AU5497399A - Monoclonal antibody inhibitor of gm-csf, il-3, il-5 and other cytokines, and uses thereof - Google Patents

Description

WO 00/09561 PCT/AU99/00659 1 MONOCLONAL ANTIBODY INHIBITOR OF GM-CSF, IL-3, IL-5 AND OTHER CYTOKINES, AND USES THEREOF FIELD OF THE INVENTION 5 This invention relates to a method of isolating monoclonal antibody inhibitors and reagents derived therefrom and other inhibitors of cytokine binding including monoclonal antibodies and reagents derived therefrom and small molecules capable of inhibiting binding of GM-CSF, IL-3 and IL-5 to the common beta receptor subunit. 10 INTRODUCTION Human interleukin (IL)-5, IL-3 and granulocyte-macrophage colony-stimulating factor (GM-CSF) are cytokines involved in hemopoiesis and inflammation (Metcalf; 1986). All three cytokines stimulate eosinophil production, function and 15 survival (Metcalf; 1986) and therefore have the ability to influence inflammatory diseases such as asthma, atopic dermatitis and allergic rhinitis where the eosinophil plays a major effector role. IL-5, being the eosinophil specific cytokine, has received most of the initial attention with IL-5 mRNA and protein levels noted to be elevated in lung tissue and bronchoalveolar lavage (BAL) fluid from 20 symptomatic asthma patients (Fukuda et al 1994). Correlation between IL-5 levels and allergen challenge and disease activity have also been seen (Sur et al, 1996). It is becoming apparent, however, that not only IL-5 but also GM-CSF and IL-3 play a role in eosinophil production and activation in asthma as there is evidence of both GM-CSF and IL-3 being synthesized at sites of allergic inflammation (Bagley 25 et al, 1997b; Allen et al 1997). It is possible that expression of these cytokines contributes to the total number of infiltrating eosinophils and the degree of eosinophil activation. Alternatively, they may be responsible for different phases of eosinophil infiltration. Recent kinetic data from patients undergoing antigen challenge showed that IL-5 levels increased between days 2-7 post challenge, 30 whilst GM-CSF peaked at day 2, and remained elevated throughout day 16. Furthermore, GM-CSF detection extended beyond the site of allergen challenge. IL-5, GM-CSF and IL-3 stimulate eosinophils and other normal and cancer cells by binding to cell surface receptors that comprise a ligand-specific a chain and a 3 35 chain which is shared by the three receptors (3c) (Bagley et al 1997a). Binding to each receptor a chain is the initial step in receptor activation, however, engagement of either a chain alone is not sufficient for activation to occur. Recruitment of 3 c WO 00/09561 PCT/AU99/00659 2 by each ligand: cx chain complex follows, a step that has two major functional consequences: firstly, it allows the binding of IL-5, GM-CSF and IL-3 to become essentially irreversible; and secondly, it leads to full receptor activation (Bagley et al 1997a). Since 3c is the major signalling component of these receptors its 5 engagement leads to the activation of JAK-2, STAT-5 and other signalling molecules culminating in the full plethora of cellular activities commonly associated with either IL-5, GM-CSF and IL-3 stimulation such as eosinophil adherence, priming for degranulation and cytotoxicity, and prolongation of viability (Bates et al, 1996). 10 In order to block or antagonize the activity of eosinophil-activating cytokines in vivo three major approaches are being tried. One of them utilizes antibodies to the implicated cytokines. For example, antibodies to IL-5 are being used in an animal model of allergen-induced asthma and have shown to have a relatively long lasting 15 effect in preventing eosinophil influx into the airways and bronchial hyperresponsiveness (Mauser et al, 1995). A second approach relies on IL-5 or GM-CSF mutants which can bind to the respective cc chains with wild type affinity but which have lost or shown reduced ability to interact with human Pc. IL-5 mutants such as E13Q, E13K and E13R, and the human GM-CSF mutant E21R 20 directly antagonize the functional activation of eosinophils by IL-5 or GM-CSF respectively (Tavernier et al 1995; McKinnon et al 1997; Hercus et al 1994b). However, at least in the case of E13K, eosinophil survival is not antagonized and in fact this mutant is able to support eosinophil survival (McKinnon et al 1997). A third approach involves the use of soluble receptor c chains which can sequester 25 circulating cytokines. However, this carries the risk of a cytokine:receptor c chain complex potentially interacting with surface-expressed P3c and triggering receptor activation. The common theme amongst these approaches is that they tackle a single receptor system involving either IL-5, GM-CSF or IL-3 leaving the other two eosinophil-acting cytokines unaffected. Although the concomitant 30 administration of IL-5 and GM-CSF antagonists may be considered, this may be clinically impracticable. An alternative approach to blocking eosinophil-activating cytokines involves targeting the common 3 chain of their receptors. Although 1 3 c does not directly 35 bind IL-5, GM-CSF or IL-3 alone, it does bind to these cytokines complexed to the appropriate receptor ca chain. Lopez et al in WO 97/28190, which is incorporated herein by reference in its entirety, have identified the major binding WO 00/09561 PCT/AU99/00659 3 sites of, 3c for the IL-5:IL-5R(x, GM- CSF:GM-CSFR(x and IL-3:IL-3Ra complexes. Significantly, these sites are utilized by all three complexes and comprise the predicted B'-C' loop and F'-G' loop in 3c. Thus targeting Pc is not only desirable but also feasible, with the added potential to allow the simultaneous 5 inhibition of IL-5, GM-CSF and IL-3 action by a single agent. These workers have shown that certain mutants in the B'-C' and the F'-G' loop fail to bind IL-5, GM-CSF and IL-3. SUMMARY OF THE INVENTION 10 The present invention results from the isolation of a monoclonal antibody (BION 1) raised against the membrane proximal domain (domain 4) of P3c which is able to block the production and activation of human eosinophils stimulated by IL-5, GM-CSF or IL-3 and blocks the growth of leukaemic cell lines. This MoAb was able to block the high affinity binding of all three cytokines to eosinophils by 15 binding to residues in the predicted B'-C' and F'-G' loops of P3c, and prevented receptor dimerization and 3 c phosphorylation. It was found that raising an antibody capable of blocking the binding of all three cytokines was possible by screening monoclonal antibody-expressing hybridoma cell lines arising from immunising mice with cells expressing only domain 4 of Pc and lacking domains 1 20 to 3 and expressing domain 4 and the transmembrane and cytoplasmic regions. Additionally this finding is likely to have implications for other members of the cytokine receptor superfamily some of which are shared subunits in a given subfamily (that is they bind several cytokines), and some which are ligand specific 25 and bind to only one cytokine. The receptor (x-chains for GM-CSF, IL-3 and IL-5 and Pc belong to the rapidly expanding cytokine receptor superfamily. Within this superfamily several sub-families are now emerging that are characterized by the sharing of a communal receptor subunit by multiple ligands: gpl30 acts as an affinity converter and signal transducer for IL-6 (Hibi et al., 1990; Taga et al., 30 1992), IL-11I (Hilton et al., 1994), oncostatin M (Liu et al., 1992), ciliary neurotrophic factor, leukaemia inhibitory factor (LIF) (Ip et al., 1992) and cardiotrophin-1(Pennica et al., 1995); the LIF receptor (LIFR) also binds ciliary neurotrophic factor (Davis et al., 1993), cardiotrophin-1 (Pennica et al., 1995) and oncostatin M in addition to LIF (Gearing et al., 1994); IL-2R P3 supports affinity 35 conversion and signalling of IL-2 and IL-15 (Giri et al., 1994); IL-2R 7 chain affinity converts IL-2 (Takeshita et al., 1992), IL-4 (Russell et al., 1993), IL-7 (Noguchi et al., 1993), IL-9 (Kimura et al., 1995) and IL-15 (Giri et al., 1994); WO 00/09561 PCT/AU99/00659 4 evidence also suggests that IL-4 and IL-13 share a receptor component (Zurawski et al., 1993) and this subunit has recently been cloned (Hilton et al., 1996). It is not known which residues in gp 130, LIFR and IL-2R p and y chains are important for ligand binding or indeed whether different ligands share or have unique sets of 5 binding determinants on these communal receptor subunits. Because these common subunits are vital for transducing signals by several ligands, the possibility arises that interfering with the ability of these common subunits to bind ligand or to form homodimers may affect the action of more than one ligand. 10 Clear similarities in structure between Ic and other cytokine receptors have been recognised and similarities in at least part of the binding site, the F'-G' loop, have been identified in Lopez et al in WO 97/28190. Accordingly it is an expectation that the means employed by the inventors to obtain a monoclonal antibody that inhibits binding of IL-3, GM-CSF and IL-5 will also lead to the isolation of 15 monoclonal antibodies that inhibit binding of other cytokines to their respective receptors. In a broad form of a first aspect the invention could be said to reside in a method of isolating a monoclonal antibody capable of inhibiting any one of IL-3, GM-CSF 20 and IL-5 binding to the common receptor Pc, or a monoclonal antibody capable of inhibiting a cytokines binding to a receptor analogous to c, said method comprising the step of immunising an animal with a cytokine receptor or portion of a cytokine receptor containing the critical binding site which portion might include the extracellular domain 4 or analogous domain in the analogous common receptor 25 or part thereof, isolating antibody producing cells from said animal and fusing antibody producing cells with a myeloma cell line, screening for a cell line that produces an antibody of the desired type. The immunisation may involve introducing a cDNA clone of a portion of or all of 30 the common receptor including the extracellular domain 4 or analogous domain in the analogous common receptor or part thereof, into a cell and proliferating said cells to form a recombinant cell line, inoculating an animal with said recombinant cell line, isolating antibody producing cells from said animal and fusing the antibody producing cell line with a myeloma cell line to form a hybridoma cell line, 35 screening for a hybridoma cell line that produces an antibody that binds to the recombinant cell line but not to the parent, and then testing for inhibition against all WO 00/09561 PCT/AU99/00659 5 three cytokines. In one form the cell into which the cDNA clone is introduced is mammalian and one commonly used mammalian cell line is a COS cell. The cDNA may encode a full or partial portion of domain 4 when it is in a 5 configuration where the F'-G' loop and/or the B'-C' loop is in its native shape. The data below show that cDNA encoding substantially only domain 4 of the extracellular portion of Pc as well as the transmembrane and the intracellular portions maintains these sites in a sufficiently integral conformation so that an antibody raised thereagainst will give the inhibition sought. It is postulated that the 10 same will be the case for analogous receptors for the cytokine superfamily. This method should be distinguished from immunising with the whole receptor since the extracellular domain 4 is likely to be covered or masked by other domains in the whole receptor. 15 Pc has two repeats of the cytokine receptor module (CRM), each of which has two discrete folding domains (CRDs), so that in total Pc has 4 domains hence named domains 1 to 4 (P31 to 4). It is postulated that domain 2 of any CRM may be an equivalent of domain 4 and therefore domain 2 may be used in the immunisation. 20 In addition the domain 4 of Pc or equivalent domain in other cytokine receptors may be expressed in isolation in a microbial host such as Escherichia coli and used to immunise animals for developing monoclonal antibodies. The analogous receptor may be any one of the cytokine superfamily receptors but 25 not limited to the group comprising Pc, LIFR, gpl30, IL-2RP, IL-4R/IL-13R, IL 2Ry, IL-3Ra, EPOR, TPOR and OBR. It will be understood that in one specific form of this aspect of the invention the method is used to isolate a monoclonal antibody that inhibits cytokine binding to a 30 common receptor subunit. The common receptor is envisaged to be selected from the group of receptors acting for more than one cytokine including but not limited to gpl30, LIFR, IL2RO/IL2Rct, and IL-4R/IL-13R in addition to Pc. It will also be understood that the invention encompasses monoclonal antibodies or 35 fragments thereof produced as a result of this first form of the invention.

