Microenvironments in haemopoietic and lymphoid differentiation

Microenvironments in haemopoietic and lymphoid differentiation

Ciba Foundation symposium 84 In honour of Professor John Humphrey, FRS

1981

Pitman London

The Ciba Foundation is an international scientific and educational charity. It was established in 1947 by the Swiss chemical and pharmaceutical company of CIBA Limited-now CIBA-GEIGY Limited. The Foundation operates independently in London under English trust law.

The Ciba Foundation exists to promote international cooperation in medical and chemical research. It organizes international multidisciplinary meetings on topics that seem ready for discussion by a small group of research workers. The papers and discussions are published in the Ciba Foundation series.

The Foundation organizes many other meetings, maintains a library which is open to all graduates in science or medicine who are visiting or working in London, and provides an information service for scientists. The Ciba Foundation also functions as a centre where scientists from any part of the world may stay during working visits to London.


Ciba Foundation symposium 84 In honour of Professor John Humphrey, FRS

1981

Pitman London

JOHN HUMPHREY, FRS

Photograph by Department of Medical Illustration, Royal Postgraduate Medical School, London

0 Ciba Foundation 1981

ISBN 272-79636-0

Published in 1981 by Pitman Books Ltd, London. Distributed in North America by CIBA Pharmaceutical Company (Medical Education Administration), Summit, NJ 07006, USA.

Suggested series entry for library catalogues: Ciba Foundation symposia.

Ciba Foundation symposium 84 x + 348 pages, 70 figures, 26 tables

British Library Cataloguing in publication data: Microenvironments in haemopoietic and lymphoid

differentiation. - (Ciba Foundation symposium;

1. Blood cells - Congresses 2. Cell differentiation - Congresses I. Porter, Ruth 111. Series 596’.01’13 QH607

84)

11. Whelan, Julie

Text set in 10/12 pt Linotron 202 Times, printed and bound in Great Britain at The Pitman Press, Bath

Contents

Symposium on Microenvironments and cell differentiation, held at the Ciba Foundation, London, 11-13 November 1980 Editors: Ruth Porter (Organizer) and Julie Whelan

MELVYN F. GREAVES (Chairman) Introduction: signals, receptors and repertoire in haemopoietic differentiation 1

L. WEISS Haemopoiesis in mammalian bone marrow 5 Discussion 15

T. M. DEXTER Self-renewing haemopoietic progenitor cells and the factors controlling proliferation and differentiation Discussion 32

22

T. D. ALLEN Haemopoietic microenvironments in vitro: ultrastructural aspects 38 Discussion 60

D. G. OSMOND, M. T. E. FAHLMAN, G. M. FULOP and D. M. Regulation and localization of lymphocyte production in the RAHAL

bone marrow 68 Discussion 82

E. C. GORDON-SMITH and M. Y. GORDON Environmental factors in haemopoietic failure in humans 87 Discussion 103

M. GREAVES, J. ROBINSON, D. DELIA, R. SUTHERLAND, R. NEWMAN and C. SIEFF Mapping cell surface antigen expression on haemopoietic progenitor cells using monoclonal antibodies 109 Discussion 121

vii

viii CONTENTS

J. W. SCHRADER, P. F. BARTLEm, I. CLARK-LEWIS and A. W. BOYD Lymphoid differentiation in vifro 130 Discussion 145

R. V. ROUSE and I. L. WEISSMAN Microanatomy of the thymus: its relationship to T cell differentiation Discussion 173

161

E. J. JENKINSON Expression and function of major histocompatibility complex antigens in the developing thymus: studies on normal and nude mice 178 Discussion 187

G. JANOSSY, J . A. THOMAS, G. GOLDSTEIN and F. J. BOLLUM The human thymic microenvironment 193 Discussion 207

J.-F. BACH and M. PAPIERNIK Cellular and molecular signals in T cell differentiation 215 Discussion 230

General discussion: tolerance and diversification of the T cell repertoire 236

P. NIEUWENHUIS, N. A. GASTKEMPER and D. OPSTELTEN Histophysiology of follicular structures and germinal centres in relation to B cell differentiation 246 Discussion 259

G. G. B. KLAUS and A. KUNKL The role of germinal centres in the generation of immunological memory 265 Discussion 275

B. M. BALFOUR, H. A. DREXHAGE, E. W. A. KAMPERDIJK and E. Ch. M. HOEFSMIT Antigen-presenting cells, including Langerhans cells, veiled cells and interdigitating cells 281 Discussion 298

J. H. HUMPHREY Differentiation of function among antigen-presenting cells 302 Discussion 313

Final general discussion T cell subsets and T cell function 322

ix CONTENTS

Decision-making in development 325

M. F. GREAVES Chairman’s summing-up 335

Index of contributors 337

Subject index 339

Participants

T. D. ALLEN Department of Ultrastructure, Paterson Laboratories, Christie Hospital & Holt Radium Institute, Wilmslow Road, Withington, Manchester, M20 9BX, UK

