Granulocyte-Macrophage Colony-stimulating Factor Receptors Alter ...

THE JOURNAL OF BIOLOGICAL CHEMISTRY

Val. 268,No. 14, Issue of May 15, pp. 10154-10159,1993 Printed in U.S.A.

0 1993 by The American Society for Biochemistry andMolecular Biolow, Inc.

Granulocyte-Macrophage Colony-stimulating Factor Receptors Alter Their Binding Characteristics during Myeloid Maturation through Up-regulation of the Affinity Converting,8 Subunit (KH97)* (Received for publication, September 9, 1992, and in revised form, November 9, 1992)

Leo M. Budell, Hans Hoogerbruggel, Karin Pouwelsl, Carin van Buitenenl, Ruud Delwelg, Bob Lowenberg$$, and Ivo P.TouwSII From the $Dr. Daniel den Hoed Cancer Center, Rotterdam, The Netherlands and the SErasmusUniversity, Rotterdam, The Netherlands

Acute myeloid leukemia blasts express dual affinity cells there is aconsiderable degree of functional overlap (high and low) granulocyte-macrophage colony-stim- between GM-CSF and interleukin-3 (IL-3) andsome to extent ulating factor(GM-CSF) binding, and the high affinityalso interleukin-5 (IL-5) (6-9). GM-CSF binding is counteracted by excess interleuA common feature of GM-CSF, IL-3, and IL-5 receptorsis a single classof GM- that thehigh affinity interaction of these growth factors with kin-3 (IL-3). Neutrophils express CSF-R with intermediate affinity that lack IL-3 cross- their receptors requires the presenceof two different receptor reactivity.Here we demonstrate the differentiation subunits, an cy and a @ subunit. The cy subunits are specific associated changes of GM-CSF binding characteristics for GM-CSF,IL-3andIL-5andbindtotheir respective in three models representative of different stages of ligands with low affinity. The affinity converting @ subunit, myeloid maturation. W e find that high affinity GMalternatively referred to asBe or KH97 (IO),is shared by the CSF binding is converted into intermediate affinity a chains of GM-CSF-R, IL-3-R, and IL-5-R in the formation binding, which still cross-reacts with IL-3, beyond the of the respective high affinity receptor complexes (9,ll). This stage of promyelocytes. During terminal maturation towards neutrophils, IL-3 cross-reactivity is gradually explains why GM-CSF can compete with IL-3 and IL-5 in the mechanism underlying high affinity binding tovarious human cell types (12-19). lost. We sought to determine Acute myeloid leukemia(AML)blast cellsexpress both the affinity conversion of the GM-CSF-R. Northern high affinity receptors(Kd -50 pmol/liter) which cross-reacts and reverse transcriptase-polymerase chain reaction analysis of GM-CSF-Ra and -PC (KH971 transcripts did with IL-3 and low affinity GM-CSF binding sites (Kd 5-20 nmol/liter) (12-15, 20, 21). In contrast, peripheralblood neunot provide indications for the involvement ofGMwith a singleintermediate CSF-R splice variants in the formation of the inter- trophils express GM-CSF receptors be mediateaffinity GM-CSFR complex. In COS-cell affinity (12-15, 18) (Kd -400 pmol/liter),thatcannot transfectants with increasing amounts of B. in the pres- competed for by IL-3. No explanation has yet been given for ence of a fixed numberof GM-CSF-Ra chains, the high the differences in ligand binding characteristics of GM-CSFaffinity GM-CSF binding converted into intermediate R between AML blasts and neutrophils. Moreover, it has not affinity GM-CSF binding. These results are discussed previously been investigated whether the GM-CSF binding in view of the concept that increasing expressionof PC characteristics progressively change during maturation from subunits may cause alternativeoligomerization of the myeloblasts towards neutrophils,or whether these alterations GM-CSF-Ra and -PC subunits resulting in the formation occur at the terminal stageof neutrophilic differentiation. of intermediate rather than high affinity GMIn the present study, we first determined at which stages of CSFRa-P, complexes. maturation the GM-CSF binding features change. For this purpose we used three leukemia models: 1)primary AML cells that mature spontaneously in culture towardsmyelocytes; 2) chronic phase CML cell samples containing myelocytes, meGranulocyte-macrophagecolony-stimulatingfactor(GMtamyelocytes, and band forms; and 3) the promyelocytic cell CSF)’ is a hematopoietic growth factorwith multilineage line HL-60 in which retinoic acid (RA) induces neutrophilic stimulative abilities (1-4), that exerts its effects on hemato- differentiation. We subsequentlyperformed a series of experpoietic cells through specific membrane receptors ( 5 ) . At the iments toanalyze which mechanisms can be held responsible immature stagesof myeloid development, GM-CSF regulates for the GM-CSF-R affinity changes that take place during proliferation and commitment. In mature neutrophils GMmyeloid maturation. CSF has a role in functional activation. In immaturemyeloid * This work was supported by The Netherlands Cancer Foundation Koningin Wilhelmina Fonds. The costs of publication of this article were defrayed in part by the payment of page charges. This article must thereforebe hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. (1 T o whom correspondence should be addressed: Dr. Daniel den Hoed Cancer Center, P. B. 5201, 3008 AE Rotterdam, The Netherlands. Tel.: 010-4391718; Fax: 010-4842008. The abbreviations used are: GM-CSF, granulocyte-macrophage colony-stimulating factor; IL, interleukin; RA, retinoic acid PCR, polymerase chainreaction; AML, acute myeloid leukemia; CML, chronic myeloid leukemia; R, mean number of binding sites.

