Identification of two potential receptor-binding sites for hGM-CSF

Identification of two potential receptor-binding sites for hGM-CSF

M.O.Eberhardt^I; R.Frank^II; R.Kratje^I; M.Etcheverrigaray^I

^ILaboratorio de Cultivos Celulares, Facultad de Bioquímica y Ciencias Biológicas.Universidad Nacional del Litoral. C.C 242, S3000ZAA, Santa Fe, Argentina

^IIMolekulare Erkennung. Gesellschaft für Biotechnologische Forschung mbH, GBF. Mascheroder Weg 1, D-38124, Braunschweig, Germany

^{Address to correspondence}

ABSTRACT

Two receptor-binding sites for hGM-CSF are described. Competitive binding ELISA using four monoclonal antibodies (MAbs) showed different epitope recognitions. The antibody combining sites were mapped using sets of overlapping peptides and hexapeptide libraries prepared by the SPOT synthesis technique. We identified the conformationally dependent epitopes A₁₈E₂₁R₂₃R₂₄F₁₁₉ and R₂₃E₂₁N₁₇W₁₃ bound by MAb CC5B5 and the nonlinear epitope P₁₁₈F₁₁₉W₁₃E₁₄ bound by MAb M1B8. The epitopes recognized by these two MAbs are very closely located on the native protein surface. The peptide L₆₁YKQGKLRGSLTK₇₂ was recognized by MAb M7E10 and the peptide A₁PAR₄, representing the N-terminal sequence of the protein, was bound by the nonneutralizing MAb CC1H7. Inhibition assays of the GM-CSF biological activity demonstrated that MAb M1B8, CC5B5 and M7E10 bind to domains which are responsible for the interaction of the cytokine with the GM-CSF receptor.

Keywords: GM-CSF, monoclonal antibodies, SPOT synthesis, epitope mapping.

INTRODUCTION

Human Granulocyte Macrophage-Colony Stimulating Factor (hGM-CSF) is a member of the four-helix-bundle family of cytokines that regulates the proliferation and differentiation of hematopoietic progenitor cells (Sieff et al., 1985; Metcalf et al., 1986). It has potential clinical utility, enhancing the rate of hematopoietic recovery after cancer chemotherapy. Bioactive hGM-CSF was expressed in distinct host systems: bacteria, yeast and mammaliam cells. All of them have biological activity; however, patients receiving either nonglycosylated or partially glycosylated rhGM-CSF develop antibodies that impair the clinical efficacy of the trial (Wadhwa et al., 1996).

The GM-CSF biological activity is exerted by a high-affinity cell surface receptor (R) (Gearing et al., 1989). It consists of a ligand-specific a subunit that confers low-affinity binding properties and a b subunit that can also associate with interleukin-3 and -5 (Hayashida et al., 1990).

Regardless of the multiple studies on GM-CSF, little is known about the essential regions of the molecule that confer bioactivity. In fact, the amino acid 14-24 sequence in the a helix of GM-CSF was predicted to be an important sequence involved in its biological activity (Clark-Lewis et al., 1988). Other approaches using site-directed mutagenesis demonstrated that Glu₂₁ residue plays a critical role in the binding to GM-CSFR (López et al., 1992).

We produced a panel of monoclonal antibodies (MAbs) following immunization with pure E.coli-derived hGM-CSF. The panel was characterized in terms of immunochemical behavior, neutralizing ability of the biological activity and epitope sequences recognized by the antibodies in order to evaluate GM-CSF epitopes involved in the bioactivity. In this study, we confirmed that the 14-24 region is essential for receptor recognition and described a new potential area in the neighborhood of the C-terminal region of the molecule that is highly necessary for its biological activity.

MATERIALS AND METHODS

Production and Purification of the Mabs

The MAb-producing hybridoma clones were established from Balb/c splenocytes following immunization with pure E. coli-derived GM-CSF using standard fusion protocols (Galfrè and Milstein, 1981). Hybridomas were grown in mice to produce intraperitoneal ascitic fluid and MAbs were purified by protein A affinity chromatography (Pharmacia, Sweden). The MAbs selected were M1B8, M7E10, CC5B5 and CC1H7.

Competitive Binding Assay

To determine the number of distinct epitopes and the relationship between the epitopes identified by each MAb, competitive binding assays were carried out. Purified antibodies were conjugated to biotin (Bayer and Wilchek, 1980) and allowed to react with 50 ng GM-CSF adsorbed onto microtiter plates in the presence of nonbiotinylated antibodies. The binding of each biotinylated antibody was measured using peroxidase-conjugated avidin (Sigma, USA).

