Cytotechnology 2: 259-267, 1989. 9 1989 KluwerAcademic Publishers.Printed in the Netherlands.

Haemopoietic growth factors: their role in cell development and their clinical use

Nydia G. Testa and T. Michael Dexter Department of Experimental Haematology, Paterson Institute for Cancer Research, Christie Hospital and Holt Radium Institute, Manchester M20 9BX, UK

Key words: cell function, differentiation, growth factors, haemopoiesis, haemopoietic regeneration, leukaemia Abstract

Mature blood cells are derived from haemopoietic stem cells which grow and proliferate to give rise to progenitor cells more restricted in their proliferation and differentiation capacity. These in turn give rise to cells belonging to any of the haemopoietic lineages. The haemopoietic growth factors interleukin 3, granulocyte-macrophage colony-stimulating factor, granulocyte colony stimulating factor, macrophage colony-stimulating factor and erythropoietin act on haemopoietic cells to promote cell survival, proliferation, differentiation and maturation, as well as many functions of the mature cells. These factors, now purified to homogeneity and molecularly cloned have recently become available. This has facilitated studies of their roles in cell production, and the range of target cells sensitive to them in vitro and in vivo in several species. The latter experimental data led to the first clinical trials where these factors have been used successfully in several clinical settings: erythropoietin to correct the anaemia of renal disease; granulocyte and granulocyte-macrophage colony-stimulating factors to accelerate haemopoietic regeneration after chemotherapy and bone marrow transplantation, and in other situations where increase in the numbers of white cells and stimulation of their function were required. The results to date allow optimism; the clinical use of growth factors not only in haematology and oncology, but in wider fields of medicine may well constitute a major breakthrough in the near future.

Introduction

All haemopoietic cells are derived from stem cells, located in the bone marrow in the adult. They are able to self-renew (to maintain their own numbers) and to give rise to more differentiated progenitor cells. These may in turn undergo further proliferation and commitment restricting them irreversibly to one of any of the blood cell lineages. All these processes will eventually, through further cell proliferation and maturation,

produce the mature functional cells circulating in the blood stream. Cell production in the haemopoietic tissue reaches numbers of the order of 4 • 101 t cells per day in a normal human adult, first to replace blood cells normally lost through senescence, and even higher numbers when there is increased demand, for example after blood loss or during infection. We now know that cell production can be considerably influenced in vivo by a family of glycosylated polypeptides collectively named

260 haemopoietic growth factors. They were first described in the sixties as colony stimulating factors (CSFs) because of their ability to stimulate clonal proliferation (which resulted in the formation of colonies) from progenitor cells in vitro. When haemopoietic colonies were first grown, the growth factors used were produced in situ by feeder cells, or were added to the cultures unpurified, as a component of media conditioned by any of several cell types. Recently, however, these factors have been purified to homogeneity, the genes cloned, and their products expressed. Recombinant molecules are now available, and are at present being evaluated in clinical trials. Here, we will review briefly the structure of the haemopoietic tissue and the role of the haemopoietic growth factors, and will discuss their possible uses in the treatment of disease.

the potential to differentiate along the myeloid (but not the lymphoid) lineages detects colony forming cells (Mix-CFC) closely related to stem cells (Metcalf, 1988). In cultures in a semisolid matrix, containing the appropriate growth factors, these cells form clonal colonies that contain mature cells of several myeloid lineages, plus undifferentiated cells which can produce more mixed colonies in vitro, or in vivo in the spleen of irradiated mice (Humphries et al., 1981 ). More mature progenitor cells, which eventually become restricted in their differentiation potential, may also be detected in vitro by their capacity to give rise to colonies that contain cells belonging just to one or two lineages (Fig. 1), if stimulated with the relevant growth factors.

