Lack of evidence that a transplanted liver causes acute graft-versus-host disease in rats. A comparison of liver and spleen grafts.

In this study, we used rat transplant models to investigate whether--and, if so, to what extent--transplanted liver had the potential to incite graft-versus-host disease, compared with the disease induced by a spleen graft. Livers from PVG(RT1c) rats were transplanted orthotopically into (DAxPVG)F1 (RT1a/c) rats and vascularized spleen grafts from PVG rats were transplanted heterotopically into (DAxPVG)F1 recipients. The intensity of the GVH disease was assessed by the recipients' morbidity and mortality, recipient-type serum class I (RT1Aa) antigen titer, and histological examination. The recipients of spleen grafts died within 14 and 23 days of transplantation; all animals had lost body weight and showed typical GVH signs, such as ear erythema, diarrhea, and alopecia. However, all the recipients bearing liver grafts survived indefinitely and did not demonstrate weight loss or the typical symptoms associated with GVH disease. The skins, tongues, and intestines of the liver-grafted rats were virtually normal at histological examination, whereas the livers, salivary glands, and skins of spleen-grafted rats were infiltrated by immunoblasts. The recipient-type serum RT1Aa antigen titer increased progressively until death in the spleen graft but not the liver graft recipients. These results provide evidence that suggest that transplanted liver is less likely than transplanted spleen to initiate the GVH disease in rats.     

Host immune suppression after small bowel/liver transplantation in rats

Simultaneous liver grafting in the Lewis (RT1¹)-to-DA (RT1^a) rat strain combination protects small intestinal grafts from rejection. The present study examined host immune responses after combined small bowel/liver transplantation (SBL) in this model. Orthotopic liver transplantation and heterotopic small intestinal transplantation were performed simultaneously and compared with isolated small bowel allografts (SBA) and isolated small bowel isografts (SBI). All rats were sacrificed on postoperative day (POD) 7 or 14 for immunological and histological studies. The mean time to rejection of the SBA was 6.6±0.3 days. Incontrast, there was no clinical or histological evidence of intestinal rejection in SBL recipients during the 14 days of follow-up. The SBL recipients showed clinical and histological evidence of graft-versushost disease (GVHD). Lmphocyte proliferation and IL-2 production in response to donor antigens were suppressed after SBL transplantation compared with the SBA or the SBI controls (P<0.05). Cell-mediated cytotoxicity and lymphocytotoxic antibody production against donor cells were also significantly inhibited in the SBL recipients compared with the SBA control group (P<0.05). We conclude that SBL transplantation in the Lewis-toDA rat strain combination: (1) suppresses host alloimmune responses, (2) prevents early intestinal rejection, and (3) favors the development of GVHD.     

A study of tolerance induced by liver transplantation on intestinal graft in the rat]

In some strain combinations of rats, orthotopic liver transplantation (OLT) permits long-term donor-specific survival of fully allogeneic kidney, heart or skin grafts. The difficulties encountered in the clinical situation to obtain tolerance of small-bowel transplantation (SBT), in spite of massive non-specific immunosuppression, led us to study possible liver-induced tolerance in SBT. Inbred DA (RTIa) and PVG (RT1c) rats were used respectively as donors and recipients and divided in two groups: group 1: SIT alone (n = 6); group 2: combined OLT/SBT (n = 6). SIT was performed 17 days after OLT. No immunosuppressive treatment was given to the recipients. Biopsies of small-bowel grafts were performed in both groups at various times after small bowel engraftment. All animals in group 1 showed evidence of acute rejection of the graft between days 6 and 9 post-graft. The histologic pattern of rejection associated lamina propria (LP) mononuclear cell infiltration, crypt lesions and villous atrophy at the end-point of rejection. In group 2, long-term survival (> 100 days) of small bowel grafts was achieved in five of the six animals in spite of strong mononuclear cell infiltration in the LP, which peaked two months after small bowel grafting but then disappeared partially. This striking mononuclear cell infiltrate contrasted with only minor epithelial damage. These data demonstrate that liver grafting can enhance the survival of a small-bowel graft from the same donor in a rat model. Histological findings show that an intense immunological reaction takes place within liver-induced tolerated small-bowel grafts.     

