sign in sign up
<<
Home , Erythropoietin, Erythropoietin receptor, Growth factor receptor, KIT (gene), Thrombopoietin, Thrombopoietin receptor

(1997) 11, 1821–1826  1997 Stockton Press All rights reserved 0887-6924/97 $12.00

Study of the in essential JJ Kiladjian1,2, N Elkassar1, G Hetet1,3, J Briere1,2, B Grandchamp1 and C Gardin1,2

1Unite´ INSERM U 409, Faculte´ de Me´decine X Bichat, Paris; 2Service d’He´matologie Clinique, Hoˆpital Beaujon, Clichy; and 3Association Claude Bernard, Paris, France

Essential thrombocythemia (ET) is a myeloproliferative dis- yocytes in vitro growth in such culture assays.13 Conflicting order associated with megakaryocytic hyperplasia and throm- results have been reported using ET-derived CD34+ BM cells, bocytosis. In this disease, in vitro autonomous growth of as persistence14 and absence of autonomous megakaryocytic megakaryocytic colonies has been demonstrated by various 12,15 investigators. This phenomenon is impaired by the inhibition growth have been reported. of the thrombopoietin/c-mpl pathway. In order to evaluate the Most studies were done before the isolation of thrombopo- potential role of mutations of the receptor in the origin of ietin (Tpo), the most lineage-specific factor associated with this autonomous growth, we compared the expression of c-mpl proliferation and maturation.16 The role of mRNA isoforms in derived from ET patients and nor- Tpo and its receptor c-mpl in this megakaryocytic proliferation mal subjects. Overlapping c-mpl PCR fragments derived from 17 four ET patients were sequenced to search for small mutations. is currently under investigation: Methia et al have shown In the 10 ET and five normal samples we studied, relative that normal megakaryocytopoiesis was severely impaired expression of the c-mpl isoforms was identical. New variants when c-mpl expression was inhibited by anti-sense oligonu- of Mpl-P and K isoforms, Mpl-P2 and K2 were detected. Cloning cleotides (ASO) derived from the c-mpl sequence. Using the of these isoforms indicated that they are produced by alterna- same ASO, Li et al18 demonstrated that the formation of tive splicing of exon 9 sequences shared by Mpl-P and K. Their megakaryocytic colonies derived from ET samples was also predicted sequence would be deleted by 24 ami- noacids, upstream of the WSSWS box of the second domain dependent on the Tpo-c-mpl pathway. In this study, no syn- of c-mpl. Two sequence variations, leading to DNA restriction thesis of Tpo was detectable by RT-PCR in serum-free liquid polymorphisms, were present in the extracellular and Mpl-K cultures of ET cells, suggesting that this prolifer- intracytoplasmic domains. Both were present in normal and ET ation was not due to autocrine or paracrine growth. Prelimi- samples, excluding mutations of c-mpl as a cause of ET. nary studies have not found decreased blood levels of Tpo in Keywords: essential thrombocythemia; thrombocytosis; mega- ET patients as opposed to the low blood level karyocytes; c-mpl commonly found in PV patients.19 Although the mechanisms of circulating Tpo regulation are still unclear, platelets play a major role in the of the circulating Tpo, as they Introduction express functional receptors able to bind and internalize Tpo.20 Both the in vitro growth data and the normal or elev- Essential thrombocythemia (ET) is a myeloproliferative dis- ated Tpo blood levels in ET patients, as measured by an ELISA order (MPD) characterized by a sustained thrombocytosis. assay,19 suggest that c-mpl receptor expression or function Clinical diagnosis requires the elimination of other myelopro- could be altered in the population of clonal hematopoietic liferative disorders vera (PV), chronic myelogen- cells usually present in ET. ous leukemia and agnogenic myeloid metaplasia, and a care- C-mpl receptor studies in human and ful search for conditions known to be associated with reactive mouse hematopoietic cells or leukemic cell lines have thrombocytosis.1,2 The chronic course of ET is associated with 3 revealed the existence of multiple mRNA isoforms (Figure 1) an increased incidence of thrombosis and bleeding. Although whose role in megakaryocytic physiology is unknown: Mpl-S the disease can evolve into acute leukemia, the frequency of (mpl-tr in mouse) a putative soluble receptor form, and Mpl- leukemia and the role played by cytotoxic drug therapy in the 4 K, a form with a truncated intracytoplasmic domain, only transition to leukemia are still debated. found in human cells, are coexpressed with Mpl-P (mpl 1 or As in other MPDs, clonal proliferation of various type I in mouse), the full-length receptor.21–24 Two other vari- hematopoietic lineages including granulocytes, megakaryo- ant forms with in-frame deletions of the mRNA 5Ј end have cytes, erythroblasts and monocytes has been demonstrated in been described in mouse cell lines with megakaryocytic fea- 5–7 most cases of ET. The mechanisms of the proliferative tures.25,26 These deletions mapped to the first c-mpl domain, advantage predominantly affecting megakaryocytes in vivo and one of them (mpl type II) is associated with a higher pro- are unclear. In vitro, endogenous erythroid colony formation, liferative activity in transfection assays than the predominantly ie in absence of added erythropoietin, can be demonstrated expressed transcript mpl type I.26,27 in 60% of patients’ blood or bone marrow (BM) samples, as 8,9 Based on these different results, we analyzed the relative is the case in PV. Similarly, autonomous growth of megakar- expression of c-mpl isoforms in platelets derived from ET yocytic colonies can be observed in blood or BM samples patients and searched for c-mpl mutations in those patients derived from ET patients in the absence of added aplastic with unexplained high counts. serum or growth factors such as 3, 6 (IL-3, IL-6) and GM-CSF.8–11 Hypersensitivity of ET-derived megakary- ocyte progenitors to IL-3 have also been demonstrated in Materials and methods vitro.12 The addition of directed against growth-fac- tors (IL-3, IL-6 or GM-CSF) did not inhibit ET-derived megakar- Patients selection and samples preparation