MATERIALSANDMETHODS

Cells-AML cells from 2 cases of AML, cytologically classified as FAB M1 (22, 23) and light density CML chronic phase cells of 3 patients were recovered from the interfaceof Ficoll-Isopaque centrifugation of white blood cellsfrom peripheral blood. The cellular composition of these samplesis shown in TableI. None of the patients had received treatment before cell samples were drawn.Normal peripheral blood granulocytes (90-95% pure) were obtainedafter sedimentation of the blood of healthy subjects in 0.1% methyl cellulose and Ficoll-Isopaque centrifugation (24) as previously described (25). The HL-60 cell line was obtained at passage level 28 (26) from Dr. Farzaneh, King Cross College, London, United Kingdom.

10154

10155

GM-CSF-R and Myeloid Maturation TABLE I Morphology of AML, CML, and HL60 cell samples Numbers indicate percentages of total cells scored. Cells

9

AML 1 dO d8 AML 2 dO dl0 CML 1 CML 2 CML 3 HL-60 HL-60 RA

+

Myeloblasts Promyelocytes

Myelocytes

67 12

12 36

12 30

58 6 10 8

29 28 16 12 31 100

51 16 17 30

Metamyelocytes Band

forms

Other” Segmented

9 15

7

9 9 31 32 30

27 28

45

33

3 22

Indicates other morphologies, i.e. of megakaryocytic, monocytic, and eosinophilic lineages. I n Vitro Culture-The HL-60 cells were maintained in RPMI,10% precipitate.Forty-eight h post-transfection,theCOS-1 cells were 1 nmol/literEDTA, gentlyforced fetal calf serum a t a density of maximally 5-6 X lo5 cells/ml. Gran- detachedandharvestedwith through a 22-gauge needle to obtaina single-cellsuspension, andused ulocytic maturation in HL-60 was induced by incubating exponentially growing cells (6 X 105/ml) in RPMI, 10% fetal calf serum with in binding studies. fff]Thyrnidine Incorporation into AML Cells-DNA synthesis of 1 p~ all-trans-retinoic acid (RA, Sigma) for 5 or 6 days. After culture AML cellswas measuredas described (6). Two X lo‘ cells were with RA, the percentage of cells with polymorphic nuclei exceeded 90% (Table I). The AML cells were cultured in RPMI,10% fetal calf cultured for 3 days in 96-well round-bottom microtiter trays in 100 serum with 100 pmol/liter IL-3, and after 8-20 days harvested for pl of serum-free medium, with or without addition of IL-3 or GMCSF. Four h before harvesting, 0.1 pCi of titrated thymidine (Amerreceptor assays. Examination of cytospin smears of cultured AML cells showed that the cells had lost their blastic appearance and had sham) was added. Cells were harvested on nitrocellulose paper using proceeded towards promyelocytic/myelocytic stages of maturation a Titertek Harvester 550 (Flow Laboratories, Isrike, United King(Table I). dom). Radioactivitywas determinedwith a scintillationcounter Radioiodination of Growth FactorsandBinding Experiments(Beckman LS 3800, Fullerton, CAI. All experiments were performed Recombinant human GM-CSF (glycosylated, Chinese hamster ovar- in triplicate and dataexpressed as mean disintegrations/min. ium cell derived; molecular mass 20-30 kDa) (27) and IL-3 (Escherichia coli derived; molecular mass 15 kDa) (28) were prepared and RESULTS purified at Genetics Institute (Cambridge, MA). GM-CSF and IL-3 Comparison of GM-CSF Receptor Numbers and Affinity in were radiolabeled with Bolton and Hunter reagent (Amersham Laboratories, Amersham, United Kingdom) (29) as described (30). Spe- Noncultured Versus Cultured AML Cells-Binding charactercific activities of radiolabeled growth factors, as determined by self- istics of GM-CSF receptors on AML blasts (case 1) before displacement analysis (31),were 5-7 X lo4cpm/ng for GM-CSF and and after culture are shown in Fig. 1 and Table 11. In line 4-6 X lo4cpm/ng for IL-3. Trichloroacetic acid precipitationshowed with previous observations (12-15, 18),the noncultured cells lessthan 5% nonprecipitableradioactivity for bothfactors. The