Donor cell infiltration of recipient tissue as an indicator of small bowel allograft rejection in the rat

This study assessed whether screening of host tissues for graft cells could be used as an effective monitor of rejection following small bowel transplantation. Allogeneic rat small bowel transplantation was performed with or without cyclosporin (CyA) immunosuppression and cellular infiltration of host tissues assessed by immunohistological staining. Without immunosuppression, grafts were completely rejected within 1 week. CyA treatment for 7 days preserved the graft for 28 days although there was histological evidence of mild rejection in some of the animals studied. Continuous CyA treatment preserved the graft for up to 56 days. The peripheral lymph nodes and spleens of untreated animals were transiently infiltrated by low numbers of donor cells that disappeared by day 6. There was a marked donor cell infiltration of the lymph nodes and spleens of 7-day, CyA-treated animals that was maintained during the administration of immunosuppressive therapy but that declined thereafter. Continuous CyA treatment sustained donor cell infiltration up to day 56. These findings suggest the presence of donor cells in recipient lymph nodes and spleen to be indicative of effective control of rejection and their disappearance to be predictive of developing rejection responses. Examination of recipient peripheral tissues for donor cells may provide an improved technique for monitoring clinical small bowel transplantation.     

Canine small bowel transplantation. A study of the immunological responses.

A canine small bowel allograft model was used to determine the effects of radiation to the graft in modifying the immunological effects of the passenger leukocytes. When untreated allografts were transplanted, death of the recipient animals occurred at a mean of nine days. The allograft was well-preserved and showed no signs of rejection. The reasons for attributing death to graft-versus-host (GVH) disease are discussed. When allografts were treated with 150 rads prior to transplantation, allograft rejection occurred, with death of the recipient animals at a mean of 9.2 days. This was the only group in which cell-mediated immunity developed. When allografts were treated with 50 rads, prolonged survival of the recipients to a mean of 28 days was noted. It is postulated that in this group a balance was struck between the allograft rejection reaction and GVH disease, with prolongation of allograft survival.     

Migration of host and donor T cells in small bowel transplantation

In this study migration of host and donor CD4⁺ and CD8⁺ T cells in a fully allogeneic model was described and compared with the migration pattern in a graft-versushost reaction (GVHR) model, where the T-cell traffic in the graft served as a physiological control. Heterotopic small bowel transplantations were performed in a rat model, with animals being sacrificed on postoperative days (POD) 2, 3, 4, 5, and 7. Graft and host mesenteric lymph nodes were harvested, homogenized, and stained with monoclonal antibodies against MHC class 1, CD4⁺, and CD8⁺ antigens. The host and donor T cell migration patterns were studied using a doublestaining flow cytometric technique. We found that during the development of rejection, the normal physiological circulation of graft and host T cells was disrupted. In the graft of the allogeneic model, a shift from host cell to graft cell dominance occurred on POD 3–4. This change in migration pattern coincided in the host with a 6 % peak in graft cell infiltration, which disappeared on POD 7. These patterns of T-cell migration may be further explored for diagnostic purposes.     

主要组织相容性复合体Ⅰ类样分子等位基因与活体小肠、肝、肾移植急性排斥反应相关性研究

目的 研究供、受者间主要组织相容性复合体(MHC)Ⅰ类样分子(MICA)等位基因匹配率与小肠、肝、肾移植急性排斥反应的相关性.方法 对收集的4例活体小肠移植供、受者血液标本及组织病理切片,5例活体肝脏部分移植供、受者及6例活体肾脏移植供、受者血液标本,提取基因组DNA并进行HLA配型,针对MICA基因最常见的8个基因型13个等位基因进行聚合酶链反应-序列特异性引物法(PCR-SSP)检测,统计供、受者间等位基因匹配率,并分析其与临床急性排斥反应的相关性.结果 所有移植病例供、受者HLA配型均为半相合状态.供、受者间MICA等位基因匹配率较高者(6/13),其受者显示出相对较轻的临床、病理排斥反应及相对较长的生存时间.而供、受体间MICA等位基因匹配率较低者(≤6/13),其受者则显示出相对较重的临床及病理排斥反应,生存时间亦较短.结论 活体小肠、肝、肾移植中供、受者间MICA等位基因的匹配率高低与移植后受者急性排斥反应强度间存呈负相关,与生存时间呈正相关.     