WO 00/09561 PCT/AU99/00659 6 In a broad form of a second aspect the invention could be said to reside in a monoclonal antibody, or fragments thereof capable of inhibiting the binding of the three cytokines IL-3, GM-CSF and IL-5 to the 3c receptor. 5 The degree of inhibition may range from complete inhibition to moderate inhibition, which inhibition will of course be dependent on the amount of monoclonal antibody or fragments thereof added to inhibit and the relative affinity of the antibody or fragment thereof to the 0c. 10 The extent of inhibition of respective ones of the three cytokines is not necessarily identical and may vary, so the different cytokines may be inhibited from binding to different degrees. The antibody fragments may be larger portions such as Fab fragments or much 15 smaller fragments of the variable region. These fragments may be used as separate molecules or alternatively may form part of a recombinant molecule which is then used for therapeutic purposes. Thus for example the monoclonal antibody may be "humanised" by recombining nucleic acid encoding the variable region of the monoclonal antibody with nucleic acid encoding non-variable regions of human 20 origin in an appropriate expression vector. The inhibition preferably leads to blocking of at least one function of all three cytokines. One of the benefits that is proposed to be derived from these antibodies or antibody fragments is their use in modifying cells stimulated by one of the three 25 cytokines, and more in one specific form modifying the activity of the three cytokines is proposed to impact greatly on eosinophil function. Therefore preferably the activity leads to inhibition of stimulation of effector cell activation and where the antibody or fragment thereof is to be used for treatment of asthma leads most preferably to inhibition of IL-5, IL-3 & GM-CSF mediated eosinophil 30 activation. It will be understood however that cells other than eosinophils are also the effectors of adverse conditions in humans and animals as a result of stimulation by these cytokines and inhibition of such stimulation is also contemplated by this invention. These include cells that express either one or all of GM-CSF, IL-3 and IL-5 receptors, the stimulation of which leads to pathology. Examples of these are 35 leukaemic cells, endothelial cells, breast cancer cells, prostate cancer cells, small cell lung carcinoma cells, colon cancer cells, macrophages in chronic inflammation WO 00/09561 PCT/AU99/00659 7 such as rheumatoid arthritis, dendritic cells for immunosuppression and neutrophils in inflammation. Thus in one form the invention may be said to reside in an inhibitor of leukaemic 5 cell growth wherein the inhibitor is capable of inhibiting the binding of one or all of IL-3, GM-CSF and IL-5 to the Pc receptor. The inhibitor may be BION- 1 or an agent capable of inhibiting BION-1 binding with ic. A number of different facets of eosinophil function might be modified so that in 10 one form IL-5, IL-3 & GM-CSF mediated eosinophil survival is inhibited or blocked. In a second form IL-5, IL-3 and GM-CSF mediated eosinophil activation is inhibited or blocked. In one form of this second aspect of the invention the monoclonal antibody or 15 fragment thereof binds to at least the F-G' loop of domain 4 of the Pc subunit. In an alternative form the monoclonal antibody or fragment thereof binds to at least the B'-C' loop of domain 4 of the Oc subunit but this alternative form is not limited to monoclonal antibodies or fragments thereof that only bind to the F-G' loop but 20 includes monoclonal antibodies or fragments thereof that perhaps binds to both the F'-G' as well as the B'-C' loop of domain 4 of the 3c. It is thought that the monoclonal antibody isolated by the inventors inhibits dimerisation of the common receptor units and thus the invention might encompass 25 an antibody or fragments thereof of the second aspect of the invention that inhibit Pc receptor dimerisation. In one very specific form the monoclonal antibody is the antibody produced by the hybridoma cell line BION-1 (ATCC HB-12525). 30 In a broad form of a third aspect the invention could be said to reside in a hybridoma cell line capable of producing a monoclonal antibody of any form of the first or second aspect of the invention. 35 In one specific form of the third aspect of the invention the hybridoma cell line is BION-1 (ATCC HB-12525).

