Induction Of Graft-versus-host Reaction Research Articles

Kinetics of CD4 + and CD8α + T-cell subsets in graft-versus-host reaction (GVHR) in ginbuna crucian carp Carassius auratus langsdorfii

We have previously demonstrated the presence of graft-versus-host reaction (GVHR) in fish employing a model system of clonal triploid ginbuna and tetraploid ginbuna-goldfish hybrids. To elucidate the role of CD8α + T cells in the induction of GVHR, we investigate the kinetics of CD4 + and CD8 + T-cell subsets in GVHR along with the pathological changes associated with GVH disease (GVHD) in ginbuna. GVHR was not induced with a leukocyte fraction lacking CD8α + T cells separated by magnetic cell sorting. Ploidy and immunofluorescence analysis revealed that CD4 + and CD8α + T cells from sensitized donors greatly increased in the host trunk kidney, constituting more than 80% of total cells 1–2 weeks after donor cell injection, while those from non-sensitized donors constituted less than 50% of cells present. The increase of CD4 + T cells was greater and more rapid than that of CD8α + T cells. The number of donor CD4 + and CD8α + T cells was highest in trunk kidney followed by spleen. Increases in donor CD4 + and CD8α + T cells were also found in liver and PBL, although the percentages were not as high. Pathologic changes similar to those in human and murine acute GVHD were observed in the lymphoid organs as well as target organs such as skin, liver and intestine, including the destruction of cells and tissues and massive leukocyte infiltration. The pathologic changes became more severe with the increase of CD8α + T cells. These results suggest that donor-derived CD8α + T cells play essential roles for the induction of acute GVHR/D in teleosts as in mammals.

Persistence of allospecific helper T cells is required for maintaining autoantibody formation in lupus-like graft-versus-host disease

Induction of a graft-versus-host (GVH) reaction (GVHR) in non-irradiated (C57BL/10ScSn x DBA/2)F1 mice (BDF1) with DBA/2 lymphoid cells leads to chronic GVH disease (GVHD). One of the pathological alterations of this type of GVHD is hyperplasia of host B cells with production of lupus-like autoantibodies. This hyperstimulation of host B cells has previously been demonstrated to be induced by alloreactive donor T helper cells that were also proposed to maintain it. We provide three pieces of experimental evidence in support of this concept. First, treatment of mice with chronic GVHD by injection of monoclonal anti-Thy-1.2 antibodies, performed at week 6 after the injection of C57BL/6 lymphoid cells into (C57BL/6 x C57BL.bm12)F1 mice led to a significant decrease in the titre of anti-nuclear antibodies. Second, CD4+ donor T cells persisted in BDF1 mice with GVHD (GVHF1) for at least 10 weeks after the induction of GVHR; these T cells showed alloreactive helper activity against H-2b MHC determinants of the opposite parent in vitro. Third, T cells of GVHF1 mice, obtained 2 months after the induction of GVHR and transferred into normal secondary recipients, induced signs of chronic GVHD in DBF1 but not in DBA/2 mice. The combined results show that persisting donor T helper cells in GVHF1 mice retain their alloreactivity towards H-2 class II antigens for a long time after the induction of GVHR and they strongly suggest that these T cells are also the driving force behind the production of lupus-like autoantibodies at the late stage of chronic GVHD.

Progression of autoimmune-mediated hepatic lesions in a murine graft-versus-host reaction by neutralizing IL-10

C57BL/6 (B6) spleen T cells which were injected into major histocompatibility complex (MHC) class II-disparate (B6. C- H- 2 bm/2 ×B6) F1 hybrid mice induced autoimmune graft-versus-host reaction (GVHR). Early production of interferon (IFN)-γ and delayed production of interleukin (IL)-10 might play an important role in the formation of GVHR hepatic lesions. To clarify whether blocking of IL-10 deteriorate autoimmune-mediated hepatic lesions induced by GVHR, and to elucidate the change of the T helper (Th)1/Th2 cytokines in the liver, anti-IL-10 monoclonal antibodies (mAbs, 500 μg) were given 4 h before the induction of GVHR. We evaluated the change of splenomegaly and GVHR-induced hepatic lesions. The changes of the expressions of IFN-γ and IL-4 mRNA isolated from liver-infiltrating lymphocytes were measured by real-time polymerase chain reaction (PCR). In GVHR with anti-IL-10 mAbs mice splenomegaly and periportal cellular infiltration was significantly increased compared with those of GVHR mice. In these mice, both IFN-γ and IL-4 mRNA expression levels were significantly elevated by the neutralization of IL-10. These findings suggest an important role of IL-10 in murine GVHR due to MHC class II disparity. IL-10 may play a crucial role in down-regulating autoimmune-related hepatic lesions.

