A monoclonal antibody that inhibits the action of GM-CSF on normal but not leukaemic progenitors

Leukemia Research Vol. 14, No. 7, pp. 637~zt4, 1990. Printed in Great Britain.

0145-2126/90 $3.00 + 0.00 Pergamon Press plc

A M O N O C L O N A L ANTIBODY THAT INHIBITS THE ACTION OF GM-CSF ON N O R M A L BUT NOT LEUKAEMIC PROGENITORS LEONIE K. ASHMAN,* ANTONY C. CAMBARERI* and JEFFREY M. EGLINTONt *Department of Microbiology and Immunology, The University of Adelaide and tDivision of Human Immunology, Institute of Medical and Veterinary Science, Adelaide, South Australia

(Received 24 November 1989. Revision accepted 18 March 1990) Abstract--Monoclonal antibody YB5.B8 was previously shown to inhibit haemopoietic colony formation in response to a complex growth factor supplement in vitro (Cambareri A. C., Ashman L. K., Cole S. R. & Lyons A. B. (1988), Leukemia Res. 12, 929). We now report studies of the effect of the antibody on colony formation by normal human bone marrow cells in response to recombinant human colony-stimulating factors GM-CSF, G-CSF and IL-3. MAb YB5.B8 significantly reduced the yield of colonies of all types examined (granulocyte-macrophage, granulocyte, macrophage and eosinophil) in response to GM-CSF but not to IL-3 or G-CSF. However, MAb YB5.B8 failed to influence the proliferation of the myelomonocytic leukaemia cell line RC-2A in response to GM-CSF, G-CSF or IL-3. Direct binding studies demonstrated the presence of low numbers of receptors for GM-CSF on RC-2A cells, however, the binding of this cytokine was not influenced by co- and/or pre-incubation with MAb YB5.B8. Therefore the antigen identified by YB5.B8 is probably not a receptor for GM-CSF and may indirectly influence the response of normal haemopoietic progenitors to this cytokine.

Key words: Haemopoietic progenitor, colony-stimulating factor, myeloid leukaemia, monoclonal antibody.

INTRODUCTION

expression at low levels on the cell surface, suggested that YB5.B8 might bind to the receptor for a haemopoietic growth factor, perhaps IL-3. Furthermore, the association of over-expression of the antigen with poor prognosis in A N L L [5] parallelled the observations with HER-2/neu, a growth factor receptor of the EGF-receptor family, in breast and ovarian cancer [6, 7]. To test the hypothesis that YB5.B8 identifies the receptor for a known haemopoietic growth factor, we studied the effect of the antibody in the CFUGM assay using recombinant GM-CSF, G-CSF or IL-3 as stimulus. In addition, the antibody was examined for its ability to modify the proliferation of myeloid leukaemia cells in response to recombinant cytokines, and to block cytokine binding to these cells. The inhibitory effect of the antibody in the C F U - G M assay was specific for GM-CSF, however, studies with myeloid leukaemia cells indicated that the antibody probably does not bind to the GM-CSF receptor.

MONOCLONAL antibody YB5.B8 [1] identifies a 150,000 mol.wt cell surface molecule which is restricted to tissue mast cells [2, 3], approximately 1% of normal bone marrow cells including haemopoietic progenitor cells which give rise to colonies in vitro [4], and blast cells from some patients with A N L L [1, 5]. Recently we showed that inclusion of the YB5.B8 antibody in the culture medium in the CFUGM assay using a complex growth factor mixture (HPCM) as stimulus resulted in up to 50% inhibition in the yield of colonies [4]. This result, together with the unusual distribution of the antigen and its

Abbreviations: ANLL, acute non-lymphoblastic leukaemia; CFU-GM, colony-forming unit granulocytemacrophage; CGL, chronic granulocytic leukaemia; EGF, epidermal growth factor; G-CSF, granulocyte colonystimulating factor; GM-CSF, granulocyte macrophage colony-stimulating factor; HPCM, human placental-conditioned me. .m; IL-3, interleukin 3; IL-4, interleukin 4; M-CSF, macrophage colony-stimulating factor; MNC, mononuclear cells; rh-, recombinant human; TNFo:, tumour necrosis factor o:. Correspondence to: Dr L. K. Ashman, Department of Microbiology and Immunology, The University of Adelaide, G.P.O. Box 498, Adelaide, 5001, South Australia.

MATERIALS AND METHODS

Cells Normal bone marrow cells were obtained from allogeneic transplant donors at the Royal Adelaide Hospital. 637

638

L.K. ASHMANet al.