Correspondence: C Gardin, U 409, Faculte´ X Bichat, 18, rue Henri All patients met the standard PVSG criteria for the diagnosis Huchard, 75018 Paris, France; Fax: 33 1 42264624 of ET.1 Nine ET patients have been studied for c-mpl mRNAs Received 10 March 1997; accepted 31 July 1997 expression. C-mpl cDNA was sequenced from one of these The in ET JJ Kiladjian et al 1822

Figure 1 Exon maps of c-mpl mRNA isoforms. Location of primers for MPL-P, K and S amplification. SP, signal ; CRD, receptor domain; TM, trans-membrane; ICD, intracytoplasmic domain.

patients, and from three other ET patients. Restriction poly- start procedure, in the presence of 5% DMSO, was used only morphisms were then analyzed in 44 ET patients. All selected for the amplification of the Mpl P,K, and S isoforms in a single patients were female patients and had blood samples separ- reaction. Reaction products were analysed on 2% agarose gels ated for X- inactivation studies of granulocytes, stained with EtBr or acrylamide 6% denaturating gels for lymphocytes, and platelets as previously described.7 All analysis of radioactive PCR products. Estimation of the relative patients gave their informed consent for use of this material intensities of each PCR product was obtained by addition of for clinical research. 2 ␮Ci of 32P-dCTP per reaction, and the dried gel was counted C-mpl mRNAs expression was also studied in five healthy on a Packard Instant Imager (Packard Instruments, Rungis, volunteers, and the c-mpl cDNA sequenced from one. Restric- France) using the supplier’s software. tion polymorphism studies were performed on 30 control sub- jects. RNA samples were also prepared from an HEL cell line subculture grown in standard conditions, and from two CD34+ Cloning and sequencing procedures purified cell fractions derived from human leukapheresis products. PCR products were cloned into the Promega (Charbonnieres, Platelets were recovered from 50 ml of blood drawn on France) pGEM-T vector after ligation of gel purified PCR frag- sodium citrate tubes (Vacutainer; Poly Labo, Strasbourg, ments with T4 DNA and transformation of E. coli DH5␣ France). Platelet rich plasma (PRP) was obtained after centri- (Library Efficiency DH5␣, Life Technology). The sequence of fugation at 900 r.p.m. for 15 min. The platelets themselves two independent clones was determined with a T7 DNA Poly- were recovered by centrifugation at 1400 r.p.m., the pellet merase (Pharmacia Biotech, Orsay, France). being directly processed for RNA preparation as described7 or kept at −80°C. Leukocyte DNA was prepared by conven- tional methods.7 C-mpl mutation search and polymorphism analysis

