Postgraduate Haematology Companion website This book has a companion website: www.wiley.com/go/hoffbrand/ph7 with: ∙ Figures and tables from the book for downloading Postgraduate Haematology EDITED BY A Victor Hoffbrand MA, DM, FRCP, FRCPath, FRCP (Edin), DSc, FMedSci Emeritus Professor of Haematology, University College London, London, UK Douglas R Higgs MD, FRCP, FRS Professor of Molecular Haematology & Director, Weatherall Institute of Molecular Medicine, John Radcliffe Hospital, Oxford, UK David M Keeling BSc, MD, FRCP, FRCPath Consultant Haematologist, Oxford Haemophilia and Thrombosis Centre, Oxford University Hospitals, Oxford, UK Atul B Mehta MA, MD, FRCP, FRCPath Consultant Haematologist, Lysosomal Storage Disorders Unit, Department of Haematology, Royal Free Hospital, London, UK Seventh Edition This edition first published 2016 © 2016, 2011, 2005 by John Wiley & Sons Ltd First published as Tutorials in Postgraduate Haematology © William Heinemann Ltd 1972 Second edition 1981 published © Butterworth Ltd Third edition 1989 published © Butterworth Ltd Fourth edition 1999 published © Butterworth-Heinmann Ltd Registered office: John Wiley & Sons, Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK Editorial offices: 9600 Garsington Road, Oxford, OX4 2DQ, UK The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK 111 River Street, Hoboken, NJ 07030-5774, USA For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell The right of the author to be identified as the author of this work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. The contents of this work are intended to further general scientific research, understanding, and discussion only and are not intended and should not be relied upon as recommending or promoting a specific method, diagnosis, or treatment by health science practitioners for any particular patient. The publisher and the author make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of fitness for a particular purpose. In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of medicines, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each medicine, equipment, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. Readers should consult with a specialist where appropriate. The fact that an organization or Website is referred to in this work as a citation and/or a potential source of further information does not mean that the author or the publisher endorses the information the organization or Website may provide or recommendations it may make. Further, readers should be aware that Internet Websites listed in this work may have changed or disappeared between when this work was written and when it is read. No warranty may be created or extended by any promotional statements for this work. Neither the publisher nor the author shall be liable for any damages arising herefrom. Library of Congress Cataloging-in-Publication Data Postgraduate haematology / edited by A. Victor Hoffbrand, Douglas R. Higgs, David M. Keeling, Atul B. Mehta. – Seventh edition. p. ; cm. Includes bibliographical references and index. ISBN 978-1-118-85432-7 (cloth) I. Hoffbrand, A. V., editor. II. Higgs, Douglas R., editor. III. Keeling, David (David Michael), editor. IV. Mehta, Atul B., editor. [DNLM: 1. Blood Physiological Phenomena. 2. Hematologic Diseases. 3. Leukemia–blood. 4. Thrombosis–blood. WH 100] RC633 616.1′ 5–dc23 2015019099 A catalogue record for this book is available from the British Library. Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Cover image: GettyImages-464401418 by Frentusha Set in 9.5/12pt MinionPro by Aptara Inc., New Delhi, India 1 2016 Contents Contributor list, vii 13 Clinical blood transfusion, 214 Shubha Allard, Marcela Contreras Preface to the seventh edition, x 14 Phagocytes, 246 Preface to the first edition, xi John Mascarenhas, Marina Kremyanskaya, 1 Stem cells and haemopoiesis, 1 Ronald Hoffman Emma de Pater, Elaine Dzierzak 15 Lysosomal storage disorders, 270 2 Erythropoiesis, 11 Atul B Mehta, Derralynn A Hughes Douglas R Higgs, Noémi Roy, Deborah Hay 16 Normal lymphocytes and non-neoplastic lymphocyte disorders, 278 3 Iron metabolism, iron deficiency and disorders of Paul Moss, Mark Drayson haem synthesis, 21 Clara Camaschella, A Victor Hoffbrand, Chaim 17 The spleen, 303 Hershko Paul Moss 4 Iron overload, 40 18 The molecular basis of haematological Clara Camaschella, A Victor Hoffbrand, Maria malignancies, 314 Domenica Cappellini Niccolo Bolli, George Vassiliou 19 Laboratory diagnosis of haematological 5 Megaloblastic anaemia, 53 neoplasms, 332 A Victor Hoffbrand Torsten Haferlach, Barbara J Bain 6 Haemoglobin and the inherited disorders of globin 20 Acute myeloid leukaemia, 352 synthesis, 72 Alan K Burnett, David Grimwade Swee Lay Thein, David Rees 21 Adult acute lymphoblastic Llukaemia, 371 7 Sickle cell disease, 98 Clare J Rowntree, Adele K Fielding Anne Marsh, Elliott P Vichinsky 22 Childhood acute lymphoblastic leukaemia, 384 8 Hereditary disorders of the red cell membrane and Ajay Vora disorders of red cell metabolism, 114 Paola Bianchi, Narla Mohandas 23 Supportive care in the management of leukaemia, 399 Eliza Gil, Vanya Gant, Panagiotis Kottaridis 9 Acquired haemolytic anaemias, 138 Modupe O Elebute, Rachel Kesse-Adu 24 Chronic myeloid leukaemia, 419 David TO Yeung, Timothy P Hughes 10 Inherited aplastic anaemia/bone marrow failure syndromes, 156 25 The myelodysplastic syndromes, 438 Inderjeet S Dokal Kavita Raj, Ghulam J Mufti 11 Acquired aplastic anaemia and paroxysmal 26 Myeloproliferative neoplasms, 474 nocturnal haemoglobinuria, 174 Peter J Campbell, Claire Harrison, Judith CW Marsh, Austin G Kulasekararaj, Anthony R Green Neal S Young, Peter Hillmen 27 Chronic lymphocytic leukaemia and other chronic 12 Red cell immunohaematology, 195 B-cell disorders, 500 Geoff Daniels, Marcela Contreras, Shubha Allard Emili Montserrat, Peter Hillmen v Contents 28 T-cell lymphoproliferative disorders, 524 41 Congenital platelet disorders, 761 Pier Luigi Zinzani, Alessandro Broccoli Maurizio Margaglione, Paul RJ Ames 29 Multiple myeloma, 537 42 Primary immune thrombocytopenia, 773 Jes ús San-Miguel, Joan Bladé Drew Provan, Adrian C Newland 30 Amyloidosis, 562 43 Thrombotic thrombocytopenic purpura and Simon DJ Gibbs, Philip N Hawkins haemolytic–uraemic syndrome (congenital and acquired), 783 31 The classification of lymphomas: updating the WHO Pier Mannuccio Mannucci, Flora Peyvandi, classification, 575 Roberta Palla Elias Campo, Stefano A Pileri 44 Heritable thrombophilia, 795 32 Hodgkin lymphoma, 601 Trevor Baglin, David Keeling Piers Blombery, David Linch 45 Acquired venous thrombosis, 809 33 Non-Hodgkin lymphoma: low grade, 614 Beverley J Hunt, Henry G Watson William Townsend, Robert Marcus 46 Antithrombotic agents, 820 34 Non-Hodgkin lymphoma: high grade, 631 Trevor Baglin, David Keeling Jessica Okosun, Kate Cwynarski 47 Management of venous thromboembolism, 830 35 Stem cell transplantation, 651 Trevor Baglin, David Keeling Charles Craddock, Ronjon Chakraverty 48 Haematological aspects of systemic disease, 838 36 Normal haemostasis, 676 A Victor Hoffbrand, Atul B Mehta Keith Gomez, John H McVey 49 Haematological aspects of tropical diseases, 854 37 The vascular function of platelets, 699 Imelda Bates, Ivy Ekem Stephen P Watson, Neil V Morgan, Paul Harrison 50 Neonatal haematology, 870 38 Haemophilia and Von Willebrand disease, 715 Irene Roberts, Subarna Chakravorty Michael A Laffan, K John Pasi 51 WHO Classification: Tumours of the 39 Rare inherited coagulation disorders, 733 Haematopoietic and Lymphoid Tissues (2008), 885 Flora Peyvandi, Marzia Menegatti 40 Acquired coagulation disorders, 743 Peter W Collins, Jecko Thachil, Cheng-Hock Toh Index, 888 vi Contributor list Shubha Allard Alan K Burnett Geoff Daniels Barts Health NHS Trust and NHS Blood and Professor, Institute of Cancer and Genetics, Cardiff International Blood Group Reference Laboratory, Transplant, London, UK University, Cardiff, UK NHS Blood and Transplant, Bristol, UK Paul R Ames Clara Camaschella Emma de Pater Department of Haematology, Haemostasis and Vita-Salute University, Milan, Italy Erasmus Stem Cell Institute and Department of Cell Thrombosis, St George’s Hospital, London, UK Biology, Erasmus Medical Center, Rotterdam, Netherlands Peter J Campbell Trevor Baglin Head, Cancer Genetics and Genomics at the Cambridge University Hospitals NHS Trust, Institute, Wellcome Trust Sanger Institute, Inderjeet S Dokal Addenbrookes Hospital, Cambridge, UK Cambridge, UK Chair of Child Health and Honorary Consultant in Haematology, Barts and The London School of Barbara J Bain Elias Campo Medicine and Dentistry, Queen Mary University of London, Barts Health NHS Trust, London, UK Professor in Diagnostic Haematology, St Mary’s Hospital Clinic, University of Barcelona, Barcelona, Hospital Campus of Imperial College Faculty of Spain Medicine, London and Honorary Consultant Maria Domenica Cappellini Haematologist, St Mary’s Hospital, London, UK Foundation IRCCS Ca’ Granda Policlinico and Ronjon Chakraverty DISSCO University of Milan, Milan, Italy Professor of Haematology and Cellular Imelda Bates Immunotherapy, University College London, Professor of Tropical Haematology, Liverpool School London, UK Mark Drayson of Tropical Medicine, Liverpool, UK Professor Clinical Immunodiagnostics, Director, Clinical Immunology Service, Honorary Consultant Subarna Chakravorty Paola Bianchi Department of Paediatrics, Imperial College University Hospitals Birmingham and Heart of England, Birmingham, UK Fondazione IRCCS Ca’ Granda Ospedale Maggiore Healthcare and Imperial College London, London, Policlinico Milano, Oncohaematology Unit, UK Physiopathology of Anaemias Unit, Milan, Italy Elaine Dzierzak Professor of Cell Biology, Erasmus Stem Cell Peter W Collins Joan Bladé Professor, Cardiff Institute of Infection & Immunity, Institute, Erasmus Medical Centre, Rotterdam, Netherlands and Centre for Inflammation Research, Hospital Clinic de Barcelona, Institut School of Medicine, Cardiff University, University University of Edinburgh, Edinburgh, UK d‘Investigacions Biomediques August Pi I Ferrer Hospital of Wales, Cardiff, UK (IDIBAPS), Barcelona, Spain Ivy Ekem Marcela Contreras Piers Blombery Professor of Transfusion Medicine, Royal Free and Department of Haematology, University of Ghana and Korle Bu Teaching Hospital, Accra, Ghana UCL Cancer Institute, School of Life and Medical University Medical Schools, London, UK Sciences, University College London, London, UK Modupe O Elebute Charles Craddock Niccolo Bolli Professor and Consultant Haematologist, Centre for Honorary Consultant, Department of Haematology, King’s College Hospital, London, UK Division of Hematology, Fondazione IRCCS Istituto Clinical Haematology, Queen Elizabeth Hospital, Nazionale dei Tumori, University of Milan, Milan, Birmingham, UK Italy Adele K Fielding Reader in Haematology, Cancer Institute, University Kate Cwynarski Alessandro Broccoli Consultant Haematologist and Honorary Senior College London, London, UK Institute of Haematology ‘L. e A. Seràgnoli’, Lecturer (UCL), Department of Haematology, Royal University of Bologna, Bologna, Italy Free Hampstead NHS Trust, London, UK vii Contributor list Vanya Gant Chaim Hershko MD Austin G Kulasekararaj University College London Hospitals NHS Trust, Department of Medicine, Shaare Zedek Medical Department of Haematology, King’s College London, UK Center; Professor Emeritus, Hebrew U Hadassah Hospital/King’s College London, London, UK Medical School, Jerusalem, Israel Simon DJ Gibbs Michael A Laffan National Amyloidosis Centre, Royal Free and Douglas R Higgs Centre for Haematology, Imperial College School of University College London Medical School, London, Professor of Molecular Haematology and Director, Medicine, Imperial College, Hammersmith Hospital, UK Weatherall Institute of Molecular Medicine, John London, UK Radcliffe Hospital, Oxford, UK Eliza Gil Swee Lay Thein University College London Hospitals NHS Trust, Peter Hillmen King’s College London, Molecular Haematology, London, UK Professor of Experimental Haematology and Faculty of Life Sciences and Medicine, and Honorary Consultant Haematologist, Leeds Department of Haematological Medicine, King’s Keith Gomez Teaching Hospitals NHS Trust, Leeds, UK College Hospital NHS Foundation Trust, London, UK Haemophilia Centre and Thrombosis Unit, Royal Free London NHS Foundation Trust, London, UK A Victor Hoffbrand Emeritus Professor of Haematology, University David Linch Anthony R Green College London, London, UK Professor of Clinical Haematology, UCL Cancer Institute, School of Life and Medical Sciences, Professor of Haemato-Oncology, University of University College London, London, UK Cambridge; Departments’ of Haematology and Ronald Hoffman Oncology, Cambridge University Hospitals NHS Tisch Cancer Institute, Icahn School of Medicine at Foundation Trust; Cambridge Institute for Medical Mount Sinai, New York, New York, USA Pier Mannuccio Mannucci Research, Wellcome Trust/MRC Stem Cell Institute, A Bianchi Bonomi Hemophilia and Thrombosis and Department of Haematology, University of Center, IRCCS Cà Granada, Ospedale Maggiore, Cambridge, Cambridge, UK Derralynn A Hughes Milan, Italy Senior Lecturer and Honorary Consultant Haematologist, RFH Lysosomal Storage Disorders David Grimwade Unit, Royal Free Hospital, London, UK Robert Marcus Professor of Molecular Haematology, Department of Consultant Haematologist, King’s College Hospital Medical and Molecular Genetics, King’s College NHS Foundation Trust, and Department of Clinical London, UK and Honorary Consultant Timothy P Hughes Haematology, King’s College Hospital, Denmark Haematologist, Guy’s and St Thomas’ NHS Haematologist and Head of Department (RAH site), Hill, London, UK Foundation Trust, London, UK Department of Haematology, SA Pathology, Adelaide; Clinical Professor, Discipline of Medicine, University of Adelaide and South Australian Health Maurizio Margaglione Torsten Haferlach and Medical Research Institute, Adelaide, Australia Medical Genetics, Department of Clinical and MLL Munich Leukemia Laboratory, Munich, Experimental Medicine, University of Foggia, Italy Germany Beverley J Hunt Professor of Thrombosis and Haemostasis and Anne Marsh Claire Harrison Consultant Haematologist, Guy’s and St Thomas’ Department of Hematology/Oncology, UCSF Professor and Consultant Haematologist, Foundation Trust, London, UK Benioff Children’s Hospital Oakland, Oakland, CA, Department of Haematology, Guy’s and St Thomas’ USA NHS Foundation Trust, London, UK David Keeling Consultant Haematologist, Oxford Haemophilia and Judith CW Marsh Paul Harrison Thrombosis Centre, Oxford University Hospitals, Professor, King’s College London, and Department Senior Lecturer, School of Immunity and Infection, Oxford, UK of Haematological Medicine, King’s College College of Medical and Dental Sciences, University Hospital, London, UK of Birmingham, Birmingham, UK Rachel Kesse-Adu Consultant, Department of Haematology, Guy’s and John Mascarenhas Philip N Hawkins St Thomas’ Hospital, London, UK Tisch Cancer Institute, Icahn School of Medicine at National Amyloidosis Centre, Royal Free and Mount Sinai, New York, New York, USA University College London Medical School, Centre for Amyloidosis and Acute Phase Proteins, London, Panagiotis Kottaridis UK Department of Haematology, University College John H McVey London, London, UK Professor of Cardiovascular Biology, School of Biosciences and Medicine, University of Surrey, Deborah Hay Guildford, UK Weatherall Institute of Molecular Medicine, John Marina Kremyanskaya Radcliffe Hospital, Oxford, UK Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New York, New York, USA viii Contributor list Atul B Mehta K John Pasi Cheng-Hock Toh Consultant Haematologist, Lysosomal Storage Professor and Consultant Haematologist, Barts and Professor of Haematology, Roald Dahl Centre, Royal Disorders Unit, Department of Haematology, Royal The London School of Medicine and Dentistry, Liverpool University Hospital, Liverpool, UK Free Hospital, London, UK Royal London Hospital London, UK William Townsend Marzia