* D. R. G. is supported by NIH Training Grant AI07019

When immunologic homeostasis is perturbed by antigen, the failure to accomodate that antigen as "self" usually results in what we observe as an immune response. In the case of tumors, however, a massive bombardment of the system with tumor antigens can induce a potent general suppression of immunity such that the tumor must often be removed before immunity can be demonstrated [ I ]. Such observations suggest that while there may be antigenic determinants on tumor cells which can serve as targets of immunity, immunoregulatory modification might be necessary for such immunity to become manifest. Thus, immune suppression induced by tumor challenge is probably the major stumbling block to effective immunity against many tumors. In terms of therapy, the activity of the suppressor circuit might serve as a target of effective imm unengineering. Alternatively, we can envision situations in which tumors of the lymphoid system may come under suppressor cell control and thus be rendered benign. For example, Rohrer and Lynch [2] have demonstrated control by suppressor T cells ofMOPC-3l5 myeloma clone growth and secretion. similar effects have been obtained by Abbas et al. [3]. Suppressor T cells appear in normal people infected with Epstein-Barr virus (EBV) [4], and such T cells have been shown to be capable of inhibiting in vitro transformation of B cells by EBV [5]. In some cases, therefore, failure to effect suppression of a proliferating cell may be a cause of cancer, so that therapy must then be aimed at enhancing suppressor cell activity. Such tumors, while rare, may be important for our development of effective tumor therapy. The ability to up or down regulate immune responses is likely to be a key factor in cancer therapy. While the role of suppressor cells in cancer is an active area of research, little is known about the role of the cells that mediate contrasuppression. Contrasuppression is an immunoregulatory T cell activity which is defined functionally as the ability to interfere with suppressor cell signals. Contrasuppressor effector cells have been shown to express a unique profile of cell surface antigens and to function, at least in part, by rendering helper T cells resistant to suppressor cell signals. Clearly, such an activity could have major consequences for our understanding and control of cancer. In this briefpaper we will discuss the evidence that contrasuppressor T cells have an active role in the immune response to cancer. This will lead us into a hypothetical consideration of the role of class I antigens in the activation of regulatory T cells and the consequences of this theory for immunomodulation and therapy. Finally we will review evidence for the possibility that in some cases, involving transformed cells of the imm une system, this regula tory activity might enhance tumor incidence by interfering with the suppressor cells capable of controlling tumor growth.

A. Immune Sequelae to the Activation of Contrasuppression

Relatively little is known about the nature of the signals which initiate contrasup pression, but the dose of immunizing antigen is certainly a key factor. Contrasuppression seems to be induced at doses of antigen optimal for immune responses ([6, 7], T. Lehner, personal communication). Certain antigen-presenting cells, such as Langerhans cells, dendritic cells, and peritoneal exudate macrophages induced by complete (but not incomplete) Freund's adjuvant, preferentially activate the cells of this circuit [8, 9]. Other factors involved in activation have been reviewed elsewhere [ 10, 11 ]. Following activation, several events have been elucidated. A circuit of T cell interactions has been defined on the basis of surface characteristics of the communicating T cells and the nature of their functional molecular products. The first subset that has been characterized as follwing activation is an I-1+, Ly-2 T cell which functions to induce contrasuppression [12]. The 1-1 determinants detected on the cells and molecules of the contrasuppressor circuit are serologically distinct from those expressed by cells and molecules of the "feedback" suppressor circuit [13]. The product of the inducer cell is a molecule(s) which bears an 1-1 subregion encoded product and can be absorbed on the immunizing antigen ( or closely related antigens). The cross-reactive nature of this antigen recognition distinguishes this molecule from suppressor factors [ 14], and is potentially extremely important. This will be discussed further in the next section. The contrasuppressor inducer factor must interact with a contrasuppressor transducer cell in order to have its effects. This transducer cell is an 1-1+, Ly-l, 2 T cell [12, 14]. The evidence at hand suggests that this interaction is restricted by genes linked to the V region of the Ig locus. The effector cell of the contrasuppressor circuit is an 1-1+, Ly-l T cell which can be positively selected by adherence to the Vicia villosa lectin [ 15], which distinguishes it from helper cells. Further, its activity can be blocked by the presence of N-acetylD-galactosamine [16]. This cell functions to render helper T cells (and probably other cells of the immune system) resistant to suppressor cell signals [ 15]. Further, this cell has the ability to block tolerogenic signals in vivo (allowing immunity to become manifest) [ 17]. Contrasuppression has been implicated in the generation and transfer of contact sensitivity 117 -19], resistance to malaria infections ([ 10], R. Mogil, personal communication), and development of the hyperimmune state [20,21]. We consider next the possibility that this activity functions in the immune response to cancer.

