Modulation of MHC class I antigen decreases pancreatic islet immunogenicity

JOURNAL

OF SURGICAL

RESEARCH

46,317-321 (1989)

Modulation of MHC Class I Antigen Decreases Pancreatic Islet lmmunogenicity PETER G. STOCK, M.D., NANCY L. ASCHER, M.D., PH.D., SALLY CHEN, B.S., GINNY L. BUMGARDNER, M.D., M. JANE FIELD, B.S., AND DAVID E. R. SUTHERLAND, M.D., PH.D. Department

of Surgery, University

of Minnesota,

Minneapolis,

Minnesota

55455

Presented at the Annual Meeting of the Association for Academic Surgery, Salt Lake City, Utah, November 16-19, 1986

Pretreatment regimens directed at reducing the immunogenicity of pancreatic islets have emphasized the elimination or alteration of the major histocompatibility complex (MIX?) class II-positive dendritic cells within the islet. Unfortunately, the efficacy of such pretreatment regimens has been extremely variable and the relative contribution of the dendritic cells to the overall immunogenicity of pancreatic islets has remained ambiguous. Recent evidence has suggested that the MHC class I antigen present on the endocrine cells within the islets may play an important role in the alloimmune response. This study utilized the in dtro mixed lymphocyte-islet coculture system to determine if pretreatment of whole islets with an anti-MHC class I monoclonal antibody specific to the donor strain would block the generation of cytotoxic T lymphocytes (CTL) in the in vitro mixed lymphocyte-islet coculture. Pretreatment of BlO.BR (H2k) and DBA/BJ (H-2d) islets with an allospecific antiMHC class I monoclonal antibody blocked the generation of allospecific CTL when the pretreated islets were placed into coculture with CW7B1/6 (H-2b) splenocytes. If such a pretreatment regimen is similarly effective in duo, it could potentially be used as an antirejection strategy in 0 iogs Academic POW, rnc. pancreatic islet allotransplantation.

INTRODUCTION The emphasis on reducing pancreatic islet immunogenicity prior to allotransplantation has principally been directed at eliminating or altering the function of the major histocompatibility complex (MHC) class II-positive passenger leukocytes or dendritic cells within the islets [l]. A variety of pretreatment regimens have been used, including ultraviolet irradiation [2], tissue culture [3, 41, and antibodies directed against donor strain MHC class II antigen [5] or dendritic cells [6]. The efficacy of such pretreatment regimens has been extremely variable, and in some models islet grafts from which Ia+ cells have been eliminated still incite a rejection response, undiminished from that seen with untreated islets [7]. The relative contribution of dendritic cells to the overall im-

munogenicity of pancreatic islets has not been precisely quantitated. In order to more clearly define the effects of depletion of MHC class II-positive dendritic cells on the immunogenicity of pancreatic islets, we developed a murine in vitro mixed lymphocyte-islet coculture (MLIC) system (see Fig. 1). With this system we can eliminate certain variables which complicate the interpretation of in vim experiments. In initial experiments, we demonstrated that isolated islets were capable of evoking an MHC-specific cytotoxic T lymphocyte (CTL) response in the MLIC [El, 91. The degree of allospecific cytotoxicity generated in vitro correlated with the rejection rates observed during in uiuo transplant experiments. BlO.BR islets, rejected by 80% of C57BL/6 recipients, stimulated a high degree of cytotoxicity when cocultured with C57BL/6 splenocytes; while C57BL/6 islets, rejected by only 20% of BlO.BR recipients, stimulated a low degree of toxicity when cocultured with BlO.BR splenocytes [lo]. Subsequent experiments demonstrated that whole islets depleted of MHC class II-positive cells by pretreatment with anti-Ia antibody stimulated a slightly decreased cytotoxic response in the MLIC when compared to untreated islets [ 111. The fact that immunodepleted islets were capable of generating a strong in vitro allospecific response was consistent with in uiuo transplant data by Gores et al. [7], in which BlO * BR islets depleted of MHC class II+ cells were rejected at the same rate as untreated islets when transplanted beneath the kidney capsule of C57BL/ 6 mice. In further MLIC experiments, designed to determine the relative immunogenicity of the individual cellular components of pancreatic islets, the cytotoxicity generated by fluorescence-activated cell sorted (FACS-IV) pure /3 cells (MHC class I-positive cells only) was compared to nonpurified dispersed islet cells (MHC class I-positive and MHC class II-positive cells) [12]. Surprisingly, the purifled fraction of @cells stimulated as strong a cytotoxic response as did the nonpurified dispersed islet cell population. Furthermore, the stimulation of CTL by both the purified p cells and the nonpurified dispersed islet cells was blocked by the addition of an allospecific monoclonal