J.-F. BACH Unite de Recherches Nephrologiques de I’Inserm U 25, HBpital Necker, 161 Rue de Skvres, 75730 Paris Cedex 15, France

B. M. BALFOUR Division of Immunological Medicine, Clinical Research Centre, Watford Road, Harrow, Middlesex, HA1 3UJ, UK

A. W. BURGESS Ludwig Institute for Cancer Research, Melbourne Tumour Biology Unit, Post Office, Royal Melbourne Hospital, Victoria 3050, Australia

T. M. DEXTER Paterson Laboratories, Christie Hospital & Holt Radium Institute, Wilmslow Road, Withington, Manchester, M20 9BX, UK

S. GORDON Sir William Dunn School of Pathology, University of Oxford, South Parks Road, Oxford, OX1 3RE, UK

E. C. GORDON-SMITH Department of Haematology, Royal Postgradu- ate Medical School, Hammersmith Hospital, Ducane Road, London W12 OHS. UK

M. F. GREAVES Membrane Immunology Laboratory, Imperial Cancer Research Fund Laboratories, PO Box No. 123, Lincoln’s Inn Fields, London WC2A 3PX, UK

P. L. GREENBERG Department of Medicine, Stanford University School of Medicine, Stanford, California 94305, USA

B. L. M. HOGAN Imperial Cancer Research Fund, Mill Hill Laboratories, Burtonhole Lane, London NW7 lAD, UK

X

PARTICIPANTS xi

J. C. HOWARD Agricultural Research Council, Institute of Animal Phy- siology, Babraham, Cambridge, CB2 4AT, UK

J. H. HUMPHREY Department of Immunology, Royal Postgraduate Medical School, Hammersmith Hospital, Ducane Road, London W12 OHS, UK

G. JANOSSY Department of Immunology, The Royal Free Hospital, Pond Street, Hampstead, London NW3 2QG, UK

E. J. JENKINSON Department of Anatomy, University of Birmingham, Medical School, Vincent Drive, Birmingham, B15 2TJ, UK

G. G. B. KLAUS Division of Immunology, National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 lAA, UK

R. G. MILLER The Ontario Cancer Institute, 500 Sherbourne Street, Toronto, Ontario, Canada M4X 1K9

P. NIEUWENHUIS Histologisch Laboratorium, Rijksuniversiteit Gro- ningen, Oostersingel 69/1, 9713 EZ Groningen, The Netherlands

D. G. OSMOND Department of Anatomy, Strathcona Anatomy and Dentistry Building, McGill University, 3640 University Street, Montreal, Quebec, Canada H3A 2B2

M. PAPIERNIK Unit6 de Recherches NCphrologiques de 1’Inserm U 25, HBpital Necker, 161 Rue de S&vres, 75730 Paris Cedex 15, France

R. V. ROUSE Department of Pathology, Stanford University School of Medicine L235, Stanford, California 94305, USA

J. W. SCHRADER The Walter & Eliza Hall Institute of Medical Research, Post Office, Royal Melbourne Hospital, Victoria 3050, Australia

L. WEISS Department of Animal Biology, School of Veterinary Medicine, University of Pennsylvania, 3800 Spruce Street, Philadelphia, Pennsylva- nia 19104, USA

Introduction: Signals, receptors and repertoire in haemopoietic differentiation

MELVYN F. GREAVES

Membrane Immunology Laboratory, Imperial Cancer Research Fund Laboratories, PO Box No. 123, Lincoln’s Inn Fields, London WCZA 3PX, UK

The outstanding, unresolved issue in biology is the nature of the molecular mechanisms underlying selective gene expression in differentiation. The haemopoietic system offers no new conceptual challenges in this respect but may provide a particularly convenient system for analysis, in addition to its interesting idiosyncrasies that are the concern of experimental haematologists and immunologists.

Developmental biologists studying embryogenesis and morphogenesis, principally in vertebrate species, have for decades argued over the role of the cellular environment in determining maturation pathways. The concepts discussed are frequently enmeshed in a tangled web of semantics and frustrated by inadequate technology. Transplantation systems, however, some of ex- traordinary elegance (Le Douarin et a1 1975), strongly indicate that cells adopt or change developmental options according to their surroundings. Almost all of these experiments involve the gross analysis of non-clonal cell populations and are subject to the obvious criticism that cell selection cannot easily be discounted. Some studies with clonally identified populations do, however, suggest that the local environment can provide a directive or even re-directive influence (e .g. Eguchi 1976). An alternative possibility is that local environments can be merely selective or ‘permissive’ and provide the appropriate environment for cells that make or re-make the ‘correct’ choice by some random process (SaxCn 1977). Although both types of mechanism may exist, a distinction between them will ultimately be important. Certainly there is no doubt that environment has a profound influence on developmental response, as evidenced most dramatically by the behaviour of teratocarci- noma cells (Mintz 1978).

