H-2-incompatible bone marrow chimeras produce ... - Europe PMC

Proc. Nati. Acad. Sci. USA Vol. 82, pp. 7063-7067, October 1985 Immunology

H-2-incompatible bone marrow chimeras produce donor-H-2-restricted Ly-2 suppressor T-cell factor(s) (radiation chimera/marrow transplantation)

M. NOGUCHI*, K. ONo0*, M. OGASAWARA*, K. IWABUCHI*, L. GENG*, K. OGASAWARA*, R. A. GOODt, AND K. MORIKAWA* *Institute of Immunological Science, Hokkaido University, Sapporo, 060, Japan; and tUniversity of South Florida, Department of Pediatrics, All Children's Hospital, 801 6th Street, South, St. Petersburg, FL 33701

Contributed by R. A. Good, May 13, 1985

ABSTRACT To study adaptive-differentiation phenomena of T lymphocytes, suppressor T-cell factors (TsF) produced by Ly-2' splenic T cells from fully allogeneic mouse bone marrow chimeras were analyzed. AKR mice irradiated and reconstituted with B10 marrow cells (B1O--AKR chimeras) produced an Ly-2' TsF after hyperimmunization with sheep erythrocytes. The TsF suppressed primary antibody responses (to sheep erythrocytes) generated with spleen cells of mice of H-2b haplotype but not those of H-2k haplotype. Thus, this suppressor factor was donor-H-2-restricted. The immunoglobulin heavy chain variable region gene (Igh-V)-restricting element was not involved in this form of suppression. Similar results were obtained when TsF from B6-+BALB/c and BALB/c-oB6 chimeras were analyzed. The TsF from B1O--AKR chimeras suppressed responses of BlO.A(3R) and BlO.A(SR) mice but not those of B1O.A(4R). This finding showed that identity between the factor-producing cells and target spleen cells is required on the left-hand side of the Ep locus of the H-2 region and that the putative I-jb locus is not involved in this form of suppression. The present results support the postulate that post-thymic differentiation in the presence of continued or repeated stimulation with antigen and donor-derived antigenpresenting cells generates donor-H-2-restricted T-cell clones that may predominate within the repertoire of the specific antigen being presented.

Adaptive differentiation is a characteristic of T lymphocytes (1-3) by which they acquire a self-specificity repertoire during differentiation. The model of the irradiation bone marrow chimera has provided a useful strategy for analyzing how hematopoietic stem cells differentiate and mature in an allogeneic microenvironment, where developing T lymphocytes acquire the ability to recognize major histocompatibility complex (MHC) antigens of recipient as self and to cooperate efficiently with antigen-presenting cells (APC), B cells, or cells of T-cell subsets bearing the MHC antigens of the recipient (2-5). Arguments against the concept of adaptive differentiation have been presented and the crucial role of recipient microenvironment has been questioned (6-8). The molecular basis of the recognition process and alterations of expression of allogeneic MHC antigen in the environment of recipient is not well understood and more penetrating studies seem indicated. In previous reports (5, 9-12), it was shown that adaptive differentiation occurs for helper T cells from H-2-incompatible chimeric mice; these cells developed so as to generate primary antibody responses to sheep erythrocytes (SRBC) in combination with cells of recipient H-2 type but not with cells

genetically identical to their original MHC. T cells from such chimeras primed with an antigen in vivo were found later to cooperate in vitro also with cells of donor type, indicating that there may be two separate stages of T-cell differentiation during which self-restriction specificity is acquired. One is responsive to intrathymic influences and not associated with antigenic stimulation, and the other responsive to postthymic stimuli that involve antigen-presentation. Here we analyze restriction specificities of a suppressor T-cell factor (TsF) produced by Ly-2+ splenic T cells from H-2-incompatible chimeras that have been hyperimmunized with SRBC according to Yamauchi et al. (13). This TsF inhibited only those primary anti-SRBC responses that were generated by spleen cells of donor type. This factor was both antigen-specific and donor-H-2-restricted. If the factor contained recognition molecules for H-2 antigens on target cells as proposed by Flood et al. (14), these findings would fit with our postulate that post-thymic differentiation involving stimulation with antigen generates donor-H-2-restricted T-cell clones that may become dominant among cells determining the repertoire for antigens with which an animal has been stimulated.