317 All

Copyright 0 1989 rights of reproduction

OOZZ-4804/89 $1.50 by Academic Press, Inc. in any form reserved.

318

JOURNAL

OF SURGICAL

C57BL/6

RESEARCH:

(H-2b

VOL. 46, NO. 4, APRIL

)

1989

DBA/PJ (H-2 di or BlO.BR (H-2 ) Y/3

Re&mndr (H-2b) SplenWYter, (isolatedfrom

SplW

U l0.0.*:

Alloqecific 5’Wabelkd target cells added in different effectohtarget ratios

1O:l

so:1

1OO:l

1

1

1

cunwx4hom

Measure release of 51Cr into Supernatant FIG. 1.

Methodology

of mixed lymphocyte-islet

antibody directed against stimulator MHC class I antigen. These initial experiments suggested that the MHC class I antigens, present on allogenic parenchymal cells, have an important role in stimulating the maturation of CTL precursors in MLIC and can do so independent of the MHC class II-positive passenger leukocytes or dendritic cells within the islets. The present experiment was designed to determine if pretreatment of whole islets with an anti-MHC class I antibody specific to the donor strain would block the generation of CTL in the MLIC. Such an approach, if successful, could potentially be used as an antirejection strategy for pancreatic islet allotransplantation. MATERIALS

AND

METHODS

Islet isolation. Murine pancreatic islets were isolated from DBA/SJ (H-2d) or BlO.BR (H-2k) mice (Jackson Laboratories, Bar Harbor, ME) using a modification of Brunstedt’s technique [13]. This method yields intact, clean pancreatic islets free of contaminating exocrine and vascular debris. Pancreas pieces were placed into vials containing 8 ml collagenase solution (1 mg/ml). After vigorous horizontal shaking (200 cpm) in a 37°C water bath for 11-13 min, the partially digested pancreases were reincubated in collagenase solution (0.5 mgjml). After each digestion, the liberated tissue suspension was washed once with sterile Hanks’ balanced salt solution (HBSS) and free islets were meticulously handpicked under a dissecting microscope (X40). After the initial islet segrega-

culture.

tion, the islet preparations were cleaned by four or five transfers between black-bottomed 100 X 15-mm petri dishes containing HBSS with 1% bovine serum albumin. This method yields loo-150 islets/mouse pancreas. Preparation of responder lymphocytes for use in the MUC. Responder lymphocytes were prepared from spleens which were removed by aseptic technique from C57B1/6 mice (H-2b; Jackson Laboratories). The spleens were teased apart in cold RPM1 1640 media. The clumps were removed by passage through nylon netting, and single cell suspensions were obtained. Mixed lymphocyte-islet coculture (MLIC) technique. Cytotoxic T lymphocytes were generated in 0.2-ml microculture wells containing 50 stimulator DBA/2J islets (H2d) or BlO.BR islets (H-2k) and 5 X lo5 responder C57Bl/ 6 splenocytes. Some of the islets were pretreated with an anti-class I monoclonal antibody prior to placement into the coculture system as described below. Cells were cultured in RPM1 1640 supplemented with 10% fetal calf serum (heat-inactivated), 1% Pen-Strep, and Hepes 15 W/liter for 5 days at 37’C, 5% CO,. Sensitized splenocytes were harvested from the MLIC at 5 days and used as effector cells to determine generation of allospecific cytotoxicity against 61Cr-labeled targets in a standard Cytotoxicity assay. Pretreatment of islets with anti-class I monoclinal antibody. The following anti-MHC class I antibodies were used to pretreat some of the islets prior to placement into the MLIC: (1) Hybridoma 34-l-23, anti-H-2KdDd (Organon Teknika, Durham, NC); and (2) Hybridoma 3-83 P,

