcant increase in TNF--producing cells in non-speciﬁc inﬂammation and UC, compared to controls; CD samples were signiﬁcantly increased above those in UC. Cyclosporine therapy in CD reduced this value. Another study examined CD and control tissues by in situ hybridization and detected increased mRNA for TNF- in the lamina propria of CD samples [149]. Additionally, mRNA-positive cells were regularly found in the macroscopically normal mucosal samples from CD patients [149], a ﬁnding in agreement with the detection of TNF- by ultrastructural immunogold analysis in endoscopically and microscopically uninvolved CD mucosal samples [133]. Arachidonic acid metabolism is involved with TNF- synthesis and function(s) (and vice versa) and both TNF- and eicosanoids play a role in IBD [139, 150, reviewed in 133]. The interface of TNF- and arachidonic acid functions in IBD was substantially strengthened by the localization of TNF- (by ultrastructural immunogold labeling) to lipid bodies in ﬁve cell lineages in intestinal biopsy specimens of patients with CD [133]. Lipid bodies are organelles newly recognized to have a key role as a non-membrane intracellular site of arachidonic acid metabolism [141, 142; see Chapter 3]. Ultrastructural autoradiographic methods have localized arachidonic acid to lipid bodies, and ultrastructural immunogold localizations show that enzymes essential for

Ultrastructure of Mast Cells and Basophils

76

prostaglandin and leukotriene synthesis also reside in these structures [141, 142]. Additional localization of a proinﬂammatory cytokine, such as TNF-, in lipid bodies implicates lipid bodies and their non-membrane stores of arachidonic acid in cytokine and eicosanoid cascades which are important to the pathophysiology of IBD. 4.2.2. Basic Fibroblast Growth Factor

We identiﬁed bFGF in non-membrane bound cytoplasmic lipid bodies which typify HMCs of lung origin (ﬁg. 50A, 51) [5, 108, 109]. Lipid bodies, large, non-membrane-bound organelles, were located among cytoplasmic secretory granules (ﬁg. 50) and in a paranuclear distribution (ﬁg. 51). Lipid bodies in HLMCs were enmeshed in a cage of intermediate ﬁlaments. Their surfaces were partially encircled with elongated elements of smooth endoplasmic reticula, some of which contained 5 nm gold-labeled bFGF (ﬁg. 51B). Focal clusters of free ribosomes appeared adjacent to and attached to lipid bodies (ﬁg. 51B). Lipid bodies that were labeled for bFGF were closely applied to the nucleus to such an extent that the encircling perinuclear cisterns appeared obliterated (ﬁg. 51A); some lipid bodies were adjacent to nuclear pores (ﬁg. 51B). Controls for the immunogold technique were negative (ﬁg. 50B). We have determined that lipid bodies are important contributors to eicosanoid synthesis [5], contain other cytokines, such as TNF- [133], and also contain much of the machinery necessary for protein synthesis [17]. In this study, close associations of bFGF-containing smooth endoplasmic reticula, ribosomes and nuclei with lipid bodies were noted in HLMCs. These associations may indicate a non-classical synthetic-secretory-storage pathway for bFGF in HMCs.

4.3. Cytokines in Granules Imaging ultrastructural contents of secretory granules in basophils and mast cells has been an objective in our laboratory for some years. By doing so, we hoped to provide proof of principle that unloading of classical secretorystorage granules during PMD occurs by vesicular transport of granule materials to the extracellular milieu [see Chapter 7]. To this end we have used post-embedding ultrastructural immunogold methods to image cytokines: bFGF [137], TNF- [133–135, 143], CLC protein [25, 151; see Chapter 5], chymase [152; see Chapter 5], and histamine [153; see Chapter 5]. We also developed new enzyme-afﬁnity-gold [154; see Chapter 6] and enzyme-inhibitor-gold [155; see Chapter 6] techniques to visualize histamine [154; see Chapter 6] and heparin [155; see Chapter 6] for the same purpose.

Subcellular Localization of the Cytokines, Basic Fibroblast Growth Factor and Tumor Necrosis Factor- in Mast Cells

77

4.3.1. Basic Fibroblast Growth Factor

bFGF is one of the cytokines recently implicated in mast cell biology [156– 161]. It is one of several cytokines that, for lack of a signal peptide sequence necessary for entry into the RER and thence to Golgi structures and the classical route for secretion, has remained an enigma regarding its cellular release and participation in extracellular events [162]. Some have claimed that bFGF is only released from cells that die [163]. bFGF has been localized in the nuclear matrix, in the cytoplasm, and associated with ribosomes in a variety of cells [164–166]. When immunoreactive bFGF was found in mast cells, it soon was determined that the speciﬁc mRNA for bFGF was also present in mast cells [156, 159–161]. Hence, mast cells have the necessary machinery to synthesize immunoreactive bFGF. Shortly after this, the stimulated release of rat mast cell (RMC) granules containing bFGF was demonstrated by an ultrastructural immunogold analysis [157]. A study using a post-embedding immunogold method to label bFGF after an unmasking step with hyaluronidase reported the presence of granule stores by electron microscopy of bFGF in HSYMCs in patients with rheumatoid arthritis [157], a disorder in which large numbers of synovial mast cells are present [34]. We did a similar study to localize bFGF within partially puriﬁed HMCs obtained from surgically removed lungs [36, 109], remote from tumor areas, in cells that were maintained for 6 h in culture [167]. In this study, we conﬁrmed the granular (ﬁg. 52A, 53A) and nuclear locations of bFGF of previous studies [157, 160, 161, 164, 165]. Also, altered granule matrices typical of a secretory form, termed piecemeal degranulation (PMD) [7; see Chapter 7], had diminished bFGF label and small (80- to 100-nm) perigranular vesicles near these releasing granules contained bFGF (ﬁg. 53B). Cytochemical controls for the immunogold stain were negative (ﬁg. 52B). Mast cell and ﬁbroblast biology are intimately interconnected, with bidirectional effects at play [168]. Mast cells are increased in tissue areas of chronic inﬂammation, ﬁbrosis, and wound healing. bFGF stored in mast cells would be expected to inﬂuence biological responses of ﬁbroblasts in the same compartments in these diseases/processes. To do so would require release/secretion of the cytokine so that it could impact ﬁbroblasts directly. Because bFGF

Fig. 50. bFGF in human mast cell lipid bodies. A The centrally located non-membrane bound lipid body (L) is labeled with both 5- and 20-nm gold. This lipid body is enmeshed in intermediate ﬁlaments; focal smooth endoplasmic reticular elements (arrowhead) are close to the lipid body. Secretory granules (G) adjacent to the lipid body also contain bFGF. B The speciﬁc antibody was omitted. The large lipid body (L) and surrounding secretory granules (G) are not labeled by the secondary antibody alone. A !92,000. B !51,000. [From 137, with permission.]

Ultrastructure of Mast Cells and Basophils

78

Subcellular Localization of the Cytokines, Basic Fibroblast Growth Factor and Tumor Necrosis Factor- in Mast Cells

79

does not contain a signal sequence characteristic of regulated secretory proteins generally, release of bFGF either from damaged or dying cells [163] or by a non-classical secretory route has been postulated [162]. These suggestions pertain to bFGF-producing cells that are not classical secretory cells like mast cells are. The presence of bFGF in classical membrane-bound secretory-storage granules of mast cells documented here and elsewhere [157] clearly places this non-classical secretory protein in the appropriate organelle for exocytosis to the extracellular milieu, given a degranulation stimulus [36, 109, 157]. We have described an alternative form of secretion – termed piecemeal degranulation (PMD) [6, 7] – that operates by emptying individual granules via vesicular transport of ‘small pieces’ of granule contents, ultimately leaving empty granule containers in place in the cytoplasm. We have also shown that these containers can be reﬁlled, a process that was dramatic in the process of wound healing [77]. By far, secretion by PMD from HMCs is the most frequently observed method of granule content release from HMCs in vivo, from all tissue sites, and in all mast cell-rich diseases [55, 169]. In our current study, we show that perigranular vesicles contain bFGF, presumably in transit from nearby granule stores that are visibly depleted of bFGF label [137]. Classically speaking, a secretory product needs a signal sequence in the primary structure deduced from the nucleotide sequence of its cDNA [170, 171] in order to enter cisterns of rough endoplasmic reticula prior to traversing Golgi structures and being sorted to the constitutive secretory or regulated secretory pathways. In the latter case, proteins enter and are stored in secretory granules prior to their secretion from the cell by a regulated event. Since bFGF lacks such a sequence, its secretory route from cells has been in question [162, 163]. Actually, mature HMCs have very little rough cytoplasmic reticulum, and Golgi structures are sparse [16]. So, the synthetic organelles through which a signal sequence-containing protein must travel are not a major feature of mast cells that have the capacity to reconstitute contents in previously emptied granules. This puzzle has intrigued us for years. Subsequently, we have determined that HMC granules contain much of the necessary machinery for Fig. 51. Human lung mast cell paranuclear lipid bodies (L) contain bFGF. A Both 5and 20-nm gold-labeled secondary antibodies label sites of bFGF in the lipid body. Note intermediate ﬁlaments, which merge with the edge of the lipid body and close clusters of ribosomes near the lipid body. The lipid body in A deeply indents the electron-dense nuclear chromatin next to it. B A 5-nm secondary antibody only was used after the bFGFspeciﬁc antibody. Both lipid bodies (L) are extensively labeled. A scroll-ﬁlled secretory granule (G) nearby is also labeled; mitochondria (M) are not. The nuclear matrix reveals bFGF label. Intermediate ﬁlaments enmesh one lipid body (top); the second lipid body shows smooth endoplasmic reticulum imperfectly encircling it (arrow). Some of these labeled with gold particles. A !50,000. B !74,000. [From 137, with permission.]

Ultrastructure of Mast Cells and Basophils

80

Subcellular Localization of the Cytokines, Basic Fibroblast Growth Factor and Tumor Necrosis Factor- in Mast Cells

81

de novo, in situ synthesis [16, 172–175; see Chapter 10]. Thus, we envision a role for these organelles in the actual production of storage materials, so that they may be thought of not only as secretory and storage structures but also as synthetic structures. The immunogold identiﬁcation of bFGF within granules indicates that this route could serve as a bypass synthetic-secretory route for proteins such as bFGF that lack the necessary signal sequence for entry into the classical synthetic-secretory route which utilizes the RER and Golgi structures and typiﬁes classical secretory granule-containing or non-granule-containing cells, such as pancreatic acinar cells and plasma cells. 4.3.2. Tumor Necrosis Factor-

TNF-, a multifunctional, proinﬂammatory cytokine, can be produced by mast cells. We used a microwave energy-assisted aldehyde ﬁxation method to prepare puriﬁed rat peritoneal mast cells for the post-embedding immunogold ultrastructural localization of TNF-. These ﬁxation methods were superior to chemical ﬁxation alone in preserving both the ultrastructural morphology and immunoreactive TNF- in RMCs [134]. The percent of TNF--positive mast cells in samples prepared with microwave-assisted ﬁxation in low (84%) and standard (81%) glutaraldehyde concentrations exceeded that for low (56%) and standard (15%) glutaraldehyde concentrations without the assistance of microwave energy. TNF- was identiﬁed in large storage granules of RMCs (ﬁg. 54). The percent of positive granules in microwave-assisted standard (44%) and low (40%) glutaraldehyde-ﬁxed samples was considerably higher than the percent of positive granules in standard (5%) and low (10%) glutaraldehydeﬁxed samples without microwave assistance. The location of TNF- in rat peritoneal mast cells suggests that this cytokine can use the regulated secretory route(s) for release from appropriately stimulated RMCs into the microenvironment. We used fast (seconds) and ultrafast (milliseconds) microwave energyassisted chemical ﬁxation protocols, post-embedding immunogold staining, and a morphometric analysis to investigate the early morphological changes and the TNF- immunoreactivity in the cytoplasmic granules of rat peritoneal mast cells that had been stimulated to secrete by exposure to compound 48/80. Exposure to compound 48/80 induced the development of increased numbers

Fig. 52. bFGF in human lung mast cell granules. In A, immunogold labeling with a secondary antibody cocktail (gold sizes: 5 and 20 nm) shows bFGF in membrane-bound secretory granules. Both 5-nm (examples are encircled) and 20-nm reporter antibodies label the granules. In B, a control irrelevant antibody, substituted for the speciﬁc primary antibody, does not label secretory granules. A !70,000. B !52,500. [From 137, with permission.]

Ultrastructure of Mast Cells and Basophils

82

Subcellular Localization of the Cytokines, Basic Fibroblast Growth Factor and Tumor Necrosis Factor- in Mast Cells

83

of cytoplasmic granules that exhibited decreased electron density; these granules often also appeared swollen. These granule alterations were accompanied by a signiﬁcantly decreased proportion of granules that were positive for TNF- immunoreactivity. We also calculated the density of TNF- labeling/ m2 in both dense (unaltered) and altered granules in specimens (ﬁg. 55) [135]. TNF- immunoreactivity was present in dense granules (regardless of whether or not the specimens had been stimulated with compound 48/80) and in cells that were ﬁxed with either fast or ultrafast microwave energy. However, altered granules exhibited a decreased density of TNF- label. These ﬁndings show that changes in the immunolocalization and/or density of TNF- immunoreactivity occur very rapidly upon stimulation of rat peritoneal mast cells with compound 48/80.

Fig. 53. bFGF in secretory granules and vesicles of human lung mast cells. A Both of the gold probes (5 and 20 nm) label a scroll granule. B A granule (G) with an altered matrix of less electron density (typical of piecemeal degranulation) than adjacent granules shows less label than the unaltered granules nearby. A single 5-nm gold labeled (encircled) secondary antibody was used. Several electron-lucent, small transport vesicles near the altered granule also contain bFGF (arrows). A !82,000. B !69,000. [From 137, with permission.]

Ultrastructure of Mast Cells and Basophils

84

Subcellular Localization of the Cytokines, Basic Fibroblast Growth Factor and Tumor Necrosis Factor- in Mast Cells

85

54

Ultrastructure of Mast Cells and Basophils

86

55

Subcellular Localization of the Cytokines, Basic Fibroblast Growth Factor and Tumor Necrosis Factor- in Mast Cells

87

Fig. 54. TNF- in rat peritoneal mast cell cytoplasmic granules demonstrated by ultrastructural immunogold analysis. Cytoplasmic granules of rat peritoneal mast cells ﬁxed for 7 s in 0.05% glutaraldehyde-containing ﬁxative with microwave irradiation (A, E, F), for 30 s in 2.5% glutaraldehyde-containing ﬁxative with microwave irradiation (B, D), or by immersion in 0.05% glutaraldehyde-containing ﬁxative for 2 h, 20 ° C, with no microwave exposure (C). The presence of TNF- is indicated in the rat cytoplasmic granule matrix by the presence of gold particles in specimens incubated with anti-TNF- antibody (A–C). Note that gold particles in the rapid microwave-ﬁxed samples (A, B) are more numerous than in the immersion-ﬁxed sample (C). Omission of the speciﬁc primary antibody to TNF- (D) or substitution of non-immune serum for the TNF- antibody (E) yielded no labeling. Prior absorption of the speciﬁc antibody to TNF- with rmTNF- resulted in an 87% reduction in the percent of labeled rat mast cell granules (F) (two gold particles are visible on this granule). Bar = 0.4 m. [From 134, with permission.] Fig. 55. Rat mast cells from control cells (A–C) or from cells stimulated with compound 48/80 for 15 s (D–G). Specimens were prepared with immunogold to demonstrate TNF- (A, B, D–F) or stained with TNF--absorbed speciﬁc antibody to TNF- – i.e., as an absorption control (C, G). (Note: in A, D–F, staining for TNF- was performed with sham-absorbed antibody to TNF-, as a positive control for the TNF- absorption; C, G.) Dense granules contain prominent TNF- immunoreactivity in cells stained with shamabsorbed (A, D–F) or unabsorbed (B) antibody to TNF-, but staining is greatly reduced in dense granules of cells stained with TNF--absorbed-anti-TNF- antibody (C). A swollen, electron-lucent granule in a cell not stimulated with compound 48/80 (arrow in A) has a small amount to TNF- immunoreactivity. Note that mast cells stimulated for 15 s with compound 48/80 (D–G) retain TNF- label in dense granules but not in swollen, electron-lucent granules (D, E). Swollen granules exhibiting a ﬁnely reticular pattern also were devoid of TNF- label (F), whereas the electron-lucent space surrounding them did contain gold (F). TNF--absorbed speciﬁc antibody did not label dense granules or swollen, altered granules above background label in compound 48/80-stimulated cells (G). Circles represent all granule-localized 10-nm gold particles in A–G. A !22,000. B !20,000. C, E !21,000. D !20,200. F !19,200. G !21,300. [From 135, with permission.]

Ultrastructure of Mast Cells and Basophils

88

Chapter 5

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Immunogold Ultrastructural Techniques Identify Subcellular Sites of Chymase, Charcot-Leyden Crystal Protein, and Histamine in Basophils and Mast Cells

5.1. Overview Ultrastructural post-embedding immunogold methods allow the detection of granule products in properly prepared samples which contain mast cells or basophils. Moreover, the probe size is sufﬁciently ﬂexible that small transport vesicles containing these products can be labeled and quantiﬁed. In addition to our studies of cytokine localizations [see Chapter 4] we used these methods to detect immunoreactive chymase (in RMCs), CLC protein (in HBs), and histamine (in RMCs). These methods and their results are reviewed here.

5.2. Chymase Rat peritoneal mast cell chymase is a serine protease that cleaves peptide bonds in the carboxyl side of aromatic amino acid residues [176–182]. Early histochemical evidence suggested that this activity was localized to cytoplasmic granules [183], a concept supported by a large amount of biochemical and cell fractionation data [176–182]. We deﬁned the ultrastructural localization of chymase in rat peritioneal mast cells (ﬁg. 56) using standard aldehyde ﬁxation and a microwave ﬁxation method [184] that provided excellent preservation of subcellular details and retention of chymase that was visualized with a post-embedding immunogold

Fig. 56. Rat peritoneal mast cell, ﬁxed for 7 s by microwave-assisted aldehyde ﬁxation and prepared for the immunogold localization of chymase. Numerous gold particles are present over the granules. Cytoplasm and mitochondria (M) are not labeled. Bar = 0.2 ␮m. [From 152, with permission.]

Ultrastructure of Mast Cells and Basophils

90

procedure [152, 185]. Thin sections were exposed ﬁrst to goat IgG anti-rat chymase and second to gold-conjugated rabbit Ig directed against goat IgG. By transmission electron microscopy, gold particles labeled the matrix of cytoplasmic granules (ﬁg. 56). Control sections treated with non-immune sera or by omission of the speciﬁc primary antibody did not exhibit labeling of mast cells. Thin sections treated simultaneously with puriﬁed RMC chymase and anti-chymase antibody showed marked reduction in granule staining.

5.3. Charcot-Leyden Crystal Protein CLCs were initially described in 1853 by Charcot and Robin [186] in the tissues of a patient with leukemia. In 1872, Leyden [187] also identiﬁed these distinctive bipyramidal hexagonal crystals in the sputum of asthmatics. Since their initial description, CLCs have been associated with eosinophils and are considered a hallmark of eosinophil-associated diseases [188]. The CLC protein, which is the sole protein constituent of CLCs [189, 190], has been puriﬁed to homogeneity, characterized biochemically and immunochemically [189–192] and the cDNA has been cloned [193]. The CLC protein is also a prominent constituent of HBs, which form intracellular CLC under hypotonic conditions [194–196]. Basophil CLC protein is immunochemically indistinguishable from eosinophil CLC protein [194]. CLC protein is abundant in both granulocytes, with eosinophils containing approximately 8 pg/cell [192] and basophils containing approximately 3 pg/cell [194]. We used an immunogold post-embedding ultrastructural procedure to stain the CLC protein (ﬁg. 8–10, 57, 58A–C) [151] as follows: 70-nm sections were placed on gold grids and ﬂoated on 50-␮l drops of reagent at 25 ° C. The following sequence of reagents was used: (a) 4% sodium metaperiodate, 15 min; (b) 3! wash, 10 min each in 0.2 mM Millipore-ﬁltered 20 mM Tris(hyroxymethyl)aminomethane buffer containing 0.9% saline, 0.1% globulin-free bovine serum albumin (BSA), pH 7.6 (TBS-BSA); (c) 5% normal goat serum in TBS-BSA, 1 h; (d) primary rabbit polyclonal afﬁnity chromatography puriﬁed anti-CLC (150 ␮g/ml in TBS-BSA containing 1% Tween-20 and 1% normal goat serum, 2 h at 25 ° C; (e) 3! wash, 10 min each in TBS-BSA; (f) secondary gold-labeled antibody (1:20 dilution of either 10-, 20-, or 30-nm colloidal gold conjugated to goat anti-rabbit IgG in TBS-BSA containing 0.1% Tween-20, 0.4% gelatin and 1% normal goat serum), 1 h; (g) 2! wash, 10 min each, in TBS-BSA, and (h) 2! wash, 10 min each, in distilled water. Speciﬁcity controls included the following alterations of the standard sequence: (1) omission of speciﬁc primary antibody; (2) substitution of non-

Immunogold Ultrastructural Techniques Identify Subcellular Sites of Chymase, Charcot-Leyden Crystal Protein, and Histamine in Basophils and Mast Cells

91

Fig. 57. Mature human basophil developing in a 7-week cord blood culture supplemented with murine 3T3 ﬁbroblast culture supernatant and prepared with immunogold to demonstrate CLC protein. Gold has labeled nuclear and cytoplasmic sites diffusely and almost all granules contain diffuse label as well. Some granules are partially empty, i.e., they have undergone PMD. Bar = 1 ␮m. [From 25, with permission.]

Ultrastructure of Mast Cells and Basophils

92

Fig. 58. A–C Human basophils prepared with an immunogold method to localize Charcot-Leyden crystal (CLC) protein. D–F Panels were prepared with an enzyme-afﬁnity-gold method to label histamine [see Chapter 6]. Gold particles, indicating CLC protein in the central CLCs (C) within the particulate matrix of granules (G) and overlying the particulate matrix, are seen in A and B. Histamine, indicated by gold particles (D, E), resides in the particulate matrix and not in the homogeneous CLCs (C) within these granules (G). The non-particulate primary granules (G) are heavily labeled for CLC protein in C but not for histamine in F. A !70,300. B !56,100. C !57,300. D !57,000. E !85,500. F !66,500. [From 201, with permission.]

Immunogold Ultrastructural Techniques Identify Subcellular Sites of Chymase, Charcot-Leyden Crystal Protein, and Histamine in Basophils and Mast Cells

93

immune normal rabbit IgG (150 ␮l) for the speciﬁc primary antibody, and (3) substitution of solid-phase CLC protein-Sepharose-absorbed primary antibody for unabsorbed speciﬁc primary antibody. Using this immunogold method, we deﬁned the subcellular locations of the CLC protein in quiescent peripheral blood eosinophils [151] and basophils [197], and their activated counterparts in vivo [125, 198], ex vivo basophils [67, 199–201], and in IL-5-induced eosinophils which developed in suspension cultures of human CBC [25, 64, 65, 202]. In addition, we have described the endocytosis of CLC protein and its intralysosomal location in monocytes and macrophages in vivo and in vitro [64, 125, 203]. Although the cellular function(s) of this protein is still obscure, the presence of such large quantities in eosinophils and basophils (ﬁg. 57) suggests an important role in normal or pathological events associated with these lineages. The cloning of the gene for the CLC protein [193] and expression of the protein product [204] aid our understanding of the biological properties of this unique basophil and eosinophil constituent. HMCs, in contrast to basophils, do not stain for the CLC protein by light microscopy [205]. Since we noted an admixture of lineages in suspension cultures of cord blood supplemented with c-kit ligand-containing additives [see Chapter 2], we examined these cells for CLC protein using an immunogold method to detect this protein. We found that mature basophils that differentiated in these long-term cultures contained CLC protein (ﬁg. 57) [25] and noted the subcellular distribution of this protein in relationship to activated phenotypes [see Chapter 2] which the mature basophils assumed over 3–14 weeks of culture. In addition to the previously identiﬁed locations (i.e., intragranular, cytoplasmic, nuclear, vesicular, and formed CLCs within granules, cytoplasm, and nucleus) (ﬁg. 8, 9, 57, 58A–C), CLC protein was also present in Golgiassociated smooth membrane-bound vesicles. In contrast, none of the cells in the mast cell lineage contained CLC protein (ﬁg. 29), thereby providing an excellent immunomorphologic difference for their distinction from basophils arising in the same cultures. The presence of CLC protein (and its absence in mast cells) is a consistent ﬁnding regardless of the source of growth factors used or the individual cord sample from which starting cells were secured or the time in culture [see Chapter 2]. CLCs have been described in HBs before [19, 194, 197, 206]. By ultrastructural analysis, we have noted CLCs within particle granules and free in the cytoplasm or nucleus of basophils [197]. Our initial immunogold studies of the distribution of CLC protein in HBs showed labeled, formed CLCs within granules and rare cytoplasmic vesicles that contained CLC protein [197]. These studies were extended to show that CLC protein is contained within cytoplasmic and nuclear CLCs of activated peripheral blood basophils [67, 199,

Ultrastructure of Mast Cells and Basophils

94

200] and during release and recovery experiments of isolated, puriﬁed peripheral blood basophils, the CLC protein was not released extracellularly [207] but was instead redistributed to cytoplasmic, nuclear, plasma membrane, and vesicular locations [67, 199, 200]. In particular, basophils that displayed the morphological phenotype of PMD had increased diffuse labeling of cytoplasm and nucleus, as well as labeled plasma membranes [67]. Later, after recovery, labeling of these locations diminished and, again, the granular compartment became predominantly labeled (ﬁg. 58A–C) [67; see Chapter 9]. We also noted the appearance of a heavily labeled, newly evident, homogeneously dense granule population (ﬁg. 58C) [67] that resembled the CLC protein-containing primary granules of eosinophils [151]. The distribution of CLC protein in the basophils that differentiate in c-kit ligand-supplemented suspension cultures of cord blood [25] resembled that described before in peripheral blood basophils [197]. As in the stimulated secretory model described above [67, 200, 206], we found increased amounts of diffuse cytoplasmic, nuclear, and plasma membrane label in cells expressing the PMD phenotype [25]. At later times in cultures in which some basophils had recovered granule contents, we noted increased numbers of CLC proteinpositive particle granules, intragranular CLCs, and CLCs free in the cytoplasm and nucleus [25]. Also noted were increased numbers of homogeneously dense, heavily labeled granules [25]. Occasionally, we found Golgi area vesicles that were positively labeled for CLC protein.

5.4. Histamine Histamine released by mast cells plays an important role in immediate hypersensitivity reactions and in other inﬂammatory processes [reviewed in 208]. Several lines of evidence, including differential centrifugation [178, 209], biochemical determinations [208, 210], and autoradiographic [211] or immunohistochemical [212, 213] studies, indicate that histamine is associated with mast cell secretory granules. We used a post-embedding immunogold approach to demonstrate the ﬁne structural localization of histamine in the granules of unstimulated rat peritoneal mast cells that were ﬁxed either by standard aldehyde ﬁxation or by a microwave-aldehyde ﬁxation method [152, 153, 184]. For histamine detection, post-embedding, immunogold staining methods were done as follows: 70-nm sections of unstimulated cells on nickel grids were ﬂoated section side down on 30-␮l drops of reagents at 20 ° C. The following reagents were used in sequence: (a) 10% sodium metaperiodate for 0, 5, 15, 30, or 60 min to solubilize osmium and unmask antigens; (b) wash in 20 mM

Immunogold Ultrastructural Techniques Identify Subcellular Sites of Chymase, Charcot-Leyden Crystal Protein, and Histamine in Basophils and Mast Cells

95

Tris buffer containing 0.9% saline, 0.2% non-immune goat IgG, pH 7.6 [TBSIg); (c) 5% non-immune goat IgG for 20 min; (d) primary antibody [guinea pig anti-histamine-BSA antiserum (Peninsula Labs, Belmont, Calif., USA), or rabbit anti-histamine-BSA antiserum (Milab, Malmo, Sweden, and Chemicon, El Segundo, Calif., USA), used neat or diluted up to 1:100, in TBS-Ig, 0.1% Tween-20, 1% normal goat serum] for 1 h at 20 ° C or for 24 h at 4 ° C; (e) wash 3!, 10 min each, in TBS-Ig; (f) secondary 5- or 10-nm gold-conjugated goat IgG directed against guinea pig or rabbit IgG (diluted 1:20–1:80 in TBS-Ig, 0.1% Tween-20, 1% normal goat serum) for 1 h at 20 ° C, and (g) wash 2! in TBS-Ig, and wash 2! in distilled water. Membrane contrast was enhanced with 0.25% lead citrate for 3 min. Controls included the following: (a) omission of primary antibody and (b) substitution of normal guinea pig IgG for the speciﬁc guinea pig anti-histamine-BSA antiserum. The speciﬁcity of the antibody-antigen reaction was further established by doing absorption and blocking controls. We absorbed the guinea pig anti-histamine-BSA antiserum with histamine bound to agarose (160–320 ␮l/ml) or to agarose alone for 1–3 h at 20 ° C. After centrifugation in a microfuge for 1 min at 1,000 rpm at 20 ° C, the supernatant was used in place of the primary antibody in the labeling experiments. For a blocking control, puriﬁed histamine hydrochloride (1–100 mg/ml) was incubated with guinea pig anti-histamineBSA antiserum (ﬁnal dilution 1:50 in TBA-Ig-Tween) for 1 h at 20 ° C and then the sections on the grids were added to the droplet for simultaneous incubation with the histamine hydrochloride and the primary antibody for 1 h at 20 ° C. By transmission electron microscopy, gold particles indicating the presence of histamine were localized to the matrix of cytoplasmic granules (ﬁg. 59). Control sections, including omission of the primary antibody, substitution of normal guinea pig IgG for the speciﬁc antibody, and exposure of the anti-histamine antiserum to either puriﬁed histamine or histamine bound to agarose were negative, or, in the histamine absorption, showed a signiﬁcant reduction (p ! 0.005) in granule staining [153].

Ultrastructure of Mast Cells and Basophils

96

Fig. 59. Post-embedding immunogold labeling for histamine in rat mast cell granule. Bar = 0.2 ␮m.

Immunogold Ultrastructural Techniques Identify Subcellular Sites of Chymase, Charcot-Leyden Crystal Protein, and Histamine in Basophils and Mast Cells

97

Chapter 6

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Ultrastructural Enzyme-Affinity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

6.1. Overview Enzyme-afﬁnity-gold methods have been used at the ultrastructural level to image enzyme substrates in tissues based on afﬁnity to their speciﬁc enzymes [214]. The ﬁrst example of this class of post-embedding methods was reported for the detection of RNA with ribonuclease (RNase)-gold (R-G) [215]. We reported that this reaction was not completely speciﬁc for RNA based on our observation that heparin-rich organelles, e.g. mast cell granules, avidly bound the R-G reagent [155]. These studies determined that the well-known biochemical property of heparin to inhibit RNase acted as an inhibitor-gold method and bound the R-G reagent, thus allowing visualization of heparin in ultrastructural samples [155]. After establishing and validating the speciﬁcity of this method, we used it to detect changes in heparin stores in HMCs during AND and recovery in short-term cultures of isolated HLMCs [216; see Chapter 9]. We next developed a new enzyme-afﬁnity-gold method to detect histamine to study the changing histamine stores in basophils and mast cells in vivo, ex vivo and in vitro [154; see Chapters 7, 9].

6.2. Histamine Histamine was ﬁrst synthesized in 1907 [217] and has been known to be associated with mast cells since the seminal observations of Riley and West [218]. Light microscopic methods to image histamine-containing mast cells are available [219, 220] and selective localization of histamine in electron-dense secretory granules of RMCs with either ultrastructural autoradiography [221] or an ultrastructural immunogold method [153; see Chapter 5] has been reported. HMCs are a good source material to test methods designed to demonstrate histamine at the ultrastructural level, since mast cells are known to contain large amounts of granule histamine (average, 3.74 pg/cell, in freshly isolated HLMCs [36]), and preservation of their varied granule patterns for ultrastructural studies offers a good challenge to methods designed to do so. 6.2.1. Diamine Oxidase-Gold (DAO-G) Method to Image Histamine

A colloidal suspension of gold was prepared according to the method of Frens [222]. Four milliliters of an aqueous 1% solution of sodium citrate was added to a boiling aqueous solution of 100 ml 0.01% tetrachloroauric acid and allowed to boil for 5 min before cooling on ice. The pH of the colloidal gold suspension so produced was adjusted to 7 with 0.2 M potassium carbonate. Preparation of the diamine oxidase-gold (DAO-G) complex was according to the method of Bendayan [215]. Three milligrams of DAO were dissolved in distilled water and placed in a polycarbonate ultrafuge tube with 10 ml of the colloidal gold suspension. The mixture was centrifuged at 25,000 rpm for 30 min at 4 ° C, in a Beckman ultracentrifuge with a #50.2 Ti rotor. The DAO-G complex formed a red sediment that was carefully recovered and re-suspended in 3 ml 0.1 M PBS containing 0.02% polyethylene glycol, pH 7.6 (ﬁnal concentration 1 mg DAO/ml). For cytochemical labeling, section-containing grids were inverted and ﬂoated on PBS drops for 5 min, followed by incubation on a drop of DAO-G at 37 ° C for 60 min. The grids were vigorously washed in distilled water and stained with dilute lead citrate for 10 min before viewing by electron microscopy. The effect of pH on preparation of the DAO-G complex was tested by preparing the reagent at different pH values, from 5.2 to 8.4, before selecting pH 7 as the routine preparation value for DAO-G. Similarly, the staining pH was varied from 6 to 8 before selecting pH 7.6 as the optimal labeling value for histamine, a value close to that optimal for the enzymatic reaction of DAO [223; and manufacturer’s literature]. The effect of temperature on labeling was assessed at 20, 37, and 70 ° C and, as expected, granule histamine labeling with

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

99

DAO-G increased with temperature. Background also increased, however, and 37 ° C was selected as the standard incubation temperature for labeling. We did a number of controls to assess the speciﬁcity of cytochemical labeling of HMC granules with DAO-G: (a) boiling of DAO-G for 10 min before use for labeling; (b) incubation of grids with colloidal gold reagent only; (c) observation of type II alveolar pneumocyte (present in the same samples of lung cells) secretory granules, lamellar bodies, which contain surfactant, not histamine; (d) preparation of agar test blocks containing a variety of compounds by ﬁxation, processing and embedding identical to that used for mast cells, and subsequent labeling of agar test block thin sections with DAO-G by identical procedures as for tissue blocks. These compounds included histamine (8 mg/150 ␮l agar), RNA (2.5 mg/150 ␮l agar), polyuridine (1.5 mg/150 ␮l agar), and polyadenine (3 mg/150 ␮l agar); (e) pre-incubation of the DAO-G reagent in solution, or of grids, with histamine before labeling grids with DAO-G; (f) pre-incubation of grids with heparin before labeling grids with DAO-G; (g) labeling of grids with DAO-G that had been passed over a variety of materials packed into Pasteur pipette columns (these included: glass wool, Sepharose beads, heparin-agarose beads, histamine-agarose beads). We also examined labeling after ﬁltering colloidal gold only over such beads as well as labeling with DAO-G ﬁltered over heparin-agarose, histamine-agarose, or Sepharose beads that had previously been ﬂushed with histamine; (h) incubation of grids for 60 min at 37 ° C in DAO, washing, and labeling with DAO-G, and (i) incubation of grids for 60 min at 37 ° C in heparinase III (Sigma), heparinase or heparinase II, washing, and labeling with DAO-G. 6.2.2. DAO-G Labels Human Mast Cell Secretory Granules

All substructural types of granules that are regularly present in HLMCs were labeled with DAO-G (ﬁg. 60–63); particle granules and particle-containing portions of mixed granules labeled with somewhat less intensity, however. Swollen, partially empty granules (ﬁg. 60A) and granules with remaining lightly dense altered matrix either did not label or were markedly reduced in labeling. Immature granules with central condensation foci [224] were sometimes devoid of label (ﬁg. 60B). DAO-G did not label lamellar bodies (ﬁg. 60L, 62B), the surfactant containing secretory granule of the type II alveolar pneumocytes, that were present in the same sections as the HLMCs, thereby providing a good internal speciﬁcity control for each sample. The nuclear heterochromatin of each cell type was regularly labeled with DAO-G (ﬁg. 63).

Ultrastructure of Mast Cells and Basophils

100

6.2.3. Enzyme-Afﬁnity Cytochemical Controls

The labeling obtained with DAO-G was tested in a number of controls. In general, controls that abrogated labeling of histamine-containing HMC granules (ﬁg. 63, 64A, 65A) included: (a) inactivation of the enzymatic activity of DAO by boiling the DAO-G reagent (ﬁg. 65I) which retained its typical red color; (b) speciﬁc removal of the labeling capacity of DAO-G by ﬁltering it over solid-support bound substrate, i.e., histamine-agarose beads (ﬁg. 64C, 65B, 66) – the histamine-agarose retained all visible red DAO-G (ﬁg. 67) and the efﬂuent used for labeling was clear, and (c) digestion of the section on the grid with DAO before staining with DAO-G (ﬁg. 65H, 68). Colloidal gold alone, or after it was ﬁltered over histamine-agarose beads, heparin-agarose beads (ﬁg. 65F) or Sepharose beads (ﬁg. 65D), did not label mast cell granules. Nuclear heterochromatin was also abrogated by these controls (ﬁg. 68). DAO-G, when passed over glass wool alone (used to pack Pasteur pipettes to hold various beads), did not lose its labeling capacity, and this process generally removed larger, unwanted gold aggregates from the reagent. DAO-G that was ﬁltered over Sepharose beads (ﬁg. 64A, 65C) or heparin-agarose beads (ﬁg. 64B, 65E) also retained granule labeling capacity, and the individual bead preparations remained white (ﬁg. 67); the efﬂuent used for labeling was red. When Sepharose beads, heparin-agarose beads, or histamine-agarose beads were ﬂushed with histamine before ﬁltering DAO-G, the histamine-agarose beads turned red, the efﬂuent was clear, and mast cell granules did not label with DAO-G; Sepharose beads remained white, the efﬂuent was red, and labeling of mast cell granules occurred; heparin-agarose beads remained white, the efﬂuent was red, and the granules were labeled with extra intensity. Increased label intensity also occurred sometimes when soluble histamine was pre-incubated with DAO-G before incubating the grid or the grid was pre-incubated with histamine before DAO-G. In each case, background also increased, clearly showing the need to use solid-support substrate to successfully block the activity of the enzyme-gold complex. In addition, the presence of free histamine in any of the controls acts to bind non-complexed DAO that might digest substrate and/or compete for binding sites with the labeled reagent, thereby potentially increasing the efﬁciency of label with DAO-G. Pre-incubation of grids with heparin, heparinase III, heparinase II, or heparinase (ﬁg. 65G) before DAO-G incubation did not abolish the speciﬁc granule labeling. Agar blocks were prepared exactly as were the cell samples for electron microscopy after incorporating histamine or a variety of macromolecules into molten agar and cross-linking these materials with aldehyde ﬁxatives. DAO-G was subsequently used to stain thin sections of the agar blocks. The histamine-agar block labeled with DAO-G (8.38 particles/in2); RNA-, polyuridine-, polyadenine-containing agar blocks and agar alone were negative. The

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

101

Ultrastructure of Mast Cells and Basophils

102

histamine-agar block labeled poorly with gold only (0.06 particles/in2) or with an irrelevant enzyme-gold complex, R-G (0.04 particles/in2), prepared according to Bendayan [155, 215]. 6.2.4. Comment regarding a New Ultrastructural Method to Detect Histamine

We described a new enzyme-afﬁnity-gold technique for ultrastructural cytochemistry, based on the original work of Bendayan [215]. We used DAO (histaminase) to image histamine in HMC granules in partially puriﬁed, short-term cultures of HLMCs (which are rich in histamine) as a model system. The results clearly showed that histamine-rich, membrane-bound secretory granules [218–221, 225] are extensively and speciﬁcally labeled. The method works on aldehyde-ﬁxed, osmicated, uranyl en bloc-stained cells dehydrated in alcohol and embedded in Epon, thereby providing the excellent preservation of ﬁne ultrastructural detail regularly associated with material prepared in this way. The optimal temperature, time and pH for staining were determined to coincide with those known for the optimal enzymatic action of DAO, and the ability to stain mast cell granules was abrogated by boiling the red DAO-G reagent before use, a procedure known to destroy the enzymatic activity of DAO [223]. The DAO-G remained red after boiling, indicating that the enzyme-gold complex was intact. Dissociation of enzyme-gold complexes is accompanied by a color change to purple when protein ﬂocculates and disperses from the red colloidal gold, as described [215, 222]. Therefore, the binding basis for this new technique to image histamine at the ultrastructural level is the afﬁnity of enzyme (DAO) and substrate (histamine) as reported for other enzyme-afﬁnity ultrastructural techniques [214].

Fig. 60. A–K DAO-G labels human mast cell granules but not type II alveolar pneumocyte lamellar bodies (L) in the same grid. A, B Full-scroll granules are labeled for histamine but a swollen scroll granule with partial loss of granule materials (arrowhead, A) and a small condensing granule with a central dense nucleoid (arrowhead, B) are not labeled. C A mixed granule with empty, longitudinally sectioned scrolls and several round, dense aggregates shows label associated with the lamellae of scrolls and the dense aggregates. Electron-lucent areas are not labeled. D–F Scroll granules cut in cross section (D, E) and longitudinal section (F) are labeled for histamine. G A scroll granule cut longitudinally contains ﬁnely granular dense material within the centers of individual scrolls (compare with empty longitudinal scrolls in C); the granule in G is heavily labeled. H, I Crystal granules showing regular parallel arrays are labeled for histamine. J Reticular granules, which contain beaded-thread arrays, contain histamine. K A granule containing an irregular, single thick thread shows gold particles attached to this reticular thread with in an otherwise electron-lucent membrane-bound granule. L A type II pneumocyte lamellar body in the same sample is not labeled with DAO-G. Bars: A, B, J 0.2 ␮m; C–I, K, L 0.1 ␮m. [From 154, with permission.]

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

103

Fig. 61. A membrane-bound, full scroll granule (A) and a depleted scroll granule of human mast cells (B), seen in cross section, are labeled with DAO-G, indicating the presence of histamine; non-membrane-bound lipid bodies (L) are not labeled. Bars = 0.2 ␮m. [From 154, with permission.]

Extensive controls were done to verify the speciﬁcity and practicality of the new enzyme-afﬁnity method for histamine. These fall into several general categories: (a) abrogation of label by speciﬁc substrate (histamine) or speciﬁc enzyme activity (DAO); (b) staining of a speciﬁc test model (histamine-agar block); (c) evaluation of non-speciﬁc sticking of colloidal gold; (d) retention of labeling after exposure of DAO-G to a variety of solid-support vehicles;

Ultrastructure of Mast Cells and Basophils

104

Fig. 62. A Human mast-cell granules are heavily labeled with DAO-G. B A cluster of lamellar bodies in an adjacent type II alveolar epithelial cell in the same grid is not labeled. Bars = 0.2 ␮m. [From 154, with permission.]

(e) evaluation of the impact of heparin (another mast cell granule constituent) and heparinase on labeling, and (f) evaluation of label of another secretory granule, the phospholipid-rich lamellar body of type II alveolar pneumocytes, cells that were present in the same preparations of HLMCs used for this study. Prior digestion of the thin sections with DAO before staining with DAOG abrogated mast cell granule staining; prior digestion with three irrelevant enzymes (heparinase, heparinase II, heparinase III) did not abrogate labeling of mast cell granules. Our initial attempts to abrogate staining by blocking with soluble-speciﬁc substrate did not succeed. For example, neither pre-incubation of sections, or the DAO-G reagent, with soluble histamine, nor incubation of grids with DAO-G that was passed over a variety of histamine-ﬂushed solid beads reliably resulted in diminution or absence of mast cell granule labeling. In some instances, labeling actually improved, perhaps because soluble substrate effectively bound any excess, dissociated, or weakly associated enzyme in the enzyme-gold reagent that would compete with the enzyme-gold complex. We

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

105

Fig. 63. The DAO-G method labels two sites in this HMC: cytoplasmic granules and condensed nuclear chromatin. Non-membrane-bound, homogeneously structured lipid bodies (L) and cytoplasm are not labeled. Bar = 0.6 ␮m. [From 154, with permission.]

Ultrastructure of Mast Cells and Basophils

106

Fig. 64. Controls for DAO-G method. DAO-G ﬁltered over Sepharose beads (A) or heparin-agarose beads (B) still extensively labels human mast-cell granules. DAO-G ﬁltered over histamine-agarose (C) no longer labels mast-cell granules. Bars = 0.4 ␮m. [From 154, with permission.]

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

107

Fig. 65. Montage of human mast cell granules shows gold-labeled (A, C, E, G) and unlabeled (B, D, F, H, I) granules with the DAO-G procedure. A DAO-G only; B DAO-G ﬁltered over histamine-agarose; C DAO-G ﬁltered over Sepharose beads; D gold only ﬁltered over Sepharose beads; E DAO-G ﬁltered over heparin-agarose; F gold only ﬁltered over heparin-agarose; G digestion of grid with heparinase (EC 4.2.27) before staining with DAO-G; H digestion of grid with diamine oxidase (histaminase) before staining with DAO-G; I heat inactivation of DAO-G by boiling for 10 min before use. Bars = 0.1 ␮m. [From 154, with permission.]

Ultrastructure of Mast Cells and Basophils

108

Fig. 66. DAO-G was ﬁltered over histamine-agarose beads before staining this mast cell. Granules no longer label after the DAO-G reagent was absorbed with solid-phase substrate (compare with ﬁg. 63). Bar = 0.6 ␮m. [From 154, with permission.]

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

109

Fig. 67. Photograph of Pasteur pipette columns ﬁlled with cotton wool and various beads through which DAO-G was ﬁltered before staining thin sections of human mast cells on grids. A DAO-G ﬁltered over Sepharose; beads remain white (reagent was red and mast cell granules were labeled). B DAO-G ﬁltered over histamine-agarose; beads are red (reagent was clear and mast cell granules did not label). C Histamine was ﬂushed over Sepharose before DAO-G ﬁltration; beads remain white (reagent was red and mast cell granules labeled). D Histamine was ﬂushed over histamine-agarose before DAO-G ﬁltration; beads are red (reagent was colorless and mast cell granules did not label). E Histamine was ﬂushed over heparin-agarose before DAO-G; beads remain white (reagent was red and mast cell granules labeled, indicating that neither histamine nor DAO-G bound to heparin-agarose). Also, DAO-G ﬁltered over heparin-agarose did not bind to heparin-agarose but did label mast cell granules (data not shown). [From 154, with permission.]

Ultrastructure of Mast Cells and Basophils

110

Fig. 68. Digestion of the grid containing this mast cell with DAO (histaminase) was done before staining with DAO-G. DAO-G no longer labels cytoplasmic granules and nuclear chromatin (compare with ﬁg. 63). Bar = 0.6 ␮m. [From 154, with permission.]

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

111

next used solid-support substrate (histamine-agarose) to block DAO-G staining of HMC granules and effectively blocked all label. Moreover, the histamineagarose served as an excellent visual indicator of effective blocking because the bound, red DAO-G was all retained on the beads, turning them red; the efﬂuent was clear, and mast cell granules did not stain. We also passed the red DAO-G over glass wool (necessary to retain the various beads in the Pasteur pipettes used), Sepharose beads, and heparin-agarose, and in each instance, the glass wool or beads did not acquire the red coloration of DAO-G; the efﬂuent was clear and mast cell granules stained. (In some experiments, focal clusters of red appeared in the glass wool, and the resultant reagent used to stain grids contained fewer unwanted aggregates.) We developed a histamine-containing agar block model to conﬁrm the speciﬁcity of the DAO-G complex. Histamine was incorporated into molten agar, ﬁxed in aldehydes, osmicated, stained en bloc with uranyl, dehydrated in alcohols, embedded in Epon, sectioned, and stained exactly as the mast cell preparations. The histamine-agar block stained with DAO-G (8.38 particles/in2) but not with colloidal gold only (0.06 particles/in2) or with an irrelevant enzyme-gold complex (0.04 particles/in2). Other test blocks of agar alone or agar blocks containing a variety of macromolecules did not stain with DAO-G. Thus, the speciﬁcity for histamine of the enzyme-gold complex DAO-G was established. In all of the experiments, a colloidal gold-only control was incorporated, and non-speciﬁc staining with each newly prepared batch of colloidal gold did not occur. Nor did staining occur when the colloidal gold was passed over a variety of solid supports including histamine-agarose. Heparin, a major constituent of HMC granules, did not block DAO-G staining, whether the reagent or the grid was pre-incubated with heparin or whether DAO-G was passed over a solid heparin-Sepharose support. Nor did the heparin-speciﬁc enzyme, heparinase, abrogate DAO-G staining of HMC granules. A phospholipid-rich secretory granule (of type II pneumocytes), present in each preparation tested, did not stain, providing an internal negative control for this procedure. In sum, the DAO-G enzyme-afﬁnity technique is speciﬁc for the demonstration of histamine, as shown by the following key facts and controls necessary to validate a new enzyme-afﬁnity-gold procedure: (a) HMC granules, known to contain histamine, are positive; type II pneumocyte phospholipidrich lamellar bodies, which do not contain histamine, are negative; (b) histamine-containing test agar blocks are positive; (c) exposure of DAO-G to speciﬁc substrate histamine (in solid form) effectively abrogates staining; (d) digestion of the sample with speciﬁc enzyme, histaminase, abrogates staining, and (e) a wide variety of controls for the speciﬁcity tests did not abrogate staining of HMC granules or the histamine-agar test blocks.

Ultrastructure of Mast Cells and Basophils

112

In addition to staining HMC granules, the DAO-G complex stained heterochromatin in mast cells as well as other cells, in concert with its known activity against the substrate putrescine [226, 227], a naturally occurring polyamine [228]. Putrescine forms covalent links to bacteriophage DNA [229, 230], and in some mammalian cells, polyamines affect the initiation of DNA synthesis [231–233]. Whether polyamines are actually in the nucleus in vivo has been examined by cell fractionation studies, which are plagued with redistribution possibilities. However, with rapid fractionation procedures, a nuclear location seems likely [234], although some authors prefer to state that the nuclear location of polyamines is not yet certain [228]. The enzyme-afﬁnity-gold staining of human cell heterochromatin we reported is not subject to the criticism of redistribution, nor are nuclei expected to contain histamine. Therefore, the nuclear localization of putrescine, a substrate for DAO and a polyamine known to interact with DNA, is suggested by the enzyme-afﬁnity-gold method we reported [154]. The initial report of an oxidative deaminating enzyme which inactivated histamine appeared in 1929 by Best [235], and the name histaminase was suggested for the enzyme responsible for the loss of the physiological activity of histamine [236]. Zeller [237] introduced the term diamine oxidase in 1938, based on the expanded substrate activities of pig kidney histaminase to putrescine and cadaverine. Although there has been controversy regarding the putative unity of these enzymes [226, 227, 238], it has been clearly established that they are indeed the same enzyme [227], and the terms diamine oxidase and histaminase are used interchangeably. Histamine was synthesized in 1907 [217]. In humans, histamine is characteristically present in the secretory granules of two cells, mast cells and basophils, and is released with rapid kinetics by regulated secretion [reviewed in 7]. Both of these granulated secretory cells also undergo a secretory process, termed piecemeal degranulation (PMD) that characteristically empties granules in place, leaving their containers intact [reviewed in 7; see Chapter 7]. Tissue mast cells and circulating basophils that enter tissues in a wide variety of circumstances release granule materials in this fashion [reviewed in 169]. We have suggested that vesicular transport may be the principal mechanism by which these granules losses are affected. For example, it is possible to load basophil and mast cell granules with eosinophil peroxidase from the exterior, a process that occurs by vesicular transport [239]. Imaging histamine-loaded vesicles during stimulated release reactions of mast cells and basophils would provide absolute proof that such a mechanism for secreting stored granule contents exists. The enzyme-afﬁnity-gold technique described here provides specific localization of histamine in ultrastructural preparations with sufﬁcient clarity to determine whether (or not) such transport vesicles exist [see Chapter 7].

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

113

Using the DAO-G method, we established that all of the varied, ultrastructurally deﬁned granule types in HMCs do indeed contain histamine. The particle granule usually labeled with less intensity than granules with other substructural patterns, a ﬁnding of interest since the major granule type in HBs also contains particles and labels less intensely than HMC non-particle granules [104].

6.3. Heparin Enzyme-afﬁnity-gold ultrastructural cytochemistry was initiated in 1981 when RNase was coupled to colloidal gold particles and used to label intracellular sites of RNA [215]. Since then, R-G has been used for this purpose in many studies [214, 240–248] and the principle extended to the identiﬁcation of other enzyme substrates by constructing new enzyme-gold probes for this purpose. Among the substrates imaged at high magniﬁcation in optimally preserved ultrastructural samples are DNA [214, 215, 240–242], collagen [214, 241, 242], elastin [214, 241, 242], glycogen [214, 241, 242], hyaluronic acid [242, 247], sialic acid [242, 247], glucosides [242, 244], histamine [29, 104, 154, 248–253], mannose [242], pectin [242], and phospholipid [242]. The initial study with R-G used rat pancreatic acinar cells (RPACs) to image RNA [215]. Precise and speciﬁc localization was reported in ribosomes and nucleoli, structures known to contain RNA. One of the necessary speciﬁcity controls for enzyme-afﬁnity cytochemistry is to show that binding is diminished or eradicated when inhibitors of the enzyme in question are used [215]. For RNase, one well-known inhibitor is heparin [254–257]. HMC granules store proteoglycans which have been shown primarily to consist of heparin (and smaller quantities of chondroitin sulfate E) [258, 259], a macromolecule said to be the most anionic substance in the body [260, 261] and to be unique to mast cell granules [254, 261]. Historically, the anionic charge of heparin and heparin-protein granule complexes is thought to be the basis for dye binding that facilitates the recognition of mast cells [262–264], as well as to provide ‘false positives’ in imaging techniques based on highly positively charged avidin, in light microscopic preparations [265, 266]. It is thought that this unique property of heparin serves to inhibit RNase in biochemical studies [254–257]. We thought that these properties and the known afﬁnity of heparin for RNase might prove useful in imaging heparin, using optimally prepared electron microscopic samples. We selected the HMC granule to study ﬁrst, since it is a unique site of large amounts of heparin and since it should not contain

Ultrastructure of Mast Cells and Basophils

114

RNA, the substrate for which the R-G method was developed [155, 216]. We also examined R-G labeling in guinea pig basophil granules [267], since subcellular isolation and biochemical analysis of their single granule population shows that they do not contain heparin but do primarily contain chondroitin sulfate [268, 269]. Cytoplasmic granules are membrane-bound organelles which subserve storage, processing, secretory and digestive functions, depending on the cell type within which they reside. Most granules are rendered electron-dense with standard ultrastructural methods; some have variegated patterns, and they vary in size from ⬃100 nm to several micrometers in diameter in standard thin sections. While none of the cell types containing these granules contains heparin (with the exception of the mast cell), most of them contain other proteoglycans. We examined a number of cell lineages with cytoplasmic granules (using the R-G method) to determine the potential universality of imaging proteoglycans in their subcellular locations. These studies reveal a wider application of the R-G method, i.e., as an inhibitor which binds to proteoglycans in ultrastructural samples. 6.3.1. Ribonuclease-Gold (R-G) Method to Image Heparin

Electron microscopic samples of tissue and isolated, puriﬁed and/or cultured mast cells (human, rat, mouse), basophils (human, guinea pig, rabbit), puriﬁed guinea pig basophil granules, eosinophils (human, guinea pig), cloned, granulated mouse lymphocytes, human platelets, guinea pig megakaryocytes, neutrophils (human, mouse, guinea pig, rat), human type II alveolar pneumocytes, human endothelial cells, guinea pig Kurloff cells, rat pancreas acinar cells, glucagon-secreting cells of the human pancreas islet, calcitonin-secreting cells of the human thyroid, neuroendocrine cells of human lung, ileum and stomach, prolactin- and growth hormone-secreting cells of human pituitary, and macrophages (human, guinea pig, mouse) were examined. The samples were ﬁxed routinely in a mixture of aldehydes, either post-ﬁxed in collidine-buffered osmium tetroxide, in potassium ferrocyanide-reduced osmium, or not treated with osmium, stained en bloc in uranyl, dehydrated in a graded series of alcohols, inﬁltrated in a propylene oxide-Epon sequence and embedded in Epon 812, which was polymerized at 60 ° C for 16 h, all as previously reported [270]. Thin sections were cut with a diamond knife and were placed on gold or nickel grids and then air-dried before staining. A colloidal gold suspension [222] was prepared as follows. Four milliliters of a 1% aqueous solution of sodium citrate were added to 100 ml of a boiling aqueous solution of 0.01% tetrachloroauric acid and boiled for 5 min. The R-G complex was prepared according to the method of Bendayan [215]. Briefly, the colloidal gold suspension was adjusted to pH 9.0 with 0.2 M K2CO3,

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

115

and 10 ml of this solution were added to a siliconized tube containing 0.1– 0.7 mg RNaseA dissolved in 0.1 ml distilled H2O. The mixture was centrifuged at 25,000 rpm for 30 min at 4 ° C, using a Beckman 50.2 Ti rotor. The dark red sediments of the reagent (and one of colloidal gold only) were separately resuspended in 3 cc of phosphate-buffered saline (PBS), pH 7.5, containing 0.2 mg/ml of polyethylene glycol. Grids containing sections were inverted and ﬂoated section-side-down on a PBS drop for 5 min, followed by incubation on a drop of R-G for 1 h at 37 ° C, pH 7.5. The grids were washed sequentially in PBS and distilled water and stained with lead citrate for 3–10 min prior to examination by electron microscopy. 6.3.2. Inhibitor-Gold Cytochemical Controls

A number of speciﬁcity controls were done [155, 216, 267]. Samples were stained with the uncomplexed colloidal gold suspension alone, with an irrelevant protein-gold complex (BSA-gold), with an irrelevant enzyme-gold complex [deoxyribonuclease-(DNase)-gold], or with an inactivated R-G complex that had been heated at 100 ° C for 10 min. Agar blocks, containing macromolecules [RNA (16 mg/ml), heparin (20–30 mg/ml), polyuridine (poly(U), 10 mg/ml), polyadenine (poly(A), 20 mg/ml), chondroitin sulfate (CS, 26 mg/ ml), and histamine (53 mg/ml)], were ﬁxed and processed as tissue blocks and stained with R-G, or with gold suspension alone. Exposure of the substrate in samples and the R-G reagent to a variety of enzymatic or acidic digestions was done. These included incubations of sections en grid or the R-G reagent with RNase (0.5–1.0 mg/ml), DNase (1 mg/ml), heparinase I (1 mg/ml), histaminase (2.2 mg/ml), proteinase K (0.25 mg/ml), pronase E (0.5 mg/ml), or 0.1 N hydrochloric acid (HCl). In some cases, incubations of samples with enzymes together or in sequential order preceded staining with R-G. Cationized ferritin (CF) (0.5 ml in 10 ml Hanks’ balanced salt solution) and cationized colloidal gold (poly-L-lysine bound to 10 nm gold) were used to stain cells directly in sections placed on grids. Some samples that were stained with either CF or cationized gold (CG) en grid were then stained with R-G. Additional samples were studied in which either the grids carrying them or the enzyme-gold reagent were incubated with the following reagents either individually or in various combinations: poly-L-lysine (0.5 mg/ml), BSA (1 mg/ml), human serum albumin (5%), normal goat serum (5%), CS (3 mg/100 ␮l), DNA type I (1 mg/ml), ␥-globulin (1 mg/ml), polyvinyl sulfate (0.1 mg/ml), protamine (1 mg/ml), guanidine HCl (1 mg/ml), sodium arsenate (0.001 or 0.002 M) histamine (1–2 mg/ml), RNA (1 mg/ml), and heparin (1–2 mg/ml). Finally, samples were stained en grid with R-G previously absorbed with heparin-agarose beads or Sepharose beads, with or without additional exposure of the R-G reagent

Ultrastructure of Mast Cells and Basophils

116

Fig. 69. HSMC in vivo stained with R-G shows gold labeling of cytoplasmic granules, nuclear chromatin, and nucleolus. Bar = 0.8 ␮m. [From 289, with permission.]

to RNA or heparin, or exposure of the heparin-agarose or Sepharose beads to RNase before adding R-G to the beads. The results of these speciﬁcity controls were quantitated for mast cell granules and expressed as density of gold particles/mm2 of granule area [155].