maximum binding capacity was estimated at 90-95% for GM-CSF expressed few high affinity GM-CSF receptors (60 sites/cell; and 55-60% for IL-3. In Scatchard calculations the “free” counts/ Kd 50 pmol/liter) and -30-fold more low affinity GM-CSF min were corrected for the maximum binding capacity (31). receptors (1,900 sites/cell; Kd 20 nmol/liter). In the presence GM-CSF and IL-3receptor analysis were performed as previously of excess (200 nmol/liter) nonlabeled IL-3, GM-CSF binding described (30). Cross-competition experiments were performed with to the high affinity GM-CSF receptors was completely pre200 nmol/liter heterologous growth factor (12). Experiments were carried out in duplicate or triplicate. Receptor numbers and binding vented (Fig. 1A). In contrast, after 8-20 days of culture, the affinities were derived from Scatchard plot analysis, using the ENZ- cells, then predominantly showing a promyelocytic/myelocyfitter program (Sigma). tic morphology, had lost the characteristic high affinity GMPCR Analysis of GM-CSF-Ra and /3 Transcripts-Total RNA was CSF binding. In return, intermediate affinity GM-CSF bindisolated from the cells after lysis in guanidium thiocyanate containing ing sites appeared (Fig. 1B) (Kd = 350 pmol/liter; R = 2,700 buffer (32) followed by cesium chloride gradient centrifugation (32). cDNA prepared from these RNAs using reverse transcriptase (Mo- sites/cell). Low affinity bindingwas markedly reduced. Excess IL-3 (200 nmol/liter) virtually prevented all GM-CSF binding loney murine leukemia virus-reverse transcriptase superscript, (81-93%; Fig. 1B). The remainder of the GM-CSF bound GIBCO-BethesdaResearchLaboratories,Gaithersburg,MD) were used as templates in the PCR analysis. Primers spanning the extra- with a low affinity. Theoretically, the presence of low affinity cellular plus transmembrane domain were: 5”TCAGATCTGCACGM-CSF binding after cross-competition with IL-3could be 5’-CCTTTTAAAGAGGAAGCCG-3’ CATGCTTCTCCTG-3’and corresponding to base pair 137-1193 of the published GM-CSF-Ra explained by the residual presence of AML blast cells after sequence (33),and 5’-CAGAGCTGACCAGGGAGATGGTGCT-3’ culture (Table I), which express excess GM-CSF-Ra chains. corresponding to base pairs However, because the presence oflow affinity binding was and 5’-GTAGATGCCACAGAAGCG-3’ 12-1426 of the published & sequence (10). not evident from the Scatchard analysis without IL-3 comConstruction of GM-CSF-Ra and p Deletion Mutants and Expres- petition, it is more likely that the residual low affinity GMsion in COS Cells-Deletion mutants of both GM-CSF-R subunits, CSF binding is caused by the GM-CSF-Ra subunits dissothat lacked the complete cytoplasmic part (GM-CSF-Rakyt, GMcomplex CSF-RpAcyt) were obtained by PCR from complete cDNAs, using ciatedfromtheintermediateaffinityGM-CSF-R after IL-3 competition.Comparable GM-CSF binding results the primers indicated above. Both truncated receptor cDNA fragments were cloned into thepCMV, expression plasmid (pCMV4-GM- before and after in vitro culture were obtained in the other CSF-RaAcyt, pCMV4-GM-CSF-Rp~cyt) (34). GM-CSF-Ra cDNA, AML case (Table 11). provided byDr. N.Gough (33), was cloned into thepCMV, expression GM-CSF Binding to CML Cells-The CML chronic phase was provided plasmid. The 6. expression plasmid (pSVhuIL-5-Rp (9)) three cases exclusively expressed GM-CSF by Drs. G. Plaetinck and J. Tavernier (Roche Research, Gent, Bel- cellsfromall receptors of intermediate affinitytype ( K d 330-390 pmol/ gium). COS-l cells were transfected with (complete or truncated) a and /3 subunits by C a p o 4precipitation (32). The expression of p, was liter), at densities that were somewhat lower than those on varied by changing the amount of & expression plasmid in the Capo4normal granulocytes (Table 11). Excess1L-3competed 20-