STOCK

ET AL.: PANCREATIC

anti-H-2KkDk (Organon Tecknika). The islets were preincubated with the anti-class I monoclonal ab (1:4 dilution) for 12 hr in RPM1 1640 supplement with 10% fetal calf serum (heat-inactivated), 1% Pen-Strep, and Hepes 15 &/liter for 12 hr at 37”C, 5% COz. The pretreated islets were then washed in RPM1 1640 3X prior to placement into the MLIC. TargetceZlsforcytotoxicityassuy. TheP815(H-2d)tumor cell line was used as the allospecific target for the DBA/BJ (H-zd) strain, and BlO.BR concanavalin A (Con A) blasts (H-2k) were used as the allospecific target for the BlO.BR (H-2k) strain. P815 tumor cell lines were maintained in vitro (minimal essential media (MEM), 37”C, 10% COz) until use in the cytotoxicity assay. BlO.BR Con A blasts (H-2k) were prepared as follows: Spleen cells (100 X 106) were mixed in a 50-ml flask with 0.25 ml of Con A (final concentration, 5.0 pg/ml), 5 ml of fetal calf serum (final concentration, lo%), and 45.4 ml of MEM. The cells were incubated for 48 hr at 37”C, washed twice in D-(methyZ)mannoside (0.3 n&f), and then washed a third time in MEM. Cytotoxicity assay. The cytotoxicity assay is modified from Brunner et al. [14]. Cells from tumor lines or from the Con A blast preparations were labeled with 51Cr (sodium chromate; New England Nuclear, Boston, MA) for 1 hr at 37°C. Following incubation, the targets were washed three times in MEM and fetal calf serum. Two thousand target cells (in 100 ~1) were placed in conical bottom microtiter plates (in triplicate), and 2.0 X lo5 or 1.0 X lo5 effector cells (in 100 ~1) harvested from MLIC at Day 5 were added to achieve effector:target ratios of 1OO:l and 50:1, respectively. The plate was centrifuged at 200g for 4 min and incubated for 4 hr at 37°C. The plate was then centrifuged at 2000g for 10 min; 100 ~1 of supernatant was removed from each well and placed in 12 X 75-mm glass tubes for counting in a Beckman Biogamma II. Maximal release from ‘Cr-labeled target cells

TABLE Generation

ISLET

319

IMMUNOGENICITY

was determined in triplicate in 100 111of supernatant from 2000 target cells lysed with cetrimide (hexadecyltrimethylammonium bromide). Another triplicate of 2000 target cells/well was incubated for 4 hr with medium alone, and 100 ~1 of supernatant was removed from each well to determine the spontaneous release of ‘*Cr. Cytotoxicity was expressed as a percentage according to the following formula: Experimental (counts/min) - spontaneous (counts/min) Maximum (counts/min) - spontaneous (counts/min)

X 100 = % cytotoxicity.