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

117

6.3.3. RNase-Gold-Labeled Sites in Human Mast Cells

HMCs stained with R-G revealed gold-labeled granules, nuclei, nucleoli, and ribosomes (ﬁg. 69). These labeled sites were evident in HMCs of diverse origin, including HLMCs in vivo and after isolation, puriﬁcation ex vivo, and short-term culture intervals in vitro and HSMCs in vivo (ﬁg. 69) and after isolation and puriﬁcation ex vivo, and development of HMCs in vitro from agranular precursors in human CBCs that were cultured with rhSCF [23]. HMCs have large, electron-dense secretory granules with a variety of ultrastructural patterns. In general, these include scrolls, particles, reticular threads, crystals, homogeneous electron-dense material, and mixtures of these patterns [54]. All such variegated patterns in HMC granules bound R-G (mean = 452.5 gold particles/␮m2 of HMC mature granule R-G label in 15 experiments that were quantiﬁed) (ﬁg. 70A, B). In contrast, colloidal gold alone had a mean granule label of 2.02/␮m2 in six experiments (ﬁg. 70C). An irrelevant protein-gold conjugate, BSA-gold (ﬁg. 70D), did not label HMC granules, nor were HMC granules stained with an irrelevant enzyme (DNase)-gold complex (ﬁg. 70E). Immature granules in developing HMCs displayed electron-dense central nucleoids surrounded by a less dense matrix [20, 23]. In these developing granules, R-G stained the electron-dense nucleoids but not the less dense matrix. 6.3.4. Experiments to Determine the Basis of R-G Staining of HMC Granules

Electron-Dense Reagents en grid. Two reagents were examined: CF [271] and CG [272]. High magniﬁcation views revealed that both reagents bound extensively to electron-dense substructural materials within mature HMC granules (ﬁg. 70J). Prior staining en grid with either CF (ﬁg. 70K, L) or CG, followed by R-G staining, resulted in reduction of R-G labeling of HMC granules. CG is prepared by binding the cationic reagent, poly-L-lysine, to colloidal gold particles [272]. When poly-L-lysine alone was placed en grid before staining with R-G, reduction in R-G granule label was also noted (ﬁg. 70G). Sample Preparations. When post-ﬁxation with OsO4 and staining with uranyl acetate en bloc were omitted from sample preparation of HMCs, the intensity of R-G staining was diminished but exceeded background levels for the samples. Physical Parameters Related to the R-G Reagent. The degree of HMC granule staining was related to the age of the R-G reagent. In general, the R-G staining level for granules up to 2 weeks after preparation of the enzyme-gold complex was intense but dropped extensively after this time. Flocculation of the R-G reagent, with a color change from red to purple [273], was associated with failure to stain HMC granules. Dilution of the R-G reagent up to 1:100 provided decreased stain levels that persisted, however, above background for

Ultrastructure of Mast Cells and Basophils

118

the same samples. Heating the R-G reagent at 100 ° C for 10 min did not alter the red color of the reagent, but HMC granule staining was eradicated (ﬁg. 70T). Staining Parameters Affecting R-G Labeling of HMC Granules. The optimal time for staining grids with R-G was 1 h and the optimal temperature was 37 ° C. When very short staining times were used (e.g., 5 min), R-G labeling of HMC granules was evident but diminished compared to the standard R-G labeling of HMC granules. Similarly, reduction in R-G staining was noted when temperatures of 20 and 4 ° C were used for a 1-hour staining interval (compared to R-G staining of granules at 37 ° C). The density of R-G/␮m2 of HMC granules was determined over a pH staining range spanning 8.5–4.5. In general, the optimal pH for R-G staining of HMC granules was 7.5, a value used for standard and comparative purposes. Signiﬁcant quantitative reductions in R-G label/ ␮m2 granule occurred as the staining pH was increased or decreased. 6.3.5. Effect of General Inhibitors, Blockers, and Enzyme Digestions on R-G Labeling of HMC Granules

We performed experiments in which the effect of R-G granule labeling of general inhibitors, blockers, and enzyme digestions (by incubating the samples en grid in the test reagent or incubating the R-G reagent in solution with the test reagent) was assessed. In addition, a single reagent was tested before, with, or after R-G staining of samples. Furthermore, combinatorial studies of multiple reagents were done together, in sequential order and reverse-sequential order. In these experiments, labeling of HMC granules was blocked or reduced, respectively, with exposure to normal goat serum or human serum, but exposure to ␥-globulin (ﬁg. 70H) or DNA (ﬁg. 70I) did not block R-G staining of HMC granules. BSA placed en grid before R-G staining quantitatively reduced HMC granule label (ﬁg. 70S). When BSA and R-G were incubated in solution before staining, a reduction (but not signiﬁcantly so) in R-G labeling of HMC granules occurred (ﬁg. 70R). Blocking with CS by incubation in solution with R-G before staining (ﬁg. 70M) or en grid before staining with R-G qualitatively reduced but did not signiﬁcantly change the R-G label of granules. Blocking R-G with polyvinyl sulfate abrogated HMC granule R-G staining altogether (ﬁg. 70N). Inhibition of R-G staining of HMC granules occurred with exposure of the R-G reagent to guanidine-HCl (ﬁg. 70O) or protamine before en grid staining (ﬁg. 70P); no signiﬁcant change in R-G granule staining was seen with sodium arsenate (ﬁg. 70Q). Digestion of samples en grid with pronase E, proteinase K, or HCl did not signiﬁcantly reduce R-G labeling of HMC granules.

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

119

6.3.6. Effect of Speciﬁc Blocking and Enzyme Digestions on R-G Labeling of HMC Granules

Blocking of the R-G reagent in solution with heparin before R-G staining yielded a marked reduction in granule R-G label (ﬁg. 71C), but large aggregates were often present. When the heparin was placed en grid before R-G staining, no signiﬁcant change in label density occurred (ﬁg. 71D). When the R-G reagent was absorbed by passage over solid-phase heparin before en grid staining, a marked reduction in granule staining occurred (ﬁg. 71I) (and the heparin-agarose turned red) (ﬁg. 72) compared to the retention of granule R-G staining when R-G was passed over Sepharose beads alone (the Sepharose beads remained white) (ﬁg. 72). Heparinase digestion of the grid before R-G staining resulted in diminished granule staining (ﬁg. 71E), whereas incubation of heparin and heparinase together with the R-G reagent before staining the grid abrogated the reduction of staining effected by heparinase. Histamine blocking in solution resulted in decreased R-G labeling of granules (ﬁg. 71G), whereas histamine en grid did not affect R-G staining. Digestion of the grid with histaminase before R-G staining did not signiﬁcantly reduce HMC granule labeling with R-G (ﬁg. 71H).

Fig. 70. Human mast cell scroll-ﬁlled granules (all panels except F) and an osmiophilic lipid body (F) from 6-hour cultured lung mast cells (A, C–E, G, J–T), isolated skin mast cell (I), and skin mast cell in vivo (B, H) samples prepared to show the impact on the R-G method of various controls. A, B, K Panels show the standard amount of R-G label for granules; a lipid body does not stain with R-G (F). The granules do not stain with gold alone (C), with a gold-labeled protein, BSA-gold (D), or an irrelevant gold-labeled enzyme, DNase-gold (E). Pre-incubation of the grid with poly-L-lysine blocks subsequent RNase staining (G). Incubation of the R-G reagent with macromolecular ␥-globulin (H) or DNA (I) does not block R-G labeling of granules. J–L Panels show that incubation of the grid in cationized ferritin (CF) causes CF to bind to the scrolls and lamellae within granules (J), and pre-incubation of the grid with CF followed by R-G staining results in visible CF binding that markedly blocks R-G staining (L) compared to standard R-G staining for this experiment (K). Incubation of the R-G reagent with chondroitin sulfate (CS) caused qualitative reduction of R-G granule labeling (M) that did not reach statistical signiﬁcance, whereas incubation of R-G with polyvinyl sulfate prior to staining completely blocked it (N). Incubation of R-G in guanidine-HCl (O) or in protamine (P) completely inhibited granule gold label, but incubation of R-G in 0.001 M sodium arsenate produced no inhibition of granule gold label (Q). Additional controls showed that BSA incubated with R-G prior to staining did not block granule staining (R), but when the grid was blocked with BSA before R-G staining, the granule staining was eradicated (S). If the R-G reagent was boiled prior to use, granule staining was absent (T). Bars: A, J 55 nm; B, H, I 95 nm; C, G, N 80 nm; D, Q 65 nm; E, L 70 nm; F 135 nm; K, S 75 nm; M, O, P, R 60 nm; T 90 nm. [From 155, with permission.]

Ultrastructure of Mast Cells and Basophils

120

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

121

Blocking of the R-G reagent in solution or by incubating the grid with RNA simultaneously, or before R-G incubation, yielded variable results, from a reduction in granule R-G staining that did not achieve statistical signiﬁcance (ﬁg. 71J) to the presence of large aggregates on the grid. When RNA en grid preceded the R-G staining, no change in label density occurred, and aggregates were excessive and random in distribution. RNase digestion of the grid before R-G staining sometimes resulted in diminished granule staining but sometimes increased it and generally produced large, random aggregates of R-G staining (as did incubation of the R-G reagent with RNase before staining the grid) (ﬁg. 71K). Combining digestions of grids with either proteinase or pronase before RNase digestion and R-G staining markedly increased HMC granule label (ﬁg. 71M). Digestion of ﬁxed HMCs with RNase before further processing for electron microscopy and followed by R-G staining en grid resulted in no change in granule label. Therefore, RNase digestions resulted in decreased or increased, positive as well as negative, staining in variable approaches of 10 experiments. Moreover, the presence of large aggregates impeded accurate quantitation (ﬁg. 71K).

Fig. 71. Human mast cell scroll-containing (A, C–P) and crystal-containing (B) granules from 6-hour cultured lung mast cell (A, C–P) and isolated skin mast cell (B) samples prepared to demonstrate the effect(s) of heparin, RNA, heparinase, and RNase on granule staining with R-G. Standard level of granule staining for lung mast cell scroll granules (A). A crystal granule, typical of many HSMC granules, does not bind as much R-G (B). Gold particles in this granule are few and are primarily located at the interface of the crystal with homogeneous granule matrix material. When R-G was incubated with heparin before staining, the granule stain was reduced (C), but incubating the grid in heparin before staining with R-G did not reduce the staining (D). Digestion of the grid with heparinase before R-G staining abrogated gold label (E). Incubation of heparin with heparinase followed by incubation of the mixture with R-G before staining resulted in granule labeling (F) as with R-G only (A). Combining histamine and R-G in solution before staining signiﬁcantly reduced granule staining (G). Digestion of the grid with histaminase before staining with R-G did not signiﬁcantly change the gold labeling of granules (H). Passage of R-G over heparin-agarose before staining abrogated the granule label (I). Incubation of R-G with RNA before staining did not signiﬁcantly reduce the granule label (J). Incubating R-G with RNase before staining produced large gold aggregates bound to the granules (K) and digestion of the grid with RNase before staining with R-G did not signiﬁcantly change the amount of granule label (L). Digestion of the grid with proteinase K followed by RNase markedly increased the amount of R-G granule label (M). Passage of RNase over Sepharose beads followed by the R-G before staining signiﬁcantly reduced granule label (N), whereas passage of RNase over heparin-agarose beads followed by R-G before staining did not diminish granule label, which was often seen in aggregates (O). Gold alone did not bind to granules in this series of experiments (P). Bars: A, J 65 nm; B, L 70 nm; C, D 80 nm; E, F, N 75 nm; G, H, K 95 nm; I 50 nm; M 105 nm; O 60 nm; P 110 nm. [From 155, with permission.]

Ultrastructure of Mast Cells and Basophils

122

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

123

We next evaluated a number of combined blocking and digestion experiments. When the R-G reagent was combined in solution with heparin and RNA before staining, it resulted in a signiﬁcant decrease of granule staining compared to standard R-G staining. However, compared to heparin blocking in solution only or to RNA blocking in solution only, signiﬁcant differences were not seen. When RNA and RNase were incubated in solution for 60 min before a second 60-minute incubation with the R-G reagent, followed by en grid staining, random aggregates formed and the granule label was undiminished. In general, solid-phase heparin was most effective in absorbing the HMC granule labeling ability of the R-G reagent. We therefore incorporated its use into several combination schemes to assess the R-G reagent. When R-G was passed over Sepharose beads before staining, the granule label was retained and the beads remained white (ﬁg. 72). Passage of the R-G reagent over heparin-agarose beads before staining yielded a marked reduction of granule label and the beads turned red (ﬁg. 71I, 72). Combination in solution of R-G and RNA for 1 h, then passage over Sepharose beads and en grid staining, resulted in an insigniﬁcant reduction of gold label in granules and the beads remained white (ﬁg. 72). Combination in solution of R-G and RNA for 1 h, then passed over heparin-agarose beads and en grid staining, resulted in a signiﬁcant reduction of gold granule label, and the beads were red (ﬁg. 72). When R-G and heparin were incubated for 60 min in solution, passed over Sepharose beads, and grids stained, the granule gold label was signiﬁcantly reduced and the beads remained white (ﬁg. 72). A double absorption with heparin, by incubation of R-G and heparin together for 60 min, followed by passage over heparin-agarose and en grid staining, resulted in the lowest density of granule gold label in the heparin absorption experiments, and the beads were stained red (ﬁg. 72). We also evaluated the initial passage of RNase over either Sepharose or heparin-agarose, preceding the passage of R-G over each type of beads before en grid staining. In this case, granule R-G label after Sepharose was signiﬁcantly reduced (ﬁg. 71N) but not after passing over heparin-agarose (ﬁg. 71O). 6.3.7. R-G Staining of Agar Blocks Containing Heparin, RNA or Histamine

Standard reagent-containing agar blocks were prepared, ﬁxed, and processed for electron microscopy identically to the osmium collidine uranyl en bloc schedule used for a majority of cell and tissue samples. Thin sections containing these test materials were stained en grid with the R-G reagent. Agar blocks containing heparin or RNA stained with the R-G reagent. Agar alone, Epon alone, and agar blocks containing CS or histamine did not bind the R-G reagent.

Ultrastructure of Mast Cells and Basophils

124

Fig. 72. Bead absorptions of R-G: 1 Sepharose bead absorption; 2 heparin-agarose bead absorption; 3 R-G incubated with RNA and passed over Sepharose beads; 4 R-G incubated with RNA and passed over heparin-agarose; 5 R-G incubated with heparin and passed over Sepharose beads; 6 R-G incubated with heparin and passed over heparin-agarose. [From 216, with permission.]

6.3.8. R-G Labels Chondroitin Sulfate (CS) in Guinea Pig Basophil Granules

Basophilic leukocytes are metachromatic granule-containing secretory granulocytes which contain a mixture of granular proteoglycans devoid of heparin. In guinea pigs, isolated granules primarily contain CS [268, 269]. We used puriﬁed guinea pig basophil preparations [274], cells known to be devoid of heparin, to image other proteoglycans with the R-G method to validate its use for this purpose [267]. Thus, it is possible that R-G binds to heparin and CS as a competitive inhibitor as well as to RNA by is enzymatic properties. We found that mature granules in guinea pig basophils were labeled with R-G (133 gold particles/␮m2) (ﬁg. 73, 74) and that blocking with heparin did not alter the density of granule gold (229/␮m2), but blocking with CS did (0.1/␮m2). Gold alone did not attach to the granules. Ultrastructural autoradiographic and biochemical analyses of guinea pig basophils have determined that radiolabeled sulfur is incorporated into the granules of these cells [268, 275]. This indicates that the granules of guinea pig basophils contained sulfur-rich glycosaminoglycans (GAGs), or proteoglycans, which are GAGs linked to a protein core. Orenstein et al. [268, 269] characterized the sulfated GAGs in guinea pig basophils and found that most of these macromolecules consisted of a mixture of CS and dermatan sulfate. Smaller amounts were identiﬁed as heparin sulfate, but heparin was absent. It was suggested that the highly charged [35S] GAG eluting from DEAE-cellulose with high salt concentrations may represent an oversulfated CS [268].

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

125

In our studies [267], we used the known proteoglycan composition of guinea pig basophil histamine-rich secretory granules that was determined by biochemical analysis of isolated pure granule fractions [268] to document that the enzyme-afﬁnity-gold ultrastructural probe, R-G, can speciﬁcally bind to CS in basophil granules, which are devoid of heparin. Thus, we conﬁrmed that the R-G method has a wider applicability in electron microscopy – i.e., for imaging of RNA and CS. 6.3.9. R-G Labels CS in Human Basophil Granules

HBs are rare granulocytes that are produced in the bone marrow, circulate in mature granulated forms and leave the blood to enter tissues in disease [53]. They are classic secretory cells which also release potent biochemical mediators of inﬂammation by non-classical, stimulated secretory routes, making use of vesicular transport to do so [105]. Basophils that have migrated to tissues (in a wide variety of diseases and experimental circumstances) may secrete granule materials into the extracellular milieu. They are effector cells in CBH, a specialized form of cell-mediated hypersensitivity [169, 276]. The large cytoplasmic secretory granules in HBs (ﬁg. 75A, B) are metachromatic in light microscopic preparations, a property related to the presence of proteoglycans (GAGs linked to a protein core) within them [53]. Ultrastructural autoradiographic studies determined that radiolabeled sulfur was incorporated into HB granules [50]. This indicates that basophil granules contain sulfur-rich proteoglycans. Biochemical studies of HBs showed that they contain primarily CS in their secretory granules [277–279]. These studies were done using basophils from patients with myelogenous leukemia [277, 278] and basophils which developed in human CBC cultures [279]. Heparin was absent from HBs [278]. Using the R-G ultrastructural method, we found that the CS-rich granules in HBs could be labeled in electron microscopic samples (ﬁg. 75A, B). Generally, the electron-dense, particle-packed granules, which are the main morphologic granule type in HBs [53], bound the R-G reagent (ﬁg. 75A), whereas the small, infrequent paranuclear granule (described by Hastie [280]) did not. Finely granular, homogeneously dense material present in some HB granules bound R-G (ﬁg. 75B), whereas CLCs embedded within particle-packed

Fig. 73. Guinea pig bone marrow basophilic myelocytes, stained with R-G, show extensive labeling of ribosomes bound to RER (cisternae of the RER do not bind R-G) (A) and to free clusters of polyribosomes (B). The labeled RER encircles an immature granule in A, which has moderate R-G staining of the granule matrix. The granule in B is a smaller, mature granule displaying a typical parallel array. This granule is extensively labeled with gold particles. A !54,000. B !61,000. [From 289, with permission.]

Ultrastructure of Mast Cells and Basophils

126

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

127

granules were conspicuously devoid of R-G label indicating proteoglycans (ﬁg. 75A). In conjunction with biochemical [277–279] and earlier cytochemical analyses [281] designed to demonstrate sulfated glycoconjugates [282], the new ﬁndings obtained with R-G support and extend the localization of CS to HB granules. 6.3.10. R-G Labels Proteoglycans in Rabbit Basophil Granules

Rabbit basophils also leave the blood vascular space to enter tissues, such as the rabbit cornea, during inﬂammation [reviewed in 169]. Rabbit basophil secretory granules have sulfated glycoconjugates, as demonstrated by cytochemistry [283]. Their contents are ultrastructurally heterogeneous, displaying mixtures of electron-dense particles, concentric and tangled thick threads, and ﬁnely granular materials [169, 283–285]. These electron-dense components of rabbit basophils were labeled with the R-G method, indicating the presence of proteoglycans within them (ﬁg. 75C, D). 6.3.11. R-G Labels Heparin in Rat Peritoneal Mast Cell Granules

It has long been known that rat peritoneal mast cells are a rich source of heparin [182, 263, 286], and immunocytochemical [287] and cytochemical [288] approaches have been used to demonstrate the granule stores of heparin in rat peritoneal mast cells. One method used heparinase digestions and a cationic-gold method to demonstrate heparin in rat peritoneal mast cell granules

Fig. 74. Guinea pig peripheral blood basophil granules, prepared with R-G. A The parallel array of an unaltered mature granule remaining in the cell that was stimulated to degranulate retains R-G label. B A membrane-free granule, released into a cytoplasmic degranulation sac in a stimulated cell, is labeled minimally. C Another membrane-free granule that has been extruded to the extracellular space from a stimulated cell is also labeled minimally. D, E The granules in these panels were obtained from puriﬁed granule preparations [Dvorak et al., J Immunol 1977;119:38–46] from partially puriﬁed basophils [274]. These preparations were prepared in two ways: (1) to preserve granule membranes (D) or (2) to remove granule membranes (E). Note that the isolated basophil granules retaining their membranes are heavily labeled (arrow, D); the eosinophil mature granule adjacent to it has less label, and the label is conﬁned entirely to the matrix compartment. The central crystal is negative. In E, the preparation of basophil granules, devoid of membranes, reveals a marked reduction in label similar to the extruded granules from stimulated cells in B and C. F, G Developing immature granules in basophils that were recovered from an 18-hour culture after degranulation are shown. F The membrane-bound immature granules have R-G label over the central electron-dense material. The label is also associated with small peripheral vesicles in the outer granule space. G The condensed, electron-dense granule matrix of a more mature granule binds R-G extensively. A !58,000. B !41,800. C !44,900. D !39,900. E !50,400. F !60,800. G !63,200. [From 289, with permission.]

Ultrastructure of Mast Cells and Basophils

128

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

129

[288] and obtained results similar to those that we have obtained using the ultrastructural enzyme-afﬁnity-gold method based on the inhibition of RNase by heparin (ﬁg. 76) [289]. As with the cationic-gold method [288], we noted that immature granules with irregular electron-dense materials were largely devoid of heparin (ﬁg. 76C) [289]. Generally, the large, homogeneously electron-dense, mature secretory granules, which typify rat peritoneal mast cells, were heavily labeled with R-G (ﬁg. 76). Occasional granules similar in appearance to these mature granules in the same mast cells were nearly devoid of label with R-G (ﬁg. 76B, C). Focal areas within mature granules that contained homogeneously electron-dense material also failed to bind R-G or bound it poorly (ﬁg. 76D). These ﬁndings suggest some variability in heparin content in what appear to be (by routine ultrastructural morphology) unaltered, mature granules and in foci within mature granules. Immature, condensing granules apparently contain less heparin than is present in mature, fully condensed granules. 6.3.12. R-G Labels CS in Cultured, Bone Marrow-Derived Mouse Mast Cell Granules

Granulated cells were developed from mouse bone marrow, liver or spleen cells in factor-containing culture media and were identiﬁed by ultrastructural analysis as immature mast cells [47, 48]. Extensive studies characterized these cells as expressing Fc⑀ receptors that bind IgE, containing histamine, incorporating 35SO4 into CS (but not into heparin), and containing radiolabeled sulfur in their prominent, immature secretory granules by electron microscopic autoradiography [47, 48]. Stimulation of these immature MMCs led to classic AND of their immature granules in conjunction with histamine secretion [47, 49]. We stained similar preparations of cultured, bone marrow-derived, immature MMCs with R-G only or with a combined method including R-G and cytochemistry to detect non-speciﬁc esterase(s) [126]. The immature MMCs contained a mixture of homogeneously electron-dense mature granules and immature granules that were ﬁlled with small vesicles, focal, electron-dense progranular material and variable proportions of homogeneously dense matrix (ﬁg. 77). A prominent esterase-positive ectoenzyme was associated with the plasma membrane of these immature mast cells (ﬁg. 77C). The R-G method stained CS in mature and immature granules in these cells which contain no heparin (ﬁg. 77). Thus, analogous to labeled granules in human and guinea pig basophils, staining in the immature MMCs further substantiates use of this technique to label CS in cell lineages that do not contain heparin.

Ultrastructure of Mast Cells and Basophils

130

Fig. 75. Human peripheral blood (A, B) and rabbit corneal (C, D) basophils prepared with R-G demonstrate gold-labeled electron-dense structures. The typical large particulate material of the secretory granule is heavily labeled (A); the homogeneous CLC within this granule (arrow) is not labeled. Less frequently found, homogeneously dense HB granules also bind R-G (B). Rabbit basophil granule electron-dense material, close to the extracellular corneal collagen but still within the cell, binds R-G extensively (C), and, at higher magniﬁcation (D), electron-dense, concentric, thick threads in granules bind R-G. A !77,000. B !61,000. C !69,500. D !90,000. [From 289, with permission.]

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

131

76

Ultrastructure of Mast Cells and Basophils

132

77

Ultrastructural Enzyme-Afﬁnity-Gold and Inhibitor-Gold Techniques Identify Subcellular Sites of Histamine and Heparin in Basophils and Mast Cells

133

Fig. 76. Rat peritoneal mast cells, prepared with the R-G method, show dense labeling of secretory granule heparin, with no labeling of mitochondria (M) and adjacent cytoplasm (A–D). Some granules with ﬁnely granular, homogeneously dense contents label poorly (arrowheads, B, C). Focal areas of granules displaying a similar content also bind very little R-G in the same granule, which is otherwise heavily labeled (arrow, D). Immature granules containing focal deposits of electron-dense material in an electron-lucent background also stain poorly with R-G (arrow, C). A !50,500. B !45,500. C !36,500. D !50,500. [From 289, with permission.] Fig. 77. Immature mouse mast cells cultured from bone marrow [47, 48] and prepared with the R-G method (A–E) or with a double method, namely R-G and cytochemistry, to detect non-speciﬁc esterase(s) (C) [126]. A mixture of homogeneously electrondense mature granules and immature granules with heterogeneous contents (arrow) binds R-G in A. Immature granules (B, C) bind R-G predominately over electron-dense granule material (B) and over progranules (arrow, C) incorporated within the immature granules. The peripherally located vesicular component of immature granules also binds R-G (arrow, B). A prominent esterase ectoenzyme, located on the plasma membrane of an immature mouse mast cell, is visible in C. Higher magniﬁcations of granules with extensive gold binding are shown in D and E. A ribosome-ﬁlled portion of cytoplasm that binds R-G has indented a granule (arrow, E). A !23,000. B, C !32,000. D !50,000. E !50,500. [From 289, with permission.]

Ultrastructure of Mast Cells and Basophils

134

Chapter 7

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

7.1. Overview Piecemeal degranulation (PMD) is a term we coined to describe the morphology of secretion from HBs in contact allergy skin biopsies [290, 291]. In a sequential biopsy study of experimentally elicited lesions, electron microscopy revealed a progressive emptying of HB secretory granules [291]. Thus, the term PMD was used to describe progressive losses of granule particulate contents in the absence of granule-to-granule or granule-to-plasma membrane fusions. The morphologic hallmark of complete release by this mechanism is the retention of empty granule containers in the cell cytoplasm (ﬁg. 78). Secretory processes have traditionally been classiﬁed as regulated or constitutive processes: regulated secretion refers to exocytosis of storage granules, and constitutive secretion refers to vesicular secretory trafﬁc directly from Golgi structures to the plasma membrane [7]. PMD could be viewed as a diminished regulated secretory event. Ultrastructural morphology does not support an upregulated constitutive secretory event, which bypasses granule storage sites, for at least two reasons: (1) granule containers visibly empty with time, and (2) Golgi structures are diminished in mature HBs, without evidence of enlargement or increased vesicular components. However, upregulation of constitutive secretion involving vesicular trafﬁc from storage granules is supported by ultrastructural images of particle-ﬁlled (or empty) vesicles fused to HB granules, as well as the presence of these vesicles in the perigranular and peripheral cytoplasm (ﬁg. 79).

Fig. 78. rhSCF-injected human skin biopsy shows basophil (B) migration in a venule of the papillary dermis. The endothelial cell (E) has large collections of cytoplasmic ﬁlaments (white arrow) and prominent nuclear irregularities. The migrating basophil rests beneath the basal lamina (open arrowhead) of the endothelial cell and above the underlying pericytes (P). Intercellular junctions (arrows) between endothelial cells remain closed. The basophil shows extensive PMD, characterized by numerous empty granules; several granules remain ﬁlled with particles (closed arrowhead). L = Lumen. !21,700. [From 106, with permission.]

Ultrastructure of Mast Cells and Basophils

136

Fig. 79. Higher magniﬁcation of a peripheral blood basophil shows two particle-containing cytoplasmic granules surrounded by particle-ﬁlled (arrows) and empty vesicles. One granule has multiple compartments delineated by dense concentric membranes. !38,500. [From 52, with permission.]

PMD of mature HBs has been documented in a wide variety of human diseases in vivo [169] and in developing [19] and releasing [206, 292] basophils in vitro. Tissue basophils that migrate into inﬂammatory and neoplastic microenvironments almost always reveal evidence of PMD in human samples (ﬁg. 78) [169]. Insight into the potential triggering of HB PMD was gained when this form of release was found to prevail in HBs developing in vitro in suspension cultures of human CBCs when those cultures were supplemented with either rhIL-3 [19], rhIL-5 (ﬁg. 80) [19], or the c-kit ligand in various forms [24, 25]. It is possible that a wide variety of cytokines initiate PMD in microenvironmental sites that are rich in blood basophils [169]. Similarly, PMD is the prevailing ultrastructural morphologic manifestation of secretion from tissue mast cells in human disease [169].

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

137

The ultrastructural morphologic evidence supports vesicular transport as the mechanism for effecting PMD from HBs in vivo and in vitro [7, 52, 53, 169, 206, 276, 290–293]. Thus, visual evidence of vesicles fused to secretory granules, free in the cytoplasm and in the peripheral cytoplasmic area and fused to plasma membranes has been published for HBs. Proof of principle that PMD is effected by vesicular transport of loaded vesicles requires visualization and kinetic analyses of granule protein-loaded, 80- to 100-nm vesicles in stimulated basophils and/or mast cells. For the greater part of the past 10 years we have pursued this goal. This pursuit required the development of numerous tools and resources. Chief among these were the isolation and puriﬁcation of circulating basophils, identiﬁcation of speciﬁc growth factors to increase the supply of this rare granulocyte, understanding of secretogogue mechanisms and reliable analyses of secreted basophil products, and the development of ultrastructural preparations allowing imaging of small vesicles and quantiﬁable electron-dense tags for granule materials in small vesicles [reviewed in 7, 53; see Chapters 2, 4, 5, 6]. Applications of these tools to well-deﬁned models of basophil (and mast cell) secretion have provided substantial proof of principle for the effector function of vesicular transport in PMD.

7.2. Cutaneous Basophil Hypersensitivity In 1970, we demonstrated in guinea pigs extensive inﬁltration of mature leukocytes from the blood into tissues during cellular immunity of delayed onset mediated by lymphocytes [4]. We named such reactions cutaneous basophil hypersensitivity (CBH) (ﬁg. 81) to distinguish them from classical delayed hypersensitivity (DH). Subsequently, we identiﬁed this invasion of tissues by basophils in a variety of circumstances and several species [169]. Substantial differences and similarities exist between CBH and DH, which is another form of cell-mediated immunity. The discovery of CBH and its distinction from DH led to the discovery of a new form of secretion, termed piecemeal degranulation (PMD), which we initially described in guinea pig CBH reactions [294], and subsequently found to be the single most frequent ultrastructural feature of basophils when they inﬁltrate tissues [169]. DH is generally induced in animals by mycobacteria-containing adjuvants, such as complete Freund’s adjuvant [3, 295]. Historically, since the work of Dienes [296], it has been known that a form of delayed cutaneous hypersensitivity could be induced to protein antigens without the use of mycobacterial adjuvants. Rafael and Newel [297] termed this form of hypersensitivity

Ultrastructure of Mast Cells and Basophils

138

Fig. 80. PMD of a human basophil (5-week cord blood cell culture containing rhIL-5 and a fraction of stimulated human T-cell-conditioned media) reveals non-fused, empty, and partially empty granule chambers and numerous cytoplasmic vesicles with granule particles. Note the polylobed nucleus with condensed chromatin. !16,000. [From 141, with permission.]

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

139

Jones-Mote reactivity, on the basis of its similarity to reactions induced in man [298], and they correctly distinguished Jones-Mote reactions from classic DH on the basis of their evanescent character, their lack of induration and their occurrence in the absence of circulating antibody. Since Jones-Mote reactivity occurs early in the course of immunization, is evanescent and does not differ signiﬁcantly from classic DH, many regarded it as a weak form of DH [3]. DH reactions, by contrast, become progressively more intense over several weeks after immunization. In the early 1970s, new morphological, immunochemical and tissue culture evidence indicated that Jones-Mote reactions represented a distinct form of cellular hypersensitivity [4, 299–303]. Because the principal feature of these reactions is the invasion of tissues by basophils, because similar reactions may be elicited by antigens other than simple proteins and because the morphology of the reactions studied by Jones-Mote was unknown, we called these reactions cutaneous basophil hypersensitivity (CBH). CBH reactions are erythematous non-indurated lesions that can be elicited within the ﬁrst few weeks after sensitization, whereas DH reactions are not erythematous but do exhibit considerable induration [4]. Typical immunizing and skin test doses of antigen are 5–10 times larger for CBH than for DH. Moreover, when antigen is administered systemically in appropriately sensitized animals, a systemic skin rash with microscopic features of CBH occurs (ﬁg. 82) [304]. The reactions of CBH are similar, with respect to time course, evolution and distribution in the skin, to those of DH. However, central pallor and necrosis, which are features of DH, are not seen in CBH. As in DH, CBH begins with diapedesis of lymphocytes, 4–6 h after skin testing. Thereafter, a marked divergence of the two types of reaction is seen histologically. CBH reactions are characterized by striking effusions of blood-borne mature granulated basophils which may comprise one-third to one-half of the total cellular inﬁltrate. This is a remarkable event, given that the normal frequency of blood basophils is about 0.5% of all circulating leukocytes. In CBH there is an increase in microvascular permeability, which is signiﬁcant but much smaller than that seen in DH, minimal deposits of ﬁbrin in the intervascular dermis and activation of the microvascular endothelium, similar to that of DH but without evidence of endothelial injury [305]. Inﬁltrating basophils in CBH are mature cells that do not divide locally [4]. Maximum numbers of basophils occur in 24–48 h lesions. Classical DH is characterized by the local accumulation of large numbers of lymphocytes and monocytes of bone marrow origin; smaller numbers of basophils than in CBH can also be present in DH. Whereas basophils are a prominent component of CBH reactions, there is little doubt that lymphocytes are the cells essential to the induction and expression of this form of hypersensitivity, just as they are in classical DH. Animals primed for CBH exhibit an expansion of the thymus-dependent paracortical

Ultrastructure of Mast Cells and Basophils

140

Fig. 81. Light micrographs of papillary dermis of a 48-hour CBH reaction to the contact sensitizing antigen arsanilic acid (ABA) conjugated with the synthetic polypeptide, L-GAT (ABA L-GAT); one-half prepared for 1-␮m Epon-embedded sections (A) and the other half for routine histology with parafﬁn embedding and staining with hematoxylin and eosin (B). Guinea pig basophils readily identiﬁed as granule-containing polymorphonuclear cells (A) are not recognizable in the routine preparation (A). Arrow indicates an epidermal cell in mitosis. A, B !420. [From 4, with permission.]

zones of draining lymph nodes [302], reactivity is inhibited by anti-lymphocyte serum [300] and reactions are passively transferred to normal recipients by viable sensitized lymph node cells. The ﬁnding that a highly speciﬁc antiT-lymphocyte serum strikingly inhibits the expression of CBH in vivo [306] is in agreement with morphological data implicating T-cell expansion in the draining lymph nodes [302] and indicates that the lymphocytes mediating CBH are probably T cells. It is now well established that basophils leave the blood and inﬁltrate tissues in CBH reactions as well as in numerous basophil-rich pathologies [4, 53, 169]. The delayed recognition of this important facet of basophil biology can be explained by the difﬁculty of recognizing basophils in tissues, their rarity

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

141

Fig. 82. Light micrographs of 1-␮m-thick Epon-embedded and Giemsa-stained sections of the rash of systemic CBH induced with ovalbumin in guinea pigs showing a venule in the papillary dermis ﬁlled with basophils (A) and numerous basophils that have migrated into the papillary dermis (B). A, B !760. [From 276, with permission.]

Ultrastructure of Mast Cells and Basophils

142

Fig. 83. Electron micrograph of a guinea pig basophil showing two nuclear lobes, short blunt surface processes and dense cytoplasmic granules. The latter containing esterases(s) are labeled with autoradiographic silver grains as a result of pre-incubation of the cells with [3H]-labeled di-isopropyl ﬂuorophosphate, a serine esterase inhibitor. !8,400. [From 7, with permission.]

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

143

Ultrastructure of Mast Cells and Basophils

144

in the circulation and confusion of these metachromatic granulated cells with another metachromatic granule-containing cell, the mast cell. Criteria to differentiate between mature cells of either basophil or mast cell lineage have been reviewed [7, 54, 293]. Our initial ultrastructural identiﬁcation of guinea pig basophils (ﬁg. 83, 84) served as the basis for the deﬁnition of CBH as a new entity in this species [4]. Later, similar ultrastructural analyses were instrumental in deﬁning the basophil-rich inﬁltrate that accompanies the tissue reaction to contact allergens in humans (ﬁg. 85) [290, 291]. Basophilic leukocytes, although rare, are present in many species. Their ability to inﬁltrate tissues in CBH and basophil-rich pathologies suggests that they play an important role under these circumstances.

7.3. Mast Cells Mast cells in three species (human, mouse, rat) provided insights into PMD. HMCs were examined in vivo and ex vivo, MMCs in vivo and RMCs ex vivo and in all of these circumstances contributed to our documentation of PMD and vesicle transport as important to the secretory repertoire of mast cells. 7.3.1. Human Mast Cells in vivo

HMCs undergo PMD in multiple organ sites in human disease [reviewed in 55, 169]. Initially, we noted the characteristic ultrastructural morphology of empty and partially empty granules in HMCs in bowel samples of patients with CD, a chronic IBD of unknown etiology [307]. Other forms of inﬂammatory diseases and host response to neoplasia are accompanied by PMD of mast cells [reviewed in 55, 169]. In particular, we noted extensive PMD of HSMCs in vivo in bullous pemphigoid and melanoma with subsequent ultrastructural evidence of recovery of granules [77; see Chapter 9]. In a large study, 117 coded intestinal biopsies were examined by electron microscopy [69]. All surgical biopsies were obtained from uninvolved sites of patients with two IBDs (UC or CD) and from patients with preneoplastic and neoplastic diseases (adenocarcinoma, rectal polyp, familial polyposis). Biopsy sites included normal ileum, colon, and rectum as well as conventional ileostomies and continent pouches constructed from the ileum. Among other observations pertaining to HMC biology (granule morphology, lipid body number,

Fig. 84. Higher magniﬁcation micrographs of guinea pig basophil granules showing the crystalline parallel array in longitudinal view (A) and in cross-sectional view (B). Numerous cytoplasmic vesicles are also present. A, B !33,800. [From 7, with permission.]

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

145

Fig. 85. Electron micrograph of a human basophil within an epidermal vesicle in a 3-day lesion of allergic contact dermatitis caused by Urushiol. Granules contain variable numbers of particles. Some granules have markedly reduced particle content. Several granules contain multiple dense concentric membranes which envelope granule particles. Cytoplasmic vesicles contain particles or appear empty. Several large empty granule spaces are present; others contain glycogen particles similar to dense cytoplasmic glycogen particles. N = Nucleus. !9,000. [From 290, with permission.]

Ultrastructure of Mast Cells and Basophils

146

Fig. 86. Mast cell (CD, ileum) showing PMD. Note that nearly all granules have completely or partially emptied their containers of dense contents, leaving non-fused containers throughout the cytoplasm. !9,450. [From 308, with permission.]

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

147

mast cell-nerve associations), this large sample of coded biopsies was evaluated for ultrastructural evidence of mast cell secretion in vivo. Sixty percent of the biopsies had such evidence. Mast cell secretion was evident in control biopsies, many of which were obtained from uninvolved tissues of patients with IBD. Biopsies of inﬂamed continent pouches from UC patients showed more mast cell secretion than non-inﬂamed UC pouch biopsies [308]. This evidence of mast cell secretion supports recent work that documents high constitutive levels of histamine in jejunal ﬂuids of CD patients [309] and suggests a proinﬂammatory role for mast cells in inﬂammation associated with pouchitis. The primary ultrastructural form of secretion from HGMCs in this study was PMD typiﬁed by variable losses of dense content from granules (ﬁg. 86) (rarely, typical images of AND were also seen [308; see Chapter 9]). Granule losses of PMD were either focal within single granules, complete losses of single granule contents, or partial to complete losses of dense material from variable numbers of, to sometimes all, cytoplasmic granules. The end result of such granule losses was the presence of non-fused, empty granule containers in undamaged mast cells. Some of these containers were larger than granules; most were of similar size. As in previous studies [307] basophils (of peripheral blood origin) were also present in the lamina propria of patients with IBD (ﬁg. 87). While being a minor component, compared to the large numbers of mast cells present, these granulocytes displayed piecemeal losses of their particulate granule contents, characteristic of PMD (ﬁg. 87) [6, 7, 307, 308, 310]. Some granules undergoing piecemeal losses contained CLCs (ﬁg. 87). We have previously suggested that mast cells and basophils with similar stored mediators of inﬂammation (i.e., histamine), while clearly of separate lineages, together represent a rapidly mobilized pool of cells ready to inﬁltrate acute inﬂammatory reactions (i.e., basophils in newly initiated contact allergy) that give way to mast cell recruitment, proliferation and maturation during chronic inﬂammatory reactions (i.e., mast cells in IBD) [53, 290, 291, 311]. Biopsy samples from human ilea similar to those examined in the large study described above [69] were prepared for the ultrastructural detection of histamine in vivo with a new enzyme-afﬁnity method [154, 253; see Chapter 6]. The principle ﬁndings of this study were as follows: (1) HGMCs have histamine-rich electron-dense granules (ﬁg. 88); (2) histamine is present in electron-dense mast cell granules regardless of their variegated underlying electrondense substructure (ﬁg. 89A–C); (3) partially and completely empty granule containers in HGMCs undergoing PMD in vivo contain little or no histamine (ﬁg. 90A, C); (4) histamine is bound to interstitial collagen in areas adjacent to secretory mast cells, whereas basal laminae of other cell lineages in the same areas do not bind histamine, and (5) histamine absorption and histaminase di-

Ultrastructure of Mast Cells and Basophils

148

Fig. 87. A basophil (B) and mast cell (M) (UC, ileostomy) seen together show essential differences between these two functionally similar cells. The smaller basophil, a granulocyte, has a polylobed nucleus with condensed chromatin and small numbers of granules, many with piecemeal losses of granule particles. Note the numerous cytoplasmic vesicles. One granule contains a Charcot-Leyden crystal (arrowhead), structures which are never present in mast cells. The larger mast cell has a single large nucleus with more dispersed chromatin and more numerous and smaller granules than are present in basophils. !6,000. [From 308, with permission.]

gestion controls (ﬁg. 89D, E, 90B) are negative for both electron-dense HGMC granules and interstitial collagen. Mast cell granules were extensively labeled for histamine in cells with a complete complement of electron-dense granules (ﬁg. 88); cells with a mixture of full and empty granules also contained histamine in electron-dense (full) granules (ﬁg. 90A). Lipid bodies (large, round, non-membrane-bound osmiophilic structures), present in smaller numbers than granules in the cytoplasm of ileal mast cells, were not labeled for histamine, thus providing an internal negative control within the same samples. Basophils are granulocytes that also contain histamine in their electron-dense granules. Although these granulo-

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

149

Fig. 88. Histamine is indicated by extensive gold label with the DAO-G stain over electron-dense granules of non-secretory mucosal mast cells in a biopsy of an ileal stoma from a patient with CD. A !20,900. B !40,900. C !59,900. [From 253, with permission.]

Ultrastructure of Mast Cells and Basophils

150

Fig. 89. High magniﬁcation micrographs of non-secretory mast cell granules (CD and UC, ileostomies) show DAO-G label indicating histamine in scroll granules (A), homogeneously dense granules (B), and particle granules (C). Scroll granules do not label when DAO-G is absorbed with solid-phase histamine before staining sample (D), when sample is digested with DAO before staining with DAO-G (E), or when gold only is used to stain sample (F). A !63,700. B !50,800. C !45,600. D !51,300. E !49,400. F !52,300. [From 253, with permission.]

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

151

Fig. 90. A At higher magniﬁcations, mast cell granules from cells undergoing PMD (CD ileostomy) show gold label for histamine in electron-dense granules (1), gold label associated with residual electron-dense material in a partially empty granule (2), and no label in an entirely empty granule (3) in a cell undergoing PMD. B Absorption of DAO-G reagent with histamine-agarose before staining abrogates all mast cell granule label. C Residual electron-dense reticular arrays contain small amounts of gold label for histamine in otherwise electron-lucent, empty, histamine-free granules of mast cell undergoing PMD. A !48,000. B !39,500. C !42,000. [From 253, with permission.]

Ultrastructure of Mast Cells and Basophils

152

cytes were rare in the material examined [308] they too underwent PMD, and in such cells the histamine was reduced in the partially empty and completely empty granules. Small, perigranular cytoplasmic vesicles were gold labeled in basophils, indicating the presence of histamine. Data obtained with the new method used here to detect substructural histamine sites in HGMCs displaying PMD document the ultrastructural morphology of HMC histamine secretion in vivo for the ﬁrst time [253]. These studies clearly demonstrate that the secretory process termed PMD is biologically relevant in in vivo HGMCs in IBD and familial polyposis. 7.3.2. Human Mast Cells ex vivo

We examined HLMCs ex vivo with an immunogold method to detect the cytokine bFGF to localize its subcellular distribution [137]. In the course of these studies [137; see Chapter 4] we found this cytokine in electron-dense secretory granules (ﬁg. 52A, 53A) and in small 80- to 100-nm cytoplasmic vesicles adjacent to them (ﬁg. 53B). Also, altered granule matrices typical of PMD had diminished label for bFGF (ﬁg. 53B). These localizations suggest that perigranular vesicles loaded with bFGF are presumably in transit from nearby granule stores that are visibly depleted from bFGF. 7.3.3. Mouse Mast Cells in vivo

We used light and electron microscopy to analyze the eyelid inﬂammation that develops in transgenic mice that overexpress IL-4 [249, 312]. IL-4 is a multifunctional lymphokine that has a broad range of biologic activities [313]. Originally described as a product of helper T cells [314], IL-4 is also produced by mast cells [315–318]. By contrast, the non-T, non-B cells that are associated with IL-4 production during some immune responses in mice in vivo [107] have been identiﬁed as basophils and their precursors [11, 30] – a cell lineage that exhibits some similarities to, but is distinct from, the mast cell lineage [7, 31, 47, 48, 66]. IL-4 has a critical role in the pathogenesis of allergic and parasitic diseases [318A]; it is necessary for regulation of Ig isotype production [319–323], and it also can stimulate the growth and function of lymphocytes, macrophages, and mast cells [324]. Recently, overexpression of a transgene for IL-4 in mice has been accomplished [312]. IL-4 transgenic mice not only exhibit high serum IgE levels, but also develop severe inﬂammation of the eyelids. Histologically, the eyelid lesions contain mixtures of mast cells and eosinophils, and these resemble lesions observed in human allergic disorders [312]. The frequency and severity of the eyelid erythema and swelling in the IL-4 transgenic mice correlated directly with the level of the IL-4 transgene expressed in their T lymphocytes [312]. These ﬁndings may have relevance to human disease, because IL-4-producing

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

153

T-cell clones can be derived from the inﬂamed conjunctiva of patients with vernal conjunctivitis, a basophil-rich inﬂammatory inﬁltrate [325], in greater numbers than from the conjunctiva of control subjects [326]. Mast cells participate in many allergic or chronic inﬂammatory disorders, as well as in the host response to neoplasia, in both humans and animals [50]. Our ultrastructural studies indicate that, in settings such as these, mature mast cells can release their granule contents but retain the granule membranes within the cytoplasm; this process has been termed piecemeal degranulation (PMD) [169] to distinguish this mode of secretion from the explosive extrusion of granules that characterizes AND [7]. Although ultrastructural evidence indicates that the process of PMD can result in alterations of granule contents with a reduction in their electron density, until recently there has been no way to determine whether the cells involved have actually secreted histamine. Histamine is a major mast cell granule component and a potent proinﬂammatory mediator [50]. We developed a new post-embedding ultrastructural enzyme-afﬁnity-gold technology that can localize this mediator at the ultrastructural level [154]. We tested the speciﬁcity of the DAO-G method on cultured HLMCs and showed that electron-dense secretory granules contained histamine and that the staining of these granules was abolished by prior digestion with DAO or by ﬁltering the DAO-G over solid-phase histamine before use [154; see Chapter 6]. However, in that initial study we did not assess whether mast cells stimulated to secrete mediators in vivo exhibited alteration of staining with DAO-G. We have now evaluated the eyelid lesions that develop in IL-4 transgenic mice [312] using light microscopy of alkaline-Giemsa-stained plastic sections (ﬁg. 91), routine transmission electron microscopy, and enzyme-afﬁnity-gold electron microscopy to detect histamine [249]. We found that the tissue mast cells exhibiting PMD in the eyelid lesions had greatly diminished staining of granules with DAO-G (ﬁg. 92), results that indicate that this morphologic expression of mast cell secretion is associated with secretion of histamine in vivo [249]. These results in vivo in a MMC model of inﬂammatory eye disease [249] together with similar ﬁndings in vivo in human IBD [253] represent the ﬁrst direct evidence for histamine secretion by mast cells in vivo [249, 253]. 7.3.4. Rat Mast Cells ex vivo

We used RMCs puriﬁed from the peritoneal ﬂuid in ex vivo studies to detect substructural sites of chymase, histamine and TNF-␣ [134, 135, 152, 153, 185; see Chapters 4, 5]. For RMCs the immunogold technique for histamine was used and was sufﬁcient to detect granule stores [153] but it was difﬁcult to assess vesicular stores due to the low signal that was prevalent with this post-embedding method (ﬁg. 59). Therefore, we directed our attention to localization of the cytokine TNF-␣ (ﬁg. 54) [134, 135] and the abun-

Ultrastructure of Mast Cells and Basophils

154

Fig. 91. Light microscopy of alkaline-Giemsa-stained plastic 1-␮m sections of eyelids in IL-4 transgenic mice shows piecemeal degranulation of mast cells. A mast cell with no altered granules is shown (A) to compare with those showing staining alterations of some (B, C) or virtually all (D) of the cytoplasmic granules. The mast cells undergoing piecemeal degranulation (B–D) show many cytoplasmic granules that stain pink, in contrast to the dark blue staining exhibited by the granules in the normal-appearing mast cell in A. However, the granules with altered staining remain within the cytoplasm. A !1,850. B !1,250. C !1,450. D !1,850. [From 249, with permission.]

dant granule protein, chymase (ﬁg. 56) [152, 185] in RMCs. Immunoreactive TNF-␣ was present in RMC granules [134] and utilization of microwave-assisted aldehyde ﬁxation protocols in seconds and milliseconds allowed us to show that the stimulated reduction of granule electron-dense material was associated with a decreased density of TNF-␣ label (ﬁg. 55) [135]. Thus, the rapid changes in RMC contents that preceded the granule extrusion stimulated by compound 48/80 were associated with losses from granule stores of TNF-␣ [135]. Next, we showed that the granule-associated protease chymase was also within cytoplasmic vesicles in appropriately stimulated rat peritoneal mast cells (ﬁg. 93) [185]. Rat peritoneal mast cells were recovered before or 1–10 s after exposure to the secretogogue compound 48/80 (10 mg/ml) and then were examined with post-embedding immunoelectron microscopy to identify the subcellular

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

155

localization of chymase [185]. In comparison to unstimulated cells, compound 48/80-stimulated cells showed an increase (15%, 28%) in the surface area of the cell and a decrease (12%, 6%) in the surface area of the total granule compartment before degranulation channel formation. These global cellular changes occurred in a background of transient but signiﬁcant (p ! 0.01) increases in the area and number of chymase-immunoreactive vesicles per ␮m2 cytoplasm (ﬁg. 93) [185]. These changes were detectable at 5 or 7 s after stimulation with compound 48/80 but returned to near prestimulation levels by 9 or 10 s after addition of compound 48/80 (total cumulative histamine release was 28% by 8 s and 47% by 14 s) [185]. These observations suggest that vesicles participate in the early stages of regulated secretion of chymase from rat peritoneal mast cells.

