by cells releasing IgM antibody to phosphorylcholine (9). Preliminary results (10) also indicated that such anti-idiotypic

Transcription

1 Proc. Nat. Acad. Sci. USA Vol. 69, No. 9, pp , September 1972 Specific Suppression of the Antibody Response by Antibodies to Receptors (plasmacytomas/phosphorylcholine/anti-idiotypic antiserum/homeostasis) HUMBERTO COSENZA AND HEINZ KOHLER Department of Pathology, University of Chicago, Chicago, Illinois Communicated by Hewson Swift, July 17, 1972 ABSTRACT Balb/c myeloma proteins, TEPC 15 and MOPC 167, bind phosphoryicholine and precipitate with the C- moiety of pneumococci R36A. Antiidiotypic antibodies to TEPC 15 myeloma raised in A/He mice prevent plaque formation by cells releasing antibody to phosphoryicholine and inhibit induction of the antibody response to phosphorylcholine in vitro and in vivo. This suppression is specific since antibodies against noncrossreacting idiotypic determinants of MOPC 167 do not prevent either formation of plaques or induction of the antibody response. Furthermore, the response to sheep erythrocytes is not suppressed by antibodies to TEPC 15. These results indicate that antibodies to phosphorylcholine and cell receptors for phosphorylcholine share similar antigen-binding sites with TEPC 15 myeloma protein. Thus, anti-idiotypic antibodies directed against TEPC 15 myeloma protein function as "anti-receptor" antibodies since they specifically prevent induction of the primary response to phosphorylcholine. It is proposed that antibodies to receptors may be involved in the homeostasis of the immune response. The presence of immunoglobulin molecules on the surface of lymphoid cells, particularly on cells derived from bone marrow, is now well established (1-4). This evidence includes the demonstration of immunoglobulin molecules on the surface of lymphoid cells by immunofluorescent techniques with antisera to immunoglobulins (1), by the adherence of immunocompetent cells to bead columns antisera to immunoglobulins (2), and by isolation of immunoglobulins from the surface of malignant and normal lymphoid cells (3, 4). Thus far, nonspecific suppression of the antibody response has been achieved by antisera directed against the heavy or light chains of immunoglobulins (5, 6). Finally, suicidal killing of receptor-bearing cells by heavily labeled antigen (7) is specific for the antigen but yields no information as to the nature and biological function of the antigen-sensitive cell. Circumstantial evidence now suggests that the combining site of the receptor molecules present on cells specifically reactive to an antigenic determinant is similar to the combining site of the antibody produced in response to the antigen (8). Our objective was to obtain antibodies directed against the receptor molecules for a single antigenic determinant and to use those antibodies to specifically suppress the response to only that antigen. We used Balb/c myeloma proteins with Abbreviations: Plasma cell tumors are designated by letters and their myeloma proteins by numbers only; i.e., tumor TEPC 15 produces protein 15; PFC, plaque-forming cell antibody activity against the hapten phosphorylcholine to prepare anti-idiotypic antibodies in A/He mice. These antiidiotypic antibodies specifically inhibited plaique formation by cells releasing IgM antibody to phosphorylcholine (9). Preliminary results (10) also indicated that such anti-idiotypic antibodies specifically suppressed induction of the immune response to phosphorylcholine. Thus, anti-idiotypic antibodies are operationally "anti-receptor" antibodies, and we refer to them as such. In the present paper, we extend these observations and suggest a possible role of such antibodies in regulation of the immune response. MATERUILS AND METHODS Animals. 