Menegatti Flora Peyvandi University College Hospitals NHS Foundation Trust A Bianchi Bonomi Hemophilia and Thrombosis A Bianchi Bonomi Hemophilia and Thrombosis and King’s College London, London, UK Center, Fondazione IRCCS Ca’ Granda Ospedale Center, IRCCS Cà Granada, Ospedale Maggiore, Maggiore Policlinico, Università degli Studi di Milan, Italy Milano, Milan, Italy George Vassiliou Haematological Cancer Genetics, Wellcome Trust Stefano A Pileri Sanger Institute Hinxton, Cambridge, UK Narla Mohandas Professor of Pathology, Bologna University School of Laboratory of Red Cell Physiology, New York Blood Medicine, and Director of the Haematopathology Center, New York, NY, USA Unit, European Institute of Oncology, Bologna, Italy Elliott P Vichinsky Medical Director, Hematology/Oncology, UCSF Benioff Children’s Hospital Oakland, Oakland, CA, Emili Montserrat Drew Provan and Professor of Pediatrics, University of California Institute of Haematology and Oncology, Hospital Centre for Haematology, Institute of Cell and San Francisco, CA, USA Clinic, University of Barcelona, Barcelona, Spain Molecular Science, Queen Mary University of London, London, UK Ajay Vora Neil V Morgan Consultant Paediatric Haematologist and Honorary Lecturer in Cardiovascular Genetics, Cardiovascular Kavita Raj Professor of Haematology (University of Sheffield), and Respiratory Sciences, School of Clinical and Professor of Haematological Oncology, King’s Department of Haematology, Sheffield Children’s Experimental Medicine, College of Medical and College London, London, UK Hospital, Sheffield, UK Dental Sciences, University of Birmingham, Birmingham, UK David Rees Stephen P Watson Consultant Paediatric Haematologist, King’s College Professor in Cardiovascular Sciences and Cellular Paul Moss London/King’s College Hospital, Department of Pharmacology, Centre for Cardiovascular Sciences, Professor and Head, School of Cancer Sciences, Haematological Medicine, King’s College Hospital, Institute for Biomedical Research, College of University of Birmingham, and Queen Elizabeth Denmark Hill, London, UK Medical and Dental Sciences, University of Hospital, Birmingham, UK Birmingham, Birmingham, UK Irene Roberts Ghulam J Mufti Professor of Paediatric Haematology, Department of Henry G Watson Department of Haematological Medicine, King’s Paediatrics and Weatherall Institute of Molecular Consultant Haematologist, Aberdeen Royal College Hospital, UK Medicine, University of Oxford, Oxford, UK Infirmary, Aberdeen, UK Adrian C Newland Clare J Rowntree David TO Yeung Professor of Haematology, Academic Haematology Department of Haematology, University Hospital of Haematologist, Department of Haematology, SA Unit, Blizard Institute; Barts and The London School Wales, Cardiff, UK Pathology, Adelaide; Clinical Associate Lecturer, of Medicine and Dentistry, Queen Mary University Discipline of Medicine, University of Adelaide, of London, London, UK Adelaide, Australia Noémi Roy Academic Clinical Lecturer, Weatherall Institute of Jessica Okosun Molecular Medicine, John Radcliffe Hospital, Neal S Young Centre for Haemato-Oncology, Barts Cancer Oxford, UK National Heart, Lung and Blood Institute, National Institute, University of London, London, UK Institutes of Health, Bethesda, MD, USA Jesús San-Miguel Roberta Palla Clinica Universidad de Navarra, Centro de Pier Luigi Zinzani Department of Pathophysiology and Investigación Medica Aplicada (CIMA), Pamplona, Professor, Institute of Haematology ‘L. e A. Transplantation, Università degli Studi di Milano, A Spain Seràgnoli’, University of Bologna, Bologna, Italy Bianchi Bonomi Hemophilia and Thrombosis Center, Milan, Italy Jecko Thachil Department of Haematology, Manchester Royal Infirmary, Manchester, UK ix Preface to the seventh edition Since the sixth edition of Postgraduate Haematology was pub- with myelofibrosis. Other advances in therapy include many lished in 2011, substantial advances have been made in our new monoclonal antibodies used for treating Hodgkin and non- understanding of the pathogenesis of inherited and acquired Hodgkin lymphomas, and new immunomodulatory and protea- haematological diseases. This progress has largely resulted from some inhibitory drugs that are increasing life expectancy in mul- the application of next generation sequencing of the relevant tiple myeloma. The more widespread use of orally active, direct exomes and genomes to identify the DNA mutations responsible inhibitors of coagulation and of the orally active iron chelating for these diseases. For example, mutation of the myeloid differ- drugs are also having a major impact on patient care. entiation primary response gene (MYD88) has been found in The seventh edition of Postgraduate Haematology reflects over 90% of cases of Waldenstrom’s macroglobulinaemia; muta- these exciting developments in the diagnosis and treatment tion of calreticulin has been found in most of the JAK2 negative of blood diseases, with revised text, new scientific diagrams cases of essential thrombocythaemia and primary myelofibro- and tables. Douglas Higgs, David Keeling and Atul Mehta sis; and multiple driver mutations have been shown to underlie have formed an Editorial team with the original Editor, Victor myelodysplasia and acute myeloid leukaemia revealing the com- Hoffbrand, and many new and previous authors have con- plexity of these diseases and the wide individual variation that tributed superb, up to date, well-illustrated, chapters. We thank is relevant to their treatment and prognosis. It seems likely that most warmly Danny Catovsky, Edward Tuddenham and Tony understanding the genetic complexity of hamatological malig- Green for their major contribution as Editors of previous edi- nancies will play an increasingly important role in providing per- tions. We also thank Claire Bonnett, Rob Blundell and Tom Bates sonalised treatment for specific tumours. of Wiley Blackwell who have been responsible for the publishing These advances have been accompanied by the introduc- process throughout the preparation of this edition and have been tion of new, effective, targeted therapies, based on the knowl- unstinting in their support, patience and professional expertise. edge that has been gained of the key signalling pathways on Thanks also to Kathy Syplywczak who project managed this edi- which the malignant cells depend for their proliferation and tion, and we are also grateful once again to Jane Fallows for her survival. For example, inhibitors of the B-cell receptor sig- superb art work and scientific diagrams. nalling pathway have proved life saving in patients with chronic lymphocytic leukaemia resistant to other therapies, and JAK2 AVH, DRH, DK, ABM inhibitors are extending survival and quality of life in patients London and Oxford x Preface to the first edition In this book the authors combine an account of the physiological nal of the Royal College of Physicians of London and the and biochemical basis of haematological processes with descrip- Quarterly Journal of Medicine for permission to reproduce tions of the clinical and laboratory features and management of Figures 4.1, 4.5, 4.10, 4.11, 4.12, 9.4 and 9.10, also the pub- blood disorders. Within this framework, each author has dealt lishers of Progress in Haematology for Figure 7.2, and many with the individual subjects as he or she thought appropriate. other publishers who, together with the authors, have been Because this book is intended to provide a foundation for the acknowledged in the text. We are particularly grateful to Pro- study of haematology and is not intended to be a reference book, fessor JV Dacie for providing material which formed the basis it reflects, to some extent, the views of the individual authors of many of the original illustrations in Chapters 4–8. We are rather than providing comprehensive detail and a full bibliog- greatly indebted to Mrs T Charalambos, Mrs J Cope and Mrs raphy. For these the reader is referred to the selected reading D Haysome for secretarial assistance and to Mrs P Schilling and given at the end of each chapter. It is hoped that the book will the Department of Medical Illustration for photomicrography, prove of particular value to students taking either the Primary or art work and general photography. the Final Part of the examination for Membership of the Royal Finally, we are grateful for the invaluable help and forbearance College of Pathologists and the Diplomas of Clinical Pathology. we have received from Mr R Emery and William Heinemann It should also prove useful to physicians wishing to gain spe- Medical Books. cial knowledge of haematology and to technicians taking the Advanced Diploma in Haematology of the Institute of Medi- London, 1972 cal Laboratory Technology, or the Higher National Certificate AVH in Medical Laboratory subjects. SML We wish to acknowledge kind permission from the editors and publishers of the British Journal of Haematology, the Jour- xi CHAPTER 1 Stem cells and haemopoiesis Emma de Pater and Elaine Dzierzak Erasmus Stem Cell Institute, Erasmus Medical Centre, Rotterdam, Netherlands and University of Edinburgh, Centre for Inflammation Research, UK 1 Introduction Hierarchical organization and lineage relationships in the adult haemopoietic Haemopoietic stem cells (HSCs) are the foundation of the adult system blood system and sustain the lifelong production of all blood lineages. These rare cells are generally defined by their abil- The haemopoietic system is the best-characterized cell lineage ity to self-renew through a process of asymmetric cell division, differentiation hierarchy and, as such, has set the paradigm for the outcome of which is an HSC and a differentiating cell. In the growth and differentiation of tissue-specific stem cells. HSCs health, HSCs provide homeostatic maintenance of the system are defined by their high proliferative potential, ability to self- through their ability to differentiate and generate the hundreds renew and potential to give rise to all haemopoietic lineages. of millions of erythrocytes and leucocytes needed each day. In HSCs produce immature progenitors that gradually and pro- trauma and physiological stress, HSCs ensure the replacement gressively, through a series of proliferation and differentiation of the lost or damaged blood cells. The tight regulation of HSC events, become restricted in lineage differentiation potential. self-renewal ensures the appropriate balance of blood cell pro- Such restricted progenitors produce the terminally differenti- duction. Perturbation of this regulation and unchecked growth ated functional blood cells. of HSCs and/or immature blood cells results in leukaemia. The lineage relationships of the variety of cells within the Over the last 50 years, great success has been achieved with adult haemopoietic hierarchy (Figure 1.1) are based on results of bone marrow transplantation as a stem cell regenerative ther- in vivo transplantation assays in irradiated/myeloablated recip- apy. However, insufficient numbers of HSCs are still a major con- ient mice and many in vitro differentiation assays that became straint in clinical applications. As the pivotal cells in this essen- available following the identification of haemopoietic growth tial tissue, HSCs are the focus of intense research to: (1) fur- factors. These assays facilitated measurement of the matura- ther our understanding of their normal behaviour and the basis tional progression of stem cells and progenitors, at or near the of their dysfunction in haemopoietic disease and leukaemia branch points of lineage commitment. Clonal analyses, in the and (2) provide insights for new strategies for improved and form of colony-forming unit (CFU) assays or single cell trans- patient-specific stem cell therapies. This chapter provides cur- plantation assays, were developed to define the lineage differen- rent and historical information on the organization of the adult tiation potential of the stem cell or progenitor, and to quantitate haemopoietic cell differentiation hierarchy, the ontogeny of the number/frequency of such cells in the population as a whole. HSCs, the stromal microenvironment supporting these cells, In general, the rarer a progenitor is and the greater its lineage dif- and the molecular mechanisms involved in the regulation ferentiation potential, the closer it is in the hierarchy to the HSC. of HSCs. In vitro clonogenic assays measure the most immature Postgraduate Haematology, Seventh Edition. Edited by A Victor Hoffbrand, Douglas R Higgs, David M Keeling and Atul B Mehta. © 2016 John Wiley & Sons, Ltd. Published 2016 by John Wiley & Sons, Ltd. 1 Postgraduate Haematology B pre-B B lymphocyte CLP T pre-T T lymphocyte HSC CFU-S MPP G-CFC Granulocyte GMP GM- CFC Figure 1.1 The adult haemopoietic GM-CFC Macrophage hierarchy. Haemopoietic stem cells are at the foundation of the hierarchy. Through CMP a series of progressive proliferation and differentiation steps the mature blood cell BFU-E CFU-E Erythrocytes lineages are produced. Haemopoietic Platelets stem cells have the greatest proliferative MEP and multilineage differentiation potential, while the mature blood cells are Meg- Megakaryocyte not proliferative and are lineage CFC restricted. While large numbers of mature cells are found in the blood and turn over Proliferative potential rapidly, the bone marrow contains long-lived quiescent haemopoietic stem Frequency and turnover rate cells at a very low frequency. progenitor CFU-GEMM/Mix (granulocyte, erythroid, of the long-term repopulating HSCs to self-renew. The clonal macrophage, megakaryocyte), bipotent progenitors CFU-GM nature of engraftment and the multilineage potential of HSCs (granulocyte, macrophage) and restricted progenitors CFU-M has been demonstrated through radiation, retroviral and bar- (macrophage), CFU-G (granulocyte), CFU-E (erythroid) and code marking of bone marrow cells. Such studies suggest that, BFU-E (burst-forming unit-erythroid). While such in vitro at steady state, several HSC clones contribute to the haemopoi- clonogenic assays measure myeloid and erythroid potential, etic system at any one time. Further analyses of bone marrow lymphoid potential is revealed only in fetal thymic organ HSCs show that this compartment consists of a limited num- cultures and stromal cell cocultures in which the appropriate ber of distinct HSC subsets, each with predictable behaviours, microenvironment and growth factors are present. Long-term as described by their repopulation kinetics in myeloablated adult culture assays (6–8 week duration), such as the cobblestone- recipients. In general, the bone marrow haemopoietic cell com- area-forming cell (CAFC) and the long-term culture-initiating partment, as measured by in vitro clonogenic assays and in vivo cell (LTC-IC) assays, reveal the most immature of haemopoietic transplantation assays, shows a progression along the adult dif- progenitors. Currently, the major hurdle in studies and clinical ferentiation hierarchy from HSCs to progenitors and fully func- applications of HSCs is the fact that HSCs cannot be expanded tional blood cells with decreased multipotency and proliferative and are poorly maintained in culture. The only way to detect a potential. bona fide HSC is in vivo. The use of flow cytometry to enrich for HSCs and the var- In vivo, the heterogeneity of the bone marrow population of ious progenitors in adult bone marrow has been instrumen- immature progenitors and HSCs is reflected in the time peri- tal in refining precursor–progeny relationships in the adult ods at which different clones contribute to haemopoiesis. Short- haemopoietic hierarchy. HSCs are characteristically small ‘blast’ term in vivo repopulating haemopoietic progenitor cells such cells, with a relatively low forward and side light scatter and as CFU-S (spleen) give rise to macroscopic erythro-myeloid low metabolic activity. Both mouse and human HSCs are neg- colonies on the spleen within 14 days of injection. Bona fide ative for expression of mature haemopoietic lineage cell-surface HSCs give rise to the long-term high-level engraftment of all markers, such