Specimens were obtained with informed consent, and approval of the relevant ethics committees. Mononuclear cell fractions were prepared and cryopreserved as described previously [8]. Cryopreserved peripheral blood CGL cells were kindly provided by the Division of Haematology, IMVS. Cells of the RC-2A myelomonocytic leukaemia line [9] were obtained and maintained as described previously [10]. The KG-1 and HL-60 cell lines were obtained from ATCC. All cell lines were mycoplasma-free.

chloride method [14]. 125I-labelled GM-CSF (125I-GMCSF) was separated from iodide ions by Sephadex G25 chromatography (Pharmacia) using phosphate-buffered saline elution buffer containing 0.02% polyoxyethylene sorbitan monolaurate (Tween 20), and stored at -20°C. Prior to use, 125I-GM-CSF was purified from non-proteinassociated radioactivity by cation exchange chromatography on a 0.3 ml carboxymethyl-Sepharose CL-6B column and stored at 4°C for up to 5 days.

Cytokines

binding medium (RPMI-1640, pH 7.4 supplemented with 20mM Hepes, 0.5% bovine serum albumin and 0.1% NAN3) at a concentration of 4 x 10 7 cells/ml with monoclonal antibodies 1B4, YB5.B8 and 3D3.3 at concentrations of 2.0 and 0.2 ~tg/ml for 5 h at 4°C on a rotating mixer. In a subsequent experiment, cells were incubated at a concentration of 5 x 10 7cells/ml with monoclonal antibodies at a concentration of 25 gg/ml in binding medium which was not supplemented with NaN 3. This incubation was performed for 90 minutes at 37°C in a shaking water bath. Cell suspensions were added directly to the binding assay reaction mixtures. Binding assays. For binding assays, 5-6 × 106 pre-incubated RC-2A cells were incubated for 16 h at 4°C in 0.15 ml binding medium containing 0.2ng 125I-GM-CSF in siliconised glass tubes on a rotating platform. Cell suspensions were cooled on ice, overlayed onto 0.2 ml foetal calf serum at 0°C, and centrifuged at maximum speed for 45 s in a Beckman Microfuge 12. The visible cell pellet was removed by cutting and the radioactivity associated with the pellet was determined in a Packard Auto-Gamma 5650. Nonspecific radioligand binding was determined by incubation in the presence of a 100-fold excess of GM-CSF. Specific binding was calculated by subtracting nonspecific from total binding. To determine that the binding of 1251-GM-CSF to RC-2A cells was specific, 20 ng of rh-IL-3 (100-fold higher concentration than 125I-GM-CSF) was added to the reaction mixture in two experiments.

Recombinant human (rh) GM-CSF produced in COS cells was obtained from Genetics Institute (Cambridge, MA) and was >99% pure as specified by the supplier, rhIL-3 and rh-G-CSF, produced in E. coli and purified to >99%, were supplied by Genetics Institute and Amgen (Thousand Oaks, CA) respectively. In some cell proliferation experiments (as indicated), IL-3 was used in the form of a transfected COS cell supernatant, rh-TNFoi was provided by Genentech (California).

Growth of haemopoietic colonies in vitro The methods have been described elsewhere [4, 11]. Briefly, MNC from bone marrow or cord blood were depleted of monocytes by adherence to plastic and, where indicated, T cells by rosetting with sheep erythrocytes. The resultant cell populations were plated at 1-2 x 10n/plate essentially as described by Metcalf [12] in a final volume of 1.0 ml of 0.3% agar containing HPCM or recombinant cytokine. After 14 days, culture plates were fixed and the agar discs dried onto slides. Colonies were stained with Luxol Fast Blue MBS (eosinophils), o:-naphthyl acetate esterase (macrophages) and chloroacetate esterase (neutrophils). MAbs for addition to the cultures, YB5.B8, 1B4 (antiHLA class I) and 3D3.3 (negative control), were all purified, dialysed extensively to remove azide, and filter sterilized (cf. 4). All are of IgG1 subclass.

lmmunofluorescence assay Binding of YB5.B8 and control antibodies 1B4 and 3D3.3 to RC-2A, KG-1 and HL-60 cells was studied by indirect immunofluorescence as previously described [13] except that the detecting reagent was fluorescein-labelled affinity-purified F(ab')2 sheep antibody to mouse immunoglobulins (Silenus, Australia). Labelled cells were fixed with 1% paraformaldehyde in phosphate-buffered saline and examined using a FACScan flow cytometer (BectonDickinson).

Proliferation assays Proliferation of RC-2A ceils in response to recombinant cytokines was measured by 3H-thymidine incorporation as described elsewhere [10]. Briefly, cells were cultured in quadruplicate at 6 x 10ffwell in fiat-bottomed microtitre plates in a final volume of 220 ~tl containing cytokine and/ or MAb as indicated. Cultures were pulsed with 1 ~tCi/well methyl-3H-thymidine (25 Ci/mmol, Amersham) for the last 16 h of the 4 day culture period.