Mean cytotoxicity and standard error were calculated for each triplicate at a given effector:target ratio. The Student t test was used to determine the significance between control and experimental groups. RESULTS The data are summarized in Table 1. DBA/BJ and BlO.BR islets were capable of stimulating the generation of an allospecific cytotoxic T cell response in the MLIC. Pretreatment of both strains of islets with an allospecific anti-MHC class I monoclonal antibody (moAb) prior to placement into the MLIC caused a significant reduction in the generation of CTL. In both cases, the generation of allospecific CTL was nearly abrogated by anti-MHC class I pretreatment. Furthermore, pretreatment of the islets with a third party anti-MHC class I moAb had no effect on stimulating the generation of allospecific CTL. DISCUSSION These data suggest that the generation of an allospecific CTL response against pancreatic whole islets can be ab-

1

of Allospecific T-Lymphocytes in the Mixed Lymphocyte-Islet Coculture Pretreatment with Anti-MHC Class I Monoclonal Antibody

following

Percentage cytotoxicity (*SE) Effector:Target Donor strains of stimulator islets DBA/PJ DBA/W DBA/ZJ BlO.BR BlO.BR

Pretreatment

regimen”

None Anti MHC Class I H-2d moAb Anti-MHC Class I H-2k moAb None Anti-MHC Class I H-2k moAb

Allospecific

target

P815 (H-2d) P815 (H-zd) P815 (H-zd) BlO.BR Con-A blast (H-2’) BlO.BR Con-A blast (H-2’)

1OO:l 30.0 3.4 35.4 30.4 6.7

f f + + f

0.05 o.ot 0.8 3.9 3.6t

at Day 5 Ratio 5o:l 21.8 + 1.7 2.3 k 1.4t 26.8 + 1.4 -

Note. Coculture contained 5 X 10’ C57BL/6 (H-2b) responder splenocytes plus indicated group of stimulators, either DBA/SJ (H-zd) or BlO.BR (H-2’). ’ Pretreatment regimen involved 12 hr of incubation with a 1:4 dilution of anti-MHC class I moAb. Islets were subsequently washed and placed into the MLIC. * P < 0.05 control vs pretreated group.

320

JOURNAL

OF SURGICAL

RESEARCH:

rogated by pretreating the islets with a monoclonal antibody directed against the MHC class I antigen of the stimulator strain. This is consistent with previous data demonstrating that generation of an allospecific CTL response against either purified /3 cells (MHC class I+, II-) or nonpurified dispersed islet cells (MHC class I+, class II+) could be blocked with a monoclonal antibody directed against stimulator MHC class I antigen [12]. It is also consistent with in uiuo [7] and in vitro [ll] data demonstrating that elimination of MHC class II+ cells from the islet failed to significantly decrease the immunogenicity of the islet. Although previous efforts at decreasing the immunogenicity of islets prior to allotransplantation have emphasized the role of the MHC class II+ dendritic cells in the initiation of the immune response, this data suggests that the MHC class I antigen may play a very important role in the initiation of the immune response. There are several studies providing evidence that donor MHC class I antigen can be recognized in the absence of MHC class II-positive cells. In vitro studies (mixed lymphocyte culture) have demonstrated that accessory cells of either stimulator’(donor) or responder (recipient) origin can present alloantigen to responder T lymphocytes [ 1517]. Depletion of donor MHC class II+ cells from the islets prior to transplantation eliminates only the donortype antigen-presenting cells, but responder strain accessory cells are still available for alloantigen presentation. Although indirect presentation of antigen by responder accessory cells has been referred to as the “alternative pathway,” this pathway is in fact the principal pathway for most immune responses, and in these experiments it is obviously responsible for generation of allospecific CTL against pancreatic islets. The fact that donor MHC class I antigen can stimulate an allospecific CTL response in the absence of donor MHC class II antigen is not surprising. The “alternative” pathway is presumably the major pathway for the generation of an immune response against environmental pathogens. Other evidence for recognition of MHC class I antigen by immunoreactive T cell populations comes from tolerance models. In these models, antigen-specific tolerance was induced by sensitizing hosts with MHC class I+, class II- cellular populations including red blood cells, platelets, and hepatocytes [la-201. The activation of suppressor T cells in these models clearly involved recognition of donor MHC class I antigen in the absence of donor MHC class II antigen. A role for MHC class I antigen in the generation of the immune response is suggested by studies from Milton et al. [21]. They demonstrated a 20-fold increase in MHC class I antigen in a rejecting kidney allograft during the first few days following transplantation. Although this may represent a response to immune injury, it is also possible that the increase in MHC class I antigen plays a role in initiating or amplifying the rejection process. Marked enhancement of MHC class I antigen has been detected in the prediabetic BB rat, and the increase in

VOL.