as those found on B lymphoid cells (CD19, haemopoietic lineages. Serial transplantations reveal the ability B220), T lymphoid cells (CD4, CD8, CD3), macrophages (CD15, 2 Chapter 1 Stem cells and haemopoiesis Mac-1) and granulocytes (Gr-1). Positive selection for mouse Endothelial HSCs relies on expression of Sca-1, c-kit, endoglin and CD150 niche markers and for human HSCs on expression of CD34, c-kit, IL- 6R, Thy-1 and CD45RA markers. Similarly, cell types at lineage branch points have been identified, including the CMP (com- mon myeloid progenitor), CLP (common lymphoid progeni- Expand Differentiate Apoptosis tor) and GMP (granulocyte macrophage progenitor). Recently, Migrate using the Flt3 receptor tyrosine kinase surface marker along with many other well-studied markers, the LMPP (lymphoid primed multipotent progenitor) has been identified within the lineage negative, Sca-1 positive, c-kit positive (LSK) enriched fraction of HSCs. These cells have granulocyte/macrophage, B lymphoid and T lymphoid potential, but little or no megakary- Self-renew HSC ocyte/erythroid potential. This suggests that the first lineage dif- ferentiation event is not a strict separation into common lym- phoid and myeloid pathways. While these cell-surface marker changes and functional restriction events are represented by dis- Endosteal niche crete cells in the working model of the haemopoietic hierarchy as depicted in textbooks and Figure 1.1, it is most likely that there is a continuum of cells between these landmarks and/or alternative Figure 1.2 The bone marrow haemopoietic niches. Haemopoietic differentiation paths. The currently identified progenitor cells in stem cells are found in the endosteal and endothelial niches of the the hierarchy represent the cells present at stable and detectable bone marrow. These niches support the maintenance, self-renewal, frequencies and for which we currently have markers and func- expansion, differentiation, migration and survival of haemopoietic tional assays. As more cell-surface markers are identified and the stem cells through local growth factor production and cell–cell sensitivity of detection is increased, additional intermediate cell interactions. subsets are likely to be identified. Together with single cell tran- scriptomic approaches, it may be possible to predict the molec- ular events needed for the HSC state and the differentiation of HSCs are also found in the mouse spleen at approximately a the entire haemopoietic system. 10-fold lower frequency and in the circulating blood at a 100- fold lower frequency. The capacity for HSCs to migrate and also be retained in bone marrow supportive niches is of rele- Sites of adult haemopoiesis vance to clinical transplantation therapies. HSCs injected intra- venously in such therapies must find their way to the bone Bone marrow, spleen, thymus and lymph nodes are the marrow for survival and effective haemopoietic engraftment. haemopoietic sites in the adult, and each tissue plays a special For example, stromal-derived factor (SDF)-1 and its recep- role in supporting the growth and differentiation of particular tor CXCR4 (expressed on HSCs) are implicated in the move- haemopoietic cell lineages and subsets. Equally important is the ment of HSCs and the retention of HSCs in the bone marrow. blood itself, which is a mobile haemopoietic tissue, with mature Indeed, HSC mobilization can be induced through AMD3100, blood cells travelling through the circulation to function in all an antagonist of SDF-1, and by the administration of G-CSF. parts of the body. Not only do the terminally differentiated cells, Mobilization strategies with G-CSF are used routinely to stim- such as erythrocytes and lymphocytes, move by means of the cir- ulate bone marrow HSCs to enter the circulation, allowing ease culation, but HSCs (at low frequency) also migrate through the of collection in the blood rather than through bone marrow circulation from the bone marrow to other haemopoietic tissues. biopsy. HSCs are mostly concentrated in the bone marrow and are found in the endosteal and vascular niches (Figure 1.2). HSCs can be induced to circulate by administration of granulocyte colony- stimulating factor (G-CSF). Recent improvements in confocal Development of HSCs microscopy have allowed the visualization of the migration of Waves of haemopoietic generation in circulating HSCs to the bone marrow endosteal niche by time- embryonic development lapse imaging in the mouse. The estimated frequency of HSCs is 1 per 104 –105 mouse bone Until the mid-1960s it was thought that blood cells were marrow cells and 1 per 20 × 106 human bone marrow cells. intrinsically generated in tissues such as the liver, spleen, bone 3 Postgraduate Haematology marrow and thymus. Survival studies in which cells from Definitive HSCs un-irradiated tissues were injected into lethally irradiated mice showed that it was the bone marrow that contains the potent cells responsible for rescue from haemopoietic failure. Later, Long-lived adult haemopoiesis through clonal marking studies, it was demonstrated that the Transient embryonic bone marrow harbours HSCs during the adult stages of life. But where, when and how are HSCs generated during ontogeny? haemopoiesis In the 1970s, examination of mouse embryo tissues suggested that adult haemopoietic cells are generated in the yolk sac, migrate and colonize initially the fetal liver and subsequently the bone marrow, where they reside throughout adult life. Primitive However, studies in non-mammalian vertebrate models (avian Progenitors and amphibian) demonstrated that the aorta region in the body of the embryo generates the long-lived adult blood Figure 1.4 Waves of haemopoietic cell emergence during system, while the yolk sac (or equivalent tissue) produces the embryonic stages. The earliest haemopoietic cells are produced transient embryonic haemopoietic system. In agreement with during the first wave of haemopoietic fate determination. The these studies, the aorta–gonad–mesonephros (AGM) region of onset of this wave occurs in the yolk sac blood islands and mammalian embryos was later found to generate the first HSCs produces transient primitive erythroid cells. This wave continues of the permanent adult blood system. with the production of erythroid-myeloid progenitors in the The development of the haemopoietic system is complex. As absence of bona fide haemopoietic stem cells. True long-lived a growing organism, the embryo itself needs rapid haemopoiesis self-renewing definitive haemopoietic stem cells (adult to allow it to thrive before the adult system is generated. Thus, repopulating stem cells) are generated in the second wave of a simple transient haemopoietic system is generated at early haemopoietic cell emergence in the AGM region. In this wave, embryonic stages to rapidly produce primitive erythroid and haemogenic endothelial cells bud into the aortic lumen as these myeloid cells. In the yolk sac, both haemopoietic and endothelial cells take on haemopoietic stem cell fate. cells are simultaneously generated from a common mesodermal precursor cell, the haemangioblast (Figure 1.3). Thereafter, other haemopoietic progenitor and differentiated cell types are istics (longevity and self-renewability) of the adult haemopoietic generated in both the yolk sac and the intraembryonic AGM system. However, some early yolk sac progenitors provide long- region to create an intermediate haemopoietic system. These lived tissue resident macrophages, such as the glial cells in the progenitors and differentiated cell types arise from a specialized brain. The independent and distinct waves of haemopoiesis that population of endothelial cells that have haemogenic potential supply the embryo and adult are likely derived from different (haemogenic endothelial cells). At both these early times in subsets of mesodermal precursor cells (Figure 1.4). ontogeny, the mouse embryo contains no HSCs. Hence, in the The adult system has its foundation in a cohort of initiating absence of HSCs, the embryo generates a haemopoietic system HSCs. The first adult HSCs are autonomously generated in the that is short-lived and lacks the important qualitative character- mouse AGM at E10.5 and in the human AGM beginning at week 4 of gestation. Recently, the process of HSC generation has been visualized in real time in the mouse embryo. This remarkable Haemangioblast Haemogenic demonstration confirms that HSCs are derived via a transdif- endothelium ferentiation event in which specialized endothelial cells lining the aorta bud into the lumen to form round cells with HSC fate (Figure 1.5), and shows that haemopoietic development is con- served between mammalian and non-mammalian species. The emerging mouse aortic HSCs are characterized by the loss of cell-surface markers for endothelium, such as Flk-1 and VE- Haemopoietic Endothelial Haemopoietic cadherin, and the gain of expression of haemopoietic markers CD41 and CD45 and HSC markers Sca1, c-kit and endoglin. The Figure 1.3 Precursors to haemopoietic cells in embryonic stages. emerging aortic HSCs are as functionally potent as bone marrow The mesodermal precursor to haemopoietic and endothelial HSCs, since these sorted cells can form a complete long-term lineages at early stages of development is the haemangioblast. haemopoietic system and self-renewing HSCs after transplanta- Later, haemogenic endothelial cells are the precursors to tion into irradiated adult recipient mice. haemopoietic stem cells and progenitor cells. Both precursors Lineage tracing experiments in the mouse embryo have indi- appear to exist during a short window of developmental time. cated that the adult haemopoietic system is generated during a 4 Chapter 1 Stem cells and haemopoiesis Haemopoietic A recent study revealed the presence and generation of HSCs cluster in the E10.5/E11 head vasculature through lineage marking. It remains uncertain whether the placenta (or the yolk sac) can Aorta generate HSCs de novo since there is no method at present by which cells can be uniquely marked in these developing tis- sues. Nonetheless, quantitative studies in which HSC numbers in each of these tissues was determined suggest that the AGM cannot generate all the HSCs that are found in the fetal liver Inductive Gonad Mesonephros microenvironment (a tissue that harbours haemopoietic cells but does not gen- erate them) and later in the adult bone marrow (Figure 1.6). Figure 1.5 Schematic diagram of the aorta–gonad–mesonephros Since the placenta at mid-gestation contains an abundance of (AGM) region and haemopoietic cell clusters emerging from the HSCs, it is possible that this highly vascularized tissue generates dorsal aorta. The haemopoietic stem-cell-inductive HSCs from haemogenic endothelium and/or that the placenta microenvironment is localized in the ventral aspect of the aorta. is a highly supportive and proliferative microenvironment for Tissues ventral to the AGM, such as the gut and mesenchyme AGM-derived HSCs. provide HSC-inducing signals, whereas dorsal tissues such as the The development of the haemopoietic system in the human notochord and the neural tube suppress HSC induction. conceptus closely parallels that in the mouse conceptus. Like the mouse placenta, the developing human placenta contains HSCs. Already at week 6 of gestation HSCs can be detected, short window of development, spanning E9–E12. Using Cre-lox as analysed by in vivo xenotransplantation into immunodefi- recombination (temporally and cell-lineage controlled) to mark cient mice; also, haemopoietic progenitors are found at these endothelial cells in the mid-gestation embryo, it was found that early stages. Phenotypic characterization shows that HSCs and almost all the blood cells in the circulation and haemopoietic tis- progenitors are in both the CD34-positive and CD34-negative sues of the adult mice were derived from VE-cadherin express- fractions at week 6 of gestation and are exclusively in the CD34- ing cells. Moreover, these cells require the Runx1 transcription positive fraction by week 19. These cells are in close association factor, as demonstrated by Runx1 conditional deletion in this with the placental vasculature. The placenta may be considered mouse model. Other lineage tracing experiments marking the a source of haemopoietic progenitors and HSCs in addition to earliest cells expressing the Runx1 and SCL transcription factor genes, showed that the progeny contributed to the bone mar- row cells in the adult. Thus, the progeny of haemogenic endothe- lial cells in the major vasculature of the embryo contribute to a Yolk sac cohort of adult bone marrow HSCs that form the foundation of haemopoiesis throughout adult life. Embryonic haemopoietic sites and haemopoietic migration AGM The AGM and yolk sac are not the only sites where haemopoi- Fetal liver etic cells are found in the early conceptus. The placenta is a highly haemopoietic tissue and much like the early-stage yolk sac, the mouse placenta can produce erythro-myeloid progeni- tors. Embryos deficient for the Ncx1 gene lack a heartbeat and Bone marrow circulation, and thus were used to study the origins of early Placenta haemopoietic progenitors. Ncx1 deficient embryos were shown to contain erythro-myeloid progenitors in the yolk sac and pla- Figure 1.6 Haemopoietic sites during development. The first centa, demonstrating that these haemopoietic progenitors are haemopoietic stem cells arise in the AGM region. Other generated by these tissues. Unfortunately, the embryos die before haemopoietic cells and progenitors are generated in the yolk sac the onset of HSC generation at mid-gestation, precluding anal- and placenta. It is as yet undetermined whether the yolk sac and ysis of HSC production in the yolk sac and placenta. In nor- placenta can generate haemopoietic stem cells. Haemopoietic cells mal embryos where the circulation is established between the generated in these three tissues migrate and colonize the fetal liver. embryo body and the extraembryonic tissues at E8.25, HSCs Subsequently, the long-lived haemopoietic cells (primarily the are detected in the placenta and yolk sac only beginning at E11, haemopoietic stem cells) migrate and colonize the bone marrow, subsequent to the first HSC generation in the AGM at E10.5. where they reside in the adult stages of life. 5 Postgraduate Haematology umbilical cord blood for preclinical studies and potential clinical lymphoid output. HSCs with a predominant myeloid output can therapies. also be found in prenatal life. Thus, the myeloid type HSC is not unique to aging – it is the prevalence to maintain these HSCs that is. HSC quiescence, proliferation and ageing It is unclear why such heterogeneity in HSCs exists. HSCs Somatic stem cells undergo lifelong self-renewal and possess generally do not undergo apoptosis in response to DNA dam- the potential to produce the differentiated cells of the tissue. age and have adopted several mechanisms to preserve stem- HSCs are considered to be relatively dormant stem cells, dividing ness rather than self-renewal, to reduce DNA damage and/or rather infrequently. They are enriched in the quiescent fraction to prevent inappropriate differentiation leading to loss of HSCs. of adult bone marrow and are resistant to 5-fluorouracil (which Both developmental and stem cell protective mechanisms may is an antimetabolite drug that results in the death of rapidly assist in providing maximum HSC fitness during reproductive dividing cells). Recent studies in mice using a label-retaining life, providing an evolutionary benefit. Altered gene expression, method for analysis of cycling versus non-cycling cells show that however, may drive lymphoid differentiation, deplete lymphoid- under homeostatic conditions, dormant HSCs cycle only once biased HSCs and thus contribute to the relative predominance of every 21 weeks. The adult mouse possesses approximately 600 myeloid-biased HSCs. dormant LSK CD150+ CD48− CD34− HSCs. Interestingly, 38% of HSCs in G0, considered to be the dormant HSCs, can be acti- vated by myelo/lymphodepletion during injury, 5-fluorouracil or G-CSF