MATERIALS AND METHODS Mice. AKR/J, SJL, and B10.A(5R) mice were obtained from The Jackson Laboratory, Bar Harbor, ME; C57BL/10 (B10), B10.BR, C57BL/6 (B6), and BALB/c mice were from the Shizuoka Laboratory Animal Cooperation, Hamamatsu, Shizuoka Prefecture, Japan. B10.A(3R) and B10.A(4R) mice originally obtained from T. Hamaoka (Osaka University) were bred and maintained in the mouse colony at Hokkaido University. Chimeras. Marrow chimeras were prepared as described (9, 10, 15). After hematopoietic reconstitution, all were maintained without further manipulation for 6-10 weeks prior to analyses for chimerism and preparation of Ly-2 TsF factors. For simplicity, chimeras prepared by injecting C57BL/10 cells into irradiated AKR/J mice will be referred to as B10-lAKR; BALB/c mice treated with C57BL/6 cells, as B6-*BALB/c; and C57BL/6 mice treated with BALB/c cells, as BALB/c--B6. Chimerism. Thymocytes and spleen cells were analyzed for susceptibility to cytolysis by use of specific anti-Thy-1 sera, anti-H-2 sera, and selected rabbit complement (9, 16). Thymocytes and spleen cells from each chimera were derived exclusively from donor bone marrow cells (data not shown). Abbreviations: APC, antigen-presenting cells; HRBC, horse eryth-

rocytes; Igh-V, genes regulating expression of the variable

portions

of the Ig heavy chain; MHC, major histocompatibility complex; PFC, plaque-forming cells; SRBC, sheep erythrocytes; TsF, suppressor T-cell factor(s); TsiF, suppressor/inducer T-cell factor(s).

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.

7063

7064

Immunology: Noguchi et al.

Antisera and TsF. Anti-Thy-1.2 monoclonal antibody (F7D5) was purchased from Olac Ltd. (Bicester, United Kingdom). Anti-Ly-1.2 and anti-Ly-2.2 monoclonal antibodies were provided by N. Tada and S. Kimura (Memorial Sloan-Kettering Cancer Center, New York). Anti-IlJb was prepared as described (13). To remove auto-antibodies, anti-IlJb antiserum was adsorbed with AKR/J thymus and spleen cells before use. Ly-2 TsF were prepared by the method of Yamauchi et al. (13). Spleen cells from mice hyperimmunized with SRBC were treated with anti-Ly-1 plus rabbit complement and then were cultivated for 48 hr at 107 cells per ml. Supernatant fluids were centrifuged and passed through Millipore filters (SLGS, 0.22-,m pore size). Plaque-Forming Cell (PFC) Responses to SRBC or Horse Erythrocytes (HRBC) in Vitro. Spleen cells (106) were cultured in quadruplicate with 5 x 105 SRBC or HRBC for 4 days in 96-well culture plates (no. 35860, Corning) in medium prepared according to Ohmori and Yamamoto (17). Collected cells were assayed by the PFC assay of Jerne (5, 9). Results are presented as PFC per 107 spleen cells. RESULTS Identity of H-2 Between Ly-2 TsF Producer and Target Cells Is Required for Ly-2 TsF to Express Suppressive Activity. Ly-2 TsF made by immune Ly-1-2+ splenic T cells inhibits primary antibody response in an antigen-specific and H-2restricted fashion (13, 18). To show that Ly-2 TsF was H-2-restricted and not contaminated by immunoglobulin heavy chain variable region gene (Igh-V)-restricted Ly-1 suppressor/inducer T-cell factor (Ly-1 TsiF), Ly-2 TsF produced by B10 (H-2b Igh-lb) or AKR (H-2k Igh-qd) mice was analyzed. Spleen cells from B10 or AKR mice hyperimmunized with SRBC were depleted of Ly-l+ cells by treating with anti-Ly-1.2 monoclonal antibody and rabbit complement and then were cultured for 48 hr; culture supernatants (Ly-2 TsF) were collected and then assayed. When TsF derived from B0O mice was added to primary anti-SRBC PFC cultures of B10, AKR, and B1O.BR (H-2k Igh-lb) spleen cells, it suppressed only responses of B10 mice but not of H-2 and Igh- V-disparate AKR or H-2-disparate but Igh-V-identical B1O.BR mice (Table 1). Ly-2 TsF derived from AKR almost completely suppressed responses of H-2identical Igh-disparate B1O.BR mice as well as syngeneic AKR mice but not those of H-2-disparate B10 mice, demonstrating that Ly-2 TsF suppresses the antibody response according to H-2 restriction, as reported by Yamauchi et al. (18). Ly-2 TsF Produced by H-2-Incompatible B10--AKR Bone Marrow Chimeras Is Donor-H-2-Restricted and Antigen-Specific. As shown in Table 2, the TsF produced by splenic Ly-2 Table 1. Identity at the H-2 region but not the Ig locus of Ly-2 TsF producer and target cells is required for Ly-2 TsF to express suppressive activity Ly-2 TsF from B10 Ly-2 TsF from AKR Assay cells* PFC per PFC per % TsFt in culture culture % TsFt culture 0 B10 2370 0 1460 5 910 (62)t 5 1370 (6) 0 AKR/J 540 0 1130 5 710 (0) 5 260 (77) B1O.BR 0 1730 0 760 5 1720 (3) 5 140 (82) *Unprimed spleen cells (106). tPercent T-cell suppressor factors (i.e., % filtered supernatant fluid from cultured TsF-producing cells) added to assay culture medium. iNumbers in parentheses represent percent suppression produced by the factors.