I981 Microenvironments in haemopoietic and lymphoid differentiation. Pitman Medical, London (Ciba Foundation symposium 84) p 1-4

1

2 GREAVES

In the haemopoietic system, as in other multi-lineage tissues (e.g. brain), it is important to distinguish, at least conceptually, ‘differentiation’ events which concern the commitment to a cell lineage from the subsequent maturation sequence in which this affiliation is expressed in cell phenotype and function. Neither event is understood in molecular terms but they could well turn out to be different. Both types of processes involve genetic ‘decisions’ that will almost certainly be subject, indirectly at least, to a fairly extensive network of control signals emanating from outside the cell. These signals will include short- and long-range messages which are inter- or intra-lineage in transmission and involve public or private transactions-for example, insulin or erythropoietin, respectively. In this general sense it is highly probable that each cell type in any lineage has its own unique ‘Merkwelt’ (sensory world) (Greaves 1975).

The sorts of general question one therefore asks about regulatory signals in a developmental system are (Table 1):

(1) What is the cellular source (transmitter) of the signal(s) and what is the responder (receiver)?

(2) What is the range and location of the signal? (3) What is the ‘informational’ content of the signal: is it dictating

(instructing) or facilitating (permissive)? (4) What is the anatomy of the signal pathway-that is, what molecular

species are used as signals and as complementary receptors for these signals?

TABLE 1 Anatomy of regulatory signalling

1. What is the signal source?

2 . What is the signal range? Cell-cell (a) Cell-surface restricted

Extracellular matrix--collagen, proteoglycans Extracellular diffusible signals

(b) Intercellular-gap junctions

(a) Soluble ligands: ‘hormonal’ regulators (short or long range) (b) Membrane vesicles (short range)

3 . What is the signal ‘content’? ‘Instructive’ or ‘permissive’ (Lamarckian or Darwinian?)

4. What is the structure of the signal and its complemenfary receptor? Same candidate cell surface receptor species as for cell-cell interactions: H-UHLA molecules;

glycosyltransferases; lectins

INTRODUCTION 3

Translating these general enquiries into the specific context of haemopoiesis, some key questions are: how good is the evidence that specific microenvironments/niches really exist and direct lineage-specific differentiation? Is the first gene to be expressed by a lineage-committed cell that for the receptor species which receives the key regulatory signals for ‘early’ maturation of that lineage? Or, alternatively, do individual uncommitted cells express many receptor species for different cell lineage regulators and then respond according to which species happens to become occupied?

I suspect that at present these are questions we are unable to answer. In addition to purified, cloned populations, genetic ‘markers’ of commitment prior to phenotypic expression are clearly required as well as exclusive gene products (e.g. haemoglobin, immunoglobulin) to serve as ideal markers. The susceptibility of particular gene loci to deoxyribonuclease 11 provides one approach (cf. Wallace et a1 1977). In the B cell lineage we may be particularly fortunate since the remarkable rearrangements or splicings of immunoglobulin genes (Brack et a1 1978) provide an excellent ‘marker’ for what may be the first and crucial step in commitment to the B cell lineage. At present this molecular approach is limited in part by the number of cells required and by the rather unlikely possibility of being able to ask the same questions of single cells.

However, progress in purifying several haemopoietic regulator molecules (colony-stimulating activity, erythropoietin, thymic ‘hormones’, etc.), in the development of cell type specific monoclonal antibodies, in the purification or enrichment of stem cells and progenitors by multiparameter cell-sorting methods (Beverley et a1 1980, Goldschneider et a1 1980a, b), and in methods for the long-term maintenance of stem cells with the reproduction of haemopoiesis in vitro, indicates that the severe technical barriers to resolving many of the fundamental questions in haemopoiesis are being rapidly dismantled.

In the immediate future there are some exciting questions to address. One that plays a central role in this symposium is the role of the major histocompatibility complex, particularly H-2 I region/HLA-DR or ‘Ia’ cell surface glycoproteins, in regulating the clonal development and function of T lymphocytes, and perhaps of other haemopoietic cells also. Another important and rich area for exploitation is the pathology of haemopoiesis. Here we are dealing with a two-edged sword. We expect, and indeed know, that pathology can provide vital clues to normal function-for example, the ‘environmental’ deficiency in the aplasia of Steel mice-and provides the (‘dirty’) equivalents to mutants in genetic research. At the same time we look forward with enthusiasm to the basic cellular and molecular biology of haemopoiesis providing new clues to the pathobiology of haemopoiesis, as expressed in aplasia, immunodeficiency and leukaemia.