8- to 12-week-old female Balb/c mice from Cumberland Farms were used in all experiments. Antigens. Strain R36A, rough pneumococci was obtained from the American Type Culture. Pneumococci were killed by exposure to heat at 900 for 60 min and preserved in 0.6% Formalin-0.15 M NaCl. Sheep erythrocytes were obtained from a single sheep maintained at the University of Chicago animal quarters. Immunization. Mice were injected intravenously with 109 heat-killed pneumococci in 0.2 ml of 0.15 M NaCl or with 0.2 ml of 20% suspension of sheep erythrocytes in 0.15 M NaCl. A drop containing 106 heat-killed pneumococci in minimal essential medium or a drop of 1.25% suspension in sheep erythrocytes in the same medium was added to spleen-cell cultures (107 cells per ml). Spleen-Cell Cultures. The preparation of spleen-cell cultures has been described (11). Briefly, suspensions of dissociated cells were prepared by gently forcing the spleens through a stainless steel screen. The suspension of single cells was washed in cold Hank's balanced salt solution before addition of monolayer-type minimal essential medium consisting of Hank's balanced salt solution supplemented with sodium pyruvate, amino acids, vitamin, glutamine, and 10% fetalcalf serum (12) (Reheis, lot no. H72208). Cultures of 107 cells suspended in 1.0 ml of medium in plastic dishes (Falcon, tissue culture grade, no. 3001) were incubated at 370 in an atmosphere of 7% 02-10% C02-83% N2; during incubation the cultures were on a rocking platform oscillating at 9 cycles/min. A drop of a nutritional mixture and a drop of fetal-calf serum were added daily to each culture (12). Hemolytic-Plaque Assay. Spleen cells synthesizing IgM hemolytic antibody were enumerated up to 5 days after immunization of mice or cultures by the hemolytic-plaque

2 2702 -Immunology: Cosenza and Kohier IN VIVO IN VITRO DAYS AFTER IMMUNIZATION FIG. 1. PFC response to phosphorylcholine. (a) Balb/c mice were immunized intravenously with 109 heat-killed pneumococci R36A; five mice were killed each day beginning at day 2 and up to 5 days after immunization. Each spleen was assayed separately for PFC against sheep erythrocytes diazotized with phosphorylcholine (@-) or pneumococcal C- (0----O). The number of plaques is given per 107 spleen cells plated, and it represents the average of plaques from five spleens (four slides per spleen). (b) Spleen cells from normal Balb/c mice were immunized in vitro with 106 heat-killed pneumococci R36A; the cultures were assayed for PFC up to day 5 after immunization against sheep erythrocytes treated as in (a). The number of plaques is given per 107 cells originally cultured, and it represents the average number of counts on six slides; duplicate slides from each of triplicate cultures were examined. technique (13), as modified for use with microscope slides (14). As target cells for detection of antibodies to phosphorylcholine, sheep erythrocytes were diazotized with phosphorylcholine or penumococcal C- (9). Preparation of Antisera. Balb/c mice bearing IgA plasmacell tumors TEPC-15 or MOPC-167 were given to us by Dr. Michael Potter. The myeloma proteins secreted by these tumors bind phosphorylcholine and also precipitate the C- of pneumococci R36A. An immunoabsorbent column (16) consisting of phosphorylcholine diazotized to sepharose beads was used to purify the myeloma proteins in their monomeric form. Antisera against TEPC 15 and MOPC 167 myeloma proteins were prepared in A/He mice as described by Potter and Lieberman (15). The antisera were heat-inactivated and absorbed with normal Balb/c serum. Each antiserum reacted only with the myeloma protein used for immunization. RESULTS Response of Balb/c mice or cultures of spleen cells to immunization with pneumococci R36A Mice were injected intravenously with pneumococci, the spleens were assayed for plaque-forming cells (PFC) with sheep erythrocytes either diazotized with phosphorylcholine or pneumococcal C- as target cells. The agar slides containing the cell suspensions were incubated in Hank's balanced salt solution for 2 hr before addition of a 1:25 dilution of guinea-pig complement. The results, Fig. la, show that PFC were