B. Evidence for Contrasuppression in Tumor Immunity

While contrasuppressor cells have not been used to modulate directly the immune response in cancer, they have been implicated in a number of systems. In this section we will discuss the involvement of contrasuppression in tumor immunity. Hamaoka et al. [22] described an immunization protocol which produced hapten-reactive T -lymphocytes in the "absence" of suppressor cells. Recently, Rozyka et al. (manuscript in preparation) have demostrated the production of a potent contrasuppressor factor from cells that were primed using Hamaoka's immunization protocol. Hamaoka et al. [23] have further demonstrated that primed animals can produce effective immunity to haptenated tumor cells. Thus, it is likely that activation of contrasuppression to interfere with suppressor cell activity is responsible for the enhanced immune response against the haptenated tumor cells. This is further supported by the observation that the immunity, with time, became cross reactive, such that after priming resistance could be demonstrated for the same tumor cells without hapten [23]. This may be a reflection of the cross-reactive nature of the contrasuppressor inducer cell discussed above [14], that is the reactions against the haptenmodified tumor-associated antigens raised contrasuppressor cells that protected the cells reacting to "unmodified" antigen from host suppressor mechanisms. Contrasuppression may be implicated in natural resistance to AKR leukemia virus. Mureullo and McDevitt [24] demonstrated that the transfer of resistance to oncogenesis was dependent upon an I-J+, Ly-l T cell, a cell with a "contrasuppressive phenotype". In addition, resistant animals could be rendered sensitive by injecting anti-Ly-l or anti-I -1 antisera in vivo. Since the effector cell of contrasuppression is an I-J+, Ly-l T cell [15], removal of this cell could account for the above observations. Cells which interfere with suppressor cell function were im plicated in genetic resistance to Friend leukemia virus (FL V) by Kumar and Bennett [25]. Susceptibility to leukemogenesis correlates with susceptibility to immunodepression by FL V [26]. Susceptibility to immunodepression was further correlated with ability to induce suppressor cells in vitro with FL V [27]. Resistance to suppressor cell induction by FL V was shown to be effected by a marrow dependent cell ("M cell"). Removal of the M cell allowed induction of suppressor cells in resistant strains [25]. (Contrasuppressor cells have been identified in bone marrow and shown to be involved in regulation of hematopoeisis [28].) Kumar and Bennett went on to descrIbe a "suppressor interfering cell" in the FL V system [29]. This will be considered in more detail in the next section. These observations support a role for contrasuppression in control of immunity to leukemia. Antibodies to certain tumor antigens may react with immunoregulatory cells [30]. Antisera against the Meth A fibrosarcoma raised in Fl animals, but not syngeneic homozygous animals, have been shown to disrupt contra suppressor activity. Production of these disruptive antibodies correlates with an increased incidence of metastasis in Fl animals over the parental strain [31 ]. While it remains to be proven that contrasuppressor cells are needed for optimum tumor immunity, the evidence is compelling that this investigative avenue is worth following. In the next section we will consider the activation of this circuit and hypothesize a role for antigen presentation in the context of class I (rather than class II) antigens.