The haemopoietic growth factors The structure of the haemopoietic tissue The stem cells are defined by their capacity to reconstitute the haemopoietic system upon transplantation into irradiated hosts. Experiments using either unique chromosome markers or inserted retroviruses have shown that a single cell may be responsible for the reconstitution of the haemopoietic tissue in mice (Abramson et al., 1979; Snodgrass and Keller, 1987). This implies that stem cells can undergo extensive cell renewal to maintain their own numbers and it is known that they do not decline with age (Schofield et al., 1986). They can also undergo differentiation along the several distinct haemopoietic lineages (Fig. 1). Little is known about the regulatory procedures which determine growth and differentiation at the stem cell level, although interactions with ceils in the haemopoietic stroma are thought to be crucial (Dexter, 1982). There is a continuum in the stem cell population, with the capacity to self-renew being progressively lost, perhaps together with the acquisition of characteristics which will lead to commitment to any of the haemopoietic lineages (Fig. 1). An assay for a cell population which still conserves some self-reproduction capacity, and

At least in vitro, the haemopoietic growth factors are essential for the survival, proliferation and differentiation of haemopoietic cells, and also for the induction of many functional characteristics of the mature cells. They are all glycosylated polypeptides, although the sugar groups are not essential for their biological activity: recombinant growth factors produced by bacteria are effective in vivo (Metcalf, 1988). Four major CSFs are at present available in recombinant form: interleukin 3 (IL-3), also called multi-CSF, granulocyte-macrophage CSF (GM-CSF), granulocyte CSF (G-CSF) and macrophage CSF (M-CSF) also called CSF-1 (Fung et al., 1984; Gough et al., 1984; Tsuchiya et al., 1986). The cloning of these genes for murine factors was shortly followed by the cloning of the human genes (reviewed by Metcalf, 1988). In addition, the human gene for erythropoietin (which is also defined as a growth factor by its role in stimulating the proliferation and haemoglobinisation of erythroid cells) was also cloned recently (Lin et al., 1985). Although the CSFs have considerable overlap in their actions on haemopoietic cells (Table 1), there is also a hierarchy in their targets which resembles the hierarchy in haemopoietic cells

261

03 0

[.-,

?

1"

?

o

1" [..,, ~o

r~

1,

U

w,

r~

ak

262 Table 1. T h e s e n s i t i v i t y o f c o l o n y f o r m i n g cells to h a e m o p o i e t i c g r o w t h factors Factor

Mix-

GM-

G-

M-

Eo-

Meg-

CFC

CFC

CFC

CFC

CFC

CFC

BFU-E

CFU-E

IL-3

+

+

+

+

+

+

+

_+

GM-CSF

+

+

+

+

+

+

+

-

-+

+

G-CSF

-

+

+

.

M-CSF

-

+

-

+

Epo

.

.

.

.

.

.

. .

. .

. .

+

.

N o m e n c l a t u r e a n d r e f e r e n c e s in t h e text.

shown in Fig. 1. Thus, IL-3 stimulates the growth of Mix-CFC (Metcalf, 1988) as well as increasing the survival of CFU-S in vivo (Spivak et al., 1985). From this, it appears that IL-3 acts on primitive pluripotential cells in the haemopoietic hierarchy as well as on the more committed progenitors. Mix-CFC stimulated by IL-3 may produce neutrophils, macrophages, eosinophils, basophils, megakaryocytes, and (if erythropoietin is present), erythrocytes. GM-CSF stimulates preferentially progenitors of granulocytes and macrophages (GM-CFC) (but may also stimulate Eo-CFC and a proportion of Mix-CFC when used at high concentration), as well as erythroid and megakaryocyte progenitors, BFU-E and MegCFC (Metcalf et al., 1980) From this it can be seen that the target cells of IL-3 and GM-CSF show considerable overlap. The latter, however, does not appear to promote the formation of basophils. G-CSF stimulates the development of neutrophils from progenitor cells (Metcalf and Nicola, 1983), although in the presence of other factors, may act synergistically and allow the growth of cells of other lineages (Heyworth et al., 1988). M-CSF stimulates GM-CFC and M-CFC to divide and proliferate along the macrophage lineage. Also, as described above for G-CSF, M-CSF can synergise with other factors. A variety of other bioregulators (interleukins 1, 4, 5 and 6) also influence growth and differentiation in the haemopoietic system. Although only IL-5 can induce colony formation (from progenitors of eosinophils, Eo-CFC, Clutterbuck and Sanderson, 1988), they all may synergise with the CSFs to stimulate and modulate cell production.