detected in amounts above background by day 2; the PFC increased exponentially up to day 4, and decreased in number 5 days after immunization. Also, cultures of normal Balb/c spleen cells immunized with pneumococci were assayed 2-5 days after initiation of cultures. The response in vitro (Fig. lb) corresponds remarkably closely to the response in vivo, except that PFC decrease rapidly by day 5 due to deterioration of culture conditions. The kinetics and magnitude of the responses either in vivo or in vitro are equivalent whether sheep erythrocytes diazotized with phosphorylcholine or pneumococcal C- are used as target cells. Apparently, phosphorylcholine is the major antigenic determinant to which the antibodies against the pneumococcal antigen are directed (9), at least as measured here. For convenience, the antibody elicited by the pneumococci is referred to as antibody directed against phosphorylcholine even though sheep erythrocytes pneumococcal C- were used as target cells in the following experiments. Differences in the specific inhibition of plaque-formation by antibodies to TEPC 15 and MOPC 167 We have recently demonstrated that anti-idiotypic antiserum directed against TEPC 15 myeloma protein inhibited plaqueformation by cells from mice or cultures immunized with pneumococci R36A (9); thus, the primary response to phosphorylcholine expresses only the TEPC 15 idiotype. To demonstrate the absence of other idiotypic determinants in the primary response to phosphorylcholine, we examined the effect of antiserum against MOPC-167 on formation of plaques against phosphorylcholine. Though this myeloma protein precipitates with the pneumococcal C-, it does not share idiotypic determinants with TEPC 15 (15); furthermore, phosphorylcholine is not the complete antigenic determinant to which it is directed (16). As shown in the following experiment, antibody against MOPC 167 does not inhibit formation of IgM plaques against phosphorylcholine. 0 Co W Q ANTI-TEPC S a. Ui z25- Proc. Nat. Acad. Sci. USA 69 (1972) 0 PHOSPHORYLCHOLINE SRBC ANTISERUM DILUTION FIG. 2. Differences in the specific inhibition of formation of plaques against phosphorylcholine by antisera to TEPC 15 and MOPC 167. Spleen-cell suspension from Balb/c mice immunized intravenously with 109 heat-killed pneumococci R36A were assayed at day 4 for PFC against phosphorylcholine with sheep erythrocytes pneumococcal C- as target cells. The slides were incubated in dilutions of antiserum to TEPC 15 or anti-mopc 167 for 1. hr, washed in 0.15 M NaCl for 15 min, and incubated with guinea-pig complement for 1 hr. As control, spleen cells from Balb/c mice immunized with sheep erythrocytes 4 days previously were assayed for PFC against sheep erythrocytes in the presence of 1: 100 dilutions of antiserum to TEPC 15 or to MOPC 167. SRBC, sheep erythrocytes.

3 Proc. Nat. Acad. Sci. USA 69 (1972) Antibodies to Receptors 2703 Spleen-cell suspensions prepared from Balb/c mice immunized with pneumococci 4 days previously were assayed for PFC to phosphorylcholine; slides were incubated in dilutions of antiserum to TEPC 15 or MOPC 167 for 1 hr. The slides were then washed in 0.15 M NaCl for 15 min and incubated in guinea-pig complement for 1 hr. The antiserum against TEPC-15 completely prevented formation of plaques at a dilution of 1:100 and inhibited about 50% of plaques at a dilution of 1:10,000 (Fig. 2). The antiserum against MOPC 167 failed to inhibit formation of plaques even at low dilutions. Neither antiserum inhibited formation of plaques to sheep erythrocytes. Presumably, the antiserum to TEPC-15 suppresses plaque formation because it combines with the region of the combining site of antibody to phosphorylcholine. By inference, the antiserum to MOPC-167 is not directed against antibody reactive with phosphorylcholine. The following experiments report the effects of both antisera on induction of the antibody response to phosphorylcholine. Specific inhibition of induction of the antibody response to phosphorylcholine by antibody to TEPC 15 The effect of antibody to TEPC 15 or MOPC 167 on the response to phosphorylcholine was tested both in vitro and in vivo. For the first set of observations, cultures of normal Balb/c spleen cells were immunized at day 0 with either pneumococci or sheep erythrocytes (see Methods). 