Suppressor function of hepatic mononuclear inflammatory cells during murine chronic graft-vs-host disease. I. Macrophage-enriched cells mediate suppression in the liver

A modified model of graft-versus-host-induced systemic sclerosis (scleroderma) exhibits all major aspects of the human disease

OBJECTIVE

Diffuse systemic sclerosis (SSc; scleroderma) is a debilitating disease characterized by excessive dermal fibrosis with later progression to internal organs. In addition to the fibrotic component, major aspects of the disease include vascular or circulatory involvement and immune dysregulation evidenced by inflammatory cells in affected tissues and production of autoantibodies. Many animal models resembling this disease have been studied, including genetic models in mice and chickens, challenge with chemicals such as bleomycin or vinyl chloride to induce fibrosis, and models of graft-versus-host (GVH)-induced disease using certain strains of mice with differences in minor histocompatibility loci. The present studies were undertaken to determine if alteration of the induction of GVH-induced scleroderma could result in a model that more fully represented the human condition.

METHODS

Disease was induced by injection of spleen cells from B10.D2 mice into BALB/c mice deficient in mature T and B cells (recombination-activating gene 2 targeted). Dermal thickening, collagen deposition, vasoconstriction, and parameters of immunity were analyzed.

RESULTS

Similar to the human disease, this modified GVH model of SSc demonstrated evidence of dermal thickening, particularly in the extremities, progressive fibrosis of internal organs, vasoconstriction and altered expression of vascularity markers in skin and internal organs, early immune activation, inflammation in skin and internal organs, and autoantibody generation.

CONCLUSION

This modified model of GVH-induced SSc exhibits all major components of human disease and is likely to contribute to better understanding of the disease mechanisms and, ultimately, improved treatments for patients.

Amelioration of steroid-resistant chronic graft-versus-host-mediated liver disease via tacrolimus treatment

Liver disease associated with chronic graft-versus-host disease (cGVHD) is a major cause of mortality and morbidity. Steroids and cyclosporin (CSA), which are the standard therapy, give rather disappointing results, and toxicity is high. Tacrolimus (FK506) is a potent macrolide lactone immunosuppressant that is used in the prevention of solid organ rejection. This study evaluated the therapeutic role of FK506 in the treatment of severe cGVHD-mediated liver disease that did not respond to combined steroids and CSA therapy. Fifteen patients with various hematological disorders who underwent allogeneic stem cell transplantation were enrolled in the study. All patients had severe cholestatic liver disease disturbances and underwent liver biopsy, which was compatible with cGVHD-mediated liver disease. All the patients were negative for markers of chronic liver disease, including viral serology. They received FK506 orally (4-20 mg/day according to serum levels), and were evaluated biweekly by physical examination and liver function tests. Patients were followed for a median of 12 months (range 3-24 months). FK506 treatment ameliorated liver functions in 9 of 15 patients (60%), 5 of whom demonstrated complete normalization of liver enzymes (33%). In 5 patients, no major effect was observed, and 1 patient showed deterioration of his liver functions. Mean GGT levels decreased from 171.5 to 55.6 within 6 months of treatment. Median time to response was 3 months (range 1-11). Side effects were generally transient. Treatment with FK506 was found to be effective in the majority of patients with steroid and CSA-resistant cGVHD-associated liver disease, with manageable side effects. In view of these findings, FK506 may yet evolve into first line therapy for cGVHD induced liver toxicity.

Autoantibodies in human chronic graft-versus-host disease after hematopoietic cell transplantation