C. Class I Antigens in Contra suppression and Tumor Immunity

In recent years it has become dogma that helper T cells recognize antigen in the context of class II surface antigens for the initiation of immune responses. Class I antigens are generally viewed as targets for effector cell (CTL) function, such as in T cell killing of transformed or virally infected targets. With few exceptions Ir gene effects mapping to class I loci mediate responses to viral antigens [32] or minor histocompatibility antigens [33]. It is becoming increasing clear, however, that class I antigen presentation in cell-mediated immunity may well involve activation of immunoregulatory subsets. Such regulation has implications for humoral immunity as well. Using H-2D region mutants, Stukart et al. [34] demonstrated a role for the H-2D halotype in regulating responses to Moloney leukemia virus, even when the effector cells were directed only at virus associated with K-end antigens. H-2D region control of immune responses has also been observed for radiation leukemia virus-induced tumorigenesis [35], Friend virus-induced splenomegaly [36], T -lymphocyte proliferative autoimmune responses to thyroglobulin [37], antibody levels and cellular infiltration in autoimmune thyroiditis [38], and ability to induce suppression for contact sensitivity with DNFB [39]. Antibody responses to equine myoglobin are regulated by complementing genes in H-2D and I-A [40]. Murine resistance to malaria may depend upon activation of contrasuppressor cells to overcome suppression ([ 10], R. Mogil, personal communication). Vaccination against fatal malaria infection is dependent upon the transfer of infected reticulocytes which display elevated levels of class I antigens [41 ]. Resistance, however, does not necessarily depend upon the parasite residing within reticulocytes, as immunization with the organism in reticulocytes leads to protective immunity against a fatal strain that proliferates only in mature red blood cells. This indicates that malarial parasites in reticulocytes are not simply better targets of effector cell activity. Class I antigens have been shown to be important in induction of immunity in several tumor systems. SJL reticulosarcoma lines bearing H-2D antigens are capable of inducing immunity to lines which lack H-2D [42]. Examination of progressor and regressor lines of a UV -induced sarcoma revealed an anitgenic difference mapping to the H-2D region of the MHC. Again, like the Hamaoka story and the immunity to malaria, the regressor line was found to be capable of inducing immunity to the progressor line which lacks the H-2D linked antigen [43]. In the FL V system, K umar and Bennett [29] examined an FL V -induced ""suppressor interfering cell" which was activated in vitro by genetic mismatch of this cell with its target. (The H-2 haplotype of the FL V induced suppressor cell was irrelevant.) This allogeneic activation was mapped to H-2D [44]. Recently, a system has been developed to analyze the activaton of contrasuppression by antigen-presenting cell subsets in vitro. Preliminary results suggest that this antigen-specific activation can be blocked by anti-class I ( especially H-2D) but not by anti-class II antibodies (in preparation). In light of the above observations, we propose that antigen presentation in the context of class I antigens, especially H-2D, may be important in initiation of contrasuppression. It may be relevant that dendritic cells, which can activate contrasuppression which leads to dominant immunity in vitro [9], are high in H-2D antigen expression [45]. If so, then a strategy for optimal tumor immunity may be elevation of class I antigen expression on the tumor cells to activate contrasuppression and allow dominant immune responsiveness over tumor-induced suppression. Experiments are in progress to test this notion.

D. Flip Side: Contrasuppression in Enhancement of Lymphoid Tumor Development

It is well established that persistent activation of target cells by their hormones can result in transformation and carcinogenesis. Regulatory factors are essentially the hormones of the immune system, and we can propose that persistent activation of their targets can result in neoplasia. Signals which inhibit activation, such as suppressor cell factors, might then serve to prevent lymphoid transformation, whereas activities like contrasuppression might, in some instances, enhance lymphoid tumorogenesis. For example, Houghton et al. [46] have described a situation in which antigenic hyperimmunization causes the appearance of tumors of cells of the immune system. The fact that several B-cell lymphomas produced in this way react with the immunizing antigen suggests direct involvement of the hyperimmunization protocol. Hyperimmunized animals have been shown to possess a potent antigen-specific contrasuppressive activity [20, 21 ]. As mentioned above, malaria infections in mice produce a potent contrasuppression coincident with recovery. Such infections can enhance oncogenesis by virus [47]. Whether there is any correlation of these effects is unknown, but suggests an exciting possibility. People infectd with EpsteinBarr virus (EBV) exhibit potent suppressor T -cell activity [4], and such cells have been shown to be capable of inhibiting EBV transformation in vitro [5]. Chronic infection with malaria, however, might induce a general contrasuppression which would interfere with this beneficial immunosuppression to allow expansion of the virus-transformed cells. This is a possible rationale for the association of EBV -induced lymphomas in malarial regions [48]. The MRL mouse is a murine model of systemic lupus erythematosis and lymphoproliferation in which autoimmunity proceeds in the face of general suppression [49]. These animals have been shown to be resistant to tolerance induction [50] and suppressor cell signals [51], probably as a result of excessive contrasuppressor activity [51]. The proliferating cells in these animals have a controlled neoplastic tendency, as suggested by the spontaneous appearance of transformed, tumorgenic lines when these cells are cloned (C. Reinisch, personal communication). An understanding of the role of immunoregulatory T cells in the control of such lymphoid tumors will great ly increase our knowledge of lymphocyte regulation and the regulation of transformed cells in general.