Proc. Natl. Acad. Sci. USA 82 (1985) T cells from B10--AKR chimeras suppressed anti-SRBC PFC responses generated only with B10 spleen cells and not with AKR or B10.BR cells and was thus donor-H-2-restricted rather than recipient-H-2-restricted as reported by Yamauchi et al. (13) and not influenced by other genetic contributions (e.g., Igh-V). This table also shows that TsF did not have an antigenically nonspecific effect (for example, in suppressor responses to HRBC). Ly-2 TsF Producedi by the H-2-Incompatible Chimeras of Other Combinations Are Also Donor-H-2-Restricted. Irradiation bone marrow chimeras show wide variations in responsiveness, which are influenced by differences in strain combinations employed to establish the chimeras and by the nature of antigens used in the assay systems (19, 20). To determine whether the results obtained from B10--AKR chimeras are limited to the B10-+AKR strain combination or whether they may be extended to other chimeras, we prepared additional H-2-incompatible chimeras, B6-+BALB/c and BALB/c--B6, and analyzed restriction specificity of Ly-2 TsF produced in these chimeras. The TsF from both B6-*BALB/c and BALB/c-+B6 chimeras suppressed antiSRBC PFC responses only in cell cultures of the donor H-2 type (Table 3). Although marginal inhibition was seen in responses of AKR cells to which the TsF from BALB/c-*B6 chimeras was added, the suppressive factors from these chimeras appeared to be predominantly donor-H-2-restricted. Ly-2 TsF from B10--AKR Chimeras Prepared After Purging of Marrow With Anti-Thy-i Plus Complement. To reduce the likelihood that the cells producing TsF were derived from a few residual donor cells that escaped killing with anti-Thy1.2 alone, B10-+AKR chimeras were prepared by use of an alternative method of purging the marrow that employed anti-Thy-1.2 plus complement. TsF from such chimeras was also found to be exclusively donor-strain-restricted (Table 4). No differences between TsF from the chimeras purged in the two different ways was seen. It thus seems unlikely that cells producing the TsF are derived from a small residual population of T cells of donor origin. Ly-2 TsF-Producing Cells from B10-+AKR Chimeras Are Donor-Derived, Ly-2+, I-J- Cells. To determine the phenotype of Ly-2 TsF-producing cells from B10--AKR chimeras, we treated Ly-1-depleted spleen cells with three kinds of antisera (anti-Thy-1.2, anti-Ly-2.2, or anti-I-Jb) and complement prior to cultivation. Table 5 compares suppressive activity of supernatants from the four kinds of spleen cell cultures. Treatment with anti-Ly-2.2 or anti-Thy-1.2 antiserum plus complement inhibited ability of the spleen cells to generate TsF. Supernatants did not inhibit anti-SRBC PFC responses of normal B10 mice. In contrast, anti-I-Jb antiserum plus complement did not effect TsF activity. The same treatment was, however, used to demonstrate the presence of I-J determinants on the Ly-1 inducers of suppression (unpublished results). Thus, these observations establish that Table 2. Ly-2 TsF produced by B10 -+ AKR chimeras is antigen-specific and donor-H-2-restricted PFC per culture in response to % Ly-2 TsF from Assay cells in culture B10-. AKR SRBC HRBC 0 4200 B10 2080 840 (80) 10 3100 (0) AKR

