Transplantation in miniature swine. VI. Factors influencing survival of renal allografts.

Renal allografts were performed between and among animals from three herds of miniature swine that were selectively inbred to homozygosity at the major histocompatibility complex, MSLA. The results suggest several genetic factors which influence the survival of renal allografts in these animals. As expected, the major histocompatibility complex (MHC) was of dominant importance, and all MSLA-mismatched grafts were rejected promptly (12 +/- 3.7 days). Some MSLA-matched grafts were also rejected (30 +/- 15.0 days), indicating that non-MSLA loci also determine antigens which can lead to kidney rejection. Other MSLA-matched grafts were accepted indefinitely. At least one immune response gene that determined ability to reject kidneys across non-MSLA differences seemed to be segregating in our swine population. Animals that had accepted MSLA-matched renal grafts for extended periods demonstrated markedly prolonged survival of subsequent donor skin grafts compared to skin graft survival across the same non-MSLA difference in normal animals. This finding suggests that failure to reject kidneys across non-MSLA differences indicates systemic tolerance, and that there may be a relationship between the induction of such tolerance and the proposed immune response gene controlling rejection.     

The long-term fate of fresh and frozen orthotopic bone allografts in genetically defined rats

Fresh and frozen orthotopic iliac crest bone grafts in rats were studied histologically for determination of the long-term effects of histocompatibility matching and the freezing process on orthotopic bone graft incorporation. Grafts exchanged between groups of inbred rats, syngeneic or differing with respect to major or minor histocompatibility loci, were studied histologically at 20, 30, 40, 50, and 150 days after bone transplantation. A numerical histologic scoring system was developed and used by three observers for evaluation of coded hematoxylin and eosin sections. All frozen graft groups had the same fate regardless of histocompatibility relations between donors and recipients, and all grafts were inferior to fresh syngeneic grafts. Both fresh allograft groups received similar scores and initially at 20 and 30 days had scores similar to those of the fresh syngeneic groups. In the later intervals, however, the fresh allografts were inferior to the fresh syngeneic grafts and similar to the frozen groups. This is consistent with an older model describing two distinct phases of osteogenesis. In the long term, frozen syngeneic and fresh and frozen allografts across major and minor histocompatibility barriers were comparable, but all were significantly inferior to fresh syngeneic bone grafts.     

Down-regulation of the immune response to histocompatibility antigens and prevention of sensitization by skin allografts by orally administered alloantigen.

The effects of oral administration of major histocompatibility antigens on the alloimmune response have not been investigated. Lymphocytes from inbred LEW (RT1u) rats that were pre-fed allogeneic WF (RT1l) splenocytes exhibited significant antigen specific reduction of the mixed lymphocyte response in vitro and delayed-type hypersensitivity response in vivo, when compared with unfed controls. In an accelerated allograft rejection model, LEW rats were presensitized with BN (RT1n) skin allografts 7 days before challenging them with (LEW x BN)F1 or BN vascularized cardiac allografts. While sensitized control animals hyperacutely reject their cardiac allografts within 2 days, animals prefed with BN splenocytes maintained cardiac allograft survival to 7 days, a time similar to that observed in unsensitized control recipients. This phenomenon was antigen-specific, as third-party WF grafts were rejected within 2 days. Immunohistologic examination of cardiac allografts harvested on day 2 from the fed animals had markedly reduced deposition of IgG, IgM, C3, and fibrin. In addition, there were significantly fewer cellular infiltrates of total white blood cells, neutrophils, macrophages, T cells, IL-2 receptor-positive T cells, and mononuclear cells with positive staining for the activation cytokines IL-2 and IFN-g. On day 6 posttransplant, the grafts from fed animals showed immunohistologic changes typical of acute cellular rejection usually seen in unsensitized rejecting controls. Feeding allogeneic splenocytes prevents sensitization by skin grafts and transforms accelerated rejection of vascularized cardiac allografts to an acute form typical of unsensitized recipients. Oral administration of alloantigen provides a novel approach to down-regulate the specific systemic alloimmune response against histocompatibility antigens.     