WO 00/09561 PCT/AU99/00659 8 Since GM-CSF, IL-3 and IL-5 need to bind their respective a chains before being able to interact with Pc, at present most screening for new inhibitors utilise cell based assays where both, x and Pc receptor units are co-expressed. Solid phase assays rely on inhibition of GM-CSF, IL-3 or IL-5 to their respective x chain only 5 since these cytokines cannot bind to Pc alone. Since BION-1, unlike these three cytokines, can directly bind to 3c we propose that it can be used as a novel solid phase screening assay. Any compound that binds the appropriate site which is likely to inhibit all three cytokines will also inhibit the binding of BION- 1. Additionally once further inhibitory compounds are uncovered these could be used 10 in the place of BION-1 in that screening process. This therefore facilitates the screening of larger number of candidate inhibitor compounds. In a broad form of a fourth aspect therefore the invention could be said to reside in a method of screening peptides, oligonucleotides and other small molecules for 15 their capacity to competitively inhibit the binding of BION-1 or the binding of an agent capable of inhibiting BION-1 binding, to the Pc subunit. Generally the screening assay involves contacting BION-1 or fragment thereof with the Pc subunit or fragment thereof as well as a candidate inhibitory 20 compound, and measuring the degree of binding. A reporting means is preferably provided to facilitate the detection of binding of BION1 or fragment thereof with Pc subunit or fragment thereof. Thus, for example, a competitive binding assay using labelled BION-1 could be used for this 25 purpose. 3c or domain 4 of 3c is immobilized on a plate or tube and several compounds added, followed by labelled or tagged BION-1 or fragments thereof. Since BION-1 binds the region of 3c involved in binding all three cytokines, any compounds that block or reduce the binding of BION- 1 or fragments thereof to 3 c or domain 4 will be considered candidate inhibitory compounds. Thus, the 30 availability of BION-1 as an agent that for the first time allows the direct binding to the cytokine binding region of 3c affords a novel test for the identification of simultaneous inhibitors of GM-CSF, IL-3 and IL-5. It will be understood that the same will apply for other cytokines and their respective receptors. 35 It will be understood that not the entire Pc subunit needs be used to screen candidate compounds, and certainly the present data indicates that a fragment of the WO 00/09561 PCT/AU99/00659 9 3 c subunit encompassing domain 4 has sufficient structure in common with the native Pc subunit to reflect the configuration of the cellular target for an inhibitor useful for an in vivo effect. 5 In a broad form of a fifth aspect, the invention could be said to reside in a cytokine inhibitor capable of simultaneously blocking the binding of Pc by IL-3, GM-CSF, and IL-5 made according to the fourth aspect of the invention. It is thought that compounds that inhibit binding of the IL-3, IL-5 and GM-CSF to 10 the Pc will be therapeutically useful for intervention in conditions where IL-3, GM CSF and IL-5 play a pathogenic role, mainly allergy, asthma, acute and chronic myeloid leukaemias, lymphoma and inflammation including rheumatoid arthritis, breast cancer and prostate cancer. 15 Similarly for other common cytokine receptors it is thought that antagonists or agonists will be therapeutically useful. gp 130 is functionally analogous to Pc in that it is a common binding sub-unit and signal transducer for the 11-6, oncostatin M (OSM), ciliary neutrotrophic factor (CNTF), leukaemia inhibitory factor (LIF) and IL- 11. It is suggested that raising an antibody against a domain analogous to 20 domain 4 of Pc will also lead to blocking of two or more of these cytokines. Antagonism of this receptor system will be useful in inflammation, leukaemia and lymphoma. Antagonists to IL2R3/IL2Rcx may be useful as immunosuppresants. Antagonists of LIFR may be useful for the prevention of implantation of embryos in uteri. Antagonists of IL-4/IL-13 will inhibit IgE production and may be useful 25 in treating asthma and allergies. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1. Flow cytometry analysis of the staining of MoAb BION-1 30 (continuous line) and an isotype matched IgG, control MoAb (dotted line) to (A) COS cells transiently transfected with pc, (B) CHO cells constitutively expressing 13c, (C) TF-1.8 cells, (D) neutrophils, (E) eosinophils and (F) monocytes. 35 Figure 2. MoAb BION-1 recognizes Pc protein. (A) Immunoprecipitation of 3 c from 1251-surface labelled CHO 3c WO 00/09561 PCT/AU99/00659 10 cell lysate. Both experiments were performed on 7.5% SDS PAGE under reducing conditions. (C) MoAb BION-1, but not another anti-o3c MoAb IC 1, recognises a 3 c mutant (3c-AQP) that contains only domain 4 in the extracellular 5 regions by immunoprecipitation of CHO AQP cells. This mutant has a flag attached so it can also be seen by anti-flag MoAb M2. The experiment was performed on 10% SDS PAGE under reducing conditions. 10 Figure 3. Dose-dependent competition for the binding of 12SI-IL-5 (50 pM), 1 2 5 I-GM- CSF (50 pM) and 125 I-IL-3 (200 pM) by MoAbs BION-1 (.), an anti-3c MoAb control (), and an IgG, control MoAb (o) to TF-1.8 cells (2x106 per point). (-- ) represents ligand binding in the presence of 200-fold excess 15 unlabelled ligand. Each point is the mean of triplicate determinations. Figure 4. A Fab fragment of BION-1 blocks high affinity binding of IL-5, GM-CSF and IL-3. Binding of 1 2 5 I-IL-5 (50 pM), 20 1 2 5 1I-GM-CSF (50 pM) and 1 2 5 1I-IL-3 (200 pM) were assessed on TFI.8 cells (2x106 per point) in the presence or absence of 3000 nM MoAb BION-1, or control anti-pc MoAb, or 4200 nM Fab fragment of BION-1 for 2 h at room temperature. ICI was used as a non blocking anti-3c control. Cells were 25 separated from unbound radioligand by spinning through FCS and the resulting cell pellet was counted. The results are expressed as a percentage of the total specific cpm bound seen in the absence of antibody. Non-specific binding was determined in the presence of 200 fold excess of cold ligand. 30 Total binding seen for 1251-IL-5, 125I-GM-CSF and 1 25 I-IL-3 were 9744.2, 2567.1 and 3379.13 cpm respectively. Blocking with Fab fragment of BION-1 was determined from a single point and BION-1 and ICI values are the mean of duplicate determinations with error bars representing 1 35 standard deviation.