E. Conclusion

In this brief discussion, we have outlined our argument that contrasuppression might play an important role in the immune response to cancer. While antigen load often induces active suppression to most tumors, induction of contrasuppression early in the response might allow protective immunity to become dominant. There is suggestive evidence that contrasuppression can be initiated by presentation of antigen in the context of class I antigens, in which case these will have a profound role in determining the outcome (positive versus negative immunity) ofa tumor challenge. Many tumors can potentially be controlled by immune responses against the tumor. Certain tumors, in addition, might be affected by the regulatory molecules of the immune system themselves, especially if the tumors are lymphoid in origin. Such tumors might behave anomalously ( on the surface), being enhanced by positive influences on immune function and controlled by suppressive signals. Nevertheless, it is clear that as our understanding of immunoregulation increases, we simultaneously improve our potential for controlling the immune response to cancer and increase our abilites to produce effective therapy.

Acknowledgments

The authors wish to thank Rona Mogil for critical discussion of this manuscript and Astrid Swanson and June Goldstein for expert secretarial assistance.

References

I. Fujimoto S, Greene MI, Sehon AH (1976) Regulation of the immune response to tumor antigens I. Immunosuppressor cells in tumor-bearing hosts. J Immuno1116:791

2. Rohrer JW, Lynch RG (1979) Immunoregulation of localized and disseminated murine myeloma: Antigen-specific regulation of MOPC-315 stem cell proliferation and secretory cell differentiation. J 123:1083

3. Abbas AK, Burokoff SJ, Getter ML, Greene MI (1980) T lymphocyte-mediated suppression of myeloma function in vitro III. Regulation of antibody production in hybrid myeloma cells by T lymphocytes. J Exp Med 152:969

4. Tosato G, Magrath I, Koski T, Dooley N, Blaese M ( 1979) Activation of suppressor T cells during Epstein-Barr virus induced infectious mononucleous. N Engl J Med 301: 1133

5. Thorley-Lawson DA (1980) The suppression of Epstein-BaIr virus infection in vitro occurs after infection but before transformation of the cell. J Immunol 124: 745

6. SchiffC (1980) Regulation of the murine immune response to sheep red blood cells: The effect of educating lymphocytes in vitro with glycoproteins isolated from the sheep erythrocyte mem brane. Medical School Thesis, Yale U Sch Med, New Haven, CT

7. Askenase PW, Hayden B, Gershon RK (1975) Augmentation of delayed-type hypersensitivity by doses of cyclophosphemide which do not effect antibody responses. J Exp Med 141:697

8. Ptak W, Rozycka D, Askenase PW, Gershon RK (1980) Role of antigen-presenting cells in the development and persistence of contact hypersensitivity. J Exp Med 151 : 363

9. Britz JS, Askenase PW, Ptak W, Steinman RM, Gershon RK ( 1982) Specialized antigen-presenting cells: Splenic dendritic cells and peritoneal exudate cells induced by mycobacteria acitvate effector T cells that are resistant to suppression, J Exp Med 155 : 1344

10. Green DR (1981) Contrasuppression, an immunoregulatory T cell activity. Doctoral Thesis, Yale University, New Haven. CT

II. Green DR, Gershon RK, (to be published) Contrasuppression, Adv Cancer Res

12. Gershon RK, Eardley DD, Durum S, Shen FW. Yamauchi K, Cantor H, Murphy DB (1981) Contrasuppression: A novel immunoregulatory activity. J Exp Med 153: 1533