Nitric oxide increases the amount of murine cytomegalovirus-DNA in mice latently infected with the virus.

At 5 wk after the intraperitoneal infection of murine cytomegalovirus (MCMV) in adult (C3H/He x BALB/c) F1 (CBF1) mice, MCMV-DNA was noted mainly in the salivary glands (SG). Since graft-versus-host reaction (GVHR) is well known as a major risk factor for CMV reactivation, parental BALB/c spleen cells were then intravenously transferred to these mice to induce GVHR. A high copy number of full-length MCMV-DNA was thus observed in the lungs as well as the hearts and livers at 4 to 6 wk after the cell transfer, whereas the copy numbers in SG after the cell transfer were comparable to those before the transfer. It was therefore suggested that the amount of MCMV-DNA noted in these organs had been increased by GVHR. However, no transcripts of MCMV-DNA could be detected. Since a marked expression of mRNA of inducible nitric oxide synthetase (iNOS) was noted in the lungs and the hearts during the course of GVHR, the role of NO in the increase of MCMV-DNA after the cell transfer was next examined. The administration of L-arginine, a substrate for NO, increased the amount of MCMV-DNA in the lungs and the hearts after the induction of GVHR. In contrast, the material which suppressed iNOS induction inhibited the emergence of MCMV-DNA. Therefore, the NO generated during the course of GVHR increased the amount of the viral DNA in the lungs and the hearts of the mice latently infected with MCMV. If the increase in the amount of viral genome is the initial step of viral reactivation from the latent state, then NO would thus trigger the initial step of the reactivation of the DNA virus. As the lungs are the major organ to produce NO after systemic stimulation, the observations of this study may explain why the lungs are the major site of CMV latency and why GVHR triggers CMV reactivation. These findings are thus considered to demonstrate that NO plays a novel role in viral infection.

Monoclonal antibody against lymphocyte function-associated antigen 1 inhibits the formation of primary biliary cirrhosis-like lesions induced by murine graft-versus-host reaction

Interaction between intercellular adhesion molecule 1 (ICAM-1) and lymphocyte function-associated antigen 1 (LFA-1) might be involved in the pathogenesis of liver diseases. We investigated whether monoclonal antibodies (mAbs) against these two adhesion molecules could inhibit the formation of primary biliary cirrhosis (PBC)-like lesions in an animal model using graft-versus-host reaction (GVHR) with major histocompatibility complex class II disparity. PBC-like hepatic lesions such as cellular infiltration of portal area and nonsupprative destructive cholangitis (NSDC) were generated by injecting spleen T cells of C57BL/6 (B6) mice into (B6. C-H-2bm12 X B6) F1 mice. In the liver of these mice, increased number of LFA-1-positive cells and enhanced expression of ICAM-1 on sinusoidal endothelial cells and bile duct epithelial cells were observed immunohistochemically, when compared with F1 mice without GVHR. Hepatic lesions of these mAb-treated mice were almost completely inhibited in these mice compared with GVHR mice. Furthermore, we studied to determine which anti-LFA-1 mAb or anti-ICAM-1 mAb was essential to inhibit the hepatic lesions. Mice solely treated with anti-LFA-1 mAb showed significant inhibition of hepatic lesions, whereas treatment with anti-ICAM-1 mAb could not inhibit the lesions. Despite the inhibition of hepatic lesions, induction of GVHR and production of antimitochondrial antibodies were not impaired in mAb-treated mice. We conclude that LFA-1 mediates cell infiltration into the liver in this murine model of GVHR and suggest a possible therapeutic role of mAbs to this adhesion molecule in selective autoimmune liver diseases. (Hepatology 1996 Oct;24(4):888-94)

Monoclonal antibody against lymphocyte function-associated antigen 1 inhibits the formation of primary biliary cirrhosis-like lesions induced by murine graft-versus-host reaction