WO 00/09561 PCT/AU99/00659 11 Figure 5. BION-1 blocks high affinity binding of IL-5, GM-CSF and IL-3 to human eosinophils. Human eosinophils (1.8x10 6 per point) were incubated with 1 2 5 I-IL-5 (10 pM or 3 pM), 1 25 I-GM-CSF (50 pM) or 1 2 5 I-IL-3 (200 pM) either alone or 5 in the presence of 1gM MoAb at room temperature for 2 h. MoAb's 9E10 and 8E4 were used as isotype matched and non-blocking anti-Pc control respectively for BION-1. Cells were separated from unbound radioligand by spinning through FCS and the resulting cell pellet was counted. The 10 results are expressed as a percentage of the total specific cpm bound seen in the absence of antibody. Non-specific binding was determined in the presence of 200 fold excess of cold ligand and was determined to be an average of 0.3% of total counts added. Total binding seen for 1 25 I-IL-5, 1 2 5I-GM-CSF 15 and 1 2 5 I-IL-3 were 765, 622 and 748 cpm respectively. Each point is the mean of duplicate determinations and error bars represent 1 standard deviation. Figure 6. MoAb BION-1 recognizes an epitope in Pc comprising at least 20 residues M363, R364, E366 and R418. Human Pc wild type and mutants (Woodcock et al, 1996) were tested for reactivity with MoAb BION-1 and 1Cl used as a control. Following expression of wild type P 3 c and 1 3 c mutants on COS cells, 3 c was immunoprecipitated by MoAb 8E4, followed by Western 25 blotting by MoAb BION-1 (top) or the non blocking MoAb IC I (bottom). Figure 7. The binding of 1 25 1I-labelled MoAb BION-1 (1 nM)to TF. 1 cells is inhibited by IL-3 (*), but not by TNF-a (o). (---) 30 Represents inhibition in the presence of 200-fold excess of unlabelled BION- 1. Figure 8. BION-1 IgG selectively inhibits IL-5, GM-CSF and 1L-3 mediated proliferation of TF 1.8 cells. The proliferation 35 experiments represent the comparison of inhibition of BION-1 IgG at maximal dosage (400 gg/ml BION-1 with IL-5 and IL-3 and 850 gg/ml with GM-CSF) against ED 5 0 WO 00/09561 PCT/AU99/00659 12 concentrations for IL-5 (0.3 ng/ml), GM-CSF (0.03 ng/ml) or 11-3 (0.3 ng/ml). An anti-ic antibody and an irrelevant IgG antibody were used as controls. The results are expressed as DPM. Each value represents the mean of 5 triplicate determinations and error bars represent the SEM. Figure 9. Fab fragment of BION-1 and BION-1 IgG inhibits IL-5, GM-CSF and 11-3 mediated proliferation of TF1.8 cells. Intact IgG or Fab fragment of BION-1 were titrated against a 10 fixed concentration of IL-5 (0.3 ng/ml), GM-CSF (0.03 ng/ml) or IL- 3 (0.3 ng/ml) in proliferation assays where TF1.8 cells at 5xl0 4 /well were incubated for 48 hours and then pulsed for 5 hours with 0.5 gCi/well 3 H-Thymidine. The results are expressed as DPM. Each value represents the 15 mean of triplicate determinations and error bars represent the SEM. Figure 10. Eosinophil survival. (A) Viability of eosinophil after 36 hours in the presence of IL-5, IL-3 and GM-CSF. (B) 20 Viability of eosinophil after 36 hours in the presence of 1L-5, IL-3 and GM-CSF (1 nM) and different concentrations of MoAb BION-1 (o) and 8E4 (.). Each point is the mean of triplicate determinations from three samples and error bars represent 1 standard deviation. 25 Figure 11. MoAb BION-1 inhibits IL-5-stimulated CD69 up-regulation on human eosinophils. (A) CD69 up-regulation in the presence of different concentrations of IL- 5, IL-3, GM-CSF and TNF-o. (B) CD69 up-regulation stimulated by 1 nM of 30 IL-5, GM- CSF, IL-3 or TNF-oa in the presence of different concentrations of MoAb BION-1 or control anti-P3c MoAb 8E4. Each point is the mean value of three replicates and error bars represent 1 standard deviation. 35 Figure 12. Inhibition of 11L-3-induced a and 03 chain dimerization and phosphorylation by MoAb BION-1. Immunoprecipitations using anti-IL-3Ro MoAb 9F5 or anti-Pc MoAb 8E4 from WO 00/09561 PCT/AU99/00659 13 MO7e cells preincubated with MoAbs BION-1, MoAb IC I or medium alone (-) for 1 min, before being stimulated (+) or not (-) with IL-3 (50 nM) for 5 min. The figure was visualised by PhosphorImaging and the position and 5 molecular weight (in thousands) of marker proteins are shown to the left of the gels. The gels were reprobed by Western blotting analysis using anti-phosphotyrosine MoAb 3-365-10 and the top panel shows the image of part of the gels in the P 3 c area. 10 Figure 13. Screening peptides for inhibition of MoAb BION-1 binding to soluble ic domain 4 adsorbed to solid phase. E. coli derived soluble Pc domain 4 (S3c#4) was coupled to Maxisorp ELISA plates at 10gg/ml in 0.1M carbonate buffer overnight and then 15 blocked with 1% BSA. (A) B45.pep (FHWWWQP GGGCDYDDDK) (+) and (B) YB 12.pep (FPFWYHAHSPWS-GGGCDYKDDDK) (*) were derived from biopanning libraries with spc#4 using an acid eluant. B45 was allowed the bond to spc#4 at 0.0125gM and YB 12 20 was allowed to bind s3c#4 at 0.025gM. BION-1 was added to the plates at a starting concentration of 5gg/ml and serial dilutions were used to titrate the BION-1 down to 0.004gg/ml. The plate was washed again and BION-1 binding to s3c#4 was detected. 25 Figure 14. BION-1 specifically inhibits the growth in vitro of chronic myelomonoctic cells (CMML). A contol antibody (1C1) does not inhibit. 30 DETAILED DESCRIPTION OF THE INVENTION Materials and Methods AQP cDNA: To express domain 4 of 3 c on the cell surface we cloned the activated 35 3c mutant, hicAQP, with an extracellular deletion removing domains 1 to 3 WO 00/09561 PCT/AU99/00659 14 (D'Andrea et al 1996), into the eukaryotic expression vector pcDNA3 (Invitrogen). Cytokines and cell lines: Recombinant human IL-3 and GM-CSF were produced 5 in E. coli as described (Barry et a11994, Hercus et al 1994b). Recombinant human IL-5 was purified from E.coli by Bresatec (Adelaide, South Australia). Recombinant EPO was purchased from Johnson & Johnson (New Jersey). TNFct was a gift from Dr. J. Gamble in the Hanson Centre for Cancer Research. COS cells were transfected with receptor cDNA as described previously 10 (Woodcock et al 1994). CHOp3c and CHOAQP cells stably expressing either full length 3c or domain 4 respectively were generated by electroporation (Hercus et al, 1994a). TF1.8 cells were a gift from Dr J. Tavernier from University of Gent, Belgium. MO7e cells, a human megakaryoblastic cell line, were from Dr P Crozier, Aukland, New Zealand. Human eosinophils were purified from the 15 peripheral blood of slightly eosinophilic volunteers via sedimentation through dextran and centrifugation through a discontinuous density gradient of hypertonic Metrizamide, as previously described (Vadas et al 1979). Eosinophils were more than 92% pure. Human neutrophils and monocytes were purified from peripheral blood as described previously (Lopezet al, 1990) with more than 95% purity. 20 Generation of anti-fic MoAbs: BALB/c mice were immunized intraperitonally with lx107 COS cells transfected with Pc or AQP expression constructs. AQP constructs express substantially only domain 4 of the extracellular domains of 3 c. The immunizations were repeated 4 times at two-weekly intervals. Four weeks 25 after the final immunization, a mouse was boosted with 2x106 COS transfectants intravenously. Three days later, splenocytes were harvested and fused with NS-1 myeloma cells as previously described (Sun et al 1996). Hybridoma supernatants were screened on CHO 3c or CHO AQP cells by flow cytometry, with untransfected CHO cells as a control. All antibodies were from single hybridoma 30 clones as selected by limiting dilution method. MoAbs were purified from ascites fluid or hybridoma supernatant by a protein A sepharose column. The isotypes of MoAbs were tested with a Mouse MoAb Isotyping Kit (Boehringer Mannheim, Germany). Fab fragments were generated using a Fab Preparation kit (Pierce, Rockford, IL) following the supplied protocol. 35 WO 00/09561 PCT/AU99/00659 15 Immunofluorescence: Freshly purified neutrophils, eosinophils, monocytes, or CHO and COS cell transfectants (5x105) were incubated with 50gl of hybridoma supernatant or 0.25mg of purified MoAb for 45-60 min at 4oC. Cells were washed twice and then incubated with FITC-conjugated rabbit anti-mouse Ig 5 (Silenus, Hawthorn, Victoria, Australia) for another 30-45 min. Cells were then washed and fixed before analysing their fluorescence intensity on an EPICS-Profile II Flow Cytometer (Courter Electronics). Two colour staining was carried out by additional incubation with another MoAb directly coupled to PK. 10 Ligand Binding Assay: IL-3 and GM-CSF were radio-iodinated by the iodine monochloride method (Contreras 1983). 1 2 5 I-IL-5 was purchased from Dupont NEN (North Sydney, NSW, Australia). Binding assays were performed as previously described (Lopez et al 1989). Briefly, 1-2x106 TF-1.8 cells were preincubated with BION-1 Fab fragments, anti-o3c or control MoAbs over a 15 concentration range of 0.06 to 4200 nM for 1 hour. Radio-labelled ligand was then added and incubated for a further two hours before the cells were separated from free label by spinning through FCS. Counts associated with the resulting cell pellets were determined by counting on a y counter (Cobra Auto Gamma; Packard Instruments Co, Meridien, CT). Non-specific binding was determined 20 for each ligand by binding in the presence of a 200 fold excess of unlabelled cytokine. MoAb binding assay: MoAbs were radio-iodinated by the chloramine-T method (McConahey 1980). Saturation binding studies were performed by incubating 25 2x10 6 TF1 cells in a range of concentrations of radio-labelled antibodies in the presence or absence of excess unlabelled antibodies. The binding affinity of each anti-o3c MoAb to its antigen was determined by Scatchard transformation (Scatchard 1949) and analysed with the ligand program (Munson and Rodbard, 1980). Competition binding experiments were set up by preincubating the TF1.8 30 cells with a range concentration of IL-3, or GM-CSF, or IL-5 prior to adding radio-labelled MoAb QP1 for two hours as per ligand binding assay. Epitope analysis was determined by testing the capacity of each unlabelled MoAb to compete for the binding of each radio-labelled MoAb to the P3c on COS cell transfectant. 35 Co-immunoprecipitation of a and P chains and the Pc phosphorylation assays: MO7e cells were surface labelled with 125I by the lactoperoxidase method as WO 00/09561 PCT/AU99/00659 16 described previously (Walsh and Crumpton, 1977). The labelled cells incubated in either medium containing IL-3 (100 ng/ml) alone or IL-3 together with the MoAb QP1, ICI (0.5 mg/ml) or 7G3 (30 mg/ml) for 5 min. Cells were lysed in lysis buffer consisting of 137 mM NaC1, 10 mM Tris-HCI (pH 7.4), 10% 5 Glycerol, 1 % Nonindet P40 with protease and phosphatase inhibitors (10 mg/ ml leupeptin, 2mM phenylmethlysulphonyl fluoride, 10 mg/ml aprotonin and 2 mM sodium vanadate) for 30 min at 4oC followed by centrifugation of the lysate at 10,000 x g for 15 min to remove cellular debris. The lysate was precleared with mouse-Ig-coupled Sepharose beads for 18 h at 4oC and incubated with 10 anti-IL-3Ra, anti-Pc MoAb beads for 2 hr at 4oC. The beads were washed 6 times with lysis buffer and immunoprecipitated proteins were separated by SDS-PAGE under reducing condition. The immunoprecipitated proteins were detected by a PhosphorImager (Molecular Dynamics, Sunnyvale, CA). The gels were then reprobed by Western blotting analysis with an anti-phosphotyrosine MoAb, 15 3-365-10 (Boehringer Mannheim, Frankfurt, Germany). TF-1.8 cell proliferation assay: TF-1.8 cells were grown in the presence of 2 ng/ml of GM-CSF. The cells were starved for 24 hours before setting up proliferation assays as described previously (Sun et al., 1996). From 20 dose-response curves the half-maximal proliferation dosage of IL-3 (0.3 ng/ml), GM-CSF (0.03 ng/ml), IL-5 (0.3 ng/ml) or EPO (5 ng/ml) was chosen to perform proliferation experiments in the presence of a range of concentrations of MoAbs. The 3 H-Thymidine incorporation of each sample was determined by liquid scintillation and expressed as disintegrations per minute (DPM). 25 Eosinophil survival assays: The maximal dose of IL-5 required to support eosinophil survival after 36 hours was determined. Eosinophils were then cultured with 1 nM of IL 5 plus anti-Pc MoAbs for 36 hours. The viability of eosinophils was quantitated by propidium iodide staining and flow cytometry analysis as 30 described (Nicoletti, 1991). CD69 expression: CD69 expression on eosinophils was measured by means of an antiCD69 monoclonal MoAb coupled to PE by flow cytometry. 35 Pc mutants and MoAb Mapping: Single amino acid substitutions in the B'-C' and F'- G' loops of domain 4 of the Pc have been described previously (Woodcock, et WO 00/09561 PCT/AU99/00659 17 al., 1994; 1996). The cDNAs for wild type Pc and each of the 3c mutants in the B'-C' and F'-G' loops were introduced into COS cells by the electroporation (Hercus et al., 1994). Cell transfectants were analysed for surface expression with 48 hours after transfection. Mutants on the B' and C' 3-strands such as 5 L356N, W358N, I374N and Y376N were expressed on FDCP1 cells from retroviral expression constructs (Jenkins et al., 1995). Epitope-mapping of anti-3c antibodies was analysed by Immunofluorescent study. The anti-c MoAbs were tested for their abilities to recognise wild type P3c and the 3 c mutants analysed by flow cytometer using standard immunofluorescence method. For each mutant, the 10 experiment was repeated at least twice. BION-1 binding inhibitory peptides: E. coli derived soluble Pc domain 4 (s3c#4) was coupled to Maxisorp ELISA plates at 10g/ml in 0.1 M carbonate buffer overnight and then blocked with 1% BSA. B45.pep (FHWWWQP 15 GGGCDYDDDK) was derived from four rounds of biopanning the Ph.D-7mer library with soc#4 using an acid eluant. YB 12.pep (FPFWYHAHSPWS GGGCDYKDDDK) was derived from biopanning the Ph.D-12mer library with sIc#4 using an acid eluant. B45 was allowed the bond to s3c#4 at 0.0125pM and YB 12 was allowed to bind spc#4 at 0.0251M. The plate was washed in TBS + 20 0.5% Tween. BION-1 was added to the plates at a starting concentration of 5gg/ml and serial dilutions were used to titrate the BION-1 down to 0.004gg/ml. The plate was washed again and BION-1 binding to s]c#4 was detected with o mouse conjugated to HRP, using a colour based reaction which was read on a plate counter by absorption. 25 BION-1 inhibition of chronic myelomonocytic cells: Peripheral blood from a patient with chronic myelomonocytic leukemia was centrifuged over Ficoll-Paque to separate the mononuclear cells. After washing and counting, the cells were plated on agar as a concentration of 105 per plate. After incubation in medium 30 containing monoclonal antibodies BION-1 or ICI, with or without IL-3, for 14 days at 37 0 C the number of arising colonies were counted by mycroscopical examination. Each cell cluster containing more than 40 cells was counted as a colony.