Histocompatible miniature swine: an inbred large-animal model

Three herds of miniature swine, each homozygous for a different set of alleles at the major histocompatibility complex (MHC), and five intra-MHC recombinant strains, have been reported by the authors' laboratory. One herd (SLAdd) was selected for further inbreeding to achieve a histocompatible line. It has undergone seven additional generations of sequential brother-sister or father-daughter matings (termed G7). To determine the level of histocompatibility of these animals, the authors performed skin and heart transplantation without immunosuppression. In contrast to MHC-matched, minor antigen-mismatched animals that rejected skin in 11 days (median survival time [MST], n=6) and hearts in 35 days (MST, n=4), G7 animals accepted skin for greater than 340 days (>340, >448, and >677 days) and hearts for greater than 265 days (>265 and >269 days). Nevertheless, rejection of second grafts could be induced by sensitization, indicating that weak minor antigens remain, requiring further inbreeding to achieve full histocompatibility. We conclude that G7 animals are sufficiently inbred to accept first set skin and heart grafts indefinitely.     

Reconsideration of the lymphatic drainage of the rat testis

Gynogenetic diploid individuals were produced in an anuran amphibian, Xenopus laevis, and their response to skin grafts exchanged among siblings was studied. All skin grafts exchanged among nongynogenetic sibling froglets, as well as those from genetically unrelated donors, were rejected within 30 days. More than half (57%) of the gynogens that received grafts from sibling partners exhibited a prolonged survival (over 30 days), including long-term survival of over 120 days in 13%. The skin grafted from genetically unrelated froglets onto Nieuwkoop and Faber stage 42-56 larvae and onto perimetamorphic stage 58-65 animals was rejected within 30 days. Similarly, most (96%) of the skin grafts from outbred sibling froglets onto larvae at these stages were rejected acutely or subacutely (12-39 days). However, the skin grafted from sibling froglets to gynogens at larval stage 42-56 and perimetamorphic stage 58-61 enjoyed a long-term survival significantly more frequently (81%) than that in the final metamorphic (stage 64-65) counterparts (57%). These results support the view that in the adult Xenopus allograft responses are reactions to a single MHC as well as to cumulative, multiple minor H-locus barriers. The results also suggest that in larval stages the responses against minor H-locus barriers are generated only mildly.     

Permanent acceptance of both cardiac and skin allografts using a mild conditioning regimen for the induction of stable mixed chimerism in mice

Patients who are receiving an organ transplant nowadays are sentenced to the life-long administration of immunosuppressive drugs, which have serious side effects. The reliable induction of donor-specific tolerance therefore remains a major goal in organ transplantation. Previously, we have developed a sublethal, non-myeloablative murine model in which permanent mixed, multilineage chimerism and donor-specific tolerance are established. Our model involves engraftment of fully allogeneic T cell depleted donor bone marrow cells in low dose irradiated and anti-CD3 treated major histocompatibility complex (MHC)-disparate recipient mice. To investigate whether vascularized organ grafts are accepted in our model, we performed heterotopic heart transplantations in our mixed chimeric mice. Chimeric mice permanently accepted hearts from the bone marrow donor (>130 days) and rapidly rejected third party-type allografts (median survival time 9 days). Untreated control recipient mice rejected both donor- and third party-type allografts. In addition, mice that accepted their cardiac grafts, donor-specific acceptance of skin grafts was observed. In conclusion, the establishment of stable mixed chimerism with this low-toxicity regimen resulted in permanent donor-specific acceptance of vascularized organ as well as skin grafts across a full MHC barrier.     

Acceptance of cardiac allografts by lethally irradiated rats repopulated with syngeneic bone marrow. Role of class I and class II alloantigens

LEW rats (RT1l) that are lethally irradiated and repopulated with syngenic bone marrow accept WF (RT1u) cardiac allografts. If bone marrow repopulation is delayed for two days after irradiation and operation, grafts containing passenger leukocytes (nonperfused grafts) are generally rejected, but perfused grafts, which have fewer passenger leukocytes, are accepted. If bone marrow is given on the same day as irradiation and surgery, then both perfused and nonperfused grafts are accepted. The difference in acceptance of grafts by recipients that are repopulated on day 0 as opposed to day 2 depends on class II alloantigens, because grafts that are similar to LEW for class II antigens are not rejected by day-2-repopulated recipients. Also, the acceptance of totally major histocompatibility complex (MHC)-mismatched cardiac allografts by day-0-repopulated recipients is influenced by a radiation-resistant host cell. Splenectomy of the irradiated and repopulated recipient prevents tolerance induction unless syngeneic irradiated spleen cells are returned to the recipient. Thus class II alloantigen disparities appear to be a major barrier to tolerance induction in the system of total body irradiation and syngeneic bone marrow reconstitution, although proper timing of bone marrow administration can minimize rejection of completely MHC-mismatched grafts.     