7.4. Basophils HBs have been puriﬁed from the blood and their stimulated release reactions measured for a number of stimuli with different kinetics [reviewed in 7]. We did detailed ultrastructural kinetic analyses of secretogogues that showed PMD as part of their secretory continuum/repertoire [206, 292, 327]. With this standard morphological-morphometric approach as background, we were then able to examine the changing subcellular locations of immunoreactive CLC protein and enzyme-afﬁnity tagged histamine in PMD. 7.4.1. Piecemeal Degranulation of Human Basophils ex vivo Is Stimulated by Bacterial Products, Tumor-Promoting Agents, and Cytokines

f-Met Peptide. We examined the kinetics of morphological change induced by stimulation of HBs with f-Met peptide using partially puriﬁed cells from normal donors [206]. Supernatants were collected at 30 and 60 min and assayed for histamine. Samples of basophils were prepared for electron microscopy at 0, 10, 20, 30 s and 1, 2, 5, and 10 min post-stimulation with f-Met peptide.

Fig. 92. Electron micrographs of DAO-G preparations of inﬂamed eyelids in IL-4 transgenic mice. A–F Electron-dense secretory granules in mast cells undergoing PMD are gold-labeled, indicating the presence of histamine in the unaltered cytoplasmic granules of these mast cells. However, there is little or no gold labeling of the swollen granules, which exhibit greatly diminished electron density and altered granule matrix (A, C, F). In one mast cell (A), two immature granules show less gold label (arrowheads) than do the mature, electron-dense granules nearby. G Eosinophil granules (note their central, electrondense cores) are not labeled with gold. A !12,500. B !39,900. C !20,000. D !36,100. E !55,100. F !27,600. G !36,100. [From 249, with permission.]

Ultrastructure of Mast Cells and Basophils

156

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

157

We found that f-Met peptide, a bacterial peptide, induced a unique sequence of morphologic events that included morphologies we have previously identiﬁed and termed piecemeal degranulation (PMD) in HBs in situ as well as those induced by IgE mechanisms ex vivo and termed anaphylactic degranulation (AND), thus supporting a general degranulation model for basophils and mast cells [6]. In addition to this degranulation continuum, we found that chambers of releasing granules underwent extraordinary increases in size as they emptied their contents and before their resolution by extrusion (ﬁg. 94). The enlarging granule chambers accumulated numerous concentric dense membranes, vesicles, and CLCs. These early changes generally preceded half-maximum histamine release, whereas the later extrusion of full granules, emptied granules and their membranous contents coincided with half-maximum histamine release [328]. Shedding of membranes from several sources accompanied extrusion of granules and intragranular CLCs. These sources included the expanded granule membranes from empty granules, granule membranes from full granules, collections of intragranular concentric dense membranes and vesicles, and surface membranes and processes. These extraordinary membrane shifts were generated and persisted over the 10-minute period examined and coincided with the later time frame within which leukotriene C4 was generated and released from HBs stimulated by f-Met peptide [328]. Viable basophils, completely free of both full and empty granules, showed some morphologic evidence of recovery of granule products by 10 min after stimulation with f-Met peptide. The unique morphology of f-Met peptide-induced degranulation of HBs is supported by the uniqueness of the biochemical events associated with this trigger. The early phase of this anatomic continuum may reﬂect relevant in situ activity of basophils, since we regularly ﬁnd PMD in tissue basophils in human disease [169]. This bacterial product may be one of many bacterial, viral, or cellular products with the capability for the induction of PMD from HBs. Tetradecanoyl Phorbol Acetate (TPA). We examined the ultrastructural kinetic morphology associated with stimulation of HBs with TPA – a tumorpromoting phobol diester known to elicit histamine (but not LTC-4) release [329, 330]. Partially puriﬁed HBs were prepared for electron microscopy and examined either after control incubations in buffer alone or at 0 time, 1, 2, 5, 10, 30, and 45 min after TPA stimulation [292]. Standard morphology and ultrastructural quantitation of vesicles and granules and contents of vesicles or alteration of granules was done. Like biochemical studies that have determined that TPA is a unique secretogogue for HBs [329, 330], the morphology stimulated by TPA and associated with histamine release was also unique [292]. For example, very few images of AND were evident. A far greater number of PMD images was seen. PMD was associated with ⬃50% alteration of cytoplasmic

Ultrastructure of Mast Cells and Basophils

158

Fig. 93. Immunogold localization of chymase in an unstimulated rat peritoneal mast cell (A) and in rat peritoneal mast cells that had been stimulated for 5 s (B), 7 s (C), or 8 s (D) with compound 48/80 (10 ␮g/ml). A The granule membranes are closely apposed to the relatively electron-dense granule matrix of these unaltered granules. Gold particles heavily label the granule matrix but not the nucleus (N), cytoplasm, or several perigranular vesicles. B The membranes of the granules are irregularly contoured and are separated from the matrix of the granules by electron-lucent regions (asterisks). Gold particles are infrequent in these electron-lucent regions but are prominent in some regions of the granule matrix. Vesicles (arrows) are attached to granular membranes. C Gold particles are present beneath out-pouched vesicles attached to a granule membrane (arrows). D At 8 s, a small perigranular vesicle contains gold label for chymase (arrow); unlabeled vesicles are also present. Bar = 0.25 ␮m. [From 185, with permission.]

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

159

granules by 45 min after TPA stimulation. This evidence of empty granules was associated with, and preceded by, a rapid, extensive, and sustained increase in particle-containing cytoplasmic vesicles, as compared with buffer controls (p ! 0.001 for each TPA stimulation time compared with unstimulated basophils) (ﬁg. 95). Cytokines – Recombinant Histamine-Releasing Factor (rHRF); Monocyte Chemotactic Protein-1 (MCP-1). An ultrastructural analysis of HBs stimulated with anti-IgE, rHRF, or MCP-1 (compared with unstimulated cells) was performed [327, 331, 332]. Partially puriﬁed peripheral blood basophils were prepared for electron microscopy at time points known to precede histamine release and at half-maximum histamine release times for each secretogogue. Activation morphologies associated with stimulation included granule-vesicle attachments (ﬁg. 96, 97), PMD, AND, and uropod formation. These features were qualitatively similar in the stimulated samples. Quantitative differences were evident, however, when stimulated samples were compared with controls or at different time points after stimulation with a single agent or when individual secretogogues were compared. All stimulated samples differed quantitatively from the control samples. Rank orders for morphologic activation events revealed that the most effective trigger for AND was anti-IgE 1 MCP-1 1 rHRF, whereas the most effective trigger for uropod formation was rHRF 1 anti-IgE 1 MCP-1. Rank orders for PMD and granule-vesicle attachments were the same: MCP-1 1 anti-IgE 1 rHRF. Important relationships among these anatomic events reveal that the development of motile conﬁgurations is not associated with the development of secretion morphologies, that granulevesicle attachment and PMD are associated and that PMD precedes and is inversely related to AND in stimulated samples.

Fig. 94. Control human basophils from peripheral blood of normal donors show fully granulated cell (A) and a cell with a mixture of full and empty granules (arrowhead) (B). Cells have polylobed nuclei, are round, and have short surface processes. Small aggregates and individual dense particles of glycogen are present in the cytoplasm. Full cytoplasmic granules (B) contain dense concentric membranes (arrow). C A 10 s post-f-Met peptidestimulated sample shows a basophil containing only discrete, enlarged, empty granule containers and empty vesicles in the cytoplasm. Several enlarged granules contain membranes and vesicles; one retains a small amount of particulate content (open arrowhead). Note that the cell surface is devoid of processes. A, B !14,000. C !19,400. [From 206, with permission.]

Ultrastructure of Mast Cells and Basophils

160

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

161

7.4.2. Ultrastructural Morphology of Secretion by FMLP- and TPA-Stimulated Human Basophils

Quantitation of total granule numbers, altered (by piecemeal losses of granule matrix) granule numbers and total vesicle numbers was done for two disparate HB secretogogues, formyl-methionyl-leucyl-phenylalanine (FMLP) [67, 104, 200] and TPA [199, 104]. These two secretogogues, FMLP, a bacterial peptide with rapid biochemical release kinetics [328], and TPA, a tumor-promoting phobol diester with slow biochemical release kinetics [329, 330], also showed different ultrastructural correlates of secretion [206, 292]. FMLP-stimulated basophils [206], for example, initially emptied granules, the containers of which remained in situ in the cytoplasm. Many of these empty containers enlarged dramatically (thereby satisfying the morphological criteria for the PMD-II phenotype, see later) before accumulated intragranular membranes and vesicles were released along with the extrusion of the empty container membranes. These early events (occurring 0–20 s post-stimulation) were morphologically analogous to PMD of HBs in vivo. In both cases, large numbers of cytoplasmic vesicles were present, and in the FMLP-stimulated model, vesicle numbers increased in conjunction with acquisition of the morphology of PMD, and preceded the development of the morphology of granule extrusions. Vesicle numbers decreased dramatically in cells displaying the morphology of AND, deﬁned by the extrusion of full granules, the membranes of granules previously emptied by PMD and intragranular CLCs. These events coincided with the half-maximum histamine release reported for this model at 1.3 min post-stimulation [328]. Granule-free cells, termed completely degranulated basophils (CDBs), analogous to those produced by anti-IgE stimulation at later times, were prevalent between 20 and 120 s. Thus, a degranulation continuum of morphologic change occurred when HBs were stimulated with FMLP. Essentially, in morphologic kinetic studies, a continuum of PMD occurred early and progressed to AND at later time points, coincident with the rapid release of histamine. TPA, on the other hand, elicits histamine release slowly [329]. Unlike the rapid kinetics associated with IgE-mediated histamine release (15 min) or FMLP-mediated histamine release (2 min), the histamine release stimulated

Fig. 95. A At 5 min after TPA stimulation, this human peripheral blood basophil has many full vesicles (arrows) in the cytoplasm. Several granules show focal loss of granule particles, and one granule has decreased particles throughout (arrowhead). B At 10 min after TPA stimulation, a large number of full vesicles are seen (arrow). Nearly all granules show focal losses in granule particles or extensively altered contents of membrane-bound, non-fused granules (arrowhead). Empty vesicles are also seen. A !44,000.0 B !54,500. [From 292, with permission.]

Ultrastructure of Mast Cells and Basophils

162

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

163

from HBs by the phorbol ester, TPA, reaches a maximum by 1 h [329]. By electron microscopic evaluation, extensive PMD was evident in multiple samples, achieving ⬃50% granule alteration by 45 min post-stimulation. This evidence of empty granules was associated with, and preceded by, a rapid, extensive and sustained elevation in particle-containing cytoplasmic vesicles, as compared to buffer-incubated controls at all time points examined [292]. There was minimal classical exocytosis that was not associated with signiﬁcant reductions in HB granule numbers over a 45-min period. Completely granule-free cells were absent – a feature that is regularly present at peak histamine release times after FMLP stimulation of replicate samples from the same donors [206]. There was extensive PMD, characterized by a change in the ratio of altered to unaltered granules, from 1 to 4 (in controls) to 1 to 2 by 45 min after exposure to TPA. In concert with these ﬁndings, a 5-fold increase in the number of cytoplasmic vesicles containing particles occurred in TPA samples, compared to unstimulated cells, at multiple times sampled, including 45 min after TPA. At no time did empty vesicles increase to levels observed in FMLP-stimulated samples from the same donors, nor did extensively enlarged empty granule containers appear. The total number of vesicles (after TPA) was stable, indicating balanced vesicular trafﬁc between the cell surface and granules, accompanying PMD and extending to 45 min after stimulation of basophils. Thus, a coordinate secretion of histamine [329] was associated with the morphology of PMD [292] in this model.

Fig. 96. High magniﬁcation views of human peripheral blood basophils 2 min (A, B) or 7 min (C–F) after stimulation with rHRF are shown. Glycogen-rich granule-vesicle attachments are illustrated (A–D). Attached vesicles show a variable degree of altered contents similar to that underlying them in the particulate granule material, and vesicles are encased with dense glycogen aggregates. Some of these attached vesicles are oriented toward the cell surface (A, B). Note the separate vesicle adjacent to one of the attached vesicles, which spans the cytoplasm to dock on the plasma membrane (arrow, B). The entire content of one granule is altered (E), as in PMD, and an extensive glycogen aggregate surrounds this granule. Granule exocytosis (F), as in AND, is illustrated (arrow). A !50,800. B !53,200. C !49,400. D !39,900. E !33,700. F !48,900. [From 327, with permission.]

Ultrastructure of Mast Cells and Basophils

164

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

165

7.4.3. Stimulated Basophils Show Differences in Ultrastructural Phenotypes, Granules and Vesicles

An ultrastructural analysis of FMLP-stimulated basophils was done in two complementary ways to maximize the obtainable information. Thus, a sample of 1100 basophils was quantitated at rapid times after stimulation (kinetics study) as well as by distinctive morphologic phenotypes. The phenotypes represent ultrastructural snapshots of several stages of PMD (termed I, II), AND (termed I, II), CDBs, and recovering basophils (RBs) (termed I, II), which roughly spanned, in order, the time sequence of 0–10 min following exposure to FMLP [67]. The piecemeal degranulating basophil I (PMD-I) phenotype (ﬁg. 98B) is characterized by focal and complete losses and alterations of the dense particle contents of the major granule population. Rarely, several granules are fused, but fusion events are not a general feature of the phenotype. Rather, numerous cytoplasmic, smooth membrane-bound vesicles (many containing electron-dense granule particles) are present. Some granules had either vesicles with electron-lucent interiors or dense particle- or dense content-ﬁlled vesicles budding from their membranes. The typical, condensed, polylobed nucleus is unchanged, as is true for all phenotypes. The surface architecture resembles that of unstimulated cells, consisting of blunt, irregular protrusions. The piecemeal degranulating basophil II (PMD-II) phenotype (ﬁg. 98C) displays non-fused, empty (or emptying) granule chambers in the cytoplasm, many of which are markedly enlarged compared to unaltered speciﬁc granules. The surface architecture of PMD-II cells is smoother, with fewer processes than are found in either unstimulated (ﬁg. 98A) or PMD-I cells (ﬁg. 98B). The characteristic feature of the anaphylactic degranulating basophil I (AND-I) phenotype (ﬁg. 98D) is the presence of intracytoplasmic degranulation channels formed by fusion of granule membranes. These degranula-

Fig. 97. A high magniﬁcation montage of human peripheral blood basophils, stimulated with MCP-1 for 5 s (A–G) or 30 s (H–I), shows glycogen-rich, granule-vesicle attachments that accompany PMD. Some of the attached, elongated vesicles contain electron-dense granule particles (arrows, A–C); some attached vesicles contain less dense (or electron-lucent) altered materials (arrows, D–I). Larger, electron-dense glycogen particles are interposed at the necks of tubular-vesicular granule extensions (A, B, E) or encase the outer surface of attached vesicles (D, F–I). Also noted are single (A–C, E, F, I), double (H), and triple (G) glycogen-encased vesicles attached to granules. Some granules, with vesicles attached, have a full complement of dense particles (A–D), focal, underlying piecemeal losses of granule particles (E, G, I), or diminished particles throughout (F). One empty granule (D) has individual glycogen particles within it. A !47,000. B !54,000. C, D !46,000. E !66,000. F, G !59,000. H !48,000. I !51,000. [From 327, with permission.]

Ultrastructure of Mast Cells and Basophils

166

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

167

98

Ultrastructure of Mast Cells and Basophils

168

98

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

169

Fig. 98. Human basophil phenotypes and their relative distribution of CLC protein. A Unstimulated buffer-incubated control basophil, with polylobed nucleus, irregularly placed short surface processes, and a full complement of particle-ﬁlled granules, shows a small number of 30-nm gold particles representing CLC protein associated with the nucleus, cytoplasm, plasma membrane and granules (representative gold particles are encircled). Several granules show focal decreases in granule particle contents; most are unchanged. Cytoplasmic glycogen particles are visible as individual particles and small aggregates (closed arrow). Cytoplasmic, empty smooth vesicles are also present (open arrow). The percentage of total cellular gold particles representing CLC protein in six cellular compartments [nucleus (NUC), cytoplasm (CYTO), granule (GRAN), vesicles (VES), plasma membrane (PM) and formed Charcot-Leyden crystals (CLC)] is illustrated graphically for combined unstimulated buffer control basophils incubated for 20 s, 1- and 10-minute intervals. Bar = 0.7 ␮m. B PMD-I basophil from a sample obtained 10 s after f-Met peptide shows distribution of 30-nm gold particles representing CLC protein in the nucleus, cytoplasm and plasma membrane. Granule label is obscured by the dense granule contents at this low magniﬁcation. Note that the PMD-I phenotype has ⬃1/2 the cytoplasmic granules that have lost dense particles but are essentially similar in size to unaltered dense granules. The percentage of total cellular gold particles for six cellular compartments of the PMD-I phenotype is shown. The data for this phenotype were collected from basophils recovered 10, 20, 30 s and 1 min after f-Met peptide stimulation. Bar = 0.7 ␮m. C PMD-II basophil from a sample obtained 20 s after f-Met peptide shows the morphological characteristic of this phenotype, i.e., enlarged, empty, non-fused granule containers. 30-nm gold particles representing CLC protein are seen in the cytoplasm, plasma membrane and formed CLCs (arrow) within an enlarged granule container. The percentage of total cellular gold particles for the PMD-II phenotype is shown. The data for this phenotype were collected from basophils recovered 10, 20, 30 s and 1 min after f-Met peptide stimulation. Bar = 0.7 ␮m. D AND-I basophil from a sample obtained 1 min after f-Met peptide shows intracellular, membrane-bound, elongated degranulation channels containing either altered granule particles or homogeneous CLCs (arrow). Several unchanged granules remain in the cytoplasm. Note the extraordinary ampliﬁcation and complexity of the surface processes. 30-nm gold particles representing CLC protein are present in the nucleus, cytoplasm, granules, vesicles, plasma membrane, formed CLCs (arrow) and channel membranes. The percentage of total cellular gold particles for eight cellular compartments in this phenotype is shown. These compartments include channel membrane (CH M) and channel lumen (CH LUM) as well as the six previously described compartments. The data for this phenotype were collected from basophils recovered 20 s, 1, 2, 5 and 10 min after f-Met peptide stimulation. Bar = 0.5 ␮m. E AND-II basophil from a sample obtained 20 s after f-Met peptide shows extrusion of membrane-free granule particles (closed arrow), CLC (open arrow) and concentric membranes (curved arrow) located in deep surface clefts and cul-de-sacs. 30-nm gold particles representing CLC protein are present in the nucleus, cytoplasm, plasma membrane and CLCs. The percentage of total cellular gold particles for the seven cellular compartments is shown. These compartments include extracellular granules (EC-G) in addition to those described previously. The data for this phenotype were collected from basophils recovered 20 and 30 s, 1, 2 and 10 min after f-Met peptide stimulation. Bar = 0.7 ␮m. F CDB basophil from a sample obtained 20 s after f-Met peptide shows complete absence of granules, a polylobed nucleus and shedding of surface processes. The cytoplasm contains mitochondria and large aggregates of glycogen (arrows). 30-nm gold particles representing CLC protein

Ultrastructure of Mast Cells and Basophils

170

are present in the nucleus, cytoplasm, and plasma membrane. The percentage of total cellular gold particles for six cellular compartments is shown. The data were collected from basophils recovered 20 and 30 s, 1, 5 and 10 min after f-Met peptide stimulation. Bar = 0.6 ␮m. G RB-I basophil from a sample obtained 2 min after f-Met peptide shows 30-nm gold particles representing CLC protein in the nucleus, cytoplasm, granules, plasma membrane and CLCs. Formed CLCs are present in particle-containing granules (closed arrowhead) and as large, round, non-membrane-bound homogeneous structures in the cytoplasm (open arrowheads). Note that this RB-I cell has a polylobed nucleus and a large number of granules, many of which are completely ﬁlled with dense particulate contents. Other granule chambers, while not enlarged, have diminished granule particles. Many of these granules are smaller than the usual 1–2 ␮m size. The percentage of total cellular gold particles for six cellular compartments is shown. The data were collected from basophils recovered 2, 5 and 10 min after f-Met peptide stimulation. Bar = 0.5 ␮m. H RB-II basophil from a sample obtained 2 min after f-Met peptide shows numerous cytoplasmic granules. Most of these are the main basophil granule type ﬁlled with dense particles, some of which contain 30-nm gold-labeled formed CLCs within them. Gold particles representing CLC protein are present in the nucleus and cytoplasm as well. The plasma membrane is nearly devoid of label. Granules ﬁlled with homogeneously dense contents are heavily and uniformly labeled with gold particles (arrowheads). These granules are as large as the main particle granule type, are membrane-bound, do not show evidence of piecemeal degranulation, and contain homogeneously dense contents devoid of particles. The percentage of total cellular gold particles for six cellular compartments is shown. The data were collected from basophils recovered 2, 5 and 10 min after f-Met peptide stimulation. Bar = 0.6 ␮m. [From 67, with permission.]

tion channels contain altered granule matrix particles, concentric intragranular membranes, formed CLCs, or they are empty. Some residual, unaltered or partially altered granules remain in the cytoplasm. The cell surface shows increased processes when compared to unstimulated cells or to PMD-I and PMD-II phenotypes. The anaphylactic degranulating basophil II (AND-II) phenotype (ﬁg. 98E) is characterized by multiple extrusion events. Various intragranular contents are extruded multiply and circumferentially around the perimeter of the cell through multiple narrow pores in the plasma membrane. The materials released consist of membrane-free granule particles, intragranular concentric membranes, intragranular vesicles, and formed, homogeneous, spherical CLCs. Residual unaltered or partially altered granules remain in the cytoplasm. Only 28% of the total remaining granules in AND-II cells are altered, a value similar to that of unstimulated basophils (29%). The cell surface of the AND-II cells is the most complex of all phenotypes stimulated by FMLP. Most of the extraordinary surface complexities and elongated processes are the result of externalization of granule and channel membranes as the contents of these spaces are extruded to the extracellular space.

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

171

The CDB phenotype (ﬁg. 98F) is characterized by the virtual absence of speciﬁc cytoplasmic granules, following their stimulated release, and no altered granules remain in the cytoplasm. Cell surfaces are either irregular or smooth. The recovering basophil I (RB-I) phenotype (ﬁg. 98G) is characterized by large numbers of cytoplasmic granules, a polylobed nucleus and a simpliﬁed surface architecture. Formed CLCs are plentiful in speciﬁc granules. A distinctive feature of the recovering basophil II (RB-II) phenotype (ﬁg. 98H) is the presence of large, uniformly dense, membrane-bound granules, which are completely devoid of the particulate, concentric membranous and formed CLC substructural components, which typify the major HB granule population. Also characteristic of RB-II cells is the presence of a full complement of the large (1–2 ␮m) particle-ﬁlled granules, which are typical for HBs. Many of these granules contain formed CLCs. From quantitative data, it is seen that the total granule number/␮m2 reﬂects the regulated secretory process, AND, which is effected by extrusion of entire granules to the cell’s exterior, a process most prevalent at 1 min (0.48/␮m2) post-stimulus and in the late AND-II and CDB phenotypes (0.25/␮m2, 0.02/␮m2). On the other hand, altered granules are a reﬂection of the PMD process. The percentage of altered granules is maximal between 10 s and 1 min post-stimulus (52–43%) and in the PMD phenotypes, early and late (45%, 62%), and the early AND phenotype (64%). The CDB contains no altered granules. By contrast, TPA induces 53% altered granules by 45 min at maximal histamine release intervals; average granule numbers in TPA-stimulated basophils do not decrease, in contrast to decreased values in FMLP-stimulated basophils. Quantitation of total vesicles (TVs) in the cytoplasm of FMLP- and TPAstimulated basophils showed interesting correlations [67, 104, 200]. Thus, TV number/␮m2 was greatest at 0 time after FMLP stimulation (3.02/␮m2) and preceded the onset of extensive PMD, characterized by altered granules. TVs fell to the lowest numbers in the CDB phenotype (0.33/␮m2), cells in which no PMD or regulated secretion by AND is ongoing. In TPA-stimulated cells, which did not develop extensive AND (the only time showing AND was at 30 min) but, instead, showed extensive PMD, vesicle numbers remained high. These quantitative studies, therefore, are supportive of a role for trafﬁcking vesicles as transport containers in secretory HBs.

Ultrastructure of Mast Cells and Basophils

172

7.4.4. Vesicular Transport of Charcot-Leyden Crystal Protein in Piecemeal Degranulation of Human Basophils

With time after stimulation of HBs, either with FMLP or TPA, the percentage of total cytoplasmic vesicles that were gold-loaded (indicating the presence of CLC protein) changed (ﬁg. 99C, D) [67, 199, 200]. CLC proteinloaded vesicles were signiﬁcantly increased at 10 and 20 s, 1 and 2 min, compared to unstimulated cells when HBs were stimulated with FMLP (ﬁg. 98C, 100) [200]. The FMLP-stimulated basophil phenotypes (ﬁg. 98) showed quite different percents of TVs that were carrying CLC protein [67]. For example, PMD-I, spanning 0 time to 1 min after stimulation, had 14% CLC protein-loaded vesicles. PMD-II spanned 10 s to 1 min after stimulation and had 7% goldloaded vesicles. The three stimulated phenotypes produced by granule extrusion – AND-I, AND-II, CDB – spanned 20 s to 10 min and had signiﬁcantly more CLC protein-labeled vesicles than unstimulated cells. The recovery phenotypes – RB-I, RB-II – spanned 2–10 min. While the early recovery phenotype (RB-I) had more vesicles loaded with CLC protein than unstimulated cells, the late recovery phenotype (RB-II) had returned to control %VG/TV (gold-loaded vesicles/total vesicles). The percent of CLC protein-positive vesicles in the PMD-I phenotype signiﬁcantly exceeded that in the PMD-II phenotype as expected, since the PMD-I phenotype is characterized by vesicle-rich basophils transporting packets of granule materials to the plasma membrane in the absence of granule extrusion, and the PMD-II phenotype is characterized by basophils experiencing greater vesicular trafﬁc in than out, leading to expanded and partially to completely empty granule chambers. The PMD-II phenotype has signiﬁcantly less VG/TV (indicating CLC protein transport) than the secretory AND-I phenotype. The AND-I and AND-II phenotypes are actively extruding entire granules; vesicular transport of CLC protein persists. The CDB phenotype is deﬁned by the virtual absence of secretory granules and ⬃14% of total cellular gold label for the CLC protein is located on the plasma membrane [67]. Endocytosis of this CLC protein source was reﬂected in a substantial peak (37%) of gold-loaded vesicles in CDB, cells that expressed the smallest number of available cytoplasmic vesicles (0.33/␮m2) of all stimulated phenotypes. The percent of CLC protein-loaded vesicles in the CDB phenotype was signiﬁcantly greater than that in the PMD-I and PMD-II phenotypes. The recovering phenotypes – RB-I, RB-II – contain substantial numbers of reconstituted particle and formed CLC-containing granules; the percent of CLC protein-loaded vesicles was signiﬁcantly less than that in the PMD-I, AND-I, AND-II and CDB phenotypes. Thus, such an analysis of basophil phenotypes stimulated by FMLP demonstrates vesicular transport of CLC protein to be a mechanism for effecting secretion in secretory phenotypes (PMD-I, PMD-II, AND-I, AND-II) as

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

173

Fig. 99. Percentage gold-labeled vesicles of the total vesicle population (%VG/V) in human peripheral blood basophils either unstimulated (UN) or stimulated with FMLP (A, C) or TPA (B, D), recovered for electron microscopy at the indicated times and prepared with a post-embedding enzyme-afﬁnity-gold method to detect histamine (A, B) or an immunogold method to detect CLC protein (C, D). Signiﬁcance values for each panel (compared to unstimulated cells) are as follows: A * p ! 0.001. B * p ! 0.025, ** p ! 0.05, *** p ! 0.001. C * p ! 0.005, ** p ! 0.001. D * p ! 0.05, ** p ! 0.01, *** p ! 0.01. Replicate samples examined for FMLP-stimulated histamine release (A, C) released 56% at 60 min, and for TPA-stimulated histamine release (B, D) released 70% at 60 min. [Adapted from 104, 199, 200 with permission.]

Ultrastructure of Mast Cells and Basophils

174

well as a mechanism for effecting restitution of granule stores of CLC protein by endocytosis in recovering phenotypes (CDB, RB-I, RB-II). The developmental kinetics of CLC protein-loaded vesicles in basophils stimulated by TPA showed progressively increased levels of gold-labeled vesicles that were signiﬁcantly greater than in unstimulated cells at 2, 5, and 10 min after stimulation (ﬁg. 99D, 101) [199]. As for FMLP-stimulated cells, levels of CLC protein-loaded vesicles decreased at later stimulation times (30 and 45 min). The mechanism of decline differs, however, for these two secretogogues. For example, decreased percent of VG/TV for CLC protein in FMLPstimulated cells is associated with reconstitution of large granule stores of CLC protein and, for TPA-stimulated cells, is associated with depletion of virtually all granule stores of CLC protein in emptied granule containers, as well as in particle-ﬁlled granules, remaining therein. 7.4.5. Vesicular Transport of Histamine in Piecemeal Degranulation of Human Basophils

Histamine transport in vesicles developing over the kinetic release reactions induced by FMLP and TPA in HBs was also quantitated. For the rapid releaser, FMLP, histamine-loaded vesicles exceeded those of unstimulated cells at all times examined (ﬁg. 99A, 102) [104]. For the slow releaser, TPA, histamine-loaded vesicles signiﬁcantly exceeded those of unstimulated cells at 1, 2, 10, 30 and 45 min (ﬁg. 99B, 103) [104]. Thus, for each secretogogue, histamine-loaded vesicles increased and remained elevated, compared to unstimulated cells, for the duration of the time examined. The mechanism for the persistence of histamine transport in vesicles differed, however, between these two different basophil secretogogues. Continued vesicular transport of histamine in FMLP-stimulated cells was associated with endocytosis of gold-labeled histamine from the plasma membrane in completely degranulated, agranular basophils (CDB) and RBs – phenotypes which populate the kinetic samples obtained between 20 s and 10 min [104]. The continued vesicular transport of histamine in TPA-stimulated cells at late times in the kinetic sequence was associated with ongoing evidence of PMD, virtually no ultrastructural evidence of secretion by AND, and no evidence of recovery [104]. 7.4.6. Comparative Analysis of Vesicle Transport of CLC Protein and Histamine in Stimulated Human Basophils

Comparisons of the %VG/TV (e.g., the proportions of TVs that are carrying histamine or CLC protein) between two unique secretogogues for HBs (ﬁg. 99) are of interest and are facilitated by ultrastructural cytochemical, morphometric and kinetic analyses [67, 104, 199, 200]. FMLP, an extremely rapid secretogogue, induces simultaneous peaks (%VG/TV) of vesicles loaded with

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

175

CLC protein or histamine; values return to control levels (in the case of CLC protein) by early recovery at 10 min (ﬁg. 99C). This coincides with reconstitution of formed CLCs within granules at 10 min [67, 200]. The %VG/TV for histamine does not return to baseline, however, but at 10 min, it is still elevated over unstimulated cells (ﬁg. 99A). Thus, the behavior of vesicle transport for these two cellular products differs considerably in FMLP-stimulated cells over a 10-minute interval. The elevated %VG/TV for histamine at 10 min most likely is a combination of uptake and synthesis [250], since histamine release, measured biochemically, is essentially over much earlier than at 10 min poststimulation with FMLP [328]. TPA (a slow secretogogue with morphological criteria indicating extensive and continuing PMD by 45 min, little evidence of AND, and none of recovery) shows a peak value of %VG/TV for CLC protein at 10 min and for histamine at 45 min, also corresponding to peak histamine release times for this trigger [329, 330]. As for FMLP, the %VG/TV for CLC protein drops extensively by 45 min (ﬁg. 99D), but the %VG/TV for histamine remains high (ﬁg. 99B). Since there is no morphologic evidence for recovery (and therefore synthesis) in TPA-stimulated basophils at 45 min, and little evidence for AND, the elevated vesicular trafﬁc for histamine likely reﬂects ongoing PMD. These ﬁndings are consistent with ultrastructural evaluations of granule contents at 45 min after TPA. For example, granule number is not reduced, and nearly 50% of the gran-

Fig. 100. FMLP-stimulated human peripheral blood basophils prepared for immunogold detection of CLC protein and studied at 20 s (A–C), 1 min (D–F) or 2 min (G–I) after activation are undergoing PMD (A, B), AND (D), or are completely degranulated basophils (C, E–G), and recovering basophils (H, I). In these high magniﬁcation micrographs, 30-nm gold-loaded empty vesicles and full vesicles (arrows) are present after stimulation for 20 s (A–C), 1 min (D–F), and 2 min (G–I) with FMLP. A The CLC protein-labeled empty vesicle is adjacent to an unaltered, particle-ﬁlled granule. B The gold-labeled empty vesicle is adjacent to an altered, partly particle-ﬁlled granule that has an expended, partly empty container. C Gold-loaded empty and full vesicles are in the granule-free cytoplasm between the gold-labeled plasma membrane and nucleus. D The basophil showing AND has an extruded granule (G) attached to the cell surface; 30-nm gold indicating CLC protein is bound to the underlying plasma membrane and peripheral cytoplasm. An empty vesicle nearby is also labeled but a full vesicle is not. Arrowheads indicated glycogen aggregates. E, F Two views of the same completely degranulated basophil show 30-nm gold particles in empty vesicles. A full vesicle is unlabeled. G–I At 2 min after stimulation, 30-nm gold-labeled full vesicles (arrows) are present in recovering (H, I) and completely degranulated (G) cells. I The granule contains a gold-labeled, formed CLC. The plasma membrane of this recovering cell is devoid of gold label. A !64,500. B !42,300. C !39,600. D !55,700. E !52,000. F !53,400. G !89,200. H !60,700. I !53,400. [From 200, with permission.]

Ultrastructure of Mast Cells and Basophils

176

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

177

Ultrastructure of Mast Cells and Basophils

178

ules are empty and are represented only by their granule-sized containers, and neither particle-ﬁlled nor empty containers have residual formed CLCs. Thus, a granule source for histamine continues to exist at 45 min post-stimulation, but one for CLC protein is virtually absent. 7.4.7. Comment regarding Basophil Secretion

In 1975, we proposed a degranulation model to explain progressive losses, occurring over days, of granule contents from HBs in experimentally induced contact allergy skin lesions [6, 290, 291]. We postulated that closely coupled endocytotic-exocytotic trafﬁc of small, smooth membrane-bound vesicles effected the emptying of secretory granule containers, in the absence of granule fusion and extrusion – a process characterized by the retention of granule containers of undiminished size in the cytoplasm. We postulated further that this steady-state secretion would be altered in an important way if either the rate or the amount of vesicular trafﬁc were changed. For example, we envisioned that a faster rate of vesicular trafﬁc would result in fusions of vesicles which would create channels between granules and plasma membrane, thus producing the anatomy of regulated secretion, or AND. The cytoplasmic channels that form could contain multiple membrane-free granules (degranulation sacs or channels) in situ as well as provide communication between a single granule and the plasma membrane – events necessary for exocytosis directly through membrane pores to the external milieu. We now present the evidence developed since the degranulation model was proposed in 1975 [6] in support of vesicular transport as a mechanism for effecting secretion from HBs. The evidence was collected in three ways: (a) direct inspection; (b) quantitation, and (c) direct labeling of expected vesicular cargo. By direct inspection, the existence of large numbers of vesicles of appropriate size was documented in basophils, and, in certain circumstances, fusion and/or budding of vesicles with/from large cytoplasmic secretory gran-

Fig. 101. TPA-stimulated human peripheral blood basophils prepared with immunogold to detect CLC protein and studied at 1 min (A), 2 min (B), 5 min (C), 10 min (D, E), and 15 min (F) after activation show gold-labeled vesicles in the cytoplasm (arrows). Some vesicles are electron-lucent; others contain particles similar to the particulate matrix of adjacent granules. CLC protein is also localized by gold particles in the nuclear matrix and nuclear membrane (C), the cytosol (A–F), the plasma membrane (A, E), a homogeneously dense primary granule (D), a formed CLC within the particulate matrix of a granule (E), and the granule membrane of an empty granule chamber (F). The arrowhead in (D) shows an example of the small minor granule population of human basophils [280] that does not contain label for CLC protein. A !62,700. B !59,900. C !39,000. D !39,900. E !42,300. F !77,000. [From 199, with permission.]

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

179

ules, termed granule-vesicle attachments, was also documented. Quantitation allowed documentation of rapidly changing numbers and contents of vesicles in HBs stimulated with different secretogogues over time and in morphological phenotypes of releasing and recovering basophils. Direct labeling of expected vesicular cargo was accomplished with ultrastructural immunogold and enzyme-afﬁnity-gold methods, which label the CLC protein and histamine, respectively. Quantitation of gold-loaded vesicular carriers in stimulated HBs directly conﬁrmed that releasing basophils transported these granule materials in cytoplasmic vesicles, as predicted by the degranulation model proposed in 1975 [6]. Classically, secretion biologists have categorized secretion as either regulated (granule exocytosis) or constitutive (vesicle transport without storage or

Fig. 102. FMLP-stimulated human peripheral blood basophils prepared with DAO-G to detect histamine (A–C, E, F, H–L) or with speciﬁcity controls (D, G) at 0 time (A–D), 10 s (E–G), 20 s (H), 30 s (I), 1 min (J, K) and 10 min (L) after activation. The open arrowhead in all panels indicates the cell surface; G = granule, EG = empty granule (typical for PMD), N = nucleus. At 0 time and in the same activated cell, both empty, electron-lucent (arrow, A) and full, electron-dense (arrow, B) vesicles are gold-labeled. Cytoplasmic glycogen particles are electron-dense (A). A nearly empty granule (typical for PMD) retains some DAO-G label (C). One cytoplasmic vesicle is also labeled. Note the vesicle attached to the empty granule. Electron-dense glycogen particles (arrows) are attached to the narrowed neck of the fused vesicle. Inset in C shows a free, electron-lucent vesicle, labeled with gold for histamine (arrow), encased in surrounding electron-dense glycogen aggregates, present in the cytoplasm of the same basophil. The speciﬁcity control (D) (DAO-G absorbed with histamine-agarose before staining) is negative for histamine in the large granule and adjacent vesicles. Electron-dense glycogen particles (arrows) remain visible. At 10 s (E), three subcellular sites are labeled with DAO-G as follows: (1) granule matrix (G) but not the intragranular CLC; (2) cytoplasmic, electron-lucent, perigranule vesicle (arrow); (3) polyamines in the nucleus (N). A vesicle is attached to the histamine-labeled granule (arrowhead). Also at 10 s (F), DAO-G extensively labels the granule matrix (G) of the full granule and shows less label in the empty granule (EG) typical for PMD. The speciﬁcity control (G) (digestion of the section with DAO before staining with DAO-G) is negative for histamine in the large granules (G), adjacent perigranular vesicles, and for polyamines in the nucleus (N). Electron-dense glycogen particles (arrow) remain visible in the cytoplasm. Granule particle-ﬁlled, DAO-G-labeled vesicles rest just beneath the plasma membrane (open arrowheads) at 20 s (H) and 30 s (I) post-stimulus. At 1 min (J), similar particle-ﬁlled cytoplasmic vesicles contain histamine. Also, at 1 min after stimulation (K), extrusion of granule particles to the cell surface (open arrowhead) is visible. The underlying cytoplasm contains gold-labeled electron-lucent vesicles (arrows). At10 min (L), DAO-G is attached to the cell surface (open arrowhead), within underlying electron-lucent and granule particle-ﬁlled vesicles, and to an adjacent recovered granule (G). A !56,000. B !88,800. C and inset !49,900. D !46,600. E !42,300. F, G !45,600. H–K !95,000. L !65,600. [From 104, with permission.]

Ultrastructure of Mast Cells and Basophils

180

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

181

stimuli) [52, 333, 334]. Our deﬁnition of PMD holds that stimuli produce secretion (in the absence of exocytosis of granules) that is mediated by vesicular transport of granule materials to the extracellular milieu. Speciﬁc stimuli that produce PMD in HBs are now being identiﬁed. Among these are FMLP, a bacterial peptide [67, 104, 200, 206], TPA, a tumor-promoting phorbol ester [104, 199, 292], IL-3, an IL with basophil activation properties [18], MCP-1, a chemokine [327], and a cytokine, rp21, an IgE-dependent histamine-releasing factor [327]. As understanding of the cell biology of basophil secretion has progressed, it has become apparent that newer paradigms for secretion in other cell systems have blurred the once classical deﬁnition of secretion [52, 333, 335–338]. Thus, evidence is accruing for non-granular secretion from acinar exocrine cells [335, 339], stimulated constitutive secretion that differs from non-stimulated basal secretion from endocrine cells [336] and a regulated secretory pathway in constitutive secretory cells [337]. Much more work remains to be done to develop a better understanding of the molecular mechanisms important to the regulation of vesicle-mediated events in HB secretion. The availability of sufﬁcient numbers of basophils (made possible by new isolation procedures and the identiﬁcation of basophil-

Fig. 103. TPA-stimulated human peripheral blood basophils prepared with DAOG to detect histamine, studied at 0 time (A), 2 min (B–D), 5 min (E–G), 10 min (H, I), 15 min (J), 30 min (K), and 45 min (L) after activation. A At 0 time, the basophil granule (G) is labeled with DAO-G. B At 5 min, three cytoplasmic vesicles near the cell surface (open arrowhead) are labeled for histamine. One gold-labeled vesicle is electron-lucent and is encased in electron-dense glycogen particles (closed arrowhead). Another gold-labeled vesicle also contains granule particles (long arrow), and a third gold-labeled vesicle is electron-lucent (short arrow). C At 2 min, note the gold-labeled vesicle ﬁlled with granule particles adjacent to the labeled granule (G). D Also at 2 min, the electron-lucent vesicle contains DAO-G. E–G At 5 min, the cell surface is indicated by open arrowheads. Note in E and F the granule particle-ﬁlled, DAO-G-labeled perigranular vesicles (open arrows) adjacent to the cell surfaces. The adjacent granules (G) also label for histamine; the granule in F has a focal, electron-lucent region (arrow) typical for PMD. Cytoplasmic glycogen particles (closed arrows, E) are electron-dense and generally larger than the ⬃20-nm gold label. G At 5 min, a DAO-G-labeled vesicle is fused to the cell surface (open arrowhead). H, I At 10 min, electron-lucent (H) and particle-ﬁlled (I) cytoplasmic vesicles (arrows) contain histamine. In H a large, empty granule (EG) (devoid of gold label and granule particles) is typical for PMD. In I the cell surface is coated with CF (used in cell processing for electron microscopy) and is indicated by the open arrowhead. J–L The granule particle-ﬁlled cytoplasmic vesicles are labeled with DAO-G (arrows) at 15 (J), 30 (K), and 45 min (L) after TPA activation. The cell surface is indicated by open arrowheads in K and L; an empty granule (EG) typical for PMD is present at 45 min post-stimulation in L. A, B !52,300. C !111,600. D !65,600. E, K !67,900. F !58,000. G !76,000. H !64,100. I, J !107,400. L !51,300. [From 104, with permission.]

Ultrastructure of Mast Cells and Basophils

182

Piecemeal Degranulation of Basophils and Mast Cells Is Effected by Vesicular Transport of Stored Secretory Granule Contents

183

speciﬁc growth factors), combined with strong support for vesicle-mediated basophil secretion from ultrastructural studies, will facilitate this work. Much has been learned about cytosolic and membrane factors necessary for the function of vesicular machines in cell biology [340–344]. These studies involve a wide-ranging number of systems and processes beyond the scope of this review. Principles determined by this large body of work should, however, prove valuable, particularly when they are geared to studies designed to better understand the role of vesicle transport in the cell biology of HBs.

Ultrastructure of Mast Cells and Basophils

184

Chapter 8

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Mast Cell-Derived Mediators of Enhanced Microvascular Permeability, Vascular Permeability Factor/ Vascular Endothelial Growth Factor, Histamine, and Serotonin, Cause Leakage of Macromolecules through a New Endothelial Cell Permeability Organelle, the Vesiculo-Vacuolar Organelle

8.1. Overview Proof of principle for a vesicular transport mechanism for effecting PMD of mast cells and basophils was offered in Chapter 7. We have also proposed that upregulation of this vesicular trafﬁc would result in vesicular fusions, forming channels (sacs) in continuity with the plasma membrane and functionally providing a conduit for granule extrusion. This well-known mode of secretion, when stimulated by IgE mechanisms in basophils and mast cells, is called anaphylactic degranulation (AND) [see Chapter 9]. Release of powerful mediators of allergic inﬂammation occurs with either of these modes of secretion. Mast cells traditionally can be found in perivascular locations and basophils, given appropriate stimuli, rapidly leave the blood vascular space by a transendothelial cell route to enter the perivascular milieu [276, 345]. Mediators released from these cells have powerful permeabilizing properties. We used ultrastructural methods to deﬁne the trafﬁc route(s) of macromolecular leakage from the microvasculature exposed to the mediators released from basophils and mast cells in allergic inﬂammation and from tumor cells in the tumor stroma. These morphologic and functional studies deﬁne a new perme-

ability organelle in endothelial cells which we have called the vesiculo-vacuolar organelle (VVO) (ﬁg. 104–106) [13–15].

8.2. What Are Vesiculo-Vacuolar Organelles (VVOs)? VVOs [13] are large collections of vesicles and vacuoles focally distributed in venule endothelial cytoplasm, often in a parajunctional location (ﬁg. 104, 105). In standard, single electron microscopic sections (⬃70- to 80-nm), their individual components number from several to hundreds [13, 14]. Connections to luminal, abluminal and lateral plasma membranes can be found in these single sections. In some single sections these were interconnected as complex transcellular structures. When the individual parts of VVOs in control vessels (ﬁg. 105, 106) and tumor vessels were quantiﬁed, they were the same [13]. Morphometrics (shape-factor analysis) showed that the shape of individual units of VVOs was more variable in tumor-associated vessels than in control vessels [13, 14]. The size of individual vesicles and vacuoles of tumor and control venular VVOs was more variable than the more uniform size of caveolae in capillaries [15]. An analysis of size distributions of the vesicles and vacuoles comprising VVOs uncovered a modal distribution [108, 346–348] comprised of multiples of the smallest, unit-sized vesicle, such that the increasing size of each mode could be accounted for by fusions of the smallest vesicles [349]. This unit vesicle size was identical to that of capillary caveolae [349]. Thus, a powerful mathematical analysis supports formation of VVOs in tall venular endothelium from multiple fusions of individual caveolae – a process that is generally lacking in thin capillary endothelia. Direct examination of single ultrastructural sections of standard thickness allows one to deﬁne individual anatomic components of the vesicles and vacuoles that comprise VVOs. Thus, they open to endothelial cell surfaces (luminal, abluminal, lateral) by stomata which are guarded by a thin, electron-dense diaphragm (ﬁg. 106) [13, 14]. Interconnected parts also are joined by stomata and diaphragms, and larger structures have multiples of such structures. Narrow necks provide ⬃11-nm-wide channels from stomata to single vesicles or between stomata of adjacent vesicles and vacuoles [14]. These substructures Fig. 104. Cross section of a venule from an inﬂamed eyelid of an IL-4 transgenic mouse. Two pericytes (P) embrace this vessel; two endothelial cell nuclei (N) are evident. Portions of ﬁve endothelial cells line the venule; individual endothelial cells are connected by interendothelial cell junctions (solid arrowheads), none of which show gaps. Focal clusters of vesicles and vacuoles, termed VVOs (open arrowheads), span the cytoplasm of individual endothelial cells that comprise this vessel. L = lumen. Bar = 3 ␮m. [From 249, with permission.]

Ultrastructure of Mast Cells and Basophils

186

Mast Cell-Derived Mediators of Enhanced Microvascular Permeability Cause Leakage of Macromolecules through a New Endothelial Cell Permeability Organelle, the Vesiculo-Vacuolar Organelle

187

Fig. 105. Control mouse skin venule. Numerous collections of cytoplasmic vesicles and vacuoles constitute the vesiculo-vacuolar organelle (VVO) (open arrowhead). Note that the interendothelial cell junctions show complex interdigitations; electron-dense tight junction areas are located on the luminal aspect of the lateral surfaces of endothelial cells (closed arrowheads). !19,200. [From 357, with permission.]

of VVOs – stomata, necks, diaphragms – are all morphologic characteristics of caveolae [13, 14, 350–356]. Central electron-dense ‘knobs’ within the diaphragms occluding stomata (ﬁg. 106) also characterize VVOs and caveolae [13, 14, 350, 351]. Thus, considerable morphologic similarities exist in caveolae and VVOs [13–15]. Earlier nomenclature for several of these caveolar structures was as follows: (i) caveola = plasmalemmal vesicle; (ii) stoma = fenestra. We use stoma, rather than fenestra, to avoid confusion with the fenestra which typiﬁes thin, fenestrated endothelia. The latter structures are also guarded by thin diaphragms but are considerably larger than the stomata of caveolae and VVOs. Ultimately, functional and developmental analogies may become apparent for both types of fenestrae, the stomata of VVOs and caveolae in continuous endothelia, and the fenestrae of fenestrated endothelia [349]. VVO formation is not inﬂuenced by variable methods of ultrastructural ﬁxation and processing. For example, VVOs are readily illustrated in mate-

Ultrastructure of Mast Cells and Basophils

188

Fig. 106. Control mouse skin venule endothelial cell shows portions of a VVO near the lumen (L). Individual vesicles and vacuoles are connected by thin diaphragms (closed arrow) and display electron-lucent stomata, many of which contain an electron-dense knob centrally (arrowhead). Numerous ribosomes ﬁll the cytoplasm. !109,000. [From 357, with permission.]

rial that is ﬁxed by immersion or perfusion at the natural blood pressures of animals used, by a rapid microwave ﬁxation protocol [357], in frozen thick sections processed for pre-embedding immunocytochemistry [358], and in frozen thin sections prepared for post-embedding immunocytochemistry [357]. Therefore, it is extremely unlikely that these structures reﬂect artifacts of ﬁxation and processing [359].

Mast Cell-Derived Mediators of Enhanced Microvascular Permeability Cause Leakage of Macromolecules through a New Endothelial Cell Permeability Organelle, the Vesiculo-Vacuolar Organelle

189

8.3. Serial Section Analysis Demonstrates the Transendothelial Cell Pathway Provided by VVOs We examined a large number of ultrathin (12- to 15-nm) serial sections and computer-assisted three-dimensional reconstructions of VVOs (ﬁg. 107, 108) [15]. These studies dispelled any further doubt that these complex organelles connected all three surfaces of endothelial cells in venules of control animals and in tumor vessels of mice, rats, and guinea pigs. Earlier documentation to support their interconnectedness was derived from electron-dense tracer studies (see later) and from specimen-tilting series of electron microscopic sections [13, 14]. The serial section analysis [15, 349] revealed that less than 1% of endothelial vesicles were free in the cytoplasm; VVOs were comprised of a large number of individual parts and occupied large segments of the cytoplasm of endothelial cells.

8.4. Electron-Dense Ultrastructural Tracer Analysis Demonstrates the Transendothelial Cell Pathway Provided by VVOs We ﬁrst probed VVO function using electron-dense tracers [351, 352, 360–363] in experimental animal tumor models that were known to produce vascular permeability factor/vascular endothelial growth factor (VPF/ VEGF) – a vascular permeabilizing factor known to be 50,000 times more potent than histamine [364–370] – and in normal controls (ﬁg. 109) [13–15]. Soluble tracers of different sizes and properties [anionic ferritin (FE), cationic ferritin (CF), and horseradish peroxidase (HRP)] and timed collections of samples after intravenous delivery of tracers constituted the study material. In some models, serial sections and three-dimensional reconstructions were done [15]. Altogether, these studies showed the rapid transit of tracers from the vascular space to the extracellular space [13–15]. HRP (⬃5-nm) passaged tall venule endothelial cell VVOs within seconds to be released as small vesicle-sized packets of electron density into the basal lamina (ﬁg. 109B–D) [13, 14]. These small bursts of HRP, no longer conﬁned to vesicles, assumed irregular shapes of reactivity in this location (ﬁg. 109D) that clearly anteceded the later (⬃5-min) passage of HRP in a paracellular route through junctions that did not show gap formation. As HRP traveled through interconnected VVOs, it could pass transcellularly through one endothelial cell and then take a transcellular route through two adjacent endothelial cells by crossing unaltered endothelial cell junctions through intercommunicating VVOs above or below the tight areas of endothelial cell junctions. Within VVOs, HRP ﬁlled open stomata (seen as electron-dense stomata) (ﬁg. 109C) or

Ultrastructure of Mast Cells and Basophils

190

Fig. 107. Two consecutive ultrathin (12-nm) sections, from a series of 61, of a control mouse skin venule. The mouse had been injected intravenously with FE 5 min prior to sacriﬁce and intradermally with Hanks’ balanced salt solution 1 min prior to sacriﬁce. Together these sections demonstrate direct, open communication between VVOs present in adjacent endothelial cells. A vesicle in the left-hand cell (A) connects with the intercellular cleft which in turn connects successively with vesicles and vacuoles (B–G) in the right-hand cell. Diaphragms close the stomata linking some adjacent vesicles/vacuoles (for example, B–C, C–D, and D–F in panel B). Closed arrow indicates area of tight junction near the luminal surface. Scattered FE particles are present in the vascular lumen (L) and in the vesicle (open arrow) immediately above vesicle labeled B. R = Red blood cell; BL = basal lamina. A, B !63,600. [From Feng et al, J Physiol (Lond) 1997; 504(3): 747–761, with permission.]

was restricted by closed stomata (seen as electron-lucent structures) (ﬁg. 109B) and easily passed through the narrow neck connecting VVO stomata. FE (⬃11-nm) has an anionic charge and is readily visible as small particles by electron microscopy (ﬁg. 109E), thus allowing quantiﬁcation of a soluble macromolecular tracer by EM [13–15]. FE moved more slowly through VVOs in tumor-bearing sites than did HRP, as expected from its larger size. At early

Mast Cell-Derived Mediators of Enhanced Microvascular Permeability Cause Leakage of Macromolecules through a New Endothelial Cell Permeability Organelle, the Vesiculo-Vacuolar Organelle

191

times (⬃1 min), VVO components on the luminal endothelial cell front contained FE which passed to mid-cytoplasm and basilar fronts over a 30-min period. Intercellular junctions never contained FE in transit from lumen to abluminal sites, except in rare instances of VVO connections to the wider adherans areas of junctions. As with HRP, individual open stomata contained the electron-dense particles of FE (ﬁg. 109E), and in closed stomata, individual FE particles remained on the luminal side of the diaphragms, seen in cross-sectional view, with no passage to the conjoined vesicle or vacuoles. FE also passed between vesicles and vacuoles through the narrow necks between them. Both of these tracers in kinetic experiments traversed this transcellular route through VVOs but with different kinetics (dictated by their size) in tumor-associated venules with continuous endothelium. They also traversed VVOs similarly in control venules, but, using a probe with quantiﬁcation possibilities (FE), the passage was far less in non-leaky control venules [13, 15]. Rather, active endocytosis of FE in coated vesicles and ﬁlling of multivesicular bodies (MVBs) in endothelial cell cytoplasm (both being components of the endocytotic organellar system) was vigorous [14].

8.5. VVOs in Tumor Vessels VPF/VEGF was originally discovered in the late 1970s because of its capacity to increase the permeability of microvessels to plasma [364, 365]. A potent vascular permeabilizing protein in tumor culture supernatants [364] was puriﬁed and named VPF [367]. Subsequently, endothelial cell mitogenic activity was found to be mediated by the same molecule, resulting in the designation of VPF/VEGF [371–374]. The permeabilizing cytokine, VPF/VEGF, is among the most potent vascular permeabilizing agents known, having a potency some 50,000 times that of the mast cell- and basophil-derived mediator, histamine Fig. 108. Computer-generated three-dimensional reconstruction of portion of a venular VVO from mouse skin injected 5 min earlier with 50 ng VPF. The interior volumes of VVO vesicles and vacuoles in successive electron micrographs were traced onto transparent overlays with reference marks to retain register. Tracings were digitized at a resolution of 5.9 nm/pixel. Since section thickness (15-nm) was greater than in-plane resolution, bicubic interpolation was performed between sections, and convolutions were used to smooth surfaces. A–F Portion of a VVO (15 consecutive serial, ultrathin EM sections illustrating 25 individual vesicles-vacuoles) reconstructed with Advanced Visual Systems (Waltham, Mass., USA) software, here viewed in successive rotations around a horizontal axis at intervals of 30° (except 15°, C–D). There are two openings (E, F) to the vascular lumen and four to the abluminal surface (A). [From 15, with permission.]

