Communicated by J. F. A. P. Miller, March 27, 1984. ABSTRACT. Monoclonal .... ing have been described by Gray and Goeddel (9)]; (iv) IFN- y from secondary ...
Proc. Natl. Acad. Sci. USA Vol. 81, pp. 4515-4519, July 1984
Monoclonal antibodies against murine y interferon (hybridomas/immune interferon/neutralization/metabolic radiolabeling/immunoprecipitation)
MARIA PRAT*, GIORGIO GRIBAUDOt, PAOLO M. COMOGLIO*, GIORGIO CAVALLOt, AND SANTO LANDOLFOtt *Institute of Histology and Embryology and
tInstitute of Microbiology,
Medical School, University of Torino, 10126-Torino, Italy
Communicated by J. F. A. P. Miller, March 27, 1984
ABSTRACT Monoclonal antibodies against murine immune interferon (IFN-y) were produced by fusing the murine nonsecreting myeloma cell line P3.X63.Ag8.653 with spleen cells from rats immunized with IFN-y-containing supernatants obtained by stimulating a T-cell lymphoma, L12-R4, with phorbol 12-myristate 13-acetate. Supernatants from a twicecloned hybridoma (AN-18.17.24) were found to neutralize and to adsorb in depletion experiments up to 27 units of mouse IFN-y but not equivalent amounts of mouse leukocyte or fibroblast IFNs. The AN-18.17.24 monoclonal antibody neutralized to the same extent mouse IFN-y from different sourcesnamely, (i) concanavalin A-stimulated spleen cells, (ii) alloantigen-stimulated spleen cells, and (iii) monkey fibroblasts transfected with the cloned gene of murine IFN-y. Moreover, the monoclonal antibody displayed species specificity, since it did not neutralize IFN-y of human origin. Binding inhibition experiments with murine IFN-y preparations exposed to enzymatic or physicochemical degradation demonstrated that the protein moiety and not the carbohydrate residues were responsible for the binding to the AN-18.17.24 monoclonal antibody. Finally, this monoclonal antibody immunoprecipitated two molecular species of IFN-y of about 16.8 and 17.8 kilodaltons, respectively, from [35S]methionine- or [3Hlglucosamine-labeled supernatants of stimulated L12-R4 cells.
Activation of T lymphocytes by alloantigens (1, 2), mitogens (3), and conventional antigens (4) induces the release of an interferon (IFN) type classified as immune (IFN-y) that acts on a variety of somatic and immunorelated target cells in a genetically unrestricted manner (5). Some in vitro and in vivo studies indicate that IFN-y has a greater antiproliferative effect on neoplastic cells than do leukocyte and fibroblast IFNs (IFN-a and IFN-p, respectively) (6). Recently, genes of both human and murine IFN-y have been successfully cloned by recombinant DNA technology. The results of these experiments show that the molecular weights of human and mouse IFN-'y are 17,100 and 15,900, respectively (7-9). A limit of this kind of technology, however, is that it does not take into account the possible posttranslational modifications of the synthesized molecules. In fact, as far as IFN-y is concerned, discrepancies on molecular weights have been reported in studies using conventional physicochemical techniques for purification (10) as opposed to those using recombinant DNA technology (7-9). Monoclonal antibodies have proved to be valuable reagents for the unambiguous identification, quantitative analysis, and large-scale purification of a variety of proteins (11). Whereas different monoclonal antibodies have already been produced and successfully used for the purification and biological characterization of IFN-a and -,/ (12-15), few and conflicting reports exist in the case of monoclonal antibodies against human IFN-y (6, 16). In previous studies we established in vitro a murine T-cell The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
lymphoma line producing, upon stimulation with phorbol 12myristate 13-acetate (PMA), huge amounts of homogeneous IFN-y apparently not contaminated by other known lymphokines (17, 18). Taking advantage of this system, we now have produced a rat monoclonal antibody recognizing murine IFN-y but not murine IFN-a or -,3.