Induction of donor-specific unresponsiveness to cardiac allografts in rats by pretransplant anti-CD4 monoclonal antibody therapy

In the present report a monoclonal antibody designated OX-38 directed against the rat CD4 molecule was tested for its ability to prolong the survival of heterotopic vascularized rat heart allografts transplanted across major histocompatibility barriers. Fluorescence-activated cell-sorter analysis showed that administration of OX-38 selectively depleted 80-95% of CD4+ cells from peripheral blood of treated rats. The immunosuppressive effects of OX-38 in vivo were verified by suppression of an antibody response against OX-38 itself as a heterologous protein immunogen. Recipient rats received OX-38 antibody as a single agent given in pretransplant regimens. Nine of 12 treated rats have maintained heterotopic abdominal heart allografts for greater than 175 days. Control rats that did not receive antibody therapy rejected their grafts within 14 days. Rats that maintained heart allografts for greater than 100 days accepted second donor strain hearts but rejected third-party heart grafts transplanted into the femoral space. Anti-CD4-induced allograft unresponsiveness persisted for at least 90 days following surgical removal of donor tissue and retransplantation of a second donor-matched heart. These results indicated that transient, pretransplant therapy with monoclonal antibodies directed against the CD4+ lymphocyte induced specific, long-lasting unresponsiveness to fully MHC-mismatched cardiac allografts in rats without additional immunosuppression.     

Orthotopic corneal transplantation in mice--evidence that the immunogenetic rules of rejection do not apply.

The fate of orthotopic corneal transplants has been studied in inbred strains of mice. Using a surgical technique that achieves > 95% success of syngeneic cornea grafts, it was determined that a high proportion of orthotopic cornea allografts were accepted indefinitely, irrespective of the degree of immunogenetic disparity between graft donor and recipient. Grafts that succumbed to irreversible rejection developed extensive corneal edema and intrastromal neovascularization as harbingers of corneal opacity and endothelial cell failure. The highest rate of rejection occurred among grafts that confronted their hosts with multiple minor histocompatibility antigens, with or without major histocompatibility antigens. Much lower rates of rejection (< 35%) were observed when the donors of the grafts differed from recipients at class I and/or class II major histocompatibility loci. Corneal grafts that confronted their hosts with class II MHC alloantigens alone experienced early, acute inflammation, and eventually developed stomal neovascularization, but only a small minority of these grafts were eventually destroyed. Allogeneic corneas that were transplanted orthotopically into eyes of presensitized mice were uniformly subjected to an acute rejection process that produced opacity within three weeks; however, in a minority of instances, the inflammation and opacity subside, and after eight weeks the grafts displayed a clear, nonvascularized appearance. The high rate of success of even grossly histoincompatible orthotopic corneal allografts in mice resembles the extraordinary success of unmatched allogeneic corneas transplanted into human eyes. The results are discussed in terms of the possible mechanisms that permit orthotopic corneal allografts to enjoy significantly better survival than orthotopic grafts of other types of solid tissues.     

Skin graft rejection caused by the maternally transmitted antigen Mta

Mta is a medial histocompatibility antigen of the mouse. It does not stimulate a primary mixed lymphocyte response and stimulates only a very weak secondary response. Primary skin grafts are rejected with a mean survival time of 59 days by Mta- NZB recipients, and 39 days by recipients of the C57BL/6 background. Rejection is accelerated in recipients primed against Mta with a skin graft or cells, especially when these differ by multiple minor histocompatibility antigens. Mta is determined by a maternally transmitted, extrachromosomal genetic element, so backcross mice reject skin from their inbred, homozygous paternal strain. Mta, therefore, constitutes a new exception to the classic laws of transplantation.     