4 GREAVES

REFERENCES

Beverley PCL, Linch D, Delia D 1980 The isolation of human haemopoietic progenitor cells using monoclonal antibodies. Nature (Lond) 287: 332-333

Brack C, Hirama M, Lenhard-Schuller R, Tonegawa S 1978 A complete immunoglobulin gene is created by somatic recombination. Cell 15: 1-14

Eguchi G 1976 ‘Transdifferentiation’ of vertebrate cells in cell culture. In: Embryogenesis in mammals. Elsevier/Excerpta MedidNorth-Holland, Amsterdam (Ciba Found Symp 40) p

Goldschneider I, Metcalf D, Battye F, Mandel T 1980a Analysis of rat hemopoietic cells on the fluorescence-activated cell sorter. I. Isolation of pluripotent hemopoietic stem cells and granulocyte-macrophage progenitor cells. J. Exp Med 152: 419-437

Goldschneider I, Metcalf D, Mandel T, Bollum FJ 1980b Analysis of rat hemopoietic cells on the fluorescence-activated cell sorter. 11. Isolation of terminal deoxynucleotidyl transferase- positive cells. J Exp Med 152: 438-446

241-253

Greaves MF 1975 Cellular recognition. Chapman & Hall, London Le Douarin NM, Renaud D, Teillet MA, Le Douarin GH 1975 Cholinergic differentiation of

presumptive adrenergic neuroblasts in interspecific chimeras after heterotopic transplantation. Proc Natl Acad Sci USA 72: 728-732

Mintz B 1978 Genetic mosaicism and in vivo analyses of neoplasia and differentiation. In: Saunders GF (ed) Cell differentiation and neoplasia. Raven Press, New York, p 27-53

Saxen L 1977 Directive versus permissive induction: a working hypothesis. In: Lash JW, Burger MM (eds) Cell and tissue interactions. Raven Press, New York, p 1-9

Wallace RB, Dube SK, Bonner J 1977 Localization of the globin gene in the template active fraction of chromatin of Friend leukemia cells. Science (Wash DC) 198: 1166-1168

Haemopoiesis in mammalian bone marrow

L. WEISS

Department of Animal Biology, School of Veterinary Medicine, University of Pennsylvania, 3800 Spruce Street, Philadelphia, Pennsylvania 19104, USA

Abstract The bone marrow supports haemopoiesis of all blood cell types and delivers mature cells to the blood. Haemopoiesis is characterized not only by the differentiation and proliferation of haemopoietic stem cells but by a number of physically associated cell types. These include macrophages, lymphocytes and, when haemopoiesis is intense, a multinucleate branched stromal cell. The venous vasculature of the bone marrow is associated with both haemopoiesis and the delivery of blood cells to the circulation. The wall of the vascular sinus consists of an endothelium lying upon a basement membrane. On the outside surface of the basement membrane lie adventitial cells or pericytes which branch out into the haemopoietic space forming a scaffolding upon which haemopoietic clusters are arranged. These cells move away from the wall of the vascular sinus to permit maturing blood cells to penetrate the endothelium and enter the circulation. Under other circumstances, adventitial cells accumulate fat, becoming the adipocytes of marrow.

Bone has an unsurpassed capacity to concentrate multipotential stem cells (CFU-S) in its marrow and to support every line of haemopoiesis. Haemo- poietic bone marrow is a densely cellular, highly vascular reticular connective tissue. Haemopoietic and connective tissue cells, stroma and vasculature are closely packed and intermembranous junctional complexes are regularly present among all cell types (Weiss 1976, Campbell 1980). The haemopoietic tissue contains a prominent system of arborized, anastomosing venous vessels, the venous sinuses, so extensive that virtually all haemopoiesis is perivenular. Arterial vessels run through the haemopoietic tissue, branching and narrowing to capillaries approximately 4 pm in diameter (Weiss 1965, 1970, 1976).

The venous sinuses are thin-walled vessels 15 to 100pm in diameter whose walls consist of three layers: endothelium, basement membrane and adventitia. The endothelium, like other endothelia, contains transcytotic vesicles,

1981 Microenvironments in haemopoietic and lymphoid differentiation. Pitman Medical, London (Ciba Foundation symposium 84) p 5-21

5

6 WEISS

microfilaments, microtubules, some mitochondria and lysosomes, and junctional complexes (Simionescu 1978). But this endothelium is distinctive. Apertures develop in the cytoplasm of its endothelial cells through which maturing blood cells produced in the perivascular haemopoietic space pass en route to the circulation. Its endothelial cells can be moderately phagocytic. The endothelial cytoplasm is quite responsive in appearance in large-scale haemopoietic cell crossing, becoming vesiculated, dense and thin, or rarified and voluminous, and fenestrated (Weiss 1970, Sakai et all981). The basement membrane is, histochemically, a carbohydrate-protein complex similar in substance to reticular fibres. It is readily extracted, and preserved best in freeze-fracture etch preparations. The adventitia consists of a simple layer of adventitial cells which cover approximately 60% of the outside surface of the venous sinuses in normal rodents and branch into the perivascular haemopoietic tissue in broad sheet-like processes (Weiss 1970). These processes form a reticulum whose interstices are crowded with haemopoietic cells. The processes are also associated with slender argyrophilic extracellular fibres, the reticulurfibres, and presumably produce them. Because these adventitial cells form a reticulum and probably produce reticular fibres they are termed udventitial reticular cells. The perivascular spaces thereby constitute a reticular connective tissue and, as such, can contain macrophages, mast cells and virtually all the cells of the connective tissues. Yet the preponderant cells in haemopoietic bone marrow are the haemopoietic cells.