Binding studies Radioiodination of GM-CSF. GM-CSF was radioiodinated to high specific radioactivity by an iodine mono-

Pre-incubation of RC-2A cells with monoclonal antibodies, In an initial experiment, cells were incubated in

RESULTS

Effect of M A b Y B 5 . B 8 in the CFU-C assay W e previously s h o w e d that inclusion of Y B 5 . B 8 , but not control M A b s , in the culture m e d i u m in the C F U - C assays reduced colony yield w h e n H P C M or P H A - L C M were used as the source of g r o w t h factors [4]. In order to analyse the m e c h a n i s m of inhibition, we have examined the effect of the a n t i b o d y on colony f o r m a t i o n in response to r e c o m b i n a n t h u m a n colony-stimulating factors, G M - C S F , G - C S F and IL3. The o p t i m u m levels of cytokines for this assay were established in preliminary experiments. G M CSF, G - C S F and IL-3 were used at final concentrations, respectively, 30 n g / m l , 100 n g / m l and 1/1000 of a transfected C O S cell s u p e r n a t a n t in the antibody inhibition experiments. T h e s e levels gave near m a x i m u m colony yields. M A b s , Y B 5 . B 8 , 1B4 ( a n t i - H L A class I) and 3D3.3 (negative control) were used at the previously d e t e r m i n e d o p t i m u m concentration of 2 ~tg/ml final [4]. T h e results of two

MAb inhibits GM-CSF induced colony formation

639

TABLE 1. EFFECT OF M A b Y B 5 . B 8 IN THE C F U - C ASSAY WITH RECOMBINANT CYTOKINES AS STIMULUS

No. of colonies/plate (mean 4- S . E . M . ; five replicates) Stimulus HPCM

Experiment 1 2

GM-CSF

1 2

IL-3

1 2

G-CSF

1 2

Added antibody

G

-YB5.B8 -YB5.B8

38.6 23.6 54.4 43.4

-YB5.B8 -YB5.B8

M

--4-

EO

1.1 4.3* 3.6 3.6t

54.6 42.4 58.2 44.4

- 1.2 +-- 1.25 4- 2.2 -+ 1.85

7.0 3.6 9.3 4.2

--+ 1.6 --+ 0.9 --+ 1.2 --- 0.7*

39.4 27.8 56.3 48.4

-+ 4+-+

-YB5.B8 -YB5.B8

3.0 1.4 4.0 4.6

--+ 1.3 4- 0.6 4- 1.3 -+- 1.3

28.6 26.6 37.0 40.0

-+ 2.7 4- 0.7 4- 1.0 --+ 2.2

-YB5.B8 -YB5.B8

18.8 15.8 11.4 8.0

4- 0.7 --- 2.6 - 0.9 -+ 0.9t

8.8 9.8 42.0 40.2

-+ 0.4 --+ 1.2 +-- 2.0 4- 0.5

1.8 2.0* 2.2 2.4t

GM

7.8 9.4 15.8 9.4

+- 1.6 -+ 2.1 --- 0.8 4- 1.2"

5.4 6.0 6.0 3.0 4.6 3.4 2.3 2.6

10.6 8.0 12.2 6.2

0.9 1.6 1.0 0.4~:

111.6 83.0 141.8 102.8

-+ 4-+ +

1.6 6.5* 6.3 6.6*

--- 0.6 --- 0.6 --- 0.6 4- 0.5t

0.2 --- 0.2 -1.3 4- 0.3 0.2 4- 0.2t

52.0 37.4 72.8 56.0

-+ -+ 4-

2.6 2.9* 3.2 2.3*

- 0.8 -+ 1.2 4- 0.7 - 0.3

0.2 - 0.2 0.2 4- 0.2 ---

36.4 31.6 43.3 47.2

4- 2.9 4- 1.9 --- 2.2 4- 2.7

-----

27.6 25.6 53.4 48.2

44-+

-----

4-+ 4-

Total

1.0 2.6 2.3 1.1

B o n e marrow M N C from two different normal donors (Experiments 1 and 2) were depleted of monocytes by adherence to plastic and plated at 2 × 10 4 ( E x p e r i m e n t 1) and 1.3 x 10 4 (Experiment 2) ceils per plate. G r o w t h factor supplements were, as indicated, H P C M , 5% v / v ; G M - C S F , 30 ng/ml; IL-3, 1/1000 dilution of a transfected C O S cell supernatant; GCSF, 100 ng/ml. Purified, sterile Y B 5 . B 8 was added to a final concentration of 2 ~tg/ml at the time of plating. The twotailed Student's t-test [14] was used for statistical comparisons. * p < 0.01; t p < 0.05; $ p < 0.001. TABLE 2. INHIBITION OF COLONY FORMATION IN RESPONSE TO G M - C S F IS INDEPENDENT OF T CELLS