10156

GM-CSF-Rand Myeloid Maturation 0.15

0.10

BIF

BIF

0.05

0.00 0

1 2 3 4 5 6 GM-CSF bound (pM)/l E6 cells

0

7

3010 20 40 GM-CSF bound (pM)/lE6 cells

50

FIG. 1. Scatchard plot of GM-CSF receptor-binding in AML cells (case 1) before and after culture. A , AML cells: fresh (6.4 x andB, after 20 days of culture (0.95 X lo6cells/tube) were incubated with '251-GM-CSF(40-5000 pmol/liter). Bindingof GMCSF was determined as described under "Materials and Methods." Scatchard plots of parallel GM-CSF binding experiments that were performed in the absence (open circles) or presence of excess IL-3 (closed circles). Data were corrected for nonspecific binding and normalized for cell numbers (data are expressed as binding perlo6 cells). Each point representsthe mean of two estimations. B/F, bound/free.

lo6cells/tube),

TABLEI1 GM-CSF binding characteristicsof AML, CML, HL60cells, and neutrophils Mean numbers of binding sites ( R ) (mean f S.D.) and dissociation constant (&) (fS.D.) were derived from radiolabeling studies with (AML and HL-60 cells). Datawere analyzed withthe program ENZfitter noncultured cells or cells that were cultured during indicated periods (Elsevier Biosoft). Without IL-3

Cells

With IL-3 (200 nmol/liter)

R

Kd

R

pmol/liter

AML 1 do" d8 d20 AML 2 do" dl0

CML 1 CML 2 CML 3

60 f 45 1,900 f 950 2,600 f 82 2,700 f 50 59 f 3 30,000 1,600 f 570 430 360 f 14 610 f 10 540 f 18 520 f 8

50 f 16 20,000 f 13,000 390 f 38 350 f 17 44 f 7 f 14,000 f 39 330 450f 20 340 220f 30 390 f 16

Kd

pmollliter

0 1,900 f 780

18,000 f 7,900

NDb 720 f 150

3,000 f 1,100

ND 610t 25 f 23 360 f 8

f 93 980 f 250 410 f 25

HL60

Experiment 1 Experiment 2 HL60 + RA d5 d6

380 f 8 200 470 f 9 530 f 28 1,230 f 36 2.100 f 180

370 f 16 310 f 9 250 f 35 450 f 33 f 53 410 -

NeutroDhils (n = 3) Binding data were in favor of dual binding characteristics. * ND, not determined. e Specific binding detectable, but toolow for Scatchard analysis.

240 f 9 t9 515 f 59 1,220 490 f 46 2.200 f 160

1,100 f 71 120 f 10 250 f 74 f 48 420 f 21

a

60% of labeled GM-CSF binding (Table 11). In cases 1 and 2, some low affinity binding appeared tobe present after crosscompetition with IL-3 (Table 11). However, in all cases the major portion of IL-3-resistant GM-CSF binding sites expressed the characteristic intermediate affinity. GM-CSF Binding to HL-60 Cells Before and After Treatment with All-trans-Retinoic Acid-HL-60 cellsexpressed 380-470 GM-CSF binding sites/cell of intermediate affinity ( K d 200-370 pmol/liter, Table 11). Excessnonlabeled IL-3 competed 30-40%of lZ5I-GM-CSF binding to HL-60 cells. The K d value of the GM-CSF binding that remained detectable after IL-3 competitionwas lower than in the absenceof IL-3, suggesting the presence oflow affinity binding. After exposure of the HL-60 cells to RA(5-6 days), IL-3 crossreactive GM-CSFbinding was completely lost(Table 11). Parallel experiments withradiolabeled IL-3 showed that upon RA treatment the HL-60 cells lost IL-3 binding sites (data not shown).