Mapping of GM-CSF-derived Epitopes with Cellulose-bound Peptides by SPOT Synthesis

All cellulose-bound sets of peptides (peptide scans and libraries) were automatically prepared according to standard SPOT synthesis protocols (Frank, 1992) using a SPOT synthesizer (Abimed GmbH, Germany).

(a) Screening of Linear Epitopes with Overlapping Peptide Scan

In order to map linear epitopes, each antibody was assayed with overlapping peptide scans (peptide length 15 aa, offset of 3 aa), spanning the GM-CSF primary sequence. Then, epitope boundary analyses were performed with overlapping peptides having an offset of one and a stepwise reduced length (from 15 mer to 4 mer). The analysis of peptide-bound antibodies was carried out as previously described by Frank and Overwin (1996).

(b) Screening of Discontinuous Epitopes with Combinatorial Libraries

Combinatorial hexapeptide libraries were prepared to delineate peptide sequences that are recognized by the antibodies and mimic conformational epitopes (mimotopes). The following strategies for the deconvolution of individual sequences were used: a dual positional scanning with 2,000 hexapeptides in one single screen and a second and third generation of an iterative search, starting with two defined positions (400 hexapeptide library). Analysis of peptide-bound antibodies was carried out using a chemiluminiscence detection system (Bio Rad, USA).

(c) Overlapping Peptide Scan Followed by Electroblotting

The technique described by Rüdiger et al. (1997) was used. The GM-CSF-derived peptide scan (peptide length 15 aa, offset of 3 aa) membranes were incubated overnight at 4°C with each antibody. The peptide membranes were washed with Tris buffered saline and then the peptide-bound antibodies were electroblotted onto polyvinylene difluoride membrane (PVDF) at a constant strength of 0.8 mA/cm² during 2 h in a semi-dried blotter (Biometra, Germany). After blocking, the PVDF membranes were incubated with alkaline phosphatase-labelled anti-mouse antibody (Jackson Immuno Research Lab., Germany). Analysis of peptide-bound antibodies was carried out with a chemiluminiscence detection system.

Inhibition of the Biological Activity by Monoclonal Antibodies

A bioassay based on the TF-1 cell line which proliferates in response to GM-CSF was used (Kitamura et al., 1989). Purified MAbs were diluted serially with RPMI 1640 (Gibco, USA) containing 5% fetal calf serum (FCS; Bioser, Argentina) and preincubated for 1 h at 37°C with an equal volume (25 µl) of E. coli-derived GM-CSF (6.25 U/ml). Exponentially growing TF-1 cells were washed four times, resuspended at a concentration of 2 × 10⁵cells/ml in RPMI 1640 supplemented with 5% FCS and added in 50 µl aliquots to each well. The plates were incubated for 48 h and cell proliferation was determined by measuring the activity of the dehydrogenase enzyme with a commercial colorimetric kit (Cell Titer 96^TM, Promega, USA).

RESULTS AND DISCUSSION

Competitive Binding Assay

The epitope specificity of the MAbs was analyzed using the competitive binding ELISA. The biotin-labelled MAbs were used as probes to study the ability of unlabelled homologous or heterologous MAbs to inhibit binding to GM-CSF (Fig.1). As expected, homologous antibodies instead of heterologous MAbs competed completely for the binding of the biotinylated probes, which produced different competition patterns. It was also observed that some heterologous MAbs enhanced the binding of the probe, indicating that a conformational change in the antigenic molecule after the binding of one antibody might enhance the binding of a second antibody. Nevertheless, the alosteric change identified antibodies that bind to independent epitopes.

Moreover, when each antibody was used to cross-block the binding of the other three antibodies, three distinct epitope patterns were found. MAb CC5B5 partially competed with MAb M1B8. However, we classified them as antibodies that map the site referred to as A (Table 1) because they showed the same binding features in the presence of the other MAbs. Except for MAb M1B8 and CC5B5, none of the remaining MAbs competed with each other for the GM-CSF binding. Therefore, MAb M7E10 and CC1H7 had to bind to different sites, referred to as site B and site C, respectively.

Mapping of GM-CSF-derived Epitopes with Cellulose-bound Peptides by SPOT Synthesis

The epitope groups were analyzed in terms of the essential amino acids needed for the binding, using cellulose-bound peptide scans and combinatorial libraries generated by the SPOT synthesis technique. MAb CC1H7 recognized the continuous epitope APARSPSPSTQPWEH representing the N-terminal sequence of the protein. Further analysis of epitope boundaries elucidated the sequence A₁PAR₄ as the linear epitope recognized by this MAb.