No. of colonies/plate (mean --+ S . E . M . ; five replicates) Treatment

M A b added

G

Monocyte depleted only

none 3D3.3 tr-HLA class 1 YB5.B8

10.6 7.6 8.4 3.6

Monocyte and T cell depleted

none 3D3.3 oc-HLA class 1 YB5.B8

7.0 6.0 8.0 5.4

4-+ 4-

M

EO

GM

Total

0.9 1.0 1.4 0.8

9.4 11.8 11.8 8.4

--- 0.6 -+ 0.6 4- 0.6 -+ 1.0

1.8 1.8 1.8 1.8

-+ 0.2 -+ 0.4 -+ 0.4 4- 0.6

-----

21.8 21.2 22.0 13.8

4- 1.0 4- 1.4 +-- 1.6 -+ 0.7

--- 1.5 -+ 0.4 --- 1.3 4- 0.8

9.4 10.0 9.6 5.4

-+ 1.1 - 0.8 4- 1.5 +-- 1.2

3.6 3.6 4.6 2.0

44-+ 4-

-----

20.0 19.6 22.2 12.8

-+ 44-+

1.0 1.2 0.7 0.9

2.3 1.0 3.1 1.4

B o n e marrow M N C were depleted of monocytes by adherence to plastic. O n e aliquot was retained for plating while another was, in addition, depleted of T cells by E-rosetting as described previously [4]. Twenty-five percent of the cells were recovered in the non-T fraction. Cell suspensions containing T cells were plated at 2 x 10a/plate while T-depleted cells were plated at 8 x 103/plate with 30 ng/ml G M - C S F as stimulus. Purified, sterile YB5.B8 and control IgG1 M A b s were added at a final concentration of 2 ~tg/ml.

experiments using monocyte-depleted bone marrow M N C f r o m t w o n o r m a l d o n o r s a r e s h o w n in T a b l e 1. A s r e p o r t e d by o t h e r s [16, 17], i n d i v i d u a l c y t o k i n e s w e r e r e l a t i v e l y p o o r at s t i m u l a t i n g c o l o n y g r o w t h compared with HPCM, giving fewer colonies and, e s p e c i a l l y in t h e c a s e o f I L - 3 , less c o m p l e t e diff e r e n t i a t i o n . A s p r e v i o u s l y r e p o r t e d [4], Y B 5 . B 8

i n h i b i t e d t h e t o t a l y i e l d o f c o l o n i e s by a p p r o x i m a t e l y 3 0 % w h e n H P C M was u s e d as t h e s o u r c e o f g r o w t h f a c t o r s . T h e y i e l d o f all c o l o n y t y p e s w a s r e d u c e d . When recombinant cytokines were used significant r e d u c t i o n s in t h e t o t a l n u m b e r o f c o l o n i e s o c c u r r e d o n l y w i t h G M - C S F as s t i m u l u s . A s w i t h H P C M , all types of colonies were affected. No significant effects

640

L . K . ASHMAN et al.

on colony growth were observed when IL-3 or GCSF were used as stimulus, except in the case of the yield of G colonies with G-CSF in one experiment (p < 0 . 5 ) . Control antibodies had no significant effect on colony yields (not shown). In order to reduce the likelihood that the action of YB5.B8 on colony-forming cells involved accessory cells, bone marrow MNC were depleted of T cells (by E-rosetting) as well as monocytes. As shown in Table 2, removal of T cells did not affect the inhibitory action of YB5.B8 on colony formation with GMCSF as stimulus. The extent of inhibition was 36% in the case ofT-depleted populations, compared with 37% for T-containing populations.

A

\\

:

•

O z

Ill O

II1 ne

c

Binding of MAb YB5.B8 to human myeloid leukaemia cell lines Because the CFU-C assay uses impure cell populations and large amounts of cytokines, and is difficult to score, we sought an alternative experimental system for further studies of the mode of action of MAb YB5.B8 on the response of haemopoietic cells to GM-CSF. In our original report [1], we were unable to detect binding of YB5.B8 to any cell lines, however, as can be seen from Fig. 1A, high sensitivity flow cytometry analysis shows that the antibody does bind to the myelomonocytic leukaemia cell line RC2A. Extremely weak, but reproducible, binding to the KG-1 myeloblastic cell line (Fig. 1B), but no significant binding to the HL-60 promyelocytic cell line (Fig. 1C), were also observed.

Enhancement of the proliferation of cell lines in response to recombinant cytokines We previously reported that the proliferation of RC-2A cells was enhanced by rh-GM-CSF or rh-GCSF indicating that the cells have receptors for these cytokines [10]. Similarly, enhanced proliferation of KG-1 cells in response to GM-CSF and IL-3 has been reported [18, 19]. In a preliminary experiment, it was found that the optimum assay time for RC-2A cells and KG-1 cells responding to GM-CSF and IL-3 was 4 days. The enhancement of proliferation was 87% and 65% for RC-2A in the presence of GM-CSF and IL-3 respectively. KG-1 cells did not respond to IL3 and displayed only 10% enhancement of proliferation in the presence of GM-CSF. Other workers have reported variability in the binding of cytokines to KG-1 sublines [20]. RC-2A cells were used for all further studies. Dose response curves were constructed for GM-CSF, G-CSF and IL-3 in the RC2A proliferation assay (Fig. 2). Maximal stimulation was observed at 5-10 ng/ml for all cytokines. The extent of proliferative enhancement (approx. 15%) obtained in this experiment with purified IL-3

