Purification and characterization of a monoclonal T-cell suppressor factor specific for poly(LGlu60LAla30LTyr10) (T-cell hybridoma/antigen-binding peptide/isolation)

A monoclonal T cell-derived suppressor factor specific for the terpolymer poly(LGlu'LAla3LTyr'0) produced by the T-cell hybridoma 258 C4.4, was purified to homogeneity. This was accomplished by fractionation of the culture medium by using a combination of affinity chromatography and reversephase and ion-exchange high-performance liquid chromatography. The purified factor is composed of a single Mr 24,000 polypeptide-chain, and the homogeneous protein maintains the ability to suppress antibody and T-cell proliferative responses to poly(LGlu60IAla30LTyr'0) specifically. The specific activity of pure suppressor factor is calculated to be 8 x 107 units/jig. The immune response to the synthetic terpolymer poly(LGlu'1AlaULTyr"0) (GAT) by inbred strains ofmice is controlled by an immune response (Ir gene) that maps to the K, I-A subregions of the H-2 gene complex (1, 2). Immunization with GAT stimulates antibody formation in vivo and in vitro and primes for T-cell proliferation to GAT in vitro in lymph node cells from mice bearing the responder H-2abdfk haplotypes (3, 4). In nonresponder (H-2P q s) mice, immunization with CAT does not stimulate GAT-specific plaque-forming-cell (PFC) or T-cell proliferative responses, unless GAT is complexed with a carrier such as methylated bovine serum. albumin (GAT-methylated albumin) (2-4). To date, our studies suggest that lack ofresponses to GAT in nonresponder mice is correlated with development of CAT-specific suppressor. T cells (5). Extracts of GAT-specific suppressor T cells from nonresponder mice contain a soluble suppressor factor(s), called GAT-TsF, that inhibits PFC responses to GAT-methylated albumin in nonresponder mice in vivo and in vitro (6, 7) and inhibits T-cell proliferative responses by nonresponder mice primed with, GAT-methylated albumin (8). Affinity-purified GAT-TsF is a protein of estimated Mr 45,000-60,000 determined by gel filtration and has the ability to bind specifically to GAT (9). The antigen-binding moiety bears determinants that appear to be encoded both by the Ig variable heavy chain gene complex and the I-J subregion of the major histocompatibility gene complex (10, 11). It has not been possible to purify GAT-TsF to homogeneity because of the extremely small quantity of active material in these extracts. Therefore, we have constructed T-cell hybridomas by fusing the AKR thymoma BW5147 and splenic T cells from GAT-primed nonresponder DBA/1 mice (12). Hybridomas were screened by bioassay for constitutive synthesis of GAT-TsF, and selected hybridomas were cloned. The medium from the hybridomas provides a rich source of GAT-TsF; each milliliter of medium contains 20-40 times the amount of suppressive material that can be obtained in an extract of all of the lymphoid cells from an entire mouse. In addition, protein purification and detection methodologies have been developed to the point that detection of picomolar quantities of protein may be considered routine (13, 14). In this communication, we report the successful purification of the monoclonal T-cell product GAT-TsF to homogeneity by use of affinity chromatography and high performance liquid chromatography (HPLC). To monitor the purification of this protein, the biological activity was determined by specific suppression of PFC or T-cell proliferative responses, or both, at each step. MATERIALS AND METHODS Mice. DBA/1 and B10.Q (H-2q) mice were bred in the Animal Resources facility at the Jewish Hospital of St. Louis and were 3-6 mo old when used. H-2q mice are nonresponders to GAT. Antigen and Immunization. GAT was purchased from Vega-Fox (Tucson, AZ); methylated bovine serum albumin was purchased from Sigma. The insoluble complex, GAT-methylated albumin, was prepared as described (3). DBA/1 or B10.Q mice were injected in the hind foot pad with 5-20 ,g of GAT as GAT-methylated albumin emulsified in complete Freund's adjuvant containing Mycobacterium tuberculosis H37 Ra (8). T-Cell Hybridoma Lines. Splenic T cells (enriched by passage through nylon wool) from GAT-primed, nonresponder DBA/1 mice were fused to the hypoxanthine/guanine phosphoribosyltransferase-deficient AKR thymoma BW5147 by the technique of Galfre et al. (15). Clones producing GAT-TsF constitutively were isolated and have been maintained in tissue culture or kept frozen in liquid nitrogen. GAT-TsF from the prototype hybridoma 258 C4.4 has been characterized extensively, and was used