13. Yamauchi K, Taniguchi M, Green D, Gershon RK (to be published) The use of a monoclonal anti-I-J antibody to distinguish cells in the feedback suppression circuit from these in the contrasuppressor circuit. Im m unogenetics b 14. Yamauchi K, Green DR, Eardley DD, Mur phy DB, Gershon RK (1981) Immunoregulatory circuits that modulate responsiveness to suppressor cell signals: The failure of B10 mice to respond to suppressor factors can be overcome by quenching the contrasuppressor circuit. J Exp Med 153: 1547

15. Green DR, Eardley DD, Kimura A, Murphy DB, Yamauchi K, Gershon RK (1981) Immunoregulatory circuits which modulate responsiveness to suppressor cell signals: Characterization of an effector cell in the contrasuppressor circuit. Eur J 1mmunol 11:973

16. Green DR, Gershon RK (in preparation)

17. Ptak W, Green DR, Durum SK, Kimura A, Murphy DB, Gershon RK (1981) 1mmunoregulatory circuits that modulate responsiveness to suppressor cell signals: Contrasuppressor cells can convert an in vitro tolerogenic signal into an immunogenic one. Eur J Immunol II :980

18. Iverson GM, Ptak W, Green DR, Gershon RK (to be published) The role of contrasuppression in the adoptive transfer ofimmunity 19. Green DR (1982) Contrasuppression: Its role in immunoregulation. In: Fefer A, Goldstein A ( eds) The potential role of T cells in cancer therapy. Raven New York

20. Green DR, Gershon RK (1982) Hyperimmunity and the decision to be intolerant. Ann NY Acad Sci 392:318

21. Green DR, Gershon RK (to be published) Immunoregulatory T cells in hyperimmune mIce

22. Hamaoka T, Toshizawa M, Yamamoto H, Kuroki M, Kitagawa M (1977) Regulatory functions of hapten-reactive helper and suppressor T lymphocytes. II. Selective inactivation of hapten-reactive suppressor T cells by hapten-nonimmunogenic copolymers of D-amino acids and its application to the study of suppressor T -cell effect on helper T-cell development. J Exp Med 146:91

23. Hamaoka T, Fujiwara H, Teshima K, Aoke H, Yamamoto H, Kitagawa M (1970) Regulatory functions of hapten-reactive helper and suppressor T lymphocytes. 111. Amplification ofa generation oftumor-specific killer T lymphocyte activities by suppressor T celldepleted hapten-reactive T lymphocytes. J Exp Med 149: 185

24. Meruelo D, Flieger N, Smith D, McDevitt HO ( 1980) In vivo or in vitro treatments with anti-I-J alloantisera abolish immunity to AKR leukemia. Proc Natl Acad Sci USA 77:2178 25.

25. Kumar V, Bennett M (1976) Mechanisms of genetic resistance to Friend virus leukemia in mice. II. Susceptibility of mitogen responsive lymphocytes mediated by marrow-dependent cells. J Exp Med 143:713

26.Ceglowski WS, Friedmann H (1969) Murine virus Leukemogenesis, ralationship between susceptibility and immunodepression. Nature 224:1318

27. Kumar V, Bennett M (1976) Mechanisms of genetic resistance to Friend virus leukemia in mice. III. Susceptibility of mitogen responsive lymphocytes mediated by T cells. J Exp Med 143:728

28. Michaelson JD ( 1982) The characteriza tion and functional analysis ofimmunoregulatory cells in murine bone marrow. Medical School Thesis. Yale U Sch of Med, New Haven

29. Kumar V, Bennett M (1977) H-2 compatibility requirements for T suppressor cell functions induced by Friend leukemia virus. Nature 265:345

30. Flood PM, DeLeo AB, Old LJ, Gershon RK (to be published) The relation of cell surface antigens on methylcholanthrene-induced fibrosarcomas to cell Igh-V linked cell interaction molecules. Proc Natl Acad Sci USA

31. Flood PM, DeLeo AB, Old LJ, Gershon RK (this volume) Antisera against tumor-associated surface antigens on methylcholanthrene-induced sarcomas inhibit the induction ofcontrasuppression

32. Zinkernagel RM, Althage A, Cooper S, Kreeb G, Klein CA, Sefton B, Flaherty L, Stimpfling J, Shreffler D, Klein J (1978) lr genes in H-2 regulated generation of antiviral cytotoxic T cells. J Exp Med 148: 592