Ultrastructure of Mast Cells and Basophils

192

[366, 370, 375, 376]. We now know that mast cells also synthesize and store VPF/VEGF [377, 378]; thus, secretory stimuli might enhance vascular permeability by several mediators released from mast cells. We observed [13] that the predominant pathway for electron-dense tracers to leak into the subcutis from the blood vascular space in vessels in tumors secreting VPF/VEGF was by way of VVOs (ﬁg. 109B–E) [14]. VVOs also occurred in the venules of the normal

Mast Cell-Derived Mediators of Enhanced Microvascular Permeability Cause Leakage of Macromolecules through a New Endothelial Cell Permeability Organelle, the Vesiculo-Vacuolar Organelle

193

subcutis of animals not bearing tumors, and these structures represented the predominant pathway by which tracers exited these normal vessels. However, both VVO labeling and tracer extravasation were much greater in tumor vessels than in control vessels (p ! 0.001). Quantiﬁcation of the individual vesicle and vacuoles that comprise VVOs in tumors compared with normal endothelial cells showed that the perimeters and areas of tumor vessel vesicles and vacuoles signiﬁcantly exceeded that of normal endothelial cells (ﬁg. 110). Morphometry revealed that VVOs were enormous cytoplasmic structures (median area 0.12–0.14 ␮m2 in a single electron micrograph). Specimen tilting provided evidence that individual VVO vesicles and vacuoles communicated with each other and with the plasma membrane of endothelial cells by stomata, some of which were closed by diaphragms composed of a single membrane. High magniﬁcation images with two tracers, ferritin (ﬁg. 109E) and HRP (ﬁg. 109C), revealed that the passage of

Fig. 109. Ultrastructural tracer studies of VVO permeability in mouse skin venules. A Several VVOs ﬁll the parajunctional endothelial cytoplasm in this control mouse ab-

dominal skin venule. L = Lumen; Weibel-Palade body (arrowhead). B–D Tumor vessel endothelial cells in mouse skin venules from mouse ovarian tumor (MOT)-carrying animals show VVOs at 10 s (B, D) and 5 min (C) after injecting horseradish peroxidase (HRP) i.v. Sections were reacted for HRP reaction product by cytochemistry but were otherwise unstained. In B, continuity of VVO vesicles with the blood vascular space (top) is made evident by electron-dense reaction product. Reaction product also extends to the basal lamina, focally (arrow). The interendothelial cell junction does not contain HRP (arrowhead). In D, an individual vesicle of partially HRP-positive VVO is fused with the abluminal plasma membrane and has released a cloud of HRP adjacent to the basal lamina and tissue front (arrow). Many endothelial cell vesicles are ﬁlled with dense reaction product similar in intensity to the overlying HRP-ﬁlled vessel lumen. C An open, HRP-containing stoma of a vesicle in contact with the abluminal plasma membrane. The stoma contains dense HRP, and the vesicle lumen is faintly HRP-positive. The abluminal surface of the endothelial cell plasma membrane is focally stained with HRP beneath this vesicle, but the basal lamina (BL) remains unstained (compare the HRP-positive stoma of this abluminal vesicle with the FE-positive stoma of a similar abluminal vesicle in E). These images indicate passage of tracers by open stomata in continuity with other vesicles of VVOs at other levels before the vesicle lumen itself becomes extensively tracer-positive (or, alternatively, after the vesicle chamber proper has emptied of tracer). Such images support interconnected, patent passages throughout VVOs that span tumor vessel endothelia. E A tumor vessel endothelial cell VVO vesicle with a stoma that contains FE is seen 30 min after i.v. injection of FE in a lead citrate-stained sample of skin venule from a MOTcarrying mouse. A central stoma is ﬁlled with FE particles. The vesicle chamber contains little FE. The vesicle in E is bounded by a trilaminar unit membrane and is connected to the underlying plasma membrane by an elongate, narrow neck (arrows). Several FE particles are present in the basal lamina (BL). A !22,800. B !45,600. C Bar = 80 nm. D !57,000. E Bar = 63 nm. [A from 358, with permission; B–E from 14, with permission.]

Ultrastructure of Mast Cells and Basophils

194

macromolecules through VVOs occurred through stomata and was regulated at the level of stomatal diaphragms, thereby demonstrating a mechanism for controlling the passage of macromolecules across endothelial cells. Compared with tumor endothelial cells, little circulating ferritin and HRP entered the VVOs of normal endothelial cells because stomata joining vesicles and vacuoles to each other and to the lumen and ablumen were closed. These ﬁndings indicate that VVOs provide a major pathway for the extravasation of circulating macromolecules across endothelial cells taller than capillary endothelial cells, and suggest that upregulated VVO function accounts for the well-known hyperpermeability of tumor blood vessels (ﬁg. 111).

Mast Cell-Derived Mediators of Enhanced Microvascular Permeability Cause Leakage of Macromolecules through a New Endothelial Cell Permeability Organelle, the Vesiculo-Vacuolar Organelle

195

Fig. 110. A total of 521 vesicles and vacuoles from tumor endothelial cells and 487 from control endothelial cells were measured. The mean areas and perimeters of tumor endothelial cell vesicles-vacuoles were signiﬁcantly larger than in normal endothelial cells. [From 357, with permission.]

8.6. VVOs in Inflammatory Eye Disease Vessels We have shown that eyelid lesions that develop in transgenic mice that overexpress IL-4 [312, 317] contain increased numbers of mast cells, that these mast cells exhibit PMD [249; see Chapter 7] and that the microvessels in the eyelid lesions have prominent VVOs (ﬁg. 104) [249]. We demonstrated the ultrastructural features of mast cell PMD using light and electron microscopic methods that we have previously used to examine PMD of human and animal mast cells in a wide variety of diseases and experimental models [see Chapter 7]. We initially described PMD in basophils that had inﬁltrated experimentally induced and sequentially biopsied cutaneous contact hypersensitivity lesions in humans [290, 291]. As characterized by light and electron microscopy, PMD of basophils evolved gradually over 3 days after elicitation of the reaction. Accordingly, we postulated that PMD represented a mechanism for the slow release of basophil or mast cell granule contents, and thus differed from the much more rapid secretion seen in classic IgE-mediated AND

Ultrastructure of Mast Cells and Basophils

196

Fig. 111. Diagram showing two VVOs within an endothelial cell. The complex transcytoplasmic pathway linking the vascular lumen with the endothelial cell basal lamina is illustrated in both; one (left) opens to the interendothelial cell junction above its narrow mid portion. Small circular structures within vesicles and vacuoles are stomata; some of these exhibit dense central knobs. [From 14, with permission.]

of mast cells and basophils [see Chapter 9]. Since these early studies, it has become apparent that the most frequently observed ultrastructural evidence of function detected in the mast cells and basophils in biopsies of human disease are those of PMD [169]. We also studied inﬂamed eyelid tissues of the IL-4 transgenic mice, using an enzyme-afﬁnity-gold method that we recently developed to localize histamine at the ultrastructural level [154]. This method is based on the afﬁnity of DAO, which binds to its substrate, histamine, even after conjugation to electron-dense gold particles [154]. Moreover, DAO-G is used as a post-embedding stain on routinely and optimally prepared ultrastructural samples, allowing excellent preservation of ﬁne structural anatomy. We found that DAO-G stained the unaltered electron-dense granules of normal MMCs, but staining was markedly reduced or absent in the swollen, altered granules of the mast cells that exhibited evidence of PMD. These ﬁndings represent the ﬁrst ultrastructural evidence for the secretion of histamine from mast cells in vivo [249; see Chapter 7].

Mast Cell-Derived Mediators of Enhanced Microvascular Permeability Cause Leakage of Macromolecules through a New Endothelial Cell Permeability Organelle, the Vesiculo-Vacuolar Organelle

197

8.7. Permeability through VVOs in Normal Vessels Is Induced by VPF/VEGF, Histamine, and Serotonin We sought to recapitulate the functional permeability changes (and anatomy reﬂecting them) produced in the associated microvasculature of VPF/ VEGF-producing tumors by directly injecting VPF/VEGF, histamine, or serotonin into appropriate sites of control animals and by sampling for electron microscopy at multiple times after injecting the tracer, FE (ﬁg. 112) [15, 349]. In some cases we examined skin samples, in others, the cremasteric muscle was studied. These studies yielded a number of important ﬁndings. Increased permeability for all three mediators occurred in skin and cremasteric muscle of either guinea pigs, mice and rats. The FE tracer clearly demonstrated the leakage to occur transcellularly through VVOs (ﬁg. 112), which showed a gradient of decreasing FE concentrations from lumina to endothelial cell to extracellular space. Widened junctions or junction ‘gaps’ did not account for the leakage of FE to the extracellular space; increased formation of fenestrated endothelia did not occur in these rapid tracer studies. Thus, three potent permeability mediators cause transcellular vascular leakage of macromolecules via the VVO route in control microvascular beds [15, 349], and increased fenestrated vessels were not rapidly induced by VPF/VEGF [379–382].

8.8. Stomatal Diameters in Venule VVOs and Capillary Caveolae Respond Differently when Exposed to Permeabilization We measured stomatal diameters within VVOs in control venules and within caveolae of control capillaries after the injection of VPF/VEGF, histamine or serotonin, and found that they were increased after exposure to permeabilizing agents, compared to saline injections, in VVOs, but that the stomata of caveolae in capillary endothelial cells did not change size [349]. These measurements lend credence to the hypothesis that VVO stomatal-restraining diaphragms are altered/lost in the presence of permeabilizing mediators, giving rise to an enhanced transcellular sieve for the passage of macromolecules [349].

8.9. Ultrastructural Localization of Receptors and Ligands to VVOs What is the evidence that VVOs display receptors and ligands important to permeability functions? Several specialized ultrastructural approaches have been devised to investigate this issue. These include the development of a new

Ultrastructure of Mast Cells and Basophils

198

Fig. 112. Mediator-induced VVO leakage demonstrated with FE in control skin venules. A Extravasation of FE by way of VVOs in guinea pig skin venules 20 min after intradermal injection of VPF/VEGF (A) or 10 min later in response to an intradermal injection of histamine (B). In A, note FE particles entering VVO vesicles-vacuoles that open to the vascular lumen (V). B FE has progressed to mid-cytoplasmic VVO components. A few FE particles have spilled into the subendothelial spaces in A and B. A !90,300. B !91,200. [From 349, with permission.]

enzyme-afﬁnity-gold probe to localize histamine in routine preparations of EM samples [154; see Chapter 6], and utilization of immunoperoxidase and immunonanogold ultrastructural methods to detect the ligand VPF/VEGF [383] and its receptor, VPF/VEGFR-2 [358].

Mast Cell-Derived Mediators of Enhanced Microvascular Permeability Cause Leakage of Macromolecules through a New Endothelial Cell Permeability Organelle, the Vesiculo-Vacuolar Organelle

199

8.9.1. Histamine, a Potent Permeability-Producing Ligand, Binds to VVOs

Our long-term interest in allergic inﬂammation prompted studies of the inﬂammatory eye disease that develops in the eyelids of transgenic mice overexpressing IL-4 [312]. These lesions displayed increased numbers of mast cells, many of which were undergoing a form of secretion termed piecemeal degranulation (PMD) [249]. This process is deﬁned by the partial to complete transport of secretory granule contents in vesicles to the extracellular milieu, characteristically leaving membrane-bound empty granules in place in the cytoplasm of mast cells [Chapter 7]. Venules in the immediate vicinity of mast cell secretion contained large collections of VVOs (ﬁg. 104) [249]. These were also evident remote from secretory mast cells. By using the enzyme-afﬁnity-gold method [154], which stains histamine based on binding to DAO-G, we showed that mast cells had released much of their histamine content [249]. Individual endothelial cell vesicles and vacuoles comprising VVOs in these eye lesions bound the gold reagent, indicating histamine, to their stomata (ﬁg. 113C); interendothelial junctions did not bind the probe, indicating the absence of histamine [249]. Gaps at junctions were not present [249]. Gap formation of interendothelial cell junctions has been reported to occur when exogenous histamine is placed on tissues [384–386]. Our studies, using similar methods, did not demonstrate such gaps. Macromolecular leakage probably occurred through VVOs in the venules of the eye model of allergic inﬂammation, an in vivo model of chronic release of histamine [249]. The histamine that was bound to VVO stomata in inﬂammatory eye disease probably indicates the presence of histamine receptors in this location that bind histamine and change the permeability properties of histamine-binding VVOs in allergic inﬂammation. 8.9.2. VPF/VEGF, a More Potent Permeability-Producing Ligand, Binds to VVOs

Another long-term interest of ours has been to deﬁne the anatomic mechanism(s) of the enhanced permeability of tumor vessels [366–370, 387]. This interest led to the discovery of a protein produced by tumors, one with permeabilizing power vastly in excess of that of histamine, which we named

Fig. 113. Ligands localized to VVOs. A Ascites mouse ovarian tumor (MOT)-induced peritoneal wall microvascular endothelia illustrating VPF/VEGF-positive VVOs. B VVO structures are not stained in a control for this pre-embedding immunoperoxidase technique. L = Lumen. C IL-4 Tg mouse sample from inﬂamed eyelid, prepared with the DAO-G technique to illustrate histamine, shows gold particles precisely bound to a stoma within a VVO (arrow). A, B Bar = 276 nm. C !63,000. [A, B From 357, with permission; C from 249, with permission.]

Ultrastructure of Mast Cells and Basophils

200

Mast Cell-Derived Mediators of Enhanced Microvascular Permeability Cause Leakage of Macromolecules through a New Endothelial Cell Permeability Organelle, the Vesiculo-Vacuolar Organelle

201

vascular permeability factor (VPF) [364–370]. Subsequently, emphasis on this protein family’s angiogenic properties led to another name – vascular endothelial growth factor (VEGF) [371, 373, 374, 388, 389]; the protein family is now often referenced as VPF/VEGF. This cytokine is produced by many animal and human tumors, activated macrophages, keratinocytes, rheumatoid synovial cells, embryonic tissues, mast cells, neutrophils, and eosinophils, and by cultured epithelial and mesenchymal cell lines [370, 387]. VPF/VEGF accumulates in the microvessels supplying tumors and certain inﬂammatory reactions in which VPF/VEGF is also overexpressed [390–392]. We used pre-embedding immunoperoxidase to detect the precise localization of VPF/VEGF by electron microscopy in tumor microvessels [383]. Intense immunoreactivity was detected on the abluminal plasma membrane of tumor-associated microvascular endothelial cells and in their VVOs (ﬁg. 113A). Some individual stomata were also labeled for VPF/VEGF, but many were not stained. It is likely that the VPF/VEGF staining represents binding of this ligand to its receptor(s) in VVOs. In contrast to labeling of abluminal plasma membrane and VVOs, endothelial cytoplasmic organelles (such as MVBs and Weibel-Palade bodies) and basal lamina were not immunoreactive for VPF/VEGF. Some VPF/ VEGF staining of the luminal plasma membranes was also noted, but the lateral plasma membranes, which comprise the intercellular junctions and clefts, did not bind VPF/VEGF. Thus, the electron microscopic evidence supports a gradient of cytokine activity from ablumen (closest to the VPF/VEGF-producing tumor cells) to lumen, and in VVOs (where macromolecular leakage also occurs) but not to intercellular junctions where macromolecular leakage does not occur [383]. 8.9.3. VPF/VEGFR-2, a High-Afﬁnity Receptor for VPF/VEGF, Is Present in VVOs

We next looked for the site(s) of a high-afﬁnity receptor for VPF/VEGF, the VPF/VEGFR-2 receptor (fetal liver kinase-1, or ﬂk-1, in rodents; kinase insert domain-containing receptor, or KDR, in humans [393–399]) in three models [358]. Since post-embedding methods of detection failed, we adapted a pre-embedding immunoperoxidase and immunonanogold method to examine (i) normal mouse kidney [VPF/VEGF is strongly expressed by glomerular epithelial cells (podocytes) and tubular epithelial cells] [390], (ii) mouse mammary carcinoma, and (iii) normal mouse angiogenesis sites induced by an adenoviral vector expressing murine VPF/VEGF164 (ADvpf) [400]. In all three models, the VPF/VEGFR-2 was evident by light microscopic immunoreactivity, allowing selection of areas to examine at high magniﬁcation by electron microscopy [358]. By doing so, we precisely localized the receptor for this ligand on glomerular endothelial cells but not on glomerular epithelial cells. Thus, podo-

Ultrastructure of Mast Cells and Basophils

202

cytes and endothelium, separated only by their respective basal laminae in glomerulae, showed precise and separate localization of the ligand, VPF/VEGF, to podocytes, and for the receptor, VPF/VEGFR-2, to the thinned, fenestrated endothelial cells which characterize the glomerular endothelium [358, 390]. Using either nanogold (ﬁg. 114B–D) or peroxidase (ﬁg. 114A) as reporters, equivalent immunoreactivity for the VPF/VEGFR-2 was observed on both the luminal and the abluminal surfaces of tumor- and adenovirus-induced vascular endothelium, but plasma membranes at interendothelial cell junctions were spared except at sites connected to VVOs [358]. VPF/VEGFR-2 was also localized to the membranes and stomatal diaphragms of VVOs (ﬁg. 114A). This staining distribution corresponds to that for VPF/VEGF and is consistent with a model in which VPF/VEGF increases microvascular permeability, after binding to its receptor, VPF/VEGFR-2, by opening VVOs to allow the transendothelial cell passage of plasma and plasma proteins. Thus, considerable evidence exists implicating the presence of ligands (histamine, VPF/VEGF) with extensive permeabilizing power and a receptor (VPF/VEGFR-2) for VPF/ VEGF to VVOs in endothelial cells [249, 358, 383].

Mast Cell-Derived Mediators of Enhanced Microvascular Permeability Cause Leakage of Macromolecules through a New Endothelial Cell Permeability Organelle, the Vesiculo-Vacuolar Organelle

203

Fig. 114. Immunocytochemical studies of VVOs. A Unstained immunoperoxidase preparation localizes VEGFR-2 to endothelial structures in TA3/St mouse mammary adenocarcinoma 9 days after injection of 1 ! 106 tumor cells into the subcutaneous space of mice. VEGFR-2 is focally present (as determined by electron-dense peroxidase reaction product) on luminal and abluminal plasma membranes and on membranes of VVOs. Some caveolae attached to luminal plasma membranes show VEGFR-2 associated with the diaphragms that close these structures (arrows). B–D Immunonanogold preparations localize VEGFR-2 to endothelial structures in angiogenic sites induced 4 days earlier in the skin of nude mice by adeno-VPF/VEGF (B, D) or at 5 days in muscle (C). The silverenhanced gold label indicating VEGFR-2 is associated with VVOs (B–D). L = Lumen. A !128,000. B Bar = 130 nm. C Bar = 70 nm. D Bar = 80 nm. [From 358, with permission.]

Ultrastructure of Mast Cells and Basophils

204

Chapter 9

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

9.1. Overview In 1991, we reviewed the ultrastructural studies of degranulation and recovery from degranulation of rat, human and mouse mast cells, and of guinea pig and human basophils [7]. That review provided ultrastructural evidence for two secretory modes, AND and PMD, and images of the recovery from stimulated AND of HMCs and guinea pig basophils in short-term cultures. In the intervening years we have examined new models of regulated secretion and recovery in these interesting cells. We review these models here using morphologic endpoints as well as those which utilize substructural localizations of the CLC protein, histamine, and heparin over the kinetic course of release and recovery events.

9.2. Murine Mast Cells 9.2.1. Spontaneous Degranulation of Cultured Bone-Marrow-Derived Immature Mast Cells from X-Linked-Immunodeﬁcient Mice

IgE-mediated, antigen-dependent stimulation of immature MMCs cultured in IL-3-containing media produces secretion by granule exocytosis [49]. Similar cultured mast cells were derived from X-linked-immunodeﬁcient (Xid) mice and examined by electron microscopy. In these cultures, Xid mast cells

were also immature [401]. In contrast to cultures obtained from control mice, 10–20% of the immature mast cells of Xid origin were undergoing secretion by granule extrusion in the absence of any secretogogue (ﬁg. 115). Spontaneous secretion may be related to disordered tyrosine kinase function and/or signal transduction pathways in the Xid mouse. 9.2.2. Anaphylactic Degranulation of Beige Rat Mast Cells

Giant secretory granules, characteristic of the Beige mutation in RMCs, became swollen, displayed altered matrix contents, developed granule-to-granule fusions and ultimately extrusions of granules through pores in the plasma membrane, when stimulated (ﬁg. 116) [402]. 9.2.3. Rat Peritoneal Mast Cells

Appropriately stimulated peritoneal RMCs underwent typical AND. Our studies were directed towards examination of extremely rapid kinetics of these secretory events that were captured with our microwave-assisted ﬁxation methods for electron microscopic studies [184, 403, 404]. Combined with ultrastructural gold labeling and morphometric analyses, these studies have shown that RMC granule protein moves in small vesicles after stimulation and that visibly altered, less dense granule matrices contained less chymase and TNF- when appropriately stained for these granule contents in stimulated cells [135, 185; see Chapters 4–7].

9.3. Human Basophils 9.3.1. Anaphylactic Degranulation

Our earliest studies of AND of HBs stimulated by an IgE-mediated mechanism [405] have been supplemented by the identiﬁcation of similar morphologic secretion mechanics stimulated by a variety of secretogogues with different mechanisms of action [reviewed in 7]. HBs more recently have been examined after stimulation with a bacterial peptide FMLP, several cytokines,

Fig. 115. Mouse mast cell cultures show control (A) and Xid (B) immature mast cells. Both cells have a single-lobed nucleus. The control cell has regular, short, narrow surface folds and numerous immature secretory granules ﬁlled with dense and vesicular materials. The Xid mast cell is actively extruding many cytoplasmic secretory granules. Their electron-dense and poorly dense vesicular contents are no longer membrane-bound, and remain close to the secreting cell; the extruded granules are enmeshed in elongated, narrow surface folds. Some membrane-bound granules remain within the cytoplasm. A, B !8,000.

Ultrastructure of Mast Cells and Basophils

206

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

207

Fig. 116. Peritoneal mast cells from a Beige rat undergoing anaphylactic degranulation of giant secretory granules. Swollen, less electron-dense secretory granules in the process of extrusion and within degranulation channels in the cytoplasm are enlarged beyond the giant, electron-dense cytoplasmic secretory granules not so involved in this cell and in adjacent mast cells. !9,000.

and a tumor-promoting agent (TPA) [206, 292, 327; see Chapter 7]. These three categories of stimuli advanced our understanding of PMD, albeit with vastly differing kinetics, provided substantial ultrastructural data in support of a degranulation continuum model spanning PMD ] AND, and provided the opportunity to image and quantitate vesicle transport of granule proteins at high magniﬁcation [see Chapter 7]. Additionally, these experiments illustrated an

Ultrastructure of Mast Cells and Basophils

208

unusual form of AND stimulated by TPA whereby secretory granules reached plasma membranes but were not secreted through them. This ‘forme fruste’ of AND (ﬁg. 117) was associated with a prolonged and continuing phase of PMD and no ultrastructural evidence of recovery from secretion [292]. Other stimuli (FMLP) caused a release reaction which rapidly evolved from PMD to AND with actual and complete granule extrusion (ﬁg. 118) and histamine release, and then rapidly entered a recovery phase of granule materials [206, 406]. 9.3.2. Recovery from Anaphylactic Degranulation

Basophils, like other granulocytes, are generally viewed as end-stage cells with no potential for renewal or prolonged functional life, particularly after migration to tissues from the peripheral blood. Basophils are secretory cells characterized by synthetic product storage in prominent cytoplasmic granules and regulated secretion of these products by exocytosis of granules [52]. Many secretory cells with similar granule storage sites and secretory programs (i.e., numerous endocrine and exocrine cells) are not end-stage cells and characteristically replenish secreted materials as needed [7]. Evidence that basophils also recover from regulated degranulation by granule reconstitution was acquired through electron microscopic studies of guinea pig basophils in short-term cultures following secretion [407]. Mast cells (cells with functional similarity to basophils) of two species (rat and human) are also known to reconstitute granules by several mechanisms following their regulated secretion by exocytosis [reviewed in 7]. We examined the ultrastructural morphology of HB recovery in shortterm experiments (up to 6 h) following FMLP stimulation. Previously, FMLP stimulation of HBs was shown to induce a rapid continuum of ultrastructural changes [206] corresponding in time to the regulated secretion of secretory products, determined biochemically [328], and characterized by early PMD merging into an AND phase. PMD in HBs is characterized by the development of partially empty and empty granules which do not fuse and remain in the cytoplasm [52, 169]; AND in HBs is characterized by the extrusion of individual granules to the cell’s exterior following granule and plasma membrane fusion (ﬁg. 118) [7, 52]. In the recovery period (10 min to 6 h), we used a combined method – postﬁxation exposure to CF [271, 405] and reduced osmium [405] – to yield optimal ultrastructural information. That is, the reduced osmium post-ﬁxation technique maximizes the imaging of glycogen, granule contents and membranes [405]. The post-ﬁxation exposure to CF allows binding of CF to negatively charged membranes in continuity with the plasma membrane, thus deﬁning such continuities that are present at section levels out of the plane of view [405]. These techniques together provide an analysis of morphological

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

209

117

Ultrastructure of Mast Cells and Basophils

210

118

Fig. 117. Focal blebs in fused granule-plasma membranes protrude outward beyond the perimeter of the cell in TPA-simulated human peripheral blood basophils at 1 min (A) and 10 min (B, C) after stimulation. The underlying granules show no change in particle packing (B), focal piecemeal losses in granule particles beneath the blebbed, fused membranes (B, C) and diminished (altered) packing of granule particles throughout (A). Note that the raised surface bleb is nearly as large as the granule in C. Granule bulges beyond the perimeter of the cell in TPA-stimulated basophils at 2 min (D–F) show a bulge of an unaltered, particle-ﬁlled granule (D) and an extensively bulged granule (E). The particle packing of the granules (E, F) is diminished. All bulged granules are restrained by overlying dense plasma and granule membranes (D–F). A !85,500. B !51,300. C !59,400. D !38,500. E !59,900. F !61,800. [From 292, with permission.] Fig. 118. FMLP-stimulated human basophil at 20 s shows separately located, extruded, membrane-free granules (open arrowheads) and spherical, homogeneous CLCs (closed arrowheads) closely associated with cell surface cul-de-sacs. The cytoplasm contains full and empty vesicles, glycogen aggregates, and two full granules. Note focal ‘piece’ emptied from one of these (arrow). !20,700. [From 206, with permission.]

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

211

Fig. 119. A paucigranular, recovering human basophil (1 h after FMLP) shows the cell surface stained with cationized ferritin, which also binds to the surface of one extruded granule (A). Another granule (same cell, B) remains in the cytoplasm and contains granule particles; a large number of vesicles (which either are electron-lucent or contain electron-dense particles) (arrows) surround this one remaining cytoplasmic granule (B). In A, note that several vesicles (arrows) are attached to the plasma membrane immediately beneath the overlying extruded granule. N = Nucleus. A !69,400. B !52,300. [From 406, with permission.]

granule reconstitution in HBs at early recovery times beyond stimulated secretion. The evident morphological mechanisms were primarily those of conservation (by re-uptake of granules and portions of granules and membrane as vesicles and vacuoles) (ﬁg. 119, 120), condensation (electron-dense foci in reﬁlling granules) (ﬁg. 121), and synthesis (Golgi expansion and progranule formation) (ﬁg. 122). Recovery from FMLP-induced HB secretion was characterized by a number of morphological events. First, there was little evidence of residual PMD, characterized by regular-sized, electron-lucent cytoplasmic containers and no evidence of PMD with enlarged, electron-lucent granule containers at the time intervals sampled, spanning 10, 20, 30 min to 1, 3 and 6 h. Rather, many granule containers displayed central condensation of electron-dense materials, often superimposed on less dense contents (ﬁg. 121). Ultimately, large numbers of cytoplasmic granules were completely ﬁlled with electron-dense material, and revealed no evidence of focal electron-lucent areas characteristic of PMD. Such granules also did not generally have large collections of dense concentric membranes.

Ultrastructure of Mast Cells and Basophils

212

Fig. 120. Recovering human basophils show extruded granules (3 h after FMLP) in various stages of internalization. Note the progressive investiture by two short cytoplasmic arms (arrows) which close and exclude the extracellular tracer, cationized ferritin. Cationized ferritin binds to the cell surfaces and the surfaces of granules that still maintain continuity to the cell surfaces (A–D). A !43,700. B !62,700. C !48,700. D !37,000. E !41,300. F !53,700. [From 406, with permission.]

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

213

Fig. 121. Recovering human basophil granules (3 h after FMLP) show central, electron-dense condensation foci superimposed on less dense peripheral particles. These foci vary in shape from rectangular (A, C) to round (B). One recovering granule (A) retains a peripheral area of focal piecemeal loss of the peripheral granule particles adjacent to the central dense core. This area of granule release is nearly electron-lucent and is intimately associated with a cluster of electron-dense glycogen particles (arrow). A !63,700. B !57,000. C !59,900. [From 406, with permission.]

Granule condensation of central dense materials on a less dense background was noted in both small and large (⬃1 m) particle-containing granules, many of which were located in the peripheral cytoplasmic area (ﬁg. 121). The central dense nucleoids resemble the general process of granule condensation seen in recovering and maturing mast cells [20, 24, 167]. The central densities in HB granules should not be confused with the central dense cores that are an integral part of the anatomy of mature human eosinophil secondary granules [60]. Most of the recondensing basophil granules displayed a single dense core, but some had several. That these granules represent reﬁlled granules, which were previously emptied by PMD [52, 169], is suggested by the presence of some granules with central condensation cores that also retained focal electronlucent piecemeal holes in their particulate material (ﬁg. 121A). We also noted elongated tubular granules ﬁlled with dense particles in the peripheral cytoplasmic area, structures that are not generally present in unstimulated cells. Paucigranular basophils were noted at recovery times. These mature cells with polylobed nuclei were not damaged. They differed from the nearly granule-free basophils, which abounded at 1 min after FMLP stimulation as a re-

Ultrastructure of Mast Cells and Basophils

214

sult of extrusion of nearly all secretory granules, in that they had large numbers of cytoplasmic smooth membrane-bound tubules and vesicles (ﬁg. 119). Other features of these cells were large masses of aggregated glycogen in their cytoplasm. The large number of cytoplasmic vesicles present in recovering HBs often formed focal collections in the peripheral cytoplasm (ﬁg. 119). These collections surrounded granules and were sometimes concentrated beneath surfaceattached, previously released granule particles (ﬁg. 119A). Most of these vesicles were smooth membrane-bound, but coated pits in the plasma membrane were seen in similar locations beneath granular particles. Some vesicles that were attached to the plasma membrane and that were located in the peripheral cytoplasm at these locations contained the post-ﬁxation tracer, CF, indicating their continuity with the cell surface. The vesicles located in these sites consisted of a mixture of electron-lucent structures and vesicles ﬁlled with dense and particulate materials analogous to adjacent cytoplasmic granules or to overlying granule particles on the plasma membrane (ﬁg. 119B). Granule-sized, membrane-bound vacuoles were seen in the peripheral cytoplasm and were in continuity with the plasma membrane, as evidenced by the presence of the electron-dense post-ﬁxation tracer, CF, within them and staining their membranes. These spaces, devoid of granule content, were less frequently present than were granules in the process of being individually enclosed by short, arm-like extensions of the cytoplasm (ﬁg. 120). Such enclosure suggested cytoplasmic re-uptake of previously extruded granules that remained intact and attached to the cell surface. Internalization of these granules resulted in their reinvestiture with a plasma membrane-derived membrane to which the granule was attached. Initially, these released granules were coated by CF when attached to the cell surface. As they became enclosed, this post-ﬁxation tracer no longer bound to the granules. The majority of basophils recovering from FMLP stimulation of histamine secretion did not show extensive increases in Golgi structures. Some cells did show increased numbers of Golgi area electron-lucent, smooth membranebound vesicles, and others had concentrations of small vesicles and larger progranules ﬁlled with electron-dense particles (ﬁg. 122). New sites and different densities of labeling for CLC protein in distinctive basophil phenotypes stimulated by f-Met peptide were deﬁned. New sites for CLC protein included nucleus, cytoplasm, degranulation channel, degranulation channel membrane, plasma membrane, and a newly recognized granule population similar to primary granules in eosinophils [60, 67]. These new sites, as well as previously documented sites of CLC protein (granules, intragranular CLCs, cytoplasmic vesicles) showed variable labeling when analyzed by phenotype. Other sites (besides intragranular CLCs) of formed CLCs

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

215

Ultrastructure of Mast Cells and Basophils

216

included cytoplasm, degranulation channel, extracellular space, and, rarely, nucleus. Analysis of cytoplasmic vesicles, total granules, and altered granules, and gold particles in subcellular compartments in seven identiﬁable phenotypes revealed f-Met peptide stimulated HBs to empty their granules by transporting CLC protein in vesicles to the plasma membrane in the absence of granule extrusion in cells exhibiting PMD (ﬁg. 123A) [67, 200; Chapter 7]. In cells exhibiting AND, gold-labeled CLCs were extruded to the cells’ exterior in concert with granule particles and concentric dense membranes contained within granules (ﬁg. 123B). Completely degranulated cells had a high density of plasma membrane gold label that was associated with numerous gold-laden endocytotic cytoplasmic vesicles. Basophils reconstituted their main granule population, within which CLCs resided, in part by endocytosis of previously released plasma membrane-bound CLC protein (ﬁg. 123C). Completely recovered cells displayed decreased CLC protein labeling of the plasma membrane and vesicle compartments, the presence of a highly labeled new granule subset that resembled CLC protein-containing primary granules in eosinophils, and the highest density of granule and intragranular CLC gold labeling of all phenotypes that developed after stimulation (ﬁg. 124). Conclusions from these studies indicated that seven individual f-Met peptide-activated HB phenotypes labeled by an ultrastructural immunogold method to detect subcellular sites of CLC protein showed changing distributions of this protein which document the capability of HBs to undergo complex release and recovery reactions [67]. We examined subcellular histamine localizations in puriﬁed HBs that were stimulated to degranulate with FMLP, using an ultrastructural enzymeafﬁnity technique [see Chapter 6]. Basophils were collected at early (0, 20 s, 1 min) and late (10 min to 6 h) time points post-stimulation and were prepared for routine ultrastructural and DAO-G cytochemical analysis. Histamine was present in unaltered cytoplasmic secretory granules (30.77 gold particles/m2; p ! 0.001 compared with background); speciﬁcity controls (histamine absorption, diamine oxidase digestion) abrogated granule label for histamine. Altered granules in stimulated cells were not signiﬁcantly labeled for histamine, as compared with background (p = not signiﬁcant); unaltered granules in the same cells contained more histamine than altered granules (p ! 0.05). During recovery times, spanning 10 min to 6 h, granules again appeared electron-dense and contained histamine (33.49/m2; p = not signiﬁcant, as

Fig. 122. Golgi areas (G) near the polylobed nuclei (N) of recovering HBs (6 h after FMLP) show large numbers of vesicles (A) and electron-dense progranules (B). A !30,500. B !32,000. [From 406, with permission.]

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

217

Ultrastructure of Mast Cells and Basophils

218

compared with unaltered granules in 1 min FMLP-stimulated cells, and p ! 0.05, compared to altered granules in 1 min FMLP-stimulated samples) [250]. Other structures devoid of histamine in actively secreting cells included extruded granules and intragranular and extruded CLCs (ﬁg. 125). RBs displayed morphologic evidence of material and membrane conservation, granule content condensation, and biosynthesis. Subcellular histamine-rich sites in actively recovering basophils included condensing granules and collections of cytoplasmic vesicles in three locations – beneath the plasma membrane, adjacent to granules, and in the Golgi region. These studies show that unaltered granules of actively releasing HBs, as well as similar granules that are reconstituted after FMLP-stimulated degranulation, contain histamine (ﬁg. 126), but that altered granules in stimulated cells undergoing degranulation are devoid of histamine. Reconstitution of histamine-rich granules is associated with DAO-G-positive cytoplasmic vesicles (ﬁg. 125D), suggesting transport of histamine derived from either new synthesis, re-uptake of released histamine, or both, to reconstituted granules [201, 250].

9.4. Human Mast Cells 9.4.1. Anaphylactic Degranulation

AND, the name for regulated secretion from basophils and mast cells, exhibits the classical secretory morphology of granule membrane fusions and extrusions from cells. Development of membrane-bound degranulation chambers within the cytoplasm that develop continuities with the cell exterior is an integral part of AND which may or may not be a part of the secretory morphology. These differences are evident in different species, depending on triggers used and whether the target cells are basophils or mast cells. Our earlier studies of AND from mast cells have been summarized in detail [7]. The species involved in those studies of mast cells included human and Fig. 123. f-Met peptide-stimulated human basophils, recovered at 10 s (A), 1 min (B) and 2 min (C), show high magniﬁcations of the PMD-I (A), AND-II (B) and RB-II (C) phenotypes. A The PMD-I basophil shows extensive labeling of peripheral cytoplasm and plasma membrane with 30-nm gold particles representing CLC protein. B The AND-II basophil shows 30-nm gold label in the nucleus, cytoplasm and in extruded, formed CLCs adjacent to the cell surface (arrowheads). An electron-lucent cytoplasmic vesicle is also labeled (arrow). C The RB-II basophil shows 30-nm gold labeling of formed CLCs within particle-ﬁlled granules. Perigranular cytoplasmic label is present, but the plasma membrane is devoid of gold label. A, C Bar = 0.6 m. B = Bar 0.4 m. [From 67, with permission.]

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

219

Fig. 124. Granule types of human basophils prepared with immunogold to demonstrate CLC protein (A–C) and control (D, omission of speciﬁc primary antibody). A ANDII phenotype obtained at 1 min post-f-Met peptide stimulation. B RB-I –1 min. C RB-I – 2 min, AND-I –20 s (inset). D PMD-I –20 s. A particle-ﬁlled granule, the major granule type in human basophils, labeled with 30-nm gold particles is illustrated (A) [note goldlabeled vesicular structure attached (open arrow) and gold-labeled full vesicle adjacent (closed arrow) to unaltered granule]. The RB-I cell in B shows an unlabeled particle-ﬁlled granule and a heavily gold-labeled, homogeneously dense granule of similar size. The RB-I cell in C shows another heavily gold-labeled granule with homogeneous contents [note the empty vesicle (open arrow) and the full vesicle (closed arrow) next to the granule]. The inset in C shows the second granule population to be described in human basophils [280] – a small granule that has homogeneous lightly dense contents, generally resides near the nucleus of the cell, and does not bind the antibody speciﬁc for CLC protein. In D, the control sample shows that the structures imaged do not artifactually bind the goldlabeled secondary antibody. These structures include the particle-ﬁlled granule, the large, homogeneously dense granule (arrowhead), the cytoplasm, empty vesicles, and plasma membrane. Bar = 1 m. [From 67, with permission.]

Ultrastructure of Mast Cells and Basophils

220

Fig. 125. Human basophil granules in cells recovering from FMLP stimulation prepared with DAO-G to demonstrate histamine show altered, nearly empty (A), partially empty (B), and unaltered full granules (C, D) in the cytoplasm. A, B The altered granules are nearly devoid of granule matrix particles and contain central CLCs (C). The empty areas are largely devoid of DAO-G labeling, and the CLCs are not labeled for histamine. Cationized ferritin stains the overlying plasma membrane (A) but not the granule membranes, indicating that these are closed to the cell surface. C The granule is intensely labeled with DAO-G. Note that the gold labeling is attached only to the particulate matrix surrounding the central, unlabeled CLCs (C). D The unaltered granule is labeled with DAO-G, as is a particle-ﬁlled vesicle adjacent to it (arrow). E Poorly DAO-G-labeled granule particles, partially enwrapped by surface processes (arrows), are stained on their plasma membrane-associated side with cationized ferritin. Such images at late recovery time intervals suggest re-uptake of previously released, histamine-poor granule materials. A !51,300. B !66,700. C !59,100. D !55,600. E !77,100. [From 250, with permission.]

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

221

murine. Since that review we have added to our body of ultrastructural knowledge studies of AND in isolated preparations of HSMCs [35], in HSMCs in vivo [252, 408, 409], as well as in observational studies of spontaneous AND in vivo in HGMCs [308]. Application of ultrastructural labeling methods for histamine and heparin [see Chapter 6] to several studies of AND provide information about their release during AND of HLMCs ex vivo [216, 251] and HSMCs in vivo [252]. Isolated HSMCs were prepared and cultured overnight before functional and electron microscopic studies were done [35]. Mast cell suspensions were examined after stimulation with anti-IgE to produce AND or examined in buffer-incubated controls. Histamine release was measured in replicate samples. Control, isolated HSMCs studied by electron microscopy were well preserved and fully granulated. Although all granule patterns reported for HMCs were found, crystal granules were the most prevalent (ﬁg. 127), as is true for HSMCs in situ. Individual mast cells containing both crystal and scroll granules occurred. Lipid bodies were rare, as in HSMCs in situ. Control, isolated mast cells did not express granule changes associated with either PMD or recovery during wound healing in situ, nor were morphologic changes of AND present. Spontaneous histamine release was 0% in control samples. AND of isolated HSMCs was accompanied by 24% maximum histamine release and characteristically showed extrusion of altered, membrane-free granules through multiple pores in the plasma membrane to the exterior of the cell (ﬁg. 7). Other morphologic aspects of AND, as expressed in isolated HLMCs [reviewed in 7], were also present. These events included granule swelling, fusion, alteration of matrix contents, degranulation channel formation, pore formation, and shedding of granules, membranes, and surface processes. The ultrastructural morphology of isolated HSMCs and their IgE-mediated degranulation shows some differences from similar studies of isolated HLMCs, and of HLMCs and HGMCs in biopsy samples. These differences include crystal granules as the predominant granule pattern, minor numbers of lipid bodies, and extrusion of granules during AND as characteristic for HSMCs. By contrast, isolated HLMCs and HGMCs have more scroll granules and particle granules, respectively, and more lipid bodies. In isolated HLMCs, AND is almost exclusively an intracellular fusion event characterized by the formation of complex degranulation channels within which altered granule matrix materials solubilize [reviewed in 7]. SCF, also known as kit ligand or MCGF, the ligand of the c-kit receptor, is a pleiotrophic growth factor that represents a major mast cell survival and developmental factor, both in murine rodents and in humans [410, 411]. SCF has many effects in MMC development: it can suppress apoptosis, both in vitro [412–414] and in vivo [414], promote the proliferation of cells in the mast

Ultrastructure of Mast Cells and Basophils

222

Fig. 126. Reconstituted, unaltered, particle- and membrane-containing human basophil cytoplasmic granules (G) in a cell recovered 6 h after FMLP stimulation contain DAO-G labeling for histamine. N = Nucleus. !40,000. [From 250, with permission.]

cell lineage at multiple stages of their maturation [10, 411], and promote certain aspects of mast cell maturation [95, 96, 415]. rhSCF can also promote the development of immature HMCs in vitro [23, 27, 416–418]. However, SCF can also induce mast cells to express secretory function. SCF can promote c-kit-dependent mast cell degranulation and mast cell-dependent local cutaneous inﬂammation in mice in vivo [419]. In vitro studies indicate that SCF can directly induce puriﬁed HSMCs to release histamine and prostaglandin D2 [420], can enhance FcRI-dependent mediator secretion by puriﬁed human lung [421] or skin [420] mast cells, and can induce MMCs to release the cytokines IL-6 and, to a lesser extent, TNF-, in the absence of signiﬁcant release of preformed mediators such as histamine or serotonin [422]. We performed an ultrastructural analysis of ten skin biopsies which had been obtained from 3 women with breast cancer who were undergoing daily subcutaneous dosing with recombinant methionyl-human stem cell factor (rhSCF) as part of a phase I clinical trial [408, 409]. The biopsies were obtained at sites of subcutaneous administration of rhSCF, within ⬃1 to 2 h of rhSCF

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

223

injection. At the same time, biopsies were taken at contralateral control sites which had not been directly injected with rhSCF. We noted that subcutaneous dosing with rhSCF in these subjects induced the local development of a wheal-and-ﬂare response, which was associated with evidence of mast cell degranulation (ﬁg. 128, 129), as well as a systemic increase in numbers of cutaneous mast cells. The electron microscopic analysis revealed that all biopsies of swollen, erythematous rhSCF-injected sites exhibited AND of both mature and immature mast cells, an acute inﬂammatory response characterized by the migration of neutrophils, basophils (some of which exhibited evidence of PMD), and eosinophils through blood vessel walls into the perivascular and extravascular spaces, and edema and ﬁbrin deposition within the interstitium. By contrast, the control biopsies contained no evidence of mast cell degranulation or acute inﬂammation. However, both control and rhSCF-injected sites exhibited mast cells which were undergoing granule building and maturation. Thus, at the doses tested in these subjects, subcutaneous injection of rhSCF induced anaphylactic-type degranulation of dermal mast cells at the injection site, with an acute inﬂammatory response that was associated with the recruitment of granulocytes. By contrast, mast cells at sites distant from those directly injected with rhSCF exhibited no evidence of enhanced secretion. We took advantage of the extensive mast cell degranulation that was present at sites of rhSCF injection to assess the ultrastructural localization of histamine in HMCs that are undergoing classical AND in vivo [252]. Although there is no question that mast cells store histamine in their granules and that this biogenic amine can be released by mast cells in response to agents that induce mast cells to degranulate [423], until recently it has not been possible to study this process morphologically at the ultrastructural level. We developed a DAO-G ultrastructural enzyme-afﬁnity technique to localize histamine [154]. This method, which can be used in conventionally prepared ultrastructural samples, is based on the afﬁnity of the enzyme, diamine oxidase (histaminase), for its substrate, histamine [154]. We used this method to detect histamine in skin biopsies obtained from patients with breast carcinoma who were receiving daily subcutaneous injections of rhSCF in a phase I study of this cytokine. We examined control biopsies obtained at sites remote from rhSCF injection as well as biopsies of rhSCFinjected skin that were obtained within 2 h and 30 min of the subcutaneous injection of rhSCF at that site. The rhSCF-injected sites (which clinically exFig. 127. High magniﬁcation micrograph of granules in the cytoplasm of an isolated human skin mast cell shows regular parallel arrays of a crystal granule (top). Homogeneous dense material and peripheral parallel lamellae comprise several mixed granules as well. !70,000. [From 35, with permission.]

Ultrastructure of Mast Cells and Basophils

224

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

225

Fig. 128. Human skin mast cell in the papillary dermis of rhSCF injection site biopsy specimen (obtained on day 13 of rhSCF dosing) reveals morphology typical of AND. Altered, swollen granules are present in intracytoplasmic degranulation chambers that open to the cell surface, and swollen, membrane-free, extruded granules (arrows) are present in the interstitium at a distance from the probable mast cell of origin. These extruded granules rest in amorphous, slightly dense pools of interstitial edema that separate adjacent bands of collagen. Electron-dense ﬁbrin fragments are also visible, some of which (open arrowhead) are attached to the mast cell surface. !11,300. [From 409, with permission.]

hibited a wheal-and-ﬂare response), but not the control sites, contained mast cells undergoing regulated secretion by granule extrusion. The DAO-G-afﬁnity method detected histamine in electron-dense granules of mast cells in control and injected skin biopsies (ﬁg. 130); however, the altered matrix of membranefree, extruded mast cell granules was largely unreactive with DAO-G (ﬁg. 131). These ﬁndings represent the ﬁrst morphologic evidence of histamine secretion by classical granule exocytosis in HMCs in vivo.

Ultrastructure of Mast Cells and Basophils

226

Fig. 129. High magniﬁcation micrograph of rhSCF injection site biopsy specimen, which was obtained on day 13 of rhSCF dosing, illustrating extracellular, membrane-free human mast cell granules in close proximity to surfaces of elongated, narrow cell processes of dermal dendrocytes. Dermal dendrocyte at top illustrates intermediate ﬁlament-packed process of dermal dendrocyte in continuity with dendrocyte cell body. Banded interstitial collagen ﬁbers are plentiful. !34,000. [From 409, with permission.]

We also examined IgE-mediated AND in isolated HLMCs with the ultrastructural method to detect histamine [251]. DAO-G uniformly labeled the electron-dense secretory granules of HLMCs. Higher magniﬁcations of the human lung mast cell (HLMC) granules labeled for histamine showed substructural patterns typical of HLMC granules beneath the gold particles (ﬁg. 132). Speciﬁcity controls for the technique were negative (ﬁg. 132E, F). HLMCs that were stimulated with antibody to IgE and sampled 5 and 20 min later for electron microscopy were also well preserved (ﬁg. 133). More-

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

227

130

Ultrastructure of Mast Cells and Basophils

228

Fig. 131. Skin biopsy from a site that had been injected with rhSCF, obtained from a patient who had received rhSCF for 1 day. The biopsy was obtained 1 h 40 min after the subcutaneous injection of 25 g/kg of rhSCF. Multiple extruded, membrane-free mast cell granules (G), which exhibit altered matrix materials, are present in the interstitium near the degranulating mast cell; these extruded granules exhibit relatively little DAO-G labeling for histamine, except in regions of the granules that contain the most electron-dense matrix material (solid arrowheads). !28,000. [From 252, with permission.]

over, classic ultrastructural images of AND (a regulated secretory event) were evident in the absence of cell injury. AND in HLMCs is characterized by the formation of cytoplasmic degranulation channels (ﬁg. 133) from the fusion of granule membranes. The content of these elongated chambers consists of altered granule matrix materials with diminished electron density prior to pore formation with the plasma membrane and release of channel contents. Sometimes individual granules fused with the cell surface and released their matrix materials directly to the cell exterior. DAO-G labeled electron-dense, unaltered cytoplasmic granules adjacent to degranulation channels in anti-IgE-stimulated mast cells (ﬁg. 133). Completely electron-lucent cytoplasmic degranulation channels were devoid of gold particles, indicating the absence of histamine in them (ﬁg. 133A). When re-

Fig. 130. Human mast cells in skin biopsies obtained 1 h 40 min (A) or 2 h (B) after the subcutaneous injection of 25 or 5 g/kg rhSCF, respectively, in patients who had received daily subcutaneous injections of rhSCF for 13 days. Reactivity for histamine is indicated by DAO-G labeling in the cytoplasmic granules of a mast cell that exhibits no morphologic evidence of secretory activity (A); DAO-G label is absent when the grid containing a section of the specimen was digested with DAO before staining with DAO-G (B). A !61,800. B !68,400. [From 252, with permission.]

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

229

Ultrastructure of Mast Cells and Basophils

230

sidual wisps of altered granule matrix materials were visible in degranulation channels as well as in the process of extrusion from them, small numbers of gold particles labeled this material, indicating some residual histamine association (ﬁg. 133B). HMCs are a rich and unique source of heparin, which is stored in cytoplasmic secretory granules and accounts for metachromasia, a staining property used to identify mast cells by light microscopy. We used a labeling method for heparin that depends on the well-known property of RNase inhibition by heparin to image subcellular sites of heparin in AND of HLMCs [155, 216]. HLMCs were isolated, partially puriﬁed, either stimulated or not stimulated to secrete with anti-IgE, and recovered 20 min (or 6 h) later for routine electron microscopy. Histamine secretion was also determined. A previously developed post-embedding, enzyme-afﬁnity-gold electron microscopic technique to image RNA with R-G, which also binds to heparin, was employed to determine the subcellular locations of heparin in non-secretory and secretory mast cells. Speciﬁcity controls for the novel use of this method and quantitation of granule labeling in these controls were performed. Heparin was labeled by R-G in electron-dense granules within non-secretory HLMCs (ﬁg. 134) and in electron-dense granules that persisted in secretory HLMCs at the maximum histamine secretion time (20 min) (ﬁg. 135A). Speciﬁcity controls showed that gold alone did not label HLMCs and that absorption with heparin signiﬁcantly reduced or abrogated HLMC granule staining with R-G, but that RNA absorption did not. Heparin stores were absent in newly formed, electron-lucent intracytoplasmic degranulation channels in secretory HLMCs (ﬁg. 135A). Electron-dense granule matrices in the process of extrusion to the cell exterior still retained heparin at the instant of cellular secretion (ﬁg. 135B).

Fig. 132. High magniﬁcation views of unstimulated human lung mast cells granules at 5-min (A) and 20-min (B) incubation in buffer and after culture for 3 h (C), 6 h (D–F), 18 h (G), and 24 h (H, I). The granules containing mixtures of scrolls and homogeneously dense matrix materials are labeled for the presence of histamine (A–C, H). Scroll-containing granules contain histamine that is localized by gold particles attached to individual lamellae, which comprise scrolls, as well as to electron-dense or electron-lucent centers of scrolls (A, C, D, G, H). Addition of individual scrolls to larger granules shows that the single scrolls (arrows) also contain histamine (H). Focal collections of lipid (open arrows) within some granules are devoid of DAO-G label for histamine (I). Small, perigranular cytoplasmic vesicles also label for histamine (solid arrows) (I). An absorption control with solid-phase histamine (E) and a digestion control with DAO (F) both show abrogation of DAO-G label for histamine. A !64,100. B !45,600. C, G !76,000. D, E !62,700. F !47,000. H !60,800. I !49,900. [From 251, with permission.]

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

231

Fig. 133. Human lung mast cells at 5 min (A) and 20 min (B) post-stimulation with anti-IgE show degranulation channels (D) that either are electron-lucent and devoid of histamine (A) or contain a few residual wisps of electron-dense matrix with some DAO-G label (B). The degranulation channel in B is open to the cell surface (arrowhead). The unaltered granules remaining in each cell are heavily labeled for histamine. A, B !26,100. [From 251, with permission.]

Ultrastructure of Mast Cells and Basophils

232

Fig. 134. Electron micrographs of human lung mast cell granules stained with absorbed R-G of a typical experiment are shown: A Sepharose bead absorption; B heparinagarose bead absorption; C R-G incubated with RNA and passed over Sepharose beads; D R-G incubated with RNA and passed over heparin-agarose; E R-G incubated with heparin and passed over Sepharose beads, and F R-G incubated with heparin and passed over heparin-agarose. Gold particle count/µm2 of granule area is on each panel. Bars: A 110 nm; B, D 130 nm; C 200 nm; E 150 nm; F 190 nm. [From 216, with permission.]

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

233

Fig. 135. Secretory human lung mast cells, 20 min (A, B) or 6 h (C) after stimulation with anti-IgE, stained with R-G, show large intracytoplasmic degranulation channels (C) devoid of electron-dense material and heparin (A). Cytoplasmic granules that are not altered contain heparin (A, C). Extruded granules with variable electron density and heparin are present at 20 min (B) and 6 h (C) after stimulation. Note that R-G label is bound to extruded, electron-dense, membrane-free granule materials at the cell surface (arrow) in the 20-min sample (B) but that two extruded granules (G) that persist in the extracellular location at 6 h are devoid of heparin. A !30,200. B !31,900. C !20,700. [From 216, with permission.]