Fresh allograft of intact growth plate into immature articular cartilage defects in rats]

In three inbred rat strains with a distinct major histocompatibility complex (MHC), MHC-matched allografts (Fisher, RT1l to Lewis, RT1l, n = 39) or MHC-mismatched allografts (Brown Norway, RT1n to Lewis, RT1l, n = 38) of fresh intact cartilage obtained from the tibial epiphyseal growth-plate were grafted into the defects in the femoral articular surface of immature rats. In the controls (n = 72), similar defect on the joint cartilage were made with no epiphyseal plate grafts. The present histological observation was continued throughout one year after transplantation. Total success rate of the repair was significantly better in MHC-matched allografts (92%) than in MHC-mismatched allografts (61%, p less than 0.001) or controls (19%, p less than 0.001). The MHC-matched allografts showed no signs of tissue rejection. However, the MHC-mismatched allografts showed evidence of lymphocyte infiltration and 13 (81%) of 16 grafts were rejected from 12 to 48 weeks after operation. These findings indicate that MHC-matched allografts of intact growth-plate have a potential to repair defects in immature articular cartilage over a long period of time, but MHC-mismatched allografts will eventually be rejected.     

Evidence that epidermal alloantigen Epa-1 is an immunogen for murine heart as well as skin allograft rejection

Epa-1 is a non-H-2 mouse alloantigen defined by MHC-restricted, CD8+ cytotoxic T cells. In vitro it is a strong determinant for the lysis of epidermal cells, fibroblasts, and macrophages but not lymphocytes, and in vivo it functions as a target for skin allograft rejection and cutaneous graft-versus-host reactions. Genetically, Epa-1 appears to be the nonpolymorphic manifestation of a loss mutation. The establishment of C3H.Epa-1 (Epa), an Epa-1+ congenic strain on the Epa-1- C3H/HeJ (C3H) inbred strain background, facilitated the investigation of the role of Epa-1 in skin and heart allograft rejection. C3H females and males rejected first-set Epa skin grafts with median survival times (MSTs) of 20 and 30 days, respectively. However, there was a strong factor of immunization, because all second-set skin allografts were rejected by hosts of both sexes within 10 days. In contrast, all Epa hosts of both sexes permanently accepted C3H skin allografts, consistent with Epa-1 arising from a loss mutation. C3H hosts of both sexes rejected primarily vascularized first-set Epa heart allografts in similar tempo to first-set Epa skin allografts, with MSTs of about 30 days. However, in contrast to the accelerated rejection of skin allografts, sensitized C3H hosts rejected Epa heart allografts in chronic fashion, with some transplants showing very prolonged survival. Thus, Epa-1 is a relatively strong determinant of skin allograft rejection but a weaker determinant of heart allograft rejection.     

Experimental vascularized bone allografting

Presented here is a compendium of studies investigating the fate of vascularized bone allografts. The first set of experiments employ the posterior rib graft in two canine models. The rib-to-mandible model was used to evaluate the rejection phenomena of vascularized bone allografts in an outbred dog model. This ascertained the time course of rejection and histological characteristics of the grafts. Immunosuppression of the graft recipients was attempted with azathioprine and cyclosporine. The results demonstrated that azathioprine was not an effective immunosuppressant, whereas cyclosporine resulted in survival of cortical osteons. The use of the vascularized rib allograft, with and without azathioprine, to bridge the defect in the dog femur was met with failure. Further studies employed a genetically defined rat model to determine the effect of different histocompatibilities on the survival of vascularized knee allografts. Grafts were transplanted from Lewis rats to syngeneic Lewis rats as isografts and to Fischer-344 rats (F-344) and Brown-Norway rats (BN) as allografts. Grafts across a major histocompatibility barrier to BN were rejected by 7 days, whereas grafts across a weak histocompatibility barrier to F-344 were rejected more slowly. The use of cyclosporine in this model abrogated the rejection response when administered to both groups continuously. However, a short course of cyclosporine was effective in preventing rejection in the F-344 animals. Efforts to induce tolerance by blood transfusions, from the donor strain or from a third-party donor, were not effective in preventing rejection.     