Interaction between intercellular adhesion molecule 1 (ICAM-1) and lymphocyte function-associated antigen 1 (LFA-1) might be involved in the pathogenesis of liver diseases. We investigated whether monoclonal antibodies (mAbs) against these two adhesion molecules could inhibit the formation of primary biliary cirrhosis (PBC)-like lesions in an animal model using graft-versus-host reaction (GVHR) with major histocompatibility complex class II disparity. PBC-like hepatic lesions such as cellular infiltration of portal area and nonsupprative destructive cholangitis (NSDC) were generated by injecting spleen T cells of C57BL/6 (B6) mice into (B6. C-H-2bm12 X B6) F1 mice. In the liver of these mice, increased number of LFA-1-positive cells and enhanced expression of ICAM-1 on sinusoidal endothelial cells and bile duct epithelial cells were observed immunohistochemically, when compared with F1 mice without GVHR. Hepatic lesions of these mAb-treated mice were almost completely inhibited in these mice compared with GVHR mice. Furthermore, we studied to determine which anti-LFA-1 mAb or anti-ICAM-1 mAb was essential to inhibit the hepatic lesions. Mice solely treated with anti-LFA-1 mAb showed significant inhibition of hepatic lesions, whereas treatment with anti-ICAM-1 mAb could not inhibit the lesions. Despite the inhibition of hepatic lesions, induction of GVHR and production of antimitochondrial antibodies were not impaired in mAb-treated mice. We conclude that LFA-1 mediates cell infiltration into the liver in this murine model of GVHR and suggest a possible therapeutic role of mAbs to this adhesion molecule in selective autoimmune liver diseases.

Experimental Studies of Immunologically Mediated Enteropathy

We have attempted to investigate the relative roles of specific cytotoxic T lymphocytes (CTL) and allospecific suppressor T cells (Ts) in the systemic and intestinal manifestations of acute graft-versus-host reaction (GvHR) in mice. Treatment of adult (C57B1/10 x DBA/2)F1 (BDF1) mice with the suppressor cell-specific toxin 2'-deoxyguanosine (dGuo) inhibited the weight loss and mortality which normally occur after induction of GvHR and C57Bl donor cells. dGuo also delayed the development of a destructive enteropathy as typified by jejunal villus atrophy. Paradoxically, dGuo completely prevented villus atrophy during an acute GvHR in neonatal (CBA x BALB/c)F1 hosts, despite having only a slight ability to inhibit the systemic disease. In both models, dGuo had no effect on the generation of splenomegaly or anti-host CTL, and dGuo-treated mice with GvHR actually had increased proliferative alterations in the intestine, as assessed by crypt hyperplasia. In parallel, dGuo prevented the loss of NK cells which normally occurs in acute GvHR. Thus dGuo inhibits many of the destructive features of systemic and intestinal GvHR without affecting the development of CTL. We conclude that a dGuo-sensitive mechanism causes the transition from a proliferative to a destructive GvHR.

Effect of cyclosporin A treatment on the enteropathy of graft-versus-host reaction in the rat: a quantitative study of intestinal morphology, epithelial cell kinetics and mucosal immune activity.

Mucosal graft-versus-host reaction (GvHR) of the small intestine exemplifies an immunologically mediated enteropathy that is associated with expansion of mucosal mast cells (MMC). Quantitative measures of intestinal morphology, epithelial cell kinetics and mucosal immune activity were used to assess the effect of the immunosuppressive agent, cyclosporin A (CyA), in ameliorating this enteropathy and on increased activity of MMC in the jejunum. GvHR was induced in two groups of PVGU x PVGC rats by irradiation (4.50 Gy) and intravenous injection of PVGC spleen cells (150 x 10(6)). One group remained untreated, while a second group of eight rats was treated with a 50 mg/kg dose of CyA subcutaneously given daily for the first 3 days and then every second day, and which had commenced the day before induction of GvHR. On day 14, all animals were killed. Treatment with CyA prevented intestinal crypt hyperplasia but did not affect villus length, and normalized the crypt cell production rate (CCPR) from 38 to 15 cells/crypt/h (P less than 0.0001). CyA reduced the number of MMC and jejunal content of the MMC associated protease, rat mucosal mast cell protease II (RMCPII). Mean serum RMCPII concentration was reduced from 302 (s.d. = 112) in GvHR animals to 10 (s.d. = 6) ng/mL in GvHR/CyA-treated rats (P less than 0.0001). We conclude that CyA ameliorates the enteropathy of GvHR and depresses the activation of MMC, as evident by the strongly depressed serum RMCPII concentration.