The pattern of rejection after combined stomach,small bowel,and pancreas transplantation in the rat

This study was designed to investigate whetherin combined stomach, small bowel, and pancreas transplantation allograft rejection occurs in the individual organs concomitantly and with the same intensity. Heterotopic enbloc transplantation of the stomach, small bowel, and pancreas was performed in a Lewis-to-Brown Norway rat combination. Group 1 animals received no immunosuppressive therapy while animals in group, 2 were treated with cyclosproin (10 mg/kg body weight, orally) daily. Grafts were histologically evaluated on the 5th (subgroups 1a and 2a) and 10th (subgroups 1b and 2b) postoperative days. The degree of rejection was defined as moderate, intermediate, or severe according to predefined criteria. The results indicate that the small bowel is more susceptible to rejection than either the stomach or the pancreas. Mucosal biopsies of the stomach are unlikely to provide a reliable guide to rejection in the small bowel.     

Effects of donor pretreatment with antilymphocyte serum and cyclosporin on rejection and graft-versus-host disease after small bowel transplantation in immunosuppressed and nonimmunosuppressed rats

After fully allogeneic small bowel transplantation, both graft-versus-host disease (GVHD) and rejection may occur. Donor pretreatment may prevent GVHD, but this sometimes leads to accelerated graft rejection. To study a possible balance between GVHD and rejection, fully allogeneic total orthotopic small bowel transplantation was performed in rats using the WAG-to-BN donorhost combination. Untreated control grafts were rejected in 16.6±2.7 days (mean ±SEM), and 35% of the animals had mild, transient GVHD. Pretreatment of the donor with antilymphocyte serum on days-2 and-1 before grafting, either intravenously or intraperitoneally, completely eliminated the occurrence of clinical GVHD but led to significantly shortened survival times (12.3±0.8 and 10.3±0.9 days, respectively). Donor pretreatment with 50 mg/kg cyclosporin (CyA) on days-2 and-1 prolonged graft survival significantly to 22.1 days but had no significant effect on the incidence of GVHD. Administration of 25 mg/kg CyA on days 0, 1, 2, 4, and 6 after grafting prolonged survival to 38.3 days with no evidence of GVHD. Pretreatment of the donor with antilymphocyte serum (ALS), combined with the same postoperative, short-term CyA regimen, increased survival to more than 50 days, again with no evidence of GVHD. When CyA was used as both donor pretreatment and postoperative therapy, there was no survival advantage compared to the use of postoperative CyA alone. These results show that an in vivo balance between GVHD and rejection exists and that abrogation of GVHD leads to accelerated rejection. Immunosuppression of the recipient may overrule this accelerated rejection while preserving the beneficial effect of donor pretreatment: elimination of clinical GVHD.     

Intestinal distribution of hyaluronan in small bowel allografting in the rat

Hyaluronan (hyaluronic acid: HA) was demonstrated and quantified in small bowel tissue at different times after small bowel transplantation. Semiallogeneic or semisyngeneic rat models were used to elicit either unidirectional graft rejection or graft-versus-host disease (GVHD). In normal rat small bowel, HA was present in the villous lamina propria and around medium-sized vessels in the interstitium of the crypt area. During graft rejection a cellular infiltrate and edema appeared in the lamina propria in the crypt area where an accumulation of HA was also demonstrated. There was progressive accumulation of HA in the small bowel during rejection, and on day 6 there was a threefold increase compared to the values in syngeneic grafts. The increase in tissue HA was paralleled by an increase in the total water content of the rejecting graft. In specimens from animals suffering from GVHD, no significant changes in water or HA content and distribution were observed until day 12. The data suggest that accumulation of HA might contribute to the pathophysiology of the transplantation edema and that HA might be of potential diagnostic value in differentiating between graft rejection and GVHD.     

短肠综合征合并高位肠瘘患者施行亲属活体小肠移植一例

目的总结短肠综合征合并高位肠瘘患者施行亲属活体小肠移植的经验和体会。方法为1例因肠系膜上动脉栓塞而切除空肠、大部分回肠及右半结肠的患者施行亲属活体小肠移植,供者为患者之子,移植回肠长度为150 cm,供肠热缺血时间1 min,冷缺血时间65 min。受者切除肠瘘,供肠动、静脉分别与受者的腹主动脉和下腔静脉行端侧吻合,供肠的近端与受者的空肠残端行端端吻合,远端侧壁与结肠残端行侧端吻合,移植小肠末端造口,作为观查窗。术后使用他克莫司、霉酚酸酯和甲泼尼龙预防排斥反应,并给予抗感染、抗凝以及胃肠外为主、肠内营养为辅的支持治疗。结果术后移植小肠功能接近正常,能胜任一般的体力劳动。术后110 d,患者因情绪变化突发心脏意外,抢救无效死亡。结论合并肠瘘的短肠综合征并非小肠移植禁忌证,术前充分准备和术后细致观察及管理是成功的关键。     