Long-term acceptance of renal allografts following prenatal inoculation with adult bone marrow

BACKGROUND: The aim was to investigate if intravascular in utero injection of adult bone marrow into swine fetuses could lead to macrochimerism and tolerance to the donor. METHODS: Outbred Yorkshire sows and boars screening negative for MHC allele SLA of MGH miniature swine were bred. A laparotomy was performed on the sows at 50 days gestation to expose the uterus. Bone marrow harvested from SLA miniature swine was T-cell depleted and injected intravascularly into seventeen fetuses. Flow cytometry was performed to detect donor cells (chimerism) in the peripheral blood after birth. Mixed lymphocyte reactions (MLR) and cell-mediated lympholysis (CML) assays were used to assess the response to donor MHC. Previously frozen skin grafts from the bone marrow donor were placed on the offspring from the first litter. Donor-matched renal transplant from SLA donors were performed on chimeric swine, with and without a short 12-day course of cyclosporine, and one nonchimeric littermate. RESULTS: Nine inoculated offspring demonstrated donor cell chimerism in the peripheral blood and lymphohematopoietic tissues. All animals with detectable chimerism within the first three weeks were consistently nonreactive to donor MHC in vitro. Animals challenged with donor skin grafts displayed prolonged graft survival without producing antidonor antibodies. All chimeric animals accepted donor-matched kidney allografts, even one without cyclosporine. The kidney in the nonchimeric littermate rejected by day 21. CONCLUSIONS: Transplantation of allogeneic adult bone marrow into immunocompetent fetal recipients resulted in chimerism. In utero inoculation led to operational tolerance to the donor's major histocompatibility antigens and long-term acceptance to organ allografts.     

The differential effects of donor versus host Langerhans cells in the rejection of MHC-matched corneal allografts

The fate of MHC-identical, multiple minor H-disparate corneal grafts was examined in the rat. Although skin grafts exchanged between LEW and F344 rats were invariably rejected, only 26% of the corresponding corneal grafts underwent rejection. The immunologic privilege of the minor H-disparate corneal grafts was due, at least in part, to the absence of donor-derived Langerhans cells. Corneal grafts were normally devoid of donor-derived Langerhans cells; however, grafts pretreated with latex beads became infiltrated with donor-derived Langerhans cells and were rejected by 59% of the naive minor H--compatible recipients. By contrast, untreated LEW corneal grafts underwent rejection in 26% of the naive F344 hosts even though the grafts became heavily infiltrated with host-derived Langerhans cells. The immunologic privilege of minor H-disparate corneal grafts was not the result of efferent blockade or suppression of the immune response. F344 hosts bearing long-term surviving LEW corneal allografts were challenged with LEW skin grafts. In all cases, orthotopic skin grafts were rejected acutely. Moreover, all previously clear corneal grafts underwent rejection following skin graft rejection. Thus, the unique absence of donor-derived Ia+ passenger cells and the avascular graft bed conspire to provide the primary minor H-disparate corneal graft with an immunologic privilege not shared by other organ grafts.     

The immediately vascularized skin allograft.

A model for immediate vascularization of skin was devised to examine one of the possible explanations for the differential survival rates of transplants of freely grafted skin and organs, i.e., the increased vulnerability of skin to early ischemic necrosis prior to revascularization. Female Fischer (F) rat skin was transplanted beneath the kidney capsule of tolerant female Lewis (LEW) recipients. This skin healed in and initially appeared to be normal grossly and microscopically. In 27 rats, after 30 days, the composite grafts were excised, and immediately transplanted by means of vascular anastomosis into normal LEW recipients (syngeneic to the kidney portion of the graft and allogeneic to the skin). For 5 days after transplantation of the composite graft, the skin appeared to be normal with minimal or no inflammation in a panel of 11 recipients. From the 6th to 11th day, an active rejection reaction developed at the skin-kidney interface in a panel of six rats. In 10 rats, in which the composite grafts remained in their secondary hosts for 12 to 21 days, rejection of the skin was complete. The renal portion of all composite grafts appeared to be normal histologically. All recipients of composite grafts rejected subsequent orthotopic F skin grafts in an accelerated manner, with median survival times of 8.2 +/- 0.3 days compared to 11.5 +/- 0.7 days in untreated F leads to LEW controls, demonstrating that the skin on the composite graft was fully immunogenic. These results demonstrate that immediately vascularized skin allografts between rats compatible at the major Ag-B1 locus will still be rejected within 2 weeks compared to survivals of up to 48 weeks in renal allografts in the same histocompatibility combinations, suggesting that anatomical factors are not sufficient to account for the differences in survival times between skin and solid organs.     