GM-CSF Induces a Proliferative Response in AML Cells Expressing Intermediate Affinity GM-CSF Receptors-We measured DNA synthesis in response to GM-CSF in AML cells before (expressing high and low affinity GM-CSF receptors) and after10 days of culture (when the cells had acquired intermediate affinity GM-CSF receptors). In both cases the cells were still able to respond to GM-CSF after 10 days of culture, indicating that the intermediate affinity GM-CSF receptor is still capableof conducting a mitogenic signal (Fig. 2). The plateau levels of GM-CSF induced DNA synthesis before and after culture did not differ significantly. However, the dose-response curves are indicative of aslightloss of sensitivity to GM-CSF after culture (EDs0before culture: -20 pmol/liter; ED,, after culture: -40 pmol/liter), possibly as a result of the lowered affinity GM-CSF receptor on the cultured cells. Reverse Transcriptase-PCR Analysis of the Extracellular

M a t u r aMt iyoenl o i d 3HTdR uptake (dpmx 1V3 )

10157

a n dG M - C S F - R 7,Pst

I

1

2

5

4

3

6

7 hPst

A

8-

hPSt

1

2

3

4

5

6 hPst

I

c GM-CSF (pM)

PHTdR uptake (dpm x l V 3 )

FIG.3. PCK amplification of extracellular domains of GM( B )from AML (lanesAI-A3,B I , a n d B 2 ) , CSF-Ra ( A )and -0,. AML day 10 (lane B 3 ) , HL-60 (lanes A4and B5),CML (lane B 4 ) , and neutrophils (lanes A 5 , A 6 , and B 6 ) . XPst indicates XPstI markers.

GM-CSF (pM)

FIG. 2. Sensitivity to GM-CSF on fresh and cultured AML cells. DNA synthesis of AML cells ( A , case 2; R, case 1) that had been cultured for 10 days (closed circles) as well as freshly thawed AML cells (open circles) was determined in the presence of titrated concentrations of GM-CSF.ED,, values for each dose-response curve are annotated.

considered the possibility that increasedavailability of PC could be held responsible for the affinity changesduring myeloid maturation. To verify this hypothesis, we performed COS cell transfectionsin which theGM-CSF-Rachain expression was held at a constant level and the expression of PC was increased (Fig. 4B). Indeed we found that increases in the expression of PC in relation to GM-CSF-Ra affected the affinity of the GM-CSF-Ra.P, complex in a manner similar to the affinity changesin the leukemia models. DISCUSSION

The data of this study show that the properties of GMCSF-R on hematopoietic cells alter with progressive myeloid Domains of the GM-CSF-Ra and -P Subunits-By Northern maturation. During theconversion from myeloblasts towards analysis, transcripts for GM-CSF-Ra and-P could be detected (pro)myelocytes, thecells lose the characteristichigh and low in AML cells, CML cells, and HL-60 cells, as well as neutro- affinity GM-CSF binding to the GM-CSF-Ra .PCcomplex and phils (data not shown). T o investigate whethersplice variants the single GM-CSF-Ra chain,respectively, and acquire interof the GM-CSF-R extracellular domains would be involved in mediate affinity GM-CSF receptors. At this stage of maturathe formation of intermediate affinity GM-CSFreceptors, we tion, cross-competition by IL-3 for PCaffects the intermediate analyzed transcripts of the GM-CSF-Ra and -P in cells a t affinity GM-CSF binding. In the phase during which the cells different stages of myeloid maturation by reverse transcrip- mature towards terminally differentiatedgranulocytes, crosscompetition by IL-3 is gradually lost. This gradual loss of tase-PCR analysis using PCR primers spanning the extracellular plus transmembrane regions of the N and P chains. We cross-competition by IL-3 is caused by the disappearance of PCR amplifiedfromall cell samples the fragments of the IL-3-Ra subunits. In agreementwith this notion, RA-treated expected lengths of the wild type receptor, 1056 and 1414 HL-60 and neutrophils were found to lack significant IL-3 nucleotides, respectively, but no alternativelysized fragments binding. We have attempted to elucidate the events that takeplace (Fig. 3). These data rule out alternative RNA splicing as a during the first stage of myeloid maturation, leading to the mechanism involved in theformation of theintermediate loss of high and low affinity GM-CSF binding and the apaffinity GM-CSF receptors. IncreasingExpression of in COS Cells Results in the pearance of intermediate affinity receptors. Two mechanisms Conversion from High Affinity Toward Intermediate Affinity have previously been reported to affect the affinity of hemoof the GM-CSF Receptor Complex-Expression of GM-CSF- poietinreceptors. Fukunaga et al. ( 3 5 ) have demonstrated Ra and-P in principle can reconstitutehigh affinity GM-CSF that structural alterationsof the extracellular domain of the from naturally receptors in COS cells (10). Truncation of the cytoplasmic human and murine G-CSF-R, either resulting part of GM-CSF-Ra and/or -PC does not affect the formation occurring alternative RNA splicing or caused by experimenof the high affinity GM-CSF receptorcomplex formed in COS tally created deletions, may result in the decreased affinity of cell transfectants (Fig. 4A). Because immature myeloblasts the G-CSF-R. We therefore investigated whether GM-CSFexpress few (50-100) high affinity GM-CSF receptors, Ra and or -PC splice variants with altered extracellular dowhereas differentiated cells expressmuch higher numbers mains could be detected in cells expressing the intermediate (2000-3000) of intermediate affinity receptors and because affinity GM-CSF-R. However, no evidence for the presence Northern analysis indicated that fi,transcripts, but not GM- of such splice variants in cultured AML cells, HL-60 cells, CSF-Ra transcripts, increase with progressive maturation, we CML cells, and neutrophils was obtained. fiC

