Interleukin 2 regulates antigen-specific immuneglobulin response of naive murine B cells

CELLULAR

IMMUNOLOGY

120,366-374 (1989)

lnterleukin 2 Regulates Antigen-Specific lmmunoglobulin of Naive Murine B Cells’ NOAH ISAKOV*,* AND PHILLIP

Response

R. MORROW~-,~

*Department ofMicrobiology and Immunology, Ben-Gurion University of the Negev, Beer Sheva 84105, Israel; and TDepartment ofImmunology, Research Institute ofscripps Clinic, La Jolla, California 92037 Received November 10,1988; accepted January 9,1989

Activation of mouse B cells with lipopolysaccharide in conjunction with anti-immunoglobulin (Ig) antibodies results in interleukin 2 (IL2) receptor (IL2-R) expression and IL2 responsiveness. In most studies on the effect of IL2 on antibody production by B cells, polyclonally activated normal B cells or B cell lines established in vitro have been used as indicator cells, thus allowing no direct correlation between the experimental findings and the actual physiological mechanism of IL2 action in antigen-specific B cell response.We employed the splenic fragment culture technique, which measures antibody response on the clonal level, to analyze the effect of purified human recombinant IL2 (rIL2) on the primary antigen-specific Ig responseof mouse B cells. Here we report that rIL2 increased the frequency of dinitrophenyl (DNP)-responsive splenic B cells and the amount of Ig secreted per clone. The anti-DNP antibody response was dependent upon interaction of naive B cells with carrier-primed T cells, which apparently provided the signal for IL2-R expression. Recombinant IL2 also facilitated Ig isotype switching by individual clones, suggestinga role for IL2 in activation, maturation, and differentiation ofantigen-specific naive B cells in their responseto T-dependent antigens. o 1989 Academic PWS, I~C.

INTRODUCTION Induction of antigen-specific immunoglobulin (Ig)4 secretion and regulation of Ig class switching involve interaction of B cells with antigen as well as T cells and their soluble mediators ( l-4). The T cells produce a variety of peptide hormones (lymphokines) that affect B cells in two major ways: (a) activation and induction of proliferation, and (b) differentiation and maturation into Ig-secreting cells. Many of the T cell-derived factors have pleotropic effects on B cells and they affect also other cell types such as monocytes, thymocytes, eosinophils, and mast cells (reviewed in Ref. (4)). Interleukin 2 (IL2), originally termed “T cell growth factor,” has for a long time been considered as a unique growth factor for activated T cells (5). However, several ’ The work reported herein was supported, in part, by grants from the Israel Academy of Sciencesand Humanities, the Israel Cancer Association, the Chief Scientist of the Ministry of Health, the Israel Cancer ResearchFund, and NIH Grants CA-35299 and AR-3541 1. * Special Fellow of the Leukemia Society of America, Inc., and recipient of a Yigal Alon Fellowship. 3 Present address:Cytotech, Inc., I 1045 Roselle Street, Suite C, San Diego, CA 92 121. 4 Abbreviations used: BSA, bovine serum albumin; DNP, dinitrophenyl; EIA, enzyme-linked immunosorbent assay; Hy, hemocyanin; Ig, immunoglobulin; IL2, interleukin 2; IL2-R, IL2 receptor; PBS, phosphate-buffered saline; rIL2, human recombinant IL2. 366 0008-8749189$3.00 Copyright 0 1989 by Academic Press, Inc. All rights of reproduction in any form reserved.