Therefore, multipotential cells and progenitor cells with more restricted potential for differentiation can respond to a variety of growth stimulators (and also to growth inhibitors, Keller et al., 1988) indicating that these cells express simultaneously the receptors to several factors. This implies that the fine regulation that results in adequate production of the cell types needed in very diverse pathological circumstances may be the result of the balance of stimulators and inhibitors of cell growth and differentiation to which cell populations are exposed in the bone marrow. In this context, it is important to notice that the bone marrow is a highly structured tissue, with discrete environmental domains in which distinct populations of stem and progenitor cells are found in different proportions in different areas (Testa et al., 1985: Lord and Testa, 1988). These compartmentalised areas may promote the preferential development of specific haemopoietic lineages, a role which is likely to be mediated largely through growth factors. The CSFs, however, are not just responsible for cell proliferation and lineage selection; they also enhance the survival of progenitors and of mature cells, and induce cellular maturation and stimulation of several functions of the mature cells (Begley et al., 1986; Stanley et al., 1983; Kitagawa et al., 1987; Kurland et al., 1978). This implies that CSFs are likely to be crucial in the response to local infection and in the reaction to allergens. Their local production by fibroblastic or endothelial cells, or in the case of M-CSF by macrophages themselves, (cells which also produce other bioregulators like interferons and tu-

263 mour necrosis factor) are likely to be of critical importance. In this context, the recent report (Wodner-Fillipowicz et al., 1989) that activated mast cells may produce IL-3 and GM-CSF, imply the capacity to prime and activate locally neutrophils, macrophages and eosinophils, cells which have a critical role in local tissue defence and in the reaction to allergens. Although there is little homology between growth factors, which suggest that they may have evolved independently, it is of interest that many are clustered in close proximity in the q23-q33 region of chromosome 5. Not only the genes for IL-3, GM-CSF and M-CSF are found there, but also c-fms, the gene encoding for the receptor to M-CSF (reviewed by Testa and Dexter, 1989). Also, the genes coding for the epidermal growth factor and for the receptor of the platelet derived growth factor are in close proximity. It is not difficult to envisage that alterations in such a cluster of growth regulatory genes will be of relevance in the development of malignancy. In fact, a relatively large frequency of deletions in the 5q region occurs in patients with primary myelodysplasia (the 5q-syndrome) or in secondary acute myeloblastic leukaemia. Break points in that area, with the loss of critical sequences may be an important step in malignant transformation leading to leukaemia, either by causing a reduced level of a gene product being produced, or by allowing the expression of a recessive allele in the homologous chromosome. Mutations in cfms, which code for the M-CSF receptor, may also give growth stimulatory signals in the absence of a ligand (Wheeler et al., 1988). The gene coding for G-CSF is located in the q21-q22 region of chromosome 17. It is of interest that a translocation (15;17) may be observed in acute myelocytic leukaemia, with the break point close to the location of this gene. The protooncogene c-erb2 (related to the oncogene v-erbB, which has close homology with the receptor for epidermal growth factor, and which induces erythroleukemia in chickens) is also located in chromosome 17, as is the gene coding for the nerve growth factor. Although alterations in those chromosomes are usually found at diagno-

sis, and may tell us little about the initial processes in the development of malignancy, experimental systems suggest that alterations in the response or in the production of growth factors may be amongst the early steps in leukemogenesis (Testa et al., 1988).

Clinical use of growth factors The availability of recombinant growth factors have made possible studies of their effect in vivo, and data are now available on their effects in experimental animals. Mice treated with IL-3 show an increase in the number of multipotential and lineage-restricted progenitor cells in the spleen, but little if any overall increase in the bone marrow, and only a minor increase in the number of circulating cells (Lord et al., 1986). The latter effect was also observed in primates (Donahue et al., 1988). However, in primates the administration of GM-CSF caused a marked rise in the numbers of neutrophils, eosinophils, monocytes and lymphocytes in peripheral blood (Donahue et al., 1987). Also, in primates, it was shown that these two growth factors act synergistically (Donahue et al., 1988). The administration of G-CSF increased markedly the numbers of circulating neutrophils in cynamolgous monkeys and alleviated the neutropaenia seen after treatment with cyclophosphamide in these monkeys, as well as in mice (Tamura et al., 1987; Welte et al., 1987). Two general principles have been observed: there is synergism of action when combination of factors are administered (Broxmeyer et al., 1987), and the effects cease when treatment is discontinued. These encouraging data led to the first clinical trials of growth factors, which were mainly concemed with clinical situations where quick regeneration of the haemopoietic tissue was required, for example after chemotherapy or following bone marrow transplantation. A list of the possible clinical uses of growth factors is shown in Table 2. Both G-CSF and GM-CSF have been used in phase I/II trials in patients with malignancy, leading to shortening of the period of critical