MATERIALS AND METHODS Immunogen Preparation. The characterization of the L12R4 lymphoma cell line and of IFN-y production have been described (17). Briefly, L12-R4 cells were resuspended at a concentration of 1 x 106 cells per ml in RPMI 1640 medium supplemented with 10% fetal calf serum, 2 mM glutamine, 50 ,uM 2-mercaptoethanol, gentamycin at 50 pgg/ml, and aphidicolin at 3 ,ug/ml. This drug was used to synchronize tumor cells and increase IFN-y yield (19). After 24 hr cells were washed three times and stimulated with PMA (0.2 ,uM) for 48 hr at 37°C in humidified 5% C02/95% air. Cells were then spun down, supernatants were harvested and filtered, and protein was precipitated with ammonium sulfate at 4°C. Solid ammonium sulfate was added to crude supernatants to give a 50% (wt/vol) saturated solution. The large precipitate was removed by centrifugation and discarded. Ammonium sulfate was then added, again to 80% saturation, and the precipitate was collected by centrifugation and dissolved in 10 ml of 0.5 M NaCl/0.05 M potassium phosphate buffer, pH 7.4. After overnight dialysis against the same buffer, this fraction was stored at -80°C until used. Immunization Procedure. Female Wistar rats were primed intrafootpad with 3 mg of protein from the concentrated ammonium sulfate fraction emulsified in complete Freund's adjuvant. Nine booster injections were given at 3-week intervals with the same amount of protein emulsified in incomplete Freund's adjuvant. It is calculated that each rat received a total of 2 x 105 IFN-y units. Of the 16 animals immunized, 4 showed a serum titer ranging from 1:256 to 1:512 in a neutralization aisay against 9 IFN-y units. Two months after the last inoculum, the animal chosen for fusion received a final subcutaneous booster injection without adjuvant and was sacrificed 4 days later. Hybridization and Selection of Hybridomas. Spleen cells (15 x 107) from the immune rat were fused with 70 x 106 P3.X63.Ag8 variant 653 (mouse myeloma) cells by using 41.7% (vol/vol) polyethylene glycol 4,000 as fusing agent and following the procedures of Hammerling et al. (20). The fused cells were distributed in 264 wells (Nunclon 24-well plates) in RPMI 1640 medium supplemented with 10% fetal calf serum and the hypoxanthine/aminopterine/thymidine (HAT) components. The hybrid cultures positive in the neutralization assay against 9 IFN-y units were cloned twice. Abbreviations: IFN-y, immune interferon; IFN-a, leukocyte interferon; IFN-,B, fibroblast interferon; Con A, concanavalin A; PMA, phorbol 12-myristate 13-acetate; kDa, kilodaltons. tTo whom reprint requests should be addressed at: Istituto di Microbiologia, Via Santena 9, 10126-Torino, Italy.
Prat et aL
IFN Assay, Neutralization, and Depletion. IFN assays were performed by microtitration of cytopathology of vesicular stomatitis virus (VSV) on mouse L929 fibroblasts. The following IFN preparations were used: (i) L12-R4 IFN-y, produced by L12-R4 tumor cells as previously described (17); (ii) concanavalin A (Con A)-induced IFN--y, produced by stimulating C57BL/6 spleen cells with Con A (2.5 gg/ml) for 48 hr in vitro; (iii) recombinant IFN--y, kindly provided by J. Ortaldo, (Biological Response Modifier Program, National Cancer Institute, Frederick, MD) recovered from COS-7 (a monkey fibroblast cell line) cultured in serum-free conditions after transfection with the cloned murine IFN-y gene [details of the methods used for the mouse IFN- y gene cloning have been described by Gray and Goeddel (9)]; (iv) IFNy from secondary mixed lymphocyte cultures of C57BL/6 spleen cells stimulated twice in vitro with DBA/2 irradiated spleen cells as described (2); (v) mixture of mouse IFN-a and -3, kindly provided by E. De Maeyer, elicited with Newcastle disease virus in monolayer cultures of Swiss mouse C-243 cells as described (15); and (vi) human IFN-y, obtained by stimulating for 72 hr peripheral blood lymphocytes with phytohemagglutinin (1%) and PMA (10 /xM). For neutralization experiments, approximately 27 units of each IFN preparation were incubated with an equal volume of hybrid superna7 tants diluted 1:4 or with an equal volume of preimmune or immune rat serum diluted 1:32. After overnight incubation at 40C, the mixtures were assayed for antiviral activity. For depletion experiments, about 27 units of murine IFNy or IFN-a/IFN-13 were incubated with 100 1.l of hybrid supernatant for 3 hr at 40C, followed by a 10-fold excess of affinity-purified rabbit anti-rat Ig bound to protein A-Sepharose CL-4B for 1 hr. The mixture was then centrifuged, and the supernatant was tested for IFN-,y activity in the biological assay and for the presence of residual rat Ig by a radioimmunoassay (21). Enzymatic and physicochemical treatments of IFN-y