Hamster-to-rat bone marrow xenotransplantation and humoral graft vs. host disease.

Bone marrow transplantation (BMT) may induce tolerance across xenogeneic barriers. We have established a xenogeneic BMT model where hamster BM is transplanted into splenectomized LEW rat recipients resulting in high levels of engraftment. Unfortunately, graft vs. host disease (GVHD) with severe dermatitis developed in all rat recipients. We were successful in treating or preventing the dermatitis of this xenogeneic GVHD by the use of the T-cell suppressant tacrolimus. However, this compound did not prevent the development of a fatal liver injury in the rat recipients. This study was designed to elucidate the pathogenesis of this liver injury appearing in T-cell suppressed rat recipients of hamster BM. Splenectomized and irradiated (10 Gy) LEW rats received 300 x 106 unfractionated hamster BM cells. These BMT recipients were divided in 3 groups: Group I recipients (n = 8) did not receive further immunosuppression. Group II animals (n = 10) received tacrolimus 1 mg/kg/d for 7 d. Group III recipients (n = 6) were given the same daily dose of tacrolimus on a long-term basis. Chimerism was detected by flow cytometry. Cytotoxicity of recipient's sera against rat and hamster lymph node cells was measured by complement-dependent cytotoxicity (CDC) test. Immunofluorescence was used to detect hamster antirat antibodies on several recipient organs. In Group I, 2 out of 8 animals engrafted (25%) and survived for a median of 21 d showing the severe dermatitis characteristic of GVHD. In group II (n = 10), 9/10 rat recipients engrafted (90%) and survival was increased to a median of 53.7 days. However, these surviving recipients developed fatal GVHD not different from that observed in Group I recipients. All animals in Group III (n = 6) engrafted and did not show the characteristic dermatitis of GVHD. Their survival, however, was shortened to a median of 30.3 d by a severe liver injury. This injury was characterized by hepatocyte necrosis in zones 1 and 2 with polymorphonuclear (PMN) cell infiltration. Deposits of hamster immunoglobulins were present around the necrotic areas and in the portal veins. Moreover, antirat antibodies appeared in the circulation. These antibodies were sensitive to dithiothreitol (DTT) treatment indicating that they were of the IgM class. This study shows that xenogeneic GVHD may have a dual presentation in the hamster-to-rat model: a classical cellular GVHD not distinct to the allogeneic one and a humoral GVHD affecting solely the recipient liver. The degree of humoral injury is potentiated by T-cell suppression.

Role of UW solution and sodium nitroprusside in reperfusion of liver xenografts from guinea-pig to rat.

Guinea-pig livers are poorly reperfused when transplanted into rats. We have observed that, in contrast to that of the rat, the guinea-pig intrahepatic portal vein (PV) has a thick layer of smooth muscle. It is possible that, after perfusion of the liver with ice-cold saline, this could go into spasm, resulting in poor reperfusion. To test this hypothesis, guinea-pig livers were perfused with different solutions stored at varying temperatures and transplanted into LEW rats. To prevent xenograft hyperacute rejection, all xenograft recipients were treated with 80 U/kg cobra venom factor (CVF) i.v. on days -1 and 0. In addition to the percentage reperfusion, PV resistance and recipient survival were also monitored. In group I, liver xenografts perfused with ice-cold saline (4 degrees C) reperfused poorly (20-30%), resulting in the development of portal hypertension (16.5 cmH2O vs. 12 cmH2O in naive LEW rats) and shortened mean survival time (11.7 +/- 4.2 h). In contrast, group II livers perfused with saline at room temperature (23 degrees C) underwent homogeneous reperfusion (98-100%) with no increase in portal vein resistance, indicating that low temperature was the main trigger for the spasm of the PV. Moreover, recipient survival in this group was significantly prolonged to a mean of 22 + 2.6 h (P < 0.01). Although UW solution (group III) and the vasodilator sodium nitroprusside (NP) (group IV) when used alone improved the degree of hepatic reperfusion, it was still not optimal. The supplementation, however, of UW solution with NP in group V animals resulted in homogeneous reperfusion (98%) with no portal hypertension and consistent prolonged graft survival of 21.0 +/- 1.7 h. Therefore, this study has determined that the riddle of the abnormal reperfusion of guinea-pig liver xenografts by rat blood is nonimmune mediated and is due to the spasm of the strong smooth muscle in the PV tree produced by cold perfusates.