REGULATION

OF B CELLS BY INTERLEUKIN

2

367

research groups have recently shown that certain B cell tumors, as well as normal activated B cells, express receptor molecules that are similar in nature to the IL2 receptors (IL2-R) expressed on T cells (6-l 3). Furthermore, IL2-R-positive B cells could respond to IL2 by proliferation (7,8, 14- 16) and antibody synthesis and secretion (15, 17-21). Thus, murine B cells activated with lipopolysaccharide (7, 11) or human B cells activated with Staphylococcus aureus Cowan I, with or without the appropriate anti-Ig antibodies (8, 15, 19, 20) expressedIL2-R and responded to IL2 by proliferation and secretion of polyclonal antibodies. These and other studies have established that B cells possessthe potential to respond to IL2. The actual physiological role of IL2 in induction of antigen-specific primary B cell responsesin vivo is still controversial because many studies employed transformed B cell lines (12, 14) or polyclonally activated heterogeneous populations of B cells (8, 11, 13, 15, 19) to study the effect of IL2 on antigen nonspecific Ig production. Here, we approximate biological conditions to demonstrate that IL2, prepared by recombinant DNA technology (rIL2), augments the antigen-specific antibody response of naive murine B cells to a T-dependent antigen. Furthermore, analysis of the effect of rIL2 on individual B cell clones indicates that rIL2 facilitates anti-dinitrophyenyl (DNP) Ig class switching. MATERIALS AND METHODS Mice. Two-month-old BALB/c mice were obtained from the breeding colonies at the Ben-Gurion University of the Negev and the Scripps Clinic and Research Foundation. Antigens and rIL2. DNP (Sigma) was conjugated to Limuluspolyphemus hemocyanin (Hy; Worthington Biochemical Corp., Freehold, NJ) and bovine serum albumin (BSA; Sigma, St. Louis, MO) as described (22). Recombinant IL2, supplied by Dr. A. Altman, was a gift from AMGen (Thousand Oaks, CA). This IL2 preparation had a specific activity of 20 U/rig when compared to a Jurkat-derived reference human IL2 (provided by the Bilogical Response Modifiers Program, Division of Cancer Treatment, National Cancer Institute) containing 500 U/ml of IL2 at a specific activity of 13.1 U/rig. IL2 titration was performed on HT2 indicator cells as described (23). T cellproliferation assay. BALB/c mice were immunized intraperitoneally or intrafootpad with 100 pg Hy in complete Freund’s adjuvant. Half of the mice were irradiated (1300 rad using a cesium source) 8 days later, and spleens or draining popliteal lymph nodes were removed after an additional 24 hr. Cells were cultured in 96-well tissue culture plates (Costar, Cambridge, MA) at 4 X lo5 nucleated cells per well in 200 ~1complete RPM1 [RPM1 1640 (M. A. Bioproducts, Walkersville, MD) supplemented with 10%heat-inactivated fetal calf serum, 200 mM L-glutamine (both from Flow Laboratories, McLean, VA), 5 X lop5 2-mercaptoethanol (Sigma), and 50 fig/ ml gentamicin (U.S. Biochemical Corp., Cleveland, OH)]. One microcurie of [3H]thymidine ([3H]TdR; 6.7 Ci/mmol; ICN Radiochemicals, Irvine, CA) was added during the last 2 hr of culture. Plates were harvested on a Skatron cell harvester and radioactivity was determined by liquid scintilation spectroscopy. Immunization, cell transfer, and spleenfragment culture. Carrier-primed BALB/c recipient mice were prepared as described (24). Briefly, mice were immunized intraperitoneally with 100 pg of Hy in complete Freund’s adjuvant, followed 4 weeks later by an intraperitoneal injection of 100 pg Hy in saline. These mice were used as

368

ISAKOV AND MORROW

recipients 6-8 weeks after the second immunization. A single cell suspension of naive donor spleen cells, prepared by splenic perfusion with Dulbecco’s modified Eagle’s medium, was injected intravenously into the carrier-primed recipients (4 X 1O6cells/ mouse) that had been irradiated (1300 rad) 4 hr earlier. The splenic fragment culture was performed as described (22, 25, 26). Fragments were prepared from recipients’ spleens 18 hr after cell transfer and were individually stimulated with DNPz6-Hy at a hapten concentration of 1O-6A4 for 3 days. Culture fluids were changed every 3-4 days and collected individually beginning on Day 10. Recombinant IL2 (5 U/ml) was added to cultures on Day 3 and on any subsequent feeding with fresh medium. In experiments where T cell-depleted naive spleen cells have been used for in vivo reconstitution of mice, the donor spleen cells have been treated with the monoclonal antibody J lj [rat anti-Thy- 1.2, IgM (27)] + guinea pig complement for 1 hr at 37°C. Detection of secreted antibodies. An enzyme-linked immunosorbent assay (EIA) was used to detect DNP-specific antibody and to determine the frequency of positive wells and the amount and isotype of secreted Ig. Polyvinyl 96-well microtiter plates (Dynatech Laboratories, Inc., Alexandria, VA) were coated with DNP,,-BSA, and then washed and blocked with a 0.5% BSA buffer solution. Supernatants from individual spleen fragment-containing wells were added to the antigen-coated wells and incubated for 1 hr at room temperature. After three washeswith phosphate-buffered saline (PBS) solution containing 0.05% Tween 20, appropriately diluted rabbit antimouse IgM + IgA + IgG (Zymed Laboratories, Inc., San Francisco, CA) was added to the plates. Plates were incubated for 1 hr at room temperature, washed three times with PBS/Tween 20, and supplied with a 1:2000 dilution of alkaline phosphataseconjugated goat anti-rabbit IgG (Tago, Inc., Burlingame, CA). After a 1-hr incubation at room temperature, the plates were washed again, and 50 ~1of 1 mg/ml p-nitrophenyl phosphate (Sigma) was added to incubate for an additional hour, followed by 50 ~1 of 3 A4 NaOH to terminate the reaction. Converted substrate was measured photometrically at 405 nm in a microplate reader (Dynatec Model MR600). To determine Ig isotypes, supernatants collected from individual positive wells at different time points were pooled and retested in duplicate on three separateantigencoated plates. These plates were than developed with one of three rabbit antisera specific for mouse IgM, IgA, or IgG (Tago, Inc.). For all assays,standard curves were generated by using known quantities of monoclonal anti-DNP antibodies of the IgM, IgA, and IgG isotypes. RESULTS AND DISCUSSION The effect of rIL2 on splenic DNP-reactive B cell clones was analyzed by using the splenic fragment culture technique (22, 25, 26). Splenic fragments from Hy-primed recipient mice that had been irradiated and then reconstituted with naive spleen cells were cultured in the presence of DNP-Hy, and half the cultures were supplemented with rIL2. The Ig responseof naive B cells in the splenic fragment culture system is dependent on the presenceof, and interaction with, carrier-primed helper T cells (22). It is possible, therefore, that exogenous IL2 added to the cultures does not affect B cells directly, but rather stimulates the Hy-primed T cells to produce B cell-specific factors which, in turn, activate B cells. To eliminate this possibility, rIL2 was added to cultures only 3 days after their initiation. Under these conditions, most of the lethally irradiated