0.1 ml of a 1:10 dilution of antiserum to TEPC 15 was added to cultures at the time of immunization or 1, 2, or 3 days later. The same amount of antiserum to MOPC 167 was added to other TABLE 1. Specific inhibition of the primary PFC response to phosphorykcholine in vitro* Antiserum added to cultures at day Anti- Anti- MPOC TEPC PFC/ Target cell culture Pneumococci R36A Sheep erythrocytes , , , ,810-3,480 Sheep erythrocytes Sheep erythrocytes 0 8,090 Sheep erythrocytes 0 9,010 Sheep erythrocytes - 8,740 * Spleen cells from normal Balb/c mice were immunized in vitro with either 106 heat-killed pneumococci R36A or with 107 sheep erythrocytes at day 0. A/He antiserum against either MOPC 167 or TEPC 15 Balb/c myeloma protein was added to cultures (1:100 final dilution) at the times indicated. The cultures were disrupted at day 4 and assayed for PFC against untreated or sheep erythrocytes pneumococcal C-. The number of plaques is given per 107 spleen cells originally cultured and it represents the average of counts on 6 slides; duplicate slides from each of triplicate cultures were TABLE 2. Specific inhibition of the primary PFC response to phosphorylcholine in vivo* Balb/c mice PFC/spleen injected with: Target cell Exp 1. Exp 2. Pneumococci R36A A/He normal serum Sheep erythrocytes 139, ,000 A/He antiserum to Sheep erythrocytes 162, ,000 MOPC 167 A/He antiserum to Sheep erythrocytes 1, TEPC 15 Sheep erythrocytes A/He normal serum Sheep erythrocytes 164, ,000 A/He antiserum to Sheep erythrocytes 247, ,000 MOPC 167 A/He antiserum to Sheep erythrocytes 106, ,000 TEPC 15 * Balb/c mice (4 mice per group) were injected intraperitoneally with 0.3 ml of A/He normal serum, antiserum to MOPC 167, or antiserum to TEPC 15. The following day the animals were immunized intravenously with 109 pneumococci in 0.2 ml of 0.15 M NaCl or with 0.2 ml of a 20% sheep-erythrocyte suspension in 0.15 M NaCl; the spleens were individually assayed for PFC against either untreated or sheep erythrocytes pneumococcal C- at day 4 after immunization. The number of plaques is given per spleen, and it represents the average of four spleens (five slides per spleen). cultures at the time of immunization. The antisera were left in the culture dishes until they were assayed at day 4; the results (Table 1) show that antiserum to TEPC 15 completely suppressed the response when added at the time of immunization. Suppression resulted from inhibition of induction rather than from inhibition of plaque-formation, since antiserum added 1 or more days after immunization had no suppressive effect. Antiserum to MOPC 167 had no inhibitory effect on the antibody response to phosphorylcholine; neither antisera suppressed the response to sheep erythrocytes. Antiserum to TEPC 15 was similarly effective in suppressing the response to phosphorylcholine in vivo. Three groups of eight Balb/c mice each were injected intraperitoneally with 0.3 ml of antiserum to TEPC 15, to MOPC 167, or with normal A/He serum. The following day, four of the mice in each group were injected intravenously with pneumococci and the other four with sheep erythrocytes (Table 2). The PFC responses were measured 4 days after immunization. Treatment of mice with antiserum to TEPC 15 completely suppressed the response to phosphorylcholine but not to sheep erythrocytes; the antiserum to MOPC 167 had no suppressive effect on either response. DISCUSSION The present results indicate that the antibodies to TEPC 15 are directed against the combining-site region of antibody to phosphorylcholine; presumably, the antibodies to TEPC 15 inhibited plaque-formation by preventing the secreted antiexamined. bodies from binding to the target cells (sheep erythrocytes