C-mpl mutation search was done by direct sequencing of C-DNA primers selection and reverse transcription overlapping PCR c-mpl cDNA fragments, using a DNA ana- conditions lyser (Genome Express, Grenoble, France). PCR conditions were the same as previously described. D1a, D1b and D1c All primers were designed from the published Mpl-K and Mpl- are used to amplify the region coding for the P sequences (Genbank No. M90102 and M90103)21,24 using and the first domain of Mpl. D2a, D2b, D2c the Oligo 4 program. Primer sequences and hybridation tem- and D2d amplify the second cytokine receptor domain, PKa peratures are listed in Table 1. Location of primers for mRNAs and Pb the intra-cellular domain of the Mpl-P isoform, PKa expression analysis is shown on the c-mpl exon map (Figure and Kb the same region of Mpl-K (Figure 2). 1). Restriction analysis was performed after amplification of Platelet cDNA was typically obtained from total platelet platelet cDNA (Mpl-P) or blood leukocyte DNA (Mpl-K) using RNA, using random hexanucleotides and an M-MLV reverse primers located on each side of polymorphisms detected by transcriptase as recommended by the supplier (Life Tech- the sequencing (PMa and PMb for Mpl-P, KMa and KMb for nology, Cergy Pontoise, France). Mpl-K). Amplified products were then digested by HindIII for Mpl-P, and KpnI for Mpl-K, according to the supplier’s instruc- tions (Life Technology) and analyzed on agarose gels. PCR assays

All reactions were performed in a Hybaid Thermal Reactor Results (Schleicher and Schuell, Cera Labo, Ecquevilly, France), in a 50 ␮l volume with 4 ␮l of platelet cDNA, 12.5 pmoles of pri- New isoforms of c-mpl mers, 0.5 mm dNTP, 1.5 mm magnesium chloride, 0.2 units of Taq polymerase (Appligene, Strasbourg, France). Thirty-five We used the primer PK6 located in the 5Ј end of exon 8 and amplification cycles were performed (15 s at 94°C for denatur- two primers PS and 2K (Figure 1) simultaneously to amplify ation, 20 s for annealing, 18 s at 70°C for elongation). A hot the three known isoforms Mpl-P, Mpl-S and Mpl-K. In addition The thrombopoietin receptor in ET JJ Kiladjian et al 1823 Table 1 Primers used for cDNA amplification

Name Orientation Sequence Position Annealing T° Isoforms (nucleotides) (°C)

PK6 s GAGATCTCCAGTGGGCATCTGG 1198–1219 56 P, K, S PS as GAGGATTTCAAGGAGGCTGGG 1693–1713 56 P, S 2K as CTCCACCTGGTCCACCGC 1593–1610 56 K D1a s CCTCCTGGGCCCTCTTCATG 5–24 60 P, K, S D1b as TGGGAGCTAGCATGGTCCTGTT 914–935 60 P, K, S D1c as GGCACACGTATCGGGTTCCA 261–280 60 P, K, S D2a s TCTTCTACCACAGCAGGGCA 941–960 55 P, K, S D2b as CTGTAAACGGTAGCGAGATCG 1360–1380 55 P, K D2c s ACTGGAAGGTGCTGGAGCCGC 1301–1320 67 P, K D2d as AGGCCCAGGACGGCGCTGA 1511–1539 67 P, K PKa s CTCAACGGCCCCACCTAC 1393–1410 56–63 P, K Pb as TCAAGGCTGCTGCCAATAG 1889–1908 56 P, S Kb as GCCCAGGTAGCTGTGAACAGTA 1692–1715 53 K PMa s AAGATGGACCAAAGCAGACCTC 656–677 56 P, K, S PMb as GCCCAGGTAGCTGTGAACAGTA 1119–1140 56 P, K, S KMa s GGTGACCGCTCTGCATCTAGTG 1482–1503 57 P, K KMb as CCAGAGGTGACGTGCAGGAAGT 1692–1713 57 K

Position is given according to the published sequences. Isoforms, c-mpl isoforms that can be amplified with the primer; s, sense; as, anti-sense.