Considering that peptide-based protein scanning procedures usually fail to identify discontinuous epitopes, we carried out combinatorial hexapeptide library approaches in order to describe linear peptide sequences that are recognized by antibodies and mimic conformational epitopes. Dual-positional scanning of a combinatorial hexapeptide library and subsequent iterative searches with two defined positions showed that MAb CC5B5 binds peptides AERRF and RERW, probably mimicking the conformational dependent epitopes A₁₈E₂₁R₂₃R₂₄F₁₁₉ and R₂₃E₂₁N₁₇W₁₃. These amino acid sequences were found close to one another on the GM-CSF molecule. MAb M1B8 primarily binds peptides PFEWE, FFEWE and WFEWE. These peptides might mimic the nonlinear sequence P₁₁₈F₁₁₉W₁₃E₁₄ in the GM-CSF molecule. Taking into account the three-dimensional structure of the native GM-CSF molecule, we observed that the nonlinear epitopes bound by both MAbs were located very close to each other on the protein surface (Fig. 2). Based on this steric hindrance, the partial competition between them could be explained.

In previous epitope screening procedures, the binding of antibodies to cellulose-immobilized peptides was detected by using a second enzyme-labeled antibody. Nevertheless, low-affinity peptide-antibody interactions are difficult to identify since the binding equilibrium of the peptide-antibody complex in the membrane is shifted towards the noncomplexed molecules during the incubation with the second antibody (Reineke et al., 1998). In order to overcome this drawback, the electroblotting of peptide-bound antibodies onto PVDF membranes was carried out. This strategy could describe the discontinuous epitope recognized by MAb M7E10, where the peptide LYKQGLRGSLTK could be part of this assembled epitope. The discontinuous feature of the MAb M7E10-bound epitope was demonstrated in previous work (Oggero Eberhardt et al., 2001).

Inhibition of the Biological Activity by Monoclonal Antibodies

Antibodies from groups A and B neutralized GM-CSF biological activity, with the latter showing the higher capacity to inhibit it (Table 1).

Antibodies from group C neutralized the activity with the lowest ability, taking into account the very high concentration needed to produce 50% of the maximal proliferation (Fig. 3 and Table 1). Therefore, epitope groups A and B are assumed to be involved in the GM-CSF-receptor interaction.

CONCLUSIONS

In this study, we identified two potential receptor-binding sites of the hGM-CSF molecule based on inhibition assays of its biological activity by a panel of MAbs.

The MAbs were characterized in terms of their epitope specificity, showing three different recognition patterns. The amino acid sequence recognized by each epitope was established by employing the SPOT synthesis method. The use of peptide scans allowed us to map the linear epitope A₁PAR₄, located in the N-terminal helix of GM-CSF. This epitope was bound by MAb CC1H7, which was considered a nonneutralizing antibody. To overcome the limitations of mapping discontinuous epitopes by use of peptides, we developed combinatorial hexapeptide libraries and peptide scans followed by electroblotting. In this way, we identified the conformational dependent epitopes A₁₈E₂₁R₂₃R₂₄F₁₁₉ and R₂₃E₂₁N₁₇W₁₃ bound by MAb CC5B5 (group A), and the nonlinear epitope P₁₁₈F₁₁₉W₁₃E₁₄ bound by MAb M1B8 (group A). Finally, we observed that peptide L₆₁YKQGLRGSLTK₇₂ was recognized by MAb M7E10 (group B). In addition, the corresponding antibodies from groups A and B showed a high capacity to neutralize GM-CSF biological activity, indicating that they are involved in GM-CSF-receptor binding. Nevertheless, MAbs M1B8 and CC5B5, whose binding domain includes amino acids from helix A and the carboxy-terminal tail of the molecule, showed a lower neutralizing ability than antibody M7E10, whose binding domain is composed of the last residues of helix B and the initial nine residues of the loop that joins helix B with helix C.

Precedent structure-function analysis of hGM-CSF demonstrated the critical role of Glu₂₁ in binding to the high affinity receptor of the cytokine (López et al., 1992). This role was confirmed in this study by the neutralizing antibody CC5B5, whose epitope comprises the amino acid Glu₂₁ in its sequence. Notably, the MAb M7E10 showed a higher neutralizing capability than MAb CC5B5 (30 times higher), pointing to a significant area of biological activity by the GM-CSF molecule that had not been described before.