B10.BR

0 10

0 10

See Table 1 footnotes for details.

1720 1730 (0)

790 1170 (0)

550 680 (0)

1080 970 (10)

Proc. Natl. Acad. Sci. USA 82 (1985)

Immunology: Noguchi et A Table 3. Ly-2 TsF from the other two kinds of the H-2incompatible chimeras are also donor-H-2-restricted Ly-2 TsF from Ly-2 TsF from B6 -. BALB/c BALB/c -. B6 PFC per PFC per Assay cells % TsF culture % TsF culture in culture 620 0 6710 0 B6 1010 (0) 10 1170 (83) 5 2090 0 1240 0 BALB/c 680 (68) 10 5 1360 (0) 1520 0 1130 0 AKR/J 1210 (20) 10 1060 (6) 5 ND 1280 0 B1O.BR 5 1780 (0) 1520 0 ND DBA/2 450 (70) 10 See Table 1 footnotes for details. ND, not done.

Ly-2 TsF-producing cells of the B10-+AKR chimeras are exclusively donor-derived, Ly-2+, I-J- T cells. Precise Analyses of H-2-Restriction Specificity of the Ly-2+ TsF Obtained From B10-*AKR Chimeras. H-2 specificities of TsF from spleen cells of B10-*AKR chimeras added to cultures of spleen cells from various mice are shown in Table 6. This TsF suppressed the primary anti-SRBC PFC responses of B10 but not those of AKR mice, a finding concordant with those (Table 2) that TsF suppressed only donor type (H-2b) cells and exhibited no suppressive activity on cells of SJL (Igh-1b) or B10.BR mice (Hf-2k Igh-lb). No Igh-Vrestricting element was exhibited. The responses of B10.A(3R) and B10.A(5R) were suppressed, however, whereas those of B1O.A(4R) were not inhibited. This finding indicates that for the TsF to show suppressive activity, identity between the factor-producing cells and target spleen cells is required to the left side of the E, locus of H-2. The putative I-J locus had no influence on this suppressor mechanism.

DISCUSSION We analyzed H-2 restriction specificities of Ly-2 TsF of allogeneic bone marrow chimeras for restriction to donor H-2 or to recipient H-2. Yamauchi et al. (13) and Flood et al. (14)

reported that TsF from spleen cells of semi-allogeneic bone marrow chimeras or that made by the cells of nude mice engrafted with F1 thymuses suppressed primary antibody responses of spleen cells bearing the same H-2 antigens as expressed in the thymus. These investigators concluded that Ly-2 TsF-producing cells had undergone adaptive differentiation in thymus. Table 4. Suppression by Ly-2 TsF of in vitro PFC response PFC per % Ly-2 TsF from Assay cells % suppressions culture BlO AKR* in culture 197 0 B10 95 9 5 69 62 10 88 0 AKR 0 102 5 1 87 10 67 0 B1O.BR 0 70 5 4 64 10 *Chimeras prepared with marrow from BlO mice that had been purged with anti-Thy-1.2 plus complement. tRelative to response with no (0%) TsF.