Ultrastructure of Mast Cells and Basophils

234

9.4.2. Recovery from Degranulation

We also examined recovering HLMCs held in short-term cultures (3, 6, 18, 24 h) following stimulation of AND. The ultrastructural morphology of these events has been reviewed [7]. Using the new ultrastructural probes for histamine and heparin we localized these mast cell products in recovering cells [216, 251]. Recovery of HLMCs from AND in vitro proceeded primarily by two mechanisms: (1) conservation by condensation, 3–6 h [167], and (2) synthesis, 18–24 h [424, 425]. In the early recovery period (3–6 h), DAO-G labeled condensing granule domains that appeared within degranulation chambers (ﬁg. 136). As separation of progressively more electron-dense granule domains enclosed within membranes occurred, the appearance of scroll foci were noted, and these structures were stained with DAO-G with corresponding increased label density. Later recovery times (18–24 h) showed a continuation of conservation mechanics as well as the appearance of large numbers of vesicles, vacuoles, progranules and Golgi structures [425]. DAO-G labeled most of these structures, whether located in Golgi areas or in the peripheral cytoplasm. Also, gold particles prevailed in electron-lucent vesicles and vacuoles as well as in electron-dense or scroll-containing progranules (ﬁg. 137). These images document the presence of histamine in synthetic organelles of HLMCs recovering from AND but do not rule out re-uptake of previously released histamine. We suggest that the DAO-G method is revealing synthesis of new histamine during recovery of HLMCs from degranulation but that re-uptake of histamine may also be occurring. Re-uptake of extruded granule materials that were labeled with DAO-G was visible in our study of HBs recovering from AND [250]. HBs generally release each granule through separate degranulation pores during AND, and each granule is not solubilized but is available to be internalized by macropinocytosis. Also, portions of granules can be internalized by micropinocytosis. This process of recovery (conservation) is somewhat analogous in principle to conservation of retained granule materials in HLMC degranulation chambers that condense and reform histamine-rich granule domains. The distribution of heparin stained with R-G in cells which have primarily used conservation for their recovery is of interest. Retained cytoplasmic degranulation channels developed increased amounts of electron-dense material that contained heparin (ﬁg. 138). Condensing degranulation channels with heparin-containing dense material were never evident in non-secretory mast cells or in secretory mast cells examined at 20 min after stimulation. Electronlucent degranulation channels did not contain heparin (ﬁg. 138, 139A, B); these chambers underwent progressive partitioning with internal membranes and development of rounded, electron-dense granule domains which did contain

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

235

Fig. 136. This recovering human lung mast cell (18 h after anti-IgE stimulation) shows extensive condensation of dense material in irregularly shaped, resolving degranulation channels and separating granules. All are equally stained with DAO-G. !31,000. [From 251, with permission.]

heparin (ﬁg. 139A, B). Ultimately, condensing, electron-dense, heparin-rich crystalline arrays developed within granule-sized, membrane-bound containers which were derived from these channels as they resolved (ﬁg. 139C, D). HLMCs that primarily extruded individual granules and that resolved their newly formed degranulation channels by inserting granule and channel membranes into the plasma membrane compartment ultimately resolved this rapid and extensive cell surface expansion by internalizing cell processes into cytoplasmic structures, termed canaliculi. HLMCs that were recovering from stimulated secretion and that utilized this mechanism of membrane conser-

Ultrastructure of Mast Cells and Basophils

236

Fig. 137. Human lung mast cells in culture for 18 h after stimulation with anti-IgE show large numbers of cytoplasmic vesicles in the peripheral cytoplasm (A) and in an expanded Golgi area (B). Many of these vesicles contain electron-dense material, indicating progranule formation (arrowheads). Vesicles, progranules, and Golgi structures are labeled with DAO-G. In A, a large, condensing granular structure is also heavily labeled for histamine. A !26,500. B !42,000. [From 251, with permission.]

vation demonstrated well-formed, newly developed, heparin-rich granules in their cytoplasm. The canalicular structures were entirely devoid of heparin. Also of interest was the distribution of heparin in HLMCs which primarily utilized a synthetic recovery mechanism as opposed to that of channel resolution. Synthetic HLMCs generally were those which had released granules and their membranes in their entirety and did not retain cytoplasmic degranulation channels, necessitating resolution by conservation. Such synthetic cells were characterized by the presence of large numbers of cytoplasmic vesicles and vacuoles. These structures were electron-lucent or contained electron-dense

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

237

material and, together with single scrolls, were scattered throughout the cytoplasm. Electron-dense, heparin-containing vesicles and progranules were evident in expanded Golgi areas (ﬁg. 140). Peripheral cytoplasmic areas in synthetic mast cells often contained large numbers of newly formed, small scroll granules, which were heavily labeled for heparin (ﬁg. 140C). Immature HMCs, arising from cord blood mononuclear cells cultured in the presence of the c-kit ligand of human or murine origin, contain mixtures of morphologically immature and mature secretory granules [23]. Use of a new cytochemical technique to localize histamine in ultrastructural samples (based on the afﬁnity of the enzyme, histaminase, for its substrate, histamine) localized this amine to mature mast cell granules of all substructural patterns present, as well as to condensation foci appearing in immature cytoplasmic granules (ﬁg. 141, 142). This cytochemical evidence of histamine bound to condensation foci during granule building in developing mast cells is analogous to evidence obtained during granule recovery of degranulated HBs and mast cells [250, 251]. Evidence for recovery from degranulation of HMCs in vivo was found in studies of wound healing in the skin [77]. We had noted previously that PMD of skin mast cells accompanied some disorders, such as bullous pemphigoid (ﬁg. 143) and melanoma [310, 426]. Re-excision of margins surrounding recently removed melanomas provided clinical samples for electron microscopy where we observed the reﬁlling of previously depleted mast cell granules in healing wounds (ﬁg. 144, 145). Typically these granules contained numerous irregular foci of markedly dense new granule products within empty or partially empty granule containers (ﬁg. 145).

Fig. 138. Secretory human lung mast cell recovered 6 h after stimulation with antiIgE and stained with R-G, shows variable heparin levels in recovering organelles. The predominant recovery pattern uses condensation and conservation in degranulation channels. Note that degranulation channels (C1), nearly devoid of electron-dense material and heparin, persist in two cytoplasmic locations. Condensing, rounded, interconnected, electron-dense granule domains within degranulation channels (C2) containing heparin are present in multiple cytoplasmic locations. Unaltered granules stain heavily for heparin, whereas lipid bodies (L, arrows) do not. Note the lack of background staining on the plastic embedment surrounding the mast cell. !18,500. [From 216, with permission.]

Ultrastructure of Mast Cells and Basophils

238

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

239

Fig. 139. Recovering, post-secretory human lung mast cells, 6 h after stimulation with anti-IgE, stained with R-G, show restructuring of retained degranulation channels as the predominant recovery mechanism. Channels with little electron-dense material contain virtually no heparin (A, B). Some of these chambers show membranous partitioning into new granule domains (arrow, B). Condensing central deposits of electron-dense material in newly partitioned granules contain heparin (C). Larger amounts of heparin-rich, electron-dense material are seen in condensing channels (A). Multiple condensed arrays of crystalline, electron-dense, heparin-rich material are seen contained within a common granule boundary (D). A !41,300. B !36,100. C !54,600. D !51,300. [From 216, with permission.]

Ultrastructure of Mast Cells and Basophils

240

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

241

Fig. 140. Secretory human lung mast cells, recovered 6 h after stimulation with antiIgE and stained with R-G, show primarily a synthetic mechanism of recovery [425]. Golgi (A, B) containing and peripheral cytoplasmic areas (A–C) of mast cells that previously extruded cytoplasmic granules and did not generally retain extensive degranulation channels are shown. A The Golgi (G) area shows expanded cisternal stacks and vesicles admixed with mitochondria. Some of the vesicles contain heparin (arrows). B The Golgi (G) area displays larger, forming granules with increased electron-dense, heparin-rich material (arrows). C The peripheral cytoplasm is ﬁlled with newly developed granules that stain heavily for heparin. A !30,000. B, C !34,000. [From 216, with permission.]

Ultrastructure of Mast Cells and Basophils

242

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

243

Fig. 141. Eight-week suspension culture of human cord blood cells, supplemented with rhSCF and prepared with DAO-G to localize histamine. Montage of high magniﬁcation micrographs of mast cell granules labeled for histamine. A The immature granule with virtually no dense matrix contains a few gold particles attached to condensing foci of dense matrix. B–D Immature granules show more label for histamine; dense condensation foci still are more heavily labeled than remaining granule areas. E The immature granules show label within central dense areas of several scrolls. F Multiple dense foci in the immature granules contain histamine label. G–I Mature granules show gold particles attached to granule scrolls (G, H) and electron-dense matrix (I). A, C !62,700. B, I !80,800. D, E !68,400. F !50,400. G !53,200. H !62,900. [From 29, with permission.]

Ultrastructure of Mast Cells and Basophils

244

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

245

142

Ultrastructure of Mast Cells and Basophils

246

143

Fig. 142. Eight-week suspension cultures of human cord blood cells, supplemented with rmMCGF and prepared with DAO-G to localize histamine. Montage of high magniﬁcation micrographs of mast cell granules labeled for histamine. Immature (A–E) and mature (B, E, F) granules show variable degrees of condensation of electron-dense matrix materials. Gold label for histamine generally is present and is associated with condensing electron-dense granule content. Mature granules (B, E, F) contain the most label for histamine. A, E, F !49,900. B !37,500. C !79,800. D !62,700. [From 29, with permission.] Fig. 143. HSMC from lesional skin of patient with bullous pemphigoid (A) shows focal, geographic losses from granules. Only swollen, altered matrix material remains in these granule portions, a process we have termed piecemeal degranulation. A high magniﬁcation micrograph (B) shows an HSMC granule from a patient with a thigh ﬁbrous histiocytoma. Note protrusion of connected, particle-ﬁlled vesicle (arrows) from a particle granule, perhaps reﬂecting early budding of a content-ﬁlled vesicle as a mechanism for some piecemeal granule losses. A !6,000. B !73,500. [From 77, with permission.]

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

247

Fig. 144. HSMCs in sample of dermal scar after reexcision of residual melanoma and associated resection margins. During wound healing, extensive amounts of collagen (A, B) and active ﬁbroblasts (arrowhead, B) develop. Interspersed are elongated mast cells (A). Some mast cells show intracytoplasmic canaliculi (arrows), structures similar to those that increase in mast cells recovering from anaphylactic degranulation in vitro. Expanded Golgi areas with centrioles and numerous vesicles are seen (A), and nuclei often contain large nucleoli (B). Nearly all mast cell granules display focal areas of condensing, very dense contents (A, B). These are central homogeneous areas in less dense but recognizable crystal, scroll, or particle granules. Many newly dense materials were composed of particulate and regular-to-irregular thick threads. Cytoplasm contains focal collections of free and membrane-bound ribosomes. One lipid body is present (solid arrowhead). A !9,600. B !17,400. [From 77, with permission.]

Ultrastructure of Mast Cells and Basophils

248

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

249

Fig. 145. HSMC granules in dermal scar following reexcisions of melanoma and resection margins show high magniﬁcation images of recovering granule matrix materials during would healing. Note background granule pattern (scrolls, particles, ﬁnely granular) upon which focal homogeneously dense and very dense irregular threads or concentric, dense, thick threads or dense particles are superimposed. The distribution of focal, homogeneously dense areas is either central or peripheral, in locations similar to where piecemeal granule losses occur in releasing tissue mast cells. A !54,000. B !47,000. [From 77, with permission.]

Ultrastructure of Mast Cells and Basophils

250

Degranulation and Recovery from Degranulation of Basophils and Mast Cells

251

Chapter 10

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

10.1. Overview

Cytoplasmic organelles, present in mature HMCs, are dominated by storage-secretory granules, whereas clusters of free ribosomes and strands of RER are minor cytoplasmic constituents in mature HMCs. Membrane-bound HMC secretory granules are characterized by a myriad of substructural details. These have been designated as scroll-containing, crystal-containing, particle-containing, and homogeneously electron-dense content-containing granules. Many secretory granules contain mixtures of these basic patterns [54]. Alteration of these granule patterns is associated with secretion [7]. When granule extrusion and complete granule release from HMCs are stimulated (a process termed anaphylactic degranulation, AND), the individual granule matrix patterns become altered and lose density, so that this process is easily recognizable by electron microscopy [see Chapter 9]. Another form of secretion, termed piecemeal degranulation (PMD), shows focal loss of density and altered granule matrices within an individual granule or involving entire granules, which retain their membranes and appear as empty, non-fused granules in HMCs [see Chapter 7]. In neither type of secretion does the involved HMCs die; in appropriate experimental systems, HMCs have been shown to recover their complement of granules and granule contents [see Chapter 9]. HMC secretory granules have been primarily conceptualized as storage organelles for the synthetic products of the cell. Routine ultrastructural stud-

ies of HMCs suggested the possibility of another function for their prominent secretion-storage granules. Many HMC granules of all types displayed irregularly shaped electron-dense particles often identical in appearance and size to free perigranular ribosomes nearby [427]. We found these granule-associated, ribosome-like particles (ﬁg. 146–148) in variable amounts, but we particularly noticed their increased numbers in cells actively undergoing synthesis during maturation, after granule losses mediated by PMD in vivo, and recovering from stimulated AND ex vivo (ﬁg. 147) [427]. These morphologic observations informed our pursuit of a possible synthetic role for HMC cytoplasmic granules. During the experiments designed to establish a role for HMC granules in RNA biology we noted that another prominent cytoplasmic organelle, lipid bodies, also had close associations with ribosomes. These were even more difﬁcult to discern than those associated with granules in routine preparations, since HMC lipid bodies are characterized by intense osmophilia which often effectively obscures internal structures and structures associated with the phospholipid-rich shell encasing lipid bodies. Nevertheless, images consistent with lipid body-ribosome association were present [17]. Are lipid bodies, like secretory granules in HMCs, also subcellular sites/ organelles that are important in RNA biology? Studies using a multimodal, ultrastructural approach also serve to implicate these organelles in the RNA biology of HMCs [17]. In aggregate, the new studies indicate HMC secretory granule and lipid body associations with ribosomes (the protein synthetic machine of cells) with ribosomal proteins, with RNA, with poly(A)-positive mRNA and with various long-lived, or short-lived, uridine-rich and poly(A)-poor RNA species with key roles in RNA processing and splicing [16, 17, 172–175, 289, 428]. These studies indicate that secretory-storage granules and lipid bodies in HMCs are also equipped for a synthetic role, and they considerably augment our vision of the role of secretory-storage granules in the cell biology of all granulated secretory cells and of lipid bodies generally in mammalian cells where they are known to occur [7].

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

253

146

Ultrastructure of Mast Cells and Basophils

254

147

Fig. 146. Secretory granules in HLMC recovered from culture 6 h after IgE-mediated degranulation. Note the numerous ribosomes within secretory granule matrices. These are morphologically identical to perigranular ribosomes in the cytoplasm between granules. A few strands of rough endoplasmic reticulum also show ribosomes identical to the electrondense, ragged intragranular ribosomes. !51,500. [From 427, with permission.] Fig. 147. A degranulated and recovering HLMC, cultured 6 h after an IgE-mediated degranulation stimulus, shows degranulation channels with varied amounts of recovering material of variable electron density. The density varies from less to more, as indicated by 1 (less condensed) to 4 (more condensed). Note that the electron-dense, particulate ribosomes in the increasingly electron-dense, recovering degranulation channel granules concomitantly increase in number. Compare these intragranular ribosomes with the plentiful cytoplasmic ribosomes in the adjacent type II alveolar pneumocyte (upper left portion of the photograph). !32,000. [From 427, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

255

Ultrastructure of Mast Cells and Basophils

256

10.2. Regressive Ethylene Diamine Tetraacetic Acid Staining of HMCs Reveals Granule- and Lipid Body-Associated Ribonuclear Proteins

Bernhard [429] developed a chelation-based stain for ultrastructural studies that he termed ‘regressive,’ based on light microscopic staining principles. This method involves sequential staining of samples with uranyl acetate (UA), ethylene diamine tetraacetic acid (EDTA), and lead citrate (LC), resulting in preferential bleaching of structures known to contain deoxyribonucleic acid (DNA), such as chromatin, with subsequent enhancement of the electron density of structures known to contain RNA, such as ribosomes. It is now generally conceded that the electron density produced with the EDTA regressive stain is based primarily on protein moieties associated with RNA [430, 431]. Use of this staining procedure was instrumental in identifying for the ﬁrst time perichromatin ﬁbrils in the nucleus [432–434]. Perichromatin ﬁbrils are located at the inner edge of condensed nuclear chromatin, are the sites of heterogeneous nuclear RNA (hnRNA) synthesis in transcriptionally active chromatin, are associated with rapidly labeled, newly synthesized extranucleolar RNA, and are the major sites for splicing and polyadenylation in nuclear mRNA processing [434]. We investigated the requirements for imaging ribonuclear proteins (RNPs) in mature HMCs with the chelating agent EDTA in routinely processed electron microscopic samples to identify more precisely secretory granule-RNP and lipid body-RNP relationships [16, 17, 174, 428]. The staining solutions were aqueous UA and LC, the latter prepared according to Reynolds [435]. Staining in UA was varied over 1–15 min and was done at room temperature and at 40 ° C. Staining with LC was done at 1, 5 or 10 min at room temperature. The chelating agent used was aqueous 0.2 M EDTA, 30 min at room temperature, 37 or 50 ° C. The staining sequence was UA ] EDTA ] LC. In multiple evaluations, it was determined that the optimal staining for our material was: (i) 15 min in 3% aqueous UA at 40 ° C; (ii) 30 min in 0.2 M aqueous EDTA disodium salt at room temperature, and (iii) 1 min in LC at room temperature. This staining procedure varies somewhat from the original procedure reported by Bernhard [429], where the UA concentration was 5 or 0.5% for 1 min, EDTA

Fig. 148. HLMC granules (cultured and recovered 18 h after an IgE-mediated degranulation stimulus) show numerous free cytoplasmic ribosomes adjacent to secretory granules and aligned along the perinuclear membranes nearby. Note that ribosome particles of similar size, shape, and electron density are present within intragranular scrolls as well as freely associated with granule matrix. A !134,500. B !103,500. [From 428, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

257

concentration was 0.2 M for 30 min to 1 h, and LC staining was 5 s to 1 min, all done at room temperature. The use of a chelating agent in the staining procedure enhanced electrondense staining of nuclear and cytoplasmic structures known to contain RNP [429, 432] in isolated, cultured HLMCs (ﬁg. 149). Thus, bleaching of DNAcontaining structures, such as condensed nuclear chromatin (ﬁg. 149A), allowed the enhanced electron density of RNA-containing nuclear structures (interchromatin granules, perichromatin granules, perichromatin ﬁbrils and nucleoli) to be readily seen (data not shown). Cytoplasmic ribosomes (ﬁg. 149) and ribosomes that were attached to the cytoplasmic face of the nuclear membrane were particularly electron-dense. Additionally, as previously shown by Bernhard [429], electron-dense, non-particulate RNA-rich material spanned nuclear pores from nuclear to cytoplasmic locations. Altogether, these structures known to contain RNA [429, 431, 432, 434, 436–438] served as positive controls in each sample of HMCs that we examined. Electron-dense cytoplasmic ribosomes showed particularly close associations with secretory granules and lipid bodies [17, 174] in HLMCs (ﬁg. 149, 150). Rarely, if at all, were ribosomes assembled on RER. When RER strands were found, they were short, non-distended and had only a few attached ribosomes, which is typical for mature HMCs. We found extensive ribosomesecretory granule and lipid body associations rather than associations with RER. These were particularly evident in osmium-free, EDTA-stained sections, since the majority of secretory granule substructural architectures and osmophilia of lipid bodies was bleached with this chelating stain (ﬁg. 149, 150). Despite the bleaching of granules, some scroll (ﬁg. 149), particle, and reticular patterns remained visible in the substance of bleached granules. Ribosomes were particularly evident, packed into intergranular cytoplasm, attached focally and aligned linearly to granule surfaces, surrounding small underlying protrusions of scroll or other granule components, focally invaginating granule substance as clusters and oriented along and within scrolls cut longitudinally (ﬁg. 149). In addition, individual ribosomes completely within granules were evident. HMC lipid bodies in the cytoplasm and in the perinuclear location adjacent to ribosome-ﬁlled nuclear pores (ﬁg. 150) had closely associated ribosomes revealed with this technique as well [17].

Fig. 149. Control HLMCs after 6 h in culture and prepared with Bernhard’s EDTA stain (modiﬁed). In each panel (A–C), scroll granules are surrounded by electron-dense ribosomes, many of which are attached to granule surfaces; some extend into longitudinally oriented granule scrolls (B, C). N = Nucleus. A !95,000. B !88,000. C !99,500. [A From 174, with permission; B from 16; C from 428.]

Ultrastructure of Mast Cells and Basophils

258

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

259

Fig. 150. HLMC prepared with the EDTA regressive stain for RNA. Electron-dense RNA and ribosomes ﬁll the interchromatinic nuclear (N) matrix, the nuclear pore (arrows) and surround and attach to cytoplasmic granules (G) and a lipid body (LB). Nuclear chromatin, granule matrix and lipid body contents are bleached with this technique allowing better visualization of adjacent and attached electron-dense ribosomes. !74,000. [From 17, with permission.]

Ultrastructure of Mast Cells and Basophils

260

Fig. 151. Six-hour control HLMCs prepared with the enzyme-afﬁnity-gold method to detect RNA using an RNase-gold (R-G) probe. A R-G binds to heparin in granules (G) and to RNA in adjacent lipid bodies. B Absorption of the R-G probe with heparin Sepharose prior to sample staining did not abrogate binding to RNA in lipid bodies (LBs) or to the numerous membrane attached and free ribosomes adjacent to LBs. C A similarly heparin-absorbed R-G probe stains the RNA in the lipid body matrix and in the vesicular structures inside the LB matrix. D Heparin absorption of the R-G probe abrogates the heparin staining of the secretory granule (G). A !43,000. B !48,000. C !79,000. D !83,500. [From 17, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

261

10.3. Ribonuclease-Gold Stains Heparin in HMC Granules and Labels RNA in HMC Lipid Bodies

R-G staining [see Chapter 6] was used to evaluate subcellular sites of RNA in HMCs [155, 289, 428]. The initial report for the R-G method used this reagent to detect the substrate, RNA, in rat pancreatic acinar cells [215]. Precise and speciﬁc localization of R-G to nucleoli and ribosomes was demonstrated [215]. We conﬁrmed these localizations in a wide variety of cell lineages including HMCs [155, 216, 289]. HMC granules posed a problem for this technique to localize RNA, however, since we conclusively showed that most of the granule signal emanated from granule heparin stores (ﬁg. 151A) [155, 216, 289; see Chapter 6]. Lipid bodies in HMCs, which are not heparin-containing organelles, could, however, be examined with R-G staining for the presence of RNA. Lipid bodies bound the R-G reagent indicating the presence of RNA in lipid bodies (ﬁg. 151B) [17]. Gold particles were present within lipid bodies, encircling them and bound to ribosome clusters nearby (ﬁg. 151B). When intralipid body electron-lucent structures with electron-dense particles were present, the R-G reagent also was bound to these structures (ﬁg. 151C). We used a variety of blocking and digestion protocols to evaluate heparin binding of the R-G reagent. We found that in all of these, R-G staining of granules was diminished or absent (ﬁg. 151D), indicating the presence of heparin [155, 216], whereas lipid body staining did not change (ﬁg. 151B, C), indicating the presence of RNA in lipid bodies [17].

10.4. Ultrastructural Autoradiography of Uridine Incorporation Labels RNA in Human Mast Cell Granules and Lipid Bodies

[3H]-uridine has been used for years to label RNA in cells, the incorporation site of which is subsequently determined by ultracentrifugation and biochemical analysis of cell fractions or by ultrastructural autoradiography [433, 434, 437, 439–454]. We used this isotope, combined with electron microscopic autoradiography, to localize RNA in isolated, puriﬁed HLMCs (ﬁg. 152). In these studies [16, 175], multiple recovery times in culture after stimulation with anti-IgE to degranulate HMCs [7] were also examined for isotope incorporation into HMCs. These studies showed that nuclear, granular and cytoplasmic compartments of HMCs were labeled with [3H]-uridine (ﬁg. 152). Of 5,812 autoradiographic grains counted in 121 cells that were cultured for 18 h with [3H]-uridine, the relative distribution of label associated with granules (20%) exceeded that for cytoplasm (10%), mitochondria (1%) and Epon (6%) (p ! 0.001); 63% of the autoradiographic grains were, as expected, in the nuclear

Ultrastructure of Mast Cells and Basophils

262

Fig. 152. Control HLMCs, cultured for 18 h with [3H]-uridine and prepared for electron microscopic autoradiography with chemical developer, show elongated, tangled silver grains (indicating uridine incorporation into RNA) in the nucleus (A) and secretory granules (A–C). Most of the grains are located over the nuclear matrix, not over condensed nuclear chromatin. A few grains are located at nuclear pores and overlying the cytoplasm (A). Both scroll (B) and mixed scroll-particle (B, C) granules have [3H]-uridine-labeled RNA. The image in A was exposed to emulsion for 6 months; those in B and C were exposed for 1 month. Note that no background silver grains are present on the support ﬁlm beside the cells (A, B). A !16,100. B !31,600. C !37,000. [From 175, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

263

compartment [175]. We have expanded these studies to examine sites of rapidly and slowly labeled RNA in cultured control HLMCs examined at intervals of 3 min to 18 h after exposure to isotope. We also addressed the impact of regulated secretion and recovery therefrom on RNA localization in HLMCs over the same times. We found that granule stores of labeled RNA dropped markedly when granules were secreted in toto and en masse from stimulated cells, and that granule stores of [3H]-uridine increased in parallel with visible reconstitution of secretory granules in recovering mast cells. As expected, nuclear sites known to contain RNA [433, 434, 437, 444] were also labeled, providing a positive technical control for each sample. The autoradiographic results included two methods of development (ﬁg. 152, 153) and variable exposure times of the samples to photographic emulsion. In general, longer exposure times were needed to visualize short isotope incubations, but the exposure times to photographic emulsions did not impact on grain distribution within cells or their subcellular compartments. The use of physical development after gold latensiﬁcation [455–457] provided small, round silver grains with no overlap between compartments. Thus, precise localization to individual organelles larger than the silver grains was possible. The distribution of label for physical development was similar to that for chemical development. Secretory granules of all ultrastructural patterns [54] incorporated isotope (ﬁg. 152, 153). In control samples that were pulsed for 3 min with isotope and recovered after 3 min for study, the total amount of label was small (⬃5 grains/ cell); 44% of the silver grains were associated with the granule compartment (p ! 0.001 compared to cytoplasm, mitochondria and Epon). Incubation for 18 h with isotope resulted in a smaller relative label in the granule compartment (20%), but this secretory compartment exceeded the cytoplasm, mitochondria and Epon background levels of label (p ! 0.001). Although the relative label, of total cellular label, in the granule compartment fell after 18 h of incubation with isotope, the extended isotope exposure resulted in an increase in the average number of silver grains in the granule compartment to ⬃10 from ⬃2 observed with a 3-min isotope exposure. In addition to labeling secretory granules, the incorporation of [3H]-uridine into other cellular compartments was noted in unstimulated control HLMCs (ﬁg. 152, 153). After incubation in isotope for 3 min, the relative grain label in the nucleus compartment was 36% (p ! 0.001; p ! 0.005 compared to mitochondria and Epon); after exposure for 18 h, the nucleus compartment contained 63% (p ! 0.001 compared to mitochondria and Epon), and the average number of nuclear silver grains increased from ⬃2 to ⬃30 per cell. The label of the cytoplasmic compartment exceeded the values for mitochondria label after exposure for 18 h to isotope (p ! 0.001) but labeled considerably less than the

Ultrastructure of Mast Cells and Basophils

264

Fig. 153. Control HLMCs, cultured for 18 h with [3H]-uridine and prepared for electron microscopic autoradiography with physical developer, show small, round or slightly elongate silver grains (indicating uridine incorporation in RNA) in the nucleus (A) and secretory granules (B–D). Most of the nuclear grains overlie the nuclear matrix (A). Particle (B), reticular (C) and crystal (D) granules incorporate [3H]-uridine into RNA. A !17,500. B !53,000. C !51,500. D !42,000. [From 175, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

265

granule or nucleus compartments after 3 min and 18 h of exposure to isotope (p ! 0.001). The average number of silver grains increased approximately 5-fold in the cytoplasm, comparing the 3-minute to the 18-hour exposure to isotope. The label in the nucleus was associated with nucleoli, was at the interface between condensed chromatin and the nuclear matrix, was over some condensed chromatin, and spanned nuclear pores. [3H]-uridine label in the cytoplasm was associated with ribosome clusters; some of these were closely related to and in contact with the outer surfaces of secretory granules. The total amount and relative proportion of cellular isotope label present in the granule compartment of degranulating and recovering HLMCs were also evaluated. HLMCs undergoing extensive degranulation 20 min after stimulation with anti-IgE (84% histamine release) were characterized by the formation of numerous cytoplasmic degranulation channels and a reduction in secretory granules in the remaining cytoplasm (ﬁg. 154A). Cells displaying this morphology of AND prevailed in the 20-minute sample (which had received a 3-minute pulse of [3H]-uridine after the degranulation stimulus). The amount of isotopic label in the granule compartment of these active secretory cells fell to 5% [p ! 0.001 compared to control cells receiving a 3-minute pulse of isotope (44%) or an 18-hour exposure to isotope (20%)]. HLMCs recovering from AND undergo reconstitution of cytoplasmic secretory granules by conservation and synthetic mechanisms [167, 224, 424, 425]. In samples prepared for electron microscopy 6 and 18 h after stimulation, many cells showed the morphologic correlates of these processes. Thus, cells contained recovering degranulation channels with variable amounts of recondensing electron-dense granule materials and increased numbers of cytoplasmic granules (ﬁg. 154B–E). A progressive increase in isotope was associated with the granule compartment when either a 1-, 2- or 6-hour [3H]-uridine pulse was given to samples recovered 6 h post-stimulus (ﬁg. 155) (7, 8 and 12%, respectively) and after 18 h of [3H]-uridine exposure in an 18-hour recovery sample (14%) (ﬁg. 156). Two samples, identically prepared with HLMCs isolated from different donors, showed identical isotope label in the granule compartment after 18 h of [3H]-uridine exposure in 18 h recovering, previously stimulated cells (e.g., 14% of the total cellular label and 6 silver grains/cell in the Fig. 154. HLMCs stimulated with anti-IgE and recovered 20 min (A) or 6 h (B–E) later after exposure to [3H]-uridine for 3 min (A), 1 h (B), 2 h (D, E) or 6 h (C) and prepared for EM autoradiography with chemical (B) or physical (A, C–E) developer. Unlabeled degranulation channels (C) in the cytoplasm contain altered granule matrix (A). Recovery of the cells 6 h after degranulation shows recovering channels (C) ﬁlled with electron-dense materials that are labeled for RNA after exposure to [3H]-uridine (B–E). The nucleus (N) is also labeled (C, D). A !24,000. B!31,000. C !28,000. D !42,000. E !40,000. [From 175, with permission.]

Ultrastructure of Mast Cells and Basophils

266

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

267

Fig. 155. Recovered HLMCs, 6 h after anti-IgE stimulation, exposed to [3H]-uridine for 2 h and prepared for EM autoradiography with chemical (A) or physical (B) developer contain RNA in granules and nuclei. A Large, tangled silver grains ﬁll the interchromatinic, electron-lucent nuclear (N) matrix and label a completely formed scroll granule that abuts the nuclear membrane. B The granules seen at higher magniﬁcation are surrounded by cytoskeletal ﬁlaments with enmeshed electron-dense ribosomes (arrow) similar to those inside adjacent granules. The granule with the most of these is labeled with three silver grains, indicating RNA. A !38,000. B !50,000. [From 175, with permission.]

granule compartment). [3H]-uridine did not label newly formed degranulation channels in the stimulated HLMCs studied at peak histamine release (20 min) (1 grain found of a total of 149) (ﬁg. 154A), but in samples containing electrondense content in recovering degranulation channels, 110 grains (reﬂecting isotope incorporation) over these structures were found (ﬁg. 154B–E).

Ultrastructure of Mast Cells and Basophils

268

[3H]-uridine labeled other cellular compartments in degranulating and recovering HLMCs. The nucleus and cytoplasm compartments generally exceeded label for the mitochondria and Epon compartments for all stimulated degranulating and recovering samples. Comparisons of the nucleus compartment and the granule compartment in these samples showed that nucleus label exceeded granule label in all of them (p ! 0.001). (Previously, we calculated that the volume fraction of HMC nuclei does not change in similar experiments but that granule volume fractions decrease in actively secreting cells [109].) The observed reduction in granule volume may be reﬂected in a larger proportion of isotope incorporation into stable nuclear volumes than in diminishing granule volumes. [In the work reviewed here [16, 175], an analysis for average nuclear (9.8 ␮m2), granular (12.8 ␮m2), lipid body (1.57 ␮m2) and Epon background areas (77.7 ␮m2) of representative prints showed the standard average relative areas of these structures in HMCs that are not actively secreting.] The relative [3H]-uridine label for the cytoplasm compartment was generally similar to that for the granule compartment in all stimulated samples except one. In this case, a 3-minute [3H]-uridine pulse was given, and the stimulated sample was recovered after 18 h. The relative label in the granule compartment signiﬁcantly exceeded the cytoplasm compartment in this case (17 vs. 9%; p ! 0.001). We examined three samples of HLMCs that received a short 3-minute exposure to [3H]-uridine. In these samples, the granule, nucleus and cytoplasm compartments generally incorporated label that exceeded that for Epon and mitochondria. Thus, rapidly incorporated [3H]-uridine could be demonstrated in the nucleus, granule (ﬁg. 156A) and cytoplasm compartments of control, degranulated and recovering HLMCs. We also studied three samples of HLMCs exposed to long intervals (18 h) of [3H]-uridine. Nucleus, cytoplasm and granule compartment grains exceeded mitochondria and Epon label in these samples (p ! 0.001). The proportion of total cellular label for granules was identical in two separate 18-hour recovering samples prepared from different donors and cultured for 18 h in the presence of isotope after the degranulation stimulus had been given but had not yet reached the amount of label for granules in control cells incubated for 18 h (ﬁg. 152, 153). The length of incubation time in [3H]-uridine impacted on the grain counts in the granule compartment in some samples. For a 3-minute pulse period, granule label was the lowest in the immediately degranulated sample (5%) because most of the granules had been released (ﬁg. 154A). This sample incorporated label primarily into the nucleus compartment (85%). When variable isotope incubation times for constant recovery times were compared, the granule label showed a difference associated with the longest exposure time to isotope. For example, 1- and 2-hour (ﬁg. 155) isotope incubations in 6-hour

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

269

Ultrastructure of Mast Cells and Basophils

270

cultured, post-stimulus samples showed a similar proportion of label in the granules (7 and 8%, respectively) of total cell grains; these values increased to 12% when the 6-hour recovered sample was incubated for an extended time (6 h) in [3H]-uridine (p ! 0.001). Cells from the same donor pulsed with isotope for 18 h after degranulation showed a further small increase in label in the granule compartment (ﬁg. 156D) (14 vs. 12%; p ! 0.05). Thus, recovering HLMCs most signiﬁcantly increased the incorporation of [3H]-uridine into granules between 2 and 6 h of exposure to uridine and recovery from degranulation, an interval during which the morphology of granule recovery also occurs [7, 167, 224, 424, 425]. Two samples from the same donor were examined in which the isotope exposure time (1 h) was held constant and the recovery time after degranulation was tripled (6 and 18 h). In this instance, the relative granule label (7%) nearly doubled (12%), concomitant with enhanced morphological granule recovery evident between these recovery intervals (p ! 0.001). This value for the granule compartment (12%) was nearly identical to that for 18-hour exposures to isotope in two different preparations (14%, 14%). All three of these poststimulation values for granules in recovering cells (ﬁg. 156B–D) (12, 14, 14%) were less than the granule compartment value in the 18-hour control sample incubated for the full 18 h in [3H]-uridine (ﬁg. 152, 153) (20%; p ! 0.001). Thus, granule recovery was not complete in all cells at the longest recovery time that we studied, and the relative proportion of granule grains indicating [3H]-uridine also had not reached the control value of fully granulated cells. In these instances, the proportion of grains in granules and nuclei exceeded that for mitochondria and Epon. Fig. 156. HLMCs recovered 18 h after anti-IgE, exposed to [3H]-uridine for 3 min (A) or 18 h (B–D) and prepared for EM autoradiography with chemical (A, B, D) or physical (C) developer, shows RNA in granules. The labeled mixed scroll-particle granule in A shows the presence of rapidly labeled RNA (3 min). The labeled granule in D contains only particles; two single, unlabeled scrolls (arrows) are present in the cytoplasm immediately adjacent to the labeled granule. Each scroll encloses dense particles centrally. Two granules are heavily labeled with silver grains (B), nearly obscuring the underlying morphologic granule structure. Other granules nearby show numerous electron-dense particles as in the labeled granules. The physical developer has produced small, round silver grains (C) which do not obscure underlying details in granules. The central granule has three silver grains, well-formed scrolls and numerous ribosomes (arrows, 1) identical to the cytoplasmic ribosomes (arrows, 2) which surround this granule. One such cluster is labeled with a silver grain (arrow, 3), as is the center of a longitudinally oriented scroll in an adjacent granule (arrow, 4). A !42,000. B !46,000. C !42,000. D !66,000. [From 175, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

271

The autoradiographic studies that we performed [16, 175] had internal positive controls for the incorporation of radiolabeled uridine into RNA. A great deal of ultrastructural literature exists in which the distribution and kinetics of uridine incorporation are recorded for nuclear and cytoplasmic sites that are known to contain RNA [433, 434, 437, 441, 444, 445, 448, 449, 454]. As in these studies, our radiolabeled HLMCs incorporated uridine into nuclear RNA domains for ribosomal RNA, pre-mRNA and hnRNA processing, mRNA splicing sites, RNA transport through nuclear pores [458–462] and free cytoplasmic polyribosomes. Generally, nuclear and cytoplasmic RNA sites, like granule sites, were signiﬁcantly labeled in excess of mitochondrial and background Epon label. The incorporation of radiolabeled uridine into granule RNA varied with the functional status of the cells. For example, granule label decreased from control values to 5% of cell label at the height of stimulated granule release during histamine secretion. Also, reconstitution of HLMC granules 18 h after the secretory stimulus, while not morphologically complete, was nonetheless extensive [7, 167, 224, 424, 425]. In these cells, degranulation channels recovering electron-dense contents were also labeled with uridine, indicating the presence of RNA in these condensing, partitioning structures as they give rise to newly ﬁlled secretory granules. Extensive literature exists on the nature of rapidly labeled and slowly labeled RNA-rich cellular organelles and domains using kinetic studies of the incorporation of radiolabeled uridine [433, 436, 441, 442, 444, 449, 451, 463, 464]. Generally, the most rapidly labeled RNA is hnRNA, some of which, after splicing and polyadenylation to form mRNA, is exported to the cytoplasm. Ribosomal RNA is more slowly processed and, thus, more slowly labeled with [3H]-uridine prior to passage as elongated electron-dense structures [458–462] through nuclear pores to the cytoplasm. Even more slowly labeled and stable RNA pools are those known as small nuclear (sn)RNAs [451], which associate with speciﬁc proteins to form snRNPs, many of which are involved with the splicing of mRNA [465–467]. Kinetic experiments to detect different RNA pools are more complex than the ones done in our studies and require the examination of radiolabeled ultracentrifugation fractions of labeled cells exposed to speciﬁc inhibitors of various programs for RNA processing. However, it is of interest that rapidly labeled RNA (3 min) was localized to granules (and, as expected, to the nucleus). Thus, the kinetic labeling criteria for hnRNA [442] (i.e., pre-mRNA) and/or mRNA in secretory granules were met. In both control and stimulated cells given short (3-minute) isotope pulses, the granule compartment label signiﬁcantly exceeded that for the cytoplasmic compartment.

Ultrastructure of Mast Cells and Basophils

272

Fig. 157. [3H]-uridine incorporation into HLMC lipid body RNA. These ultrastructural autoradiographs of HLMCs recovered in culture 6 h after degranulation show electron-dense silver grains in the nuclear (N) matrix in A and in lipid bodies in A and B. One cytoplasmic silver grain is also present (B). M = Mitochondria. A !36,000. B !55,000. [From 17, with permission.]

In addition to nuclear, cytoplasmic and granule sites of RNA in HMCs, cytoplasmic lipid bodies also incorporated [3H]-uridine (ﬁg. 157). We noted the propensity for lipid bodies in close proximity to nuclear envelopes to incorporate [3H]-uridine into RNA (ﬁg. 157) and that not all lipid bodies incorporated [3H]-uridine. Thus, ultrastructural studies identiﬁed two new organellar sites of RNA – secretory granules and lipid bodies [16, 17, 175].

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

273

Fig. 158. HLMC secretory granules and intervening cytoplasmic ribosomes are labeled for the presence of uridine with 20-nm gold particles attached to the secondary antibody. !61,000. [From 173, with permission.]

Ultrastructure of Mast Cells and Basophils

274

Fig. 159. Higher magniﬁcation views of HLMC scroll granules show gold-labeled uridine bound to granule scrolls and cytoplasmic ribosomes (arrow). Note that ribosomes and the 20-nm gold particles are of similar size. A !98,000. B !72,000. [From 173, with permission.]

10.5. Ultrastructural Immunogold Imaging of Uridine in Human Mast Cell Granule and Lipid Body RNA

A post-embedding immunogold protocol, utilizing a rabbit polyclonal antibody to uridine that recognizes RNA and is known to work on aldehydeﬁxed material, labeled HMCs [16, 17, 173]. Secretory granules (ﬁg. 158, 159), cytoplasmic ribosomes (ﬁg. 158, 159A), lipid bodies (ﬁg. 160–162) and nuclei (ﬁg. 162) were labeled, indicating the presence of uridine, a requisite RNA precursor.

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

275

Ultrastructure of Mast Cells and Basophils

276

Anti-uridine-stained lipid bodies were closely adjacent to elements of smooth endoplasmic reticulum (ﬁg. 160A), RER, free ribosomes (ﬁg. 160B, 161, 162A), and intermediate ﬁlament bundles that were also labeled for uridine (ﬁg. 160B, C, 161). Collections of free ribosomes nearby were heavily labeled (ﬁg. 161). Lipid bodies intimately associated with perinuclear membranous boundaries contained uridine-labeled RNA (ﬁg. 162). In some instances, linearly aligned gold particles ﬁlled the visible space between lipid body surfaces and the endoplasmic reticulum-invested perinuclear cisterns (ﬁg. 162). Lipid bodies with visible internal vesicular and electron-dense structures were present regardless of the unmasking procedure used. These vesicle-like structures, when present, particularly bound the anti-uridine label to their surfaces indicating the presence of RNA in these sublipid body domains (ﬁg. 160C). In experiments where uridine incorporation into lipid bodies had been previously assessed with autoradiography, we noted enhanced expression of immunoreactive uridine in lipid bodies with the immunogold method. This was particularly evident when previously degranulated HMCs were studied during late (e.g., 6 h) recovery intervals compared to control cells. Variable unmasking schedules with sodium metaperiodate, HCl, hydrogen peroxide, hyaluronidase, and combinations of several of these were tried. It was determined that hyaluronidase unmasking was the best and most reliable, and most samples were then prepared with this method [156, 157]. Permeabilization with Triton-X was also helpful for granule uridine detection but unmasking with sodium metaperiodate or Triton-X did not allow detection of uridine in lipid bodies due to the fragility of these structures [17]. Positive controls for the antibody to uridine were present in the same samples. In addition to free cytoplasmic ribosome clusters in mast cells (ﬁg. 159A), ribosomes that were attached to extensive arrays of endoplasmic reticulum in plasma cells were labeled (ﬁg. 163A). The smooth-contoured 20-nm gold particles, bound to the secondary antibody, were attached to endoplasmic reticulum as were the approximately 20- to 25-nm, electron-dense, ragged ribosome particles (ﬁg. 163A).

Fig. 160. Immunoreactive uridine in HLMC lipid bodies. These immunogold preparations of 6 h cultured control HMCs demonstrate RNA in lipid bodies (A–C), in lipid body (LB) matrix vesicles (C), in smooth endoplasmic reticulum next to a lipid body (A), and in ribosomes and intermediate ﬁlaments which encase lipid bodies (B, C). The unmasking procedure for these images was hyaluronidase. A !47,600. B !78,200. C !91,500. [From 17, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

277

Fig. 161. Immunoreactive uridine indicating RNA in a HLMC recovered in culture 6 h after degranulation. Note extensive labeling of three cytoplasmic lipid bodies (LB) encased in intermediate ﬁlaments studded with ribosomes. A large cytoplasmic collection of ribosomes is also extensively labeled (arrowhead). !80,000. [From 17, with permission.]

Ultrastructure of Mast Cells and Basophils

278

Negative controls for the immunogold procedure were also obtained (ﬁg. 163B–D, 164A, B). These included omission of the speciﬁc primary antibody to uridine (ﬁg. 163B, 164A), substitution of normal rabbit serum (ﬁg. 163C) or normal rabbit IgG for the speciﬁc primary antibody, and an absorption control with polyuridine [poly(U)] beads of the speciﬁc primary antibody prior to staining en grid (ﬁg. 163D, 164B). A control for this absorption control was absorption of the primary antibody with polyadenine [poly(A)] beads. In this last instance, HMC granules, ribosomes, lipid bodies and nuclei retained their gold label for uridine. These results show that HMC granules and lipid bodies contain uridine, an ingredient of RNA species generally, and complement our ﬁnding of the incorporation of [3H]-uridine by autoradiography into HMC granules and lipid bodies [16, 17, 175]. These two disparate means of detection leave little doubt that HMC granules and lipid bodies contain uridine, which is integral to all RNA species.

10.6. Ultrastructural Cytochemical, Immunocytochemical and in situ Hybridization Methods Detect mRNA in Human Mast Cell Granules and Lipid Bodies

We developed (or modiﬁed [468–470]) ﬁve different methods to detect poly(A)-positive mRNA at the ultrastructural level [16, 17, 172, 428]. A series of afﬁnity probes was prepared, and the direct binding of these probes to thin sections of HMCs en grid [as well as in in situ hybridization (ISH) protocols] to localize mRNA [16, 17, 172], based on the afﬁnity of poly(U) for the poly(A) tail of mRNA [453], was tested [16, 17, 172]. Ultrastructural ISH has been developed to provide precise subcellular localization of speciﬁc mRNAs [470–485]. In our work [16, 17, 172, 428], we prepared general poly(U) probes to detect poly(A)-positive mRNA to image the site(s) of mRNA in mature HMCs. The poly(U) probes that we used included tritium-labeled poly(U), gold-labeled poly(U), biotinylated poly(U), BSA-conjugated, gold-labeled poly(U), and unlabeled poly(U). Detection methods used to discern either direct binding or ISH-facilitated binding to poly(A)-positive mRNA included ultrastructural autoradiography [to detect [3H]-poly(U)], direct inspection [for poly(U)-gold and BSA-conjugated, gold-labeled poly(U)], immunogold imaging systems [to detect the photobiotinylated poly(U) and unlabeled poly(U)]. The results were quantiﬁed by counting particulate reporters in individual mast cell compartments and calculating the percent of total label for each compartment and for the Epon background. Appropriate controls for all of these procedures were similarly quantiﬁed [16, 172].

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

279

ISH with [3H]-poly(U) en grid resulted in ultrastructural images which showed autoradiographic grains overlying HMCs (ﬁg. 165, 166A, B). Nearly half of the observed silver grains reﬂected hybridization of [3H]-poly(U) to secretory granules (ﬁg. 165A). In contrast, only 3% of the silver grains were found over the background Epon; with increased exposure time for detecting the label by autoradiography, this background value increased. The nuclear (36%) (ﬁg. 165A) and cytoplasmic (13%) compartments exceeded background (3%) but were both less than the granule (48%) (ﬁg. 165) compartment at 7 days’ exposure. Lipid bodies also were labeled with this ISH method (ﬁg. 166A, B) indicating the presence of mRNA in these organelles as well as in granules. When the digestion with DNase was omitted (a step needed for speciﬁc annealing in the ISH protocol), no mast cell compartments were labeled. Omission of digestion with RNase (a step needed to remove unbound [3H]-poly(U) from the grid) resulted in excessive and random silver grains overlying all cell compartments as well as the Epon background. Direct binding of the probe, poly(U)-gold (as well as binding after ISH) to granules and lipid bodies (ﬁg. 166C) was demonstrated. Both fused, electron-dense, lipid bodies (ﬁg. 166C) and lipid bodies illustrating an internal honeycombed pattern had poly(U)-gold bound to them. Passage of poly(U)gold over poly(A) beads before exposing the sample to poly(U)-gold removed lipid body labeling. The poly(U)-gold reagent was used with an ISH protocol described for light microscopic ISH [469], modiﬁed for electron microscopic ISH, to determine whether speciﬁc signal for mRNA could be enhanced in HMC compartments. These results showed that secretory granules and nuclei did bind poly(U)-gold after ISH (ﬁg. 167A). When counts of gold particles were compared to a variety of controls, the degree of label in granules (p ! 0.001) and nuclei (p ! 0.05) signiﬁcantly exceeded controls. The controls that used bead absorptions allowed a visual determination of the expected gold label. For example, poly(U) beads did not bind poly(U)-gold, remained white, and granules that were stained with the efﬂuent retained their gold label (ﬁg. 167B), whereas poly(A) beads did bind poly(U)-gold, turned red and abrogated granule and nuclear binding of gold when mast cells were stained with the efﬂuent from the beads (ﬁg. 167C). FurFig. 162. Six-hour control cultured HMC lipid bodies (LB) show immunoreactive uridine indicating RNA. These preparations were unmasked with hyaluronidase and show intensively labeled nuclei (N), LBs, and granules (G). Note the linear array of gold particles along LB surfaces and intensive packing of label at the LB-nuclear interface in the region of the nuclear membranes and perinuclear cisternae. A !80,000. B !99,000. [From 17, with permission.]

Ultrastructure of Mast Cells and Basophils

280

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

281

ther speciﬁcity for the poly(U)-gold reagent for ISH was determined by staining thin sections containing either poly(U)-agar or poly(A)-agar. In this case, as expected, the poly(U)-gold signiﬁcantly bound to poly(A)-agar (6.9 8 0.6) but not to poly(U)-agar (1.1 8 0.4) (p ! 0.001). The use of poly(U)-gold as a simple afﬁnity reagent to detect mRNA in electron microscopic samples allowed the detection of label in the granular (11.5 8 2.4) or nuclear (8.2 8 1.4) compartments; the efﬁciency of detection in the granule or nuclear compartment was enhanced when an ISH protocol was used with this reagent (granule: 28.4 8 5.7; nucleus: 16.4 8 3.2). The photobiotinylated poly(U) probe also labeled subcellular structures in HMCs (ﬁg. 168–171). To test the authenticity of this labeling, two independent ISH protocols [469, 470] were used, as well as four independent reporter systems [469, 470, 486, 487] for imaging the biotinylated poly(U) bound to mRNA in subcellular locations of HMCs. The impact of proteinase K digestion, omission of hybridization and speciﬁc controls for the individual reporter systems were also evaluated quantitatively. These data show that, with an ISH protocol, the photobiotinylated poly(U) probe consistently labeled the nuclear, cytoplasmic, granular and lipid body compartments (ﬁg. 168A, B, 169A, 170A, B, 171A) [17, 172]. Quantitative evaluation revealed that the lipid body compartment label signiﬁcantly exceeded that for mitochondria, another cytoplasmic organelle in HMCs (p ! 0.001) and that granules exceeded background Epon and cytoplasm label (p ! 0.001) and was equal to nuclear label. When the ISH protocol was omitted to determine the ability of the biotinylated poly(U) probe to bind directly to the samples, similar overall trends to those with ISH occurred. When samples without ISH were examined with or without proteinase K digestion, very little difference was noted in labeling. We quantitated the gold label in HMCs imaged with a second reporter system – protein A-gold – after the ISH protocol and in omission controls where the goat antibiotin antibody was deleted or the biotinylated poly(U) probe was deleted (ﬁg. 169, 170). With protein A-gold after the ISH protocol and exposure to the biotinylated poly(U) probe, granule (58%) and lipid body (44%) label exceeded Epon background (2%) (p ! 0.001); granule label exceeded that of lipid Fig. 163. Controls for the immunogold detection of uridine. A A high magniﬁcation view of rough endoplasmic reticulum in the cytoplasm of a plasma cell shows gold bound to the membrane-bound ribosomes. B The speciﬁc primary antibody was omitted. Note that the mast cell granule does not bind the gold-labeled secondary antibody. C Normal rabbit serum was substituted for the polyclonal rabbit antibody speciﬁc for uridine. Only a few gold particles of the secondary antibody are present, constituting a negative control. D When the speciﬁc primary antibody was absorbed with poly(U) beads prior to use, as shown, all cytoplasmic and granule labeling is removed. A !66,500. B !84,500. C !65,500. D !73,000. [From 173, with permission.]

Ultrastructure of Mast Cells and Basophils

282

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

283

Fig. 164. Controls for the anti-uridine immunogold stain of 6 h cultured HLMCs (A, B). A The speciﬁc anti-uridine antibody was omitted. No gold particles bind to the lipid body (LB), granules (G) or to cytoplasmic ribosomes nearby. B The speciﬁc antibody to uridine was absorbed with polyuridine before staining the sample. The LB and G are not labeled. A !54,000. B !72,000. [From 17, with permission.]

Ultrastructure of Mast Cells and Basophils

284

Fig. 165. [3H]-poly(U) electron microscopic ISH of HLMCs. A A low magniﬁcation image after 1 month exposure to emulsion en grid shows a large number of silver grains over a nucleus (N) and cytoplasmic secretory granules. Granule label is shown at higher magniﬁcation. In B a single silver grain spans a nuclear pore (dashed line indicates location of perinuclear membrane in this preparation which was not osmicated or stained en bloc with uranyl acetate) overlying the nucleus (N) below and a granule (G) in the cytoplasm above. In C the silver grain labels the underlying granule. A !15,500. B !110,500. C !76,000. [From 172, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

285

Fig. 166. Six-hour control cultured HLMC lipid bodies show label indicating poly(A)positive mRNA. A, B Autoradiographs of tritiated poly(U) hybridized to the samples show silver grains overlying lipid bodies. The samples were not osmicated or exposed to uranyl en bloc. This method of preparation preserves little of the osmiophilic lipid contents of LBs, leaving a thick electron-dense outer shell to the organelle. G = Granule. C A poly(U)gold probe was used to directly stain the lipid bodies which are joined by a narrow lipidrich fusion area between them. This is a standard ultrastructural preparation (compare with A and B). A, B !52,000. C !77,000. [From 17, with permission.]

Ultrastructure of Mast Cells and Basophils

286

Fig. 167. HLMC granules labeled with a poly(U) method to identify mRNA. Isolated, puriﬁed HLMCs were cultured for 6 h, ﬁxed, and processed for electron microscopy and prepared to detect poly(A) in mRNA by ISH using poly(U)-gold (A), by ISH with poly(U)-gold after absorption of poly(U)-gold with poly(U)-Sepharose beads (B), or by absorption of poly(U)-gold with poly(A)-Sepharose beads (C). Granules are labeled in A. Absorption with poly(U) did not remove labeling with the poly(U)-gold probe in the ISH procedure (B), but poly(A) absorption did not abrogate granule labeling (C). A !126,500. B !113,000. C !122,500. [From 16, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

287

bodies (p ! 0.001) [172]. Use of either streptavidin-gold or goat-antibiotin-gold (ﬁg. 171A) as reporters also labeled HMC granules and lipid bodies. A variety of controls were quantitated to evaluate speciﬁcity of the electron microscopic ISH using the biotinylated poly(U) probe. When the biotinylated poly(U) probe plus the primary antibody to biotin were omitted, all cellular and granular and lipid body label was eradicated. Thus, the Photoprobe (long arm) biotin was essential to image mRNA in HMCs, and the gold-labeled secondary antibody did not artifactually bind to the samples but was speciﬁcally bound to biotin in the poly(U) probe. When the primary antibiotin antibody was omitted using the protein A-gold reporter system, protein A-gold did not bind to the biotinylated poly(U) probe (but did increase background levels of label). When the biotinylated poly(U) probe was omitted, the primary antibody to biotin did not stick to HMCs, and when the biotinylated probe and the primary antibiotin antibody were omitted, protein A-gold did not stick to the sample (ﬁg. 170C). Omission of the biotinylated poly(U) probe in conjunction with the streptavidin-gold or goat-antibiotin-gold reporters also was negative (ﬁg. 171B). A BSA-poly(U)-gold probe was constructed for use in localizing mRNA by electron microscopic ISH. The probe was made by cross-linking BSA to poly(U) with glutaraldehyde, preparing a colloidal gold suspension, and attaching the BSA-poly(U) to gold. This general probe for mRNA labeled the granular, nuclear, cytoplasmic and lipid body compartments [17, 172] in HMCs isolated from surgically removed lung specimens (ﬁg. 172–176). To make this determination, we evaluated a number of procedural variations in three different ISH protocols as well as in a number of controls. [To properly evaluate the lipid body compartment, we found that the short ISH protocol was needed since the long protocol generally extracted lipid body content from samples (ﬁg. 176A).] Controls included BSA-gold with and without ISH [the BSA used for conjugation to gold was treated with glutaraldehyde identically to the treatment for cross-linking BSA to poly(U)] and gold alone with and without ISH (ﬁg. 173E,

Fig. 168. Biotinylated poly(U) electron microscopic ISH [biotinylated poly(U) prepared with Photoprobe (long arm) biotin] and direct binding of biotinylated poly(U) to human mast cells imaged with an immunogold reporter. High magniﬁcation images of secretory granules show biotinylated poly(U) directly bound to granule scrolls (A, B) and a labeled scroll granule after ISH with biotinylated poly(U) (C). Speciﬁcity controls for the immunogold reporter system are negative (D, E). D Omission of the biotinylated poly(U) probe in the ISH protocol. E Omission of the biotinylated poly(U) probe in the ISH protocol and the goat-antibiotin antibody from the immunogold reporter. A !101,000. B!154,000. C !145,000. D !112,500. E !55,500. [From 172, with permission.]