administration, and can return to the dormant state after the re-establishment of homeostasis. Haemopoietic-supportive The maintenance of HSC dormancy is thought to be an microenvironments important strategy for preventing stem cell exhaustion during Adult bone marrow microenvironment adult life. Serial transplantations in the mouse demonstrate that HSC self-renewal is limited to about six rounds of transplanta- Most tissue-specific stem cells are maintained in special tion and that there is a progressive decrease in the ability of the microenvironments/niches that support long-term cell growth transplanted stem cells to repopulate/self-renew. It has been pro- and self-renewal. To provide the continuous production of posed that accumulating DNA mutations and loss of telomere human blood over the many decades of adulthood, HSCs are repeats adversely affect HSC function. Studies of chromosome maintained in the specialized haemopoietic-supportive niches shortening in human HSCs suggest that self-replication is lim- of the adult bone marrow (Figure 1.2). The importance of the ited to about 50 cell divisions. Recently, it was found that HSC bone marrow haemopoietic niche and the interactions between characteristics are changed in aged mice. Comparison of vari- supportive cells and HSCs was first demonstrated in mice. In ous inbred mouse strains has shown that the rate of haemopoi- transplantation studies of anaemic mouse strains naturally defi- etic cell cycling is inversely correlated with their mean lifespan. cient in the c-kit receptor tyrosine kinase (W mice) or kit-ligand The decrease in HSC quality was due to cell-intrinsic genetic (KL; Steel mice) it was revealed that bone marrow from W or epigenetic factors. Causative genes were identified by tran- mutant mice could not repopulate the haemopoietic system of scriptional profiling comparisons between the HSCs of the dif- wild-type irradiated recipient mice, whereas bone marrow from ferent strains. Of particular interest are chromatin modifiers Steel mutant mice could. In contrast, W mutant mice could be involved in prevention of HSC exhaustion through maintenance repopulated by wild-type donor bone marrow cells, whereas of a stem-cell-specific transcriptional programme. Changes Steel recipients were defective for repopulation by wild-type in chromatin structure associated with high HSC turnover donor cells. It was proposed that a receptor–ligand interaction would result in stem cell senescence (which is thought to pro- was involved to support HSCs within the bone marrow microen- tect stem cells from malignant transformation by oncogenic vironment. It was subsequently shown that HSCs express c-kit events). and bone marrow stromal cells express KL. The development Transplantations of single HSCs from both the fetal liver and of ex vivo culture systems to study this complex microenviron- adult bone marrow have revealed HSC heterogeneity in lineage ment allowed further dissection of the cellular and molecular differentiation output related to developmental stage and aging: aspects of the bone marrow microenvironment. These studies some HSCs give a balanced lineage differentiation output of were aided by the isolation of mesenchymal stromal cells. myeloid and lymphoid cells, whereas others yield a predominant Stromal cell lines have been derived from the adult mouse lymphoid or myeloid cell lineage output. During fetal stages, bone marrow and fetal tissues. These are generally of mesenchy- HSCs with a balanced lineage output are at a higher frequency mal lineage, as determined by cell-surface marker expression than in adult BM. During aging the frequency of BM HSCs with and their osteogenic and adipogenic potentials. Although widely a predominant myeloid output increases as compared to the fre- heterogeneous in their ability to support haemopoiesis, some quency of HSCs with a balanced lineage output or predominant stromal lines (MS5 and AFT024, for example) have been shown 6 Chapter 1 Stem cells and haemopoiesis to support the growth and/or maintenance of HSCs in cocul- cell (VSMC) hierarchy, in between a mesenchymal stem cell and tures for long periods. Moreover, they have been instrumen- a VSMC. Other niche cells include cells of the nervous system tal in further characterization of these haemopoietic-supportive and endothelial cells. niches. Comparative transcriptional profiling and database anal- Stromal cell lines established from the AGM region, placenta ysis of HSC supportive and non-supportive stromal cell lines and fetal liver can support immature haemopoietic progenitors has revealed a complex genetic programme involving a wide and HSCs and are more supportive as compared to adult bone variety of known molecules and molecules whose function in marrow cell lines. Some can also support the haemopoietic dif- haemopoiesis is currently under investigation. ferentiation of embryonic stem (ES) cells. Such stromal cell lines The in vivo bone marrow microenvironment is very complex, in a re-aggregate culture system have been able to support the containing osteoblastic niches and vascular niches localized differentiation of cells with a haemogenic endothelial pheno- within the trabecular regions of the long bones. HSCs are main- type (VE-Cad+ CD45− CD41+ cells from mouse embryos before tained in close association with the so-called ‘stromal cells’ of the the onset of HSC generation) into long-term repopulating HSCs. niches (osteoblasts and vascular endothelial cells). Along with This highlights that in an ex vivo controlled environment, cells KL, some of the key molecular regulators within the bone mar- with a potential to become HSCs, can be influenced to do so by row niches include N-cadherin and CD150, and signalling path- other cells. However, it is still unknown whether the inductive way molecules SDF1, Notch, Wnt, Hedgehog, Tie2/angiopoietin, factors in the stromal/re-aggregate cultures are the same factors transforming growth factor (TGF), bone morphogenetic pro- produced in the in vivo physiologic HSC-inductive microenvi- tein (BMP) and fibroblast growth factor (FGF). These regula- ronment. It is likely that HSC induction is a complex process tors are implicated in a variety of cellular processes, such as HSC requiring a variety of spatial and temporal cues emanating from maintenance, differentiation, self-renewal and homing. Indeed, several cell types in the niches of the embryo. live tracking of haemopoietic progenitor/stem cells in the mouse Within the normal physiology of the embryo, the AGM model has shown the homing ability of these cells to bone mar- lies between the ventral tissue that includes mesenchyme and row niches, and mouse models as well as in vitro culture sys- the endoderm-derived gut, and the dorsal tissue including the tems are beginning to reveal the specific molecular mechanisms notochord and the ectoderm-derived neural tube (Figure 1.5). involved. Mouse AGM explant culture experiments have shown that dorsal tissues/signals repress AGM HSC activity and ventral tissues/signals enhance HSC emergence. In both mouse and Microenvironments important for human AGM regions, cells expressing HSC markers are closely haemopoietic development in the conceptus adherent to the vascular endothelium on the ventral aspect of the Prior to the establishment of an adult haemopoietic-supportive aorta. In the mouse, at precisely E10.5, single endothelial cells microenvironment, the embryo contains several haemopoietic bud into the lumen as they take on HSC identity (Figure 1.5). microenvironments that are supportive and/or inductive. The Importantly, HSC activity, as determined by functional trans- extraembryonic yolk sac and placenta, and the intraembryonic plantation assays, is localized exclusively to the ventral aspect AGM generate haemopoietic progenitor cells, whereas the AGM of the mouse mid-gestation aorta. Thus there is a strong posi- region generates the first adult repopulating HSCs (Figure 1.6). tive ventral positional influence on HSC generation in the AGM, Little is known about the differences between the microenvi- and morphogens and local signals emanating from the ventral ronments of the embryonic haemopoietic tissues. However, the endodermal tissues may be responsible for establishing the HSC- AGM microenvironment is the most-well characterized due to inductive microenvironment. the simplicity of its structure, with the aorta at the midline of the Haemopoietic transcription factors required for HSC gener- embryo and the laterally located gonads and mesonephroi (Fig- ation such as Gata2 and Runx1 are expressed in cells of the ure 1.5). The avian AGM microenvironment contains different ventral aortic clusters and endothelium. Deletion of Gata2 and types of mesenchymal cells and a population of aorta-associated Runx1 genes in mice leads to mid-gestation embryonic lethal- stem cells called ‘mesoangioblasts’ that contribute to cartilage, ity, with complete absence of adult haemopoiesis (although bone and muscle tissues, and also to blood. In the mouse AGM embryonic haemopoiesis occurs), thus demonstrating that these region, cells more typical of mesenchymal stromal cells have two pivotal transcription factors promote the HSC genetic pro- been found. Interestingly, mapping and frequency analysis in the gramme. Zebrafish and frog embryos have been useful mod- mouse conceptus show that mesenchymal progenitors, with the els for dissecting the cascade of upstream events that lead to potential to differentiate into cells of the osteogenic, adipogenic HSC induction. Developmental growth factor signalling path- and/or chondrogenic lineages, reside in most of the sites har- ways, such as the BMP, Hedgehog and Notch pathways, converge bouring haemopoietic cells, suggesting that both the HSC and to activate expression of the two transcription factors in aortic mesenchymal stromal cell microenvironment develop in par- haemopoietic cells and promote the HSC programme. In both allel. Phenotypic characterization of haemopoietic-supportive the mouse and human embryo, BMP4 is expressed in the mes- AGM stromal lines places them in the vascular smooth muscle enchyme underlying the ventral aspect of the aorta at the time 7 Postgraduate Haematology of haemopoietic cluster formation. Culture experiments have young, being harvested at the neonatal stage of development, demonstrated the positive influence of BMP4 exposure to mouse thus circumventing concerns about the ageing of HSCs; UCB and human HSC-containing cell populations. BMP4 has been transplantation induces less frequent and less severe GVHD, found to act directly on HSCs in the AGM and, in addition, may since UCB contains many fewer activated T cells than adult bone stimulate the microenvironment to produce HSC effectors. Sim- marrow; also, UCB HSCs are highly proliferative. However, only ilarly, Hedgehog signalling regulates HSCs in the AGM region, relatively small numbers of cells are harvested (approximately likely in an indirect way through VEGF. Other ventrally local- 10-fold lower than those in adult bone marrow) and this lim- ized HSC regulators include the Notch signalling molecules, as its their use to paediatric patients, unless multiple UCB units well as Wnt3a and interleukin (IL)-1. are transplanted. Despite increases in the number of UCB units High-throughput chemical screens offer a means of identify- (400 000) stored in cord blood banks (>50) around the world ing molecules involved in HSC growth, maintenance and expan- (catalogued and recorded by EUROCORD and other coordinat- sion. Through such a screen in zebrafish embryos, prostaglandin ing efforts) and HLA donor-cell selection for rare haplotypes, the E2 (PGE2) was recently identified as a regulator of HSC num- supply of HSCs is still limited. ber. When tested in the murine transplantation model, ex vivo exposure of bone marrow cells to PGE2 enhanced short-term Gene therapy and gene editing for repopulation by haemopoietic progenitors and increased the fre- haemopoietic disease quency of long-term repopulating bone marrow HSCs. PGE2 modifies the Wnt signalling pathway, which in turn is thought to Monogenic disorders of the blood are the first targets of gene control HSC self-renewal and bone marrow repopulation. Extra- therapy approaches. To effect a cure for a haematologic dis- cellular environmental cues, such as blood flow, also affect HSC ease in which a single gene or regulatory element is mutated, generation. A zebrafish chemical screen identified modulators a viral vector containing a normal copy of the gene is used to of blood flow such as nitric oxide synthetase (NOS). Inhibition introduce and express the gene in HSCs. Gene therapy for β- or deficiency of NOS reduces murine bone marrow HSC num- haemoglobinopathies, such as β-thalassaemia and sickle cell dis- ber/function. Thus, together with general physiological cues, ease, were among the first proposed and tested in mouse models. such as the haemopoietic growth factors, KL, IL-3, Flt3 and Primary immunodeficiencies (PID) are also monogenic disor- thrombopoietin, chemical modulators and developmental regu- ders and result in the absence of (parts of) the innate and adap- lators may be useful for expansion of HSC number and enhance- tive immune system. Patients can be cured with allogeneic HLA ment of HSC function for therapeutic purposes. matched (related) HSC transplantation. However, donor avail- ability is limited. For patients without an allogeneic donor, gene therapy of their own bone marrow HSCs and subsequent autol- ogous transplantation is the only option for curative treatment. Haemopoietic regenerative and Lentiviral vector infection offers an efficient mode of delivery replacement therapies of a functional copy of the mutated gene into the genome of the patient’s own HSCs used for transplantation. In initial gene ther- Stem cell transplantation apy trials of immunodeficient patients, lenti-viral vector inser- For over 50 years, HSC transplantation has been the most suc- tions in the genome of some transplanted HSCs resulted in acti- cessful and significant clinical cell regenerative therapy (see vation of oncogenes, the selective growth of these HSC clones Chapter 35). Initially, whole bone marrow was the source of and the onset of leukaemia. More recent trials have incorpo- cells used in clinical transplantation, but through experience and rated a safety feature in the lentivirus that reduces (but has much research new and/or improved sources of transplantable not completely eliminated) the unwanted activation of onco- HSCs were found. These now include the CD34+ CD38− frac- genes in the case of viral insertion. In 2010 a new gene therapy tion of adult bone marrow, mobilized peripheral blood HSCs clinical trial was initiated for Wiscott–Aldrich (WAS) patients and the CD34+ CD38− fraction of umbilical cord blood. The who suffer from thrombocytopenia, eczema, recurrent infec- cumulative data from the large number of patients worldwide tions, autoimmune disorders and high susceptibility to develop receiving a bone marrow transplant provide valuable informa- tumours. To date, all patients are alive and show significant tion on the success of autologous versus allogeneic transplan- increase in platelets and T cells, although long-term follow- tation, the number of human leucocyte antigen (HLA) differ- up is required. Similar results have been obtained for adeno- ences that are tolerated by the recipient, the incidence of graft- sine deaminase-deficient severe combined immunodeficiency versus-host disease (GVHD), and the unexpected and advanta- (ADA SCID) where 40 patients have been treated since 2000 geous graft-versus-leukaemia effect. in Italy, the UK and the US without any reports of malignant Interestingly, umbilical cord blood (UCB) appears to offer a occurrences. X-linked SCID and chronic granulomatous disease beneficial source of HSCs for several reasons: UCB HSCs are (CGD) patients have also undergone gene therapy treatment, 8 Chapter 1 Stem cells and haemopoiesis albeit with less success. The treatment for CGD was impaired tional sources of HSCs. Furthermore, such a cell