WO 00/09561 PCT/AU99/00659 18 Results Development of MoAb BION-1 Previous experiments have shown that the putative F'-G' loop of Pc contains a common binding site for IL-5, GM-CSF and IL-3 (Woodcock et al, 1996; WO 5 97/28190). We have now produced a blocking compound, represented by MoAb BION-1, by immunizing mice with COS cells transfected with a cDNA coding for domain 4 of Oc. Screening of hybridoma supernatants was performed on a CHO cell line expressing domain 4 of 3c. One hybridoma cell line was identified which produced a MoAb which specifically recognized this cell line and not a parental 10 CHO cell line not expressing domain 4 of pc. This MoAb was termed BION-1 and was characterized in biochemical, binding and biological experiments. BION-1 recognizes domain 4 as well as wild type pc. MoAb BION- 1 was tested for reactivity against cell lines transfected with Pc and 15 against primary cells known to express IL-5, GM-CSF and IL-3 receptors. BION-1 recognized COS cells transiently transfected with Pc, CHO cells permanently transfected with 3 c, the erythroleukaemic TF-1 cell line, and purified peripheral blood human neutrophils, eosinophils and monocytes (Figure 1). 20 The antigen recognized by BION-1 was confirmed to be domain 4 of Pc, by biochemical analysis of transfected cells. Figure 2A shows that BION-1 immunoprecipitated a surface 1 25 I-labelled protein of about 120,000 MW consistent with the size of 3c. Similarly, BION-1 recognized a protein of 120,000 MW by Western blotting using lysates of CHO cells expressing full length wild type 3c 25 (Figure 2B). The size of these bands also corresponded to the bands recognized by a previously developed anti-3c MoAb (Korpelainen et al 1993; Woodcock et al, 1996). To formally show that BION-1 recognized domain 4 of Pc we also tested BION-1 for its ability to immunoprecipitate domain 4 expressed on the surface of CHO cells. As a positive control we incorporated a short polypeptide to the N 30 terminus of domain 4 (flag epitope) to which a MoAb has been previously developed. As a negative control, we used the anti-oc MoAb IC1 which recognizes an epitope located elsewhere in Pc. Figure 2C shows that BION-1 immunoprecipitated a band of about 80,000 MW from 1 2 5 -surface labelled-domain 4-expressing CHO cells consistent with the expected size of domain 4. MoAb M2 35 against the flag epitope added to domain 4 of 3c also precipitated a similar size protein. In contrast, MoAb 1C I failed to immunoprecipitate domain 4. These WO 00/09561 PCT/AU99/00659 19 experiments show that BION-1 can specifically recognize domain 4 of Pc on the surface of cells and following denaturation of the protein. BION-1 inhibits the high affinity binding of IL-5,. GM-CSF and IL-3 to TF-1 cells 5 and to human eosinophils Given that domain 4 of Pc is crucial for the high affinity binding of IL-5, GM-CSF and 1L3, we examined whether BION-1 was able to affect this binding. We found that BION-1 inhibited in a dose-dependent manner the binding of 125 1-IL-5, 125 GM-CSF and 1 25 I-IL-3 to the human erythroleukaemic cell line TF-1. For each 10 radioligand we used the smallest possible concentration to maximize the possibility of measuring high affinity. This can be more readily achieved with IL-3 and GM-CSF for which the difference between the low affinity component (provided by each a chain alone) and the high affinity component (provided by co-expressing Pc with each a chain) is about 1,000 fold and 30 fold respectively. 15 In the case of IL-5, the affinity conversion of Pc is only in the 25 fold range, hence, high and low affinity binding cannot be clearly separated. This is likely to explain why BION-1 shows complete inhibition of 1 2 5I-GM-CSF and 1 2 5 I-IL-3 binding (Figure 3). The residual 1 2 5 1I-IL-5 binding seen with high concentrations of BION-1 is likely to be the result of low affinity 1251-IL-5, binding (a chain) 20 which BION-1 would not be expected to inhibit. This is consistent with BION-1 inhibition of 1 2 5 I-IL-5, binding reaching a plateau beyond which no further inhibition can be detected (Fig 3a). Other anti-pc MoAb (anti-,3c control) and the IgG, MoAb control did not inhibit 1251-1L-5, 1 25 I-GM-CSF and 1 2 5 I-IL-3 binding to TF-1 cells (Figure 3). 25 The blocking effect of BION- 1 was seen whether the MoAb was used as purified IgG or as Fab' fragment. Figure 4 shows that the Fab' fragment of BION-1 blocked the binding of 50 pM 1 2 5 I-IL-5, 50 pM 1 25 I-GM-CSF and 200 pM 1 2 5