preparations. For inhibition binding experiments (see below) IFN-,y preparations were exposed to the following treatments. (i) Trypsin digestion was performed by mixing 100 units of IFN-,y to twice-crystallized trypsin (100 ,g/ml) from bovine pancreas (Sigma, Munich, Federal Republic of Germany) for 30 min at 37°C. The reaction mixture was stopped by adding soybean trypsin inhibitor (50 ,g/ml). (ii) Glycosidase digestion was performed by incubating 100 units of IFN- y for 4 hr at 37°C as described by Otto et al. (22) with a glycosidase mixture from Turbo cornutus (Seikagaku Kogyo, Tokyo, Japan), which was free of proteases and Contained a-mannosidase (116 units), /3-glucosidase (48 units), ,B-galactosidase (93 units), a-L-fUcosidase (50 units), and ,B N-acetylglucosaminidase (204 units). (iii) Exposure to high temperature was performed by heating IFN-y preparations (100 units/ml) at 65°C or 45°C for 10 min. After each treatment, the equivalent of 27 units of IFN- y was incubated with a pretitrated amount of AN-18.17.24 monoclonal antibody for 3 hr at 4°C, and then the whole mixture was tested for the residual capability to inhibit 27 units of untreated IFN-y in the biological assay. Internal labeling and immunoprecipitation of mouse IFNy. Radiolabeled IFN-y was produced by incubating PMAstimulated L12-R4 cells (1 x 106 cells per ml) for 24 hr in methionine-free RPMI 1640 medium containing [35S]methionine (25 ,uCi/ml) (Amersham, Buckinghamshire, U.K.; 1050 Ci/mmol; 1 Ci = 37 GBq). [3H]Glucosamine labeling was performed by incubating for 24 hr PMA-stimulated L12-R4 cells in RPMI 1640 medium containing 100 ,Ci of [3H]glucosamine (Amersham; 34.6 Ci/mmol). After incubation, culture supernatants were clarified at 20,000 x g for 20 min, phenylmethylsulfonyl fluoride (final concentration, 2 mM) and aprotinin (6 1.l/ml; Sigma, Uppsala, Sweden) were added as protease inhibitors, and the treated supernatants were
Proc. Natl. Acad. Sci. USA
used for immunoprecipitation or kept frozen in liquid nitrogen. A two-step immunoprecipitation procedure was used, the first removing nonspecifically bound radiolabeled material by means of normal Ig-Sepharose 4B and the second treating with specific rat antibodies, followed by affinity-purified rabbit anti-rat Ig and protein A-Sepharose CL-4B as previously described (23). Immunocomplexes were then washed three times with 10 ml of 50 mM Tris, pH 7.4/150 mM sodium chloride/5 mM EDTA/0.5% Nonidet P-40/0.5% sodium deoxycholate (NET buffer). Samples were then eluted with 50 A.l of Laemmli buffer (see below) containing 2% NaDodSO4. NaDodSO4/PAGE was carried out in 10-16% acrylamide gradient slab gels by the procedures described by Laemmli (24). Gels were subjected to fluorography as described by Laskey and Mills (25), dried, and autoradiographed. The following radiolabeled molecular size standards were used: phosphorylase B [93 kilodaltons (kDa)], ovalbumin (46 kDa), carbonic anhydrase (30 kDa), and cytochrome c (12.5 kDa) (Amersham).
RESULTS The sera of only four out of the 16 immunized animals contained high titers of antibodies (from 1:256 to 1:512) that neutralized 27 units of mouse IFN-y in the standard assay used. One-third of spleen cells from a good responder rat were fused with the nonsecreting murine myeloma 4 days after the last immunization, and hybrid cells were distributed in 264 wells in Nunclon trays in selective medium. Growth of hybrids was observed in 262 wells. Supernatants were tested for IFN-neutralizing antibodies when cultures were about 30-40% confluent (12-14 days after fusion). Among the 262 cultures, the supernatants from three hybrids, AN-18, AN44, and AN-45, displayed neutralizing activity against 27 units of mouse IFN-y. The last two, however, lost their activity before cloning and were discarded. The AN-18 hybrid was cloned twice by limiting dilution, and the specificity of the secreted antibody was tested against different types of mouse and human IFNs. As shown in Table 1, AN-18.17.24 monoclonal antibody neutralized 27 units of mouse IFN-y but not equivalent amounts of mouse IFN-a and -/3 or human IFN-y. A similar pattern of reactivity was observed with serum from the immune rat used for the fusion. By contrast no neutralizing activity was observed with normal rat serum. To rule out the possibility that the AN-18.17.24 monoclonal antibody could interact with IFN-y receptors or interfere with the activity in some other ways, experiments of IFN-y depletion by immunoprecipitation were undertaken. As shown in Table 2, incubation of murine IFN-y preparation with the monoclonal antibody, followed by affinity-purified Table 1. Neutralization of mouse IFN--y by AN-18.17.24 monoclonal antibody IFN units* IFN type Antibody 27 MuIFN-y 27 Normal rat serumt