culture system since the earliest trials did not make use of myeloablation to would make possible the use of novel gene editing approaches for enhance chimerism of the gene-manipulated graft. Gene therapy monogenetic disorders. These gene correction approaches could trials for β-thalassaemic patients are ongoing and encouraging, be used in combination with patient-derived iPS cells. Studies but for successful treatment, higher levels of gene expression and using mouse and human ES cells have optimized culture con- HSC chimerism will be needed. ditions to include temporally changing combinations of growth Clinical trials with gene therapy are promising. However, factors (ActivinA, BMP4, VEGF, etc.) and signalling pathway lentiviral vector insertions that may result in malignant clone antagonists to control differentiation to the mesodermal, vascu- outgrowth remain a risk. New gene editing techniques offer new lar and thereafter the haemopoietic lineage. The ES-cell-derived hope for gene correction directly within the gene of concern. haemopoietic cells arise from haemangioblasts and/or primi- Gene editing makes use of endonucleases (zinc finger nucle- tive endothelial-like cells that express PECAM-1, FLK-1 (KDR) ases, TALENs or CRISPR/Cas9) to target a specific genomic and VE-cadherin, and are thought to represent the types of site and repair the mutated gene or insert a functional gene precursors, progenitors and differentiated cells found normally under the control of its own promoter. This method leaves no in the yolk sac. These results have strengthened the idea that extra genetic modulation. At present this approach requires pro- ES cell differentiation proceeds via a ‘haemogenic endothelial’ longed cell culture and a selection step for the corrected HSCs, differentiation step, before definitive haemopoietic cells can be and thus requires further research developments in HSC growth produced and require activation of the Wnt-β-catenin pathway and expansion before it will be clinically useful. Gene ther- (Figure 1.7). apy for genetic disorders of coagulation proteins is discussed in An alternative approach to produce HSCs ex vivo has recently Chapter 38. emerged. This molecular reprogramming approach aims to reprogramme non-haemopoietic or differentiated haemopoi- etic cells directly to HSCs, without going through a pluripotent New sources of HSCs for transplantation stem cell state. Such induced HSCs, or iHSCs would be gener- The ability to expand HSCs ex vivo is a theoretically practical ated through reprogramming directed by transcription factors and attractive means to obtain an accessible and limitless source pivotal to HSC generation and/or growth. Several laboratories of HSCs for transplantation therapies. Unfortunately, despite have been able to generate haemopoietic progenitors or stem many years of research using different culture systems and cells from more differentiated cells using four to eight different combinations of haemopoietic growth factors and proliferation- haemopoietic transcription factors previously identified from stimulating agents, ex vivo expansion of HSCs has not been HSC transcriptome databases. In one case, mouse endothelial- achieved. However, HSC developmental studies have begun to like precursor cells (Sca1+, Prominin 1+ and expressing a provide new insights into the processes directing the generation human CD34 reporter) have been converted into haemopoi- and growth of HSCs. If cells such as the haemogenic endothe- etic progenitors using the factors Gata2, Gfi1b, cFos and Etv6. lial cells of the embryonic aorta are present in the adult vascula- A human myeloid precursor (CD34+CD45+) cell has been ture or could be obtained from ES/iPS cells, they could provide converted into a haemopoietic progenitor cell (CD34+CD38−) a novel source of inducible HSC precursors, particularly if they using HOXA9, SOX4, RORA, and MYB; however, neither study can be sustained and expanded to large numbers in culture. was able to generate long-term repopulating HSCs. A study con- verting mouse committed B cell progenitors using a mix of eight transcription factors (Run1t1, Hlf, Lmo2, Prdm5, Pbx1, Embryonic stem cells and induced pluripotent Zfp37, Myc-n and Meis1) has resulted in long-term repopulat- stem cells ing HSCs. In this method, B cell progenitors are transduced with Pluripotent embryonic stem (ES) and induced pluripotent stem the eight factors, and immediately transplanted into irradiated (iPS) cells have been used to generate differentiated cells in recipients. In this way the native bone marrow niche preserves many tissue systems, including the haemopoietic system. Devel- the new iHSCs and allows them to be maintained and function opmental studies revealing the temporally and spatially limited in the physiologic context of the recipient (Figure 1.7). Whereas production of HSCs in the embryonic vasculature, the compo- this study demonstrates that transcription factor transduction nents of the specific microenvironment, and the knowledge of of haemopoietic cells can yield HSCs, this approach is limited in the molecular programme of endothelial to haemopoietic cell applications for research or therapy. The fact that each of these transition have yielded insight into how HSCs may be induced studies uses a completely different panel of transcription fac- and/or expanded without undergoing differentiation in such tors to make induced HSCs or HPCs indicates that there may cultures. Haemopoietic-directed differentiation of human iPS be more than one way to reprogramme cells to the haemopoi- cells towards endothelial cells, haemogenic endothelial cells and etic lineage and that a further understanding of HSC biology is HSCs would be a potentially attractive alternative to conven- required. 9 Postgraduate Haematology Re-programming De-differentiation (4 transcription factors) (8 transcription factors) pre-B Fibroblast CLP pre-T ES cell Haemogenic HSC CFU-S MPP endothelial cell GMP In vitro differentiation (growth factors) iPS cell CMP (gene corrected) Gene editing Re-programming MEP iPS cell (disease mutation) Figure 1.7 Several experimental approaches to generate HSCs progenitors, but not HSCs. In vitro haemopoietic differentiation of de novo. De-differentiation of pre-B cells with eight transcription ES and iPS cells relies on the addition of developmental and factors and immediate transplantation in vivo, allows for the haemopoietic growth factors to induce the progressive production of multilineage, self-renewing HSCs due to the differentiation of these pluripotent cells to mesoderm, endothelial, presence of functional HSC niches that are as yet not attainable in haemogenic endothelial and haemopoietic fates. As this culture in vitro cultures. To date, direct reprogramming of B cell system improves, it may be possible to make iPS cells from patients progenitors and immediate transplantation into irradiated mouse with monogenic disease and correct the gene mutation by gene recipients has been the only study successful in generating HSCs. editing. These cells may then be differentiated to HSC fate and Reprogramming with four pivotal haemopoietic transcription used for clinical treatment. factors has yielded haemogenic endothelial cells and haemopoietic Morrison SJ, Scadden DT (2014) The bone marrow niche for Selected bibliography haematopoietic stem cells. Nature 505: 327–34. Nienhuis AW (2013) Development of gene therapy for blood disorders: Boisset JC, Cappellen G, Andrieu C, Galjart N, Dzierzak E, Robin an update. Blood 122: 1556–64. C (2010) In vivo imaging of hematopoietic stem cells emergence. Pereira C-F, Lemischka I, Moore K (2014) ‘From blood to blood’: de- Nature 464: 116–20. differentiation of hematopoietic progenitors to stem cells. The EMBO Dzierzak E, Speck NA (2008) Of lineage and legacy: the development Journal 33(14): 1511–13. of mammalian haematopoietic stem cells. Nature Immunology 9: Snoeck, H-W (2013) Aging of the hematopoietic system. Current Opin- 129–36. ion in Hematology 20: 355–61. Gluckman E, Rocha V (2009) Cord blood transplantation: state of the art. Haematologica 94: 451–4. 10 CHAPTER 2 Erythropoiesis Douglas R. Higgs, Noémi Roy and Deborah Hay Weatherall Institute of Molecular Medicine, John Radcliffe Hospital, Oxford, UK 2 endothelial cells are thought to arise from a common progen- Introduction itor, the haemangioblast, which has the potential to form both blood and vessels (Chapter 1). Erythropoiesis occurs in waves Erythropoiesis is the generation of red blood cells carrying the that emerge from several sites in the developing embryo, and respiratory pigment haemoglobin, for the transport of oxygen begins at the same time as development of the circulatory sys- to the tissues. This process, from the erythroid commitment of tem. Primitive erythropoiesis is first evident at around three multipotent haemopoietic stem cells (HSCs), through the mat- weeks of gestation, and arises from the blood islands of the uration of erythroblasts, to the terminal differentiation of red extraembryonic yolk sac. A second wave of haemopoietic activ- blood cells, is governed by complex transcriptional and epige- ity emerges from the yolk sac at approximately 4 weeks’ ges- netic programmes, in response to extracellular signalling. Ery- tation, and marks the onset of definitive erythropoiesis. Ery- thropoiesis normally maintains the steady state of an individual’s throid progenitors released into the circulation at this time pass red cell mass, producing 1011 –1012 new cells per day to replace to the liver, which becomes the main site of erythropoiesis in the those that are lost through senescence or premature destruction. fetus. A final wave of haemopoietic activity occurs in the aorta– Furthermore, erythropoiesis must be able to respond rapidly to gonad–mesonephros (AGM) region, the placenta and the major erythroid stress such as haemorrhage and haemolysis. Perhaps vessels at approximately 4–6 weeks of gestation. By 10–12 weeks, unsurprisingly, this system is remarkably sensitive to systemic haemopoiesis starts to migrate to the bone marrow, where blood disease, with anaemia being a common manifestation of a wide formation becomes established during the last three months of range of inherited and acquired clinical disorders. Understand- gestation (Chapter 1). ing the basic biology of erythropoiesis provides a logical basis Primitive and definitive erythropoietic cells are distinguished for the diagnosis and treatment of the inherited and acquired by their morphology, cytokine responsiveness, growth kinet- anaemias that are so frequently encountered in clinical prac- ics, transcription factor programmes, epigenetic programmes tice. In this chapter, we outline the normal mechanisms underly- and patterns of gene expression. Importantly, the types of ing erythroid specification, differentiation and maturation, and haemoglobin produced are quite distinct in embryonic (Hb highlight some of the ways in which this complex system may Gower I ζ2 ε2 , Hb Gower II α2 ε2 and Hb Portland ζ2 γ2 ), fetal fail in erythroid diseases. (HbF α2 γ2 ) and adult (HbA α2 β2 and HbA2 α2 δ2 ) erythroid cells. These specific patterns of globin expression provide crit- ical markers for identifying the developmental stages of ery- The origins of erythroid cells during thropoiesis. It remains unclear whether primitive and definitive development haemopoiesis in mammals have entirely separate origins or if they are both derived from common stem cells that arise during Both primitive (embryonic) and definitive (fetal/adult) HSCs early development. Accurately defining the embryological ori- arise in close association with endothelial cells. HSCs and gins of these cells (Chapter 1) is important for understanding Postgraduate Haematology, Seventh Edition. Edited by A Victor Hoffbrand, Douglas R Higgs, David M Keeling and Atul B Mehta. © 2016 John Wiley & Sons, Ltd. Published 2016 by John Wiley & Sons, Ltd. 11 Postgraduate Haematology the normal mechanisms that establish and maintain HSCs and defined as discrete cell populations when assayed by cell-surface how these programmes are subverted in common haematologi- markers. CFU-Es defined in these culture systems most closely cal disorders. correspond in vivo to pronormoblasts (also known as proery- throblasts), the earliest morphologically recognizable erythroid precursor in the bone marrow. Specifying the erythroid lineage Expression of critical transcription factors At a cellular level, the precise mechanism by which HSCs differ- specifies the erythroid lineage entiate into lineage-committed progenitors remains unknown and is currently under intense investigation. However, it is clear Over the past few years, it has emerged that key haemopoietic that as HSCs differentiate, they initially form multipotential pro- transcription factors play a major role in regulating the forma- genitor cells such as CFU-GEMM – colony-forming units that tion, survival, proliferation and differentiation of multipotent have the ability to produce granulocytes, erythrocytes, mono- stem cells as they undergo the transition to erythroid cells. These cytes and megakaryocytes. Such cells retain short-term repop- transcription factors may operate on their own or as members ulating ability, but lose long-term repopulating potential. Fur- of multicomponent complexes involved in the activation and/or ther differentiation progressively restricts the lineage potential repression of gene expression. At present, the key transcription of these cells, and reduces their proliferative capacity, resulting factors known to be involved in the specification and mainte- in bipotential progenitors with the ability to form megakary- nance of HSCs include RUNX1, TAL1, LMO2, TEL, MLL and ocytes or erythroid cells (MEPs). MEPs further differentiate GATA2 (Figure 2.1). During normal erythroid development, either into megakaryocytes and platelets or into fully commit- GATA2 probably initiates the erythroid programme and plays an ted erythroid precursors and red blood cells. These cells are important role in the expansion and maintenance of haemopoi- functionally defined by their growth potential and character- etic progenitors. It is replaced during terminal erythroid matu- istics as assayed by in vitro cultures: this explains the names ration by GATA1, with the level of GATA2 declining as GATA1 ‘burst-forming’ erythroid units (BFU-E) and ‘colony-forming’ increases. GATA1 is first expressed in MEPs and is essential for erythroid units (CFU-E) (Figures 2.1 and 2.2). They are not the terminal differentiation and maturation of both megakary- morphologically recognizable in the bone marrow, but can be ocytes and erythroid cells. Lymphoid lineages Megakaryocytes Self-renewal Other myeloid lineages HSC Erythrocytes CFU-GEMM MEP BFU-E CFU-E TAL1 RUNX1 Key GATA1 KLF1 TEL transcription FOG1 Gfi-1b LMO2 factors NFE2 TAL1 MLL GATA2 Figure 2.1 Summary of some steps in self-renewal, lineage colony-forming unit - erythroid; TAL1, T-cell acute lymphoblastic specification and differentiation of haemopoietic stem cells to red leukaemia 1; TEL, translocation Ets leukaemia; LMO2, LIM cells. Some of the key transcription factors involved in this process domain only 2; MLL, mixed lineage leukaemia; GATA1, are summarized beneath the diagram. HSC, haemopoietic stem GATA-binding factor 1; GATA2, GATA-binding factor 2; FOG, cell; CFU-GEMM, colony-forming unit - granulocyte erythrocyte friend of GATA; NFE2, nuclear factor erythroid-derived 2; KLF1, monocyte megakaryocyte; MEP, megakaryocyte-erythroid Kruppel-like factor 1 (erythroid); GFi-1b, growth-factor progenitor; BFU-E, burst-forming unit - erythroid; CFU-E, independent 1b. 12 Chapter 2 Erythropoiesis 188 h 60 h 30 h 50 h 48 h 1 2 4 8 16 Erythron Progenitors Precursors HSC CFU-GEMM BFU-E BFU-E CFU-E Pro Bas Early Late Retic RBC Early Late Pol Ort Frequency per 104 ~1 ~1 4–10 20–60 50–60 400 500 1000 nucleated bm cells % cells in cycle Low 15–20 30–40 60–70 60–70 80 0 CD34 ++ +++ ++ ± – – – – CD71 (TfR) ± ± ± ++ +++ +++ +++++ ++ + EpoR – – – ± ++ ++ ++ ± ± – GPA – – ± + ++++ ++++ ++++ ++++ ++++ Globin mRNA – – – – – + + +++ +++++ ++ Hb – – – – – ± + +++ +++++ +++++ Figure 2.2 The specification and terminal differentiation of red blood cells (RBCs). The number of divisions from erythroid cells from haemopoietic stem cells. At the top, the pronormoblasts to orthochromatic normoblasts (1–16) is also estimated times for maturation of terminally differentiating cells shown. Some examples of the expression patterns of key are shown. The precursors are as follows: pronormoblasts (Pro), cell-surface markers are shown below. TfR, transferrin receptor; basophilic erythroblasts (Bas), polychromatic erythroblasts (Pol), EpoR, erythropoietin receptor; GPA, glycophorin A; bm, bone orthochromatic erythroblasts (Ort), reticulocytes (Retic), mature marrow. One emerging principle in our understanding of lineage com- precursor pool, so that each newly formed pronormoblast mitment is that factors affiliated with different lineages such as develops into 16 red cells (Figure 2.2). As a small amount of GATA1 (erythroid) and PU.1 (lymphocytes and granulocytes) cell death (ineffective erythropoiesis) normally occurs, the are both present in uncommitted progenitors, reflecting the average amplification is slightly less than 16-fold. The majority potential of these cells to develop along alternative pathways (so- (60–80%) of pronormoblasts, basophilic normoblasts and early called multilineage priming). It is now known that GATA1 and polychromatic normoblasts are in cell cycle. By contrast, late PU.1 interact and cross-antagonize each other. Therefore, as cells polychromatic/orthochromatic erythroblasts are postmitotic, differentiate, reinforcement of the transcriptional programme non-dividing cells. In the final stages of terminal maturation, of one lineage may actively suppress those of the alternative the nucleus condenses further and is eventually extruded. This lineages. produces the mature reticulocyte, which has no nucleus, but retains a few mitochondria and ribosomes. The cytoplasm Terminal maturation of committed of reticulocytes is predominantly pink on Wright–Giemsa erythroid cells staining because of the high concentration of haemoglobin, but it has a greyish tint due to the presence of ribosomes. When Once the erythroid programme has been specified, the final stained supravitally, the ribosomes precipitate into basophilic phase of erythropoiesis involves the maturation of committed granules or a reticulum. Reticulocytes continue to synthesize erythroid progenitors to fully differentiated red cells. The earli- haemoglobin for 24–48 hours after leaving the bone marrow. est recognizable erythroid precursor in the bone marrow is the pronormoblast. Division of pronormoblasts leads to progres- Changes in the expression of transcription sively smaller basophilic normoblasts, early polychromatic and factors during terminal maturation finally late polychromatic/orthochromatic normoblasts (Fig- ure 2.3). It has been estimated that, on average, four divisions Once progenitor cells have committed to become erythroid occur within the morphologically recognizable proliferating cells, GATA1 and its cofactor FOG-1 (friend of GATA1) are 13 Postgraduate Haematology among the factors needed for them to proceed through ter- minal differentiation. There are GATA-binding motifs in the promoters and/or enhancers of virtually all erythroid-specific genes studied to date, including the globin genes, haem biosyn- thetic enzymes, red cell membrane proteins (e.g. blood group antigens) and erythroid transcription factors, such as TAL1, KLF1, Gfi1b, NFE2 and GATA1 (Figure 2.1). TAL1 is a basic (i) (ii) helix–loop–helix transcription factor, which plays an impor- tant role in both the stem cell compartment and in developing the erythroid programme. KLF1 is a zinc-finger-like transcrip- tion factor expressed only in erythroid cells. Its binding sites are found in the regulatory elements of many erythroid-specific genes, including the β-globin gene. All of these transcription (iii) (iv) factors play key roles in coordinating erythroid maturation (a) and globin gene regulation. Mutations of GATA1 and KLF1 Orthochromatic are rare, but have been described in families with abnormali- (Late) ties of haemoglobin synthesis (Chapter 6), disorders of the red cell membrane (Chapter 8), abnormal haem synthesis (Chap- Polychromatic (Intermediate) ter 3) and other abnormalities of erythropoiesis manifesting, for example, as congenital dyserythropoietic anaemia, Diamond– GPA Blackfan anaemia (Chapter 10) and sideroblastic anaemia (Chapter 3). We anticipate that, with genome-wide sequencing, mutations in the other erythroid transcription factors (FOG1, Pronormoblasts TAL1, Gfi1b, NFE2) will be found underlying some rare forms (early) of anaemia. CD71 (TfR) (b) Changes in the expression of erythroid proteins during terminal maturation As multipotent progenitors enter terminal differentiation, the expression of many genes (∼6,000) is downregulated, reflecting the commitment to a single specialized lineage. By contrast ∼600 mRNAs encoding proteins that characterize the red cell pheno- type, are, in general, upregulated. Examples include blood group antigens, red cell membrane proteins, red cell glycolytic pathway enzymes, carbonic anhydrase and enzymes of the haem synthe- sis pathway. A full catalogue of these changes in gene expression can be found at https://cellline.molbiol.ox.ac.uk/eryth/index. html (Human Erythroid Maturation database). Changes in gene expression are reflected in the cell-surface (c) phenotypes of erythroid progenitors and precursors, in turn set- Figure 2.3 (a) Examples of pronormoblasts; (i) basophilic and ting up the different signalling programmes of erythroid cells as polychromatic erythroblasts, and (ii) polychromatic and they differentiate. Receptors for the key erythroid hormone ery- orthochromatic erythroblasts (iii and iv). All these different cell thropoietin (Epo, discussed in more detail below) first appear types can also be conveniently viewed at http://hsc.virginia.edu/ in small numbers on late BFU-Es, increasing in CFU-Es and medicine/clinical/pathology/educ/innes/text/nh/mature.html. pronormoblasts and subsequently declining and disappearing (b) An example of early (pronormoblasts), intermediate in later erythroid precursors (Figure 2.2). Similarly, CD71 (the (polychromatic erythroblasts) and late (orthochromatic transferrin receptor, TfR), which allows transferrin-bound iron erythroblasts) erythroid precursors separated on the basis of their to be taken into the cell, is present on early haemopoietic cells cell-surface markers (CD71 and GPA). (c) An erythroblastic island but is considerably upregulated on cells that are actively synthe- with its central macrophage surrounded by erythroid progenitors sizing haemoglobin, reaching a peak of 800,000 molecules per at various stages of differentiation. cell on polychromatic normoblasts. CD71 levels diminish in the 14 Chapter 2 Erythropoiesis late phase of terminal differentiation and the receptor is unde- similar cis-regulatory elements, which are bound by the tran- tectable on mature erythrocytes, which no longer have a need scription factors GATA1, SCL and KLF1. As will be explained for iron uptake. In addition, developing erythroid cells express in Chapter 3, one of the main mechanisms by which haem and cell-surface adhesion molecules that interact with the extracel- globin syntheses are coordinated is at the level of mRNA and lular matrix at high levels in early precursors. These are lost as translation via iron-responsive elements (IREs). There is also a maturation proceeds, freeing erythroid cells from the bone mar- need to coordinate the availability of iron to the requirements row niche (see below) to enter the circulation. of erythropoiesis, by controlling iron absorption from the gut Key among the transcripts upregulated during erythroid mat- and release of iron from its stores. This is achieved by the mas- uration are those of α and β globin. The globins are first ter regulator of iron, hepcidin. This aspect of haemopoiesis and expressed in pronormoblasts and early basophilic erythroblasts, the diseases arising from abnormalities in iron metabolism are with the number of transcripts reaching 20,000 molecules per reviewed in Chapters 3 and 4. cell in late polychromatic and orthochromatic erythroblasts. The high levels of protein synthesis and rapid cell prolifer- During the later stages of erythroid cell maturation, the amount ation that characterize the erythroid compartment, even in a of RNA per cell and the rate of total protein synthesis decline, steady state, render these cells exquisitely sensitive to perturba- but the unusual stability of globin mRNA ensures that globin tions in levels of the key substrates required for erythropoiesis. remains the predominant protein made in late erythroblasts and Among these are a variety of nutritional factors and cofactors, reticulocytes (Figure 2.2). Disorders of α and β globin struc- particularly iron (Chapters 3 and 4), vitamin B12 and folate ture and synthesis are the most frequent causes of inherited (Chapter 5), but also manganese, cobalt, vitamin C, vitamin anaemia throughout the world and are discussed in detail in E, vitamin B6 (pyridoxine), thiamine, riboflavin, pantothenic Chapters 6 and 7. acid and amino acids. Absolute or relative deficiencies of any The individual components of the haemoglobin synthetic of these factors can impair normal erythropoiesis and result in pathway (iron, free porphyrins, haem and monomeric globin anaemia. chains) are all toxic to the cell, and feedback loops have evolved to ensure that cells are not damaged by these intermediates. In particular, the synthesis of globin is accurately matched with the Control of erythropoiesis via synthesis of haem, in which some steps occur in the cytoplasm cell signalling and others in the mitochondria (Figure 2.4). mRNAs encoding many components of the haem biosynthetic pathway (e.g. ALAS The process of erythropoiesis must also be sensitive to changes and porphobilinogen deaminase) are coordinately upregulated in the circulating capacity for oxygen carriage and varying phys- in terminal erythroid differentiation and their genes contain iological demands. This precise regulation hinges on sensing IRE/IRP Haemoglobin Iron storage α-Globin β-Globin IRE/IRP Transferrin receptor Fe2+ HRI Haem HRI IRE/IRP Succinyl CoA Haem Glycine Ferrochelatase ALA synthase Protoporphyrin δ-Aminolaevulinic acid Protoporphyrinogen Coproporphyrinogen δ-Aminolaevulinic acid PBG deaminase Coproporphyrinogen Porphobilinogen Uroporphyrinogen Figure 2.4 Coordination of globin synthesis, haem synthesis and synthesis are shown in black boxes. ALA, δ-aminolaevulinic acid; iron regulation. Blue lines indicate some of the known regulatory HRI, haem-regulated eIF2alpha kinase; IRE, iron-responsive feedback systems. The red shaded box indicates reactions element; IRP, iron-responsive binding protein; PBG, occurring in the mitochondria. Rate-limiting controls of haem porphobilinogen. 15 Postgraduate Haematology hypoxia and a tight control over the supply of erythroid pre- HRE of the Epo gene (and other hypoxia-sensitive genes), to cursors. Over the past 25 years there has been great progress increase transcription and therefore increase serum Epo levels. in understanding the mechanisms by which cells sense hypoxia The positive effect of HIF is greatly increased by two cofactors, and orchestrate their response. The most important mediator of HNF-4, and the coactivator CBP/p300, also under hypoxic con- this response is the transcription factor HIF (hypoxia-inducible trol. Linking Epo production to tissue oxygenation ensures that factor), which activates genes influencing adaptive responses to when there is reduced ambient oxygen tension, blood loss or hypoxia (Figure 2.5). These include the genes for erythropoi- shortened red cell survival, the level of Epo rises, stimulating red etin (Epo) to boost erythropoiesis, glycolytic pathway enzymes cell production and ultimately providing a greater source of local to maintain energy availability despite hypoxia, the transferrin oxygen delivery. receptor to ensure increased iron availability for erythropoiesis, The effects of Epo on red cell production are mediated by and VEGF to promote angiogenesis. In rare cases of inherited both increasing proliferation and reducing apoptosis of ery- polycythaemia, constitutive mutations in HIF or vHL (von Hip- throid precursors. To respond to erythroid stress, the marrow pel Lindau; Figure 2.5) result in deregulated oxygen sensing and must continuously produce an excess of erythroid precursors an erythropoietic drive in the absence of hypoxia (Chapter 35). that can be called upon to differentiate into mature red cells Epo is a 166-amino-acid 34.4-kDa glycoprotein, found in whenever an immediate increase in erythroid output is needed. serum at baseline levels of 5–25 iU/L that can be elevated 1000- The low numbers (20–50) of Epo receptors (EpoR) on BFU-E fold by severe anaemia. It contains about 40% carbohydrate, is explain the relative Epo unresponsiveness of these cells; much rich in sialic acid residues, and has a half-life of 7–8 hours in higher levels (∼1000) are found in CFU-E, pronormoblasts and plasma, whereas non-glycosylated Epo is rapidly cleared from basophilic erythroblasts (Figure 2.2). Late BFU-E, CFU-E and the circulation. The main site of Epo production is the intersti- pronormoblasts may all require continuous signalling via the tial cells of the kidney. Under normoxic conditions, little or no EpoR to prevent their apoptosis, with signals from EpoR, via the Epo mRNA is detectable in the kidneys; hypoxia results in the JAK2–STAT5 pathway, inducing or stabilizing expression of the rapid induction of its transcription such that levels may increase antiapoptotic protein Bcl-XL (Figures 2.6 and 2.7). The overall up to 200-fold over baseline within 30 minutes. effect is that, secondary to a hypoxia-driven rise in Epo, a popu- Epo upregulation is accomplished through a hypoxia- lation of erythroid progenitors normally destined for apoptosis response element (HRE) at the 3′ -end of the Epo gene. Under at steady-state receive antiapoptotic signals, survive and expand hypoxic conditions, HIF-1α is stabilized and can bind to the the erythroid precursor pool. While the most important effect of Hypoxia High O2 Ub Prolyl hydroxylase Ub oxygen sensor Ub vHL Ub HIF1-α OH Ub HIF1-β HIF1-α Ub EpoR TfR VEGF Glycolytic enzymes Erythropoiesis Supply of iron Angiogenesis Supply of energy Degradation via Oxygenation proteasome Figure 2.5 The oxygen-sensing system. Ub, ubiquitination; vHL, and vHL cannot bind and ubiquitinate HIF-1α, the half-life of von Hippel–Lindau protein; HIF, hypoxia inducible factor. vHL is which is therefore greatly increased. As a result, HIF-1 is able to an E3 ubiquitin ligase. In the context of normal oxygenation, carry out its function as a transcription factor, upregulating the HIF-1α is hydroxylated, providing a binding site for vHL, which expression of its target genes, such as EpoR, TfR, VEGF, and ubiquitinates it, thereby targeting it for degradation by the glycolytic enzymes in response to hypoxia. proteasome. At low oxygen tension, hydroxylation cannot occur 16 Chapter 2 Erythropoiesis Dimerization or ligand-induced conformational change Epo Epo Epo Epo P JAK2 JAK2 P P JAK2 JAK2 P P JAK2 JAK2 P P JAK2 JAK2 P P P P P P P P P P P P P P STAT P P STAT P P P SHP1 P P SHP1 P STAT STAT P P STAT STAT P Target gene Figure 2.6 A summary of signalling via the erythropoietin (Epo) followed by phosphorylation of the Epo receptor. This is followed receptor as described in the text. P denotes regions of by binding and phosphorylation of STAT5. Binding of SHP1 (far phosphorylation. The diagram shows Epo-induced dimerization right) to the Epo receptor activates its phosphatase activity, which or conformational change with trans-phosphorylation of JAK2, can then dephosphorylate JAK2 and terminate signalling. Epo is to increase the number of progenitor cells, which rapidly receptor), all-trans retinoic