Figure 2 Location of primers used to sequence MPL-P and MPL-K.

to the Mpl-P, K and S fragments of expected size (516, 413 and 258 bp, respectively), two additional bands were observed on denaturating gels (Figure 3) in all platelet samples analyzed. Each additional band was cloned in pGEM-T and Search for mutations sequenced. As indicated in Figure 4, both fragments were likely to result from the use of an alternative acceptor splicing Four platelet cDNA samples from ET patients were amplified site AG*CTG (nucleotides 1380–1384) in exon 9. The with primers designed to generate overlapping fragments of resulting 72 bp deletion in both Mpl-P2 and Mpl-K2 should the Mpl-P and Mpl-K isoforms (Figure 2). In all samples, the theoretically lead to a 24 amino acid (aa) deletion of the sequence of Mpl-K cDNA differed from the published second c-mpl domain in the same reading frame, starting 38 sequence: GGCC was repeatedly found at position 1616 aa upstream of the WSSWS box. The Mpl-P2 isoform was also instead of CG. The accuracy of our sequencing was checked expressed in the HEL cell line and in CD34+ cells. Mpl-K2 by restriction analysis using StuI which cuts the GGCC was expressed in HEL cells while an additional unidentified sequence. According to this new sequence, the predicted aa band was observed in both HEL and CD34+ cells (Figure 3). sequence of Mpl-K intracytoplasmic domain is 30 aa shorter Using a different set of primers, we were unable to detect iso- than the published sequence (Figure 5). forms of the first cytokine receptor domain, in human platelets A neutral transition (A→G in position 690) was detected in samples or in the HEL or DAMI cell lines (data not shown). one patient. This transition is associated with the loss of a HindIII restriction site. of amplified fragments indi- cated that this sequence variation was present at the hetero- Relative expression of these isoforms zygous state in one of five ET samples but also in one of nine control samples. As shown in Figure 3, Mpl-P is the major c-mpl transcript in A second transition (C→T) was found in one ET sample in human platelets derived from normal subjects or ET patients. position 1570 of the Mpl-K sequence. This mutation is pre- All isoforms were similarly expressed in control and ET dicted to be responsible for an Arg for Cys change in the intra- platelet samples. cytoplasmic region. As the mutation is associated with the loss A similar proportion of Mpl-P, Mpl-K and Mpl-P2 was of a KpnI restriction site, 40 additional blood DNA samples expressed in platelets, CD34+ cells and the HEL cell line. derived from ET patients and 30 DNA from blood donors were The thrombopoietin receptor in ET JJ Kiladjian et al 1824

Figure 3 Expression of mpl isoforms. (a) Simultaneous amplification of the c-mpl mRNA isoforms by PCR using PK6, PS and 2K primers. Lane 1, amplification of HEL cells cDNA; lane 2, CD34+ cells; lane 3, platelet cDNA of a normal subject; lane 4, platelet cDNA of an ET patient. (b) Relative expression of c-mpl isoforms in human platelets. Results are expressed as percentage of total radioactivity per lane (mean ± s.e.m.).

Figure 4 MPL-P2 and MPL-K2. (a) DNA sequence and deduced aa sequence. (b) Exon maps of MPL-P2 and MPL-K2. The deleted sequence is underlined. The thrombopoietin receptor in ET JJ Kiladjian et al 1825

Figure 5 New DNA sequence of Mpl-K and deduced aa sequence. The StuI site created by the insertion of two nucleotides, and the new TGA stop codon are figured in bold characters. The modified aa sequence is underlined. screened by restriction after amplification of that region. One c-mpl receptor mutations which would favorize bind- of 30 blood donors and three of 44 ET patients were hetero- ing, dimerization of the receptor itself, or . zygotes for this sequence variation.