FL 1

FLUORESCENCE FIG. 1. Binding of MAb YB5.B8 to human myeloid leukaemia cell lines RC-2A, KG-1 and HL-60. Binding of MAb YB5.B8 ( . . . . . ) and control IgG1 MAbs 3D3.3 (negative control) ( ); and 1B4 (anti-HLA class I; positive control) ( . . . . . ) was assessed by indirect immunofluorescence and flow cytometry. The figure shows frequency versus fluorescence intensity for RC-2A (A), KG1 (B), and HL-60 (C) cells.

(produced in E. coli) was lower than that obtained with the transfected COS cell supernatant (cf. above).

Effect of MAb YB5.B8 on the proliferation of RC2A cells in response to cytokines YB5.B8 and control antibodies 1B4 and 3D3.3 were included at a final concentration of 2 ~tg/ml in RC-2A cultures containing doubling dilutions of GMCSF, IL-3 or G-CSF over the range 0-20 ng/ml. In no case was a significant effect on 3H-thymidine incorporation by RC-2A cells observed. The data for GM-CSF are shown in Fig. 3. Similar experiments with blast cells from a patient with C G L which proliferated in response to IL-3, GM-CSF and to a lesser extent G-CSF [cf. 21] also failed to demonstrate an effect of YB5.B8.

Binding of GM-CSF to RC2A cells Experiments performed for 16 h at 4°C showed that 125I-GM-CSF bound to RC-2A cells. Scatchard analysis indicated a single class of binding sites, K d

MAb inhibits GM-CSFinducedcolonyformation 80

641

to RC-2A cells was specific, competititive binding experiments with GM-CSF, IL-3 and TNFa~ (all 6.7 nM) were performed at 4°C in the presence of NAN3. The binding of t25I-GM-CSF was completely inhibited by GM-CSF and partially inhibited (by 70 - 6.4.%) by IL-3 while TNFo: had no effect.

IL-3 60

40

20"

DISCUSSION

860"}

=

o

-20"

0..

G-CSF

ILl Z

60

i,

40

U,I

= I,Z gl

.20

~1

80

60

40

20

0

001

004

0,1

0.2

0.3

0.S

1.3

2.5

S

10

20

CYTOKINE CONCENTRATION (ng/ml)

FIG. 2. Proliferative response of RC-2A cells to recombinant human GM-CSF, G-CSF and IL-3. Proliferation of RC-2A cells was measured by 3H-thymidine incorporation as described in Materials and Methods. Values shown are means -+ S.E.M. derived from four replicates. The basal level of 3H-thymidine incorporation was 4.0 x 104 cpm. 33.3 pM, and an average of 50 sites/cell (Fig. 4). In a repeat experiment the values obtained were Kd = 31.2 pM, 85 sites/cell. Monoclonal antibodies YB5.B8, 1B4 and 3D3.3 did not significantly reduce the binding of 125I-GM-CSF under experimental conditions designed to detect direct competition for GMCSF receptors (i.e. 4°C in the presence of NAN3) (Fig. 5A). To determine whether YB5.B8 or control MAbs could down-modulate GM-CSF receptors [22], RC-2A cells were pre-incubated with antibodies, at 37°C in the absence of NAN3. In this experiment, the binding of radioligand was identical with and without pre-incubation with YB5.B8 or control MAbs (Fig. 5B).

Specificity of GM-CSF binding to RC2A cells To determine that the binding of 125I-GM-CSF

Because of its low level of expression, its unusual and highly restricted distribution, and its association with poor prognosis in ANNL, we postulated that the antigen identified by MAb YB5.B8 might be a receptor for a haemopoietic growth factor. This was supported by the finding that the antibody inhibited the growth of colonies in the CFU-C assay when complex growth factor supplements were used [4]. In order to clarify the role of the YB5.B8 antigen in haemopoiesis, the effect of the antibody on the growth of colonies in response to individual colonystimulating factors in the CFU-C assay was examined. MAb YB5.B8 reduced the yield of colonies derived from bone marrow progenitor cells in response to GM-CSF, but not IL-3 or G-CSF, in the 14 day CFU-C assay. This result suggested that YB5.B8 might bind to the GM-CSF receptor. The molecular weight of the antigen is 145-150,000 [2] which is similar to that reported for the murine GMCSF receptor (130 or 180,000, depending on cell type) [23]. However, further studies using myeloid leukaemic cells and cell lines failed to support the hypothesis that MAb YB5.B8 binds to the GM-CSF receptor. The antibody binds to cells of the myelomonocytic leukaemia line RC-2A, but failed to influence the enhancement of their proliferation brought about by GM-CSF, IL-3 or G-CSF. Similarly, the antibody did not affect the proliferation of CGL cells in response to these factors. Direct binding studies using 125IGM-CSF indicated that RC-2A cells have approximately 50-100 receptors for this cytokine. The extent of binding of YB5.B8 to RC-2A cells was assessed by indirect immunofluorescence. Although not quantifiable, the results indicate that there are of the order of a thousand binding sites per cell. Furthermore, MAb YB5.B8 did not influence the binding of 125I-GM-GSF to RC-2A cells with pre-incubation at 37°C or 4°C. Interestingly, the binding of GM-CSF to RC-2A cells was inhibited by IL-3. Recently, similar observations have been reported for other primary human cell populations and cell lines [20, 24, 25]. The exact mechanism for the partial inhibition of GM-CSF binding by IL-3 is not clear, although it suggests that these two cytokines interact at the surface of RC-2A cells.