33. Wettstein PJ, Frelinger JS (1977) H-2 effects on cel1-cell interactions in the response to single non-H-2 alloantigens. J Exp Med 146:1356

34. Stukart MJ, Vos A, Boes J, Melvold RW, Bailey DW, Melief CJM (1982) A crucial role of the H-2D locus in the regulation of both the D- and the K-associated cytotoxic T Iym phocyte response against Moloney leukemia virus, demonstrated with two D b mu tants. J Immunol 128: 1360

35. Meruelo D. Lieberman M, Gington N, Deak B, McDevitt HO ( 1977) Genetic control of radiation leukemia virus-induced tumorogenesis. I. Role of the murine major histocompatibility complex, H-2. J Exp Med 146:1079

36. Chesebro B, Wehrly K, Stimpfling J (1974) Host genetic control of recovery from Friend leukemia virus-induced splenomegaly. J Exp Med 140: 1457

37. Christadoss P, Kong YM, Elkehewy M, Rose NR, David CS (1978) Genetic control of T lymphocyte proliferative autoimmune response to thyroglobulin in mice. In: Rose NR, Bigazzi PE, Warner NL (eds) Genetic control of autoimmune disease. Elsevier North-Holland, Inc New York, p 445

38. Kong YM, David CS, Giraldo AA, ElRehewy M, Rose NR (1979) Regulation of autoimmune response to mouse thy roglobulin: Influence of H-2-D-end genes. J Immunol123: 15

39. Moorhead JW (1977) Soluble factors in tolerance and contact hypersensitivity to DNFB in mice. J Immunol119: 1773

40. Berzofsky JA, Buckenmeyer GK, Hicks G (1982) Genetic control of the immune response to myoglobins. VI. Distinct Ir genes for different myoglobins: Complementing genes in I-A and H-2D fo)r equine myoglobin. J Immunol 128:737

41. Jayawardena AN, Gershon RK, Mogil RJ, Murphy DB, Burger D (to be published) Enhanced expression of H-2K and H-2D antigens on reticulocytes infected with P. yoelii. Nature

42. Kuhn M, Dyer DA, Mayo L, Jagdish Babu K, Righthand VF, Lightbody J, Beisel KW, Lerman SP ( 1982) Properties of SJL strain lymphomas which fail to express membrane H-2Ds. Fed Proc 41:822 43. Daynes RA, Fernandez PA, Woodward JG (1979) Cell-mediated immune response to ultraviolet light-induced tumors. II. The properties and antigenic specifities of cytotoxic T lymphocytes generated in vitro following removal trom syngeneic tumor-immunized mice. Cell Immunol45: 398

44. Kumar V, Bennett M (1979) Immunosuppression by Friend leukemia virus in H-2 restricted by alloreactive T lymphocytes. Proc Nat Acad Sci USA 76:2415

45. Nussenzweig MC, Steinman RM, Unkeless JC, Witmer MD, Gutchinov B, Cohn ZA ( 1981) Studies of the cell surface of mouse dendritic cells and other leukocytes. J Exp Med 154: 168

46. Lynes MA, Lanier LL, Babcock GF, Wettstein PJ. Haughton G (1978) Antigen-induced murine B cell lymphomas. I. Induction and characterization of CH 1 and CH2. J Immunol121 : 2352

47. Wedderburn N (1970) Effect of concurrent malarial infection on development of virusinduced lymphoma in Balblc mice. Lancet II: 1114

48. Miller G (1976) Epidemiology of Burkitt's lymphoma. In: Evans AS (ed) Viral infection of humans. Plenum, New York

49. Gershon RK, Horowitz M, Kemp JD, Murphy DB, Murphy DB (1978) The cellular site of immunoregulatory breakdown in the lpr mouse. In: Rose NR, Bigazzi RE, Warner NL (eds) Genetic control of autoimmune disease. Elsevier IN orth Holland, New York, p 223

50. Amagal T, Cinader B ( 1981) Resistance of MRLlmp-lprllpr mice to tolerance induction. Eur J Immunolll : 923

51. Gershon RK (1980) Immunoregulation: Some comments on the state of the art. J Allergy Clin Immuno166: 18