Immune privilege in the testis. I. Basic parameters of allograft survival

The fate of solid tissue allografts--skin or parathyroid glands--implanted in the interstitial tissue of the testis was investigated using inbred rats. The results affirm that the testis is an immunologically privileged site despite its efficient lymphatic drainage. Skin allografts survived at least several days longer than orthotopic grafts of similar size, whether major histocompatibility complex (MHC)-compatible or MHC-incompatible, and failed to induce an alloantibody response in most recipients or to prime for secondary antibody responses on rechallenge. Further assessment employed parathyroid grafts that allowed appraisal of their function and its duration, by monitoring serum calcium levels. Most intratesticular MHC-incompatible parathyroids survived for at least twice as long as control grafts in nonprivileged sites, with a median survival time (MST) of 41 days--and one-third of the grafts functioned at 100 days. MHC-compatible grafts fared even better (MST of 60 days), some surviving more than 400 days. Most F1 hybrid grafts survived virtually indefinitely. Splenectomy 5-23 days prior to implantation had a beneficial rather than a detrimental effect on the privilege afforded intratesticular parathyroid allografts. Allograft rejection was accompanied by antibody production in only one-half the animals. Grafts that had undergone functional rejection at the time of recovery usually had an intense mononuclear cell infiltrate, and long-term surviving grafts displayed varying degrees of cellular infiltration among cords of healthy, functional chief cells. Accepted parathyroids were destroyed by active immunization of the host but were unaffected by passively administered alloantibodies. The possible mechanisms controlling graft rejection in this unique privileged site are discussed.     

Transplantation in miniature swine. VII. Evidence for cellular immune mechanisms in hyperacute rejection of renal alografts.

Renal allografts were performed in miniature swine that were identical at their major histocompatibility locus and were presensitized by skin grafts from their prospective renal donors. All of these renal grafts were rejected in a hyperacute or markedly accelerated manner compared to the survival of comparable grafts in nonsensitized animals. Studies directed at the mechanism of this rejection revealed no circulating recipient antidonor antibodies by several serological assays. In contrast, mixed lymphocyte cultures (MLCs) and cell-mediated lympholysis (CML) assays demonstrated marked recipient antidonor lymphocyte reactivity that appeared after skin grafts, diminished during the tenure of the renal graft in the host circulation, and reappeared after removal of the rejected kidney. These results suggest that cellular immune mechanisms may plan a role in the accelerated rejection of major histocompatibility complex (MHC)-identical renal allografts.     

Experimental studies on vascularized allogeneic joint transplantation in rats]

Examination was made on the effects of Cyclosporine A, an immunosuppressive drug, on vascularized joint allografts. Genetically inbred DA and Lewis rats were used and a model of knee joint orthotopical transplantation across the major histocompatibility barrier was developed. Nine non-immunosuppressive allografts failed to acquire bony union. Twenty-seven short term immunosuppressive rats acquired solid union, but following withdrawal of immunosuppression, the grafts were soon rejected and joint function was lost due to pathological fractures of the grafted bone or joint instability. Thirteen allografts of continuous immunosuppression at a dose of 5 mg/kg/day showed no rejection, though the bone marrow did, and there was no impairment of function. The continuous administration of low and non-toxic dose of Cyclosporine A was necessary for maintaining grafted joint functions.     

Marrow grafts between canine litter-mates homozygous or heterozygous for lymphocyte-defined histocompatibility antigens.

Hemopoietic grafts following 1,200 R of total body irradiation were carried out between canine littermates homozygous or heterozygous for lymphocyte defined (LD) antigens of the major histocompatibility complex (MHC). In all but five of the 25 pairs studied, LD homozygous dogs were also homozygous for serologically defined (SD) antigens of the MHC. Results of transplants were compared with previous results obtained in littermate pairs matched or mismatched for the MHC. Two groups of recipients were studied. Of 14 LD homozygous recipients in group 1 given grafts from LD heterozygous donors, 12 died between 8 and 193 days and two survived more than 238 and 627 days. The most frequent cause of death was graft-versus-host disease (GVHD). Survival was significantly shorter (P less than 0.005) than that of dogs given grafts from matched littermates and not different (P congruent to 0.5) from that of dogs given grafts from mismatched littermates. Survival of 11 LD heterozygous recipients in group 2 given grafts from LD homozygous donors was not different from that of dogs in group 1 (P congruent to 0.5). The results indicate that the LD loci detected by mixed leukocyte culture are not the principal determinants within the MHC that are responsible for GVHD.