Adventitial cells have many of the cytological features of their endothelium. Patches of concentrated microfilaments occur in the subplasmalemmal cytoplasm, adjacent to reticular fibres. Adventitial cells are displaced from the vascular wall by haemopoietic cells approaching the wall preparatory to their transmural passage. During heavy cell passage the cover of adventitial reticular cells may be reduced to 20% and the cells may show cytological changes similar to those in endothelium (Weiss 1970, Sakai et al 1981). Adventitial cells may also become fatty and thereby become the adipocytes of marrow. These adipocytes occupy space, displacing haemopoietic cells and dominating the appearance of non-haemopoietic, yellow marrow. They are active metabolically, aromatizing androgens to oestrogens (Frisch et a1 1980), and they may induce granulocytopoiesis (Allen & Dexter 1978).

Haemopoiesis occurs in the perivascular reticular connective tissues. There are as yet no cytochemical markers for the selective demonstration of early haemopoietic stem cells, as CFU-E, CFU-GM, CFU-MEG, CFU-EO, or CFU-S (colony-forming unit-erythroid,-granulocyte-macrophage,-megakaryocyte,-eosinophil or-spleen). These stem cells are among the cells of the marrow morphologically identifiable by conventional transmission electron microscopy only as lymphocytes. The more precisely identifiable haemopoietic cells may have characteristic locations. Megakaryocytes develop on the

BONE MARROW HAEMOPOIESIS 7

adventitial surface of the wall of vascular sinuses, undergoing polyploidy, nuclear polymorphism and other signs of differentiation there. They remain at the wall of vascular sinuses through their life, discharging platelets through mural apertures. Late stage erythroblasts (normoblasts) and metamyelocytes lie near the sinus wall and their early stages lie deeper in the haemopoietic tissue. In general, erythroblastic clusters are closer than granulocytic clusters to the vascular sinuses. In general, moreover, haemopoiesis is most active in the marrow near bone. Where haemopoiesis is induced ectopically-beneath the renal capsule or in the rectus sheath-bone is characteristically produced first and then haemopoietic tissue develops in the marrow. In marrow recovering from aplasia induced by irradiation, haemopoietic colonies first appear contiguous to bone (R. Lambertsen & L. Weiss, unpublished observations 1980) and in autotransplants of bone marrow, haemopoiesis appears after bone develops (Tavassoli & Crosby 1968).

Macrophages, lymphocytes and adventitial reticular cells are in close association with virtually all haemopoietic cell types, perhaps through gap junctions (Campbell 1980). Macrophages have the evident function of phagocytosing haemopoietic cells in ineffective haemopoiesis. They may also produce colony-stimulating factors of various sorts (Metcalf 1979). Cells identifiable in conventional transmission electron microscopy as lymphocytes may include T cells regulating haemopoiesis. Thus the eosinophilopoiesis of infectious disease is dependent upon T helper cells (Sakai et al 1981) and erythropoiesis under certain circumstances may be modulated by T lymphocytes (Weiss 1980, Lipton et al 1980). T lymphocytes can be identified in tissue sections by immunocytochemical or enzymic methods but, as pointed out above, haemopoietic stem cells cannot yet be selectively identified. Therefore lymphocytes in bone marrow studied by conventional transmission electron microscopy are a diverse group of cells which may include T lymphocytes, haemopoietic stem cells, and other null cells. In all phases of haemopoiesis, adventitial reticular cell processes closely surround haemopoietic cells.

Lymphocytes, macrophages, and adventitial cells are cell types associated with haemopoiesis in virtually every cell line and at any level of haemopoiesis (Sakai et al 1981). Whenever haemopoiesis is intense, as in the marked erythropoiesis of spectrin-deficient mice (Brookoff et a1 1981), or in eosinophilopoiesis after a sensitized animal is reinfected with Ascaris suum (Sakai et al 1981), another cell type is present in close association with the haemopoietic cells. This is a branched cell whose perinuclear cisternae are widely dilated and penetrate the cytoplasm as an extensive system of endoplasmic reticulum. This cell type appears to undergo a life cycle. which results in a multinucleate branched cell whose cytoplasm is distinctively dense. This cell is neither fibroblastic, since it is not associated with

8 WEISS

extracellular fibres, nor phagocytic, because it seldom contains phagolyso- somes. While it possesses moderate numbers of lysosomes early in its life cycle, in late stages they are scanty or absent. It differs from the distinctive dendritic cells in murine lymphoid tissue described by Steinman et al (1975) and from the interdigitating cells described by van Ewijk et al (1974) associated with T lymphocytes in thymic-dependent zones of lymphatic tissues. It may represent a subtype of macrophage because of its capacity to form giant cells. At present we recognize it as one of the branched stromal cells of the haemopoietic tissues and are attempting to characterize it more definitively (see Figs. 1-4).