264 Table 2. Cfinical uses of haemopoietic growth factors

Acute regeneration of haemopoiesis Post-chemotherapy Post-bone marrow transplantation (autologous, allogeneic) Agranulocytosis Idiopathic Congenital Iatrogenic Granulocytopaenia e.g. AIDS Infections (treatment or prevention) Bacterial Fungal After burn s or major trauma Major surgery Resistance to antibiotics Myelodysplasia Aplastic Anaemia Acute Myeloblastic Leukaemia Anaemia of renal disease References in the text.

Growth factors after chemotherapy 10.000

E

1.ooo.

m

lOO o

4

8 Days

12

16

1. Shorter (or no) critical period 2. Lower risk of infection 3. Allows intensification of treatment [higher doses | [_shorter time

Fig. 2. Schematic representation of the results of the administxation of G-CSF to a patient with small cell lung carcinoma. The white blood cell counts after 2 courses of chemotherapy (one followed by administration of G-CSF) are shown. Curves drawn from Bronehud et al., 1987.

neutropaenia, a faster rate of recovery to normal numbers of circulating cells and a decrease in the number of serious infections following chemotherapy. The use of G-CSF resulted in a specific response in the neutrophilic lineage, but GM-CSF also caused increases in eosinophils and monocytes (Bronchud et al., 1987; Morstyn et al., 1988; Gabrilove et al., 1988). The beneficial effects are illustrated in Fig. 2. GM-CSF also improved regeneration after autologous bone marrow transplantation following high dose chemotherapy (Brandt et al., 1988; Nemunaitis et al., 1988). Encouraging results have also been reported in the alleviation of the granulocytopaenia of the AIDS syndrome following treatment with GMCSF: dose-dependent increases in neutrophils, eosinophils, monocytes and lymphocytes were observed, as well as rises in reticulocytes and platelets. The roles of growth factors in stimulating not only cell production, but also function of the

265 mature ceils (described above) have therapeutic implications not only for the treatment of granulocytopaenias (Table 2), but also for the treatment of syndromes, like myelodysplasia, where poor cell production may also be accompanied by defective function of the mature cells (VadhanRaj et al., 1987). However, an increase in blast ceils was observed in some patients, requiring cessation of the administration of GM-CSF (reviewed by Steward et al., 1989). The treatment of acute myeloblastic leukaemia with growth factors is the subject of current debate (Testa and Dexter, 1989). Leukaemic cells may have receptors for growth factors and may respond to them with proliferation (Metcalf, 1986). However, differentiation (which may lead to extinction of the leukaemic clone) has also been observed. A therapeutic role for the growth factors may also be envisaged if leukaemic cells induced to proliferate in synchrony can then be killed by cell cycleactive agents. An alternative approach could be to use antibodies to a specific growth factor if receptors for that growth factor only are present in the leukaemic cells. It may be speculated that in that case, normal haemopoiesis would still be possible due to the overlap in the window of action of the different growth factors (Table 1 ). Clearly, however, these possible treatments require caution and thought. There is a clear indication for the use of growth factors in palliative treatment of granulocytopaenias and aplastic anaemia during episodes of acute infection. The cure of these syndromes however, appears more problematic: little is known about long-term effects of the administration of growth factors that may presumably be required. Furthermore, as discussed above, our knowledge about the regulation of stem cells, whose normal function (and increase in numbers) would be required for the cure of aplastic anaemia, is far from extensive. The influx of mature cells to the site of infection, together with the local production of growth factors (Wodnar-Fillipowicz et al., 1989) and other bioregulators in situ suggest a role for the local application of growth factors, for example after trauma, extensive burns or local infection.