7065

Table 5. TsF-producer cells from B10-.AKR chimeras are donor-derived Ly-2' I-J T cells Treatment of TsF producer % suppression PFC per culture cells from B10 -. AKR 1160 Control (no TsF) 0 1200 Anti-Ly 2.2 84 190 Anti-I-Jb 0 1170 Anti-Thy-1.2 53 550 Complement control Ly-1+-depleted spleen cells from B10-+AKR chimeras were treated with anti-Ly-2.2, anti-I-Jb, or anti-Thy-1.2 plus complement prior to cultivation. In this assay, PFC culture media were supplemented with 5% supernatants from each producer culture. Each assay culture consisted of 106 unprimed spleen cells from B10 mice.

Our present results differ with these. The Ly-2 TsF obtained from fully H-2-incompatible bone marrow chimeras antigen-specifically suppressed primary anti-SRBC PFC responses of spleen cell cultures of donor H-2 type and showed no suppressive activity on cells of recipient H-2 type or those of H-2-disparate mice. TsF did not inhibit responses of Igh-identical but H-2-disparate mice, suggesting that Ly-1 TsF (13) was not involved in the suppression. It seems that Ly-2+ suppressor/effector T cells of H-2-incompatible chimeras have not undergone adaptive differentiation in the recipient thymus and do not recognize the H-2 phenotype of the recipient mouse as self. Alternatively, T cells involved seemed to have been generated under the influence of an extrathymic environment, as proposed earlier (5, 11). Repeated or continued stimulation of T cells with antigen (SRBC) plus APC from donor bone marrow fosters donorH-2-restricted helper T-cell clone development (post-thymic or thymus-independent differentiation). Although expansion by hyperimmunization of a few mature T cells has not been ruled out completely, this does not seem the case, because donor T-cell functions against even the most immunogenic host antigens (9) were not seen in our studies or those of Muto et al. (21). Further, experiments to purge T cells from marrow inoculum even more completely, using B10 marrow purged with anti-Thy-1.2 plus complement, did not give different results. Apparent discrepancies between the present findings and Table 6. Genetic specificity of restricting element of Ly-2 TsF obtained from B10 -- AKR chimeras PFC per culture I-region haplotype Assay cells* Exp. 2t in culture K AP A,,, Ep J Ec. S D Exp. lt 1840 860 b b b b b b bb B10 170 (91) 490 (43)§ 500 k k k k k k kk 1460 B10.BR 520 (0) 1130 (23) ND 630 k k k k k k k k AKR/J 580 (8) ND 810 s s s s s s ss SJL 880 (0) 910 B10.A(3R) b b b b b k d d 1980 130 (86) 720 (64) 950 B10.A(4R) k k k k b b b b 760 710 (25) 1080 (0) 1140 B10.A(5R) b b b b k k d d 1710 10 (99) 510 (70) ND, not done. *Unprimed spleen cells (106). tTsF (10%) from B10 -. AKR chimeras was added. tTsF (5%) from B10 -. AKR chimeras was added. §Numbers in parentheses represent percent suppression produced by TsF.