in the studies reported here. Cell Culture and Assay of Hybridoma T-Cell Supernates. Development of primary splenic PFC responses was measured in vitro under modified Mishell-Dutton conditions as described (5-7). PFC responses in cultures stimulated with GATmethylated albumin or sheep erythrocytes were assayed on day 5 with GAT-sheep erythrocyte complexes or sheep erythrocytes as indicator cells. T-cell proliferative responses to GAT by lymph node cells from mice immunized with GAT-methylated albumin were measured as described (4). The relative suppressive activity was determined by titration, and the specificity Abbreviations: GAT, poly(LGlu6rOAlaWOLTyrl0); PFC, plaque-forming cell(s); HPLC, high-performance liquid chromatography; S50,o 50% suppression of the response; GAT-methylated albumin, GAT complexed with methylated bovine serum albumin; GAT-TsF, soluble suppressor factor(s) from GAT-specific suppressor T cells. t To whom all reprint requests should be addressed. 1254 The publication costs ofthis article were defrayed in part by page charge payment; This article must therefore be hereby marked "advertisenent" in accordance with 18 U. S. C. §1734 solely to indicate this fact. Proc. Natl. Acad. Sci. USA 79 (1982) 1255 was verified by addition ofGAT-TsF to cultures stimulated with an irrelevant antigen. The proliferation assay and the antibodyforming-cell assay give linear titration curves over the dilutions tested (e.g., 10-3-10-'), allowing direct estimation of relative potency (6, 12). Supernatants from the parent thymoma BW5147 and control T-cell hybridomas contained no detectable GAT-TsF at dilutions of 1:1000 (12). The data have been expressed as the inverse of the final dilution of supernate or purified factor that causes 50% suppression of the response (S50 units/ml). Titration of suppressive activity was carried out at dilutions ranging from 5 x 10-3 to 1 x 10-'. As the material was purified, lower dilutions were eliminated. Thus, small amounts of biological activity (e.g., less than 1000 S50 units) would not be detected. Preparation and Use of Immunosorbents. To remove cell debris and serum proteins that absorb nonspecifically to antigen-coupled Sepharose, culture supernatants were prefiltered through bovine serum albumin-Sepharose at 40C. The effluent, containing GAT-TsF, was collected and applied to GAT-Sepharose, which was washed extensively and subsequently eluted with 2.0 M KCl (4). HPLC. Affinity-purified GAT-TsF was applied to a column containing Lichrosorb RP-8 (E. Merck, Darmstadt, Federal Republic of Germany) in a reverse-phase buffer of 0.5 M acetic acid/1.0 M pyridine, pH 5.5. The absorbed proteins were eluted with a stepwise gradient of 0-50% 1-propanol (16). An automated fluorescent detection system using fluorescamine (Hoffmann-La Roche) was used for monitoring protein in the column effluent (17). Ion-exchange HPLC was carried out on a column of 10-gm DEAE-Nugel DE-200 (Separation Industries, Orange, NJ). Proteins were eluted from DEAE columns with a linear gradient from 0.1 M acetic acid/0.2 M pyridine to 3 M acetic acid/6 M pyridine at a flow rate of 13 mVhr. Dansylation of GAT-TsF. Proteins were labeled with [3H]dansyl chloride by lyophilizing the sample and reconstituting it in 50 p1 of 0.1 M lithium borate. [3H]Dansyl chloride (specific activity, 13.7 Ci/mmol; 1 Ci = 3.7 x 1010 becquerels; Amersham-Searle) was lyophilized and reconstituted in acetone. Proteins were added to 1.1 nmol of [3H]dansyl chloride and allowed to react for 1 hr. The reaction was stopped by the addition of ammonium bicarbonate (0.1 M; 10 A1). Bovine serum albumin (25 pg) was added to each sample as the carrier protein. For NaDodSO4/polyacrylamide gel electrophoresis, the labeled samples were lyophilized and resuspended in a reducing buffer of 0.09 M Tris HCl, pH 6.8/1.0% NaDodSOJ 5.0% 2-mercaptoethanoV10% glycerol/7 M urea/0.001% bromophenol blue (18). Gel Electrophoresis. Purified, tritiated samples or unlabeled samples were analyzed by NaDodSOJpolyacrylamide gel electrophoresis under reducing or nonreducing conditions in 15% polyacrylamide gels (18). The gels that contained radioactive samples were fixed either in methanoVacetic acid/water, 10:7:83 (vol/vol), or in EN3HANCE (New England Nuclear). The gels were dried and placed against Kodak x-ray film (XAR5) for autoradiography. Amino Acid Analysis. Samples were dried in vacuo and hydrolyzed in 6 M HCV0. 1% thioglycollic acid at 110°C for 24 hr. After drying, the samples were dissolved in pH 2.2 buffer and analyzed in a fluorescamine amino acid analyzer as described