Primary biliary cirrhosis (PBC) and graft-versus-host disease (GVHD) are thought to have common immunopathologic features and previous studies have reported that 5.2 to 81% of patients with chronic GVHD after allogeneic hematopoietic cell transplant have antimitochondrial antibodies (AMA). We studied a total of 89 patients with chronic GVHD and 60 controls for AMA reactivity by ELISA and immunoblotting using recombinant PDC-E2, BCOADC-E2, and OGDC-E2, immunoblotting of beef heart mitochondrial proteins, and reactivity to nuclei, smooth muscle (ASMA), ribonucleoprotein JO-1, extractable nuclear antigen, nuclear proteins SSA/ SSB, ribonucleic P proteinase III, cardiolipin (ACA), liver kidney microsomal, thyroid microsomal, myeloperoxidase, and the reactivity of rheumatoid factor. A subset of 60 chronic GVHD sera were tested for reactivity to gp210 and LBR. Finally, liver tissue from patients with chronic GVHD and PBC was studied by immunohistochemistry to determine whether there was comparable abnormal apical staining of biliary epithelial cells using PDC-E2-specific monoclonal antibodies. Surprisingly, there were no AMA found in the sera from the 89 patients with chronic GVHD. Review of published data on AMA in GVHD suggests that previous results were primarily false positives. In contrast, sera from the patients with GVHD did have a variety of other autoantibodies and, in particular, 20/89 (22.4%) positive ANA, 23/89 (25.8%) positive ASMA, and 9/89 (10.1%) positive ACA. The other autoantibodies assayed were not statistically different from controls. Finally, abnormal biliary epithelial luminal staining of bile ducts was found, as expected, in liver tissue of patients with PBC but was absent in chronic GVHD.

Contributions of donor CD4 and CD8 cells to liver injury during murine graft-versus-host disease

We have determined the capacity of donor CD4 and CD8 T cells to mediate liver injury in the B10.D2 (donor) into BALB/c (host) chronic graft-versus-host disease (GVHD) model. First, we compared the effects of treating GVHD mice with anti-CD4 or anti-CD8 versus no treatment on the liver histology scores and elevated serum IgE levels in this model. We also examined the abilities of purified donor total T, CD4, and CD8 cells to mediate hepatic GVHD lesions. Anti-CD4 and anti-CD8 treatments caused profound depletion of peripheral CD4+ and CD8+ cells, respectively, and produced a relative enrichment of the CD8+ and CD4+ cells in the liver. Hepatic GVHD lesions and elevated serum IgE concentrations were both suppressed by anti-CD4 treatment. Anti-CD8 treatment had no effect on the severity of hepatic lesions and caused a significant increase in serum IgE levels. Attempts to induce hepatic GVHD with purified donor CD4 and CD8 cells were inconclusive because the onset of liver lesions was delayed and the lesions in both groups were contaminated by the opposite subset. Altogether, our results indicate that both hepatic lesions and elevated serum IgE concentrations in this GVHD model are dependent on donor CD4 cells. Donor CD4 cells mediated hepatic GVHD in the absence of CD8 cells. Donor CD8 cells did not produce hepatic GVHD in the absence of CD4 cells and appeared to be dependent on CD4 cells.

Liver allograft rejection in rats depleted of CD8+ cells

The mechanism(s) of rejection or tolerance induction is a competitive, complex process that presumably involves interactions between multiple subpopulations of T lymphocytes. We investigated the roles of CD8+ cytolytic and CD4+ helper T cells in rat strains that tolerate liver allografts and that differ at both the major histocompatibility complex (MHC) (RT1) and minor histocompatibility genes. Orthotopic liver transplantation (OLT) with arterial reconstruction was performed with Brown Norway (BN) (RT1n) donors and Lewis (RT1(1)) recipients, some of which were untreated, others treated with anti-CD8 antibody, and still others treated with anti-CD4 antibody. Liver graft rejection was monitored for 28 days on the basis of two criteria: (1) serum levels of AST enzyme at 3-day intervals and (2) liver biopsies at weekly intervals and at the time of sacrifice at the end of the study period. In the untreated control group, an elevation of AST was found to peak at day 6 after grafting, and it remained elevated until day 28 (AST 542 +/- 72 U/l). Histologically, signs of severe rejection were first observed on day 9; these changed to moderate rejection about day 21 and to mild rejection by day 28, when the animals were sacrificed. Recipients pre-treated with anti-CD8 demonstrated a significant elevation of AST within 6 days that, unlike in the control recipients, continued to rise sharply through the observation period (AST 1127 +/- 181 U/l, P = 0.009 vs control group). Liver biopsies showed mild rejection at day 9 and moderate rejection at days 21 through 28. Recipients pretreated with anti-CD4 showed a time course of enzyme elevation and severity of rejection that was not significantly different from that observed in the control group. However, anti-CD4 treatment resulted in only 75% depletion of CD4+ cells in peripheral blood as compared to complete elimination of CD8+ cells following anti-CD8 treatment. Functional studies of spleen and liver-infiltrating lymphocytes obtained after 28 days showed low proliferative response in mixed lymphocyte culture with both BN and PVG stimulator spleen and lymph node cells. These results suggest that in this donor/recipient combination, removal of CD8+ cells increases the severity of rejection as demonstrated by a progressive rise in AST and histology. Moreover, OLT in this combination results in a profound, nonspecific inhibition of proliferative T-cell responses to MHC alloantigens.