7066


those of Yamauchi and Flood and their co-workers (13, 14) is probably due to differences in the models used. The mice analyzed by these investigators were semi-allogeneic chimeras, whereas those we studied were fully allogeneic chimeras. In our mice, APC bearing the same H-2 molecules as those expressed in the recipient thymus were not present. Therefore, Ly-1 Qa-1 T-cell clones restricted to the recipient H-2, which originally might exist as a major population but might not be activated in vivo, could not initiate the immunoregulatory circuit in question here. Instead, T cells restricted to donor H-2 appeared to be activated and to initiate the suppressor circuit. In the semi-allogeneic chimeras studied by Yamauchi and Flood and their co-workers (13, 14), T cells could encounter the same H-2 molecules that they had learned to recognize as self within the thymus. This explanation may not be applicable to the parent--F1 chimeras reported by Yamauchi et al. (13). In such mice, APC bearing the opposite parental H-2 in the F1 cross would not be expected to exist if full chimerism had been achieved. Nonetheless, Ly-2' T cells of the semi-allogeneic chimeras eventually suppressed responses of the spleen cells of both parental H-2 types. Thus, suppressor T-cell clones specific for the H-2 molecules of the opposite parent strain may have been activated by an unknown pathway without APC or other recognized partner cells. The immunoregulatory circuit is comprised of various and successive series of T-cell subsets. Ly-1+2- I-J+ Qa-1 T cells produce Ly-1 TsiF which in turn can induce another T-cell set (Ly-1+2- I-J+ Qa-1) to either amplify the activity and/or act as a precursor for Ly-1-2+ suppressor/effector cells (18). To better understand mechanisms that permit interpretation of the several discrepancies, precise analyses of the individual steps of each cell-cell interaction must be performed. Preliminary experiments, in which restriction specificity of Ly-1 TsiF obtained from B10-*AKR have been analyzed, indicate that the TsiF also may be donor-restricted. However, in this case, the restricting elements are not H-2-linked genes but probably Igh-V-linked genes. These points will be analyzed definitively in a separate paper. Several suppressor factors have been described and heterogeneity among them has been reported. Some are antigenspecific and others nonspecific. Most require genetic identity between the factors and target cells for expression of suppressive activities, but some function without apparent genetic regulation. Identity at H-2 appears to be required for interaction between Ly-2 TsF and target cells. Flood et al. (14) suggested that the Ly-2 TsF binds carrier chain by anti-I-J receptors (I-J restriction) and that this binding is required for expression of suppressive activities. However, precise mapping of the part of the H-2 region that must be shared between the TsF-producing and the target cells has not yet been completed. In the final experimnent reported herein, we attempted to map the subregion of the H-2 MHC of the target cells to which the Ly-2 TsF was restricted. The H-2b-restricted Ly-2 TsF that was obtained from B10-*AKR chimeras suppressed the responses of B10.A(3R) (A E1 J) and B10.A(5R) (AP a3Jk), but not that of B10.A(4R) (A4k Ek Jb). Thus, [-Jb does not appear to be involved in the factor-cell interaction we describe. Instead, compatibility on the left side of H-2 is

required. Our data are not compatible with the conclusion of Flood et al. (14) or with other reports that stress a role for the I-J locus in suppressor T-cell functions (23-25). However, they may be compatible with the recent report by Hayes et al. (26), who found that T cells from B10.A(3R) mice as well as B1O.A(5R) mice did not express I-J molecules. Therefore, the similarity of responses to the TsF with cells of these two strains supports the idea that the I-J locus is not involved in the suppression investigated here.

Proc. Natl. Acad. Sci. USA 82 (1985)

The murine immune-response subregion I-J, which had been thought to encode a suppressor T-cell marker, remains controversial (23-28). Recently, Ikezawa et al. (27) proposed that the I-J determinants are carried by the modified Ep and/or A,3 chains. Our present result would suggest that Ap and/or Ep3 can exert an influence on suppression. However, the possibility that these are characteristics unique to chimeras between H-2b and H,2k mice is not excluded because Ly-2 TsF produced by the spleen cells from normal B10 or AKR mice showed wide variation in suppression patterns and we could not map the precise subregions in the H-2. Molecular analyses of the Ly-2 TsF will be required to resolve the paradox. We wish to thank Miss Michiyo Konishi for excellent technical assistance. These investigations were aided from Scientific Research Grants 5657043 and 58480184 from the Ministry of Education, Science, and Culture, Japan, the March of Dimes Birth Defects Foundation Grant 1-789, and Public Health Service Grants A122360 and AG05628. 1. Katz, D. H., Katz, L. R., Bogwitz, C. A. & Skidmore, B. (1979) J. Exp. Med. 149, 1360-1370. 2. Zinkernagel, R. M., Callahan, G. N., Althage, A., Cooper, S., Klein, P. A. & Klein, J. (1978) J. Exp. Med. 147, 882-896. 3. Singer, A., Hathcock, K. S. & Hodes, R. J. (1981) J. Exp. Med. 153, 1286-1301. 4. Bevan, M. (1977) Nature (London) 269, 417-418. 5. Onod, K., Yasumizum, R., Noguchi, M., Iwabuchi, K., Ogasawara, M., kakinuma, M., Okuyama, H., Good, R. A. & Morikawa, K. Immunobiology, in press. 6. Matzinger, P. & Mirkwood, G. (1978) J. Exp. Med. 148, 84-92. 7. Stockinger, H., Pfizenmaier, K., Hardt, C., Rodt, H., Rollinghoff, M. & Wagner, H. (1980) Proc. Natl. Acad. Sci. USA 77, 7390-7394. 8. Aizawa, S., Sado, T., Muto, M. & Kubo, E. (1981) J. Immunol. 127, 2426-2431. 9. Onod, K., Fernandes, G. & Good, R. A. (1980) J. Exp. Med. 151, 115-132. 10. Onod, K., Yasumizu, R., Oh-ishi, T., Kakinuma, M., Good,