Although little data are at present available on the combined use of growth factors, it is logical to expect that combination of factors may be designed to fulfil specific therapeutic needs, choices which will be dictated by the windows of action of each factor, and by their specific effects on cell proliferation and differentiation. Toxic effects after administration of GM-CSF and G-CSF have been observed. However, the most serious side-effects (intense bone pain, capillary leak syndrome and pericarditis) after GM-CSF, seem to have been related to 'bolus' administration, as they were not observed after continuous intravenous infusion (Steward et al., 1989). G-CSF also appears to be well tolerated after intravenous infusion or subcutaneous administration. Another factor to be considered and advises caution, is that G- and GM-CSF may influence the growth of non-haemopoietic cells: endothelial cells proliferate in response to them (Bussolino et al., 1989). Also, cell lines derived from lung carcinoma, breast carcinoma and osteosarcoma show receptors for GM-CSF (Dedhar et al., 1989). Data on primary tumour cells, however, are not yet available. On the other hand, tumour regression has been observed during Phase I trials with GM-CSF (W. Steward, personal communication) perhaps through activation of local infiltrating macrophages, which may then exert a variety of antitumour actions, including, for example, production of tumour necrosis factor. The only clear indication for replacement therapy is in the treatment of the anaemia of renal disease, where the diseased kidneys produce subnormal amounts of erythropoietin. The therapeutic use of erythropoietin in this situation was proposed many years before the recombinant molecule became available. At present, about 400 patients have been treated with outstanding results. Dose-dependent rises in haemoglobin and haematocrit levels lead to transfusion-independency in practically all the patients treated, with clearly improved quality of life and increased sense of well-being (Winnearls et al., 1986: Esback et al., 1987). The possible use of erythropoietin to alleviate thrombocytopaenia induced

266

by chemotherapy or observed after bone marrow transplantation is supported by the observations that at high doses, it stimulates megakaryopoiesis (McDonald et al., 1987). In conclusion, the large amount of information gathered on the last two years allows optimism about many different clinical uses of the haemopoietic growth factors. Although many questions remain to be answered, it is already apparent that these factors will have a great impact not only in oncology and haematology, but also in wider medical fields.

Acknowledgements The authors are supported by the Cancer Research Campaign, UK.

References 1. Abramson S, Miller RG, Phillips RA (1979) The identification in the adult bone marrow of pluripotent and restricted stem cells of the myeloid and lymphoid systems. J. Exp. Med. 145: 1587-1579. 2. Begley CG, Lopez AF, Nicola NA, Warren DJ, Vadas MA. Sanderson CJ, Metcalf D (1986) Purified colony stimulating factors enhance the survival of human neutrophils and eosinophlls in vitro: a rapid and sensitive microassay for colony stimulating factors. Blood 68: 102-108. 3. Bran& SJ, Peters WP, Atwalter SK, Kmtzberg J, Borowitz MJ, Jones RB, Shapall EJ, Bast RE, Colleen JG, Gette DH (1988) Effect of recombinant human granulocyte-macrophage colony stimulating factor on haemopoietic reconstitution after high dose chemotherapy and autologous bone marrow transplantation. New Eng. J. Med. 318: 869-876. 4. Bronehud MH, Scarffe JH, Thatcher N, Crowther D, Souza LM, Alton NR, Testa NG, Dexter TM (1987) Phase lflI study of recombinant human granulocyte colony stimulating factor in patients receiving intense chemotherapy for small cell lung carcinoma. Br. J. Cancer 56: 809-813. 5. Broxmeyer HE, Williams DE, Hangoc G, Cooper S, Crillis RK, Shadduck RK, Bicknele DC (1987) Synergistic mielopoietic actions in vivo after administration to mice of combinations of purified natural murine interleukin-3 and recombinant murine granulocyte-macrophage colony stimulating factor. Proc. Nail. Acad. Sci. USA 84: 3871-3875. 6. Bussolino F, Wang JM, Defilipi P, Turini F, Sanavio F, Edgell CJS, Aglieta M, Arese P, Mantovani A (1989) Crranulocyte and granulocyte-macrophage colony stimulating factors induce human endothelial cells to migrate and proliferate. Nature 337: 471--473.