Development and characterization of a rodent model of immune-mediated cholangitis

The cholangiopathies are a group of hepatobiliary diseases in which intrahepatic bile duct epithelial cells, or cholangiocytes, are the target for a variety of destructive processes, including immune-mediated damage. We tested the hypothesis that cholangitis could be induced in rodents by immunization with highly purified cholangiocytes. Inbred Wistar rats were immunized with purified hyperplastic cholangiocytes isolated after bile duct ligation from either syngeneic Wistar or allogeneic Fischer 344 rats; control rats were immunized with bovine serum albumin (BSA) or hepatocytes. After immunization with cholangiocytes, recipient animals developed histologic evidence of nonsuppurative cholangitis without inflammation in other organs; groups immunized with BSA or hepatocytes showed no cholangitis. Immunohistochemical studies revealed that portal tract infiltrates around bile ducts consisted of CD3-positive lymphocytes, some of which expressed major histocompatibility complex class II antigen; B cells and exogenous monocytes/macrophages were essentially absent. Transfer of unfractionated ConA-stimulated spleen cells from cholangiocyte-immunized (but not BSA-immunized) rats into recipients also caused nonsuppurative cholangitis. Moreover, these splenocytes from cholangiocyte-immunized (but not BSA-immunized) rats were cytotoxic in vitro for cultured rodent cholangiocytes; no cytotoxicity was observed against a rat hepatocyte cell line. Also, a specific antibody response in sera of cholangiocyte-immunized rats was demonstrated by immunoblots against cholangiocyte proteins. Finally, cholangiograms in cholangiocyte-immunized rats showed distortion and tortuosity of the entire intrahepatic biliary ductal system. This unique rodent model of experimental cholangitis demonstrates the importance of immune mechanisms in the pathogenesis of cholangitis and will prove useful in exploring the mechanisms by which the immune system targets and damages cholangiocytes.

Suppressor function of liver mononuclear cells isolated during murine chronic graft-vs-host disease. II. Role of prostaglandins and interferon-gamma

Mononuclear inflammatory cells (MC) isolated from the livers and spleens of mice with chronic graft-vs-host disease (CGVHD) to minor histocompatibility antigens (B10.D2----BALB/c) show defective proliferation when stimulated with Con A and LPS. In turn, both CGVHD liver and spleen cells suppress the proliferation of mitogen-stimulated normal spleen cells in a genetically unrestricted manner. The suppressor activity of CGVHD spleen cells is mediated by plastic nonadherent null (natural suppressor) cells and involves a soluble suppressor factor(s). In contrast, the suppressor activity of CGVHD liver cells is mediated by macrophages (M phi). In the current studies we show that the suppressor activity of CGVHD liver cells is also mediated by soluble factors and compare the roles of prostaglandins and interferon (IFN)-gamma in mediating defective proliferation and suppressor activities of CGVHD liver and spleen MC. Monoclonal antibody to IFN-gamma partially reversed the defective mitogen-stimulated proliferation of CGVHD spleen MC but had no effect on proliferative response of CGVHD liver MC. Indomethacin did not alter the low proliferative response of either CGVHD liver or spleen MC. Anti-IFN-gamma inhibited the ability of CGVHD spleen cells to suppress proliferation of Con A and LPS-stimulated B10.D2 spleen cells. In contrast, anti-IFN-gamma resulted in a small decrease in the ability of liver MC to suppress Con A (but not LPS)-stimulated cell proliferation. Indomethacin decreased the ability of both CGVHD liver and spleen cells to suppress Con A-stimulated proliferation but had inconsistent effects on LPS-stimulated proliferation. These results show that IFN-gamma and prostaglandins partially mediate the suppressor activity of CGVHD spleen MC. The suppressor activity of CGVHD liver MC also involves prostaglandins but is relatively independent of IFN-gamma.