REGULATION

OF B CELLS BY INTERLEURIN

369

2

TABLE 1 Effect of Lethal Irradiation (1300 rad) on the in Vitro Proliferative Response of Hy-Primed T Lymphocytes” [‘H]TdR uptake in responseto Irradiation

Day

None

HY

Con A

-

1

4234 1448

13,110 2,314

2

3428 218 2120 232

42,563 1,162 56,847 546 61,916 268

6,754 1,134 39,494 478 67,836 406 38,9 10 244

+ + + +

3 4

1574 212

’ BALB/c mice were immunized with Hy in complete Freund’s adjuvant either intraperitoneally or intrafootpad and 8 days later half of the mice were irradiated at 1300 rad. Spleens and draining popliteal lymph nodes were removed after 24 hr and cells from each organ were cultured in four identical sets of microplates in triplicate wells in the presence of medium, the immunizing antigen (Hy, 12.5 &ml), or Con A (2.5 &ml). The four plates were used to measure t3H]TdR uptake into cells on 4 consecutive days, starting on Day 1 after culture initiation. [3H]TdR was added to cultures 2 hr before harvesting. Results demonstrate [3H]TdR incorporation into lymph node cells. Similar results were obtained when primed spleen cells were stimulated with Con A but only a marginal Hy-specific responsewas observed.

carrier-primed T cells that were restimulated in vitro with the priming carrier antigen die in culture within 24-48 hr (28) because, after 2 days, they do not respond to the antigen (Hy) or a T cell specific mitogen (Con A) by either proliferation (Table 1) or IL2 production (Table 2). Because the IL2-indicator cells (HT2) also proliferate in responseto IL4 (29), we assumethat neither IL2 nor IL4 is produced by the irradiated carrier-primed T cells. TABLE 2 Effect of Lethal Irradiation (1300 rad) on IL2 Production by Hy-Primed T Lymphocytes” IL2 production in responseto Irradiation

Day

None

HY

-

1

+ +

2


14 6 3
+

3

tl <1

” Lymph node cells from irradiated (1300 rad) and nonirradiated spleen cells from Hy-primed BALB/c mice were prepared and cultured as described in the legend to Table 1. Supematants were collected 24,48, or 72 hr after culture initiation. In the two later time points, culture medium was replaced by fresh medium 24 hr prior to collection of the supematants. Titration of IL2 was performed in a bioassay using HT2 as indicator cells.

370

ISAKOV AND MORROW

r

1 :_

l-

I-

rlL2:

-

+

Exp 1

I ~

Exp 2

FIG. 1. Recombinant IL2 increases the number of DNP-responsive clones in a splenic fragment assay. Splenic fragments were cultured in the presence of DNP-Hy with or without rIL2 (5 U/ml; 192 wells/ group) that was added to half of the cultures on Day 3. Supernatants were collected on Days 10, 13, 17, and 20, and tested separately with an EIA on DNP-BSA coated plates. Spleen fragments were considered positive for anti-DNP antibody production if a specific response was obtained in supernatants of at least two consecutive collections.