R. A. & Morikawa, K. (1981) J. Exp. Med. 153, 1009-1014. 11. Onod, K., Yasumizu, R., Geng, L., Iwabuchi, K., Ogasawara, M., Kakinuma, M., Okuyama, H., Good, R. A. & Morikawa, K. Immunobiology, in press. 12. Iwabuchi, K., Onod, K., Ogasawara, M., Yasuniizu, R. & Morikawa, K. (1983) Hokkaido J. Med. Sci. 58, 232-237. 13. Yamauchi, K., Flood, P. M., Singer, A. & Gershon, R. K. (1983) Eur. J. Immunol. 13, 285-291. 14. Flood, P., Yamauchi, K. & Gershon, R. K. (1982) J. Exp, Med. 156, 361-371. 15. Onod, K., Yasumizu, R., Oh-ishi, T., Kakinuma, M., Good, R. A., Fernandes, G. & Morikawa, K. (1982) J. immunol. Methods 49, 315-322. 16. Onod, K., Fernandes, G., Shen, F. W. & Good, R. A. (1982) Cell. Immunol. 68, 207-219. 17. Ohmori, H. & Yamamoto, I. (1982) J. Exp. Med. 155, 1277-1290. 18. Yamauchi, K., Murphy, D., Cantor, H. & Gershon, R. K. (1981) Eur. J. Immunol. 11, 913-918. 19. Yasumizu, R., Onod, K., Iwabuchi, K., Ogasawara, M., Geng, L. & Morikawa, K. Immunology, in press. 20. Zinkernagel, R. M., Sado, T., Althage, A. & Kamisaku, H. (1984) Eur. J. Immunol. 14, 14-23. 21. Muto, M., Sado, T., Aizawa, S., Kamisaku, H. & Kubo, E. (1981) J. Immunol. 127, 2421-2425. 22. Singer, A., Kruisbeek, A. M. & Andrysiak, P. M. (1984) J. Inmmunol. 132, 2199-2209. 23. Murphy, D. B., Herzenberg, L. A., Okumura, K., Herzenberg, L. A. & Mcflevitt, H. 0. (1976) J. Exp. Med. 144, 699-712. 24. Tada, T., Taniguchi, M. & David, C. S. (1976) J. Exp. Med. 144, 713-725. 25. Taniguchi, M., Tokuhisa, T., Itoh, T. & Kanno, M. (1984) J. Exp. Med. 159, 1096-1104. 26. Hayes, C. E., Klyczek, K. K., Krum, D. P., Whitcomb,


R. M., Hullett, D. A. & Cantor, H. (1984) Science 223, 559-563. 27. Ikezawa, Z., Baxevanis, C. N., Arden, B., Tada, T., Waltenbaugh, C. R., Nagy, Z. A. & Klein, J. (1983) Proc.

Proc. Nadl. Acad. Sci. USA 82 (1985)

7067

Natl. Acad. Sci. USA 80, 6637-6641. 28. Steinmetz, M., Minard, K., Horvath, S., McNicholas, J., Srelinger, J., Wake, C., Long, E., Mach, B. & Hood, L. (1982) Nature (London) 300, 35-42.