acid (retinoic acid receptor) and respond by proliferating and differentiating into viable pronor- 9-cis-retinoic acid can promote erythroid differentiation. Such moblasts, it has also been suggested that Epo is able to speed up observations are consistent with reports showing that many the rate of terminal differentiation by shortening the cell cycle endocrine disorders (hypothyroidism, hypopituitarism, Addi- and maturation times of erythroblasts. son’s disease and male hypogonadism) can be associated with Recent work in the mouse has shown that Epo stimulation normochromic normocytic anaemia. through the JAK-STAT pathway causes upregulation of the pro- tein erythroferrone, which mediates suppression of hepcidin synthesis. In this way, the signals that result in higher num- The erythroid niche bers of erythroid progenitors are matched with those result- ing in increased iron absorption (see also Chapter 3). Current The control of erythropoiesis ultimately depends on integration work is investigating whether this control loop also exists in of information delivered by cell–cell contacts, local growth fac- humans. tors and systemic hormonal signalling. Although erythropoiesis Erythropoiesis is also influenced by pathways other than Epo– can be largely recapitulated using liquid cultures, it is signifi- EpoR. Erythroid progenitors express receptors for SCF (stem cantly less efficient in this context than in vivo, particularly in cell factor), insulin-like growth factor (IGF-1) and insulin. After terms of proliferation and terminal enucleation, suggesting that Epo, the second most important signalling system for erythro- erythropoiesis in its natural setting is modified by additional poiesis involves SCF (stem cell factor also called c-Kit). SCF factors. was originally identified by its ability to stimulate proliferation Inspection of the bone marrow architecture shows that matur- of multipotent haemopoietic progenitors, but it is also effective ing erythroblasts are not randomly distributed; rather, they are in supporting growth of committed progenitors, including ery- arranged in characteristic ‘erythroid islands’ made up of a cen- throid progenitors, acting synergistically with Epo. tral macrophage surrounded by up to 30 erythroid cells at var- In addition to SCF and Epo, stimulation of the nuclear hor- ious stages of differentiation (Figure 2.3). This cellular struc- mone receptors for dexamethasone (glucocorticoid receptor) ture is referred to as the erythroid niche. Erythroid cells cul- and oestrogen (oestrogen receptor) produces sustained prolif- tured in vitro with macrophages proliferate threefold more than eration of erythroid progenitors. Furthermore, the nuclear hor- those cultured alone, and macrophages secrete soluble factors mone receptors for thyroid hormone (c-ErbA/thyroid hormone such as IGF-1 and burst-promoting activity that may influence 17 Postgraduate Haematology Kidney Epo P P Oxygenation JAK2 P P Pl3-K Ras STAT P P MAPK GATA-1 Bcl-XL Anti-apoptotic Fas ligand Fas Figure 2.7 A summary of the apoptotic pathways (Epo and Fas) in erythroid progenitors. These cells (BFU-E and CFU-E) undergo apoptosis in the absence of Epo signalling or in the presence of Fas signalling. Bcl-XL may be the key pathway through which these effects are mediated. erythroid proliferation. Furthermore, direct contacts between the macrophages and surrounding erythroblasts (e.g. by cell Red cell senescence and clearance adhesion molecules) appear to be critical for red cell matura- Mature red cells live for approximately 120 days in the circula- tion, stress erythropoiesis and red cell clearance. The central tion under normal conditions, suggesting that mechanisms exist macrophage also phagocytoses and degrades the extruded red to monitor their senescence and control their removal from the cell nuclei, and may play a role in the delivery of iron to the circulation. Since mature red cells have no nucleus, they lack the maturing erythroblast. capacity to synthesize new cellular components. Their ability to The juxtaposition of mature and maturing erythroblasts in maintain cellular integrity becomes compromised with age, and erythroid islands also permits a further mechanism for the reg- characteristic features of the ageing red cell include increased ulation of erythroid expansion. In addition to the Epo-Bcl-XL glycation of haemoglobin and deamination of cytoskeletal com- pathway, apoptosis of erythroid precursors can also occur via ponents such as protein 4.1. Microvesiculation, which repre- activation of the Fas receptor (FasR), which is present on both sents an effective means of removing damaged or ineffective early and late erythroid precursors. Crucially, its activating lig- red cell components, results in the continual loss of small frag- and (FasL) appears only on late erythroblasts. Binding of FasL ments of the red cell, producing cells that are more dense to FasR activates proteolytic caspases that cleave intracellular and less deformable in the microcirculation than their younger proteins, including GATA1. The absence of GATA1 leads to counterparts. apoptosis of erythroid precursors, and may also prevent induc- As well as these changes, however, specific cues for clear- tion of the antiapoptotic protein Bcl-XL . Thus, FasL-activated ance of the aged red cell from the circulation are thought to GATA1 cleavage results in loss of Bcl-XL and increased apop- exist. Phosphatidylserine and phosphatidylethanolamine are key tosis of erythroid precursors – but only in the context of ade- constituents of the red cell membrane that are normally con- quate numbers of their more mature counterparts. The result fined to the inner aspect of the lipid bilayer. This asymme- is a negative feedback loop to control erythroid proliferation try is maintained by an ATP-dependent aminophospholipid (Figure 2.7). translocase. In senescent cells, phosphatidylserine is found in 18 Chapter 2 Erythropoiesis the outer leaflet, where it is able to bind to macrophages in form of the receptor that circulates in a complex with transferrin. the liver and spleen and prompt erythrophagocytosis. Other Erythroblasts rather than the reticulocytes are the main source senescence-related neo-antigens on the red cell surface may be of soluble TfR (Figure 2.2) and, when iron stores are adequate generated by the clustering of membrane proteins such as band and available, measuring the level of soluble TfR (normal range 3, thought to occur in response to oxidative change. Follow- 5.0 ± 1.0 mg/mL) is a good guide to the total level of erythro- ing erythrophagocytosis, the red cell components, including the poiesis. Soluble TfR levels are increased when erythropoiesis is iron from its haem groups, are recycled for subsequent red cell stimulated and decreased when it is diminished. The interpre- synthesis. tation of soluble TfR levels is complicated in iron deficiency as It is possible to compensate for a small decrease in the life- this condition independently raises the level of soluble TfR. The span of mature red cells through increased Epo production and reticulocyte count (0.5–2.0% or 25–75 × 109 /L) is raised in pro- a reduction in apoptosis. Even when compensatory mechanisms portion to the degree of anaemia when erythropoiesis is effective are able to drive increased red cell synthesis, more significant (e.g. uncomplicated response to bleeding), but is relatively low reductions in red cell survival will lead to haemolytic anaemia, when erythropoiesis is ineffective (e.g. β thalassaemia [Chap- as discussed in Chapters 6–9. Systemic illness may also limit ter 6], myelodysplasia [Chapter 25] and CDA [Chapter 10]) or red cell survival, with the short red cell lifespan in uraemia and when an extrinsic abnormality prevents a normal response (e.g. the anaemia of inflammation being well described, if not well nutritional deficiency; Figure 2.8). understood. The output of the system, the red cell mass, can be accurately measured by radioactive dilution techniques using 51 Cr, but can often be reliably estimated from the haematocrit or con- Assessing erythropoiesis centration of haemoglobin. Changes in red cell size, shape and haemoglobin content, often reflected in the red cell morphol- Erythropoiesis is disturbed in a wide range of primary haemato- ogy, may provide important guides to specific abnormalities in logical conditions and, to some extent, in almost all multisystem red cell maturation (e.g. haemoglobinopathies, thalassaemia, diseases. Defects may arise in the production of committed nutritional deficiencies). If the red cell mass is appropriate to erythroid progenitors; in the response of the oxygen sensing meet the demands for oxygenation, then Epo production will be system, mediated via Epo and EpoR, and its effect on terminal suppressed and the serum level will be in the normal range. If erythroid differentiation and maturation; or in the red cell there is inadequate oxygenation, the level of Epo will generally output achieved, which in turn exerts a major influence on be raised in proportion to the degree of anaemia, unless there the production of Epo, thus completing the regulatory loop is some impediment to Epo production (e.g. chronic renal (Figure 2.8). failure, anaemia of chronic disease). For any given degree of Simple tools are available to test the circuit in a logical manner. anaemia the level of Epo in the blood may vary, depending on Erythroid activity may be estimated by the ratio of myeloid pre- the underlying pathology. For example, levels tend to be very cursors to erythroid precursors in the marrow (normally about high in aplastic anaemia and less than anticipated in thalas- 4:1, but with a very broad normal range). In the past, total ery- saemia. This may reflect the different numbers of precursors thropoiesis was measured accurately using radioactive (59 Fe) in the marrow that are able to bind available Epo molecules, ferrokinetic assays. The plasma iron turnover measures the total thus altering the number of free Epo molecules that are (i.e. effective and ineffective) amount of erythropoiesis, whereas measured. the red cell iron utilization assay measures the degree of effective In clinical practice, erythropoiesis can be assessed by exam- erythropoiesis. To a large extent, these two parameters can now ination of the full blood count and red cell indices, the reticu- be assessed much more simply by measuring the levels of sol- locyte count and examination of the peripheral blood film and uble TfR and the reticulocyte count. Soluble TfR is a truncated bone marrow morphology. The usefulness of these and other Epo HIF Oxygenation BFU-E Haematinics CFU-E Retic Fe, B12, folate Red ME cell Hct Figure 2.8 Summary of the regulation of erythropoiesis with mass the key points for assessment boxed in blue. ME denotes assessment of the myeloid/erythroid ratio in the bone marrow. Hct, haematocrit. TfR 19 Postgraduate Haematology specialized tests in the investigation of anaemia will depend on the initial assessment of the patient by a detailed history and Selected bibliography physical examination. The specific tests used to diagnose the inherited or acquired disorders that may perturb each phase of Palis J. (2014) Primitive and definitive erythropoiesis in mammals. Frontiers in Physiology 28(5): 3. erythropoiesis are highlighted in the relevant subsequent chap- Kerenyi MA, Orkin SH (2010) Networking erythropoiesis. Journal of ters as each anaemia is discussed in detail. Experimental Medicine 207(12): 2537–41. Crispino JD, Weiss MJ. (2014) Erythro-megakaryocytic transcription factors associated with hereditary anemia. Blood 123(20): 3080–8. Haase VH. (2013) Regulation of erythropoiesis by hypoxia-inducible Conclusions factors. Blood Reviews 27(1): 41–53. Dzierzak E, Speck NA (2008) Of lineage and legacy: the development The process of erythropoiesis needs to be tightly controlled to of mammalian haematopoietic stem cells. Nature Immunology 9: 129–36. ensure adequate oxygen delivery to the tissues across a range Orkin SH, Zon LI (2008) Hematopoiesis: an evolving paradigm for stem of physiological and pathological conditions. An appreciation cell biology. Cell 132: 631–44. of the mechanisms of erythroid specification, maturation and Cantor AB, Orkin SH (2002) Transcriptional regulation of erythro- feedback has already led to an improved understanding of the poiesis: an affair involving multiple partners. Oncogene 21: 3368–76. pathophysiology of many inherited and acquired anaemias, and Unger FE, Thompson AM, Blank MJ et al. (2010) Erythropoiesis- provides a framework for the logical investigation and treatment simulating agents: time for a reevaluation. The New England Journal of anaemia in the clinical setting. of Medicine 362: 189–192. 20 CHAPTER 3 Iron metabolism, iron deficiency and disorders of haem synthesis 3 Clara Camaschella1 , A Victor Hoffbrand2 and Chaim Hershko3 1 Vita-Salute University, Milan, Italy 2 University College London, London, UK 3 Shaare Zedek Medical Center, Jerusalem, Israel Introduction Distribution of body iron Iron is essential for many metabolic processes. It shares with The amount of iron in the adult human body is normally about other transition metals two properties of particular importance 50 mg/kg in males and 40 mg/kg in females. The largest compo- in biology: the ability to exist in more than one relatively sta- nent is circulating haemoglobin, with 450 mL (1 unit) of whole ble oxidation state and the ability to form many complexes. Its blood containing about 200 mg of iron (Figure 3.1). Much of ability to exist in both ferric and ferrous states underlies its role the remainder is contained in the storage proteins ferritin and in critical enzyme reactions concerned with oxygen and elec- haemosiderin. These are found mainly in the macrophages of the tron transport and the cellular production of energy. As well as liver, spleen and bone marrow (which gain iron from breaking physiologically active iron compounds, many of which are haem down red cells), and in parenchymal liver cells (which normally proteins, there are also specialized proteins of iron absorption, obtain most of their iron from the plasma iron-transporting transport and storage. The latter are necessary to enable iron to protein transferrin). remain in solution at neutral pH, at which ferric iron is insolu- ble, and to limit the potential toxicity of this reactive metal. The insolubility of ferric iron also means that although the Earth’s crust contains approximately 4% iron and iron may be plenti- Proteins important in iron metabolism ful in the diet, much of this is not bioavailable. As a result, the Haem proteins and iron-containing enzymes body is limited in the adjustments it can make to excessive loss of iron, which frequently occurs due to haemorrhage, and iron Haemoglobin contains four haem groups linked to four globin deficiency is the most common cause of anaemia throughout chains, and can bind four molecules of oxygen. Myoglobin the world, affecting in the order of 2 billion people. The gen- accounts for 4–5% of body iron and has a single haem group eral need to conserve the metal is reflected in the absence of any attached to its one polypeptide chain. It has a higher affinity for physiological mechanism for excretion of iron, control of iron oxygen than haemoglobin and behaves as an oxygen reserve in balance being at the level of iron absorption. This is important muscles. The mitochondria contain a series of haem and non- in the rarer, but potentially fatal disorders of iron overload (see haem iron proteins (including the cytochromes a, b and c, suc- Chapter 4). cinate dehydrogenase and cytochrome oxidase) that form an Postgraduate Haematology, Seventh Edition. Edited by A Victor Hoffbrand, Douglas R Higgs, David M Keeling and Atul B Mehta. © 2016 John Wiley & Sons, Ltd. Published 2016 by John Wiley & Sons, Ltd. 21 Postgraduate Haematology Iron homeostasis Macrophages 500 mg Hepatocytes 250 mg Gut RE stores Liver stores Absorption 1 mg/day Ineffective erythropoiesis Plasma (transferrin) 4 mg Loss 1 mg/day Figure 3.1 Iron homeostasis. The