A study on graft-versus-host reaction (GVHR) by Simonsen's splenomegaly assay. Cells and antigen systems involved in induction of GVHR.

Cells and histocompatibility antigen systems involved in graft-versus-host reactions (GVHR) were analyzed using Simonsen's splenomegaly assay employing various combinations of donor and F1 hybrid recipients mice. Most of the cells proliferating in spleens of mice undergoing GVHR were J11d+, and had histological features of cells of the hematopoietic lineage. The proportions of CD3+ T cells were decreased in the spleens. Disparity at minor histocompatibility determinants of AKR, I-E and H-2D regions between B10.A(4R) donors and (4R X AKR) F1 recipients evoked only negligible GVHR. On the contrary, disparity at H-2K and/or I-A regions appeared to be sufficient to permit induction of full GVHR. When surface markers of donor spleen cells were analyzed, it was shown that Thy-1+ and/or MEL-14+ cells caused a strong effect on GVHR. Further, either CD4+ or CD8+ T cell subset could induce significant GVHR. However, synergistic influences of these two T cell subsets on one another in GVHR were observed. The present results raise the possibility of using Simonsen's assay along with a number of reagents to identify the contribution of subsets of T lymphocytes and in analyzing precise contributions of cellular components from both donor and recipient, and also of the target antigen systems of the recipient that contribute to early events involved in GVHR.

Experimental studies of immunologically mediated enteropathy. Development of cell mediated immunity and intestinal pathology during a graft-versus-host reaction in irradiated mice.

The intestinal component of a graft-versus-host reaction (GvHR) provides a useful experimental model to elucidate the pathogenesis of clinical enteropathies which cause villus atrophy and crypt hyperplasia and which are associated with a local immune response. One to three days after induction of GvHR in heavily irradiated (CBAxBALB/c)F1 mice, a proliferative form of enteropathy developed. Compared with controls, these mice had increased counts of jejunal intraepithelial lymphocytes and had a four-fold increase in crypt cell production rate as well as an increase in crypt length. These changes were accompanied by a marked enhancement of splenic natural killer cell activity. After day three, the crypt cell production rate fell to zero and cytotoxic T lymphocytes (CTL) which could lyse targets of host origin appeared. In parallel, mice with GvHR developed significant villus shortening and their clinical condition deteriorated. Further experiments showed that increased counts of intraepithelial lymphocytes, villus atrophy and crypt hyperplasia also occurred in grafts of fetal CBA intestine implanted under the kidney capsule of (CBAxBALB/c)F1 mice with GvHR. As these grafts are syngeneic to the injected CBA spleen cells, they should not be attacked by anti-host cytotoxic T lymphocytes. We suggest that the proliferative and destructive components of enteropathy in GvHR are caused by lymphokines released by an anti-host delayed type hypersensitivity reaction.

Structural Changes of the Murine Bone Marrow in Graft-Versus-Host Reaction

A graft-versus-host reaction (GVHR) was induced in neonatal (OH/He x C57B1/6)F1 mice by injection of adult OH/He spleen cells, and the effect of GVHR on the bone marrows was examined with light and electron microscopes. All the bone marrows became hypoplastic or aplastic from the beginning of the 2nd week after induction of GVHR. Degenerative changes of the hematopoietic cells and phagocytic activities of the macrophages and reticular cells, which were noticed as early as on the 3rd day, were most obvious in the 2nd week. Erythropoiesis appeared depressed through the course of observation but regeneration came forth in the 4th week. Granulopoiesis was also reduced. The number of megakaryocytes was apparently decreased. But no correlation was recognized between the megakaryocyte counts and bone marrow cellularity. Mitosis was inhibited in each hematopoietic cell series. In the markedly hypoplastic bone marrow, fibrin deposits and an increase of extracellular microfibrils were recognized, and the spaces left by the decrease of hematopoietic cells were filled with markedly dilated sinusoids. Proliferation of lymphocytes and plasma cells was not observed. Extramedullary hematopoiesis in the spleen and liver sustained no suppression. The pathogenesis of hematopoietic failure in the bone marrow of the GVHR and its implication to aplastic anemia in man are discussed.