The number of B cells that were transferred into the Hy-primed mice was predetermined by limiting dilution and selecting a transfer dose that yielded 5- 10% positive clones. In three separate experiments (Fig. 1) in which spleen fragments were stimulated in vitro with DNP-Hy, we found that the presence of rIL2 in the culture medium increased the number of positive clones by 2. I-, 2.5-, or 2.7-fold, respectively. Thus, positive clones were obtained in 74 of 576 (12.85%) rIL2-containing cultures, whereas only 3 1 of 576 (5.38%) cultures were positive when rIL2 was absent. To address the possibility that rIL2 stimulates naive donor T cells to produce factors which, in turn, affect B cell responsesto DNP-Hy, we performed an experiment in which anti-Thy-l antibody + complement-treated naive spleen cells were transferred into the Hy-primed recipient mice. We found that this depletion of T cells from spleen cell transfers did not abolish the increased frequency of positive B cell clones obtained in the presence of rIL2 (13.0% and 5.2% of the spleen fragments responded to DNP when cultured in the presence or absence of rIL2, respectively, from a total of 192 spleen fragments per group). The possibility that IL2 affects B cehs indirectly by inducing B cell factor production by non-T accessorycells or even B cells is unlikely, but cannot be completely ruled out. The presenceof such lymphokines may be difficult to detect if, for example, they are produced by B cells and act as aatocrine growth factors (30, 3 1) or if they function in a membrane-associated nonsecreted form such as interleukin 1 (32). We then compared the amount of antibodies produced by individual clones in the two experimental groups. As shown in Table 3, the DNP-reactive clones from rIL2-

REGULATION

OF B CELLS BY INTERLEUIUN

371

2

TABLE 3 Effect of IL2 on the Quantity of Anti-DNP IgM Antibody Produced by Individual Clones” Number of clones tested

Expt number

rIL2

1

-

10

+

23

224 + 12 333 t 31

-

10

361 +_54

+

20

553 f 43

2

IgM (w/ml)

a Cultured supernatants collected at four time points from individual wells that were found to contain anti-DNP antibodies were combined, and the total amount of IgM antibody produced in each culture was determined.

containing cultures produced an average of 443 rig/ml IgM, while clones from cultures lacking rIL2 produced an average of 292 rig/ml IgM. Thus, the rIL2-mediated increase in Ig production may reflect either an increase replication rate of DNP-responsive precursor B cells or increased antibody production, on a per cell basis. We attempted to measure [3H]TdR uptake by proliferating cells in a whole spleen fragment; however, this was not feasible because only a few cells responded, and even under optimal conditions (in the presence of antigen and IL2), their response was at the background level. A primary immune response is characterized in its initial phase by production of mostly antigen-specific IgM antibodies, and only after further differentiation of the plasma cells does the Ig isotype switch occur. To test whether rIL2 affects B cell differentiation in a primary immune response, we determined the Ig isotype in the splenic fragment cultures lacking or containing rIL2 by using isotype-specific rabbit anti-mouse Ig sera. The results (Fig. 2) demonstrated that a high percentage of clones

Number of Clones Secreting AntiJlNP Antibodies ---!J

a

Y

10

20

+ + + +

+

+

+ + +

+++ P

-rlLZ

b

+rlL2

FIG. 2. Isotype analysis of DNP-specific antibodies in spleen fragments cultured with or withont rIL2. Culture supernatants from DNP-responsive clones were tested with an EIA on three setsof plates that were developed with a rabbit antiserum specific for mouse I&I, IgG, or IgA. The data represent the number of clones producing each isotype or combination of isotypes.