I-IL

3 to TF-1 cells. 30 Since one of the major clinical utilities of blocking IL-5, GM-CSF and IL-3 binding is likely to be in asthma, a disease in which eosinophils are believed to play a major role, it was important to test whether BION-1 could block the binding of IL-5, GM-CSF and IL-3 to these cells. As shown in Figure 5, BION-1 35 inhibited the binding of all three radio-labelled cytokines to purified human WO 00/09561 PCT/AU99/00659 20 eosinophils. In contrast, other anti-pc MoAb or the IgG 1 MoAb control failed to do so. Epitope mapping of BION-1 5 The fact that BION-1 inhibited the binding of 1251-IL-5, 12 5 I-GM-CSF and 125I-IL 3 to TF-1 cells and eosinophils suggested that it might be binding to the critical region in 3c to which these cytokines bind or at least in close proximity to it. To try to define the region/epitope in 3 c recognized by BION-1, we used several mutants of P 3 c and examined whether substitutions of individual amino acids in the 10 predicted B'-C' loop or F'-G' loop impaired BION-1 binding. Two sets of experiments were carried out. In the first instance we immunoprecipitated wild type Pc from transfected COS cells with a MoAb anti-Ne. The immunoprecipitates were then tested for reactivity with the control anti- 3c MoAb IC1, or BION-1. The results shows that Pc mutants carrying the substitutions M363A/R364A, or 15 E366A or R418A were not recognized by BION-1 (Figure 6). In a second set of experiments, the direct binding of radio-labelled BION-1 was measured on transfectants expressing the same mutants. Similar results were obtained in that whilst ICI bound with similar affinity to wild type Pc and the Pc mutants, BION-1 binding was eliminated by the M363A/R364A, E366A and R418A mutants (Table 20 I). These results suggest that the epitope recognized by BION-1 is formed, at least in part, by M363 and/or R364, E366 and R418. This is consistent with the disclosure in WO 97/28190 that agents that bind the putative F'-G' loop (of which R418 is part of) will be antagonists of IL-5, GM-CSF and IL-3. 25 BION-1 and IL-3 reciprocally inhibit each other's binding To confirm that the epitope recognized by BION-1 was the same or close to the binding site utilized by IL-5, GM-CSF and IL-3, we performed the reverse experiment, in which BION-1 was radio-labelled and increasing concentrations of IL-3 used to compete for 1 2 5 I-BION-1 binding. The results showed (Figure 7) that 30 IL-3 competed for 1 2 5 I-BION-1 binding in a dose-dependent manner emphasizing the close and intimate proximity of BION-1 and IL-3 binding epitopes in 13c BION-1 specifically inhibits the function of IL-5, GM-CSF and IL-3 including their stimulation of eosinophil production and activation. 35 To ascertain whether the inhibition of IL-5, GM-CSF and 1L-3 binding by BION-1 was translated into inhibition of IL-5, GM-CSF and IL-3 stimulation we used the factor dependent TF-1 cell line. This cell line proliferates in the presence WO 00/09561 PCT/AU99/00659 21 of either IL-5, GM-CSF, IL-3 or erythropoietin (EPO) (Figure 8). As shown in Figure 8 MoAb BION-1 but not other MoAb anti-Pc nor an IgG control MoAb inhibited the stimulation of TF-1 cell proliferation by IL-5, GM-CSF and 11-3. In contrast, the stimulating ability of erythropoietin was not inhibited showing 5 specificity of BION-1 for the IL-5/GM- CSF/IL-3 receptors system. Titration experiments showed that BION-1 inhibited cytokine-mediated TF-1 cell proliferation in a dose-dependent manner with an ED 50 of about 100-300 nM (Figure 9). Figure 9 also shows that other anti-oc MoAb were not inhibitory, and 10 that Fab fragments of BION-1 behaved similarly to BION-1 as a whole IgG with virtually overlapping ED 50 values. Since eosinophils are believed to be the major effector cells in asthma and they respond to IL-5, GM-CSF and IL-3, we examined BION-1 for its ability to block 15 eosinophil production, eosinophil survival and eosinophil activation in response to these three cytokines. We found that BION-1 but not MoAb 8E4 inhibited the ability of IL-5, GM-CSF and IL-3 to stimulate the formation of eosinophil colonies from human bone marrow cells (Table II). 20 Importantly, BION-1 inhibited the pro survival activity of IL-5, IL-3 and GM-CSF on purified peripheral blood human eosinophils. Whilst these cytokines are essential for maintaining eosinophil viability (Figure 10A), blocking of 3c by MoAb BION-1 promotes eosinophil cell death to levels similar to those observed in the absence of cytokines (Figure 10B). Eosinophils can be activated by IL-5, 25 GM-CSF and IL-3 as well as by tumour necrosis factor (TNF-x), a factor that operates through the TNF-a receptor. A sign of eosinophil activation is the upregulation of the CD69 surface antigen, a phenomenon induced by all four cytokines (Figure 11 A). Using this activation system we found that BION-1 inhibited the activation of eosinophils by IL-5, GM-CSF and 1L-3 (Figure 11B). 30 Other MoAb anti-pc or IgGcontrols failed to do so. In addition the blocking effect of BION-1 was found to be specific in that the stimulating activity of TNF-C was not inhibited (Figure 11B). BION-1 specifically inhibits IL-3 receptor dimerization and activation 35 In order to define the mechanism of BION-1 antagonism we examined BION-1 for its ability to influence receptor dimerization and activation. We have previously shown that IL-3 or GM-CSF or IL-5 induce dimerization of the respective a WO 00/09561 PCT/AU99/00659 22 chains with Oc, a phenomenon that leads to receptor activation as measured by tyrosine phosphorylation of Oc. This is confirmed here, with Figure 12 showing that in the absence of cytokines antibodies to the a chain (left panel), or P 3 c (right panel), immunoprecipitate their appropriate antigens (a chain and Pc respectively). 5 In the presence of IL-3, dimerization of a and Pc takes place allowing either anti-a chain or anti 3 c MoAb to immunoprecipitate both receptor subunits. This is accompanied by tyrosine phosphorylation of 3c (top panel). We show in this figure that pre-incubation of the cells with BION-1 blocks receptor dimerization and tyrosine phosphorylation of Pc. As a control we used the anti 0c MoAb ICI 10 which was unable to prevent receptor dimerization and activation. BION-1 specifically inhibits chronic mvelomonocytic cell growth BION- 1 is shown to inhibit the activity of one or all of IL-5, IL-3 & GM-CSF mediated effectors of leukaemic cells. In particualr BION-1 inhibits growth in 15 vitro of chronic myelomonocytic cells (CMML), whereas a control antibody (1C 1) does not (Figure 14). Furthermore, BION-1 inhbits even in the presence of IL-3 whereas the control does not. Screening and isolation of new inhibitory compounds 20 A large range of potential therapeutic compounds that might act as antagonists, or perhaps agonists of IL-3, GM-CSF and IL-5 individually or collectively, can be readily screened. The screening is initially to determine whether the binding of BION-1 or a fragment thereof to 3c receptor or fragment is inhibited. The nature of these inhibitory compounds will not be limited, and the methods used for a 25 binding assay can be any one of the many techniques known to those skilled in the art. Such methods may include affinity selection chromatography, ultrafiltration assays, the scintillation proximity assay, interfacial optical techniques, the quartz crystal microbalance, the jet ring cell, interferometric assays using porous silicon to immobilise the receptor. Reference to such techniques can be found in 30 Woodbury et al 1999, which reference is incorporated herein in its entirety. The range of therapeutic compounds may include peptides, oligonucleotides, or other small organic or inorganic molecules. Figure 13 shows the results of screening 7-mer and 12-mer peptide libraries using soluble Pc domain 4 supported 35 on ELISA plates.