Increased resistance of (CBA x C57BL/6)F1 hybrids to the systemic graft versus host reaction during regeneration of the spleen

Partial splenectomy was performed on (CBA x C57BL/6)F1 hybrid mice 2, 12, 15, and 20 days before induction of the graft versus host reaction (GVHR) and 2 and 10 days after injection of parental immunocompetent cells. Recipients with an intact spleen, and those undergoing total splenectomy or a mock operation served as the control. Hybrids with a regenerating spleen up to 12 days of regeneration were shown to have increased resistance to GVHR, whereas splenectomy increased the resistance of the hybrids to GVHR only if carried out 2 days before induction of GVHR or 2 days after it.

Niridazole inhibition of local graft-versus-host reaction in rat muscle

The immunosuppressive properties of niridazole, an antischistosomal drug, were investigated in a model of local graft-versus-host reaction (GVHR) which involves injections of lymphoid cells into the muscle of rats immunosuppressed with cyclophosphamide. Rats were pretreated with niridazole before induction of GVHR: 100 mg/kg on Day −1, 100 mg/kg on Days −3 and −1, or 200 mg/kg on Days −5, −3, and −1. Control rats received niridazole alone, cyclophosphamide alone, or no drug treatment. Seven days after the intramuscular injection of 10 × 10 6 human peripheral blood lymphocytes, the lesions in the injected muscle sites were measured and examined histologically. Results showed that donor cells elicited a GVHR in the muscle of cyclophosphamide-treated rats. A macroscopic lesion had developed at the inoculated area, and microscopy revealed the characteristic invasion and destruction of muscle tissue by the injected lymphoid cells. In cyclophosphamide-treated rats given niridazole, GVHR was abolished or diminished in intensity. Histopathology of the lesions showed minimal invasion of muscle tissue with no proliferation of the infiltrating cells or muscle destruction. The injected cells were unable to induce GVHR in rats receiving niridazole or in untreated rats. In these two groups, gross lesions were absent, but histologically a cellular infiltrate consisting mainly of macrophages with no evidence of tissue damage was observed. These results suggest that niridazole treatment modifies the local GVHR presumably by action on donor T cells.

CYTOTOXICITY IN GRAFT-VERSUS-HOST REACTION

Graft-versus-host (GVH) reactions were induced in adult F1 hybrid mice with the i.p. injection of parental strain spleen cells. Peritoneal exudate and spleen cells of the F1 hybrids taken 8 days after the induction of GVH reaction had a nonspecific in vitro cytotoxic effect which was measured by using 51Cr-labeled target cells of parental genotype. The cytotoxic cells in the peritoneal exudates were shown to be macrophages which adhered to plastic surfaces and were sensitive to the toxic action of crystalline silica particles. Moreover, the injection of partially purified syngeneic macrophages into the F1 hybrids undergoing GVH reactions increased the cytotoxic activity of the peritoneal exudate cells obtained from these animals. These results suggest that during GVH reaction host macrophages are activated into a state of nonspecific cytotoxicity.

Allogeneic stimulation of anti-hapten response in mice.

(CBA × C57B1)F<sub>1</sub> mice were primed to the hapten NNP on either of the non-cross-reacting carriers fowl <i>γ</i>-globulin (F<i>γ</i>G) or ovalbumin (OV) and 4 weeks later were given 50 × 10<sup>6</sup> CBA or (CBA × C57B1)F<sub>1</sub> spleen cells. When challenged on the same day with NNP.F<i>γ</i>G, the two groups of mice undergoing a graft-versus-host reaction (GVHR) showed higher numbers of antibody-producing cells to both NNP and F<i>γ</i>G determinants. The antibody-forming cells were of host origin, and the effect could be induced by T lymphocytes, not by B. When mice were challenged with NNP.F<i>γ</i>G at various intervals after the induction of GVHR, it was found that by 8 days the increased response was seen only in mice primed to hapten on the heterologous carrier (NNP.OV). Increased antibody formation was not seen in primed CBA mice injected with (CBA × C57B1)F<sub>1</sub> spleen cells, or in normal (CBA × C57B1)F<sub>1</sub> mice injected with primed CBA cells. When mice undergoing GVHR were lethally irradiated, they did not increase the response of primed (CBA × C57B1)F<sub>1</sub> cells transferred to them. The relevance of these findings to the mechanism of interaction between T and B cells is discussed.