4 2704 Immunology: Cosenza and Kohler pneumococcal C-). Furthermore, antibodies to TEPC 15 also seem to be specifically directed against the binding-site region of receptor molecules onantigensensitive cells, since they prevent induction of the antibody response only to phosphorylcholine. Thus, the cell receptor for phosphorylcholine shares similar antigen-binding structures with antibodies to phosphorylcholine and with TEPC-15 myeloma. Inhibition of induction of the response to phosphorylcholine is probably due to binding of the antibodies to TEPC 15 to the receptor molecules preventing stimulation by the pneumococcal antigen. Thus, we believe that these observations permit us to conclude that the antiserum to TEPC 15 contains "anti-receptor" antibodies. Cell receptors are thought to be of IgM or of IgG classes (3, 5), possibly sharing the K class of light chain (6). Thus, the combining-site region of the receptor for phosphorylcholine, regardless of its class of heavy and light chains, must be similar to the combining site of the IgM antibody synthesized in the primary response to phosphorylcholine and of the TEPC-15 IgA myeloma. The general implication is that antibodies and antigen receptors belonging to different immunoglobulin classes but having specificity for the same antigen may share similar binding-site regions encoded by identical variable genes (17). The complete inhibition of the antibody response to phosphorylcholine by antibodies to TEPC 15 also implies that, in an individual, there are clones of cells bearing receptors of restricted specificity for an antigen, since antibodies directed against the receptor molecules for phosphorylcholine do not affect the antibody response to unrelated antigens, i.e., sheep erythrocytes. Thus, the receptor-bearing cell, as well as the antibody synthesized by the antibody-producing cell (18), may interact with only a single antigen. Our results also indicate that "triggering" of the cell receptor by antigen takes place shortly after immunization, since addition of antibodies to TEPC 15 1 day after immunization of cultures no longer prevents induction of the response to phosphorylcholine. Thus, "anti-receptor" antibodies do not interfere with proliferation of antibodyproducing cells once the precursor cells have been stimulated by antigen. Furthermore, we have observed that to inhibit induction of the response, the antibodies to TEPC 15 must be present in cultures during the entire incubation period. Thus, when suspensions of spleen cells were incubated with antiserum to TEPC 15 for 3 hr and the unbound "antireceptor" antibody was eliminated by thoroughly washing the cells with Hank's balanced salt solution, a normal PFC response to phosphorylcholine resulted upon immunization of cultures with pneumococci. It appears, then, that there is a fairly rapid turnover of receptor molecules on the surface of immunocompetent cells (6); as a result, the antibodies bound to the receptor molecules are insufficient to block the interaction between newly synthesized receptors and antigen. The present results demonstrate that "anti-receptor" antibodies specifically suppressed the immune response in vivo; it will be important now to determine whether the titer of such antibodies ever increase in the normal course of immunization. We have recently observed significant suppression of both Balb/c and A/He mice given a single injection of antibodies to TEPC 15 up to 1 month before challenge with pneumococci. It is possible, then, that "anti-receptor" antibodies may be generally used to specifically suppress the Proc. Nat. Acad. Sci. USA 69 (1972) antibody response to one antigen without inhibiting the immune response to other antigens (19). Furthermore, since A/He mice produce antibodies to TEPC-15 myeloma, their reactive cells must have receptors for the combining site of the myeloma antibodies. Recently, it has been shown that Balb/c mice can