WO 00/09561 PCT/AU99/00659 23 DEPOSIT OF CELL LINE The cell line BION-1 was deposited on the April 29th, 1998 in the American Type Culture Collection (ATCC) at 101801 University Boulevard, Manassas, Virginia, United States of America and has been designated ATCC HB-12525.

WO 00/09561 PCT/AU99/00659 24 Table I Epitope mapping of Bion-1. Binding affinities of MoAb Bion-1 tested on COS cells transfected with wild type Pc or mutants of 3 c 5 P3c wild type: Bion-1 KD 1CI1 KD 49.3* 4.4 Pc mutated in the B'-C' loop: M363A/R364A 0 t3.8 Y365A 69.41.5 10 E366A 02.4 H367A 27.02.8 1368A 21.72.4 D369A/H370A 32.43.6 Pc mutated in the F'-G' loop: 15 R418A 0 1.8 T419A 23.3 3.2 G420A 53.0 1.5 Y421A 38.9 2.3 * KDin nM 20 Ot = not detectable binding WO 00/09561 PCT/AU99/00659 25 Table II Inhibition of 1L-5, GM-CSF and IL-3 mediated eosinophil colony formation by BION-1 Medium [MoAb 8E4] [MoAb BION-1] 5 (100 RM) (4M) 0.1 1 10 100 IL-5 (lnM) 13 + 4* 15+3 13+4 8+2 2+2 2+0 GM-CSF (2nM) 9 + 4 18 + 4 20 + 4 13+4 2+2 0+0 IL-3(2nM) 4+2 8+1 8+2 4+1 1 + 1 0+1 10 NONE 0+0 0+0 0+0 0+0 2+0 0+0 *Number of day 14 eosinophil colonies per 105 seeded bone marrow cells. Values shown are the mean from triplicate determination ± SEM.

WO 00/09561 PCT/AU99/00659 26 References Allen et al 1997, Am J Otolaryngol 18:239-246 Bagley et al 1997a, Blood 89:1471 5 Bagley et al, 1997b; JAllergy Clin immunol; 99;725-728 Barry,S.C., et al (1994) J. Biol. Chem., 269, 8488-8492. Bates et al, 1996; J Imunol, 156:711-718 Contreras,M.A. et al (1983) Methods Enzymol., 92, 277-292. D'Andrea RJ et al, (1996) Blood 87:2641-2648 10 Davis,S. et al (1993) Science, 260, 1805-1810. Fukuda et al 1994, JAllergy Clin Immunol 94, 584 Gearing,D.P. et al (1994) Proc. Natl. Acad. Sci. USA, 91, 1119-1123. Giri,J.G., et al (1994) EMBO J., 13, 2822-2830. Hercus, Blood (1994a) Blood, 83:3500 15 Hercus,T.R. et al (1994b) Proc. Natl. Acad. Sci. USA, 91, 5838-5842. Hibi,M. et al (1990) Cell, 63, 1149-1157. Hilton,D.J. et al (1994) EMBO J., 13, 4765-4775. Hilton,D.J. et al (1996) Proc. Natl. Acad. Sci. USA, 93, 497-501. Jenkins et al (1995)EMBO J 14:4276 20 Kimura,Y. et al (1995) Int. Immunol., 7, 115-120. Korpelainen et al (1993) Proc Nat. Acad. Sci USA, 90, 11137-11141 Korpelainen et al (1995) Blood 86, 176-182. Liu,J. et al (1992) J. Biol. Chem., 267, 16763-16766. Lopez et al, (1990) Int JAller gClin Immunol 85, 99-102 25 Lopez AF et al (1989). Proc Natl Acad Sci USA 86, 7022-7026. Mauser et al, (1995), Am J Respir Crit Care Med 152; 467 McKinnon et al (1997), J Exp Med 186:121-129 McConahey et al (1980) Methods Enzymol 70:210. Metcalf; (1986) Blood 67:257 30 Munson,P.J. and Rodbard,D. (1980) Anal. Biochem.,107, 220-239. Nicoletti, I. et al (1991) J Immunol. Methods, 139:271. Noguchi,M. et al (1993) Science, 262, 1877-1880. Pennica,D. et al (1995) J. Biol. Chem., 270, 10915-10922. Russell,S.M. et al (1993) Science, 262, 1880-1883. 35 Scatchard (1949) Ann N.Y Acad Sci 51, 660-663 Sun et al (1996) Blood 87, 83-92 Sur et al, 1996, J Allergy Clin Immunol 97; 1272 WO 00/09561 PCT/AU99/00659 27 Taga,T. et al (1992) Proc. Natl. Acad. Sci. USA, 89, 10998-11001. Takeshita,T. et al (1992) Science, 257, 379-382. Tavernier et al 1995, Proc NatlAcad Sci U S A 23:5194-5198 Vadas et al (1979)J Immunol 122, 1228-1236 5 Walsh and Crumpton, (1977) Nature 269:307. Woodbury, C.P. et al (1999) J Chromatogr B Biomed Sci Appl 725(1):113-37. Woodcock et al (1994) EMBO J13:5176 Woodcock et al, (1996) J Biol Chem 271, 25999-26006 Zurawski,S.M. et al (1993) EMBO J., 12, 2663-2670. 10 WO 00/09561 PCT/AU99/00659 28 INDICATIONS RELATING TO DEPOSITED MICROORGANISM OR OTHER BIOLOGICAL MATERIAL (PCT Rule 13bis) A. The indications made below relate to the deposited microorganism or other biological material referred to in the description onpage 23 ,line 1-4 B. IDENTIFICATION OF DEPOSIT Further deposits are identified on an additional sheet Name of depositary institution American Type Culture Collection Address of depositary institution (including postal code and country) 10801 University Blvd Manassas, VA 20110-2209 United States of America Date of deposit Accession Number 29 April 1998 HB-12525 C. ADDITIONAL INDICATIONS (leave blank if not applicable) This information is continued on an additional sheet D. DESIGNATED STATES FOR WHICH INDICATIONS ARE MADE (if the indications are not for all designated States) E. SEPARATE FURNISHING OF INDICATIONS (leave blank if not applicable) The indications listed below will be submitted to the International Bureau later (specify the general nature ofthe indications e.g., "Accession Number ofDeposit") For receiving Office use only For International Bureau use only This sheet was received with the international application F This sheet was received by the International Bureau on: Authorized officer (Mrs) Anne HAMMETT Authorized officer A N k (02) 6283 2225 r.-.-. D 'T"/M r 1 I. .. II..I Ano