Acknowledgements Discussion This work was supported by a grant from the Fondation pour In this report we describe a new splice variant of the c-mpl la recherche Me´dicale. receptor, expressed in normal as well as in ET-derived plate- lets. The functional role of c-mpl isoforms of the extracellular domain has been analyzed for the mpl type II murine isoform, References that is deleted of 60 aa in the first domain, upstream of the 26,27 WSXWS box as compared with the mpl type I isoform. 1 Pearson T. Primary thrombocythaemia: diagnosis and manage- This mpl type II isoform induces a higher proliferative activity ment. Br J Haematol 1991; 78: 145–148. in transient transfection assays, and is unable to bind the large 2 Kutti J, Wadenvik H. Diagnostic and differential criteria of essen- form of Tpo. Experimental deletions of the extracellular tial thrombocythemia and reactive thrombocytosis. Leuk Lym- domains, associated with defective ligand-binding, have been phoma 1996; 22 (Suppl. 1): 41–45. described for the ,28 and for the ␤ chain 3 Murphy S. Thrombosis and thrombocythemia. Clin Haematol 29 1983; 12: 89–106. of the IL-3 receptor. In the latter case, abnormal dimeriz- 4 Nand S, Stock W, Godwin J, Fisher SG. Leukemogenic risk of ation of the receptor is the putative mechanism of ligand-inde- hydroxyurea therapy in , essential thrombo- pendent activation of the deleted receptor. Mpl-P2 and mpl cythemia, and myeloid metaplasia with myelofibrosis. Am J type II may share the ability to differentially bind Tpo forms Hematol 1996; 52: 42–46. and transduce proliferative signals. The absence of isoforms 5 Fialkow P, Faguet G, Jacobson R, Vaidya K, Murphy S. Evidence of the first domain in human platelets, and the similar location that essential thrombocythemia is a clonal disorder with origin in a multipotent . Blood 1981; 58: 916–919. of both deletions relatively to the WSXWS boxes of each 6 Tsukamoto N, Morita K, Maehara T, Okamoto K, Sakai H, Kara- domain are in agreement with this hypothesis, which should sawa M, Naruse T, Omine M. Clonality in chronic myeloprolifer- be explored by expression studies of Mpl-P2. ative disorders defined by X-chromosome linked probes: demon- The relative expression of truncated intracytoplasmic iso- stration of heterogeneity in lineage involvement. Br J Haematol forms, and of the isoforms of the extracellular domain, pos- 1994; 86: 253–258. sibly altered in their ligand binding ability, was similar in nor- 7 El Kassar N, Hetet G, Brie`re J, Grandchamp B. Clonality analysis of hematopoiesis in essential thrombocythemia: advantages of mal and ET samples. A truncated variant of the erythropoietin studying T-lymphocytes and platelets. Blood 1997; 89: 128–134. receptor, with a shorter intracytoplasmic domain, was shown 8 Westwood NB, Pearson T. Diagnostic application of haemopoietic to be the predominant form of the receptor in early erythro- progenitor culture techniques in polycythaemias and thrombo- blasts.30 Predominant expression of truncated isoforms of c- cythaemias. Leuk Lymphoma 1996; 22 Suppl. 1): 95–103. mpl was not observed in the two CD34+ cell fractions we stud- 9 Juvonen E, ILkkala E, Oksanen K, Ruutu T. and ied, although megakaryocytic progenitors are responsible for erythroid colony formation in essential thrombocythaemia and reactive thrombocytosis: diagnostic value and correlation to com- the c-mpl expression observed in that population. plications. Br J Haematol 1993; 83: 192–197. No mutation was detected in the clonal platelet population 10 Han ZC, Brie`re J, Abgrall JF, Sensebe L, Parent D, Guern G. derived from four ET patients. This is in agreement with the Characteristics of megakaryocyte colony formation in normal indi- absence of mutations of c-mpl exons in nine ET patients viduals and in primary thrombocythemia: studies using an optimal recently reported in an abstract.14 Two nucleotide changes cloning system. Exp Hematol 1989; 17; 46–52. were detected in some patients but also in normal samples. 11 Komatsu N, Suda T, Sakata Y, Eguchi M, Kaji K, Saito M, Miura Y. Megakaryocytopoiesis in vitro of patients with essential throm- These are therefore simple polymorphisms unrelated to the ET bocythaemia: effects of plasma and serum on megakaryocytic col- phenotype. Although the most frequently detected polymor- ony formation. Br J Haematol 1986; 64: 241–252. phism is associated with a drastic amino acid change such as 12 Kobayashi S, Teramura M, Hoshino S, Motoji T, Oshimi K, Mizo- Arg for Cys, this change occurs only in the Mpl-K intracyto- guchi H. Circulating megakaryocyte progenitors in myeloprolifer- plasmic domain which has no similarity with the intracyto- ative disorders are hypersensitive to interleukin-3. Br J Haematol plasmic domains currently known to be associated with 1994; 83: 539–544. 