Ultrastructure of Mast Cells and Basophils

288

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

289

174C, 175B, 176C). These controls showed markedly reduced attachment to subcellular compartments of HMCs compared to samples exposed to the BSApoly(U)-gold probe. Generally, Epon background gold counts were increased after ISH schedules, compared to direct binding of the probe in the absence of ISH. Increased Epon background was similarly noted when the control BSAgold probe underwent ISH. Secretory granules in HMCs bound BSA-poly(U)-gold directly but with less efﬁciency quantitatively than when an ISH protocol was used (911 vs. 4,305; p ! 0.001). The controls, BSA-gold and gold alone bound less frequently to granules than BSA-poly(U)-gold in either direct binding experiments (50 vs. 911; p ! 0.001) or after an ISH protocol (788 vs. 4,305; p ! 0.001). When comparisons of the BSA-poly(U)-gold-labeled granule compartment (911) were made to other cellular organelles (nucleus: 466, cytoplasm: 263) in the same experiments, granules contained more gold label than other organelles (p ! 0.001, p ! 0.001), whereas controls were similar among compartments. RNA-rich regions of the HMC nucleus and cytoplasm also labeled with BSA-poly(U)-gold ISH. In the nucleus, label generally was at the interface between peripheral condensed chromatin and central dispersed chromatin (ﬁg. 172, 174, 175). In the cytoplasm, focal aggregates of free ribosomes, often adjacent to granules, were gold-labeled. BSA-poly(U)-gold ISH-labeled nuclei more than the control BSA-gold ISH and gold only did (1,757 vs. 461; p ! 0.005), and direct binding of the BSA-poly(U)-gold probe to nuclei exceeded the BSA-gold and gold-only controls (466 vs. 23; p ! 0.05). Similarly, cytoplasmic gold counts for samples exposed either directly or with ISH to the BSA-poly(U)-gold probe exceeded those for controls (263 vs. 32; p ! 0.05; 1,451 vs. 797; p ! 0.005). A less destructive procedure for imaging ISH of lipid bodies was devised [17, 428]. This involved hybridization (short protocol) with unlabeled poly(U) and subsequent imaging of speciﬁcally bound poly(A)-positive mRNA label with a speciﬁc antibody to uridine and a gold-labeled secondary antibody directed toward the primary anti-uridine antibody. With this procedure, expected mRNA-containing structures, secretory granules and lipid bodies (and

Fig. 169. Biotinylated poly(U) electron microscopic ISH [biotinylated poly(U) prepared with Photoprobe (long arm) biotin] of human mast cell granule label imaged with a protein A-gold reporter (A). Note everal ribosomes adjacent to the labeled granule are also labeled with protein A-gold (A). Speciﬁcity controls for the protein A-gold reporter system are negative (B, C). B Omission of the biotinylated poly(U) probe in the ISH protocol. C Omission of the biotinylated poly(U) probe from the ISH protocol and the goat-antibiotin antibody from the protein A-gold reporter. A !120,000. B !90,000. C !79,000. [From 172, with permission.]

Ultrastructure of Mast Cells and Basophils

290

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

291

Ultrastructure of Mast Cells and Basophils

292

Fig. 171. Lipid bodies, prepared as in ﬁgure 170, show poly(A)-positive mRNA. In this case the reporter system used after ISH of the biotinylated poly(U) probe was a gold-labeled antibiotin antibody. The lipid body (A) is labeled with gold particles; the biotinylated poly(U) probe was omitted and the lipid body is not labeled by the gold-labeled reporter antibody (B). A !45,000. B !55,000. [From 17, with permission.]

Fig. 170. Six-hour control cultured HLMC lipid bodies show label indicating poly(A)positive mRNA. A, B ISH of biotinylated poly(U), imaged with sequential incubation of an antibiotin antibody and gold-labeled protein A, shows gold particles overlying large lipid bodies. C A control for protein A-gold sticking since ISH of the biotinylated poly(U) probe and the anti-biotin antibody were omitted. Lipid bodies are not labeled by protein A-gold alone. A !52,000. B !79,000. C !49,500. [From 17, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

293

Fig. 172. BSA-poly(U)-gold direct binding to human mast cells. Gold particles are present in the nucleus at the interface of condensed and dispersed chromatin (A), directly in nuclear pores (B) and overlying secretory granules (C). A !50,000. B !87,000. C!71,500. [From 172, with permission.]

Ultrastructure of Mast Cells and Basophils

294

Fig. 173. High magniﬁcation views of human mast cell granules directly binding the BSA poly(U)-gold probe (A–D) or a probe omission control of gold alone binding (E), which is negative. Note the electron-dense, ragged ribosomes surrounding and within granules (arrows, A). A !88,000. B !84,000. C !127,000. D !75,000. E !53,000. [From 172, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

295

Fig. 174. BSA-poly(U)-gold electron microscopic ISH of HLMCs. The modiﬁed long ISH protocol [470] was used; note that gold particles are plentifully bound to the nucleus and secretory granules (A, B), but not when ISH with a BSA-gold probe lacking poly(U) was used (C). A !42,000. B !57,500. C !122,000. [From 172, with permission.]

their internal vesicular substructures) were labeled indicating the presence of poly(A)-positive mRNA (ﬁg. 177, 178). These studies, with multiply labeled and imaged poly(U) probes, showed binding to poly(A)-positive mRNA in HMC granular, cytoplasmic, nuclear and lipid body compartments. These results substantially enhance the hypothesis that site-speciﬁc synthesis in secretory-storage granules may occur in secretory cells generally and that lipid bodies in mammalian cells may also participate in site-speciﬁc synthesis.

Ultrastructure of Mast Cells and Basophils

296

Fig. 175. BSA-poly(U)-gold electron microscopic ISH of human mast cells (A) and a gold-only ISH control (B). The modiﬁed short ISH protocol [469] was used. The distribution of label is in the nucleus and secretory granules with the BSA-poly(U)-gold probe; gold alone, in the probe omission short ISH protocol control, does not bind to nucleus or granules (B). A !76,200. B !75,000. [From 172, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

297

10.7. Ultrastructural Immunogold Cytochemistry of HMC with Specific Autoimmune Human Sera Detects RNA-Associated Protein Antigens of Different RNA Species in and Associated with Lipid Bodies and Granules

Autoantibodies circulate in humans during speciﬁc diseases [488–491]. Some of these are speciﬁc for different types of ribonucleoproteins and occur regularly in patients with systemic lupus erythematosus and related diseases. We selected several of these well-characterized human autoantibodies to examine subcellular localization(s) in HMCs with post-embedding immunogold methods [16, 17, 173]. Primary antibodies used for post-embedding immunogold staining included the puriﬁed IgG fractions of screened and characterized human autoimmune disease plasma containing anti-Sm, anti-RNP, antiSS-A (Ro), anti-SS-B (La), anti-ribosomal P and anti-mitochondrial activity (all from ImmunoVision, Springdale, Ariz., USA). Reference primary antibodies for SS-B, SS-A, Sm, and RNP were also obtained from the Arthritis Foundation/Center for Disease Control (Atlanta, Ga., USA) and used for post-embedding immunogold labeling. HMC granules, lipid bodies and nuclei were stained with the immunogold procedure with human sera speciﬁc for Ro (a small cytoplasmic RNP subclass of La) [17, 173, 492] (ﬁg. 179A, 180A, B), for La (precursors of 5S ribosomal RNA and tRNA) [17, 173] (ﬁg. 179B, 180C), for RNP (indicating U1, a snRNA [16, 17, 173], essential for splicing) [465–467, 493] (ﬁg. 181, 182A), for Sm (indicating U2, U5, U4/U6 snRNAs) [17] (ﬁg. 180D), and for P (representing three phosphoproteins (called P0, P1, P2) of the 60S ribosomal subunit) [494, 495] (ﬁg. 183A, 184A, B). Perigranular ribosomes also were labeled (ﬁg. 183A) with the sera speciﬁc for the ribosomal P antigen [173]. Controls for the speciﬁcity of the granule and ribosome labeling with speciﬁc autoantibody-containing human sera were done (ﬁg. 182B, 183B, 184C, 185). These included omission of speciﬁc primary antibody (ﬁg. 183B, 184C), substitution of normal human serum for the speciﬁc primary antibody (ﬁg. 182B, 185A), and substitution of an irrelevant autoantibody-containing human serum speciﬁc for mitochondria for the primary antibody [in this case, a positive control included speciﬁcally labeled mitochondria (ﬁg. 185B), whereas mast cell granules and lipid bodies were negative]. Lamellar bodies, which are surfactant-containing secretory granules, in type II alveolar pneumocytes, also present in the samples, did not bind the serum speciﬁc for anti-RNP (ﬁg. 185C). Variable combinations of unmasking procedures for these stainings were done. In this case, sodium metaperiodate provided the best unmasking results and was used to demonstrate RNP, Sm, Ro, and La. Ribosomal P antigen was

Ultrastructure of Mast Cells and Basophils

298

Fig. 176. ISH of 6 h control cultured HLMCs. The probe used here was BSA-poly(U)gold. An overnight (long) ISH schedule was used (A). Note that the nucleus (N) is heavily gold labeled for poly(A)-positive mRNA and/or their precursors but that this destructive ISH schedule has caused the lipid body to be torn from the thin section leaving rolled up edges of lipid material to surround the empty LB-sized space (star). A short (3.5 h) ISH schedule was used (B, C), either with the BSA-poly(U)-gold probe present (B) or absent (C). In the short ISH schedule LBs are not destroyed (B, C) and when the speciﬁc probe is present lipid bodies bind it as indicated by gold particles overlying the LB (B). C Omission of the probe abrogates LB label. A !88,000. B !83,500. C !92,500. [From 17, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

299

Fig. 177. ISH of HLMC granules with poly(U) imaged with an antibody to uridine and gold-labeled secondary antibody. Note that granules and perigranular ribosomes (arrowhead) are labeled. !74,000. [From 428, with permission.]

Ultrastructure of Mast Cells and Basophils

300

readily demonstrable with either sodium metaperiodate, with hyaluronidase or combinations of these unmasking procedures. These studies show that HMC granules and lipid bodies contain RNPs known to associate with uridine-rich, non-polyadenylated, stable snRNAs, which are important for splicing mRNA precursors [465, 496–501] and with ribosomes, the organelle responsible for protein synthesis [494, 495]. Positive speciﬁcity for the primary reagents was shown by labeling the expected nuclear sites involved with RNA processing, as well as cytoplasmic ribosomes [502– 504]. Use of the human sera speciﬁc for ribosomal proteins (anti-P) [494, 495] as the primary antibody in immunogold staining labeled HMC granules, lipid bodies and cytoplasmic ribosomes [17, 173], complementing the imaging of RNPs associated with ribosomes with the EDTA stain [16, 17, 174]. snRNAs are involved in splicing of pre-mRNAs, to produce mRNAs for subsequent protein synthesis. The snRNAs were localized by immunogold staining with speciﬁc human antisera [16, 17, 173]. They are uridine-rich [496] (and, thus, would be imaged with [3H]-uridine [16, 17, 175] and anti-uridine [16, 17, 173]), but they do not contain poly(A) [496] (and, thus, would not be imaged with the poly(U)-based methods which we used [17, 172]). The human antisera to snRNPs are present in patients with certain autoimmune diseases [488, 489, 491, 505–508]. Sera containing antibody to Sm are speciﬁcally present in patients with systemic lupus erythematosus and, in the antisera that we used, are reactive with polypeptides shared by U2 and U4–6 snRNP [488]. Sera containing antibody to RNP are present primarily in patients with mixed connective tissue disease; the sera we used were speciﬁc for proteins associated only with U1 snRNP. U1 has been speciﬁcally implicated as essential in the splicing of mRNA precursors [466, 493, 509–512]. Thus, small RNPs, important in the processing of mRNA precursors, are located in HMC secretory granules and lipid bodies, as well as in expected nuclear sites. It has been suggested previously that these snRNAs actually spend brief periods in the cytoplasm, prior to returning to the nucleus where they are long-lived [513, 514]. Speciﬁcally, cytoplasmic sites of U1 RNA processing and RNP assembly have been proposed [515]. Our data suggest that secretory granules and lipid bodies may be such sites in HMCs [16, 17, 173]. We also tested human sera containing antibodies to Ro and La [492, 505, 516–521]. These are present primarily in patients with Sjögren’s syndrome. Our sera for Ro were speciﬁc, being non-reactive to La, RNP and Sm; but the sera for La had low reactivity with Ro as well. These small RNPs are said to be primarily cytoplasmic (Ro) or nuclear (La), but no localization studies for secretory granules and lipid bodies have been reported. A possible role for Ro in translation of subsets of mRNAs in heart and brain has been proposed [518]; nucleoli and ribosome-rich cytoplasmic cellular regions suggested to other in-

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

301

Ultrastructure of Mast Cells and Basophils

302

vestigators a role(s) in ribosome synthesis, assembly or transport [492]. Thus, localization of Ro in HMC secretory granules and lipid bodies adds these locations to the possible repertoire for this small cytoplasmic RNP [17, 173]. La, another small RNP, recognized by antiserum to La, labels precursor forms of 5S ribosomal rRNA and certain transfer (t)RNAs (e.g., those for the amino acids Met, Asp, Gly, Glu, Asn) [522]. Initially thought to be a snRNP, La has been localized to nucleolar and cytoplasmic sites, particularly to RER and to polyribosomes. Cytoplasmic RNAs bound to La include the signal recognition particle (srp) RNA. Proposed functions of this small RNP include involvement in protein synthesis via tRNAs and the srpRNA, modulation of mRNA translation, participation in functional aspects of the biogenesis of polymerase III-transcribed RNAs, in nuclear retention of RNA polymerase III transcripts, and in ribosomal biosynthesis in the nucleolus [522–525]. Our immunoelectron microscopic localization of La in HMC secretory granules and lipid bodies (with positive nuclear labeling) identiﬁes additional RNA species in the granule and lipid body locations [17, 173].

10.8. Comment regarding Human Mast Cell Granules and Lipid Bodies Containing RNA Machinery Important for Protein Synthesis

Ultrastructural observations identiﬁed close associations of ribosomes, granules and lipid bodies in resting mature HMCs – associations that suggest non-traditional sites for protein synthesis in secretory cells poorly endowed with RER and Golgi structures [16, 17, 427]. The observed intragranular particles approximated the size, shape and electron density of ribosomes in the perigranular and perilipid body cytoplasm of HMCs. Particulate content within mixed granules and a subset of granules, termed particulate granules, in HMCs, could be a source of interpretive confusion regarding the presence of ribosomes within granules [54]. However, the particulate content of the mixed and particle granules consists of larger and more uniformly shaped electrondense structures than the ragged, ⬃25-nm electron-dense ribosomes. The latter intragranular structures also decreased and increased in recovery from func-

Fig. 178. ISH of 6 h control cultured HLMCs slow poly(A)-positive mRNA in LBs. The probe used here for hybridization was unlabeled poly(U) that was visualized by an immunogold reporter. The short hybridization schedule allowed excellent preservation of LBs, which illustrate overlying gold particles in all panels. Intralipid body vesicles (B, C) are also gold-labeled indicating the presence of poly(A)-positive mRNA. N = Nucleus; G = granule. A !74,000. B !84,000. C !74,000. [From 17, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

303

Ultrastructure of Mast Cells and Basophils

304

tional, and during developmental, processes in HMCs [16, 175, 427], much as ribosome numbers do in other functionally and developmentally engaged cells. Another difﬁculty in the recognition of granule or lipid body-associated electron-dense ribosomes in electron microscopic preparations of HMCs is that the granules, lipid bodies and ribosomes are often equally electron-dense, making visualization of the small particles (⬃25-nm) within the larger granules and lipid bodies (⬃1 ␮m2) difﬁcult at best. These observations informed our decision to search further for granule- and lipid body-ribosome relationships in the biology of RNA and HMCs. Initially we used a new enzyme-afﬁnity-gold method to detect RNA based on the afﬁnity of a gold-labeled enzyme, RNase, for its substrate, RNA [215, 242] to explore the possible relationship(s) of HMC granules, lipid bodies and RNA [155, 216]. This method provided excessive granule labeling that qualitatively was diminished when appropriate controls for RNA speciﬁcity were done [155, 216]. These studies did not, however, reach statistical signiﬁcance [155, 216]. We established that the ‘noise’ in this system did reach statistical signiﬁcance for heparin, the proteoglycan speciﬁc for HMC granules and known to inhibit RNase activity. Thus, a new ultrastructural probe, based on afﬁnity-gold detection of an enzyme (RNase) inhibitor (heparin), was established, allowing for accurate detection of heparin in resting, functional and recovering HMCs but disallowing detection of RNA stores in HMC granules with this approach [155, 216]. In samples of HMCs prepared similarly with R-G staining, another prominent organelle, cytoplasmic lipid bodies, was stained. Unlike the extensive studies showing that this staining indicated heparin in granules, lipid bodies appeared to bind the R-G by virtue of contained RNA [17]. That is, prior absorption of the reagent with heparin removed granule staining (e.g., heparin stores) but not lipid body staining, indicating RNA [17]. Thus, a new use for the ultrastructural enzyme-afﬁnity-gold method originally designed to detect RNA, i.e., to detect heparin in HMC granules, was found [155]. The studies also conﬁrmed, as reported [215], that the method detects RNA in ribosomes and nucleoli of HMCs. These ﬁndings provide a new use for this enzyme-afﬁnity-gold method in mast cell studies – one derived from the known biochemical properties of heparin as an RNase inhibitor, and identify a new subcellular location for RNA in HMCs – in lipid bodies [17].

Fig. 179. Immunogold stain to detect Ro (A) and La (B) with speciﬁc autoimmune human sera. In A, HMC granules contain a small degree of labeling for Ro; in B, both granule (G) and nucleus (N) display labeling for La. A !62,500. B !85,000. [From 173, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

305

Ultrastructure of Mast Cells and Basophils

306

Uridine, an RNA precursor, was demonstrated in HMC granules and lipid bodies by two disparate ultrastructural methods [17, 173, 175]. The ﬁrst method demonstrated the ex vivo incorporation of [3H]-uridine into granules and lipid bodies in living cells by ultrastructural autoradiography [16, 17, 175]. The second method identiﬁed the presence of intragranule and intralipid body uridine by post-embedding immunogold stains [17, 173]. Ribosome-associated RNP(s) were imaged in and adjacent to HMC granules and lipid bodies, likewise by two disparate ultrastructural techniques [16, 17, 173, 174]. The ﬁrst of these was an EDTA regressive chelation staining method [16, 17, 174], and the second was an immunogold method which detects ribosome-speciﬁc RNPs with human autoimmune sera (P0, P1, and P2) [17, 173]; both of these methods yielded identical localization results in HMCs [16, 17, 173, 174]. We used direct binding and ultrastructural ISH of general poly(U) probes, prepared in diverse ways, to detect poly(A)-positive mRNA stores in HMCs [16, 17, 172]. These methods demonstrated that poly(A)-positive mRNA species were contained not only in the expected nuclear and cytoplasmic sites but also in the secretory granules and lipid bodies of HMCs. RNA processing in the nucleus involves splicing of precursor mRNA, polyadenylation of pre-mRNAs, and transport of speciﬁc mRNAs from nucleus to cytoplasmic sites – mRNAs that associate with cytoplasmic ribosomes to direct speciﬁc protein syntheses [460, 526–528]. The electron microscopic ISH localizations of granule and lipid body stores of poly(A)-containing mRNAs, as well as nuclear stores of poly(A)-containing precursors of mRNA, indicate that site-speciﬁc localization of mRNAs in cytoplasmic secretory granule and lipid bodies (in addition to cytoplasmic ribosomes) exist. We investigated the presence of individual RNA species in HMCs using speciﬁcity-proven autoimmune sera from humans with autoimmune disorders and performing post-embedding immunogold staining [16, 17, 173]. These sera allowed us to visualize granule and lipid body sites of several RNA species. These included ribosomal RNP (see earlier), snRNPs (RNP and Sm) which associate with uridine-rich, poly(A)-negative snRNAs that are either essential for the splicing of mRNA precursors (U1) or associated with splicing events in RNA processing (U2, U4–6). Additionally, granule sites for a small cytoplasmic RNP, Ro (thought to operate in translation of mRNA subsets and in ribosome synthesis, assembly and transport), and La (representative of 5S riboFig. 180. Immunogold studies with speciﬁc human autoimmune sera show gold labeled LBs in 6 h control cultured HMCs. Sera with speciﬁcity for Ro (A, B), La (C), SM (D) label cytoplasmic LBs. N = Nucleus; G = granule. A !40,000. B !71,000. C !57,600. D !54,500. [From 17, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

307

Fig. 181. Immunogold preparation to detect RNP in HMCs with speciﬁc autoimmune human sera. A Numerous cytoplasmic granules (G) and the nucleus (N) are labeled. B, C Higher magniﬁcation views show gold label for RNP in secretory granules. A !54,500. B !70,000. C !94,000. [From 173, with permission.]

Ultrastructure of Mast Cells and Basophils

308

somal rRNA and certain tRNAs), were found in HMCs. La also is postulated to function widely in RNA biology, including in protein synthesis, mRNA translation, biogenesis and nuclear retention of polymerase III-transcribed RNAs, biogenesis of ribosomes, and has been localized to nucleus, cytoplasm, RER, polyribosomes, and to srpRNA in cells [reviewed in 17, 173].

10.9. RNA: Site-Specific Synthesis

Granule and lipid body stores of mRNA could reﬂect en site synthetic capacities that deﬁne a new role(s) for secretory granules and lipid bodies in general. A large body of evidence has recently deﬁned RNA transport and localization as a mechanism for facilitating site-speciﬁc synthesis in various subcellular organelles and compartments in cells of diverse origin from multiple species [529–538]. Such traveling RNA has been intimately associated with all three major cytoskeletal elements, i.e., microﬁlaments, intermediate ﬁlaments, and microtubules [453, 529, 530, 533, 538–542], and speciﬁc zip codes for localizing individual mRNAs at subcellular sites have been postulated [530, 535, 543, 544]. Secretory granules and lipid bodies have not been considered as such sites, but the evidence that we present suggests that they must also be included in the travels of mRNA [532, 545, 546] in mast cell biology. While secretory granules and lipid bodies have only recently been implicated in organelle-speciﬁc RNA metabolism [16, 17, 172–175], RNA transport, targeting, localization and compartment-speciﬁc synthesis to localize protein products where desired (or to prevent protein products from accumulating where not desired) have been proposed in a number of circumstances [464, 527, 529–538, 545, 547–555]. RNA sorting in mammalian cells has become a hot topic in cell biology with the recognition that focal accumulations of mRNA with certain organelles, portions of organelles, and cytoplasmic sites exist [16, 17, 477, 480, 527, 529–534, 538–540, 550, 552–562]. In many instances, the identiﬁcation of these localizations was preceded by ultrastructural visualization of ribosomes in these sites. These sightings, coupled with determinations that mRNA and, in some cases, their speciﬁc proteins are colocalized in areas of potential need, have suggested that speciﬁc localization of mRNAs may function to concentrate proteins locally where their function occurs, and/or to prevent product localization where they are of no use. Few of all these reports have implicated mRNA accumulation in classic storage granules of secretory cells and in lipid bodies generally. Some reports do suggest such a relationship, one in granular vesicles of the hypothalamus of lactating rats [560] and our communications regarding HMC granules and

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

309

lipid bodies [16, 17]. Polar granules in germ cells of Caenorhabditis elegans are said to contain embryonic RNAs that are poly(A)-positive [563], and poly(A)binding proteins have been localized in the same polar granules [564], as well as in the dual function, i.e., secretory-lysosome, granules of cytolytic lymphocytes [565]. Similar parallels to our studies [16, 17, 172–175] exist in the realization that site-speciﬁc synthesis plays a role in neurobiology [454, 546, 556, 560, 566–569]. Initially, morphologic studies recorded preferential localization of polyribosomes under the base of dendritic spines of neuronal cells and at synaptic sites [556], despite the fact that most organelles responsible for protein synthesis are located in neuronal cell bodies [570]. Specialized techniques to localize RNA precursors and poly(A)-positive mRNA followed and provided further evidence of site-speciﬁc protein synthesis at synapses [454, 566, 569]. Protein-speciﬁc mRNAs were then localized at synaptic sites [560, 567, 568], and direct visualization of RNA motility in living neurons suggested the existence of a cellular trafﬁcking system for RNA targeting [546]. The accumulating evidence of synthetic machinery in secretory-storage granules and lipid bodies suggests a more comprehensive role for these organelles in secretion-synthetic processes than previously recognized. In mast cells, this role may also implicate them in the regulation of mRNA levels to rid cells of excessive or redundant mRNAs concomitant with secretion of granules in toto by AND or their contents in part by PMD. This mechanism may complement, or function as an alternative to mRNA degradation (to lower mRNA levels) as needed. Thus, mRNA sorting to secretory granules in mast cells may serve to regulate protein supplies spatially and temporally.

Fig. 182. Immunogold studies with speciﬁc human autoimmune sera show goldlabeled LBs in 6 h control cultured HMCs. Human autoimmune sera with speciﬁcity for RNP (A) label the large lipid body (L), the secretory granule (G) beside it and the nucleus (N) beneath it. When normal human sera (NHS) was substituted for the speciﬁc autoimmune sera (B), only three gold particles bound to the lipid body (L). A !49,500. B !53,000. [From 17, with permission.]

Ultrastructure of Mast Cells and Basophils

310

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

311

Fig. 183. Immunogold stain to detect ribosomal P antigen (A) and control (B) in human mast cells. Gold particles label scroll granules and clusters of perigranular ribosomes (A); the control (B) (omission of the speciﬁc primary antibody) has no gold bound to granules or ribosome clusters nearby. A !78,000. B !76,000. [From 173, with permission.]

Ultrastructure of Mast Cells and Basophils

312

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

313

Ultrastructure of Mast Cells and Basophils

314

Fig. 185. Controls for the immunogold procedures with speciﬁc autoimmune human sera. A No label is present in a human mast cell when normal human sera is substituted for the speciﬁc primary autoimmune sera. B Mitochondria in a type II alveolar pneumocyte present in the same preparation are labeled (but human mast cell granules were not) when a speciﬁc autoimmune human antibody for mitochondria is substituted for the primary antibody. C The secretory granules (lamellar bodies, arrowheads) of a type II alveolar pneumocyte are not labeled with human autoantibody serum speciﬁc for RNP. A !63,600. B !69,200. C !49,500. [From 173, with permission.]

Fig. 184. Immunogold studies with speciﬁc human autoimmune sera for P show extensively gold labeled lipid bodies (A, B); omission of the speciﬁc antisera abrogates lipid body label (C). A !66,000. B !58,000. C !59,000. [From 17, with permission.]

Mast Cell Secretory Granules and Lipid Bodies Contain the Necessary Machinery Important for the in situ Synthesis of Proteins

315

Chapter 11

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Concluding Statement

In 1991, I reviewed the ultrastructural work concerning basophils and mast cells that I had completed in the previous 20 years (1970–1990) [7]. At that time several large themes ran through these studies, many of which continue to this day. These themes included identiﬁcation of basophils (and mast cells) in various circumstances and species. This resulted in the identiﬁcation and description of a new form of cell-mediated immunity, which we named cutaneous basophil hypersensitivity (CBH) [4]. In the present review essentially covering works spanning about 10 years (1991–2001) new identity issues are covered in Chapter 2. Altogether, these ultrastructural studies identiﬁed new sources of basophils and mast cells for the future work of our laboratory, as well as in those of numerous investigators. Another large theme presented earlier in detail [7] was our ultrastructural identiﬁcation of a new form of secretion, which we named piecemeal degranulation (PMD) [6]. In that presentation we established the capacity for vesicular transport of extracellular tracers to reach secretory granules, as part of an effort to implicate vesicular transport as the mechanism for effecting PMD. A large part of the effort of my laboratory since that review has been directed toward proof of principle of this concept [see Chapter 7]. The third large theme covered in the 1991 review [7] involved recognition that mast cells and basophils contained cytoplasmic lipid bodies, in addition to secretory granules. Our early studies implicated lipid bodies in new function(s) as key players in the arachidonic acid cascade(s) in which basophils and mast cells participate [5].

Newer studies which solidify these role(s) for lipid bodies are covered here in Chapter 3. The fourth large theme reviewed in 1991 [7] concerned our ultrastructural observations that basophils and mast cells have the capacity to survive and recover from either explosive degranulation, characterized by AND, or from PMD. An update on these processes using speciﬁc labels for granule contents such as histamine [251] and heparin [216] is presented in Chapter 9. Recognition that these recovery events do occur, combined with our ultrastructural recognition that ribosomes populated granule (and lipid body) domains, informed our pursuit of evidence for site-speciﬁc synthetic machinery [see Chapter 10]. Finally, a topic not covered at all in the 1991 review [7] blossomed between 1991 and 2001, and is presented in detail in Chapter 8. In this body of work our recognition and deﬁnition of a new endothelial cell organelle, which we termed the vesicular-vacuolar organelle (VVO), and its capacity to be altered functionally by potent permeabilizers (of basophil, mast cell, and tumor cell origin) has enhanced our understanding of how mast cell or basophil secretion causes vascular leakage in vivo in disease [13–15]. Organelles of interest that have been the focus of many of our ultrastructural studies reviewed here include granules, lipid bodies, and VVOs. We have entertained (and in some cases answered) many questions about the function(s) of these organelles. For secretory granules the question of how they empty is answered in detail by our description of a degranulation model that is a continuum from PMD to AND. If they empty in place and retain their containers, as deﬁnes PMD, do these empty containers reﬁll? Substructural labeling of two granule constituents, histamine and heparin, show that they do reﬁll [Chapter 9]. If these containers reﬁll do they have the capacity to act in a synthetic capacity? Our identiﬁcation of ribosomes and RNA metabolic machinery in and about granules and lipid bodies [16, 17] supports a vast published literature documenting site-speciﬁc synthesis in other circumstances [Chapter 10]. Lipid bodies and secretory granules also contain key cytokines [133–135, 137; Chapter 4], hence may participate in biological reactions by release of such potent agents. Another newly recognized organelle of interest, VVOs in endothelial cells, provided an answer to the regulation of leakage from the vascular compartment to the extravascular space in allergic inﬂammation [Chapter 8]. Studies of these key themes and of organelles of interest in basophils, mast cells and endothelial cells were greatly facilitated by the development and application of new ultrastructural techniques, as well as of older well-documented electron microscopic methods. Chapters 3–6, 8, and 10 describe these methods in detail. Thus, we made extensive use of ultrastructural autoradiography, post-embedding immunogold, pre-embedding immunonanogold and immunoperoxidase, enzyme-afﬁnity-gold, inhibitor-afﬁnity-gold, tracer techniques,

Concluding Statement

317

serial sectioning and computer-assisted three-dimensional reconstructions in pursuit of our aims. Development and application of this myriad of methods has allowed proof of principle for at least three major cell biological mechanisms, which we initially proposed, based on standard electron microscopic preparations and ultrastructural observations. These are as follows: (1) PMD is effected by vesicular transport of granule contents [Chapter 7]; (2) VVOs in endothelial cells are sessile cytoplasmic structures that allow regulated leakage from microvessels [Chapter 8], and (3) machinery for site-speciﬁc synthesis is present in and around secretory granules and lipid bodies of HMCs [Chapter 10].

Ultrastructure of Mast Cells and Basophils

318

U U U U U U U U U U U U U U U U U U U U U U U U U U U

References

1 2 3 4

5

6 7 8

9 10 11

12

Ehrlich P: Beiträge zur Kenntnis der Anilinfärbungen und ihrer Verwendung in der mikroskopischen Technik. Arch Mikrosk Anat 1877;13:263–277. Ehrlich P: Beiträge zur Kenntnis der granulierten Bindegewebszellen und der eosinophilen Leukocyten. Arch Anat Physiol 1879;3:166–169. Dvorak HF: Delayed hypersensitivity; in Zweifach BW, Grant L, McCluskey RT (eds): The Inﬂammatory Process, ed 2. New York, Academic Press, 1974, vol III, pp 291–345. Dvorak HF, Dvorak AM, Simpson BA, Richerson HB, Leskowitz S, Karnovsky MJ: Cutaneous basophil hypersensitivity. II. A light and electron microscopic description. J Exp Med 1970; 132: 558– 582. Dvorak AM, Dvorak HF, Peters SP, Schulman ES, MacGlashan DW Jr, Pyne K, Harvey VS, Galli SJ, Lichtenstein LM: Lipid bodies: Cytoplasmic organelles important to arachidonate metabolism in macrophages and mast cells. J Immunol 1983;131:2965–2976 [republished in J Immunol 1984;132: 1586–1597]. Dvorak HF, Dvorak AM: Basophilic leucocytes: Structure, function and role in disease; in Lichtman MA (ed): Granulocyte and Monocyte Abnormalities. London, Saunders, 1975, vol 4, pp 651–683. Dvorak AM: Basophil and mast cell degranulation and recovery; in Harris JR (ser ed)/Dvorak AM (vol ed): Blood Cell Biochemistry. New York, Plenum Press, 1991, vol 4, pp 1–415. Ishizaka T, Dvorak AM, Conrad DH, Niebyl JR, Marquette JP, Ishizaka K: Morphologic and immunologic characterization of human basophils developed in cultures of cord blood mononuclear cells. J Immunol 1985;134:532–540. Furitsu T, Saito H, Dvorak AM, Schwartz LB, Irani AM, Burdick JF, Ishizaka K, Ishizaka T: Development of human mast cells in vitro. Proc Natl Acad Sci USA 1989;86:10039–10043. Rodewald HR, Dessing M, Dvorak AM, Galli SJ: Identiﬁcation of a committed precursor for the mast cell lineage. Science 1996;271:818–822. Seder RA, Paul WE, Dvorak AM, Sharkis SJ, Kagey-Sobotka A, Niv Y, Finkelman FD, Barbieri SA, Galli SJ, Plaut M: Mouse splenic and bone marrow cell populations that express high-afﬁnity Fc⑀ receptors and produce interleukin-4 are highly enriched in basophils. Proc Natl Acad Sci USA 1991;88: 2835–2839. Dvorak AM, Monahan-Earley RA, Estrella P, Kissell S, Donahue RE: Ultrastructure of monkey peripheral blood basophils stimulated to develop in vivo by recombinant human interleukin-3. Lab Invest 1989;61:677–690.

13

14

15

16

17 18

19

20

21

22 23

24

25

26 27

28

29

30

31

Kohn S, Nagy JA, Dvorak HF, Dvorak AM: Pathways of macromolecular tracer transport across venules and small veins. Structural basis for the hyperpermeability of tumor blood vessels. Lab Invest 1992;67:596–607. Dvorak AM, Kohn S, Morgan ES, Fox P, Nagy JA, Dvorak HF: The vesiculo-vacuolar organelle (VVO): A distinct endothelial cell structure that provides a transcellular pathway for macromolecular extravasation. J Leukoc Biol 1996;59:100–115. Feng D, Nagy JA, Hipp J, Dvorak HF, Dvorak AM: Vesiculo-vacuolar organelles and the regulation of venule permeability to macromolecules by vascular permeability factor, histamine and serotonin. J Exp Med 1996;183:1981–1986. Dvorak AM, Morgan ES, Lichtenstein LM, Weller PF, Schleimer RP: RNA is closely associated with human mast cell secretory granules, suggesting a role(s) for granules in synthetic processes. J Histochem Cytochem 2000;48:1–12. Dvorak AM, Morgan ES, Weller PF: RNA is closely associated with human mast cell lipid bodies. Histol Histopathol 2003;18:943–968. Dvorak AM, Ishizaka T, Galli SJ: Ultrastructure of human basophils developing in vitro. Evidence for the acquisition of peroxidase by basophils and for different effects of human and murine growth factors on human basophil and eosinophil maturation. Lab Invest 1985;53:57–71. Dvorak AM, Saito H, Estrella P, Kissell S, Arai N, Ishizaka T: Ultrastructure of eosinophils and basophils stimulated to develop in human cord blood mononuclear cell cultures containing recombinant human interleukin-5 or interleukin-3. Lab Invest 1989;61:116–132. Dvorak AM, Furitsu T, Kissell-Rainville S, Ishizaka T: Ultrastructural identiﬁcation of human mast cells resembling skin mast cells stimulated to develop in long-term human cord blood mononuclear cells cultured with 3T3 mouse skin ﬁbroblasts. J Leukoc Biol 1992;51:557–569. Dvorak AM, Furitsu T, Ishizaka T: Ultrastructural morphology of human mast cell progenitors in sequential cocultures of cord blood cells and ﬁbroblasts. Int Arch Allergy Immunol 1993; 100: 219– 229. Dvorak AM, Mitsui H, Ishizaka T: Stimulation of partial development of human mast cells by supernatant ﬂuid from mouse ﬁbroblast cultures. Clin Exp Allergy 1994;24:649–659. Dvorak AM, Mitsui H, Ishizaka T: Human and murine recombinant c-kit ligands support the development of human mast cells from umbilical cord blood cells: Ultrastructural identiﬁcation. Int Arch Allergy Immunol 1993;101:247–253. Dvorak AM, Mitsui H, Ishizaka T: Ultrastructural morphology of immature mast cells in sequential suspension cultures of human cord blood cells supplemented with c-kit ligand; distinction from mature basophilic leukocytes undergoing secretion in the same cultures. J Leukoc Biol 1993;54:465–485. Dvorak AM, Ishizaka T, Letourneau L, Albee EA, Mitsui H, Ackerman SJ: Charcot-Leyden crystal protein distribution in basophils and its absence in mast cells that differentiate from human umbilical cord blood precursor cells cultured in murine ﬁbroblast culture supernatants or in recombinant human c-kit ligand. J Histochem Cytochem 1994;42:251–263. Dvorak AM, Estrella P, Mitsui H, Ishizaka T: c-kit ligand induction of immature neutrophils in cultures of human umbilical cord blood. Eur J Cell Biol 1993;62:422–431. Mitsui H, Furitsu T, Dvorak AM, Irani AM, Schwartz LB, Inagaki N, Takei M, Ishizaka K, Zsebo KM, Gillis S, Ishizaka T: Development of human mast cells from umbilical cord blood cells by recombinant human and murine c-kit ligand. Proc Natl Acad Sci USA 1993;90:735–739. Dvorak AM, Furitsu T, Estrella P, Ishizaka T: Human lung-derived mature mast cells cultured alone or with mouse 3T3 ﬁbroblasts maintain an ultrastructural phenotype different from that of human mast cells that develop from human cord blood cells cultured with 3T3 ﬁbroblasts. Am J Pathol 1991; 139:1309–1318. Dvorak AM, Morgan ES: Ultrastructural detection of histamine in human mast cells developing from cord blood cells cultured with human or murine recombinant c-kit ligands. Int Arch Allergy Immunol 1996;111:238–244. Dvorak AM, Seder RA, Paul WE, Kissell-Rainville S, Plaut M, Galli SJ: Ultrastructural characteristics of Fc⑀R-positive basophils in the spleen and bone marrow of mice immunized with goat anti-mouse IgD antibody. Lab Invest 1993;68:708–715. Dvorak AM, Seder RA, Paul WE, Morgan ES, Galli SJ: Effects of interleukin-3 with or without the c-kit ligand, stem cell factor, on the survival and cytoplasmic granule formation of mouse basophils and mast cells in vitro. Am J Pathol 1994;144:160–170.

Ultrastructure of Mast Cells and Basophils

320

32

33 34 35 36

37

38 39 40

41 42

43

44

45 46

47 48

49 50

51 52

53

Verbeek W, Lekstrom-Himes J, Park DJ, Dang PMC, Vuong PT, Kawano S, Babior BM, Xanthopoulos K, Koefﬂer HP: Myeloid transcription factor C/EBP⑀ is involved in the positive regulation of lactoferrin gene expression in neutrophils. Blood 1999;94:3141–3150. Dvorak AM: The mouse basophil, a rare and rarely recognized granulocyte. Blood 2000; 96: 1616– 1617. Kopicky-Burd JA, Kagey-Sobotka A, Peters SP, Dvorak AM, Lennox DW, Lichtenstein LM, Wigley FM: Characterization of human synovial mast cells. J Rheumatol 1988;15:1326–1333. Dvorak AM, Massey W, Warner J, Kissell S, Kagey-Sobotka A, Lichtenstein LM: IgE-mediated anaphylactic degranulation of isolated human skin mast cells. Blood 1991;77:569–578. Dvorak AM, Schulman ES, Peters SP, MacGlashan DW Jr, Newball HH, Schleimer RP, Lichtenstein LM: Immunoglobulin E-mediated degranulation of isolated human lung mast cells. Lab Invest 1985; 53:45–56. Massey WA, Guo CB, Dvorak AM, Hubbard WC, Bhagavan BS, Cohan VL, Warner JA, Kagey-Sobotka A, Lichtenstein LM: Human uterine mast cells. Isolation, puriﬁcation, characterization, ultrastructure, and pharmacology. J Immunol 1991;147:1621–1627. Fox CC, Dvorak AM, Peters SP, Kagey-Sobotka A, Lichtenstein LM: Isolation and characterization of human intestinal mucosal mast cells. J Immunol 1985;135:483–491. Fox CC, Dvorak AM, MacGlashan DW Jr, Lichtenstein LM: Histamine-containing cells in human peritoneal ﬂuid. J Immunol 1984;132:2177–2179. Saito H, Hatake K, Dvorak AM, Leiferman KM, Donnenberg AD, Arai N, Ishizaka K, Ishizaka T: Selective differentiation and proliferation of hematopoietic cells induced by recombinant human interleukins. Proc Natl Acad Sci USA 1988;85:2288–2292. Dvorak AM, Ishizaka T: Human eosinophils in vitro. An ultrastructural morphology primer. Histol Histopathol 1994;9:339–374. Dvorak AM, Ishizaka T: Ultrastructural identiﬁcation of human mast cells arising in in vitro culture systems; in Miyamoto T, Okuda M (eds): Progress in Allergy and Clinical Immunology. Göttingen, Hogrefe & Huber, 1992, vol 2, pp 485–494. Ogawa M, Nakahata T, Leary AG, Sterk AR, Ishizaka K, Ishizaka T: Suspension culture of human mast cells/basophils from umbilical cord blood mononuclear cells. Proc Natl Acad Sci USA 1983;80: 4494–4498. Ishizaka T, Saito H, Furitsu T, Dvorak AM: Growth of human basophils and mast cells in vitro; in Galli SJ, Austen KF (eds): Mast Cell and Basophil Differentiation and Function in Health and Disease. New York, Raven Press, 1989, pp 39–47. Ishizaka T, Mitsui H, Yanagida M, Miura T, Dvorak AM: Development of human mast cells from their progenitors. Curr Opin Immunol 1993;5:937–943. Ishizaka T, Saito H, Hatake K, Dvorak AM, Leiferman KM, Arai N, Ishizaka K: Preferential differentiation of inﬂammatory cells by recombinant human interleukins. Int Arch Allergy Appl Immunol 1989;88:46–49. Nabel G, Galli SJ, Dvorak AM, Dvorak HF, Cantor H: Inducer T lymphocytes synthesize a factor that stimulates proliferation of cloned mast cells. Nature 1981;291:332–334. Galli SJ, Dvorak AM, Marcum JA, Ishizaka T, Nabel G, Der Simonian H, Pyne K, Goldin JM, Rosenberg RD, Cantor H, Dvorak HF: Mast cell clones: A model for the analysis of cellular maturation. J Cell Biol 1982;95:435–444. Dvorak AM, Galli SJ: Antigen-induced, IgE-mediated degranulation of cloned immature mast cells derived from normal mice. Am J Pathol 1987;126:535–545. Galli SJ, Dvorak AM, Dvorak HF: Basophils and mast cells: Morphologic insights into their biology, secretory patterns, and function; in Ishizaka K (ed): Progress in Allergy. Mast Cell Activation and Mediator Release. Basel, Karger, 1984, vol 34, pp 1–141. Dvorak AM, Estrella P, Ishizaka T: Vesicular transport of peroxidase in human eosinophilic myelocytes. Clin Exp Allergy 1994;24:10–18. Dvorak AM: Ultrastructural analysis of anaphylactic and piecemeal degranulation of human mast cells and basophils; in Foreman JC (ed): Immunopharmacology of Mast Cells and Basophils. London, Academic Press, 1993, pp 89–113. Dvorak AM: Morphologic and immunologic characterization of human basophils, 1879 to 1985. Riv Immunol Immunofarmacol 1988;8:50–83.

References

321

54 55

56 57 58

59

60

61 62

63

64

65

66 67

68

69

70 71 72

Dvorak AM: Human mast cells; in Beck F, Hild W, Kriz W, Ortmann R, Pauly JE, Schiebler TH (eds): Advances in Anatomy, Embryology, and Cell Biology. Berlin, Springer, 1989, vol 114, pp 1–107. Dvorak AM: Human mast cells. Ultrastructural observations of in situ, ex vivo, and in vitro sites, sources, and systems; in Kaliner MA, Metcalfe DD (eds): The Mast Cell in Health and Disease. New York, Dekker, 1992, vol 62, pp 1–90. Fung MC, Hapel AJ, Ymer S, Cohen DR, Johnson RM, Campbell HD, Young IG: Molecular cloning of cDNA for murine interleukin-3. Nature 1984;307:233–237. Nabel G, Greenberger JS, Sakakeeny MA, Cantor H: Multiple biologic activities of a cloned inducer T-cell population. Proc Natl Acad Sci USA 1981;78:1157–1161. Yokota T, Lee F, Rennick D, Hall C, Arai N, Mosmann T, Nabel G, Cantor H, Arai K-I: Isolation and characterization of a mouse cDNA clone that expresses mast-cell growth-factor activity in monkey cells. Proc Natl Acad Sci USA 1984;81:1070–1074. Dvorak AM: Morphologic expressions of maturation and function can affect the ability to identify mast cells and basophils in man, guinea pig, and mouse; in Befus AD, Bienenstock J, Denburg JA (eds): Mast Cell Differentiation and Heterogeneity. New York, Raven Press, 1986, pp 95–114. Dvorak AM, Ackerman SJ, Weller PF: Subcellular morphology and biochemistry of eosinophils; in Harris JR (ed): Megakaryocytes, Platelets, Macrophages and Eosinophils. London, Plenum Publishing, 1991, vol 2, pp 237–344. Zucker-Franklin D: Ultrastructural evidence for the common origin of human mast cells and basophils. Blood 1980;56:534–540. Otsuka T, Miyajima A, Brown N, Otsu K, Abrams J, Saeland S, Caux C, de Waal Maleﬁjt R, de Vries J, Meyerson P, Yokota K, Gemmel L, Rennick DM, Lee FD, Arai N, Arai K, Yokota T: Isolation and characterization of an expressible cDNA encoding human IL-3. Induction of IL-3 mRNA in human T cell clones. J Immunol 1988;140:2288–2295. Yokota T, Coffman RL, Hagiwara H, Hagiwara H, Rennick DM, Takebe Y, Yokota K, Gemmell L, Shrader B, Yang G, Meyerson P, Luh J, Hoy P, Pene J, Briere F, Banchereau J, deVries J, Lee FD, Arai N, Arai KI: Isolation and characterization of lymphokine cDNA clones encoding mouse and human IgA-enhancing factor and eosinophil colony-stimulating factor activities: Relationship to interleukin-5. Proc Natl Acad Sci USA 1987;84:7388–7392. Dvorak AM, Furitsu T, Letourneau L, Ishizaka T, Ackerman SJ: Mature eosinophils stimulated to develop in human cord blood mononuclear cell cultures supplemented with recombinant human interleukin-5. I. Piecemeal degranulation of speciﬁc granules and distribution of Charcot-Leyden crystal protein. Am J Pathol 1991;138:69–82. Dvorak AM, Ackerman SJ, Furitsu T, Estrella P, Letourneau L, Ishizaka T: Mature eosinophils stimulated to develop in human cord blood mononuclear cell cultures supplemented with recombinant human interleukin-5. II. Vesicular transport of speciﬁc granule matrix peroxidase, a mechanism for effecting piecemeal degranulation. Am J Pathol 1992;140:795–807. Dvorak AM, Nabel G, Pyne K, Cantor H, Dvorak HF, Galli SJ: Ultrastructural identiﬁcation of the mouse basophil. Blood 1982;59:1279–1285. Dvorak AM, MacGlashan DW Jr, Warner JA, Letourneau L, Morgan ES, Lichtenstein LM, Ackerman SJ: Localization of Charcot-Leyden crystal protein in individual morphologic phenotypes of human basophils stimulated by f-Met peptide. Clin Exp Allergy 1997;27:452–474. Levi-Schaffer F, Austen KF, Caulﬁeld JP, Hein A, Gravallese PM, Stevens RL: Co-culture of human lung-derived mast cells with mouse 3T3 ﬁbroblasts: Morphology and IgE-mediated release of histamine, prostaglandin D2, and leukotrienes. J Immunol 1987;139:494–500. Dvorak AM, McLeod RS, Onderdonk AB, Monahan-Earley RA, Cullen JB, Antonioli DA, Morgan E, Blair JE, Estrella P, Cisneros RL, Cohen Z, Silen W: Human gut mucosal mast cells: Ultrastructural observations and anatomic variation in mast cell-nerve associations in vivo. Int Arch Allergy Immunol 1992;98:158–168. Kawanami O, Ferrans VJ, Fulmer JD, Crystal RG: Ultrastructure of pulmonary mast cells in patients with ﬁbrotic lung disorders. Lab Invest 1979;40:717–734. Colvin RB, Dvorak AM, Dvorak HF: Mast cells in the cortical tubular epithelium and interstitium in human renal disease. Hum Pathol 1974;5:315–326. Omi T, Kawanami O, Honda M, Akamatsu H: Human fetal mast cells under development of the skin and airways (in Japanese). Arerugi [Jpn J Allergol] 1991;40:1407–1414.

Ultrastructure of Mast Cells and Basophils

322

73 74

75

76

77 78 79 80 81

82

83 84 85 86

87

88

89

90 91

92

Ginsburg H: The in vitro differentiation and culture of normal mast cells from the mouse thymus. Ann NY Acad Sci 1963;103:20–39. Ginsburg H, Lagunoff D: The in vitro differentiation of mast cells. Cultures of cells from immunized mouse lymph nodes and thoracic duct lymph on ﬁbroblast monolayers. J Cell Biol 1967; 35: 685– 697. Levi-Schaffer F, Austen KF, Caulﬁeld JP, Hein A, Bloes WF, Stevens RL: Fibroblasts maintain the phenotype and viability of the rat heparin-containing mast cells in vitro. J Immunol 1985;135:3454– 3462. Levi-Schaffer F, Austen KF, Gravallese PM, Stevens RL: Coculture of interleukin-3-dependent mouse mast cells with ﬁbroblasts results in a phenotypic change of the mast cells. Proc Natl Acad Sci USA 1986;83:6485–6488. Dvorak AM, Kissell S: Granule changes of human skin mast cells characteristic of piecemeal degranulation and associated with recovery during wound healing in situ. J Leukoc Biol 1991;49:197–210. Fedorko ME, Hirsch JG: Crystalloid structure in granules of guinea pig basophils and human mast cells. J Cell Biol 1965;26:973–976. Geissler EN, Ryan MA, Housman DE: The dominant-white spotting (W) locus of the mouse encodes the c-kit proto-oncogene. Cell 1988;55:185–192. Chabot B, Stephenson DA, Chapman VM, Besmer P, Bernstein A: The proto-oncogene c-kit encoding a transmembrane tyrosine kinase receptor maps to the mouse W locus. Nature 1988;335:88–89. Besmer P, Murphy JE, George PC, Qiu F, Bergold PJ, Lederman L, Snyder HW Jr, Brodeur D, Zuckerman EE, Hardy WD: A new acute transforming feline retrovirus and relationship of its oncogene v-kit with the protein kinase gene family. Nature 1986;320:415–421. Yarden Y, Kuang WJ, Yang-Feng T, Coussens L, Munemitsu S, Dull TJ, Chen E, Schlessinger J, Francke U, Ullrich A: Human proto-oncogene c-kit: A new cell surface receptor tyrosine kinase for an unidentiﬁed ligand. EMBO J 1987;6:3341–3351. Russell ES: Hereditary anemias of the mouse: A review for geneticists. Adv Genet 1979; 20: 357– 459. Kitamura Y, Go S, Hatanaka K: Decrease of mast cells in W/Wv mice and their increase by bone marrow transplantation. Blood 1978;52:447–452. Kitamura Y, Go S: Decreased production of mast cells in Sl/Sld anemic mice. Blood 1979; 53: 492– 497. Galli SJ, Arizono N, Murakami T, Dvorak AM, Fox JG: Development of large numbers of mast cells at sites of idiopathic chronic dermatitis in genetically mast cell-deﬁcient WBB6F1-W/Wv mice. Blood 1987;69:1661–1666. Dvorak AM, Wiberg L, Monahan-Earley RA, Galli SJ: A simple technique to facilitate the ultrastructural analysis of cells in soft agar culture systems: demonstration of the development in vitro of morphologically mature mast cells and phagocytic macrophages from the bone marrow cells of genetically mast cell-deﬁcient W/Wv or congenic normal mice. Lab Invest 1990;62:774–781. Zsebo KM, Williams DA, Geissler EN, Broudy VC, Martin FH, Atkins HL, Hsu RY, Birkett NC, Okino KH, Murdock DC, Jacobsen FW, Langley KE, Smith KA, Takeishi T, Cattanach BM, Galli SJ, Suggs SV: Stem cell factor is encoded at the Sl locus of the mouse and is the ligand for the c-kit tyrosine kinase receptor. Cell 1990;63:213–224. Zsebo KM, Wypych J, McNiece IK, Lu HS, Smith KA, Karkare SB, Sachdev RK, Yuschenkoff VN, Birkett NC, Williams LR, Satyagal VN, Tung W, Bosselman RA, Mendiaz EA, Langley KE: Identiﬁcation, puriﬁcation, and biological characterization of hematopoietic stem cell factor from Buffalo rat liver-conditioned medium. Cell 1990;63:195–201. Flanagan JG, Leder P: The kit ligand: A cell surface molecule altered in steel mutant ﬁbroblasts. Cell 1990;63:185–194. Huang E, Nocka K, Beier DR, Chu TY, Buck J, Lahm HW, Wellner D, Leder P, Besmer P: The hematopoietic growth factor KL is encoded by the Sl locus and is the ligand of the c-kit receptor, the gene product of the W locus. Cell 1990;63:225–233. Williams DE, Eisenman J, Baird A, Rauch C, Van Ness K, March CJ, Park LS, Martin U, Mochizuki DY, Boswell HS, Burgess GS, Cosman D, Lyman SD: Identiﬁcation of a ligand for the c-kit protooncogene. Cell 1990;63:167–174.