372

ISAKOV AND MORROW TABLE 4 Effect of IL2 on lsotype Switching of DNP-Specific B Cell Clones Ig isotype’ rIL2

+

Number of clones analyzed

kM

IgA

W

31 74

90.3 86.5

25.8 41.9

3.2 14.9

’ Numbers represent the percentage of clones producing the isotype indicated.

in rIL2-containing or control cultures produce only anti-DNP IgM antibodies. However, a larger proportion of the rIL2-containing clones switched classesand produced two or more isotypes. Thus, 27 of 74 clones (36.4%) in the rIL2-containing cultures secretedmore than one Ig isotype compared with 5 of 3 1 clones (16.1%) in the control cultures (Table 4). In addition, 4 1.9 and 14.9%of positive clones in the rIL2-containing cultures produced IgA and IgG antibodies, respectively, compared with only 25.8 and 3.2%, respectively, in cultures lacking rIL2 (Table 4). The assumption that IL2 may signal B cells to differentiate is supported also by previous findings in different experimental systems (17-2 1). Pike et al. have demonstrated that rIL2 synergizes with the “T cell-independent” antigens fluorescein-conjugated polymerized flagellin or fluorescein-Ficoll in promoting proliferation and differentiation of hapten-specific murine B cells (33). In the murine BCL, B cell line, IL2 did not affect proliferation or p heavy chain gene transcription but induced transcription and production of the J chain protein, thus enabling assembly of the IgM pentamer (34). To test whether the presence of rIL2 in the splenic fragment cultures could overcome the need for carrier-primed T cells, we performed an experiment in which nonimmune mice were used as recipients of the naive splenic B cells. Under these conditions, no antibodies were detectable in the DNP-Hy-primed cultures even in the presence of rIL2 (not shown). Apparently, the carrier-primed T cells provide B cells with an essential signal(s) that cannot be mimicked or replaced by rIL2 alone. We postulate that interaction with the Hy-primed T cells results in IL2-R expression on the naive B cells, rendering them IL2 responsive. This hypothesis is supported by recent findings that selected mutagenized sublines of the murine EL4 thymoma can induce IL2-R and IL2 responsivenessin both murine and human B cells (35). Since the frequency of naive DNP-specific splenic B cells is predetermined, we assume that the rIL2-mediated increase in the number of anti-DNP responsive wells results from the increased amount of anti-DNP antibodies produced per cell and/or IL2-mediated expansion of anti-DNP antibody-producing cells. Thus, cultures that, in the absence of IL2, do not respond, or produce amounts of antibodies below the sensitivity threshold of the EIA, should be detectable in the presenceof rIL2. Alternatively, binding of rIL2 to its receptor on B cells may (a) increase the number and/or affinity of the antigen-specific cell surface Ig, thereby amplifying the activation signal and the resulting response, or (b) induce growth factor production and activation via an autocrine pathway. Unfortunately, the small number of B cells within splenic fragments makes it very difficult to purify a sufficient number of B cells for more direct and detailed analyses.

REGULATION

OF B CELLS

BY INTERLEUIUN

2

373

Effects of IL2 on antigen-specific B cell responsesare difficult to analyze in bulk cultures because of the low frequency of B cells possessingthe specific cell surface receptors for a given antigen. As a result, polyclonal B cell activators have been used in the past to demonstrate the effect of IL2 on antigen nonspecific Ig responses.Although these experiments demonstrated that B cells possessthe potential to express IL2-R and respond to IL2, their relevance to the physiological immune response is questionable. The advantage of the splenic fragment culture system is that it provides a physiological microenvironment and enables selection and analysis of antigen-specific B cell cultures that contain monoclonal cell progeny. Furthermore, it allows analysis of antibody switching on the level of a single cell, thus allowing clonal studies on B cell maturation and differentiation. Our findings suggest a role for IL2 in a primary antigen-specific Ig response in which interaction of antigen-stimulated B cells with the carrier-primed helper T cells presumably results in IL2-R expression on B cells. Binding of IL2 to such receptors provides further activation stimuli to the B cells, resulting in enhanced proliferation rates and/or increased antibody production and enhanced Ig isotype switching. ACKNOWLEDGMENTS We thank Dr. M. Goodman for reviewing the manuscript, Drs. L. Souza and B. Altrock for providing the rIL2, P. Minick, M. K. Occhipinti, and C. Simon for editorial assistance,and Dr. A. Altman for his continuous support.