R.Frank

E-mail: [e-mail]

Received: March 5, 2002

Accepted: August 30, 2002

• Bayer, E.A., Wilchek, M. (1980). The Use of Avidin-Biotin Complex as a Tool in Molecular Biology. Methods Biochem. Anal., 26, 1-45.
• Clark-Lewis, I., Lopez, A. F., To, L. B., Vadas, M. A., Schrader, J. W., Hood, L., Kent, S. B. H. (1988). Structure-Function Studies of Human Granulocyte- Macrophage Colony-Stimulating Factor. Identification of Residues Required for Bioactivity. J. Immunol., 141, 881-889.
• Frank, R. (1992). SPOT-Synthesis: An Easy Technique for the Positionally Addressable, Parallel Chemical Synthesis on a Membrane Support. Tetrahedron, 48, 9217-9232.
• Frank, R., Overwin, H. (1996). SPOT Synthesis. Epitope Analysis with Arrays of Synthetic Peptides Prepared on Cellulose Membranes. Methods in Molecular Biology, vol. 66: Epitope Mapping Protocols, 15, 149-169.
• Galfrè, G., Milstein, C. (1981). Preparation of Monoclonal Antibodies: Strategies and Procedures. Methods in Enzymology, 73, 3-46.
• Gearing, D. P., King, J. A., Gough, N. H., Nicola, N. A. (1989). Expression and Cloning of a Receptor for Human Granulocyte-Macrophage Colony-Stimulating Factor. EMBO J., 8, 3667-3676.
• Hayashida, K., Kitamura, T., Gorman, D. M., Arai, K., Yokota, T., Miyajima, A. (1990). Molecular Cloning of a Second Subunit of the Receptor for Human Granulocyte-Macrophage Colony-Stimulating Factor (GM-CSF): Reconstitution of a High-affinity GM-CSF Receptor. Proc. Natl. Acad. Sci. USA, 87, 9655-9659.
• Kitamura, T., Tange, T., Terasawa, T., Chiba, S., Kuwaki, T., Miyagawa, K., Piao, Y-F., Miyazono, K., Urabe, A., Takaku, F. (1989). Establishment and Characterization of a Unique Human Cell Line that Proliferates Dependently on GM-CSF, IL-3 or Erythropoietin. J. Cell. Physiol., 140, 323-334.
• López, A. F., Shannon, M. F., Hercus, T., Nicola, N. A., Cambareri, B., Dottore, M., Layton, M. J., Eglinton, L., Vadas, M. A. (1992). Residue 21 of Human Granulocyte-Macrophage Colony- Stimulating Factor is Critical for Biological Activity and for High but not Low Affinity Binding. EMBO J., 11, 909-916.
• Metcalf, D., Begley, C. G., Johnson, G. R., Nicola, N. A., Vadas, M. A., Lopez, A. F., Williamson, D. J., Wong, G. C., Clark, S. G., Wang, E. A. (1986). Biologic Properties in vitro of a Recombinant Human Granulocyte-Macrophage Colony- Stimulating Factor. Blood, 67, 37-45.
• Oggero Eberhardt, M. R., Frank, R., Kratje, R., Etcheverrigaray, M. (2001). Neutralization of the Biological Activity of Glycosylated and Non-glycosylated hGM-CSF by Monoclonal Antibodies. In: Animal Cell Technology - From Target to Market (E. Lindner Olsson, N. Chatzissavidou, E. Lüllau, eds.), ISBN 1-4020-0264-5, Kluwer Academic Publishers, Dordrecht, The Netherlands, 241-243.
• Reineke, U., Sabat, R., Volk, H. D., Schneider- Mergener, J. (1998). Mapping of the Interleukin- 10/Interleukin-10 Receptor Combining Site. Protein Sci., 7, 951-960.
• Rüdiger, S., Germeroth, L., Schneider-Mergener, J., Bukau, B. (1997). Substrate Specificity of the DnaK Chaperone Determined by Screening Cellulose-Bound Peptide Libraries. EMBO J, 16,  1501-1507.
• Sieff, C. A., Emerson, S. G., Donahue, R. E., Nathan, D. G., Wang, E. A., Wong, G. G., Clark, S. C. (1985). Human Recombinant Granulocyte -Macrophage Colony-Stimulating Factor: A Multilineage Hematopoietin. Science,  230,  1171- 1173.
• Wadhwa, M., Bird, C., Fagerberg, J., Gaines-Das, R., Ragnhammar, P., Mellstedt, H., Thorpe, R. (1996). Production of Neutralizing Granulocyte-Macrophage Colony-Stimulating Factor (GM- CSF) Antibodies in Carcinoma Patients Following GM-CSF Combination Therapy. Clin. Exp. Immunol, 104, 351-358.

Publication Dates

• Publication in this collection
  19 Mar 2003
• Date of issue
  Mar 2003

History

• Received
  05 Mar 2002
• Accepted
  30 Aug 2002