TAF support cancer cell invasion through the secretion of soluble mediators of inflammation and cell migration [7
]. Moreover, TAF invasion and modification of the extracellular matrix facilitates cancer cell invasion [9
]. Direct targeting of TAF also reduces cancer progression in vivo
and sensitizes tumors to chemotherapy [25
], further justifying additional mechanistic studies to understand the role of TAF in cancer pathogenesis. Unique clinical features of oral KS, including its propensity for chemotherapeutic resistance and its prediction of less desirable outcomes for patients [3
], warrant further mechanistic studies to elucidate novel KSHV-oral host cell interactions, including those involving OF. KSHV induces production of pro-inflammatory and pro-migratory factors by skin-derived fibroblasts following de novo
], and KSHV-infected, skin-derived fibroblasts promote intrinsic invasiveness for uninfected endothelial cells through paracrine mechanisms [11
]. However, determination of OF susceptibility to KSHV infection and the functional consequences of KS-OF interactions have not been previously described.
We found both HGF and PDLF permissive for KSHV infection, with predominant expression of latent genes in both cell types. Interestingly, although expression of KSHV lytic genes was relatively limited, PDLF exhibited an increase in expression of vIL-6 relative to HGF. Although the meaning and functional consequences of this observation were not addressed in our studies, it suggests a need for broader characterization of KSHV gene expression in OF, and that different KSHV-infected OF may serve different roles in oral KS pathogenesis. KS involvement of the periodontal ligament is exceedingly rare [27
]. However, vIL-6 induces VEGF secretion [28
], and it is interesting to speculate whether KSHV infection of PDLF may provide a supportive role for pathogenesis in the oral KS microenvironment.
We found that KSHV infection induced OF secretion of several soluble factors associated with KS pathogenesis, including VEGF, IL-8, IL-6 and IL-10. We and others have previously demonstrated mechanisms for KSHV induction of these cytokines in other cell types [11
]. Although not addressed in our study, secretion of IL-6 and IL-10 by KSHV-infected OF has implications for suppression of KSHV-specific immune responses and proliferation of KSHV-infected tumor cells in the oral cavity [31
]. We did not observe an effect of KSHV infection on secretion of IL-1β or TNF-α by OF. In addition, secretion of VEGF by OF was emmprin-dependent while secretion of IL-8 was not. Furthermore, we noted KSHV/emmprin-dependent increases in OF expression of MMP-1 and MMP-9, but not MMP-2 which contrasts our observed increases in KSHV/emmprin-dependent activation of all three MMPs in endothelial cells [20
]. Additional work will determine whether KSHV induction of emmprin-dependent pathogenesis differs for OF due to unique KSHV-host cell interactions occurring in these cells.
We found that direct targeting of emmprin, Sp1, or Egr2 suppressed KSHV-induced VEGF secretion and invasiveness for OF, and that targeting either Sp1 or Egr2 suppressed KSHV-induced expression of emmprin and related MMPs. Moreover, OF secretion of VEGF-A and invasiveness were restored with emmprin overexpression in the context of Sp1 or Egr2 targeting. Taken together, these data implicate KSHV induction of Sp1- and Egr2-mediated transcriptional activation of emmprin as an important mechanism for OF pathogenesis following KSHV infection of these cells. Additional experiments are needed to confirm direct interactions between these transcription factors and emmprin in KSHV-infected OF, as we cannot categorically exclude the possibility that Sp1 and/or Egr2, rather than binding the emmprin promoter directly, induce transcriptional activation of other cellular or viral genes that subsequently regulate the activation of emmprin. Interestingly, we also observed that KSHV infection of HGF increased expression of Egr2, but not Sp1. In contrast, we have not observed an increase in expression of either Egr2 or Sp1 following infection of primary endothelial cells (data not shown). Our data revealing upregulation of emmprin and OF invasiveness with ectopic expression of LANA is consistent with results from a recent study indicating that LANA is capable of direct interaction with Sp1 and facilitating Sp1-mediated transcriptional activation of telomerase [32
]. However, additional work is needed to determine whether LANA facilitates direct binding of Sp1 and Egr2 to the emmprin promoter in OF. Collectively, these observations support the concept that KSHV regulates TAF pathogenesis through unique transcription control mechanisms, and that elucidation of these pathways may provide clues for observed clinical differences between oral KS and KS localized elsewhere.
In summary, this report provides the first evidence for successful establishment of latent KSHV infection within human primary OF and the induction of a TAF-like phenotype for these cells following de novo infection. Moreover, we provide evidence for KSHV regulation of emmprin transcription as one mechanism for this process. These results support a putative role for TAF in KS pathogenesis, provide rationale for ex vivo studies to determine whether primary OF from oral KS lesions harbor infectious KSHV, and rationale for development of novel studies to determine whether targeting TAF and emmprin-related pathways reduces progression of “KS-like” lesions in vivo.