Kaposis sarcoma (KS), the most common neoplasm in AIDS patients, typically presents with multiple skin lesions characterized by spindle cells, the vast majority of which are infected with KSHV (Kaposis sarcoma herpes virus, also named HHV-8). In patients with AIDS, the presence of cell-associated KSHV-DNA in blood is predictive of subsequent KS development, but the mechanisms by which circulating KSHV-infected cells contribute to AIDS-KS pathogenesis are unclear. Our previous studies had detected selective expression of the chemokine SDF-1 in the capillary endothelium of skin, lymph nodes and bone marrow sinusoids, but not usually other sites. Other studies had detected SDF-1 production in angiogenic endothelium. We now found that SDF-1, which is constitutively expressed by skin capillary endothelium and displayed on the endothelial cell surface in association with heparan sulfate, can trigger specific arrest of KSHV-infected cells under physiological shear flow conditions. Moreover, we discovered that SDF-1 expressed on the endothelial barrier promotes transendothelial migration of KSHV-infected cells in the presence of tissue chemokine gradients. By triggering specific adhesion of circulating KSHV-infected cells and favoring their entry into the extravascular cutaneous space, endothelial cell-associated SDF-1 in cutaneous capillaries may dictate the preferential occurrence of KS in the skin. A condition for tissue SDF-1 gradients to be effective and promote cell adhesion and migration is that SDF-1 is either absent/present at very low concentrations or inactive in the circulation. We measured SDF-1 levels in blood of normal individuals and patients with AIDS with or without Kaposis sarcoma and determined SDF-1 levels to be approximately 16 ng/ml in normal and patient blood. In addition, we found SDF-1 to be cleaved and inactive in the circulation due to the removal of two N-terminal residues (KP) due to the action of a dipeptidase, and the removal of the C-terminal K by the enzyme Carboxypeptidase N. Heparan sulpate proteoglycans, abundant on the cell surface can avidly bind SDF-1 and protect it from enzymatic degradation. Thus, SDF-1 is present at nanogram/ml concentration in blood and is found as an active molecule bound to the endothelial cell surface at selected sites. In the skin, SDF-1 likely contributes to the arrest and transmigration of circulating KSHV-infected cells to the extracellular space thereby contributing to disease pathogenesis. Thus, blocking SDF-1/receptor interactions may represent an important therapeutic approach to the prevention of KSHV diseases in infected individuals. In related studies, we have explored the potential role of SDF-1/CXCR4 in the pathogenesis of Primary Effusion Lymphoma (PEL). Primary effusion lymphoma (PEL) is a fatal viral malignancy occurring predominantly in AIDS patients. The tumor cells are of B-cell lineage and are usually infected with KSHV alone or in conjunction with EBV. Typically, PEL presents as a malignant effusion confined to body cavities without recognizable tumor masses, but the malignancy disseminates such that most patients display systemic disease at the time of death. The survival of patients with PEL is estimated to be less than 1 year despite high-dose chemotherapy; thus, there is a need for innovative therapies. Since PEL cells express functional CXCR4 receptors that mediate SDF1 (stromal-derived factor-1)-induced chemotaxis, and the mesothelium from body cavities produces SDF1, we sought to prevent PEL dissemination by blocking CXCR4 function. We found that the CXCR4 competitive inhibitors AMD3100 and CTCE9908 blocked SDF1-induced Akt phosphorylation and chemotaxis in PEL cells from culture, but inhibition in mice was not sustained. Thus, in a pre-clinical model of PEL, these compounds were only marginally effective at reducing disease progression. In additional experiments, we found that Rapamycin is a potent inhibitor of Akt phosphorylation induced by SDF-1. Unlike AMD3100 and CTCE9908, Rapamycin treatment of mice bearing PEL ascites induced a sustained block of SDF1-induced Akt phosphorylation. In a murine model of human PEL, Rapamycin significantly extended mouse survival, reduced PEL-tumor formation and prevented ascites accumulation, but did not eradicate the disease. Akt is not constitutively active in PEL, and Rapamycin did not promote PEL cell death or block PEL cell proliferation, but likely reduced disease progression in mice by preventing Akt activation in vivo and reducing VEGF secretion. The successful use of Rapamycin to inhibit Akt activation in PEL cells to halt their colonization illustrates a novel application of mTOR (mammalian target of Rapamycin) targeting and a new principle for treatment of PEL