The presence of the haemopoietic cells themselves must be taken into account in assessing the nature of the haemopoietic microenvironment. The presence of lymphocytes and monocyte-macrophages has been noted above. Other cell types appear to influence haemopoiesis. Neutrophils, for example, produce lactoferrin which in addition to its antimicrobial properties appears to inhibit granulocytopoiesis. The distinctive haemopoietic properties of bone marrow, therefore, may depend upon the interaction of each of its cell types, and of cells in bone.

The bone marrow strikingly resembles the red pulp of spleens having vascular sinuses in being a haemopoietic tissue made of a reticular meshwork, and possessing distinctive venous sinuses (Blue & Weiss 1981). But important structural and functional differences occur. The reticular meshwork of the spleen is associated with far heavier reticular fibres than is that of marrow. The circulation through the red pulp of the spleen is anatomically open in that arterial terminals open into the reticular meshwork of the red pulp and the blood flows through this part of the pulp before entering the splenic sinuses. There is no endothelial continuity between artery and venous sinus. In the spleen, moreover, blood is regularly present in the reticular meshwork. But, physiologically, blood may flow as if in closed vessels. In experiments that include washout of the isolated perfused spleen, several types of circulation have been shown to exist, the major one being a functionally closed circulation in which blood flows as efficiently and rapidly as through skeletal muscle (Song & Groom 1971a, b). The marrow, on the other hand, appears to have an anatomically closed circulation with endothelial continuity between arterial vessels and venous sinuses, from scanning electron microscopic studies of plastic casts of the vasculature (Irino et al 1975), vital studies of the

FIG. 1 . Murine bone marrow in heightened eosinophilopoiesis due to secondary infection with Ascaris suum (see Sakai et a1 1981). A dark branched stromal cell, its nucleus at the right margin of the field, extends its cytoplasmic processes among closely surrounding eosinophils. The stromal cell contains two nuclei and various inclusions, vesicles and granules in its cytoplasm. x 7500. (Reduced by a factor of 90%.)

10 WEISS

FIG. 2. Murine bone marrow in heightened eosinophilopoiesis after secondary infection with Ascaris suum (see Sakai et al 1981). The large mononuclear cell in the upper half of the field is a megakaryocyte precursor. A branched stromal cell lies in the lower half. It contains two nuclei, each containing a nucleolus. There is a continuity of nuclear cisternae with rough endoplasmic reticulum (arrows), lysosomes (ly) and swollen mitochondria (m), all characteristic of this cell type. X 150000. (Reduced by a factor of 75%.)


FIG. 3. Murine bone marrow in heightened eosinophilopoiesis after secondary infection with Ascuris suurn (see Sakai et a1 1981). Eosinophil myelocytes surround a macrophage. A branched stromal cell is in the right lower corner. X 15 OOO. (Reduced by a factor of W%.)


marrow circulation (Brlnemark 1959) and, most important, observations that neither blood nor intravascularly injected particles are present in extravascu- lar spaces (Weiss 1965, 1970, 1976). The inter-endothelial slits of the sinuses in spleen are part of the vascular pathway-blood cells regularly pass through them. In the venous sinus of marrow, in contrast, transmural passage of cells is through cytoplasmic apertures in the endothelium and serves the delivery of cells to the circulation, and not blood flow. Blood flow is through an anatomically connected system of blood vessels which includes the venous sinuses as postcapillary venous vessels. The mural structure of marrow sinuses and of splenic sinuses, moreover, is quite different. Further, the red pulp of the spleen is primarily erythroclastic or phagocytic. It can support haemopoiesis, favouring erythropoiesis over granulocytopoiesis. As a haemopoietic tissue the marrow is more diverse than the spleen. It is far more effective in concentrating stem cells, and it supports haemopoiesis in every blood cell line but favours granulocytopoiesis over erythropoiesis (Trentin 1970).

While the bone marrow is the major mammalian haemopoietic organ, providing the microenvironments to support each type of haemopoiesis, the differentiation of the blood cells is not always completed there (Weiss 1981). Monocytes complete their differentiation in spleen, lymph nodes or other connective tissues where they undergo transformation into macrophages, epithelioid cells and giant cells. T lymphocytes undergo much of their differentiation in the thymus and complete it in the spleen. Erythrocytes are delivered from the bone marrow as reticulocytes and before they enter the general circulation undergo some terminal maturation in the spleen (Song & Groom 1971~). Eosinophils in some species, such as the rat, spend several days in the spleen after their release from bone marrow and before entering the general circulation (Spry 1971). Platelets may also complete their