Claims (31)

1. A method of isolating a monoclonal antibody capable of inhibiting any one of IL-3, GM-CSF and IL-5 binding to the common receptor Pc, or a receptor analogous to c, said method including the steps of: 5 immunising an animal with a cytokine receptor or portion of a cytokine containing at least the extracellular domain 4 or analogous domain in the analogous common receptor or part thereof, isolating antibody producing cells from said animal, fusing antibody producing cells with a myeloma cell line, and 10 screening for a cell line that produces the monoclonal antibody of capable of inhibiting any one of IL-3, GM-CSF and IL-5 binding to the common receptor Pc, or a receptor analogous to Pc.

2. A method as in claim 1 wherein the immunisation involves introducing a 15 cDNA clone of a portion of or all of the common receptor including the extracellular domain 4 or analogous domain in the analogous common receptor or part thereof, into a cell and proliferating said cells to form a recombinant cell line, inoculating an animal with said recombinant cell line, isolating antibody producing cells from said animal and fusing the antibody producing cell line with a myeloma 20 cell line to form a hybridoma cell line, screening for a hybridoma cell line that produces an antibody that binds to the recombinant cell line but not to the parent, and then testing for inhibition against all three cytokines.

3. A method as in claim 2 wherein the cell into which the cDNA clone is 25 introduced is mammalian.

4. A method as in claim 3 wherein the mammalian cell line is a COS cell.

5. A method as in claim 2 wherein the cDNA encodes a full or partial portion 30 of domain 4 when it is in a configuration where the F'-G' loop and/or the B'-C' loop is in its native shape.

6. A method as in claim 2 wherein the domain 4 of Pc or equivalent domain in other cytokine receptors is expressed in isolation in a microbial host and used to 35 immunise animals for developing monoclonal antibodies. WO 00/09561 PCT/AU99/00659 30

7. A method as in claim 2 wherein the analogous receptor is any one of the cytokine superfamily receptors from the group including Ic, LIFR, gp 130, IL 2R, IL-4R/IL-13R, IL-2Ry, IL-3Ra, EPOR, TPOR and OBR. 5

8. A method as in claim 2 wherein the method is used to isolate a monoclonal antibody that inhibits binding of all of the said receptors to a common receptor.

9. A method as in claim 7 wherein the common receptor is selected from the group of receptors acting for more than one cytokine including but not limited to 10 gpl30, LIFR, IL2R/IL2Rt, IL-4R/IL-13R and Oc.

10. A monoclonal antibody, or fragments thereof capable of inhibiting the binding of the cytokines IL-3, GM-CSF and IL-5 to the Pc receptor. 15

11. A monoclonal antibody as in claim 10 wherein the monoclonal antibody or fragment thereof binds to at least the F'-G' loop of domain 4 of the Ic subunit.

12. A monoclonal antibody as in claim 10 wherein the monoclonal antibody or fragment thereof binds to at least the B'-C' loop of domain 4 of the Pc subunit. 20

13. A monoclonal antibody as in claim 10 wherein the monoclonal antibody or fragments thereof binds to both the F'-G' as well as the B'C' loop of domain 4 of the Oc. 25

14. A monoclonal antibody as in claim 10 wherein the monoclonal antibody inhibits ic receptor dimerisation.

15. A monoclonal antibody as in claim 10 wherein nucleic acid encoding the variable region of the monoclonal antibody is recombined with nucleic acid 30 encoding non-variable regions of human origin in an expression vector.

16. A monoclonal antibody as in claim 10 wherein the inhibition leads to blocking of at least one function of all three cytokines. 35

17. A monoclonal antibody as in claim 10 wherein the activity leads to inhibition of stimulation of effector cell activation or survival. WO 00/09561 PCT/AU99/00659 31

18. A monoclonal antibody as in claim 17 wherein the antibody or fragment thereof is used for treatment of asthma and leads to inhibition of IL-5, IL-3 & GM CSF mediated eosinophil activation. 5

19. A monoclonal antibody as in claim 17 wherein the antibody or fragment thereof is used for treatment of asthma and leads to inhibition of IL-5, IL-3 & GM CSF mediated eosinophil survival.

20. A monoclonal antibody as in claim 17 wherein the effector cell is selected 10 from the list including leukaemnic cells, endothelial cells, breast cancer cells, prostate cancer cells, small cell lung carcinoma cells, colon cancer cells, macrophages in chronic inflammation, and dendritic cells for immunosuppression.

21. A monoclonal antibody as in claim 17 wherein the monoclonal antibody is 15 the antibody produced by the hybridoma cell line BION-1 (ATCC HB-12525).

22. A hybridoma cell line capable of producing the monoclonal antibody of claim 10. 20

23. A hybridoma cell line as in claim 22 wherein the hybridoma cell line is BION-1 (ATCC HB-12525).

24. A method of isolating an inhibitor capable of competitively inhibiting the binding of BION-1 or the binding of an agent capable of inhibiting BION-1 25 binding, to the Pc subunit, the method including the steps of contacting BION-1 or fragment thereof with the 3c subunit or fragment thereof as well as a candidate inhibitory compound, and measuring the degree of binding. 30

25. A method as in claim 24 wherein a reporting means is provided to facilitate the detection of binding of BION 1 or fragment thereof with Pc subunit or fragment thereof.

26. A method as in claim 24 wherein the inhibitor is a peptide or a nucleotide 35 molecule.

27. An inhibitor isolated by the method of claim 24. n -.. L I I - \ -r A TT WO 00/09561 PCT/AU99/00659 32

28. A cytokine inhibitor that simultaneously blocks the binding of Ic by IL-3, GM-CSF, and IL-5. 5

29. An inhibitor of leukaemic cell proliferation wherein the inhibitor inhibits binding of IL-3, GM-CSF and IL-5 with a subunit or fragment thereof.

30. An inhibitor as in claim 29 wherein the proliferation of the cell is cytokine dependent. 10

31. An inhibitor as in claim 29 wherein the inhibitor is BION-1 or an agent capable of inhibiting BION-1 binding with 3c subunit or fragment thereof.