10158

GM-CSF-R and Myeloid Maturation 0.4

0.3

Bff 0.2

0.1

0.0 0

100

200

GM-CSF bound (pM)

300

0

200

400

GM-CSF bound (pM)

FIG. 4. Scatchard plots of GM-CSF receptor binding in COS transfectants ( A ) GM-CSF-R cytoplasmic deletion mutants expressed in COS cells reconstitute high affinity GM-CSF binding. Scatchard plot of GM-CSF receptor binding in COS-1 cells cotransfected with GM-CSF-RaAcyt and GM-CSF-RBAcyt (&I = 79 pmol/liter, R1 = 2,100 sites/cells, K d z = 18 nmol/liter, RZ = 25,000 sites/cell).B, the affinity of the GM-CSF receptor complex in COS cellsalters as a result of increased 8, expression. Scatchard plotof GMCSF binding to COS-1 cells cotransfected with 10 pg of GM-CSF-RaAcyt expression plasmid and10 pg (closed circles; K d l = 83 pmol/liter, Rl = 2,600 sites/cells, KdZ = 24 nmol/liter, R2 = 37,000 sites/cell), or 50 pg (open circles; K d = 680 pmol/liter, R = 12,000 sites/cells) Bc expression plasmid. The second mechanism that has been shown to influence one /3 subunit, which could become an intermediate affinity receptor affinity is the involvement of additional receptor or receptor when increased availability of PC permits aPP comreceptor-linked structures. Recently a third receptor chain, plexes. The latter alternative of receptor complex formation involved in the formation and function of the high affinity has recently also been proposed for the IL-6-R andleukemia IL-2 receptor complex has been identified and molecularly inhibitory factor-R (41). Further investigations of the intercloned (36). COS cell transfectants expressing human IL-5- active domains of GM-CSF and the GM-CSF-R structures Ra and -P showed intermediate rather thanhigh affinity IL- are required to solve this issue. 5 binding (9).Similarly, IL-6 binds with intermediate affinity So far, no functional differences between high and interto IL-6-R/gp130 COS cell transfectants (37). On this basis, mediate affinity GM-CSF-R have been identified. AML cells the possible involvement of additional receptor structures in expressing the intermediate affinity GM-CSF-R after culture the formation of high affinity IL-5-R as well as the IL-6-R, were capable of responding to GM-CSF with only slightly leukemia inhibitory factor-R, and oncostatin M-R complex decreased sensitivity as compared with the fresh cells that has been postulated (9, 37). In contrast, it hasbeen reported, expressed the high affinity GM-CSF-R. Moreover, identical and here confirmedby us, that COS aswell as NIH-3T3cells tyrosine phosphorylation patterns were observed after GM-PC CSF stimulation of M07e cells, which express high affinity (data not shown), cotransfected with GM-CSF-Ra and GM-CSF-R and neutrophils (intermediate affinity GM-CSFcDNAscanexpress high affinityGM-CSFbinding(10). Transfection of COS-cellswith GM-CSF-Ra and -PC cyto- R) (42). What then could be the physiological relevance of plasmic deletion mutants also resulted in high affinity GM- alternative oligomerization? Increasing evidence has become CSF binding, proving that the intracellular domains are not available that upon ligand induced di- or oligomerization of critical for the formation of high affinity GM-CSF receptors. receptor subunits, interaction between the intracellular doThus, no experimental data have beenforwarded to indicate mains of the receptor structuresis required for growth factor -P receptor activation (35, 41, 43). Soluble forms of GM-CSFthat a componentadditionaltotheGM-CSF-Raand structures is involved in the formation of the high affinity Ra and IL-5-Ra have been reported to antagonize the actiGM-CSF-R complex. On the other hand, our datafrom COS vation of the membrane receptors. These observations indicell transfection experiments established that the level of PC cate that the intracellular domains of the GM-CSF-Ra and IL-5-Ra have a role in receptor activation. In contrast, assoexpression relative to GM-CSF-Ra directly determines the ciation of a soluble form of the IL-6-Ra (lacking the transaffinity of the GM-CSF-R complex. Therefore,itappears most likely that duringmyeloid maturation, GM-CSF-R bind- membrane andcytoplasmic parts) with the signal transducing ing characteristics are alteredsimply by up-regulation of the gp130 molecule results in the formationof a functional IL-6R. Hence it is likely that gp130 subunit dimerization, rather PC subunit. The exact configuration of the high and intermediate affin- than IL-6-Ra/gpl30 dimerization activates the IL-6 signallobservations, we now assume ity GM-CSF-R is still unknown. Kaushansky et al. (38) and ing process (41). Based on these Shanafelt et al. (39) have shown that two regions (residues that, depending on the availability of PC subunits, GM-CSF 14-24 and 77-94) in the C terminus of human GM-CSF are signalling can be mediated through ap or through aP and PP important for biological activity. In addition, the involvement association. Whether these alternativeways of receptor actiof regions 40-77 and 110-127 in receptor binding has been vation have qualitative implications for GM-CSF-mediated suggested by Nice et al. (40). These results predict that GM- responses (other thanmitogenic activation and tyrosine phosCSF can bind to several distinct binding sites of the GM- phorylation) or simply represent another example of receptor redundancy (44) remains to be answered. Hence, future inCSF-R complex andopenthe possibility thatalternative vestigations will focus on therole of the cytoplasmic domains oligomerization of GM-CSF-Ra and -PC subunits determine the different affinitiesof the receptor complex. For instance, of the GM-CSF-Ra and -PC subunits in signal transduction one could hypothesize that in the case of low availability of activated by the different oligomeric GM-CSF receptor comPC, a high affinity complex is formed by one a subunit and plexes.