References

323

93

94 95

96

97 98

99

100 101 102

103 104 105 106 107

108

109

110

111 112

Martin FH, Suggs SV, Langley KE, Lu HS, Ting J, Okino KH, Morris CF, McNiece IK, Jacobsen FW, Mendiaz EA, Birkett NC, Smith KA, Johnson MJ, Parker VP, Flores JC, Patel AC, Fisher EF, Erjavec HO, Herrera CJ, Wypych J, Sachdev RK, Pope JA, Leslie I, Wen D, Lin CH, Cupples RL, Zsebo KM: Primary structure and functional expression of rat and human stem cell factor DNAs. Cell 1990;63: 203–211. Ulich TR, del Castillo J, Yi ES, Yin S, McNiece I, Yung YP, Zsebo KM: Hematologic effects of stem cell factor in vivo and in vitro in rodents. Blood 1991;78:645–650. Tsai M, Shih LS, Newlands GFJ, Takeishi T, Langley KE, Zsebo KM, Miller HRP, Geissler EN, Galli SJ: The rat c-kit ligand, stem cell factor, induces the development of connective tissue-type and mucosal mast cells in vivo. Analysis by anatomical distribution, histochemistry, and protease phenotype. J Exp Med 1991;174:125–131. Tsai M, Takeishi T, Thompson H, Langley KE, Zsebo KM, Metcalfe DD, Geissler EN, Galli SJ: Induction of mast cell proliferation, maturation, and heparin synthesis by the rat c-kit ligand, stem cell factor. Proc Natl Acad Sci USA 1991;88:6382–6386. Kitamura Y, Shimada M, Hatanaka K, Miyano Y: Development of mast cells from grafted bone marrow cells in irradiated mice. Nature 1977;268:442–443. Kasugai T, Tei H, Okada M, Hirota S, Morimoto M, Yamada M, Nakama A, Arizono N, Kitamura Y: Infection with Nippostrongylus brasiliensis induces invasion of mast cell precursors from peripheral blood to small intestine. Blood 1995;85:1334–1340. Huff TF, Lantz CS, Ryan JJ, Leftwich JA: Mast cell-committed progenitors; in Kitamura Y, Yamamoto S, Galli SJ, Geaves MW (eds): Biological and Molecular Aspects of Mast Cell and Basophil Differentiation and Function. New York, Raven Press, 1995, pp 105–117. Rodewald HR: Pathways from hematopoietic stem cells to thymocytes. Curr Opin Immunol 1995;7: 176–187. Dvorak AM, Dvorak HF, Galli SJ: Ultrastructural criteria for identiﬁcation of mast cells and basophils in humans, guinea pigs, and mice. Am Rev Respir Dis 1983;128:S49–S52. Dvorak AM, Galli SJ, Marcum JA, Nabel G, Der Simonian H, Goldin J, Monahan RA, Pyne K, Cantor H, Rosenberg RD, Dvorak HF: Cloned mouse cells with natural killer function and cloned suppressor T cells express ultrastructural and biochemical features not shared by cloned inducer T cells. J Exp Med 1983;157:843–861. Dvorak AM: Ultrastructural evidence for vesicular transport in basophil secretion. Fundam Clin Immunol 1995;3:17–28. Dvorak AM, MacGlashan DW Jr, Morgan ES, Lichtenstein LM: Vesicular transport of histamine in stimulated human basophils. Blood 1996;88:4090–4101. Dvorak AM: Cell biology of the basophil. Int Rev Cytol 1998;180:87–236. Dvorak AM: A role for vesicles in human basophil secretion. Cell Tissue Res 1998;293:1–22. Ben-Sasson SZ, Le Gros G, Conrad DH, Finkelman FD, Paul WE: Cross-linking Fc receptors stimulate splenic non-B, non-T cells to secrete interleukin-4 and other lymphokines. Proc Natl Acad Sci USA 1990;87:1421–1425. Hammel I, Dvorak AM, Peters SP, Schulman ES, Dvorak HF, Lichtenstein LM, Galli SJ: Differences in the volume distributions of human lung mast cell granules and lipid bodies: Evidence that the size of these organelles is regulated by distinct mechanisms. J Cell Biol 1985;100:1488–1492. Dvorak AM, Hammel I, Schulman ES, Peters SP, MacGlashan DW Jr, Schleimer RP, Newball HH, Pyne K, Dvorak HF, Lichtenstein LM, Galli SJ: Differences in the behavior of cytoplasmic granules and lipid bodies during human lung mast cell degranulation. J Cell Biol 1984;99:1678–1687. Dvorak AM, Morgan E, Schleimer RP, Ryeom SW, Lichtenstein LM, Weller PF: Ultrastructural immunogold localization of prostaglandin endoperoxide synthase (cyclooxygenase) to non-membranebound cytoplasmic lipid bodies in human lung mast cells, alveolar macrophages, type II pneumocytes and neutrophils. J Histochem Cytochem 1992;40:759–769. Fawcett DW: Cytoplasmic inclusions – Lipid; in Fawcett DW (ed): The Cell, ed 2. Philadelphia, Saunders, 1981, pp 655–667. Dvorak AM: Biochemical contents of granules and lipid bodies – two distinctive organelles found in basophils and mast cells; in Harris JR (ser ed): Blood Cell Biochemistry. Dvorak AM (vol ed): Basophil and Mast Cell Degranulation and Recovery. New York, Plenum Press, 1991, vol 4, pp 27–65.

Ultrastructure of Mast Cells and Basophils

324

113

114

115 116 117

118

119 120

121 122

123

124 125

126

127

128 129 130 131

Galli SJ, Dvorak AM, Peters SP, Schulman ES, MacGlashan DW Jr, Isomura T, Pyne K, Harvey VS, Hammel I, Lichtenstein LM, Dvorak HF: Lipid bodies – widely distributed cytoplasmic structures that represent preferential nonmembrane repositories of exogenous [3H]-arachidonic acid incorporated by mast cells, macrophages, and other cell types; in Bailey JM (ed): Prostaglandins, Leukotrienes, and Lipoxins. Biochemistry, Mechanism of Action, and Clinical Applications. New York, Plenum Press, 1985, pp 221–239. Weller PF, Ryeom SW, Dvorak AM: Lipid bodies: Structurally distinct, nonmembranous intracellular sites of eicosanoid formation; in Bailey JM (ed): Prostaglandins, Leukotrienes, Lipoxins and PAF. New York, Plenum Press, 1991, pp 353–362. Dvorak AM, Monahan-Earley RA: Diagnostic ultrastructural pathology. I. Boca Raton, CRC Press, 1991, pp 1–250. Weller PF, Ackerman SJ, Nicholson-Weller A, Dvorak AM: Cytoplasmic lipid bodies of human neutrophilic leukocytes. Am J Pathol 1989;135:947–959. Weller PF, Ryeom SW, Picard ST, Ackerman SJ, Dvorak AM: Cytoplasmic lipid bodies of neutrophils: Formation induced by cis-unsaturated fatty acids and mediated by protein kinase C. J Cell Biol 1991;113:137–146. Dvorak AM, Hammond ME, Morgan ES, Dvorak HF: Ultrastructural studies of macrophages: In vitro removal of cell coat with macrophage inhibition factor-containing lymphocyte culture supernatants; chloroform extraction, phospholipase digestion, and autoradiographic studies. J Reticuloendothel Soc 1980;27:119–142. Weller PF, Dvorak AM: Arachidonic acid incorporation by cytoplasmic lipid bodies of human eosinophils. Blood 1985;65:1269–1274. Peters SP, MacGlashan DW Jr, Schulman ES, Schleimer RP, Hayes EC, Rokach J, Adkinson NF Jr, Lichtenstein LM: Arachidonic acid metabolism in puriﬁed human lung mast cells. J Immunol 1984; 132:1972–1979. Scott WA, Zrike JM, Hamill AL, Kempe J, Cohn ZA: Regulation of arachidonic acid metabolites in macrophages. J Exp Med 1980;152:324–325. Dvorak AM, Schleimer RP, Dvorak HF, Lichtenstein LM, Weller PF: Human lung mast cell and alveolar macrophage cytoplasmic lipid bodies contain arachidonic acid and prostaglandin endoperoxide synthase (cyclooxygenase), the substrate and enzyme necessary for prostaglandin production. Int Arch Allergy Immunol 1992;99:208–217. Weller PF, Monahan-Earley RA, Dvorak HF, Dvorak AM: Cytoplasmic lipid bodies of human eosinophils. Subcellular isolation and analysis of arachidonate incorporation. Am J Pathol 1991; 138: 141–148. Monahan RA, Dvorak HF, Dvorak AM: Ultrastructural localization of nonspeciﬁc esterase activity in guinea pig and human monocytes, macrophages, and lymphocytes. Blood 1981;58:1089–1099. Dvorak AM, Weller PF, Monahan-Earley RA, Letourneau L, Ackerman SJ: Ultrastructural localization of Charcot-Leyden crystal protein (lysophospholipase) and peroxidase in macrophages, eosinophils and extracellular matrix of the skin in the hypereosinophilic syndrome. Lab Invest 1990; 62: 590–607. Dvorak AM, Monahan-Earley RA, Dvorak HF, Galli SJ: Ultrastructural cytochemical and autoradiographic demonstration of nonspeciﬁc esterase(s) in guinea pig basophils. J Histochem Cytochem 1987;35:351–360. Monahan-Earley RA, Isomura T, Garcia RI, Galli SJ, Dvorak HF, Dvorak AM: Nonspeciﬁc esterase activity expressed in Weibel-Palade bodies of cloned guinea pig aortic endothelial cells [ﬁg. 3–8 republished in J Histochem Cytochem 1987;35:1171–1176]. J Histochem Cytochem 1987;35:531–539. DeWitt DL: Prostaglandin endoperoxide synthase: Regulation of enzyme expression. Biochim Biophys Acta 1991;1083:121–134. Harvima RJ, Schwartz LB: Mast cell-derived mediators; in Foreman JC (ed): Immunopharmacology of Mast Cells and Basophils. London, Academic Press, 1993, pp 115–138. Okayama Y, Bradding P, Tunon-de-Lara JM, Holgate ST, Church MK: Cytokine production by human mast cells. Chem Immunol 1995;61:114–134. Galli SJ, Gordon JR, Wershil BK: Cytokine production by mast cells and basophils. Curr Opin Immunol 1991;3:865–872.

References

325

132 133 134

135

136 137 138

139 140 141 142 143

144

145

146

147 148

149

150 151

152

Triggiani M, Oriente A, de Crescenzo G, Marone G: Metabolism of lipid mediators in human basophils and mast cells. Chem Immunol 1995;61:135–147. Beil WJ, Weller PF, Peppercorn MA, Galli SJ, Dvorak AM: Ultrastructural immunogold localization of subcellular sites of TNF-␣ in colonic Crohn’s disease. J Leukoc Biol 1995;58:284–298. Beil WJ, Login GR, Galli SJ, Dvorak AM: Ultrastructural immunogold localization of tumor necrosis factor-␣ to the cytoplasmic granules of rat peritoneal mast cells with rapid microwave ﬁxation. J Allergy Clin Immunol 1994;94:531–536. Beil WJ, Login GR, Aoki M, Lunardi LO, Morgan ES, Galli SJ, Dvorak AM: Tumor necrosis factor-␣ immunoreactivity of rat peritoneal mast cell granules decreases during early secretion induced by compound 48/80: An ultrastructural immunogold morphometric analysis. Int Arch Allergy Immunol 1996;109:383–389. Dvorak AM, Beil WJ: Recent studies on the role of tumor necrosis factor-␣ in small intestinal diseases. Curr Opin Gastroenterol 1996;12:141–146. Dvorak AM, Morgan ES, Weller PF: Ultrastructural immunolocalization of basic ﬁbroblast growth factor in lipid bodies and secretory granules of human mast cells. Histochem J 2001;33:397–402. Breese EJ, Michie CA, Nicholls SW, Murch SH, Williams CB, Domizio P, Walker-Smith JA, MacDonald TT: Tumor necrosis factor-␣-producing cells in the intestinal mucosa of children with inﬂammatory bowel disease. Gastroenterology 1994;106:1455–1466. Casellas F, Papo M, Guarner F, Antolin M, Armengol JR, Malagelada JR: Intraluminal colonic release of immunoreactive tumour necrosis factor in chronic ulcerative colitis. Clin Sci 1994;87:453–458. Carswell EA, Old LJ, Kassel RL, Green S, Fiore N, Williamson B: An endotoxin-induced serum factor that causes necrosis of tumors. Proc Natl Acad Sci USA 1975;72:3666–3670. Dvorak AM: Similarities in the ultrastructural morphology and developmental and secretory mechanisms of human basophils and eosinophils. J Allergy Clin Immunol 1994;94:1103–1134. Weller PF, Dvorak AM: Lipid bodies: Intracellular sites for eicosanoid formation. J Allergy Clin Immunol 1994;94:1151–1156. Beil WJ, Weller PF, Tzizik DM, Galli SJ, Dvorak AM: Ultrastructural immunogold localization of tumor necrosis factor-␣ to the matrix compartment of eosinophil secondary granules in patients with idiopathic hypereosinophilic syndrome. J Histochem Cytochem 1993;41:1611–1615. Costa JJ, Matossian K, Resnick MB, Beil WJ, Wong DTW, Gordon JR, Dvorak AM, Weller PF, Galli SJ: Human eosinophils can express the cytokines tumor necrosis factor-␣ and macrophage inﬂammatory protein-1␣. J Clin Invest 1993;91:2673–2684. Finotto S, Ohno I, Marshall JS, Gauldie J, Denburg JA, Dolovich J, Clark DA, Jordana M: TNF-␣ production by eosinophils in upper airways inﬂammation (nasal polyposis). J Immunol 1994; 153: 2278–2289. Tan X, Hsueh W, Gonzalez-Crussi F: Cellular localization of tumor necrosis factor (TNF)-␣ transcripts in normal bowel and in necrotizing enterocolitis. TNF gene expression by Paneth cells, intestinal eosinophils, and macrophages. Am J Pathol 1993;142:1858–1865. Cappello M, Keshav S, Prince C, Jewell DP, Gordon S: Detection of mRNAs for macrophage products in inﬂammatory bowel disease by in situ hybridization. Gut 1992;33:1214–1219. Rubin BY, Anderson SL, Sullivan SA, Williamson BD, Carswell EA, Old LJ: Nonhematopoietic cells selected for resistance to tumor necrosis factor produce tumor necrosis factor. J Exp Med 1986;164: 1350–1355. Woywodt A, Neustock P, Kruse A, Schwarting K, Ludwig D, Stange EF, Kirchner H: Cytokine expression in intestinal mucosal biopsies. In situ hybridization of the mRNA for interleukin-1␤, interleukin-6 and tumour necrosis factor-␣ in inﬂammatory bowel disease. Eur Cytokine Netw 1994;5:387–395. Donowitz M: Arachidonic acid metabolites and their role in inﬂammatory bowel disease. An update requiring addition of a pathway. Gastroenterology 1985;88:580–587. Dvorak AM, Letourneau L, Login GR, Weller PF, Ackerman SJ: Ultrastructural localization of the Charcot-Leyden crystal protein (lysophospholipase) to a distinct crystalloid-free granule population in mature human eosinophils. Blood 1988;72:150–158. Login GR, Galli SJ, Morgan E, Arizono N, Schwartz LB, Dvorak AM: Rapid microwave ﬁxation of rat mast cells. I. Localization of granule chymase with an ultrastructural postembedding immunogold technique. Lab Invest 1987;57:592–599.

Ultrastructure of Mast Cells and Basophils

326

153

154

155 156

157

158

159

160 161 162

163

164 165 166 167

168 169

170

171

172

Login GR, Galli SJ, Dvorak AM: Immunocytochemical localization of histamine in secretory granules of rat peritoneal mast cells with conventional or rapid microwave ﬁxation and an ultrastructural post-embedding immunogold technique. J Histochem Cytochem 1992;40:1247–1256. Dvorak AM, Morgan ES, Schleimer RP, Lichtenstein LM: Diamine oxidase-gold labels histamine in human mast-cell granules: A new enzyme-afﬁnity ultrastructural method. J Histochem Cytochem 1993;41:787–800. Dvorak AM, Morgan ES: Ribonuclease-gold labels heparin in human mast cell granules: New use for an ultrastructural enzyme afﬁnity technique. J Histochem Cytochem 1998;46:695–706. Qu Z, Huang X, Ahmadi P, Stenberg P, Liebler JM, Le AC, Planck SR, Rosenbaum JT: Synthesis of basic ﬁbroblast growth factor by murine mast cells. Regulation by transforming growth factor-␤, tumor necrosis factor-␣, and stem cell factor. Int Arch Allergy Immunol 1998;115:47–54. Qu Z, Kayton RJ, Ahmadi P, Liebler JM, Powers MR, Planck SR, Rosenbaum JT: Ultrastructural immunolocalization of basic ﬁbroblast growth factor in mast cell secretory granules: Morphological evidence for bFGF release through degranulation. J Histochem Cytochem 1998;46:1119–1128. Qu Z, Liebler JM, Powers MR, Galey T, Ahmadi P, Huang XN, Ansel JC, Butterﬁeld JH, Planck SR, Rosenbaum JT: Mast cells are a major source of basic ﬁbroblast growth factor in chronic inﬂammation and cutaneous hemangioma. Am J Pathol 1995;147:564–573. Qu Z, Huang XN, Ahmadi P, Andresevic J, Planck SR, Hart CE, Rosenbaum JT: Expression of basic ﬁbroblast growth factor in synovial tissue from patients with rheumatoid arthritis and degenerative joint disease. Lab Invest 1995;73:339–346. Reed JA, Albino AP, McNutt NS: Human cutaneous mast cells express basic ﬁbroblast growth factor. Lab Invest 1995;72:215–222. Inoue Y, King TE Jr, Tinkle SS, Dockstader K, Newman LS: Human mast cell basic ﬁbroblast growth factor in pulmonary ﬁbrotic disorders. Am J Pathol 1996;149:2037–2054. Mignatti P, Morimoto T, Rifkin DB: Basic ﬁbroblast growth factor, a protein devoid of secretory signal sequence, is released by cells via a pathway independent of the endoplasmic reticulum-Golgi complex. J Cell Physiol 1992;151:81–93. Schweigerer L, Neufeld G, Friedman J, Abraham JA, Fiddes JC, Gospodarowicz D: Capillary endothelial cells express basic ﬁbroblast growth factor, a mitogen that promotes their own growth. Nature 1987;325:257–259. Renko M, Quarto N, Morimoto T, Rifkin DB: Nuclear and cytoplasmic localization of different basic ﬁbroblast growth factor species. J Cell Physiol 1990;144:108–114. Quarto N, Finger FP, Rifkin DB: The NH2-terminal extension of high molecular weight bFGF is a nuclear targeting signal. J Cell Physiol 1991;147:311–318. Maher PA: Nuclear translocation of ﬁbroblast growth factor (FGF) receptors in response to FGF-2. J Cell Biol 1996;134:529–536. Dvorak AM, Schleimer RP, Schulman ES, Lichtenstein LM: Human mast cells use conservation and condensation mechanisms during recovery from degranulation. In vitro studies with mast cells puriﬁed from human lungs. Lab Invest 1986;54:663–678. Levi-Schaffer F: Mast cell/ﬁbroblast interactions in health and disease. Chem Immunol 1995; 61: 161–185. Dvorak AM: Basophils and mast cells: piecemeal degranulation in situ and ex vivo: A possible mechanism for cytokine-induced function in disease; in Coffey RG (ed): Granulocyte Responses to Cytokines. Basic and Clinical Research. New York, Dekker, 1992, pp 169–271. Abraham JA, Mergia A, Whang JL, Tumolo A, Friedman J, Hjerrild KA, Gospodarowicz D, Fiddes JC: Nucleotide sequence of a bovine clone encoding the angiogenic protein, basic ﬁbroblast growth factor. Science 1986;233:545–548. Abraham JA, Whang JL, Tumolo A, Mergia A, Friedman J, Gospodarowicz D, Fiddes JC: Human basic ﬁbroblast growth factor: Nucleotide sequence and genomic organization. EMBO J 1986;5:2523– 2528. Dvorak AM, Morgan ES: Ultrastructural cytochemical, immunocytochemical and in situ hybridization methods with polyuridine probes detect mRNA in human mast cell granules. Histochem J 2000; 32:423–438.

References

327

173

174 175

176 177 178 179 180

181 182 183 184 185

186 187 188 189 190

191 192 193

194 195 196 197

198

Dvorak AM, Morgan ES: Ultrastructural immunogold cytochemistry with autoimmune human sera and an antibody to uridine implicate human mast cell granules in RNA biology. Histochem J 2000; 32:685–696. Dvorak AM, Morgan ES: Ribosomes and secretory granules in human mast cells: Close associations demonstrated by staining with a chelating agent. Immunol Rev 2001;179:94–101. Dvorak AM, Morgan ES, Lichtenstein LM, Schleimer R: Ultrastructural autoradiographic analysis of RNA in isolated human lung mast cells during secretion and recovery from secretion. Int Arch Allergy Immunol 2000;122:124–136. Benditt EP, Arase M: An enzyme in mast cells with properties like chymotrypsin. J Exp Med 1959; 110:451–460. Lagunoff D, Benditt EP: Proteolytic enzymes of mast cells. Ann NY Acad Sci 1963;103:185–198. Lagunoff D, Pritzl P: Characterization of rat mast cell granule proteins. Arch Biochem Biophys 1976; 173:554–563. Schwartz LB, Austen KF: Enzymes of the mast cell granule. J Invest Dermatol 1980;74:349–353. Schwartz LB, Riedel C, Caulﬁeld JP, Wasserman SI, Austen KF: Cell association of complexes of chymase, heparin proteoglycan, and protein after degranulation by rat mast cells. J Immunol 1981;126: 2071–2078. Woodbury RG, Everitt M, Sanada Y, Katunuma N, Lagunoff D, Neurath H: A major serine protease in rat skeletal muscle: Evidence for its mast cell origin. Proc Natl Acad Sci USA 1978;75:5311–5313. Yurt R, Austen KF: Preparative puriﬁcation of the rat mast cell chymase: Characterization and interaction with granule components. J Exp Med 1977;146:1405–1419. Gomori G: Chloroacyl esters as histochemical substrates. J Histochem Cytochem 1953;1:469–470. Login GR, Dvorak AM: Microwave energy ﬁxation for electron microscopy. Am J Pathol 1985;120: 230–243. Login GR, Aoki M, Yamakawa M, Lunardi LO, Digenis EC, Tanda N, Schwartz LB, Dvorak AM: Immunocytochemical localization of chymase to cytoplasmic vesicles after rat peritoneal mast cell stimulation by compound 48/80. J Histochem Cytochem 1997;45:1379–1391. Charcot JM, Robin C: Observation de leucocythémie. Mém Soc Biol (Paris) 1853;5:44. Leyden E: Zur Kenntniss des Bronchial-Asthma. Arch Pathol Anat 1872;54:324–352. Weller PF: The immunobiology of eosinophils. N Engl J Med 1991;324:1110–1118. Ackerman SJ, Loegering DA, Gleich GJ: The human eosinophil Charcot-Leyden crystal protein: Biochemical characteristics and measurement by radioimmunoassay. J Immunol 1980;125:2118–2126. Gleich GJ, Loegering DA, Mann KG, Maldonado JE: Comparative properties of the Charcot-Leyden crystal protein and the major basic protein from human eosinophils. J Clin Invest 1976;57:633– 640. Weller PF, Bach D, Austen KF: Human eosinophil lysophospholipase: The sole protein component of Charcot-Leyden crystals. J Immunol 1982;128:1346–1349. Weller PF, Bach DS, Austen KF: Biochemical characterization of human eosinophil Charcot-Leyden crystal protein (lysophospholipase). J Biol Chem 1984;259:15100–15105. Ackerman SJ, Corrette SE, Rosenberg HF, Bennett JC, Mastrianni DM, Nicholson-Weller A, Weller PF, Chin DT, Tenen DG: Molecular cloning and characterization of human eosinophil Charcot-Leyden crystal protein (lysophospholipase): Similarities to IgE-binding proteins and the S-type animal lectin superfamily. J Immunol 1993;150:456–468. Ackerman SJ, Weil GJ, Gleich GJ: Formation of Charcot-Leyden crystals by human basophils. J Exp Med 1982;155:1597–1609. Archer GT, Blackwood A: Formation of Charcot-Leyden crystals in human eosinophils and basophils and study of the composition of isolated crystals. J Exp Med 1965;122:173–180. Bessis M, Tabuis J: Formation of crystals in basophelic leukocytes (in French). C R Seances Soc Biol Fil 1955;149:873–875. Dvorak AM, Ackerman SJ: Ultrastructural localization of the Charcot-Leyden crystal protein (lysophospholipase) to granules and intragranular crystals in mature human basophils. Lab Invest 1989; 60:557–567. Dvorak AM, Letourneau L, Weller PF, Ackerman SJ: Ultrastructural localization of Charcot-Leyden crystal protein (lysophospholipase) to intracytoplasmic crystals in tumor cells of primary solid and papillary neoplasm of the pancreas. Lab Invest 1990;62:608–615.

Ultrastructure of Mast Cells and Basophils

328

199

200

201

202

203

204

205

206 207 208 209 210 211

212

213

214

215 216

217 218 219 220

Dvorak AM, Ackerman SJ, Letourneau L, Morgan ES, Lichtenstein LM, MacGlashan DW Jr: Vesicular transport of Charcot-Leyden crystal protein in tumor-promoting phorbol diester-stimulated human basophils. Lab Invest 1996;74:967–974. Dvorak AM, MacGlashan DW Jr, Warner JA, Letourneau L, Morgan ES, Lichtenstein LM, Ackerman SJ: Vesicular transport of Charcot-Leyden crystal protein in f-Met peptide-stimulated human basophils. Int Arch Allergy Immunol 1997;113:465–477. Dvorak AM: Human basophil recovery from secretion. A review emphasizing the distribution of Charcot-Leyden crystal protein in cells stained with the electron-dense tracer, cationized ferritin. Histol Histopathol 1996;11:711–728. Dvorak AM, Furitsu T, Estrella P, Letourneau L, Ishizaka T, Ackerman SJ: Ultrastructural localization of major basic protein in the human eosinophil lineage in vitro. J Histochem Cytochem 1994;42: 1443–1451. Brennick JB, Dvorak AM, Ackerman SJ: Ultrastructural immunogold localization and radioimmunoassay demonstrate endocytosis of eosinophil Charcot-Leyden crystal protein (lysophospholipase) by isolated, puriﬁed peripheral blood monocytes (abstract). J Cell Biol 1991;115:4779. Zhou ZQ, Tenen DG, Dvorak AM, Ackerman SJ: The gene for human eosinophil Charcot-Leyden crystal protein directs expression of lysophospholipase activity and spontaneous crystallization in transiently transfected COS cells. J Leukoc Biol 1992;52:588–595. Leiferman KM, Gleich GJ, Kephart GM, Haugen HS, Hisamatsu KI, Proud D, Lichtenstein LM, Ackerman SJ: Differences between basophils and mast cells: Failure to detect Charcot-Leyden crystal protein (lysophospholipase) and eosinophil granule major basic protein in human mast cells. J Immunol 1986;136:852–855. Dvorak AM, Warner JA, Kissell S, Lichtenstein LM, MacGlashan DW Jr: f-Met peptide-induced degranulation of human basophils. Lab Invest 1991;64:234–253. Golightly LM, Thomas LL, Dvorak AM, Ackerman SJ: Charcot-Leyden crystal protein in the degranulation and recovery of activated basophils. J Leukoc Biol 1992;51:386–392. West GB: Histamine, mast cells and basophils; in Waksman BH (ed): Chem Immunol. New York, Karger, 1990, vol 49, pp 121–141. Hagen P, Barrnett RJ, Lee FL: Biochemical and electron microscopic study of particles isolated from mastocytoma cells. J Pharmacol Exp Ther 1959;126:91–108. Anderson P, Rohlich P, Uvnäs B: The inﬂuence of calcium on morphology and histamine content of isolated intact rat mast cell granules. Acta Physiol Scand 1979;105:350–358. Anderson P, Uvnäs B: Selective localization of histamine to electron-dense granules in antigen-challenged sensitized rat mast cells and to similar granules isolated from sonicated mast cells. An electron microscope autoradiographic study. Acta Physiol Scand 1975;94:63–73. Panula P, Happola O, Airaksinen MS, Auvinen S, Virkamaki A: Carbodiimide as a tissue ﬁxative in histamine immunohistochemistry and its application in developmental neurobiology. J Histochem Cytochem 1988;36:259–269. Steinbusch HWM, Mulder AH: Immunohistochemical localization of histamine in neurons and mast cells in the rat brain; in Björkland A, Hökfolt T (eds): Handbook of Chemical Neuroanatomy, vol 3:Classical Transmittors and Transmittor Receptors in the CNS. Part II. New York, Elsevier, 1984, p 126. Bendayan M: The enzyme-gold technique: A new cytochemical approach for the ultrastructural localization of macromolecules; in Bullock CR, Petrusz P (eds): Techniques in Immunocytochemistry. London, Academic Press, 1985, vol 3, pp 179–201. Bendayan M: Ultrastructural localization of nucleic acids by the use of enzyme-gold complexes. J Histochem Cytochem 1981;29:531–541. Dvorak AM, Morgan ES: Ribonuclease-gold ultrastructural localization of heparin in isolated human lung mast cells stimulated to undergo anaphylactic degranulation and recovery in vitro. Clin Exp Allergy 1999;29:1118–1128. Beaven MA: Histamine. N Engl J Med 1976;294:30–36. Riley JF, West GB: The presence of histamine in tissue mast cells. J Physiol 1953;120:528–537. Lagunoff D, Phillips M, Benditt EP: The histochemical demonstration of histamine in mast cells. J Histochem Cytochem 1961;9:534–541. Juhlin L, Shelley WB: Detection of histamine by a new ﬂuorescent o-phthalaldehyde stain. J Histochem Cytochem 1966;14:525–528.

References

329

221

222 223 224

225 226 227 228 229 230 231 232 233

234 235 236 237 238 239

240 241 242 243 244 245 246 247 248

Anderson P, Uvnäs B: Selective localization of histamine to electron-dense granules in antigen-challenged sensitized rat mast cells and to similar granules isolated from sonicated mast cells. An electron microscope autoradiographic study. Acta Physiol Scand 1975;94:63–73. Frens G: Controlled nucleation for the regulation of the particle size in monodisperse gold suspensions. Nat Phys Sci 1973;241:20–22. McBride P, Bradley D, Kaliner M: Evaluation of a radioimmunoassay for histamine measurement in biologic ﬂuids. J Allergy Clin Immunol 1988;82:638–646. Dvorak AM: Recovery of human lung mast cells from anaphylactic degranulation utilizes a mixture of conservation and synthetic mechanisms; in Galli SJ, Austen KF (eds): Mast Cell and Basophil Differentiation and Function in Health and Disease. New York, Raven Press, 1989, pp 119–148. Rabenstein DL, Ludowyke R, Lagunoff D: Proton nuclear magnetic resonance studies of mast cell histamine. Biochemistry 1987;26:6923–6926. Buffoni F: Histaminase and related amine oxidases. Pharmacol Rev 1966;18:1163–1199. Zeller EA: Identity of histaminase and diamine oxidase. Fed Proc 1965;24:766–768. Tabor CW, Tabor H: Polyamines. Annu Rev Biochem 1984;53:749–790. Kropinski AMB, Bose RJ, Warren RAJ: 5-(4-aminobutylaminomethyl)uracil, an unusual pyrimidine from the deoxyribonucleic acid of bacteriophage ⌽W-14. Biochemistry 1973;12:151–157. Neuhard J, Maltman KL, Warren RAJ: Bacteriophage ⌽W-14-infected Pseudomonas acidovorans synthesizes hydroxymethyldeoxyuridine triphosphate. J Virol 1980;34:347–353. Boynton AL, Whitﬁeld JF, Isaacs RJ: A possible involvement of polyamines in the initiation of DNA synthesis by human WI-38 and mouse BALB/3T3 cells. J Cell Physiol 1976;89:481–488. Heby O, Marton LJ, Wilson CB, Gray JW: Effect of methylglyoxal-bis(guanylhydrazone), an inhibitor of spermidine and spermine synthesis, on cell cycle traverse. Eur J Cancer 1977;13:1009–1017. Sunkara PS, Pargac MB, Nishioka K, Rao PN: Differential effects of inhibition of polyamine biosynthesis on cell cycle traverse and structure of the prematurely condensed chromosomes of normal and transformed cells. J Cell Physiol 1979;98:451–457. Mach M, Ebert P, Popp R, Ogilvie A: Compartmentalization of polyamines in mammalian cells. Biochem Biophys Res Commun 1982;104:1327–1334. Best CH: The disappearance of histamine from autolysing lung tissue. J Physiol 1929;67:256–263. Best CH, McHenry EW: The inactivation of histamine. J Physiol 1930;70:349–372. Zeller EA: Über den enzymatischen Abbau von Histamin und Diaminen. Helv Chim Acta 1938;21: 880–890. Schayer RW: Catabolism of physiological quantities of histamine in vivo. Physiol Rev 1959;39:116– 126. Dvorak AM, Klebanoff SJ, Henderson WR, Monahan RA, Pyne K, Galli SJ: Vesicular uptake of eosinophil peroxidase by guinea pig basophils and by cloned mouse mast cells and granule-containing lymphoid cells. Am J Pathol 1985;118:425–438. Bendayan M: Ultrastructural localization of nucleic acids by the use of enzyme-gold complexes: Inﬂuence of ﬁxation and embedding. Biol Cell 1982;43:153–156. Bendayan M: Enzyme-gold electron microscopic cytochemistry: A new afﬁnity approach for the ultrastructural localization of macromolecules. J Electron Microsc Tech 1984;1:349–372. Bendayan M: The enzyme-gold cytochemical approach: A review; in Hayat MA (ed): Colloidal Gold. Principles, Methods, and Applications. San Diego, Academic Press, 1989, vol 2, pp 117–147. Cheniclet C, Carde JP: Ultrastructural localization of RNAs in plastids with an RNase-gold method. Biol Cell 1987;59:79–88. Bendayan M, Benhamou N: Ultrastructural localization of glucoside residues on tissue sections by applying the enzyme-gold approach. J Histochem Cytochem 1987;35:1149–1155. Cheniclet C, Bendayan M: Comparative pyrimidine- and purine-speciﬁc RNAse-gold labeling on pancreatic acinar cells and isolated hepatocytes. J Histochem Cytochem 1990;38:551–562. Brooks EM, Binnington KC: Gold labeling with wheat germ agglutinin and RNAse on osmicated tissue embedded in epoxy resin. J Histochem Cytochem 1989;37:1557–1561. Londoño I, Bendayan M: High-resolution cytochemistry of neuraminic and hexuronic acid-containing macromolecules applying the enzyme-gold approach. J Histochem Cytochem 1988;36:1005–1014. Dvorak AM, Morgan ES, Lichtenstein LM, MacGlashan DW Jr: Activated human basophils contain histamine in cytoplasmic vesicles. Int Arch Allergy Immunol 1994;105:8–11.

Ultrastructure of Mast Cells and Basophils

330

249

250

251

252

253

254 255

256 257 258

259 260 261 262

263 264

265 266 267 268 269

270 271 272

Dvorak AM, Tepper RI, Weller PF, Morgan ES, Estrella P, Monahan-Earley RA, Galli SJ: Piecemeal degranulation of mast cells in the inﬂammatory eyelid lesions of interleukin-4 transgenic mice. Evidence of mast cell histamine release in vivo by diamine oxidase-gold enzyme-afﬁnity ultrastructural cytochemistry. Blood 1994;83:3600–3612. Dvorak AM, MacGlashan DW Jr, Morgan ES, Lichtenstein LM: Histamine distribution in human basophil secretory granules undergoing FMLP-stimulated secretion and recovery. Blood 1995; 86: 3560–3566. Dvorak AM, Morgan ES, Schleimer RP, Lichtenstein LM: Diamine oxidase-gold ultrastructural localization of histamine in isolated human lung mast cells stimulated to undergo anaphylactic degranulation and recovery in vitro. J Leukoc Biol 1996;59:824–834. Dvorak AM, Costa JJ, Morgan ES, Monahan-Earley RA, Galli SJ: Diamine oxidase-gold ultrastructural localization of histamine in human skin biopsies containing mast cells stimulated to degranulate in vivo by exposure to recombinant human stem cell factor. Blood 1997;90:2893–2900. Dvorak AM, Morgan ES: Diamine oxidase-gold enzyme-afﬁnity ultrastructural demonstration that human gut mucosal mast cells secrete histamine by piecemeal degranulation in vivo. J Allergy Clin Immunol 1997;99:812–820. Lindahl U, Höök M: Glycosaminoglycans and their binding to biological macromolecules. Annu Rev Biochem 1978;47:385–417. Mendelsohn SL, Young DA: Inhibition of ribonuclease. Efﬁcacy of sodium dodecyl sulfate, diethyl pyrocarbonate, proteinase K and heparin using a sensitive ribonuclease assay. Biochim Biophys Acta 1978;519:461–473. Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ: Isolation of biologically active ribonucleic acid from sources enriched in ribonuclease. Biochemistry 1979;18:5294–5299. Jaques LB: Heparins – Anionic polyelectrolyte drugs. Pharmacol Rev 1980;31:99–166. Stevens RL, Fox CC, Lichtenstein LM, Austen KF: Identiﬁcation of chondroitin sulfate E proteoglycans and heparin proteoglycans in the secretory granules of human lung mast cells. Proc Natl Acad Sci USA 1988;85:2284–2287. Thompson HL, Schulman ES, Metcalfe DD: Identiﬁcation of chondroitin sulfate E in human lung mast cells. J Immunol 1988;140:2708–2713. Anderson NG, Wilbur KM: Studies on isolated cell components. J Gen Physiol 1951;34:647–656. Salmivirta M, Lidholt K, Lindahl U: Heparan sulfate: A piece of information. FASEB J 1996; 10: 1270–1279. Uvnäs B, Åborg CH, Bergendorff A: Storage of histamine in mast cells. Evidence for an ionic binding of histamine to protein carboxyls in the granule heparin-protein complex. Acta Physiol Scand Suppl 1970;336:1–26. Lagunoff D: Analysis of dye binding sites in mast cell granules. Biochemistry 1974;13:3982–3986. Wingren U, Enerbäck L: Mucosal mast cells of the rat intestine: A re-evaluation of ﬁxation and staining properties, with special reference to protein blocking and solubility of the granular glycosaminoglycan. Histochem J 1983;15:571–582. Bussolati G, Gugliotta P: Nonspeciﬁc staining of mast cells by avidin-biotin-peroxidase complexes. J Histochem Cytochem 1983;31:1419–1421. Tharp MD, Seelig LL Jr, Tigelaar RE, Bergstresser PR: Conjugated avidin binds to mast cell granules. J Histochem Cytochem 1985;33:27–32. Dvorak AM, Morgan ES: Ribonuclease-gold labels chondroitin sulfate in guinea pig basophil granules. Histochem J 1998;30:603–608. Orenstein NS, Galli SJ, Dvorak AM, Silbert JE, Dvorak HF: Sulfated glycosaminoglycans of guinea pig basophilic leukocytes. J Immunol 1978;121:586–592. Orenstein NS, Galli SJ, Dvorak AM, Dvorak HF: Glycosaminoglycans and proteases of guinea pig basophilic leukocytes; in Becker EL, Simon AS, Austen KF (eds): Biochemistry of the Acute Allergic Reactions: Fourth International Symposium. New York, Liss, 1981, pp 123–143. Dvorak AM: Monograph – Procedural guide to specimen handling for the ultrastructural pathology service laboratory. J Electron Microsc Tech 1987;6:255–301. Danon D, Goldstein L, Marikovsky Y, Skutelsky E: Use of cationized ferritin as a label of negative charges on cell surfaces. J Ultrastruct Res 1972;38:500–510. Skutelsky E, Roth J: Cationic colloidal gold – A new probe for the detection of anionic cell surface sites by electron microscopy. J Histochem Cytochem 1986;34:693–696.

References

331

273 274 275 276

277

278 279

280 281 282 283 284 285 286 287

288 289 290

291 292

293 294

295 296 297

Horisberger M: Evaluation of colloidal gold as a cytochemical marker for transmission and scanning electron microscopy. Biol Cell 1979;36:253–258. Dvorak HF, Selvaggio SS, Dvorak AM, Colvin RB, Lean DB, Rypysc J: Puriﬁcation of basophilic leukocytes from guinea pig blood and bone marrow. J Immunol 1974;113:1694–1702. Dvorak AM: Biology and morphology of basophilic leukocytes; in Bach MK (ed): Immediate Hypersensitivity – Modern Concepts and Developments. New York, Dekker, 1978, pp 369–405. Dvorak AM, Dvorak HF: Cutaneous basophil hypersensitivity – A 20-year perspective, 1970–1990; in Foreman JC (ed): Immunopharmacology of Mast Cells and Basophils. London, Academic Press, 1993, pp 153–180. Galli SJ, Orenstein NS, Gill PJ, Silbert JE, Dvorak AM, Dvorak HF: Sulphated glycosaminoglycans synthesised by basophil-enriched human leukaemic granulocytes; in Pepys J, Edwards AM (eds): The Mast Cell. Its Role in Health and Disease. Tunbridge Wells, Pitman Medical, 1979, p 842. Metcalfe DD, Bland CE, Wasserman SI: Biochemical and functional characterization of proteoglycans isolated from basophils of patients with chronic leukemia. J Immunol 1984;132:1943–1950. Ishizaka T, Conrad DH, Huff TF, Metcalfe DD, Stevens RL, Lewis RA: Unique features of human basophilic granulocytes developed in in vitro culture. Int Arch Allergy Appl Immunol 1985;77:137– 143. Hastie R: A study of the ultrastructure of human basophil leukocytes. Lab Invest 1974;31:223–231. Parmley RT, Spicer SS, Komiyama A, Dow LW, Austin RL: Ultrastructural cytochemistry of basophils in chronic myelocytic leukemia. Exp Mol Pathol 1979;30:41–54. Spicer SS: Diamine methods for differentiating mucosubstances histochemically. J Histochem Cytochem 1965;13:211–234. Hardin JH, Spicer SS: Ultrastructural localization of dialyzed iron-reactive mucosubstance in rabbit heterophils, basophils, and eosinophils. J Cell Biol 1971;48:368–386. Horn RG, Spicer SS: Sulfated mucopolysaccharide and basic protein in certain granules of rabbit leukocytes. Lab Invest 1964;13:1–15. Wetzel BK, Horn RG, Spicer SS: Fine structural studies on the development of heterophil, eosinophil, and basophil granulocytes in rabbits. Lab Invest 1967;16:349–382. Uvnäs B, Åborg CH: On the cation exchanger properties of rat mast cell granules and their storage of histamine. Acta Physiol Scand 1977;100:309–314. Oliani SM, Freymüller E, Takahashi HK, Straus AH: Immunocytochemical localization of heparin in secretory granules of rat peritoneal mast cells using a monoclonal anti-heparin antibody (ST-1). J Histochem Cytochem 1997;45:231–235. Skutelsky E, Shoichetman T, Hammel I: An histochemical approach to characterization of anionic constituents in mast cell secretory granules. Histochem Cell Biol 1995;104:453–458. Dvorak AM, Morgan ES: Ribonuclease-gold labels proteoglycan-containing cytoplasmic granules and ribonucleic acid-containing organelles – A survey. Histol Histopathol 1999;14:597–626. Dvorak HF, Mihm MC Jr, Dvorak AM, Johnson RA, Manseau EJ, Morgan E, Colvin RB: Morphology of delayed type hypersensitivity reactions in man. I. Quantitative description of the inﬂammatory response. Lab Invest 1974;31:111–130. Dvorak AM, Mihm MC Jr, Dvorak HF: Degranulation of basophilic leukocytes in allergic contact dermatitis reactions in man. J Immunol 1976;116:687–695. Dvorak AM, Warner JA, Morgan E, Kissell-Rainville S, Lichtenstein LM, MacGlashan DW Jr: An ultrastructural analysis of tumor-promoting phorbol diester-induced degranulation of human basophils. Am J Pathol 1992;141:1309–1322. Dvorak AM: The ﬁne structure of human basophils and mast cells; in Holgate ST (ed): Mast Cells, Mediators and Disease. Dordrecht, Kluwer Academic, 1988, pp 29–97. Dvorak HF, Dvorak AM, Churchill WF: Immunologic rejection of diethylnitrosamine-induced hepatomas in strain 2 guinea pigs. Participation of basophilic leukocytes and macrophage aggregates. J Exp Med 1973;137:751–775. Dvorak AM, Dvorak HF: Structure of Freund’s complete and incomplete adjuvants. Relation of adjuvanticity to structure. Immunology 1974;27:99–114. Dienes L: The immunological signiﬁcance of the tuberculous tissue. J Immunol 1928;15:141–152. Raffel S, Newel JM: The delayed hypersensitivity induced by antigen-antibody complexes. J Exp Med 1958;108:823–841.

Ultrastructure of Mast Cells and Basophils

332

298

Jones TD, Mote JR: The phases of foreign protein sensitization in human beings. N Engl J Med 1934; 210:120–123. 299 Richerson HB, Dvorak HF, Leskowitz S: Cutaneous basophilic hypersensitivity: A new interpretation of the Jones-Mote reaction. J Immunol 1969;103:1431–1434. 300 Richerson HB, Dvorak HF, Leskowitz S: Cutaneous basophil hypersensitivity. I. A new look at the Jones-Mote reaction, general characteristics. J Exp Med 1970;132:546–557. 301 Bast RC Jr, Simpson BA, Dvorak HF: Heterogeneity of the cellular immune response. II. The role of adjuvant, lymphocyte stimulation in cutaneous basophil hypersensitivity. J Exp Med 1971; 133: 202–215. 302 Dvorak AM, Bast RC Jr, Dvorak HF: Morphologic changes in draining lymph nodes and in lymphocyte cultures after sensitization with complete or incomplete Freund’s adjuvant. Correlation with immunologic events in vivo and in culture. J Immunol 1971;107:422–435. 303 Richerson HB: Cutaneous basophil (Jones-Mote) hypersensitivity after ‘tolerogenic’ doses of intravenous ovalbumin in the guinea pig. J Exp Med 1971;134:630–641. 304 Dvorak HF, Hammond ME, Colvin RB, Manseau EJ, Goodwin J: Systemic expression of cutaneous basophil hypersensitivity. J Immunol 1977;118:1549–1557. 305 Dvorak HF, Galli SJ, Dvorak AM: Cellular and vascular manifestations of cell-mediated immunity. Hum Pathol 1986;17:122–137. 306 Stadecker MJ, Leskowitz S: The inhibition of cutaneous basophil hypersensitivity reactions by a heterologous anti-guinea pig T cell serum. J Immunol 1976;116:1646–1651. 307 Dvorak AM, Monahan RA, Osage JE, Dickersin GR: Crohn’s disease: Transmission electron microscopic studies. II. Immunologic inﬂammatory response. Alterations of mast cells, basophils, eosinophils, and the microvasculature. Hum Pathol 1980;11:606–619. 308 Dvorak AM, McLeod RS, Onderdonk A, Monahan-Earley RA, Cullen JB, Antonioli DA, Morgan E, Blair JE, Estrella P, Cisneros RL, Silen W, Cohen Z: Ultrastructural evidence for piecemeal and anaphylactic degranulation of human gut mucosal mast cells in vivo. Int Arch Allergy Immunol 1992;99: 74–83. 309 Knutson L, Ahrenstedt Ö, Odlind B, Hällgren R: The jejunal secretion of histamine is increased in active Crohn’s disease. Gastroenterology 1990;98:849–854. 310 Dvorak AM, Mihm MC Jr, Osage JE, Kwan TH, Austen KF, Wintroub BU: Bullous pemphigoid, an ultrastructural study of the inﬂammatory response: Eosinophil, basophil and mast cell granule changes in multiple biopsies from one patient. J Invest Dermatol 1982;78:91–101. 311 Dvorak HF, Dvorak AM: Basophils, mast cells, and cellular immunity in animals and man. Hum Pathol 1972;3:454–456. 312 Tepper RI, Levinson DA, Stanger BZ, Campos-Torres J, Abbas AK, Leder P: IL-4 induces allergiclike inﬂammatory disease and alters T cell development in transgenic mice. Cell 1990;62:457–467. 313 Paul WE: Interleukin-4/B cell stimulatory factor 1: One lymphokine, many functions. FASEB J 1987; 1:456–461. 314 Mosmann TR, Coffman RL: TH1 and TH2 cells: Different patterns of lymphokine secretion lead to different functional properties. Annu Rev Immunol 1989;7:145–173. 315 Burd PR, Rogers HW, Gordon JR, Martin CA, Jayaraman S, Wilson SD, Dvorak AM, Galli SJ, Dorf ME: Interleukin-3-dependent and -independent mast cells stimulated with IgE and antigen express multiple cytokines. J Exp Med 1989;170:245–257. 316 Plaut M, Pierce JH, Watson CJ, Hanley-Hyde J, Nordan RP, Paul WE: Mast cell lines produce lymphokines in response to cross-linkage of Fc⑀RI or to calcium ionophores. Nature 1989;339:64–67. 317 Paul WE: Interleukin-4: A prototypic immunoregulatory lymphokine. Blood 1991;77:1859–1870. 318 Bradding P, Feather IH, Howarth PH, Mueller R, Roberts JA, Britten K, Bews JPA, Hunt TC, Okayama Y, Heusser CH, Bullock GR, Church MK, Holgate ST: Interleukin-4 is localized to and released by human mast cells. J Exp Med 1992;176:1381–1386. 318A Min B, Prout M, Hu-Li J, Zhu J, Jankovic D, Morgan ES, Urban JF Jr, Dvorak AM, Finkelman FD, LeGros G, Paul WE: Basophils produce IL-4 and accumulate in tissues after infection with a Th2-inducing parasite. J Exp Med 2004;200:507–517. 319 Coffman RL, Ohara J, Bond MW, Carty J, Zlotnik A, Paul WE: B cell stimulatory factor-1 enhances the IgE response of lipopolysaccharide-activated B cells. J Immunol 1986;136:4538–4541.

References

333

320 321 322

323 324 325 326

327

328

329

330

331

332 333 334 335 336 337 338 339

340 341

342 343 344

Snapper CM, Paul WE: Interferon-␥ and B cell stimulatory factor-1 reciprocally regulate Ig isotype production. Science 1987;236:944–947. Snapper CM, Finkelman FD, Paul WE: Differential regulation of IgG1 and IgE synthesis by interleukin-4. J Exp Med 1988;167:183–196. Finkelman FD, Katona IM, Urban JF Jr, Snapper CM, Ohara J, Paul WE: Suppression of in vivo polyclonal IgE responses by monoclonal antibody to the lymphokine B-cell stimulatory factor 1. Proc Natl Acad Sci USA 1986;83:9675–9678. Kuhn R, Rajewsky K, Muller W: Generation and analysis of interleukin-4 deﬁcient mice. Science 1991;254:707–710. Paul WE: Pleiotropy and redundancy: T cell-derived lymphokines in the immune response. Cell 1989; 57:521–524. Collin HB, Allansmith MR: Basophils in vernal conjunctivitis in humans: An electron microscopic study. Invest Ophthalmol Vis Sci 1977;16:858–864. Maggi E, Biswas P, Del Prete G, Parronchi P, Macchia D, Simonelli C, Emmi L, De Carli M, Tiri A, Ricci M, Romagnani S: Accumulation of Th-2-like helper T cells in the conjunctiva of patients with vernal conjunctivitis. J Immunol 1991;146:1169–1174. Dvorak AM, Schroeder JT, MacGlashan DW Jr, Bryan KP, Morgan ES, Lichtenstein LM, MacDonald SM: Comparative ultrastructural morphology of human basophils stimulated to release histamine by anti-IgE, recombinant IgE-dependent histamine-releasing factor, or monocyte chemotactic protein-1. J Allergy Clin Immunol 1996;98:355–370. Warner JA, Peters SP, Lichtenstein LM, Hubbard W, Yancey KB, Stevenson HC, Miller PJ, MacGlashan DW Jr: Differential release of mediators from human basophils: Differences in arachidonic acid metabolism following activation by unrelated stimuli. J Leukoc Biol 1989;45:558–571. Schleimer RP, Gillespie E, Lichtenstein LM: Release of histamine from human leukocytes stimulated with the tumor-promoting phorbol diesters. I. Characterization of the response. J Immunol 1981;126: 570–574. Schleimer RP, Gillespie E, Daiuta R, Lichtenstein LM: Release of histamine from human leukocytes stimulated with the tumor-promoting phorbol diesters. II. Interaction with other stimuli. J Immunol 1982;128:136–140. MacDonald SM: Histamine-releasing factors and IgE heterogeneity; in Middleton E, Reed CE, Ellis EF, Adkinson NF, Yunginger JW, Busse WW (eds): Allergy – Principles and Practice, ed 4. St Louis, Mosby-Year Book, 1993, pp 1–11. MacDonald SM, Rafnar T, Langdon J, Lichtenstein LM: Molecular identiﬁcation of an IgE-dependent histamine-releasing factor. Science 1995;269:688–690. Palade G: Intracellular aspects of the process of protein synthesis. Science 1975;189:347–358. Kelly RB: Pathways of protein secretion in eukaryotes. Science 1985;230:25–32. Castle JD: Sorting and secretory pathways in exocrine cells. Am J Respir Cell Mol Biol 1990;2:119– 126. Matsuuchi L, Kelly RB: Constitutive and basal secretion from the endocrine cell line, AtT-20. J Cell Biol 1991;112:843–852. Chavez RA, Miller SG, Moore H-PH: A biosynthetic regulated secretory pathway in constitutive secretory cells. J Cell Biol 1996;133:1177–1191. Cleves AE, Cooper DNW, Barondes SH, Kelly RB: A new pathway for protein export in Saccharomyces cerevisiae. J Cell Biol 1996;133:1017–1026. Login GR, Quissell DO, Bryan KP, Dvorak AM: A role for cytoplasmic vesicles in amylase release from secretory granules in isolated rat parotid gland acinar cells. Int Congress Electron Microsc 1994; 13:83–84. Bock JB, Scheller RH: Protein transport. A fusion of new ideas. Nature 1997;387:133–135. Wilson DW, Wilcox CA, Flynn GC, Chen E, Kuang WJ, Henzel WJ, Block MR, Ullrich A, Rothman JE: A fusion protein required for vesicle-mediated transport in both mammalian cells and yeast. Nature 1989;339:355–359. Nichols BJ, Ungermann C, Pelham HRB, Wickner WT, Haas A: Homotypic vacuolar fusion mediated by t- and v-SNAREs. Nature 1997;387:199–202. Novick P, Garrett MD: Vesicular transport. No exchange without receipt. Nature 1994;369:18–19. Pfeffer SR: GTP-binding proteins in intracellular transport. Trends Cell Biol 1992;2:41–46.

Ultrastructure of Mast Cells and Basophils

334

345 346 347 348 349

350 351 352

353 354 355 356

357

358

359 360 361 362 363 364

365

366

367

Sciuto TE, Feng D, Nagy J, Dvorak HF, Dvorak AM: Basophils exit venules by a transcellular pathway in cutaneous basophil hypersensitivity in guinea pigs. In preparation, 2005. Hammel I, Lagunoff D, Bauza M, Chi E: Periodic, multimodal distribution of granule volumes in mast cells. Cell Tissue Res 1983;228:51–59. Hammel I, Dvorak AM, Galli SJ: Defective cytoplasmic granule formation. I. Abnormalities affecting tissue mast cells and pancreatic acinar cells of beige mice. Lab Invest 1987;56:321–328. Hammel I, Lagunoff D, Wysolmerski R: Theoretical considerations on the formation of secretory granules in the rat pancreas. Exp Cell Res 1993;204:1–5. Feng D, Nagy JA, Pyne K, Hammel I, Dvorak HF, Dvorak AM: Pathways of macromolecular extravasation across microvascular endothelium in response to VPF/VEGF and other vasoactive mediators. Microcirculation 1999;6:23–44. Bruns RR, Palade GE: Studies on blood capillaries. I. General organization of blood capillaries in muscle. J Cell Biol 1968;37:244–276. Bruns RR, Palade GE: Studies on blood capillaries. II. Transport of ferritin molecules across the wall of muscle capillaries. J Cell Biol 1968;37:277–299. Graham RC Jr, Karnovsky MJ: The early stages of absorption of injected horseradish peroxidase in the proximal tubules of the mouse kidney: Ultrastructural cytochemistry by a new technique. J Histochem Cytochem 1966;14:291–302. Palade GE: Fine structure of blood capillaries. J Appl Physiol 1953;24:1424. Palade GE: The microvascular endothelium revisited; in Simionescu N, Simionescu M (eds): Endothelial Cell Biology in Health and Disease. New York, Plenum Press, 1988, pp 3–22. Palade GE, Bruns RR: Structural modulations of plasmalemmal vesicles. J Cell Biol 1968; 37: 633– 649. Schnitzer JE, Oh P, Jacobson BS, Dvorak AM: Caveolae from luminal plasmalemma of rat lung endothelium: Microdomains enriched in caveolin, Ca2+-ATPase, and inositol triphosphate receptor. Proc Natl Acad Sci USA 1995;92:1759–1763. Feng D, Nagy JA, Dvorak HF, Dvorak AM: Ultrastructural studies deﬁne soluble macromolecular, particulate, and cellular transendothelial cell pathways in venules, lymphatic vessels, and tumor-associated microvessels in man and animals. Microsc Res Tech 2002;57:289–326. Feng D, Nagy JA, Brekken RA, Pettersson A, Manseau EJ, Pyne K, Mulligan R, Thorpe PE, Dvorak HF, Dvorak AM: Ultrastructural localization of the vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) receptor-2 (FLK-1, KDR) in normal mouse kidney and in the hyperpermeable vessels induced by VPF/VEGF-expressing tumors and adenoviral vectors. J Histochem Cytochem 2000;48:545–555. Severs NJ: Caveolae: Static inpocketings of the plasma membrane, dynamic vesicles or plain artifact? J Cell Sci 1988;90:341–348. Clementi F, Palade GE: Intestinal capillaries. I. Permeability to peroxidase and ferritin. J Cell Biol 1969;41:33–58. Clementi F, Palade GE: Intestinal capillaries. II. Structural effects of EDTA and histamine. J Cell Biol 1969;42:706–714. Dvorak AM, Dvorak HF, Karnovsky MJ: Uptake of horseradish peroxidase by guinea pig basophilic leukocytes. Lab Invest 1972;26:27–39. Karnovsky MJ: The ultrastructural basis of capillary permeability studied with peroxidase as a tracer. J Cell Biol 1967;35:213–236. Dvorak HF, Dvorak AM, Manseau EJ, Wiberg L, Churchill WH: Fibrin gel investment associated with line 1 and line 10 solid tumor growth, angiogenesis, and ﬁbroplasia in guinea pigs. Role of cellular immunity, myoﬁbroblasts, microvascular damage, and infarction in line 1 tumor regression. J Natl Cancer Inst 1979;62:1459–1472. Dvorak HF, Orenstein NS, Carvalho AC, Churchill WH, Dvorak AM, Galli SJ, Feder J, Bitzer AM, Rypysc J, Giovinco P: Induction of a ﬁbrin-gel investment: An early event in line 10 hepatocarcinoma growth mediated by tumor-secreted products. J Immunol 1979;122:166–174. Dvorak HF, Nagy JA, Berse B, Brown LF, Yeo KT, Yeo TK, Dvorak AM, Van De Water L, Sioussat TM, Senger DR: Vascular permeability factor, ﬁbrin, and the pathogenesis of tumor stroma formation. Ann NY Acad Sci 1992;667:101–111. Senger DR, Galli SJ, Dvorak AM, Perruzzi CA, Harvey VS, Dvorak HF: Tumor cells secrete a vascular permeability factor that promotes accumulation of ascites ﬂuid. Science 1983;219:983–985.