642

L. K. ASHMANet A

al.

8"

io. 8z

I,-

~o

0

0.04

0.1

0.2

0.3

0.6

1.3

2.5

5

10

20

GM-CSF (ng/ml)

FIG. 3. MAb YB5.B8 does not inhibit the enhancement of proliferation of RC-2A cells in response to GM-CSF. RC2A cells were cultured in 4-fold replicate with the indicated concentrations of GM-CSF with or without purified, sterile MAb: • none; [] YB5.B8; [] 1B4; B 3D3.3. Proliferation measured by 3H-thymidine incorporation during the last 16 h of the 4 day culture period. The simplest explanation for the inhibitory effect of YB5.B8 on the response to GM-CSF in the CFUC assay but not on the interaction of GM-CSF with RC-2A cells is that the former is mediated by an accessory cell. However, depletion of T lymphocytes as well as monocytes from the bone marrow cell population did not influence the inhibition of the response to GM-CSF observed in this assay. Accessory cells are believed to be necessary for optimal responses to individual CSFs [17]. The relatively poor colony-stimulating activity which we observed with individual recombinant cytokines compared with H P C M indicates minimal input from accessory cells in our system. The lack of accessory cells may have been due in part to the use of cryopreserved bone marrow specimens. Strife and coworkers [16] used cryopreservation as part of their purification procedure to obtain progenitors free from accessory cells. The expression of the YB5.B8 antigen by only about 2% of bone marrow MNC, including progenitors which grow in the CFU-C assay [4], also argues against the possibility that the effects of the antibody are mediated by accessory cells. An alternative explanation is that the YB5.B8 antibody acts as an agonist or an antagonist of another factor which influences the growth of normal bone marrow colony-forming cells, but not RC-2A cells or C G L cells. M-CSF (also known as CSF-1) is known to strongly synergise with GM-CSF in the CFU-C assay [26], and the M-CSF receptor has a molecular weight of 150,000, similar to that of the YB5.B8 antigen [27]. Recently, Sherr and coworkers [28] reported the use of transfected rat cells expressing human M-CSF receptors to prepare rat MAbs to

A E D. D Z

800"

A

600'

O U. (n

400'

¢3

200'

? i

~ .

i 0.05

. 0.10

i 0.15

[ 1251-rhGM-CSFJ (nM)

0.100 -

UJ IM If: I/. 0 Z O

B

0.075 -

0.050

-

0,025 -

i

i

1

2

[ 12Sl-rhGM-CSF BOUND] (pM)

FIG. 4. Binding of 125I-GM-CSF to RC-2A cells. A: 1251GM-CSF was incubated with RC-2A cells for 16 h at 4°C in the presence of NaN3 as described in Materials and Methods. Non-specific binding was determined in the presence of 100-fold excess of unlabelled GM-CSF. B: Scatchard analysis of the specific binding data shown in A.

MAb inhibits GM-CSF induced colony formation

643

120

Z

_o Fw a.

100

80

o

60

Z w 0

40

w

20

0

0

--INIL

GM-CSF

YB5.B8

1B4

303

NIL

GM-CSF Y85.B8

1B4

3D3

FIG. 5. MAb YB5.B8 does not inhibit the binding of rhGM-CSF to RC-2A ceils. RC-2A cells were pre-incubated at 4°C (A) or 37°C (B) with YB5.B8 or control MAbs 3D3.3 or 1B4 (anti-HLA class I) at 2 (11) or 0.2 ([]) ~tg/ml (A) or 25 ~tg/ml (B) prior to measurement of the binding of 125I-labelled GM-CSF (67pM) at 4°C. Total specific 125I-GM-CSF binding was 610 -+ 33.1 cpm (A) and 402 -+ 14.8 cpm (B).