7. Dedhar S, Gaboury L, Galloway P, Eaves C (1988) Human granuloeyte colony-stimulating factor is a growth factor active on a variety of cell types of non hemopoietic origin. Proc. Natl. Acad. Sci. USA 85: 9253-9257. 8. Clutterbuek EJ, Sanderson CJ (1988) Human eosinophil production in vitro started by means of routine eosinophil differentiation factor (IL-5): Production of functionally active eosinophils from normal human bone marrow. Blood 71; 646-651. 9. Dexter TM (1982) Stromal cell associated haemopoiesis. J. Cell Physiol. Suppl. 1: 87-92. 10. Donahue RE, Wang EA, Stone DK, Kamen R, Wong GG, Sehgal PK, Nathan DG, Clark SC (1987) Stimulation of haemopoiesis in primates by continuous infusion of recombinant human GM-CSF. Nature 321: 872-875. 11. Donahue RE, Sheera J, Metzger M, Lefebvre D, Rock B, Carbone S, Nathan DG, Garnick M, Sehgal PK, Laston D, Lavallie E, McCoy J, Schendel PF, Norton C, Turner K, Young YC, Clark SC (1988) Human IL-3 and GM-CSF act synergistically in stimulating haematopoiesis in primates. Science 241: 1820-1823. 12. Esbach JW, Egrie JC, Downing MR, Browne K, Adamson JW (1987) Correction of the anemia of end-stage renal disease with recombinant human erythrepoietin. Results of a Phase I and II clinical trial. New Eng. J. IVied. 316: 73-78. 13. Fang MC, Mapel AS, Ymer S, Cohen DR, Johnson RM, Campbell HD, Young IG (1984) Molecular cloning of cDNA for routine interleukin-3. Nature 307: 233-237. 14. Gabtilove JL, Jakubowski A, Sher H, Sternberg C, Wong G, Grous J, Yagoda A, Fain K, Moore MAS, Clarkson B, Oettgen I-IF,Alton K, Welte K, Souza LM (1988) Effect of granulocyte colony-stimulating factor on neutrepaenia and associated morbidity due to transitional cell carcinoma of the urothelium. New Eng. J. Med. 318: 1414-1422. 15. Gough NM, Gough J, Metcalf D, Kelso A, Grail D, Nicola NA, Burgess AW, Dunn All (1984) Molecular cloning of cDNA encoding a murine haematopoietic growth regulator, granulocyte-macrophage colony-stimulating factor. Nature 309: 763-767. 16. Heyworth CM, Ponting ILO, Dexter TM (1988) The response of haemopoietic cells to growth factors: developmental implications of synergistic interactions. J. Cell Sci. 91 : 239-247. 17. Humphries RK, Evans AC and Eaves CI (1981) Selfrenewal of hemopoietic stem cells during mixed colony formation in vitro. Proc. Natl. Aead. Sci. USA 78: 36293633. 18. Keller JR, Mantel C, Sing GK, Ellingworth LR, Ruscetti SK, Ruscetti FW (1988) Transforming growth factor selectively regulates early murine hematopoietic progenitors and inhibits the growth of IL-3 dependent myeloid leukaemia. J. Exp. Med. 168: 737-750. 19. Kitagawa S, Yuo A, Souza LM, Saito M, Miura Y, Takaku F (1987) Recombinant human granuloeyte colony stimulating factor enhances superoxide release in human granulocytes stimulated by a chemotactic peptide. Biochem. Biophys. Res. Commun. 144: 1143-1146.