major compartments of iron in a 70-kg man. Iron supply for Red cell Bone Tissues erythropoiesis and release of iron from senescent red haemoglobin marrow cells dominate internal iron exchange. RE, Iron-containing 2500 mg Erythroblasts enzymes 150 mg reticuloendothelial. The dotted line indicates the low 150 mg Myoglobin 300 mg proportion of iron that derives from ineffective Iron receptor tissues and functional iron compounds erythropoiesis in normal subjects. electron transport pathway responsible for the oxidation of small amount of ferritin normally present in serum contains intracellular substrates and the simultaneous production of little iron and consists almost exclusively of L subunits. It is adenosine triphosphate (ATP). Haem is an essential compo- also heterogeneous, owing to glycosylation. This glycosylation nent of microsomal and mitochondrial cytochrome P450, which and the direct relationship of serum concentration to storage is concerned with hydroxylation reactions (including drug iron in macrophages suggest that serum ferritin is secreted by detoxification by the liver), and of cyclooxygenase, involved macrophages in response to changing iron levels. in prostaglandin synthesis. Other haem proteins include the Haemosiderin, unlike ferritin, is a water-insoluble, crystalline, enzymes catalase and lactoperoxidase, which are concerned with protein–iron complex that is visible by light microscopy when peroxide breakdown, and tryptophan pyrrolase, involved in the stained by the Prussian blue (Perls’) reaction. It has an amor- oxidation of tryptophan to formylkynurenine. There is a smaller phous structure, with a higher iron/protein ratio than ferritin, group of iron sulfur proteins (e.g. xanthine oxidase, reduced and is probably formed by the partial digestion of ferritin aggre- nicotinamide adenine dinucleotide dehydrogenase and aconi- gates by lysosomal enzymes. In normal subjects, the majority of tase). Iron is also necessary for the function of ribonucleotide storage iron is present as ferritin, and haemosiderin is predom- reductase, a key enzyme in DNA synthesis. inantly found in macrophages rather than hepatocytes. In iron overload, the proportion present as haemosiderin increases con- siderably in both cell types. Ferritin and haemosiderin Ferritin is the primary iron storage protein and provides a Transferrin and transferrin receptors reserve of iron. It consists of an approximately spherical apopro- tein shell enclosing a core of ferric hydroxyphosphate (up to 4000 Transferrin is a single-chain polypeptide present in plasma (1.8– iron atoms). Human ferritin is made up from 24 subunits of 2.6 g/L) and extravascular fluid (Table 3.1). It has a plasma half- two immunologically distinct types: H and L. There are multiple life of 8–11 days. The protein is synthesized predominantly by gene copies, which are mostly pseudogenes. An intronless gene the liver, synthesis being inversely related to iron stores. Two codes for mitochondrial ferritin, an H-type ferritin. The internal atoms of ferric iron can bind to each molecule. Although trans- cavity of the ferritin molecule communicates with the exterior ferrin contains only about 4 mg of body iron at any time, it is via six channels, through which ferrous iron may enter (to inter- vital to iron transport, with over 30 mg iron passing through act with a ferroxidase centre on the ferritin H subunit) or leave this compartment each day (Figure 3.1). The uptake of iron (after reduction, e.g. by dihydroflavins or ascorbic acid). from transferrin requires that the protein is attached to spe- The way in which ferritin iron is mobilized is poorly under- cific receptors on the cell surface. The transferrin receptor gene stood, and a process in which the entire ferritin molecule is (TFRC) codes for TFR1, a transmembrane protein (identified degraded within lysosomes prior to iron release has been sug- as CD71), each molecule of two subunits binding one trans- gested. Variation in the proportion of H to L subunits explains ferrin molecule. A second receptor, TFR2, also binds trans- the heterogeneity of ferritin from different tissues on isoelec- ferrin (Table 3.1). Through their binding with HFE, TFR1 tric focusing: L-rich ferritins (from spleen and liver) are more and TFR2 are involved in regulating hepcidin synthesis (see basic than H-rich ferritins (from heart and red cells). The Figure 3.2). 22 Table 3.1 Iron transport proteins, oxidoreductases, storage proteins and regulators. Protein Chromosome Tissue Mutations (gene) location expression Structure Function Regulation and disease Duodenal cytochrome 2q31 Duodenal TMP, 6TMD Ferric reductase Fe Not known b 1 (CYBRD1) enterocytes (hepcidin) DMT1 (SLC11A2) 12q13 Widespread TMP, 568 aa Fe uptake Fe (3′ -IRE) Mk mouse, Belgrade rat Human microcytic anaemia Hemojuvelin (HFE2) 1q21.2 Liver, heart, muscle Membrane-bound receptor or Regulator of hepcidin ? Juvenile HC secreted protein synthesis Frataxin (FXN) 9q21.11 Heart, spinal cord, Mitochondrial protein, Mitochondrial iron ? Friedreich ataxia cerebellum donor FLVCR (FLVCR1) 1q32.3 Erythroid Major facilitator family Receptor for feline ? Not known leukaemia virus C; haem export Ferroportin 2q32 Liver, spleen, TMP, 571 aa, 9TMD Fe export Fe (5′ -IRE) Human HC, autosomal (SLC11A3) enterocyte dominant Hepcidin (HAMP) 19q13.1 Plasma (liver) Mature peptide of 25 aa Regulator of iron Fe (BMP6), Juvenile HC (digenic homeostasis IL6 HC) Hephaestin (HEPH) Xq11–q12 Enterocyte TMP, 1TMD, copper protein Fe2+ oxidase – Sla mouse: iron with homology to deficiency anaemia caeruloplasmin Haemochromatosis 6p21.3 Widespread HLA class I heavy chain Regulates TFRC, iron ? Human HC, autosomal (HFE) uptake and hepcidin recessive expression Mitoferrin 1 8p21 Erythroid, liver, Mitochondrial inner Mitochondrial iron ? Not known (SLC25A28) skeletal muscle, membrane importer heart STEAP3 2q14.2 Erythroid, placenta Six-transmembrane epithelial Ferric reductase (also ? Human microcytic (with TFR1) antigen of the prostate-3 reduces copper) anaemia Transferrin receptor 3q26.2–qter Widespread: TMP dimeric polypeptide Binds transferrin Fe (3′ -IRE) Lethal in knockout (TFR) highest number mouse in erythroblasts Transferrin receptor 2 7q22 Liver, erythroid 60% similarity in extracellular Binds transferrin, iron No IRE Human HC, autosomal (TFR2) cells domain to TFR homeostasis, recessive regulator of hepcidin synthesis Transferrin (TF) 3q21 Plasma, Single-chain polypeptide, Iron transport Iron stores Atransferrinaemia, extravascular glycoprotein autosomal reeessive space (Continued) Table 3.1 (Continued) Protein Chromosome Tissue Mutations (gene) location expression Structure Function Regulation and disease Ferritin heavy 11q13 Widespread, Subunit of ferritin Iron storage (catalytic Fe (5′ -IRE) Autosomal dominant Fe chain (FTH1) cytosolic subunit for iron overload (very rare) incorporation) Ferritin light chain 19q13.3–q13.4 Widespread, Subunit of ferritin Iron storage Fe (5′ -IRE) Hyperferritinaemia and (FTL) cytosolic cataract syndrome Neuroferritinopathy∗ IRP1 (ACO1) 9p21.1 Widespread Cytoplasmic, with 4Fe–4S Regulation of synthesis Cell iron Not known cluster of FTH, FTL, TFRC, DMT1, ferroportin, ALAS2 IRP2 (IREB2) 15 Widespread Cytoplasmic, no 4Fe–4S As IRP1 Cell iron Not known cluster Matriptase-2 22q13.1 Mainly liver Type II transmembrane Cleaves hemojuvelin Unknown Human IRIDA, (TMPRSS6) serine protease autosomal recessive Erythroferrone 2q37.3 Erythroid, muscle Secreted protein Hepcidin inhibitor Erythropoietin Not known ∗ In hyperferritinaemia/cataract syndrome, mutations affect the 5′ -UTR IRE. In neuroferritinopathy, mutations affect FTL coding sequences. aa, amino acid; BMP, bone morhphogenetic protein 6; HC, haemochromatosis; IL6, interleukin 6; IRIDA, iron refractory iron deficiency anaemia; IRE, iron response element; IRP, iron regulatory protein; TMD, transmembrane domain; TMP, transmembrane protein. Chapter 3 Iron metabolism, iron deficiency and disorders of haem synthesis Erythropoiesis Other tissues Iron FPN Duodenum absorption Transferrin saturation Iron release from macrophage FPN Figure 3.2 Stimulatory and inhibitory signals of ? hepcidin regulation. Hepcidin, as well as TFR1 TFR2 HJV BMP6 hemojuvelin (HJV), transferrin receptor 2 (TFR2) TF/HFE HFE and HFE, are all produced in the hepatocyte. High plasma iron and inflammation stimulate hepcidin SMAD1, 5, 8 Matriptase synthesis. This is mediated by SMADs and 2 +4 STAT3, respectively. Conversely, low plasma iron, Hepcidin increased rates of erythropoiesis (including Hepatocyte ineffective erythropoiesis) and hypoxia inhibit hepcidin production. This is mediated by matriptase and ERFE. Hepcidin binds ferroportin (FPN), causing its destruction and so inhibits iron STAT3 absorption and iron release from macrophages into plasma and from intracellular compartments. BMP, bone morphogenetic protein; ERFE, IL-6 ?ERFE erythroferrone; The ? indicates uncertainty of the ERFE function in humans; GDF-15 may be the human equivalent of ERFE. Erythroblasts Lactoferrin is a glycoprotein that is structurally related to divalent metal cations, including Mn2+ , Co2+ , Zn2+ , Cu2+ and transferrin. It is found in milk and other secretions and in Pb2+ . neutrophils. It has a bacteriostatic action at secreting surfaces by depriving microorganisms of the iron needed for their Ferroportin growth. This transmembrane domain protein is the basolateral trans- porter of iron, essential for iron release from macrophages, the intestinal absorptive enterocyte, hepatocytes and placental syn- Divalent metal transporter 1 cytiotrophoblasts. It is also present in intracellular compart- Divalent metal transporter (DMT)1 is an electrogenic pump that ments. Caeruloplasmin is required for the cell-surface local- requires proton cotransport in order to transfer Fe2+ across cell ization of ferroportin, whose concentration is controlled by membranes. This occurs at the apical membrane and subapical hepcidin, which triggers its internalization and degradation in endosomes of the duodenal enterocyte and the transferrin-cycle lysosomes. endosome, both of which have a low pH. The intestinal DMT is Other proteins produced by different mRNA splicing from that which produces endosomal DMT1. DMT1 expression is upregulated in iron defi- The roles in iron metabolism of hemojuvelin (HJV), bone ciency (see later) and may be involved in absorption of other morphogenetic protein-6 (BMP-6), small mothers against 25 Postgraduate Haematology decapentaplegic (SMADs), ferrioxidative and reduction of the hepcidin gene by activating STAT3 (signal transducer and enzymes, and caeruloplasmin are discussed under the head- activator of transcription 3) and its binding to a regulatory ele- ings of hepcidin regulation, iron absorption, iron uptake by ment in the hepcidin promoter. It may converge on a final shared erythroid cells and haem synthesis. SMAD-4-dependent pathway. The hepcidin response is remarkably rapid. In humans, iron ingestion results in a sharp increase in urinary hepcidin Hepcidin excretion within 12–24 hours of starting treatment. Likewise, Hepcidin has a central role in the regulation of iron metabolism infusion of recombinant IL-6 results in significant increase and absorption (Figure 3.2). A product of the HAMP gene in urinary hepcidin and decreased serum iron and trans- (Table 3.1), it is a small peptide (25 amino acids) released ferrin saturation within 2 hours of infusion. These obser- from a large prepropeptide of 84 amino acids. It is predomi- vations imply that hepcidin expression is directly controlled nantly expressed in the liver. It regulates iron homeostasis by by serum iron (probably by transferrin saturation) and IL- binding to cell-surface ferroportin, causing its degradation in 6 and not by long-term gradual accumulation of iron in lysosomes. It therefore acts to inhibit iron absorption, iron tissues. release from macrophages and iron transport across the pla- centa. It is bound in plasma to α2 -macroglobulin and the Response to anaemia and hypoxia major route of clearance is the kidney. Hepcidin can be mea- Hepcidin levels are reduced or undetectable in iron deficiency sured in serum or urine by ELISA or mass spectrometry-based anaemia. Iron absorption is accelerated in iron deficiency, techniques. ineffective erythropoiesis and hypoxia (Figure 3.2) likely through common mechanisms mediated by the ‘erythroid regu- Regulation of hepcidin expression lator’. Erythroid precursors secrete growth differentiation factor- The regulation of hepcidin expression is transcriptional. Hep- 15 (GDF-15) and TWSG1, two cytokines of the TGF-β family, cidin expression is increased in response to raised serum iron, which have been proposed to inhibit hepcidin production in the iron overload and inflammation, and is suppressed by iron defi- liver. Serum concentrations of GDF-15 are greatly increased in ciency, hypoxia and increased erythropoietic activity. Under thalassaemia major and other conditions associated with inef- basal conditions, expression depends on signalling through fective erythropoiesis. However, the role of these cytokines in the BMP/SMAD pathway (Figure 3.2). HJV is a member of vivo is uncertain. Recently erythroferrone, a novel protein, has the repulsive guidance molecules (RGM) family that is highly been identified that plays the role of physiological erythroid reg- expressed in liver, skeletal muscles and the heart. It is either ulator in mice. It is released by erythroid precursors and sup- associated with cell membranes through a glycosylphosphatidy- presses hepcidin after bleeding and erythropoietin treatment in linositol anchor or released as a soluble form. Membrane-bound order to increase intestinal iron absorption and macrophage iron HJV participates in the pathway regulating hepcidin expression release according to the erythropoietic needs. It remains to be as a BMP coreceptor, whereas soluble HJV antagonizes BMP- seen if this hormone plays a similar role in human erythro- 6. BMP-6 is the master hepcidin activator in vivo in murine poiesis. models. Hypoxia directly increases iron absorption through increased HFE and TFR2 are also involved in hepcidin expression (Fig- duodenal iron importer DMT1 and indirectly through hep- ure 3.2). HFE is able to bind TFR1 and TFR2. During low or cidin suppression by increased erythropoiesis. The same mech- basal serum iron conditions, HFE and TFR1 exist as a complex at anism may contribute to the harmful accumulation of iron the plasma membrane, TFR1 serving to sequester HFE to silence in response to chronic anaemia associated with ineffective its activity. Diferric serum transferrin (Fe2+ -TF) competes with erythropoiesis in thalassaemia and other dyserythropoietic HFE for binding to TFR1. Increased serum transferrin satura- anaemias. tion therefore results in dissociation of HFE from TFR1. Acting as an iron sensor, HFE then binds to TFR2 and conveys the Fe2+ - Matriptase-2 TF status to the signal transduction effector complex. HJV binds to BMP, then phosphorylates SMADs to form a SMAD-1/-5/-8– This is a type 2 member of the transmembrane serine SMAD-4 complex, which translocates to the nucleus and stimu- protease family mainly expressed in the liver. Membrane- lates hepcidin production by activating its promoter. In keeping bound matriptase-2 regulates hepcidin expression by cleaving with this model, genetic mutations of HFE, TFR2, HJV and hep- membrane-bound HJV, releasing soluble HJV fragments. The cidin all result in haemochromatosis with low serum hepcidin factors that regulate matriptase-2 expression need to be elu- levels (see Chapter 4). Iron levels also seem to control BMP-6 cidated. Matriptase-2 activity over-rides all known activating production. stimuli of hepcidin synthesis. Homozygous or compound het- A second type of transcriptional hepcidin regulation occurs in erozygous TMPRSS6 mutations in humans and homozygous inflammation. Interleukin (IL)-6 and IL-1β induce transcription inactivation of the gene in mice result in marked upregulation 26
Enter the password to open this PDF file:
-
-
-
-
-
-
-
-
-
-
-
-