the receptor. Transfected cell lines expressing high levels of the M-CSF receptor, detectable by binding of rat anti-M-CSF receptor M A b s , nevertheless failed to bind M A b YBS.B8 (R. A. A s h m u n and L. K. A s h m a n , unpublished results). Thus, the YB5.B8 antigen is distinct from the M-CSF receptor. Interleukin 4 (IL-4) was recently shown to inhibit macrophage colony growth from h u m a n bone m a r r o w in response to G M - C S F [29]. As for the inhibition of C F U growth by M A b YB5.B8, no requirement for accessory cells could be demonstrated. The h u m a n IL-4 receptor has a molecular weight of 139,000 [30] which is similar to the YB5.B8 antigen. It will be important to test M A b YB5.B8 for effects on IL-4 dependent functions and the binding of IL-4 to its receptor. In conclusion, M A b YBS.B8, which binds to h u m a n haemopoietic progenitor cells, inhibited the growth of normal bone m a r r o w colony forming cells in response to G M - C S F , but not G - C S F or IL-3, in vitro. H o w e v e r , the antibody did not influence the binding of G M - C S F to myeloid leukaemia cells or affect their proliferation in response to this factor. Cross-competition experiments with IL-3 on different cell types indicate that there m a y be m o r e than one type of receptor for G M - C S F [13, 25]. H o w e v e r , it seems unlikely that the antibody binds directly to a G M - C S F receptor. Thus, the mechanism of its inhibitory action in the C F U - C assay remains unclear but m a y involve triggering of a receptor pathway which indirectly modifies the response of normal progenitors to G M - C S F . Acknowledgements--This work was supported by grants

from the Rotary Peter Nelson Leukaemia Research Fund and the National Health and Medical Research Council of Australia. We thank Drs C. A. Juttner and L. B. To for providing normal bone-marrow specimens and Dr A. F. Lopez for valuable discussions.

REFERENCES 1. Gadd S. J. & Ashman L. K. (1985) A murine monoclonal antibody specific for a cell-surface antigen expressed by a subgroup of human myeloid leukaemias. Leukemia Res. 9, 1329. 2. Ashman L. K., Gadd S. J., Mayrhofer G., Spargo, L. D. J. & Cole S. R. (1987) A murine monoclonal antibody to an acute myeloid leukaemia-associated cell surface antigen identifies tissue mast cells. In Leucocyte Typing II1. White Cell Differentiation Antigens, (McMichael, A., Ed.), p. 726. Oxford University Press. 3. Mayrhofer G., Gadd S. J., Spargo L. D. J. & Ashman L. K. (1987). Specificity of a mouse monoclonal antibody raised against acute myeloid leukaemia cells for mast cells in human mucosal and connective tissues. Immun. Cell Biol. 65, 241. 4. Cambareri A. C., Ashman L. K., Cole S. R. & Lyons A. B. (1988) A monoclonal antibody to a human mast cell/myeloid leukaemia-specific antigen binds to normal haemopoietic progenitor cells and inhibits colony formation in vitro. Leukemia Res. 12, 929. 5. Ashman L. K., Roberts M. M., Gadd S. J., Cooper S. J. & Juttner C. A. (1988) Expression of a 150 kDa cell surface antigen identified with monoclonal antibody YB5.B8 is associated with poor prognosis in acute nonlymphoblastic leukemia. Leukemia Res. 12, 923. 6. Slamon D., Clark G. M., Wong S. G., Levin W. J., Ullrich A. & McGuire W. L. (1987) Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 235, 177. 7. Slamon D. J., Godolphin W., Jones L. A., Holt J. A., Wong S. G., Keith D. E., Levine W. J., Stuart S. G., Udove J., Ullrich A. & Press M. F. (1989) Studies of

644

L.K. ASHMANet al.