GM-CSF-R and Adyelc3id Maturation REFERENCES Metcalf, D. (1985) Science 229,16-22 Metcalf, D. (1986) Blood 67,257-267 Clark, S. C., and Kamen, R. (1987) Science 236, 1229-1237 Sieff, C. A. (1987) J. Clin. Invest. 97,1549-1557 Nicola, N. A. (1987) Immuml. Today 8, 134-140 Delwel, R., Salem, M., Pellens, C., Dorssers, L., Wagemaker, G., Clark, S. C., and Lowenherg, B.(1988) Blood 72,1944-1949 7. Bot, F. J., Eijk, L., Schipper, P., and Lowenherg, B. (1989) Exp. Hematol.

1. 2. 3. 4. 5. 6.

25. Budel, L. M., Touw, I. P., Delwel, R., and Lowenherg, B. (1989) Blood 74, 2668-2673 26. DafF>g:AD., Linskens, M. H. K., and Farzaneh, F. (1989) Leuk. Res. 13, 4UI-415

21. Wong, G. G., Witek, J. S., Temple, P. A,, Wilkens, K. M., Leary, A. C.,

28.

17, 292-259

8. Bot, F. J., Dorssers, L., Wagemaker, G., and Lowenherg, B. (1987) Blood

71, 1609-1614 9. Tavernier, J., Devos, R., Cornelis, S., Tuypens, T., Van der Heyden, J., Fiers, W., and Plaetinck, G.(1991) Cell 66,1175-1184 10. Hayashida, K., Kitamura, T., Gorman, D. M., Arai, K., Yokota, T., and Miyajima, A. (1990) Proc.,Natl. Acad. Sci. U. S. A. 87,9655-9659 11. Kitamura, T., Sato, N., Arat, K., and Miyajima, A. (1991) Cell 66, 11651174 12. Budel, L. M., Elhaz, O., Hoogerhrugge, H., Delwel, R., Mahmoud, L. A., Lowenherg, B., and Touw, I. P. (1990) Blood 75,1439-1445 13. Onetto-Pothier, N., Aumont, N., Haman, A,, Park, L., Clark, S. C., De LBan, A,, and Hoang, T. (1990) Leukemia 4,329-336 14. Chiha, S., Tojo, A,, Kitamura, T., Urahe,A,, Miyazono, K., and Takaku, F. (1990) Leukemia 4.29-36 15. Elliott, M. J., Vadas, M.A,, Eglinton, J. M., Park, L. S., To, L. B., Cleland, L. G., Clark, S. C., and Lopez, A. F. (1989) Blood 74, 2349-2359 16. Park, L. S., Friend, D., Price, V., Anderson, D., Singer, J., Prickett, K. S., and Urdal, D. L. (1989) J. Biol. Chem. 264,5420-5427 17. Tamura, S., Sugawara, M., Tanaka, H., Tezuka, E., Nihira, S., Miyamoto, C., Suda, T., and Ohta,Y. (1990) Blood 76,501-507 18. Cannistra, S. A., Konigsmann, M., DiCarlo, J., Groshek, P., and Griffin,J. D. (1990) J. Biol. Chem. 265,12656-12663 19. Lopez, A. F., Eglinton,J. M., Gillis, D., Park, L. S., Clark, L. S., and Vadas, M. A. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 7022-7026 20. Onetto-Pothier, N., Aumont, N., Haman, A,, Bigras, C., Wong, G.G., Clark, S. C., De LQan,A., and Hoang, T. (1990) Blood 75,59-66 21. Murohashl, I., Tohda, S., Suzuki, T., Nagata,K., Yamashita, Y., and Nara, N. (1989) Leuk. Res. 13, 599-604 22. French-American-British (FAB) Cooperative Group (1985) Ann. Intern. Med. 103,620-625 23. French-American-British (FAB) Cooperative Group (1976) Br. J. Hoematol. 33,451-458 24. Boyum, A. (1968) S c a d . J. Clin. Lab. Invest. 21, 77-89

