During past decades, considerable progress has been made in the field of allogeneic hematopoietic stem cell transplantation (allo-HSCT), enabling the procedure as a curative treatment to an increasing number of patients suffering from hematopoietic malignancies or autoimmune diseases. Despite numerous improvements of allo-HSCT, the rates of post-treatment complications are still high: more than half of the transplanted patients develop Graft-versus-Host Disease (GvHD), with an overall mortality up to 40%, making GvHD the major treatment sequelae after allo-HSCT. GvHD can occur acute (aGvHD) or chronic (cGvHD) as a systemic inflammatory disease, affecting various organs and tissues. Chronic inflammation subsequently leads to fibrosis and severe morbidity in cGvHD patients. Common therapeutic approaches aim to suppress alloreactive T cells, the inflammatory mediators of GvHD, but the use of immunosuppressive steroids is accompanied by GVHD-associated complications, as treatment toxicity, tumor relapse or fatal infections. Therefore, an urgent need for alternative therapies exists, which effectively prevent and attenuate GvHD without eliminating the necessary immune responses. Recent studies proposed the endothelium as a novel target for treatment approaches of GvHD. While angiogenesis, the formation of new blood vessels, is a cause of inflammation during early GvHD, the endothelium is damaged from the persisting inflammation at later GvHD stages. Due to the lack of suitable mouse models, the endothelial participation especially in the cGvHD pathology is inadequately characterized, resulting in limited anti-angiogenic therapies and novel therapeutic endothelial targets. This thesis intends to elicit the mechanisms leading to endothelial GvHD and identify such potential target structures for GvHD treatment. In the first part of this study, we examined metabolic genes that are differentially regulated during pathologic angiogenesis. By genetic deletion with CRISPR/Cas9, the enzymes Enolase 3 and Glucose-6-phosphate dehydrogenase (G6pdx) were identified to participate in the endothelial regulation under allogeneic conditions and were successfully intervened by specific pharmacologic inhibitors in vitro. Therapy using the G6pdx-inhibitor Polydatin resulted in attenuated inflammation in footpad swelling assays, ameliorated the GvHD morbidity and normalized the vascular integrity in vivo. The second part of the project deals with the endothelium during cGvHD. Two novel murine and one humanized mouse model of cGvHD were established, which closely resemble the clinical features of cGvHD in patients. Characterizing these models, a decelerated immune and hematopoietic cell reconstitution was found, followed by a massive infiltration of T and B cells and severe tissue sclerosis/ fibrosis in GvHD target organs. Determining the endothelial condition during late cGvHD, a prominent vascular damage and dysfunction was found: Next to a loss of endothelial CD31 and VE-Cadherin and an increase of von-Willebrand factor, the endothelial integrity was disturbed by the loss of tight junction protein Zonula occludens protein-1 and pericyte marker Neuron-glial antigen 2. Additionally there was occurrence of circulating endothelial cells and endothelial progenitor cells in cGvHD. Severe endothelial fibrosis by analysis of alpha-smooth muscle actin and Fibroblast-specific protein-1 was quantified. Hence for the first time, a transformation of endothelial towards a mesenchymal cell type (Endothelial-to-mesenchymal transition) was detected, a process which has not been described in the context of GvHD before. This transition was successfully prevented by administration of Pirfenidone, which normalized the CD31+ vessel quantity and ameliorated the endothelial fibrosis in vivo. Furthermore, a panel of five novel endothelium-related biomarkers (Endostatin, Connective tissue growth factor, Endoglin, Follistatin, Endocan), which were increased in blood serum of cGvHD patients, was proposed and might be beneficial for prognosis and diagnosis of cGvHD in the clinic. This study helps to provide knowledge about the connection between the endothelium and the GvHD pathology. It identified potential therapeutic targets to ameliorate GvHD without interfering the anti-tumor response after allo-HSCT, normalizing the disturbed vascular function and prevent endothelial fibrosis.