the HER-2/neu proto-oncogene in human breast and ovarian cancer. Science 244, 707. 8. O'Keefe D. E. & Ashman L. K. (1982) Peanut agglutinin: a marker for normal and leukaemic cells of the monocyte lineage. Clin. exp. lmmun, 48, 329. 9. Bradley T. R., Pilkington G., Garson M., Hodgson G. S. & Kraft N. (1982) Cell lines derived from a human myelomonocytic leukaemia. Br. J. Haemat. 51, 595. 10. Lyons A. B. & Ashman L. K. (1988) The effect of recombinant cytokines on the proliferative potential and phenotype of cells of the human myelomonocytic leukaemia line, RC-2A. Leukemia Res. 12, 659. 11. Lyons A. B., Cooper S. J., Cole S. R. & Ashman L. K. (1988). Human myeloid differentiation antigens identified by monoclonal antibodies to the myelomonocytic leukaemia cell line RC-2A. Pathology 20, 137. 12. Metcalf D. (1984) Clonal Culture o f Hemopoietic Cells: Techniques and Applications. Elsevier, Amsterdam. 13. Gadd S. J. & Ashman L. K. (1983) Binding of mouse monoclonal antibodies to human leukaemic cells via the Fc receptor: a possible source of 'false positive" reactions in specificity screening. Clin. exp. lmmun. 54, 811. 14. Contreras M. A., Bale W. F. & Spar I. L. (1983) Iodine monochloride (ICI) iodination techniques. Meth. Enzym. 92, 277. 15. Roscoe J. T. (1975) Fundamental Research Statistics for the Behavioural Sciences (2nd edition), p. 217. Holt, Rinehart & Winston, New York. 16. Strife A., Lambek C., Wisniewski D,, Gulati S., Gasson J. C., Golde D. W., Welte K., Gabrilove J. L. & Clarkson B. (1987) Activities of four purified growth factors on highly enriched human hematopoietic progenitor cells. Blood 69, 1508. 17. Bot F. J., Dorssers L., Wagemaker G. & L6wenberg B. (1988) Stimulating spectrum of multi-CSF (IL-3) on human marrow precursors: importance of accessory cells. Blood 71, 1609. 18. Wong G. G., Witek J. S., Temple P. A., Wilkins K. M., Leary A. C., Luxenberg D. P., Jones S. S., Brown E. L., Kay R. M., Orr E. C., Shoemaker C., Golde D. W., Kaufman R. J., Hewick R. M., Wang E. A. & Clark S. C. (1985)Human GM-CSF: molecular cloning of the complementary DNA and purification of the natural and recombinant proteins. Science 228, 810. 19. Yang Y-C., Ciarletta A. B., Temple P. A., Chung M. P., Kovacic S., Witek-Gianotti J. S., Leary A. C., Kriz R., Donahue R. E., Wong G. G. & Clark S. C. (1986) Human IL-3 (multi-CSF) identification by expression cloning of a novel hematopoietic growth factor related to murine IL-3. Cell 47, 3. 20. Park L. S., Friend D., Price V., Anderson D., Singer

21.

22. 23.

24.

25.

26.

27.

28.

29.

30.

J., Prickett K. S. & Urdal D. L. (1989) Heterogeneity in human Interleukin-3 receptors. A subclass that binds human granulocyte/macrophage colony stimulating factor. J. biol. Chem. 264, 5420. Griffin J. D., Sullivan R., Beveridge R. P., Larcom P. & Schlossman S. F. (1984) Induction of proliferation of purified human myeloid progenitor cells: a rapid assay for granulocyte colony-stimulating factors. Blood 63, 904. Walker F., Nicola N. A., Metcalf D. & Burgess A. W. (1985) Hierarchical down-modulation of hemopoietic growth factor receptors. Cell 43, 629. Park L. S., Friend D., Gillis S. & Urdal D. L. (1986) Characterization of the cell surface receptor for granulocyte-macrophage colony-stimulating factor. J. biol. Chem. 261, 4177. Lopez A. F., Eglinton J. M., Gillis D., Park L. S., Clark S. & Vadas M. A. (1989) Reciprocal inhibition of binding between interleukin 3 and granulocytemacrophage colony-stimulating factor to human eosinophils. Proc. natn. Acad. Sci. U.S.A. 86, 7022. Elliott M. J., Vadas M. A., Eglinton J. M., Park L. S., To L. B., Cleland L. G., Clark S. C. & Lopez A. F. (1989) Recombinant human interleukin 3 (IL-3) and granulocyte-macrophage colony stimulating factor (GM-CSF) show common biological effects and binding characteristics on human monocytes. Blood 74, 2349. Caracciolo D., Shirsat N., Wong G. G., Lange B., Clark S. C. & Rovera G. (1987) Recombinant human macrophage colony-stimulating factor (M-CSF) requires subliminal concentrations of granulocyte/ macrophage (GM)-CSF for optimal stimulation of human macrophage colony formation in vitro. J. exp. Med. 166, 1851. Rettenmier C. W., Sacca R., Furman W. L., Roussel M. F., Holt J. T., Nienhuis A. W., Stanley E. R. & Sherr C. J. (1986). Expression of the human c-fins proto-oncogene product (colony-stimulating factor-1 receptor) on peripheral blood mononuclear cells and choriocarcinoma cell lines. J. clin. Invest. 77, 1740. Sherr R. J., Ashmun R. A., Downing J. R., Ohtsuka M., Quan S. G., Golde D. W. & Roussel M. F. (1989) Inhibition of colony-stimualting factor-1 activity by monoclonal antibodies to the human CSF-1 receptor. Blood 73, 1786-1793. Jansen J. H., Wientjens G-J. H. M., Fibbe W. E., Willemze R. & Kluin-Nelemans H. C. (1989) Inhibition of human macrophage colony formation by interleukin 4. J. exp. Med. 170, 577. Park L. S., Friend D., Sassenfeld H. M. & Urdal D. L. (1987) Characterization of the human B cell stimulatory factor 1 receptor. J. exp. Med. 166, 476.