xistence of a subpopulation of bone marrow resident cells in which KSHV infection would have an oncogenic outcome. Herein, we sought to identify whether such a mECK36 like cell type could be created by de novo infection leading to productively infected tumors using a two-step approach. First, we evaluated whether a recombinant infectious KSHV, rKSHV.219, could replace BAC36 in exactly the same cellular environment to yield BCTC chemical information tumorigenic cells and whether, in these cells, rKSHV.219 could complete a full replicative cycle. Second, we tested whether we could reproduce this by direct infection of marrow resident cells isolated in the same manner as the mECs from our original model of KSHV-induced tumorigenesis. By replacing the BAC36 construct in our mECK36 model with rKSHV.219, we generated a tumorigenic cell line that when injected in immunodeficient mouse forms KS-like tumors displaying herpesvirus-like particle as detected by electron microscopy. Next, we found that de novo rKSHV.219 infection of primary bone marrow-derived cells lead to cellular immortalization in culture and transformation in vivo with robust, KS-like tumor development in immunocompromised mice, opening up the possibility of using genetically modified mice to elucidate host contributions to KS pathogenesis. Finally, using a clinically relevant histone deacetylase inhibitor, we show that, KSHV reactivation can be enhanced in vivo, a result desirable to test many emerging antiviral therapeutic approaches. Results The loss of BAC36 from the tumorigenic mECK36 cells results in the non-tumoigenic population, mECKnull We overcame the limitations of our mECK36 cells, by replacing the transfected BAC36 construct with an infection of rKSHV.219, a fully replication competent recombinant KSHV. This was done sequentially. First, we cultured the mECK36 cells without hygromycin selection, allowing for the loss of the episomal BAC36. The loss of the BAC36 was confirmed by flow cytometry for GFP showing gradual loss of GFP until the population was totally GFP negative and by a negative PCR for LANA. This resulted in the non-tumorigenic KSHV-null mECK36 cell line, mECKnull, first described in Mutlu et al. and reproduced in these sets of experiments. Two major pieces of evidence indicated that the mECKnull were not transformed: there was no tumor formation up to three months after subcutaneous injection of 3 million cells into nude mice, while an equivalent number of mECK36 cells form subcutaneous tumors that are palpable after just two weeks, the mECKnull cells lost the ability to 
form colonies in a soft agar assay while the mECK36 cells remained colony forming in soft agar and, the cells remained contact inhibited in vitro. These assays confirmed a concurrent loss of tumorigenicity with loss of the BAC36 KSHV genome and indicated that mECKnull cells remained an appropriate substrate for studies of KSHV-induced tumorigenicity. Replacing BAC36 in mECK36 cells for PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19649022 rKSHV.219 preserves KSHV-dependent tumorigenesis mECKnull cells were infected with rKSHV.219 at a multiplicity of infection of 1 to generate the cells we termed mECKnull.rK. rKSHV.219 is a construct created from wild type KSHV isolated from JSC-1 cells and modified to constitutively express GFP driven by an EF-1a promoter, along with the pac gene allowing for puromycin selection. The recombinant virus also contains RFP driven by the KSHV replication and transcriptional activator -dependent lytic polyadenylated nuclear RNA promot