FIG. 4. A schematic rendition of haemopoietic marrow showing characteristic cellular associations in haemopoiesis and the delivery of blood cells into the vasculature. The haemopoietic tissue closely surrounds a vascular sinus. The sinus consists of endothelium (end), basement membrane, and adventitial cells (adv). Apertures are present in the endothelium and haemopoietic cells, en route to the circulation, are passing through them. A macrophage (mcp) in the haemopoietic space rather typically extends a process into the lumen of the vascular sinus. Adventitial cells, the outermost components of the sinus wall, extend processes out into the perivascular haemopoietic tissue, closely surrounding developing cells. Megakaryocytes (meg) lie against the outside surface of the vascular sinus. An erythroblastic islet is on the right (erb islet). It consists of a macrophage surrounded by circlets of erythroblasts which deeply indent its cytoplasm. The more mature erythroblasts are more peripheral in the islet and lie towards the vascular sinus. Macrophages are also associated with granulocyte development, as shown in the left (gran islet). A branched cell type-in addition to macrophages and adventitial cells, which also branch-is the dark stromal cell (str). The enlarged perinuclear space continuous with endoplasmic reticulum is typical. The cell type is shown in this drawing in the multinucleate-giant cell phase of its life cycle.

14 WEISS

maturation in the spleen and are stored there. B lymphocytes, like T lymphocytes, require a brief antigen-dependent phase before final maturation and their entrance into the recirculating pool of lymphocytes. This step in their differentiation is taken in the spleen, after their release from the bone marrow.

REFERENCES

Allen TD, Dexter TM 1978 Cellular interrelationships during in v i m granulopoiesis. Differentia- tion 6: 191-194

Blue J, Weiss L 1981 Electron microscopy of the red pulp of the dog spleen including vascular arrangements, periarterial macrophage sheaths (ellipsoids), and the contractile, innervated reticular meshwork. Am J Anat, in press

Brookoff D, Bernstein S, Weiss L 1981 An electron microscopic study of spectrin-deficient (sphlsph and hdha) mice. Blood, in press

Branemark P 1959 Vital microscopy of bone marrow in rabbit. Scand J Clin Lab Invest 1l:suppl

Campbell F 1980 Gap junctions between cells of bone marrow: an ultrastructural study using tannic acid. Anat Rec 196: 101-117

Frisch RE, Canick JA, Tulchinsky D 1980 Human fatty marrow aromatizes androgen to estrogen. J. Clin Endocrinol Metab 51:394-396

Irino ST, Ono K, Watanabe K, Royata J, Uno N, Murakami T 1975 SEM studies on microvascular architecture, sinus wall, and transmural passage of blood cells in the bone marrow by a new method of injection: replica and non-coated specimens. In: Johari 0, Corvin I (eds) Proc VIII Annual Scanning Electron Microscope Symposium. IIT Research Institute, Chicago, Part I, p 267

Lipton JM, Reinherz EL, Kudisch M, Jackson PL, Schlossman SF, Nathan DG 1980 Mature bone marrow erythroid burst-forming units do not require T cells for induction of erythropoietin-dependent differentiation. J Exp Med 152: 350-360

Metcalf D 1979 Production of colony stimulating factors by lymphoid tissues. In: Cohen S et al (eds) Biology of the lymphokines. Academic Press, New York, p 515-540

Sakai N, Johnstone C, Weiss L 1981 Bone marrow cells associated with heightened eosinophilopoiesis: an electron microscopic study of murine bone marrow stimulated by Ascaris suum. Am J Anat, in press

Simionescu N 1978 The microvascular endothelium: segmented differentiations; transcytosis; selective distribution of anionic sites. In: Weissmann G et a1 (eds) Advances in inflammation research. Raven Press, New York vol 1: 61-70

Song SH, Groom AC 1971a The distribution of red cells in the spleen. Can J Physiol Pharmacol 49: 734-743

Song SH, Groom AC 1971b Storage of blood cells in spleen of the cat. Am J Physiol220: 779-784 Song SH, Groom AC 1971c Sequestration and possible maturation of reticulocytes in the normal

Spry CJF 1971 Mechanisms of eosinophilia. VI. Eosinophil mobilization. Cell Tissue Kinet 4:

Steinman RM, Lustig DS, Cohn ZA 1975 Identification of a novel cell type in peripheral lymphoid organs of mice. IV. Identification and distribution in mouse spleen. J Exp Med 141: 804-820

38: 1-82

spleen. Can J Physiol Pharmacol 50: 400-406

365-373


Tavassoli M, Crosby WH 1968 Transplantation of marrow to extramedullary sites. Science (Wash

Trentin JJ 1970 Influence of hematopoietic organ stroma (hematopoietic inductive microenvironments) on stem cell differentiation. In: Gordon AS (ed) Regulation of hematopoiesis. Appleton-Century-Crofts, New York, vol 1: 161-186

van Ewijk W, Verzijden JHM, van der Kwast ThH, Luijck-Meijer SWM 1974 Reconstitution of the thymus dependent area in the spleen of lethally irradiated mice. A light and electron microscopical study of the T cell microenvironment. Cell Tissue Res 149: 43-60