Mucosal recipient-type mononuclear repopulation and low-grade chronic rejection occur simultaneously in indefinitely surviving recipients of small bowel allografts

Lewis rat recipients of long-term, surviving, orthotopic Brown-Norway rat intestinal allografts, initially treated with cyclosporin A (CyA) or FK 506, were evaluated for their functional capacity and morphology over 1 year after the immunosuppressive therapy had been discontinued. Functional parameters such as nitrogen and fat balances, maltose absorption, blood chemistry, hematologic studies, and the weight gained by the allografted animals did not differ from those of syngeneically grafted or agematched normal animals. Immunohistochemical studies showed that the lamina propria of the allografts was repopulated with recipient MHC class II+mononuclear cells and that a normal distribution of T helper, T suppressor/killer, and IgA+plasma cells had occurred. However, fibrous replacement of the mesenteric lymph nodes and Peyer's patches were detected in all, and an inflammatory obliterative arteriolopathy developed in the mesenteric vasculature of half of the allografted animals. No such findings were observed in recipients of syngeneic grafts. These results demonstrate that the limited use of potent immunosuppressive agents immediately after transplantation averts rejection and is followed by recipient-type mucosal lymphocytic repopulation. Simultaneously, a clinically not recognizable chronic rejection evolves. This suggests that the timely diagnosis of chronic rejection may not be possible with the use of standard tests of gut function and random mucosal biopsies alone.This study was presented in part at the 32nd Annual Meeting of the Society for Surgery of the Alimentary Tract, 21–22 May 1991, New Orleans, Louisiana     

Endotoxin in the peripheral blood during acute intestinal allograft rejection

Intestinal rejection is associated with increased gut permeability and bacterial translocation. The present study examined endotoxin and proinflammatory cytokines in the peripheral circulation during acute intestinal rejection. Heterotopic intestinal transplants were performed using Lewis rats (RT1¹) as donors and DA rats (RT1^a) as recipients. DA rats with intestinal isografts were used as controls. Serum samples were obtained at sacrifice on postoperative days (POD) 7 and 14. Lipopolysaccharide (LPS) was measured using the limulus amoebocyte lysate assay. Interleukin-1 (IL-1) and 6 (IL-6) and tumor necrosis factor- (TNF-) were measured using bioassays. Large amounts of LPS were detected in the serum of intestinal allograft recipients concurrent with the development of graft rejection. Serum IL-6 and TNF- levels were significantly elevated in the allograft recipients on both POD 7 and 14 when compared to DA isografts (P<0.05). Serum IL-1 activity was not detected in the allograft or isograft recipients at either of the two time points. Further studies are warranted to determine the role of intraluminal bacteria and their products in the pathophysiology of intestinal allograft rejection.     

Facial tissue allograft transplantation

A hemifacial allograft transplant model was used to investigate the rationale for development of functional tolerance across an MHC barrier. Thirty hemiface transplantations were performed in five groups of six Lewis (RT1(1)) rat recipients each. Isografts were performed in group 1. Transplants were obtained from semiallogenic LBN(RT1(1+n)) in group 2 and from fully allogenic ACI(RT1(a)) in group 3 donors, which served as allograft rejection controls. Group 4 grafts using LBN donors and group 5 using ACI donors in addition received CsA monotherapy (16 mg/kg/d for 1 week) and maintained at 2 mg/kg/d. Signs of graft rejection were sought daily. Isograft controls survived indefinitely. All nontreated allografts were rejected within 5 to 8 days posttransplant. Eighty-three percent of face-transplant recipients from LBN donors and 67% from ACI donors did not show any signs of rejection up to 270 days and 200 days, respectively. Flow cytometry at day 63 in LBN recipients showed the presence of donor-specific chimerism for MHC class I RT1(n) antigens, namely 3.39% CD4/RT1(n); 1.01% CD8/RT1(n) T-lymphocytes; and 3.54% CD45RA/RT1(n) B-lymphocytes. In ACI recipients the chimerism test revealed 10.55% CD4/RT1(a) and 4.59% of CD8/RT1(a) T-lymphocytes. MLR assay at day 160 posttransplant revealed suppressed responses against LBN donor antigens in group 4, but moderate reactivity to ACI donor antigens in group 5. Functional tolerance toward hemifacial allograft transplants induced across MHC barrier using a CsA monotherapy protocol was associated with the presence of donor-specific chimerism in T- and B-cell subpopulations.     