References

335

368 369

370

371 372 373 374

375

376

377

378

379 380 381 382

383

384 385 386 387

388

Senger DR, Perruzzi CA, Feder J, Dvorak HF: A highly conserved vascular permeability factor secreted by a variety of human and rodent tumor cell lines. Cancer Res 1986;46:5629–5632. Senger DR, Connolly DT, Van De Water L, Feder J, Dvorak HF: Puriﬁcation and NH2-terminal amino acid sequence of guinea pig tumor-secreted vascular permeability factor. Cancer Res 1990;50: 1774–1778. Senger DR, Van De Water L, Brown LF, Nagy JA, Yeo KT, Yeo TK, Berse B, Jackman RW, Dvorak AM, Dvorak HF: Vascular permeability factor (VPF, VEGF) in tumor biology. Cancer Metastasis Rev 1993;12:303–324. Ferrara N, Henzel WJ: Pituitary follicular cells secrete a novel heparin-binding growth factor speciﬁc for vascular endothelial cells. Biochem Biophys Res Commun 1989;161:851–858. Ferrara N, Houck K, Jakeman L, Leung DW: Molecular and biological properties of the vascular endothelial growth factor family of proteins. Endocr Rev 1992;13:18–32. Leung DW, Cachianes G, Kuang W-J, Goeddel DV, Ferrara N: Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 1989;246:1306–1309. Gospodarowicz D, Abraham JA, Schilling J: Isolation and characterization of a vascular endothelial cell mitogen produced by pituitary-derived folliculo-stellate cells. Proc Natl Acad Sci USA 1989;86: 7311–7315. Dvorak HF, Nagy JA, Feng D, Brown LF, Dvorak AM: Vascular permeability factor/vascular endothelial growth factor and the signiﬁcance of microvascular hyperpermeability in angiogenesis. Curr Top Microbiol Immunol 1999;237:97–132. Brown LF, Detmar M, Claffey K, Nagy JA, Feng D, Dvorak AM, Dvorak HF: Vascular permeability factor/vascular endothelial growth factor: A multifunctional angiogenic cytokine; in Goldberg ID, Rosen EM (eds): Regulation of Angiogenesis. Basel, Birkhäuser, 1997, pp 233–269. Boesiger J, Tsai M, Maurer M, Yamaguchi M, Brown LF, Claffey KP, Dvorak HF, Galli SJ: Mast cells can secrete vascular permeability factor/vascular endothelial cell growth factor and exhibit enhanced release after immunoglobulin E-dependent upregulation of Fc⑀ receptor I expression. J Exp Med 1998; 188:1135–1145. Grützkau A, Krüger-Krasagakes S, Baumeister H, Schwarz C, Kögel H, Welker P, Lippert U, Henz BM, Moller A: Synthesis, storage, and release of vascular endothelial cell growth factor/vascular permeability factor (VEGF/VPF) by human mast cells: Implications for the biological signiﬁcance of VEGF206. Mol Biol Cell 1998;9:875–884. Esser S, Wolburg K, Wolburg H, Breier G, Kurzchalia T, Risau W: Vascular endothelial growth factor induces endothelial fenestrations in vitro. J Cell Biol 1998;140:947–959. Roberts WG, Palade GE: Increased microvascular permeability and endothelial fenestration induced by vascular endothelial growth factor. J Cell Sci 1995;108:2369–2379. Roberts WG, Palade GE: Neovasculature induced by vascular endothelial growth factor is fenestrated. Cancer Res 1997;57:765–772. Roberts WG, Delaat J, Nagane M, Huang S, Cavenee WK, Palade GE: Host microvasculature inﬂuence on tumor vascular morphology and endothelial gene expression. Am J Pathol 1998;153: 1239– 1248. Qu-Hong, Nagy JA, Senger DR, Dvorak HF, Dvorak AM: Ultrastructural localization of vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) to the abluminal plasma membrane and vesiculovacuolar organelles of tumor microvascular endothelium. J Histochem Cytochem 1995; 43:381–389. Majno G, Palade GE: Studies on inﬂammation. I. The effect of histamine and serotonin on vascular permeability: An electron microscopic study. J Biophys Biochem Cytol 1961;11:571–605. Majno G, Palade GE, Schoeﬂ GI: Studies on inﬂammation. II. The site of action of histamine and serotonin along the vascular tree: A topographic study. J Biophys Biochem Cytol 1961;11:607–626. Majno G, Shea SM, Leventhal M: Endothelial contraction induced by histamine-type mediators: An electron microscopic study. J Cell Biol 1969;42:647–672. Dvorak HF, Brown LF, Detmar M, Dvorak AM: Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am J Pathol 1995;146:1029– 1039. Ferrara N: Vascular endothelial growth factor: Molecular and biological aspects. Curr Top Microbiol Immunol 1999;237:1–30.

Ultrastructure of Mast Cells and Basophils

336

389 390

391

392

393

394 395 396

397

398 399

400

401

402 403 404 405 406 407

408

Ferrara N, Davis-Smyth T: The biology of vascular endothelial growth factor. Endocr Rev 1997; 18: 4–25. Brown LF, Berse B, Tognazzi K, Manseau EJ, Van de Water L, Senger DR, Dvorak HF, Rosen S: Vascular permeability factor mRNA and protein expression in human kidney. Kidney Int 1992; 42: 1457–1461. Brown LF, Yeo KT, Berse B, Teo TK, Senger DR, Dvorak HF, Van De Water L: Expression of vascular permeability factor (vascular endothelial growth factor) by epidermal keratinocytes during wound healing. J Exp Med 1992;176:1375–1379. Dvorak HF, Sioussat TM, Brown LF, Nagy JA, Sotrel A, Manseau E, Van De Water L, Senger DR: Distribution of vascular permeability factor (vascular endothelial growth factor) in tumors: Concentration in tumor blood vessels. J Exp Med 1991;174:1275–1278. Achen MG, Jeltsch M, Kukk E, Makinen T, Vitali A, Wilks A, Alitalo K, Stacker SA: Vascular endothelial growth factor D (VEGF-D) is a ligand for the tyrosine kinases VEGF receptor 2 (Flk1) and VEGF receptor 3 (Flt4). Proc Natl Acad Sci USA 1998;95:548–553. Brekken RA, Huang X, King SW, Thorpe PE: Vascular endothelial growth factor as a marker of tumor endothelium. Cancer Res 1998;58:1952–1959. De Vries C, Escobedo JA, Ueno H, Houck K, Ferrara N, Williams LT: The fms-like tyrosine kinase, a receptor for vascular endothelial growth factor. Science 1992;255:989–991. Millauer B, Wizigmann-Voos S, Schnurch H, Martinez R, Moller NP, Risau W, Ullrich A: High afﬁnity VEGF binding and developmental expression suggest Flk-1 as a major regulator of vasculogenesis and angiogenesis. Cell 1993;72:835–846. Quinn TP, Peters KG, De Vries C, Ferrara N, Williams LT: Fetal liver kinase-1 is a receptor for vascular endothelial growth factor and is selectively expressed in vascular endothelium. Proc Natl Acad Sci USA 1993;90:7533–7537. Terman BI, Carrion ME, Kovacs E, Rasmussen BA, Eddy RL, Shows TB: Identiﬁcation of a new endothelial cell growth factor receptor tyrosine kinase. Oncogene 1991;6:1677–1683. Terman BI, Dougher-Vermazen M, Carrion ME, Dimitrov D, Armellino DC, Gospodarowicz D, Bohlen P: Identiﬁcation of the KDR tyrosine kinase as a receptor for vascular endothelial cell growth factor. Biochem Biophys Res Commun 1992;187:1579–1586. Pettersson A, Nagy JA, Brown LF, Sundberg C, Morgan E, Jungles S, Carter R, Krieger JE, Manseau EJ, Harvey VS, Eckelhoefer IA, Feng D, Dvorak AM, Mulligan RC, Dvorak HF: Heterogeneity of the angiogenic response induced in different normal adult tissues by vascular permeability factor/vascular endothelial growth factor. Lab Invest 2000;80:99–115. Dvorak AM, Miura T, Letourneau L, Ishizaka T, Kawakami T: Spontaneous degranulation of cultured bone-marrow-derived immature mast cells from X-linked immunodeﬁcient (Xid) mice. Int Arch Allergy Immunol 1996;111:118–125. Dvorak AM: Beige peritoneal rat mast cell undergoing anaphylactic degranulation. Blood 1997; 89: 4057. Login GR, Dvorak AM: Methods of microwave ﬁxation for microscopy. A review of research and clinical applications: 1970–1992. Prog Histochem Cytochem 1994;27:1–127. Login GR, Stavinoha WB, Dvorak AM: Ultrafast microwave energy ﬁxation for electron microscopy. J Histochem Cytochem 1986;34:381–387. Dvorak AM, Newball HH, Dvorak HF, Lichtenstein LM: Antigen-induced IgE-mediated degranulation of human basophils. Lab Invest 1980;43:126–139. Dvorak AM, Warner JA, Fox P, Lichtenstein LM, MacGlashan DW: Recovery of human basophils after FMLP-stimulated secretion. Clin Exp Allergy 1996;26:281–294. Dvorak AM, Galli SJ, Morgan E, Galli AS, Hammond ME, Dvorak HF: Anaphylactic degranulation of guinea pig basophilic leukocytes. II. Evidence for regranulation of mature basophils during recovery from degranulation in vitro. Lab Invest 1982;46:461–475. Costa JJ, Demetri GD, Harrist TJ, Dvorak AM, Hayes DF, Merica EA, Menchaca DM, Gringeri AJ, Schwartz LB, Galli SJ: Recombinant human stem cell factor (kit ligand) promotes human mast cell and melanocyte hyperplasia and functional activation in vivo. J Exp Med 1996;183:2681–2686.

References

337

409

410 411 412 413

414 415

416

417

418

419

420

421 422 423

424 425

426 427 428

429

Dvorak AM, Costa JJ, Monahan-Earley RA, Fox P, Galli SJ: Ultrastructural analysis of human skin biopsy specimens from patients receiving recombinant human stem cell factor: Subcutaneous injection of rhSCF induces dermal mast cell degranulation and granulocyte recruitment at the injection site. J Allergy Clin Immunol 1998;101:793–806. Besmer P: The kit ligand encoded at the murine Steel locus: A pleiotropic growth and differentiation factor. Curr Opin Cell Biol 1991;3:939–946. Galli SJ, Zsebo KM, Geissler EN: The kit ligand, stem cell factor. Adv Immunol 1994;55:1–96. Mekori YA, Oh CK, Metcalfe DD: IL-3-dependent murine mast cells undergo apoptosis on removal of IL-3. Prevention of apoptosis by c-kit ligand. J Immunol 1993;151:3775–3784. Yee NS, Paek I, Besmer P: Role of kit-ligand in proliferation and suppression of apoptosis in mast cells: Basis for radiosensitivity of White spotting and Steel mutant mice. J Exp Med 1994;179:1777– 1787. Iemura A, Tsai M, Ando A, Wershil BK, Galli SJ: The c-kit ligand, stem cell factor, promotes mast cell survival by suppressing apoptosis. Am J Pathol 1994;144:321–328. Gurish MF, Ghildyal N, McNeil HP, Austen KF, Gillis S, Stevens RL: Differential expression of secretory granule proteases in mouse mast cells exposed to interleukin-3 and c-kit ligand. J Exp Med 1992;175:1003–1012. Kirshenbaum AS, Goff JP, Kessler SW, Mican JM, Zsebo KM, Metcalfe DD: Effect of IL-3 and stem cell factor on the appearance of human basophils and mast cells from CD34+ pluripotent progenitor cells. J Immunol 1992;148:772–777. Valent P, Spanblöchl E, Sperr WR, Sillaber C, Zsebo KM, Agis H, Strobl H, Geissler K, Bettelheim P, Lechner K: Induction of differentiation of human mast cells from bone marrow and peripheral blood mononuclear cells by recombinant human stem cell factor/kit-ligand in long-term culture. Blood 1992; 80:2237–2245. Irani AM, Nilsson G, Miettinen U, Craig SS, Ashman LK, Ishizaka T, Zsebo KM, Schwartz LB: Recombinant human stem cell factor stimulates differentiation of mast cells from dispersed human fetal liver cells. Blood 1992;80:3009–3021. Wershil BK, Tsai M, Geissler EN, Zsebo KM, Galli SJ: The rat c-kit ligand, stem cell factor, induces c-kit receptor-dependent mouse mast cell activation in vivo. Evidence that signaling through the c-kit receptor can induce expression of cellular function. J Exp Med 1992;175:245–255. Columbo M, Horowitz EM, Botana LM, MacGlashan DW Jr, Bochner BS, Gillis S, Zsebo KM, Galli SJ, Lichtenstein LM: The human recombinant c-kit receptor ligand, rhSCF, induces mediator release from human cutaneous mast cells and enhances IgE-dependent mediator release from both skin mast cells and peripheral blood basophils. J Immunol 1992;149:599–608. Bischoff SC, Dahinden CA: c-kit ligand: a unique potentiator of mediator release by human lung mast cells. J Exp Med 1992;175:237–244. Gagari E, Tsai M, Lantz CS, Fox LG, Galli SJ: Differential release of mast cell interleukin-6 via c-kit. Blood 1997;89:2654–2663. Schwartz L, Huff T: Biology of mast cells and basophils; in Middleton E Jr, Reed CE, Ellis EF, Adkinson NF Jr, Yunginger JW, Busse WW (eds): Allergy: Principles and Practice. St Louis, Mosby, 1993, pp 135–168. Dvorak AM, Schleimer RP, Lichtenstein LM: Morphologic mast cell cycles. Cell Immunol 1987;105: 199–204. Dvorak AM, Schleimer RP, Lichtenstein LM: Human mast cells synthesize new granules during recovery from degranulation. In vitro studies with mast cells puriﬁed from human lungs. Blood 1988; 71:76–85. Dvorak AM, Mihm MC Jr, Osage JE, Dvorak HF: Melanoma. An ultrastructural study of the host inﬂammatory and vascular responses. J Invest Dermatol 1980;75:388–393. Dvorak AM: Ultrastructure of human mast cells. Int Arch Allergy Immunol 2002;127:100–105. Dvorak AM, Morgan ES: The case for extending storage and secretion functions of human mast cell granules to include synthesis; in Graumann W, Bendayan M, Bosman FT, et al (eds): Prog Histochem Cytochem. Jena, Urban & Fischer, 2002, vol 37, pp 229–320. Bernhard W: A new staining procedure for electron microscopical cytology. J Ultrastruct Res 1969; 27:250–265.

Ultrastructure of Mast Cells and Basophils

338

430 431 432 433 434 435 436 437 438 439 440 441 442

443 444 445 446 447 448

449 450 451 452

453 454 455 456

Biggiogera M, Fakan S: Fine structural speciﬁc visualization of RNA on ultrathin sections. J Histochem Cytochem 1998;46:389–395. Biggiogera M, Bottone MG, Pellicciari C: Nuclear RNA is extruded from apoptotic cells. J Histochem Cytochem 1998;46:999–1005. Monneron A, Bernhard W: Fine structural organization of the interphase nucleus in some mammalian cells. J Ultrastruct Res 1969;27:266–288. Fakan S, Bernhard W: Localisation of rapidly and slowly labelled nuclear RNA as visualized by highresolution autoradiography. Exp Cell Res 1971;67:129–141. Fakan S: Perichromatin ﬁbrils are in situ forms of nascent transcripts. Trends Cell Biol 1994; 4: 86– 90. Reynolds ES: The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J Cell Biol 1963;17:208–212. Fakan S: High-resolution autoradiography as a tool for the localization of nucleic acid synthesis and distribution in the mammalian cell nucleus. J Microsc 1976;106:159–171. Fakan S, Puvion E: The ultrastructural visualization of nucleolar and extranucleolar RNA synthesis and distribution. Int Rev Cytol 1980;65:255–299. Smetana K, Lejnar J, Vlastiborova A, Busch H: On interchromatinic dense granules of mature human neutrophil granulocytes. Exp Cell Res 1971;64:105–112. Amano M, LeBlond CP: Comparison of the speciﬁc activity time curves of ribonucleic acid in chromatin, nucleolus and cytoplasm. Exp Cell Res 1960;20:250–253. Scherrer K, Darnell JE: Sedimentation characteristics of rapidly labelled RNA from HeLa cells. Biochem Biophys Res Commun 1962;7:486–490. Soeiro R, Birnboim HC, Darnell JE: Rapidly labeled HeLa cell nuclear RNA. II. Base composition and cellular localization of a heterogeneous RNA fraction. J Mol Biol 1966;19:362–372. Warner JR, Soeiro R, Birnboim HC, Girard M, Darnell JE: Rapidly labeled HeLa cell nuclear RNA. I. Identiﬁcation by zone sedimentation of a heterogeneous fraction separate from ribosomal precursor RNA. J Mol Biol 1966;19:349–361. Liau MC, Perry RP: Ribosome precursor particles in nucleoli. J Cell Biol 1969;42:272–283. Fakan S, Bernhard W: Nuclear labelling after prolonged 3H-uridine incorporation as visualized by high resolution autoradiography. Exp Cell Res 1973;79:431–444. Gambetti P, Autilio-Gambetti L, Shafer B, Pfaff L: Quantitative autoradiographic study of labeled RNA in rabbit optic nerve after intraocular injection of [3H]-uridine. J Cell Biol 1973;59:677–684. Abelson HT, Johnson LF, Penman S, Green H: Changes in RNA in relation to growth of the ﬁbroblast: II. The lifetime of mRNA, rRNA, and tRNA in resting and growing cells. Cell 1974;1:161–165. Eliceiri GL: Short-lived, small RNAs in the cytoplasm of HeLa cells. Cell 1974;3:11–14. Angelier N, Bouteille M, Charret R, Curgy JJ, Delain E, Fakan S, Geuskens M, Guelin M, Lacroix JC, Laval M, Steinert G, Van Assel S: Detection of nucleic acids by means of ultrastructural autoradiography. J Microsc Biol Cell 1976;27:215–230. Fakan S, Puvion E, Spohr G: Localization and characterization of newly synthesized nuclear RNA in isolated rat hepatocytes. Exp Cell Res 1976;99:155–164. Herman R, Weymouth L, Penman S: Heterogeneous nuclear RNA-protein ﬁbers in chromatin-depleted nuclei. J Cell Biol 1978;78:663–674. Zieve G, Penman S: Subnuclear particles containing a small nuclear RNA and heterogeneous nuclear RNA. J Mol Biol 1981;145:501–523. Fey EG, Krochmalnic G, Penman S: The non-chromatin substructures of the nucleus: The ribonucleoprotein (RNP)-containing and RNP-depleted matrices analyzed by sequential fractionation and resinless section electron microscopy. J Cell Biol 1986;102:1654–1665. Ornelles DA, Fey EG, Penman S: Cytochalasin releases mRNA from the cytoskeletal framework and inhibits protein synthesis. Mol Cell Biol 1986;6:1650–1662. Davis L, Banker GA, Steward O: Selective dendritic transport of RNA in hippocampal neurons in culture. Nature 1987;330:477–479. Kopriwa B: A comparison of various procedures for ﬁne grain development in electron microscopic radioautography. Histochemistry 1975;44:201–224. Kopriwa BM: A reliable, standardized method for ultrastructural electron microscopic radioautography. Histochemie 1973;37:1–17.

References

339

457

458 459 460 461

462

463

464 465 466 467

468 469

470

471

472 473 474 475 476 477 478 479

Kopriwa BM, Levine GM, Nadler NJ: Assessment of resolution by half distance values for tritium and radioiodine in electron microscopic radioautographs using Ilford L4 emulsion developed by ‘solution physical’ or D-19b methods. Histochemistry 1984;80:519–522. Stevens BJ, Swift H: RNA transport from nucleus to cytoplasm in Chironomus salivary glands. J Cell Biol 1966;31:55–77. Mehlin H, Lönnroth A, Skoglund U, Daneholt B: Structure and transport of a speciﬁc premessenger RNP particle. Cell Biol Int Rep 1988;12:729–736. Mehlin H, Daneholt B, Skoglund U: Translocation of a speciﬁc premessenger ribonucleoprotein particle through the nuclear pore studied with electron microscope tomography. Cell 1992;69:605–613. Visa N, Izaurralde E, Ferreira J, Daneholt B, Mattaj IW: A nuclear cap-binding complex binds Balbiani ring pre-mRNA cotranscriptionally and accompanies the ribonucleoprotein particle during nuclear export. J Cell Biol 1996;133:5–14. Sun X, Alzhanova-Ericsson AT, Visa N, Aissouni Y, Zhao J, Daneholt B: The hrp23 protein in the Balbiani ring pre-mRNP particles is released just before or at the binding of the particles to the nuclear pore complex. J Cell Biol 1998;142:1181–1193. Penman S, Vesco C, Penman M: Localization and kinetics of formation of nuclear heterodisperse RNA, cytoplasmic heterodisperse RNA and polyribosome-associated messenger RNA in HeLa cells. J Mol Biol 1968;34:49–60. Lewin B: Units of transcription and translation: The relationship between heterogeneous nuclear RNA and messenger RNA. Cell 1975;4:11–20. Hinterberger M, Pettersson I, Steitz JA: Isolation of small nuclear ribonucleoproteins containing U1, U2, U4, U5, and U6 RNAs. J Biol Chem 1983;258:2604–2613. Padgett RA, Mount SM, Steitz JA, Sharp PA: Splicing of messenger RNA precursors is inhibited by antisera to small nuclear ribonucleoprotein. Cell 1983;35:101–107. Pettersson I, Hinterberger M, Mimori T, Gottlieb E, Steitz JA: The structure of mammalian small nuclear ribonucleoproteins. Identiﬁcation of multiple protein components reactive with anti(U1)ribonucleoprotein and anti-Sm autoantibodies. J Biol Chem 1984;259:5907–5914. Lamb MM, Laird CD: Increase in nuclear poly(A)-containing RNA at syncytial blastoderm in Drosophila melanogaster embryos. Dev Biol 1976;52:31–42. Capco DG, Jeffery WR: Differential distribution of poly(A)-containing RNA in the embryonic cells of Oncopeltus fasciatus. Analysis by in situ hybridization with a [3H]-poly(U) probe. Dev Biol 1978; 67:137–151. Binder M, Tourmente S, Roth J, Renaud M, Gehring WJ: In situ hybridization at the electron microscope level: Localization of transcripts on ultrathin sections of Lowicryl K4M-embedded tissue using biotinylated probes and protein A-gold complexes. J Cell Biol 1986;102:1646–1653. Jacob J, Todd K, Birnstiel ML, Bird A: Molecular hybridization of [3H]-labelled ribosomal RNA with DNA in ultrathin sections prepared for electron microscopy. Biochim Biophys Acta 1971;228: 761–766. Jacob J, Gillies K, MacLeod D, Jones KW: Molecular hybridization of mouse satellite DNA-complementary RNA in ultrathin sections prepared for electron microscopy. J Cell Sci 1974;14:253–261. Geuskens M, May E: Ultrastructural localization of SV40 viral DNA in cells, during lytic infection, by in situ molecular hybridization. Exp Cell Res 1974;87:175–185. Hutchison NJ, Langer-Safer PR, Ward DC, Hamkalo BA: In situ hybridization at the electron microscope level: Hybrid detection by autoradiography and colloidal gold. J Cell Biol 1982;95:609–618. Binder M: In situ hybridization at the electron microscope level. Scanning Microsc 1987; 1: 331– 338. Puvion-Dutilleul F, Puvion E: Ultrastructural localization of viral DNA in thin sections of herpes simplex virus type 1 infected cells by in situ hybridization. Eur J Cell Biol 1989;49:99–109. Singer RH, Langevin GL, Lawrence JB: Ultrastructural visualization of cytoskeletal mRNAs and their associated proteins using double-label in situ hybridization. J Cell Biol 1989;108:2343–2353. Thiry M, Thiry-Blaise L: In situ hybridization at the electron microscope level: An improved method for precise localization of ribosomal DNA and RNA. Eur J Cell Biol 1989;50:235–243. Visa N, Puvion-Dutilleul F, Bachellerie JP, Puvion E: Intranuclear distribution of U1 and U2 snRNAs visualized by high-resolution in situ hybridization: Revelation of a novel compartment containing U1 but not U2 snRNA in HeLa cells. Eur J Cell Biol 1993;60:308–321.

Ultrastructure of Mast Cells and Basophils

340

480

481

482

483 484 485

486 487

488

489 490 491

492

493

494 495

496 497 498 499 500 501 502

Bassell GJ, Powers CM, Taneja KL, Singer RH: Single mRNAs visualized by ultrastructural in situ hybridization are principally localized at actin ﬁlament intersections in ﬁbroblasts. J Cell Biol 1994; 126:863–876. Sibon OCM, Humbel BM, De Graaf A, Verkleij AJ, Cremers FFM: Ultrastructural localization of epidermal growth factor-receptor transcripts in the cell nucleus using pre-embedding in situ hybridization in combination with ultra-small gold probes and silver enhancement. Histochemistry 1994; 101:223–232. Macville MVE, Wiesmeijer KC, Dirks RW, Fransen JAM, Raap AK: Saponin pre-treatment in preembedding electron microscopic in situ hybridization for detection of speciﬁc RNA sequences in cultured cells: A methodological study. J Histochem Cytochem 1995;43:1005–1018. Yi J, Michel O, Sassy-Prigent C, Chevalier J: Electron microscopic location of mRNA in the rat kidney: Improved post-embedding in situ hybridization. J Histochem Cytochem 1995;43:801–809. Chevalier J, Yi J, Michel O, Tang XM: Biotin and digoxigenin as labels for light and electron microscopy in situ hybridization probes: Where do we stand? J Histochem Cytochem 1997;45:481–491. Punnonen EL, Fages C, Wartiovaara J, Rauvala H: Ultrastructural localization of ␤-actin and amphoterin mRNA in cultured cells: Application of tyramide signal ampliﬁcation and comparison of detection methods. J Histochem Cytochem 1999;47:99–112. Bendayan M, Zollinger M: Ultrastructural localization of antigenic sites on osmium-ﬁxed tissues applying the protein A-gold technique. J Histochem Cytochem 1983;31:101–109. Hiriyanna KT, Varkey J, Beer M, Benbow RM: Electron microscopic visualization of sites of nascent DNA synthesis by streptavidin-gold binding to biotinylated nucleotides incorporated in vivo. J Cell Biol 1988;107:33–44. Conner GE, Nelson D, Wisniewolski R, Lahita RG, Blobel G, Kunkel HG: Protein antigens of the RNA-protein complexes detected by anti-SM and anti-RNP antibodies found in serum of patients with systemic lupus erythematosus and related disorders. J Exp Med 1982;156:1475–1485. Tan EM: Autoantibodies in pathology and cell biology. Cell 1991;67:841–842. Galperin C, Coppel RL, Gershwin ME: Use of recombinant proteins in the diagnosis of autoimmune connective tissue diseases. Int Arch Allergy Immunol 1996;111:337–347. Steiner G, Skriner K, Smolen JS: Autoantibodies to the A/B proteins of the heterogeneous nuclear ribonucleoprotein complex: Novel tools for the diagnosis of rheumatic diseases. Int Arch Allergy Immunol 1996;111:314–319. Farris AD, Puvion-Dutilleul F, Puvion E, Harley JB, Lee LA: The ultrastructural localization of 60-kDa Ro protein and human cytoplasmic RNAs: Association with novel electron-dense bodies. Proc Natl Acad Sci USA 1997;94:3040–3045. Krämer A, Keller W, Appel B, Luhrmann R: The 5´ terminus of the RNA moiety of U1 small nuclear ribonucleoprotein particles is required for the splicing of messenger RNA precursors. Cell 1984; 38: 299–307. Elkon KB, Parnassa AP, Foster CL: Lupus autoantibodies target ribosomal P proteins. J Exp Med 1985;162:459–471. Elkon K, Skelly S, Parnassa A, Moller W, Danho W, Weissbach H, Brot N: Identiﬁcation and chemical synthesis of a ribosomal protein antigenic determinant in systemic lupus erythematosus. Proc Natl Acad Sci USA 1986;83:7419–7423. Busch H, Reddy R, Rothblum L, Choi YC: SnRNAs, SnRNPs, and RNA processing. Annu Rev Biochem 1982;51:617–654. Reddy R, Busch H: Small nuclear RNAs and RNA processing. Prog Nucleic Acid Res Mol Biol 1983; 30:127–162. Konarska MM, Sharp PA: Interactions between small nuclear ribonucleoprotein particles in formation of spliceosomes. Cell 1987;49:763–774. Maniatis T, Reed R: The role of small nuclear ribonucleoprotein particles in pre-mRNA splicing. Nature 1987;325:673–678. Reed R, Grifﬁth J, Maniatis T: Puriﬁcation and visualization of native spliceosomes. Cell 1988; 53: 949–961. Lobo SM, Hernandez N: A 7-bp mutation converts a human RNA polymerase II snRNA promoter into an RNA polymerase III promoter. Cell 1989;58:55–67. Fakan S, Leser G, Martin TE: Ultrastructural distribution of nuclear ribonucleoproteins as visualized by immunocytochemistry on thin sections. J Cell Biol 1984;98:358–363.

References

341

503 504 505 506 507

508 509 510 511 512 513 514 515 516

517

518 519 520 521

522 523 524 525 526 527 528 529 530

Puvion E, Viron A, Assens C, Leduc EH, Jeanteur P: Immunocytochemical identiﬁcation of nuclear structures containing snRNPs in isolated rat liver cells. J Ultrastruct Res 1984;87:180–189. Nyman U, Hallman H, Hadlaczky G, Pettersson I, Sharp G, Ringertz NR: Intranuclear localization of snRNP antigens. J Cell Biol 1986;102:137–144. St Clair EW: Anti-La antibodies. Rheum Dis Clin North Am 1992;18:359–376. Schur PH, Moroz LA, Kunkel HG: Precipitating antibodies to ribosomes in the serum of patients with systemic lupus erythematosus. Immunochemistry 1967;4:447–453. Hardin JA, Rahn DR, Shen C, Lerner MR, Wolin SL, Rosa MD, Steitz JA: Antibodies from patients with connective tissue diseases bind speciﬁc subsets of cellular RNA-protein particles. J Clin Invest 1982;70:141–147. White PJ, Billings PB, Hoch SO: Assays for the Sm and RNP autoantigens: The requirement for RNA and inﬂuence of the tissue source. J Immunol 1982;128:2751–2756. Mount SM, Pettersson I, Hinterberger M, Karmas A, Steitz JA: The U1 small nuclear RNA-protein complex selectively binds a 5´ splice site in vitro. Cell 1983;33:509–518. Ruby SW, Abelson J: An early hierarchic role of U1 small nuclear ribonucleoprotein in spliceosome assembly. Science 1988;242:1028–1035. Heinrichs V, Bach M, Winkelmann G, Lührmann R: U1-speciﬁc protein C needed for efﬁcient complex formation of U1 snRNP with a 5´ splice site. Science 1990;247:69–72. Kuo HC, Nasim FUH, Grabowski PJ: Control of alternative splicing by the differential binding of U1 small nuclear ribonucleoprotein particle. Science 1991;251:1045–1050. Eliceiri GL, Gurney T Jr: Subcellular location of precursors to small nuclear RNA species C and D and of newly synthesized 5S RNA in HeLa cells. Biochem Biophys Res Commun 1978;81:915–919. Patton JG, Wieben ED: U1 precursors: Variant 3´ ﬂanking sequences are transcribed in human cells. J Cell Biol 1987;104:175–182. Madore SJ, Wieben ED, Pederson T: Intracellular site of U1 small nuclear RNA processing and ribonucleoprotein assembly. J Cell Biol 1984;98:188–192. Hendrick JP, Wolin SL, Rinke J, Lerner MR, Steitz JA: Ro small cytoplasmic ribonucleoproteins are a subclass of La ribonucleoproteins: Further characterization of the Ro and La small ribonucleoproteins from uninfected mammalian cells. Mol Cell Biol 1981;1:1138–1149. Elkon KB, Culhane L: Partial immunochemical characterization of the Ro and La proteins using antibodies from patients with the sicca syndrome and lupus erythematosus. J Immunol 1984; 132: 2350–2356. Wolin SL, Steitz JA: The Ro small cytoplasmic ribonucleoproteins: Identiﬁcation of the antigenic protein and its binding site on the Ro RNAs. Proc Natl Acad Sci USA 1984;81:1996–2000. McNeilage LJ, Whittingham S, Jack I, Mackay IR: Molecular analysis of the RNA and protein components recognized by anti-La(SS-B) autoantibodies. Clin Exp Immunol 1985;62:685–695. Ben-Chetrit E, Chan EKL, Sullivan KF, Tan EM: A 52-kD protein is a novel component of the SS-A/Ro antigenic particle. J Exp Med 1988;167:1560–1571. Simons FHM, Pruijn GJM, van Venrooij WJ: Analysis of the intracellular localization and assembly of Ro ribonucleoprotein particles by microinjection into Xenopus laevis oocytes. J Cell Biol 1994;125: 981–988. Rinke J, Steitz JA: Precursor molecules of both human 5S ribosomal RNA and transfer RNAs are bound by a cellular protein reactive with anti-La lupus antibodies. Cell 1982;29:149–159. Bachmann M, Pfeifer K, Schröder HC, Müller WEG: The nucleocytoplasmic shuttling of the La antigen in CV-1 cells. Mol Biol Rep 1987;12:239–240. Yoo CJ, Wolin SL: The yeast La protein is required for the 3´ endonucleolytic cleavage that matures tRNA precursors. Cell 1997;89:393–402. Sobel SG, Wolin SL: Two yeast La motif-containing proteins are RNA binding proteins that associate with polyribosomes. Mol Biol Cell 1999;10:3849–3862. Perry RP: RNA processing comes of age. J Cell Biol 1981;91:28S–38S. Mattaj IW: Splicing stories and poly(A) tales: An update on RNA processing and transport. Curr Opin Cell Biol 1990;2:528–538. Gray MW, Cedergren R: The new age of RNA. FASEB J 1993;7:4–6. Gottlieb E: Messenger RNA transport and localization. Curr Opin Cell Biol 1990;2:1080–1086. Singer RH: The cytoskeleton and mRNA localization. Curr Opin Cell Biol 1992;4:15–19.

Ultrastructure of Mast Cells and Basophils

342

531

532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560

Singer RH: RNA zipcodes for cytoplasmic addresses. Intracellular localization of mRNAs appears to be determined by sequences in their 3´ untranslated regions that are composed of multiple elements. Curr Biol 1993;3:719–721. Wilhelm JE, Vale RD: RNA on the move: The mRNA localization pathway. J Cell Biol 1993; 123: 269–274. Bassell G, Singer RH: mRNA and cytoskeletal ﬁlaments. Curr Opin Cell Biol 1997;9:109–115. Bassell GJ, Oleynikov Y, Singer RH: The travels of mRNAs through all cells large and small. FASEB J 1999;13:447–454. Etkin LD, Lipshitz HD: RNA localization. FASEB J 1999;13:419–420. Jansen RP: RNA-cytoskeletal associations. FASEB J 1999;13:455–466. Lasko P: RNA sorting in Drosophila oocytes and embryos. FASEB J 1999;13:421–433. Mowry KL, Cote CA: RNA sorting in Xenopus oocytes and embryos. FASEB J 1999;13:435–445. Lenk R, Ransom L, Kaufmann Y, Penman S: A cytoskeletal structure with associated polyribosomes obtained from HeLa cells. Cell 1977;10:67–78. Fulton AB, Wan KM, Penman S: The spatial distribution of polyribosomes in 3T3 cells and the associated assembly of proteins into the skeletal framework. Cell 1980;20:849–857. Isaacs WB, Fulton AB: Cotranslational assembly of myosin heavy chain in developing cultured skeletal muscle. Proc Natl Acad Sci USA 1987;84:6174–6178. Bassell GJ: High resolution distribution of mRNA within the cytoskeleton. J Cell Biochem 1993;52: 127–133. Hill MA, Schedlich L, Gunning P: Serum-induced signal transduction determines the peripheral location of ␤-actin mRNA within the cell. J Cell Biol 1994;126:1221–1230. Kislauskis EH, Zhu X, Singer RH: Sequences responsible for intracellular localization of ␤-actin messenger RNA also affect cell phenotype. J Cell Biol 1994;127:441–451. Dreyfuss G, Philipson L, Mattaj IW: Ribonucleoprotein particles in cellular processes. J Cell Biol 1988;106:1419–1425. Knowles RB, Sabry JH, Martone ME, Deerinck TJ, Ellisman MH, Bassell GJ, Kosik KS: Translocation of RNA granules in living neurons. J Neurosci 1996;16:7812–7820. Green MR: Pre-mRNA processing and mRNA nuclear export. Curr Opin Cell Biol 1989; 1: 519– 525. Izaurralde E, Mattaj IW: Transport of RNA between nucleus and cytoplasm. Semin Cell Biol 1992;3: 279–288. Zapp ML: RNA nucleocytoplasmic transport. Semin Cell Biol 1992;3:289–297. Rings EHHM, Büller HA, Neele AM, Dekker J: Protein sorting versus messenger RNA sorting? Eur J Cell Biol 1994;63:161–171. Curtis D, Lehmann R, Zamore PD: Translational regulation in development. Cell 1995; 81: 171– 178. St Johnston D: The intracellular localization of messenger RNAs. Cell 1995;81:161–170. Schnapp BJ, Arn EA, Deshler JO, Highett MI: RNA localization in Xenopus oocytes. Semin Cell Dev Biol 1997;8:529–540. Hazelrigg T: The destinies and destinations of RNAs. Cell 1998;95:451–460. Oleynikov Y, Singer RH: RNA localization: Different zipcodes, same postman? Trends Cell Biol 1998; 8:381–383. Steward O, Levy WB: Preferential localization of polyribosomes under the base of dendritic spines in granule cells of the dentate gyrus. J Neurosci 1982;2:284–291. Jeffery WR, Tomlinson CR, Brodeur RD: Localization of actin messenger RNA during early ascidian development. Dev Biol 1983;99:408–417. Lawrence JB, Singer RH: Intracellular localization of messenger RNAs for cytoskeletal proteins. Cell 1986;45:407–415. Dix DJ, Eisenberg BR: Myosin mRNA accumulation and myoﬁbrillogenesis at the myotendinous junction of stretched muscle ﬁbers. J Cell Biol 1990;111:1885–1894. Jirikowski GF, Sanna PP, Bloom FE: mRNA coding for oxytocin is present in axons of the hypothalamo-neurohypophysial tract. Proc Natl Acad Sci USA 1990;87:7400–7404.

References

343

561

562

563 564 565

566 567 568 569 570

Ainger K, Avossa D, Morgan F, Hill SJ, Barry C, Barbarese E, Carson JH: Transport and localization of exogenous myelin basic protein mRNA microinjected into oligodendrocytes. J Cell Biol 1993;123: 431–441. Hamill D, Davis J, Drawbridge J, Suprenant KA: Polyribosome targeting to microtubules: Enrichment of speciﬁc mRNAs in a reconstituted microtubule preparation from sea urchin embryos. J Cell Biol 1994;127:973–984. Seydoux G, Fire A: Soma-germline asymmetry in the distributions of embryonic RNAs in Caenorhabditis elegans. Development 1994;120:2823–2834. Kawasaki I, Shim YH, Kirchner J, Kaminker J, Wood WB, Strome S: PGL-1, a predicted RNA-binding component of germ granules, is essential for fertility in C. elegans. Cell 1998;94:635–645. Tian Q, Streuli M, Saito H, Schlossman SF, Anderson P: A polyadenylate binding protein localized to the granules of cytolytic lymphocytes induces DNA fragmentation in target cells. Cell 1991; 67: 629–639. Merlie JP, Sanes JR: Concentration of acetylcholine receptor mRNA in synaptic regions of adult muscle ﬁbres. Nature 1985;317:66–68. Garner CC, Tucker RP, Matus A: Selective localization of messenger RNA for cytoskeletal protein MAP2 in dendrites. Nature 1988;336:674–677. Jirikowski GF, Sanna PP, Maciejewski-Lenoir D, Bloom FE: Reversal of diabetes insipidus in Brattleboro rats: Intrahypothalamic injection of vasopressin mRNA. Science 1992;255:996–998. Martone ME, Pollock JA, Jones YZ, Ellisman MH: Ultrastructural localization of dendritic messenger RNA in adult rat hippocampus. J Neurosci 1996;16:7437–7446. Peters A, Palay SL, de F Webster H: The Fine Structure of the Nervous System: The Neurons and Supporting Cells. Philadelphia, Saunders, 1976.

Ultrastructure of Mast Cells and Basophils

344

U U U U U U U U U U U U U U U U U U U U U U U U U U U

Subject Index

Anaphylactic degranulation (AND) human basophils induction 206, 208, 209 recovery 209, 212, 214, 215, 217, 219 human mast cells recovery 235–238 ultrastructure labeling 219, 222–224, 226, 227, 229, 231 mouse mast cells 205, 206 phenotypes 166, 171 rat mast cells 206 ultrastructural features 1, 2, 8, 9, 11, 56 Anionic ferritin (FE), electron-dense ultrastructural tracer analysis of vesiculo-vacuolar organelles 191, 192 Arachidonic acid (AA), human mast cell lipid body localization with ultrastructural autoradiography 69, 70 Autoimmune disease, immunogold staining of RNA-associated protein antigens in lipid bodies and secretory granules 298, 301, 303, 307, 309 Autoradiography, see Ultrastructural autoradiography

Basic ﬁbroblast growth factor (bFGF) human mast cell lipid body localization 77 piecemeal degranulation studies in human mast cells 153 secretory granule localization 78, 80, 82 Bullous pemphigoid, human mast cell piecemeal degranulation 238 C/EBP, knockout mice 55, 56 Charcot-Leyden crystals (CLCs) eosinophil protein 91 human basophil expression overview 14, 15 recovering human basophil expression 215, 217 human mast cell expression 34, 39 immunogold staining of protein 91, 94, 95 protein vesicular transport in piecemeal degranulation comparison of Charcot-Leyden crystal protein and histamine transport 175, 176, 179 human basophil studies 173, 175 modeling 179, 180, 182, 184

Chondroitin sulfate (CS), ribonucleasegold imaging guinea pig basophil granules 125, 126 human basophil granules 126, 128 mouse mast cell granules 130 rat mast cell granules 128, 130 Chymase, immunogold staining 89, 91 c-Kit ligand, see Mast cell growth factor; Stem cell factor mouse basophil expression 47, 51, 55 Completely degranulated basophils (CDBs), human basophil phenotype 162, 166, 172 Cutaneous basophil hypersensitivity (CBH) basophil inﬁltration of tissues 141, 145 delayed hypersensitivity comparison 138, 140 discovery 316 lymphocyte roles 140, 141 Cyclooxygenase, human mast cell lipid body localization with immunogold staining 70, 71 Delayed hypersensitivity (DH), features 138, 140 Development circulating mast cell precursor in embryonic mice 39, 42 human basophils 11–15, 17 human mast cells 17, 18, 20, 22, 24, 25, 32, 34, 39 Diamine oxidase-gold, histamine imaging cytochemical controls 101, 103 human mast cell secretory granules 100 inﬂammatory eye disease, vessels 197 optimization 103–105, 112–114 overview 99, 100 Enzyme-afﬁnity-gold technique diamine oxidase-gold imaging of histamine cytochemical controls 101, 103 human mast cell secretory granule labeling 100

Subject Index

optimization 103–105, 112–114 overview 99, 100 overview 98 ribonuclease-gold imaging of chondroitin sulfate guinea pig basophil granules 125, 126 human basophil granules 126, 128 mouse mast cell granules 130 rat mast cell granules 128, 130 ribonuclease-gold imaging of heparin agar block staining 124 cytochemical controls 116, 117 human mast cells 114, 115, 118 optimization 118–120, 122, 124 overview 114, 115 rat mast cell granules 128, 130 FcÂR human mast cell expression 34 mouse basophil expression 42, 46, 47, 51, 55, 56 mouse mast cell expression 56 Fibroblast growth factor-2, see Basic ﬁbroblast growth factor f-Met peptide (FMLP), piecemeal degranulation induction in human basophils 156, 158 Granules, see Secretory granules Guinea pig basophil, ribonuclease-gold imaging of chondroitin sulfate in granules 125, 126 Heparin human mast cell anaphylactic degranulation studies 231, 235–238 ribonuclease-gold imaging agar block staining 124 cytochemical controls 116, 117 human mast cell imaging 114, 115, 118 optimization 118–120, 122, 124 overview 114, 115 rat mast cell granules 128, 130

346

Histamine diamine oxidase-gold imaging cytochemical controls 101, 103 human mast cell secretory granule labeling 100 optimization 103–105, 112–114 overview 99, 100 human mast cell anaphylactic degranulation studies 224, 226, 227, 229, 231, 235 immunogold staining 95, 96 inﬂammation role 95 overview of imaging techniques 99 piecemeal degranulation in human mast cells 148, 149 recovering human basophil expression 217, 219 vesicular transport in piecemeal degranulation comparison of Charcot-Leyden crystal protein and histamine transport 175, 176, 179 human basophil studies 175 modeling 179, 180, 182, 184 vesiculo-vacuolar organelle permeabilization 198, 200 Histamine-releasing factor (HRF), piecemeal degranulation induction in human basophils 160 Horseradish peroxidase (HRP), electron-dense ultrastructural tracer analysis of vesiculo-vacuolar organelles 190, 191 Human basophil anaphylactic degranulation induction 206, 208, 209 recovery 209, 212, 214, 215, 217, 219 Charcot-Leyden crystals, immunogold staining of protein 91, 94, 95 development 11–15, 17 piecemeal degranulation completely degranulated basophils 162, 166, 172 granule and vesicle ultrastructure phenotypes 166, 171, 172 recovering basophils 166, 172 stimulants

Subject Index

f-Met peptide 156, 158 histamine-releasing factor 160 monocyte chemotactic protein-1 160 tetradecanoyl phorbol acetate 158, 160 total vesicle number 172 ultrastructural morphology of stimulated cells 162, 164 vesicular transport Charcot-Leyden crystal protein transport 173, 175 comparison of Charcot-Leyden crystal protein and histamine transport 175, 176, 179 histamine transport 175 modeling 179, 180, 182, 184 ribonuclease-gold imaging of chondroitin sulfate in granules 126, 128 ultrastructural veriﬁcation in cell samples from organs and ﬂuids 62, 64 Human mast cell anaphylactic degranulation recovery 235–238 ultrastructure labeling 219, 222–224, 226, 227, 229, 231 development 17, 18, 20, 22, 24, 25, 32, 34, 39 diamine oxidase-gold imaging of histamine cytochemical controls 101, 103 human mast cell secretory granule labeling 100 optimization 103–105, 112–114 overview 99, 100 lipid bodies, see Lipid bodies piecemeal degranulation studies ex vivo 153 in vivo 145, 148, 149, 153 ribonuclease-gold imaging of heparin agar block staining 124 cytochemical controls 116, 117 human mast cell imaging 114, 115, 118 optimization 118–120, 122, 124 overview 114, 115

347

secretory granule association with ribosomes biological signiﬁcance of RNA storage and processing 309, 310 in situ hybridization detection of messenger RNA 279, 280, 282, 288, 290, 296, 307 overview 252, 253 ribonuclear protein staining with regressive EDTA staining 257, 258 RNA-associated protein antigen immunogold staining 298, 301, 303, 307, 309 uridine immunogold staining of RNA 275, 277, 279, 307 uridine ultrastructural autoradiography for RNA detection 262, 264, 266, 268, 269, 271–273, 307 tissue sources 62, 64 ultrastructural veriﬁcation in cell samples from organs and ﬂuids 62, 64 Immunogold staining Charcot-Leyden crystals 91, 94, 95 chymase 89, 91 cyclooxygenase in human mast cell lipid bodies 70, 71 histamine 95, 96 uridine in lipid bodies and secretory granules 275, 277, 279, 307 Inﬂammatory bowel disease (IBD), piecemeal degranulation studies in human mast cells 145, 148 In situ hybridization (ISH), messenger RNA detection in human mast cell lipid bodies and secretory granules 279, 280, 282, 288, 290, 296, 307 Interleukins cell culture 12–14, 51 piecemeal degranulation studies of mouse mast cells 153, 154 La, immunogold staining of RNA-associated protein antigens in human mast cell lipid bodies and granules 298, 301, 303

Subject Index

Lipid bodies arachidonic acid localization with ultrastructural autoradiography 69, 70 basic ﬁbroblast growth factor localization 77 characteristics 68, 69 cyclooxygenase localization with immunogold electron microscopy 70, 71 ribosome association in human mast cells biological signiﬁcance of RNA storage and processing 309, 310 in situ hybridization detection of messenger RNA 279, 280, 282, 288, 290, 296, 307 overview 252, 253 ribonuclear protein staining with regressive EDTA staining 257, 258 ribonuclease-gold labeling of RNA 262, 305 RNA-associated protein autoantigen immunogold staining 298, 301, 303, 307, 309 uridine immunogold staining of RNA 275, 277, 279, 307 uridine ultrastructural autoradiography for RNA detection 262, 264, 266, 268, 269, 271–273, 307 tumor necrosis factor-␣ localization 73, 74, 76, 77 Lymphocyte factor A (LFA), cell culture 12, 13 Lymphocyte factor B (LFB), cell culture 12, 13 Mast cell growth factor (MCGF), cell culture 11, 34 Melanoma, human mast cell piecemeal degranulation 238 Messenger RNA, see RNA Monkey basophil, ultrastructural identiﬁcation 56, 62 Monocyte chemotactic protein-1 (MCP-1), piecemeal degranulation induction in human basophils 160

348

Mouse basophil lymphocyte comparison 42, 46, 47, 51, 54, 55 neutrophil comparison 55, 56 ultrastructural identiﬁcation 42, 46, 47, 51, 54, 55 Mouse mast cell anaphylactic degranulation 205, 206 circulating precursors in embryonic mice 39, 42 piecemeal degranulation studies in vivo 153, 154 ribonuclease-gold imaging of chondroitin sulfate in granules 130 Piecemeal degranulation (PMD) discovery 139, 316 diseases 137, 238 human basophils completely degranulated basophils 162, 166, 172 granule and vesicle ultrastructure phenotypes 166, 171, 172 recovering basophils 166, 172 stimulants f-Met peptide 156, 158 histamine-releasing factor 160 monocyte chemotactic protein-1 160 tetradecanoyl phorbol acetate 158, 160 total vesicle number 172 ultrastructural morphology of stimulated cells 162, 164 overview 135 stages 166 ultrastructural features 1, 2, 6, 7, 12, 56, 135 vesicular transport of loaded vesicles human basophil studies Charcot-Leyden crystal protein transport 173, 175 comparison of Charcot-Leyden crystal protein and histamine transport 175, 176, 179 histamine transport 175 modeling 179, 180, 182, 184

Subject Index

human mast cell studies ex vivo 153 in vivo 145, 148, 149, 153 mouse mast cell studies in vivo 153, 154 overview 138 rat mast cell studies ex vivo 154–156 Rabbit basophil, ribonuclease-gold imaging of proteoglycans 128 Rat mast cell anaphylactic degranulation 206 chymase immunogold staining 89, 91 histamine immunogold staining 95, 96 piecemeal degranulation studies ex vivo 154–156 ribonuclease-gold imaging of chondroitin sulfate in granules 128, 130 Recovering basophils (RBs), human basophil phenotype 166, 172 Regressive EDTA staining, ribonuclear proteins in lipid bodies and secretory granules 257, 258 Ribonuclear proteins (RNPs) immunogold staining in human mast cell lipid bodies and granules 298, 300, 301, 303 regressive EDTA staining in lipid bodies and secretory granules 257, 258 Ribonuclease-gold (R-G) chondroitin sulfate imaging guinea pig basophil granules 125, 126 human basophil granules 126, 128 mouse mast cell granules 130 rat mast cell granules 128, 130 heparin imaging agar block staining 124 cytochemical controls 116, 117 human mast cells 114, 115, 118 optimization 118–120, 122, 124 overview 114, 115 rat mast cell granules 128, 130 proteoglycan detection in rabbit basophils 128 RNA labeling in lipid bodies 262, 305

349

Ribosomal P antigen, immunogold staining of RNA-associated protein antigens in human mast cell lipid bodies and granules 298, 301 RNA, detection in lipid bodies and secretory granules biological signiﬁcance of RNA storage and processing 309, 310 in situ hybridization detection of messenger RNA 279, 280, 282, 288, 290, 296, 307 ribonuclear protein staining with regressive EDTA staining 257, 258 ribonuclease-gold labeling 262, 305 uridine immunogold staining 275, 277, 279, 307 uridine ultrastructural autoradiography 262, 264, 266, 268, 269, 271–273, 307 Ro, immunogold staining of RNA-associated protein antigens in human mast cell lipid bodies and granules 298, 301, 303 Secretory granules basic ﬁbroblast growth factor localization 78, 80, 82 Charcot-Leyden crystal protein, see Charcot-Leyden crystals chymase imaging, see Chymase enzyme-afﬁnity-gold staining, see Enzyme-afﬁnity-gold technique histamine imaging, see Histamine overview of imaging techniques 77 ribosome association in human mast cells autoantibody immunogold staining of RNA-associated protein antigens 298, 301, 303, 307, 309 biological signiﬁcance of RNA storage and processing 309, 310 in situ hybridization detection of messenger RNA 279, 280, 282, 288, 290, 296, 307 overview 252, 253 ribonuclear protein staining with regressive EDTA staining 257, 258 uridine immunogold staining of RNA 275, 277, 279, 307

Subject Index

uridine ultrastructural autoradiography for RNA detection 262, 264, 266, 268, 269, 271–273, 307 tumor necrosis factor-· localization 82, 84 Serotonin, vesiculo-vacuolar organelle permeabilization 198 Sm, immunogold staining of RNA-associated protein antigens in human mast cell lipid bodies and granules 298, 301 Stem cell factor (SCF), see also Mast cell growth factor human mast cell anaphylactic degranulation promotion 222–224 kit ligand 32 mouse injection effects 32 recombinant human stem cell factor for cell culture 14 Tetradecanoyl phorbol acetate (TPA), piecemeal degranulation induction in human basophils 158, 160 Tumor necrosis factor-· (TNF-·), localization in human mast cell lipid bodies 73, 74, 76, 77 localization in secretory granules 82, 84 piecemeal degranulation studies of rat mast cells 154, 155 Ultrastructural autoradiography arachidonic acid localization in lipid bodies 69, 70 uridine ultrastructural autoradiography for RNA detection 262, 264, 266, 268, 269, 271–273, 307 Uridine, RNA detection in lipid bodies and secretory granules immunogold staining 275, 277, 279, 307 ultrastructural autoradiography 262, 264, 266, 268, 269, 271–273, 307 Vascular endothelial growth factor, see Vascular permeability factor/vascular endothelial growth factor

350

Vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) discovery 192 tumor vessel leakage via vesiculovacuolar organelles 193, 194 vascular permeabilization potency 192, 193 vesiculo-vacuolar organelle permeabilization 198, 200, 202, 203 Vesicular transport, see Piecemeal degranulation Vesiculo-vacuolar organelle (VVO) formation 188, 189 inﬂammatory eye disease vessels 196, 197 overview of features 186, 188, 317 permeability inducers 198 receptors and ligands histamine 200 vascular permeability factor/ vascular endothelial growth factor 200, 202, 203 stomatal diameters 198 transendothelial cell pathway studies electron-dense ultrastructural tracer analysis 190–192 serial section analysis 190 tumor vessels 192–195

Subject Index

351