46, NO. 4, APRIL

1989

MHC class I antigen was noted before the appearance of MHC class II antigen expression and insulitis [22]. Finally, Sollinger et al. [23] have recently reported a correlation between the decreased immunogenicity of thyroid grafts after culture in a high oxygen environment with down-regulation of MHC class I antigen. In summary, we have demonstrated that pretreatment of pancreatic islets with an allospecific MHC class I monoclonal antibody directed against donor islets abrogated the generation of CTL in the MLIC. If such a pretreatment regimen is similarly effective in uiuo, it could be potentially used as an antirejection strategy. Modulating the MHC class I antigen with a noncomplement fixing anti-MHC class I antibody might block the MHC class I signal and avoid complement-mediated destruction of the endocrine cells in uiuo. Another possibility could involve blocking the MHC class I signal with an F(ab):! fragment of the appropriate anti-MHC class I antibody. Although most of the previous attempts at decreasing pancreatic islet immunogenicity prior to transplantation have been directed at modulating or eliminating the MHC class II-positive dendritic cells within the islets, our data suggest that the dendritic cells are not the only or even the major contributor to the immunogenicity of the islet. Rather, the MHC class I-positive endocrine cells themselves are immunogenic, and future attempts at decreasing the immunogenicity of islets should involve modulation of class I antigens. ACKNOWLEDGMENTS This research was sponsored in part by NIH funding from the Juvenile Diabetes Foundation.

Grant DK 13083 and

REFERENCES Lafferty, K. J., Prowse, S. J., and Simeonovic, C. J. Immunology of tissue transplantation: A return to the passenger leukocyte concent. Annu. Rev. Immunol. 1: 143,1983. of rat islet 2. Lau, H., Reetsma, K., and Hardy, M. Prolongation allograft survival by direct ultraviolet irradiation of the graft. Science 223: 607,1984. of pancreatic islets. 3. Lacy, P. E., and Davie, J. M. Transplantation Annu. Rev. Zmmunol. 2: 183,1984. 4. Bartlett, S. T., Naji, A., Silvers, E. K., and Barker, C. F. Influence of culturing on the functioning of major-histocompatibility-complex-compatible and incompatible islet grafts in diabetic mice. Transplantation 36: 687, 1983. 5. Faustman, D., Hauptfeld, W., Lacy, P., and Davie, J. Prolongation of murine islet allograft survival by pretreatment of islets with antibody directly to Ia determinants. Proc. Natl. Acad. Sci. USA 78: 5156,19&l. 6. Faustman, D. L., Steinman, R. M., Gabel, H. M., Hauptfeld, V., Davie, J. M., and Lacy, P. E. Prevention of rejection of murine islet allografts by pretreatment with antidendritic cell antibody. Proc. Natl. Acad. Sci. USA 81: 3864, 1984. 7. Gores, P. F., Sutherland, D. E. R., Platt, J. L., and Bach, F. H. Depletion of donor Ia cells before transplantation does not prolong islet allograft survival. J. Zmmunol. 137: 1482, 1986. 8. Stock, P., Ascher, N., Kaufman, D., and Sutherland, D. Mixed 1.

STOCK

ET AL.: PANCREATIC

ISLET

10.

11.

12.

13.

14.

15.

16.