10159

29. 30.

31.

Luxenherg, D. P., Jones, S. S., Brown, E. L., Kay, R. M., Orr, E. C., Schoemaker, C., Golde, D. W., Kaufman, R. J., Hewick, R. M., Wang, E. A,. and Clark, S. C. (1985) Science 228,810-815 Yang, Y. C., Ciarletta, A. B., Temple, P. A., Chung, M. P., Kovacic, S., Witek-Giannotti, J. S., Leary, A. C., Kritz, R., Donahue, R. E., Wong, G. G., and Clark, S. C. (1986) Cell 47,3-10 Bolton, A. E., and Hunter, W. M. (1973) Biochem. J . 133,529-539 Budel, L. M., Touw, I. P., Delwel, R., Clark, S., and Lowenherg, B. (1989) Blood 74,565-571 Calvo, J. C., Radicella, J. P., and Charreau, E.H. (1983) Biochem. J. 212,

259-264 32. Samhrook, J., Fritch, E. F., and Maniatis, T. (1989) Molecular Cloning: A

Laboratov Manual,Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

33. Gearing, D. P., King, J. A., Gough, N. M., and Nicola, N. A. (1989) EMBO J . 8,3667-3676 34. Anderson, S., Davis, D. L., Dahlhack, H., Jornvall, H., and Russell, D.W. (1989) J. Biol. Chem. 264,8222-8229 35. Fukunaga,R.,Ishizaka-Ikeda, E., Pan, C. X., Seto, Y., and Nagata, S. (1991) EMBO J . 10, 2855-2865 36. Takeshita, T., Asoa, H., Ohtani, K., Ishii, N., Tanaka, N., Munakata, H., Nakamura, M., and Sugamura, K. (1992) Science 257, 379-380 37. Gearing, D. P., Comeau, M. R., Friend, D. J., Gimpel, S. D., Thut, C. J.,

McGourty, J., Brasher, K. K., K,ing, J. A., Gillis, S., Mosley, B., Ziegler,

S. F., and Cosman, D. (1992) Scwzce 255, 1434-1437 38. Kaushansky, K., Shoemaker, S. G., Alfaro, S., and Brown, C. (1989) Proc. Natl. Acad. Sci. U. S. A. 86, 1213-1217 39. Shanafelt. A. B.. Mivaiima. A.. Kitamura. T.., and Kastelein. R. A. , L . J . , L , EMBOh. 10,’4105-i112 40. Nice, E., Dempsey, P., Layton, J., Morstyn, G., Gui, D. F., Simpson, R., Fahri, L., and Burgess, A. (1990) Growth Factors 3, 159-169 41. Ip, N. Y., Nye, S. H., Boulton, T. G., Davis, S., Taga, T., Li, Y., Birren, S. I

~~

~

~

(’“”’

J., Yasukawa, K., Kishimoto, T., Anderson, D. J., Stahl, N., and Yancopoulos, G. D. (1992) Cell 69, 1121-1132 42. Okuda, K., Sanghera,J. S., Pelech, S. L., Kanakura, Y., Hallek, M., Griffin, J. D., and Druker, B. J. (1992) Blood 79, 2880-2887 43. Cunningham, B. C., Ultch, M., de Vos, A. M., Mulkerrin, M. G., Clauser, K. R., and Wells, J. A. (1991) Science 254,821-825 44. Nicola, N. A., and Metcalf, D. (1991) Cell 67, 1-4