Role of STAT1 Expression during Graft-Versus-Host Disease (GVHD) in the Gastrointestinal Tract: Association with Lamina Propria Cell Infiltration and Tissue Cytokine/Chemokine Expression.

The role of the IFN-g in the development of GVHD remains enigmatic. Whereas it has been shown that GVHD can occur in the absence of IFN-g, there are partially contradicting results how IFN-g can modulate GVHD. Thus, it has been suggested that blocking IFN-g might ameliorate gut GVHD in rodent BMT models and its direct cytopathic effects in human ex vivo skin explant models. However, there is also data supporting the notion that IFN-g is important for limiting GVHD by induction of activation induced cell death (AICD) in donor T cells. Furthermore, it has been demonstrated in preclinical BMT models that IFN-g may accelerate or mitigate GVHD depending on the conditioning intensity. We have recently demonstrated that mitigation of GVHD by treatment with the Histone deacetylase Inhibitor (HDACi) suberonylanilide hydroxamic acid (SAHA) is associated with early downregulation of STAT1 phosphorylation in the spleen and host epithelial GVHD target organs. To further understand the role of STAT1 in the development of GVHD we studied STAT1 and p-STAT1 (Tyr701) expression by immunohistochemistry in the GVHD target organs liver, small bowel and colon following induction of GVHD and correlated these findings with the presence of lamina propria (LP) lymphocytes, typical features of GVHD-induced tissue damage (crypt cell apoptosis, crypt regeneration) and expression of tissue cytokines/chemokines. GVHD was induced in the fully MHC mismatched BALB/c to B6 strain combination following lethal irradiation with 975 rad and animals were sacrificed on days +1, +3 and +6. As detected by western blots p-STAT1 expression became detectable on day +1 in the spleen and on days +3 in the liver, small bowel and colon. Compared to untreated controls immunohistochemical p-STAT1 staining became apparent on day +3 post-BMT in the small bowel and colon of syngeneic controls and GVHD animals. In syngeneic controls p-STAT1 expression decreased again on day +6. In contrast, a further significant increase in p-STAT1 staining was observed in animals with GVHD in the colon and small bowel on day +6. In the colon this significant increase in crypt cell p-STAT1 staining was associated with the presence of LP infiltrating lymphocytes and coincided with the maximal features of tissue damage (luminal sloughing, crypt destruction and crypt apoptosis). In line with these results IFN-g protein expression became detectable in colon tissue lysates on day +6 supporting the role of IFN-g producing infiltrating donor T cells in causing STAT1 activation and tissue damage. We conclude that in comparison to untreated controls STAT1 activation can be observed in the colon and small bowel starting on day +3 in animals with GVHD and syngeneic controls. Whereas pSTAT1 staining peaks at day +3 in syngeneic controls and declines thereafter, maximal STAT1 activation occurs in GVHD animals on day +6, coincides with detectable IFN-g expression and is accompanied by LP infiltration and features of severe GVHD-related tissue damage. To fully understand the role of IFN-g in the development of gut GVHD further studies are warranted to delineate the role of STAT1 dependent and independent signaling pathways in the development of GVHD.