Integrated intestinal capacity and nutritional status following small bowel transplantation

Successful small bowel transplantation requires normal functional capacity of the graft and unaltered metabolism of the host. Weight gain and wet weight of muscle groups and intra-abdominal fat pads were compared between transplanted, sham-operated, short bowel-operated, and normal rats that were fed either standard chow or fat-enriched (15 %) pellets. Weight gain and wet weight of muscle groups and fat pads for the control, transplanted, and sham-operated rats were identical, while short bowel animals showed reduced weight. Transplanted rats receiving fat-enriched food had lower wet weight of fat pads than control animals on the high-fat diet. We conclude that small bowel transplantation makes it possible to overcome the intestinal failure associated with short bowel syndrome, leading to overall normal weight gain and development of the recipient. However, altered fat metabolism, reflected in changed body composition, was observed in transplanted animals on the high-fat diet. Received: 19 November 1996 Received after revision: 10 April 1997 Accepted: 13 May 1997     

Migration of host and donor T cells in small bowel transplantation

Abstract In this study migration of host and donor CD4⁺ andCD8⁺ T cells in a fully allogeneic model was described and compared with the migration pattern in a graft-versus-host reaction (GVHR) model, where the T-cell traffic in the graft served as a physiological control. Heterotopic small bowel transplantations were performed in a rat model, with animals being sacrificed on postoperative days (POD) 2, 3, 4, 5, and 7. Graft and host mesenteric lymph nodes were harvested, homogenized, and stained with monoclonal antibodies against MHC class I, CD4 ⁺, and CD8 ⁺ antigens. The host and donor T cell migration patterns were studied using a double-staining flow cytometric technique. We found that during the development of rejection, the normal physiological circulation of graft and host T cells was disrupted. In the graft of the allogeneic model, a shift from host cell to graft cell dominance occurred on POD 3–4. This change in migration pattern coincided in the host with a 6 % peak in graft cell infiltration, which disappeared on POD 7. These patterns of T-cell migration may be further explored for diagnostic purposes.     

Graft-versus-host reaction in small bowel transplantation and possibilities for its circumvention

To describe GVHR in small bowel transplantation and its underlying mechanisms and to find methods for circumventing that response, accessory small bowel transplantation was carried out in the rat model. Animals not treated with cyclosporine, irradiation, or removal of the mesenteric lymph nodes of the graft died within 22 days postoperatively due to graft versus host disease. Mesenteric lymph nodes of the graft and recipient spleen and peripheral lymph nodes showed strong immunologic stimulation histologically and high antihost T-cell-mediated cytotoxic antihost reactivity. Seventy-one percent of the animals that had received 15 mg of cyclosporine per kilogram body weight orally survived 150 days after transplantation. After donor irradiation with 50 rads, 77 percent of the recipients survived 120 days. After microsurgical removal of the mesenteric lymph nodes of the graft, 89 percent survived 120 days. We conclude that GVHR plays an important role in small bowel transplantation and that the experimental regimens of donor, graft, and recipient treatment described herein have proved their efficacy for circumventing GVHR.     

Small-bowel transplantation in the rat with a nonsuture cuff technique: Technical and immunological considerations