Weiss L 1965 The structure of bone marrow. Functional interrelationships of vascular and hematopoietic compartments in experimental hemolytic anemia: an electron microscopic

Weiss L 1970 Transmural cellular passage in vascular sinuses of rat bone marrow. Blood 36:

Weiss L 1976 The hematopoietic microenvironment of the bone marrow: an ultrastructural study of the stroma in rats. Anat Rec 186: 161-184

Weiss L 1980 The haemopoietic microenvironment of bone marrow: an ultrastructural study of the interactions of blood cells, stroma and blood vessels. In: Blood cells and vessel walls: functional interactions. Excerpta Medica, Amsterdam (Ciba Found Symp 71) p 3-19

Weiss L 1981 Life cycles of the blood cells. In: Weiss L (ed) Histology, 5th edn. McGraw-Hill, New York, in press

DC) 161: 54-56

study. J Morphol 117: 467-537

189-208

DISCUSSION

Greaves: Dr Weiss, how good are we at identifying precursor cells in sections with biochemical or molecular probes?

Weiss: We are at a point where, as morphologists, we shall further exploit cytochemistry, particularly immunofluorescence and immunoperoxidase methods. With reagents now available one can get a good deal of information. D. Brookoff and I, for example, have followed the development of transferrin receptors on erythroblasts using an immunoperoxidase method for transferrin (unpublished). There are now, in addition, many cell surface markers known for lymphocytes (and T lymphocytes may have a role in regulating erythropoiesis and granulocytopoiesis). But many new cytochemic- a1 reagents are being developed from monoclonal antibodies, as the papers in this symposium will show. It is going to be possible to characterize precursor and regulating cells, not only in cell suspensions but in a tissue, where their cellular associations are preserved. It will soon be possible to make highly specific characterizations of subsets of these cells.

Dexter: If we ascribe an inductive role to the stroma in the marrow, are we also able to do that for the fetal liver, which is a major site of haemopoiesis in the mammalian embryo? Have you looked for, and found, reticular cells and macrophages in fetal liver?

Weiss: Yes. In fetal liver and in other haemopoietic organs such as the spleen, in regenerating bone marrow after irradiation, and also in fetal bone

16 DISCUSSION

marrow, one finds macrophages, lymphocytes and haemopoietic cells, and reticular cells. Reticular cells are associated with reticular fibres, and therefore are presumably fibroblastic. As fibroblast-like cells, they have features in common with a large company of inductive cells, shown to be involved in induction in various tissues, such as the submandibular gland. In every place that I have looked, there is this line of reticular cells associated with haemopoiesis.

Hogan: Do you find them at an even earlier stage, in the yolk sac, which is also a haemopoietic tissue?

Weiss: I have not studied this myself, but a number of these cell types are there. In the yolk sac, haemopoiesis is closely associated with the vasculature. Fibroblastic cells and macrophages, as well as other cell types, are always present in haemopoietic tissues. I don’t think the yolk sac will be an exception.

Miller: What about the haemopoietic spleen colony, where one finds both erythroid and granulopoietic development?

Weiss: I have not looked closely at spleen colonies. R. Lambertsen and I (unpublished) have studied bone marrow colonies after recovery from lethal irradiation in a mouse saved by a protected femur. We see haemopoietic colonies as soon as two days after irradiation. They tend in the early stages to be in only one line of haemopoiesis. With the exception of megakaryocyte colonies, the colonies contain macrophages and are closely associated with fibroblastic cells. There are lymphocytes as well. The lymphocytes in these colonies may be relatively mature cells, as T cells, and also haemopoietic stem cells of lymphocyte-like appearance, but one cannot distinguish these except with selective stains. So in bone marrow colonies, and probably in spleen colonies too, we find a number of different cell types associated with haemopoiesis even in the early stages.

After bone marrow is depleted of haemopoietic cells, whether by treatment with saponin in rabbits or by irradiation in mice, at the time haemopoiesis is just reappearing (in the first 1-2 days) we see reticulum cells with ramifying extended processes. Close up against these processes are cells with the appearance of stem cells. My impression is that haemopoiesis depends on the association of reticular cells with stem cells.

Osmond: Do you see regional differences in the reticular cells between various parts of the marrow? Patt & Maloney (1976) showed that the peripheral reticular cells, associated with the bone, are particularly efficient in reconstituting the marrow stroma after mechanical depletion of the marrow cavity. In our department, Dr E. Daniels (1980) finds that bone-associated cells are very effective in reconstituting a stromal matrix in vitro. Is this apparent stromal cell differentiation reflected morphologically?

Weiss: There are marked regional differences among haemopoietic col-

Microenvironments in haemopoietic and lymphoid differentiation

Documents

Transcript of Microenvironments in haemopoietic and lymphoid differentiation