267 20. Kurland SI, Bockman RS, Broxmeyer JE, Moore MAS (1978) Limitation of excessive myelopoiesis by the intrinsic modulation of maerophage derived prostaglandin E. Science 199: 552-555. 21. Lim FK, Suggs S, Lin CM, Browne JK, Smalling R, Egrie JC, Chen KK, Fox GM, Martin F, Stabinski Z, Badhauri SM, Lai PH, Goldwasser E (1985) Cloning and expression of the human erythropoietin gene. Proc. Natl. Acad. Sci. USA 82: 7580-7584. 22. Lord BI, Testa NG (1988) The hemopoietic system: structure and regulation. In: Testa NG, Gale RP (eds.) Hematopoiesis: Longterm effects of chemotherapy and radiation. M. Dekker, New York, pp. 1-26. 23. Lord BI, Molineux G, Testa NG, Kelly M, Spooncer E, Dexter TM (1986) The kinetic response of hemopoietic precursor cells in vivo, to highly purified recombinant Interleuldn-3. Lympholdne Res. 5: 97-104. 24. Metcalf D (1986) The molecular biology and functions of the granulocyte-macrophage colony stimulating factors. Blood 67: 257-267. 25. Metcalf D (1988) The molecular control of blood cells. Harvard University press, Cambridge, Mass. 26. 26. Metcalf D, Johnson GR, Burgess AW (1980) Direct stimulation by purified GM-CSF of the proliferation of multipotential and erythroid precursor cells. Blood 55: 138-147. 27. Metcalf D, Nicola NA (1983) Proliferative effects of purified granulocyte colony-stimulating factor (G-CSF) on nor~mal mouse hemopoietic cells. J. Cell Physiol. 116: 198206. 28. McDonald TP, Cottrell MB, Clift RE, Cullen WC, Lin FK (1987) High doses of recombinant erythropoietin stimulate platelet production in mice. Exp. Hematol. 15: 719--721. 29. Schofield R, Dexter TM, Lord BI, Testa NG (1986) Comparison of haemopoiesis in young and old mice. Mech. Aging Dev. 34: 1-12. 30. Snodgrass R, Keller G (1987) Clonal fluctuation within the haemopoietic system of mice reconstituted with retrovirusinfected stem cells. EMBO J. G: 3955-3880. 31. Spivak JL, Smith RR, Ihle JN (1985) Interleukin 3 promotes the in vitro proliferation of murine pluripotent stem cells. J. Clin. Invest. 76: 1613-1621. 32. Stanley ER, Guilbert LJ, Tushinsky RJ, Bartlemez SH (1983) CSF-I: a mononuclear phagocyte lineage-specific hemopoietic growth factor. J. Cell Biochem. 21: 151-159. 33. Steward WP, Scarffe JH, Dexter TM, Testa NG (1989) Clinical applications of haemopoietic growth factors. In: Dexter TM, Garland J, Testa NG (eds.) Cellular and molecular biology of the colony stimulating factors. M. Dekker, New York. (In press).

34. Tamura M, Hattori K, Nomura H, Oheda M. Kubota N, Imazebi I, Ono M, Oeyama Y, Nagata S, Shinafuji N, Asano S (1987) Induction of neutrophilic granulocytosis in mice by administration of purified human native granulocyte colony-stimulating factor (G-CSF). Biochem. Biophys. Res. Commun. 142: 454--460. 35. Testa NG, Hendry J, Molineux G (1985) Long-term bone marrow damage in experimental systems and in patients after radiation or chemotherapy. Anticancer Res. 5: 101. 36. Testa NG, Onions D, Lord BI (1988) Pre-leukemic abnermalities caused in cats by infection with a new isolate of Feline Leukemia Virus (FeLV AB/GM). Hematologica 73: 317-320. 37. Testa NG, Dexter TM (1989) Haemopoietic growth factors: their role in acute myeloblastic leukaemia. Brit. Med. Soc. Bull. 45: 582-599. 38. Tsuchiya M, Asano S, Kaziro Y, Nagata S (1986) Isolation and characterisation of the cDNA for murine granulocyte colony stimulating factor. Proc. Natl. Acad. Sci. USA 83: 7633-7637. 39. Vadan-Raj S, Buescher S, Le Maistre A, I-littelman WN, McCmdie K, Trujillo JM, Broxmeyer HE, Henney C, Gutterman JU (1987). Effects of human granuloeytemacrophage colony-stimulating factor in patients with myelodysplastic syndromes. New Eng. J. Med. 317: 15451552. 40. Welte K, Bonilla MA, Gillio AP, Boone TC, Potter GK, Gabrilove JL, Moore MAS, O'Reilly R J, Souza LM (1987) Recombinant human granulocyte colony-stimulating factor: effects on hematopoiesis in normal and cyclophosphamide treated primates. J. Exp. Med. 165: 941-948. 41. Wheeler EF, Rettenmier CW, Look AT, Sherr CJ (1986) Nature 324: 377-380. 42. Winnearls CF, Pippard MH, Downing M, 0liver DO, Reid C, Cotes PM (1986) Effect of human erythrepoietin derived from recombinant DNA on the anaemia of patients maintained by chronic hemodialysis. Lancet ii: 1175-1178. 43. Wodnar-Fillipowicz A, Heusser CM, Moroni C (1989) production of the haemopoietic growth factors GM-CSF and interleukin-3 by mast cells in response to IgE receptormediated activation. Nature 339: 150-152.

Address for offprints: N.G. Testa, Department of Experimental Haematology, Paterson Institute for Cancer Research, Christie Hospital and Holt Radium Institute, Manchester M20 9BX, UK