Stock, P. G., Ascher, N. L., Kaufman, D. B., Chen, S., Meloche, M., Field, M. J., and Sutherland, D. E. R. In uitro generated allospecific cytolytic T lymphocytes injure pancreatic islets. J. Surg. Res. 46: 74, 1988. Stock, P. G., Sutherland, D. E. R., Kaufman, D., Meloche, M., Chen, S., Field, J., and Ascher, N. L. Mixed lymphocyte-islet coculture predicts islet allograft rejection in Go. Transplant Proc. 2O(Suppl. 1): 47, 1988. Stock, P. G., Ascher, N. L., Platt, J. L., Kaufman, D. B., Chen, S., Field, M. J., and Sutherland, D. E. R. Effect of immunodepletion of major histocompatihility complex class II-positive cells from pancreatic islets on generation of cytotoxic T-lymphocytes in mixed islet-lymphocyte coculture. Diabetes 38(Suppl 1):157, 1989. Stock, P. G., Meloche, M., Ascher, N. L., Chen, S., Bach, F. H., and Sutherland, D. E. R. Generation of allospecific cytolytic Tlymphocytes stimulated by pure pancreatic B-cells in absence of Ia+ dendritic cells. Diabetes 38(Suppl 1):161, 1989. Brundstedt, J. Rapid isolation of functionally intact pancreatic islets from mice and rata by Percoll gradient centrifugation. D&&e Metub. 6: 87, 1980. Brunner, K., Mandel, J., Cerottini, J., et al. Quantitative assay of the lytic action of immune lymphoid cells of 51Cr-labelled allogenic target cells in vitro; Inhibition by isoantibody and by drugs. Zmmurwtbgy 14: 181,1968. Golding, H., and Singer, A. Role of accessory cells processing and presentation of shed H-2 alloantigens in allospecific T lymphocyte responses. J. Zmmunol. 133: 597, 1984. Weinberger, O., Germain, R. N., Springer, T., and Burakoff, S. J.

321

Role of syngeneic Ia+ accessory cells in the generation of allospecific CTL responses. J. Zmmunol. 129: 694, 1982.

islet lymphocyte culture as a model for pancreatic islet immunogenicity and cell mediated injury. Transplant Proc. 19: 4345,1987. 9.

IMMUNOGENICITY

17. Mizouchi, T., Golding, J., Rosenberg, A. S., Glimcher, L. H., Malek, T. R., and Singer, A. Both L3T4t and Lyt-2+ helper T cells initiate cytotoxic T lymphocyte responses against allogeneic major histocompatibility antigens but not against trinitrophenyl-modified self. J. Ewp. Med. 162: 427, 1985. 18. Foster, S., Cranston, D., Wood, K. J., and Morris, P. J. Production of indefinite renal allograft survival in the rat by pretreatment with viable and nonviable hepatocyte or liver membrane extracts. Transplantation 45(l): 228, 1988. 19. Wood, K. J., Evins, J., and Morris, P. J. Suppression of renal allograft rejection in the rat by class I antigens on purified erythrocytes. Transplantation 39(l): 56, 1985. 20. Goeken, N. E. Human suppressor cell induction in uitro: Preferential activation by class I MHC antigen. J. Zmmunol. 132(5): 2291,1984. 21. Milton, A. D., Spencer, S. C., and Fabre, J. W. Detailed analysis and demonstration of differences in the kinetics of induction of class I and class II major histocompatibility complex antigens in rejecting cardiac and kidney allografts in the rat. Transpluntation

41:499,1986. 22.

Ono, S. J., Issa-Chergui, B., Colle, E., Seemayer, T., Guttmann, R. D., and Fuks, A. Class I major histocompatibility complex gene expression is enhanced in the pancreatic islets of prediabetic BB rats. Transplant. Proc. 19: 3187, 1987.

23,

Sollinger, M. W., Landry, A. S., and Hullett, D. A. Mechanism of enhanced thyroid allograft survival after organ culture. In “Proceedings, 14th Annual Meeting of the American Society of Transpkmt Surgeons. [Abstract]