21,23 13 Li Y, Hetet G, Maurer AM, Chait Y, Dhermy D, Brie`re J. Spon- growth factors signalling. taneous megakaryocyte colony formation in myeloproliferative In vitro growth of ET megakaryocytes in the absence of disorders is not neutralizable by antibodies against IL3, IL6 and added growth factors cannot be explained by the existence of GM-CSF. Br J Haematol 1994; 87: 471–476. The thrombopoietin receptor in ET JJ Kiladjian et al 1826 14 Debili N, Le Couedic JP, Ternisien C, Picard F, Wendling F, Vain- brecht S, Souyri M. Molecular cloning and characterization of chenker W, Casadevall N. Autonomous growth of megakaryocyte MPL, the human homolog of the v-mpl : identification progenitors is not due to a mutation of c-mpl. Blood 1996; 88 of a member of the hematopoietic receptor super- (Suppl. 1): 98A. family. Proc Natl Acad Sci USA 1992; 89: 5640–5644. 15 Li Y, Kiladjian JJ, Gardin C, Grandchamp B, Brie`re J. Spontaneous 23 Skoda R, Seldin D, Chiang MK, Peichel C, Vogt T, Leder P. Murine CFU-MK formation in essential thrombocythemia is not due to c-mpl: a member of hematopoietic super- autocrine tpo secretion and disappears with cellular dilution and family that transduces a proliferative signal. EMBO J 1993; 12: use of selected CD34+ cells. Blood 1995; 86 (Suppl. 1): 48A. 2645–2653. 16 Kaushansky K. Thrombopoietin: the primary regulator of platelet 24 Mignotte V, Vigon I, Boucher de Cre`vecoeur E, Romeo PH, Lemar- production. Blood 1995; 86: 419–431. chandel V, Chre´tien S. Structure and transcription of the human 17 Methia N, Louache F, Vainchenker W, Wendling F. Oligodeoxy- c-mpl gene (MPL). Genomics 1994; 20: 5–12. nucleotides antisense to the proto-oncogene c-mpl specifically 25 Alexander W, Dunn A. Structure and transcription of the genomic inhibit in vitro megakaryocytopoiesis. Blood 1993; 82: 1395– encoding murine c-Mpl, a receptor for thrombopoietin. 1401. Oncogene 1995; 10: 795–803. 18 Li Y, Hetet G, Kiladjian JJ, Gardin C, Grandchamp B, Brie`re J. 26 Lofton-Day C, Buddle M, Berry J, Lok S. Differential binding of Proto-oncogene c-mpl is involved in spontaneous megakaryocyto- murine thrombopoietin to two isoforms of murine c-mpl. Blood poiesis in myeloproliferative disorders. Br J Haematol 1996; 92: 1995; 86 (Suppl. 1): 594A. 60–66. 27 Sabath DF, Lofton-Day C, Lok S, Broudy VC. Functional character- 19 Pitcher L, Taylor K, Nichol J, Hornhohl A, Rodwell R, Marty J, istics of naturally occurring isoforms of murine mpl. Blood 1996; Taylor D, Wright S, Moore D, Kelly C, Rentoul A. Thrombopoietin 88 (Suppl. 1): 93A. levels in essential thrombocythaemia: dysregulation and lack of 28 Schimmenti LA, Blechert G, Harris KW, Winkelmann JC. Localiz- diagnostic specificity in thrombocytosis. Br J Haematol 1996; 93 ation of an essential ligand binding determinant of the human (Suppl. 2): 168. erythropoietin receptor to a domain N-terminal to the WSXWS 20 Fielder PJ, Gurney AL, Stefanich E, Marian M, Moore MW, Carver- motif: implications for soluble receptor function. Exp Hematol Moore K, de Sauvage FJ. Regulation of thrombopoietin levels by 1995; 23: 1341–1346. c-mpl-mediated binding to platelets. Blood 1996; 87: 2154–2161. 29 D’Andrea R, Barry S, Moretti P, Jones K, Ellis S, Vadas M, Goodall 21 Vigon I, Florindo C, Fichelson S, Guenet JL, Mattei M, Souyri M, G. Extracellular truncations of h beta c, the common signaling Cosman D, Gisselbrecht S. Characterization of the murine Mpl subunit for interleukin-3 (IL-3), granulocyte–macrophage colony proto-oncogene, a member of the hematopoietic cytokine recep- stimulating factor (GM-CSF), and IL-5, lead to ligand-independent tor family: molecular cloning, chromosomal location and evi- activation. Blood 1996; 87: 2641–2648. dence for a function in cell growth. Oncogene 1993; 8: 2607– 30 Nakamura Y, Komatsu N, Nakauchi H. A truncated erythropoietin 2615. receptor that fails to prevent programmed cell death of erythroid 22 Vigon I, Mornon JP, Cocault L, Mitjavila MT, Tambourin P, Gissel- cells. Science 1992; 257: 1138–1141.