We have continued to use our Friend spleen focus-forming virus (SFFV)-induced mouse model for leukemia to understand the molecular changes associated with pathogenesis. Friend SFFV causes a multi-stage erythroleukemia in mice that parallels in many respects the biphasic development of myeloid leukemias in humans. The first stage is characterized by abnormal erythroid progenitor cell expansion with differentiation primarily in the spleen, resulting in death from splenomegaly. In this stage, virus-infected erythroid cells proliferate and differentiate in the absence of erythropoietin (Epo). This is the result of expression of the SFFV envelope protein, which interacts with the Epo receptor and the receptor tyrosine kinase sf-Stk to cause constitutive activation of signal transduction pathways needed for erythroid cell proliferation and differentiation. In the second stage, proviral insertional activation of the PU.1 gene and the resulting block in differentiation affords full transformation to a subpopulation of erythroid cells in the spleen, which can be isolated and expanded in vitro. We recently demonstrated that transplantation of these SFFV-transformed erythroleukemia cells (SFFV-MEL) to syngeneic, normal mice results in their preferential proliferation in the BM as well as extramedullary proliferation in the meningeal spaces of the central nervous system (CNS), resulting in rapid hind limb paralysis and death due to meningeal leukemia. TRAP staining of bone sections showed evidence of osteoclastogenesis at putative invasion sites into the meninges. Gene expression profiling of SFFV-MEL cells relative to non-transformed, SFFV-infected spleen cells revealed up-regulation of genes involved in extracellular matrix processing (Olfm1, ADAMTS7, THBS2) as well as oncogenes (Sfpi-1, Kit), angiogenic factors (FGF-13, FGF-11, VEGFA), and a bone marrow adhesion integrin (ITGA). Downregulated genes included those encoding the anti-angiogenic protein thrombospondin-1 and its receptor, CD36. Consistent with a gene expression profile that should favor angiogenesis, animals that developed meningeal leukemia after injection of SFFV-MEL cells showed evidence of pathological angiogenesis in the bone marrow as determined by CD31 immunohistochemistry. In addition, SFFV-MEL cells were shown to secrete high levels of VEGF. We further demonstrated that SFFV-MEL cells preferentially adhere to fibronectin and that this is mediated by the fibronectin receptor integrin alpha 5- beta 1. Thus, SFFV-transformed erythroleukemia cells appear to cause meningeal leukemia because they have characteristics that favor their proliferation and retention in the bone marrow, such as overexpression of genes involved in angiogenesis, osteoclastogenesis, extracellular matrix processing and bone marrow adhesion as well as oncogenes that block differentiation. Since meningeal invasion is a complication of leukemia in humans, this unique animal model for meningeal leukemia should be valuable for studying the mechanism of engraftment and proliferation of leukemic cells in the BM and their invasion of the CNS as well as for pre-clinical evaluation of experimental therapeutics for hematological malignancies with CNS involvement. In addition to investigating the molecular basis for the development of various stages of leukemia, we have also been carrying out studies to identify drugs that block leukemia cell growth. In collaboration with the CCR laboratory of Larry Keefer, we have been studying the nitric oxide (NO) prodrug JS-K, a promising anti-cancer agent active against human leukemia and several other cancers. Our data shows that the drug is also cytotoxic to SFFV-transformed erythroleukemia cells at a low IC50 value. JS-K consists of a diazeniumdiolate group necessary for the release of NO as well as an arylating ring. To understand the mechanism of cytotoxicity of JS-K against SFFV-MEL cells and determine the roles of NO and arylation in this process, we compared the effects of JS-K with CDNB, which contains an arylating ring analogous to that of JS-K without the diazeniumdiolate group necessary to release NO. Our studies indicate that both JS-K and CDNB inhibit proliferation of SFFV-MEL cells with low IC50 values and induce caspase-associated apoptosis (involving activation of caspases 8, 9 and 3 and DNA fragmentation) as well as cell cycle arrest at G2/M within 24 hrs. Although both JS-K and CDNB initially led to activation of the PI3-kinase/Akt pathway in these cells and phosphorylation of the tumor suppressor FoxO3a, within 24 hrs both the PI3-kinase/Akt and MAP kinase pathways were blocked, and this was associated with dephosphorylation and activation of FoxO3a and the subsequent upregulation of the cyclin-dependent kinase inhibitor p27- Kip1. We conclude from our studies that JS-K kills SFFV-MEL cells by inactivating the tumor suppressor FoxO3a, causing cell cycle arrest and apoptosis, and that the arylating capability of JS-K is sufficient for inducing these biological effects. Studies are in progress to determine the mechanism by which FoxO3a is activated by JS-K and to determine which proteins in the erythroleukemia cells are arylated. Also, studies will be carried out to determine if JS-K will block the development of meningeal leukemia using our SFFV-MEL-induced mouse model. Finally, we have continued our studies on a second retroviral model system: the neurodegenerative disease induced in rats by PVC-211 murine leukemia virus (MuLV). Diseased brain and spinal cord from PVC-211 MuLV-infected rats exhibit the spongiform pathology characteristic of some human neurodegenerative diseases such as HTLV-1-associated myelopathy/tropical spastic paraparesis and transmissible spongiform encephalopathy. We previously demonstrated that PVC-211 MuLV, due to subtle changes in its envelope protein, can efficiently infect brain capillary endothelial cells (BCEC), allowing it to be expressed at high levels in the neonatal rodent brain and resulting in a rapid neurodegenerative disease. To clarify the mechanism by which PVC-211 MuLV expression in BCEC induces neurological disease, we examined virus-infected rats at various times during neurological disease progression for vascular and inflammatory changes. Our studies indicate that early in the course of disease, morphologically abnormal and leaky blood vessels can be observed in the regions of the brain where neurodegeneration later occurs. This is likely the result of increased production of vascular endothelial growth factor (VEGF), which we detect in the brain 1-2 weeks after virus infection and after in vitro infection of primary BCEC cultures with PVC-211 MuLV. Furthermore, we showed that the brain and serum of rats injected 2 weeks previously with PVC-211 MuLV express high levels of MIP-1 alpha, a chemokine involved in recruitment of microglia to the brain. This was followed 3 weeks after virus infection by a marked accumulation of microglia in the diseased areas of the brain. Further studies demonstrated that depletion of microglia from rat brains blocks neurodegeneration induced by PVC-211 MuLV and that treatment with antiserum to MIP-1 alpha or splenectomy, both of which reduce the number of activated microglia in the brain, can delay disease, clearly demonstrating the importance of activated microglia in the development of PVC-211 MuLV-induced neurodegeneration. Current studies are focused on understanding how viral infection of BCEC results in the activation of VEGF and testing of pharmacological inhibitors of VEGF, MIP-1 alpha or microglia activation for their ability to block or mitigate PVC-211 MuLV-induced neurodegeneration.