REFERENCES 1. Moller, G., Ed., Immunol. Rev. 78, 1984. 2. Melchers, F., and Andersson, J., Annu. Rev. Irnmunol. 4, 13, 1986. 3. Hamaoka, T., andOno, S., Annu. Rev. Immunol. 4, 167, 1986. 4. O’Garra, A., Umland, S., DeFrance, T., and Christiansen, J., Immunol. Today 2,45, 1988. 5. Smith, K. A., and Ruscetti, F. W., Adv. Immunol. 31, 137, 1981. 6. Malek, T. R., Robb, R. J., and Shevach, E. M., Proc. Natl. Acad. Sci. USA 80,5694, 1983. 7. Zubler, R. H., Lowenthal, J. W., Erard, F., Hashimoto, N., Devos, R., and MacDonald, H. R., J. Exa. Med. 160,1170, 1984. 8. Mingari, M. C., Gerosa, F., Carra, G., Acolla, R. S., Moretta, A., Zubler, T. H., Waldman, T. A., and Moretta, L., Nature (London) 312,64 1, 1984. 9. Osawa, H., and Diamanstein, T., J. Zmmunol. 132,2445, 1984. 10. Tsudo, M., Uchiyama, T., and Uchino, H., J. Exp. Med. 160,612, 1984. 1I. Lowenthal, J. W., Zubler, R. H., Nabholz, M., and MacDonald, H. R., Nature (London) 315, 669, 1985.

12. Waldmann, T. A., Goldman, C. K., Robb, R. J., Depper, J. M., Leonard, W. J., Sharrow, S. O., Bongiovanni, K. F., Korsmeyer, S. J., and Greene, W. C., J. Exp. Med. 160,1450, 1984. 13. Jung, L. K. L., Hara, T., and Fu, S. M., J. Exp. Med. 160, 1597, 1984. 14. Lantz, O., Grillot-Courvalin, C., Schmitt, C., Fermand, J.-P., and Brout, J.-C., J. Exp. Med. 161, 1225, 1985. 15. Jelinek, D. F., Splawski, J. B., and Lipsky, P. E., Eur. J. Immunol. 16,925, 1986. 16. Mittler, R., Rao, P., Olini, G., Westberg, E., Newman, W., Hoffman, M., and Goldstein, G., J. Immunol. 134,2393, 1985. 17. Teranishi, T., Hirano, T., Lin, B.-H., and Onoue, K., J. Immunol. 133,3062, 1983. 18. Nakanishi, K., Malek, T., Smith, K., Hamaoka, T., Shevach, E., and Paul, W. E., J. Exp. Med. 160, 1605,1984. 19. Ralph, P., Jeong, G., Welte, K., Mertelsman, R., Rabin, H., Henderson, L. E., Souza, L. M., Boone, T. C., and Robb, R. J., J. Immunol. 133,2442, 1984. 20. Nakagawa, T., Hirano, T., Nakagawa, N., Yoshizaki, K., and Kishimoto, T., J. Zmmunol. 134,959, 1985.

374

ISAKOV AND MORROW

21. Kishi, H., Inui, S., Muraguchi, A., Hirano, T., Yamamura, Y., and Kishimoto, T., J. Immunol. 134, 3104, 1985. 22. Klinman, N. R., J. Exp. Med. 136,241, 1972. 23. Gillis, S., Ferm, M. M., Ou, W., and Smith, K. A., J. Zmmuno/. 120,2027, 1978. 24. Pierce, S. K., and Klinman, N. R., J. Exp. Med. 146,509, 1977.

25. Klinman, N. R., and Press,J., Transplant. Rev. 24,41, 1975. 26. Accolla, R. S., Gearhart, P. J., Sigal, N. H., Cancro, M. P., and Klinman, N. R., Eur. J. Immunol. 7, 876,1977. 27. Bruce, J., Symington, F. W., McKearn, T. J., and Sprent, J., J. Immunol. 127,2496, 198 1.

28. Anderson, R. E., and Warner, N. L., Adv. Immunol. 24,2 15,1977. 29. Mosman, T. R., Cherwinski, H., Bond, M. W., Giedlin, M. A., and Coffman, R. L., J. Zmmunol. 136, 2348, 1986. 30. Scala, G., Morrone, G., Tamburrini, M., Alfinito, F., Pastore, C. I., D’Alessio, G., and Venuta, S., J.

Immunol. 138,2527,1987. 31. Kupper, T., Horowitz, M., Lee, F., Robb, R., and Flood, P. M., J. Zrnmunol. 138,4280, 1987. 32. Kurt-Jones, E. A., Beller, D. I., Mizel, S. B., and Unanue, E. R., Proc. Natl. Acad. Sci. USA 82, 1204, 1985. 33. Pike, B. L., Raubitschek, A., and Nossal, G. J. V., Proc. Natl. Acad. Sci. USA S&7917,

1984.

34. Blackman, M. A., Tiges, M. A., Minie, M. E., and Koshland, M. E., Cell47,609, 1986. 35. Zubler, R. H., Erard, F., Lees, R. K., VanLaer, M., Mingari, C., Moretta, L., and MacDonald, H. R., J. Immunol. 134,3662, 1985.