Abstract. Small-bowel transplantation (SBT) using an nonsuture cuff technique was carried out on 137 rats. Preparation of the donor graft was carried out according to conventional procedures. Graft perfusion was done at a fixed pressure of 35 cm water. The left renal vessels of the recipient were dissected, the native kidney removed, and the graft was connected to the vessels by a nonsuture cuff technique. Of the animals, 92% survived for at least 5 days posttransplant. Three different combinations were investigated: (1) isografts; (2) semisyngeneic grafts from nontreated Lewis(LewisDA) Fl hybrids; and (3) semisyngeneic grafts from rabbit antilymphocyte globulin (ALG)-pretreated Lewis(LewisDA) Fl. In group 1, 80% of the grafts were unaffected after 1 month; flow studies showed slight or no impairment of circulation in the graft. In group 2, the recipients developed clinical signs of graft-versus-host disease (GVHD) after 1 week, and at the end of the 2nd week the animals showed signs of severe illness, leading to death due to GVHD. There was also a higher percentage of complications in this group. In group 3, 65% of the animals died. However, 27% showed intact grafts and no signs of GVDH after 1 month, indicating that antibody pretreatment of the donor may successfully prevent GVHD SBT.     

Inhibition of host-versus-graft and graft-versus-host responses after small bowel transplantation in rats by rapamycin.

The effect of rapamycin (RAPA) on both host-versus-graft (HVG) and graft-versus-host (GVH) immune responses was examined in small bowel transplant models using strongly histoincompatible donor-recipient combinations. Normal Wistar Furth (WFu; RT-1u) recipients rejected Buffalo (BUF; RT-1b) small bowel allografts within a mean survival time (MST) of 10.5 +/- 0.5 days. Administration of RAPA (0.8 mg/kg) by continuous intravenous infusion for 14 days via an osmotic pump prolonged graft survival to 25.0 +/- 4.6 days (P = 0.01). In a second strain combination, the 12.5 +/- 2.2 day survival of Brown Norway (BN; RT-1n) small bowel allografts in Lewis (RT-1l) recipients was prolonged to 21.6 +/- 2.0 and 28.5 +/- 2.8 days by 14 days of i.v. RAPA at doses of 0.8 and 1.6 mg/kg, respectively. In this model RAPA is five times more effective than cyclosporine, which at 4.0 mg/kg prolongs BN small bowel allografts in Lewis recipients to 21.6 +/- 6.3. To isolate HVG and GVH immune responses, (BN x Lewis)F1 hybrid rats served as the graft donor or host, respectively. In the HVG model, (BN x Lewis)F1 small bowel allografts, which were rejected by normal Lewis recipients at 12.2 +/- 3.6 days, were prolonged to 40.8 +/- 5.8 days (P = 0.001) by RAPA (0.8 mg/kg x 14 days). In the GVH model, the ability of Lewis small bowel allografts to produce severe GVH disease in untreated (BN x Lewis)F1 recipients at 12.3 +/- 2.8 days was delayed to 21.3 +/- 5.2 days by 0.8 mg/kg RAPA (P = 0.025). Thus, RAPA protects small bowel allografts more effectively against HVG than GVH immune responses.     

In vitro allograft irradiation prevents graft-versus-host disease in small-bowel transplantation

In small-bowel transplantation, the transfer of large numbers of donor lymphocytes with the intestinal allograft may provoke a lethal graft-versus-host reaction. The effectiveness of allograft irradiation in vitro as a method of preventing graft-versus-host disease (GVHD) was studied in a rat model of small-bowel transplantation, with the Lewis----Lewis X Brown Norway F1 hybrid strain combination. Cold harvested small-bowel allografts were irradiated immediately prior to heterotopic or orthotopic transplantation. Animals that had received heterotopic allografts irradiated with 0, 250, or 500 rad all died of GVHD after 14.4 +/- 3.0, 15.0 +/- 1.3, and 14.2 +/- 1.9 days, respectively. None of the animals that had received allografts treated with 1000 rad developed clinical or pathologic evidence of GVHD, however, and all survived for more than 6 months (P less than 0.001). Allograft function was studied in animals that underwent orthotopic transplantation. Recipients of nonirradiated orthotopic allografts all died of GVHD after 14.0 +/- 0.7 days, whereas recipients of allografts irradiated with 1000 rad all survived for more than 5 months (P less than 0.001). After 120 days, weight gain (51.8 +/- 11.7%), serum albumin (3.9 +/- 0.7 g/dl), serum triglycerides (67.0 +/- 24.3 mg/dl), CBC, and differential in these animals were not statistically different from those in either age-matched isograft recipients or normal animals, and when the rats were sacrificed, irradiated allografts showed no changes suggestive of radiation injury. These results indicate that irradiation of small-bowel allografts in vitro prevents development of GVHD, and that this can be achieved at a dose which does not cause injury to or malfunction of the allograft.