synthesize anti-idiotypic antibodies to other Balb/c myeloma proteins with antibody activity (20, 21). If cells within an individual have receptors for the combining-site region of antibody produced by other cells in the same individual, then the individual must have the capability of synthesizing "anti-receptor" antibody. Such "anti-receptor" antibody could act specifically along with antibody to the antigen for the regulation of immune responses. This work was reported in part at the Federation of American Societies for Experimental Biology, Atlantic City, N.J., April 9-14, We thank Dr. Donald A. Rowley for his generous support and advice. The expert technical assistance of Miss Helga Tremmel and Miss Ingrid Nebl is gratefully acknowledged. This work was supported by ACS Grant IC-21 and by USPHS Grants AI-10242, AI-09268, and AI Pernis, B., Forni, L. & Amanto, L. (1970) "Immunoglobulin spots on the surface of rabbit lymphocytes," J. Exp. Med. 132, Sell, S. & An, T. (1971) "Studies of rabbit lymphocytes in vitro. XIV. Fractionation of rabbit peripheral blood lymphocytes by antibody-coated polyacrylamide beads," J. Immunol. 107, Vitetta, E. S., Baur, S. & Uhr, J. S (1971) "Cell surface immunoglobulin. II. Isolation and characterization of immunoglobulin from mouse splenic lymphocytes," J. Exp. Med. 134, Sherr, C. J., Baur, S., Grundke, I., Zeligs, J., Zeligs, B. & Uhr, J. W. (1972) "Cell surface immunoglobulin. III. Isolation and characterization of immunoglobulin from nonsecretory human lymphoid cells, J. Exp. Med. 135, Manning, D. D. & Jutila, J. W. (1972) "Immunosuppression of mice injected with heterologous anti-immunoglobulin heavy chain antisera," J. Exp. Med. 135, Lesly, J. & Dutton, R. W. (1970) "Antigen receptor molecules: inhibition by antiserum against kappa light chains," Science 169, Ada, G. L. & Byrt, P. (1969) "Specific inactivation of antigen reactive cells with "2'I-labelled antigen," Nature 222, Makela, 0. (1970) "Analogies between lymphocyte receptors and the resulting humoral antibodies," Transplant. Rev. 5, Cosenza, H. & Kohler, H. (1972) "Specific inhibition of plaque formation to phosphorylcholine by antibody against antibody," Science 176, Cosenza, H. (1972) "Anti-receptor antibodies. II. Specific inhibition of induction of an antibody response to a pneumococcal antigen," Fed. Proc. 31, Cosenza, H., Leserman, L. D & Rowley, D. A. (1971) "The third cell type required for the immune response of spleen cells in vitro," J. Immunol. 107, Mishell, R. & Dutton, R. W. (1966) "Immunization of normal mouse spleen cell suspensions in vitro," Science 153, Jerne, N. K. & Nordin, A. A. (1963) "Plaque formation in agar by single antibody-producing cells," Science 140, Plotz, P. H., Talal, N. & Asofsky, R. (1968) "Assignment of direct and facilitated hemolytic plaques in mice to specific immunoglobulin classes, J. Immunol. 100, Potter, M. & Lieberman, R. (1970) "Common individual antigenic determinants in five of eight Balb/c IgH myeloma proteins that bind phosphorylcholine," J. Exp. Med. 132,

5 Proc. Nat. Acad. Sci. USA 69 (1972) 16. Chesebro, B. & Metzger, H. (1972) "Affinity labeling of a phosphorylcholine binding mouse myeloma protein," Biochemistry 11, Kuhler, H., Shimizu, A., Paul, C, Moore, V. & Putnam, F. W. (1970) "Three variable-gene pools common to IgM, IgG and IgA immunoglobulins," Nature 227, Makela, 0. (1967) "The specificity of antibodies produced by single cells," Cold Spring Harbor Symp. Quant. Biol. 32, Hart, D. A., Wang, A. L., Pawlak, L. L. & Nisonoff, A. Antibodies to Receptors 2705 (1972) "Suppression of idiotypic specificities in adult mice by administration of antiidiotypic antibody," J. Exp. Med. 135, Sirisinha, S. & Eisen, H. N. (1971) "Autoimmune-like antibodies to the ligand-binding sites of myeloma proteins," Proc. Nat. Acad. Sci. USA 68, Yakulis, V., Bhoopalam, N. & Heller, P. (1972) "The production of anti-idiotypic antibodies to Balb/c plasmacytoma globulins in Balb/c mice," J. Immunol. 108,