Role of CCAAT/enhancer-binding protein alpha (C/EBPα) in activation of the Kaposi's sarcoma-associated herpesvirus (KSHV) lytic-cycle replication-associated protein (RAP) promoter in cooperation with the KSHV replication and transcription activator (RTA) and RAP

The Kaposi's sarcoma-associated herpesvirus (KSHV)-encoded replication-associated protein (RAP, or K8) has been shown to induce both CCAAT/enhancer binding protein alpha (C/EBPαs) and p21^CIP-1 expression, resulting in G₀/G₁ cell cycle arrest during the lytic cycle. RAP and C/EBPα are also known to interact strongly both in vitro and in lytically infected cells. We recognized two potential consensus C/EBP binding sites in the RAP promoter and performed electrophoretic mobility shift assay (EMSA) analysis with in vitro-translated C/EBPα; this analysis showed that one of these sites has a very high affinity for C/EBPα. Luciferase (LUC) assays performed with a target RAP promoter-LUC reporter gene confirmed that C/EBPα can transcriptionally activate the RAP promoter up to 50-fold. Although RAP had no effect on its own promoter by itself, the addition of RAP and C/EBPα together resulted in a threefold increase in activity over that obtained with C/EBPα alone. Importantly, the introduction of exogenous Flag-tagged C/EBPα triggered RAP expression in BCBL-1 cells latently infected with KSHV, as detected by both reverse transcription-PCR and double-label immunofluorescence assay analyses, suggesting the presence of a self-reinforcing loop with C/EBPα and RAP activating each other. The RAP promoter can also be activated 50- to 120-fold by the KSHV lytic-cycle-triggering protein known as replication and transcription activator (RTA). C/EBPα and RTA together cooperated to elevate RAP promoter activity four- to sixfold more than either alone. Furthermore, the addition of RAP, C/EBPα, and RTA in LUC reporter cotransfection assays resulted in 7- to 15-fold more activation than that seen with either C/EBPα or RTA alone. Site-specific mutational analysis of the RAP promoter showed that the strong C/EBP binding site is crucial for C/EBPα-mediated transactivation of the RAP promoter. However, the C/EBP binding site also overlaps the previously reported 16-bp RTA-responsive element (RRE), and the same mutation also both reduced RTA-mediated transactivation and abolished the cooperativity between C/EBPα and RTA. Furthermore, in vitro-translated RTA, although capable of binding directly to the polyadenylated nuclear RNA (PAN) RRE motif, failed to bind to the RAP RRE and interfered with RRE-bound C/EBPα in EMSA experiments. Partial RTA responsiveness but no cooperativity could be transferred to a heterologous promoter containing added consensus C/EBP binding sites. A chromatin immunoprecipitation assay showed that all three proteins associated specifically with RAP promoter DNA in vivo and that, when C/EBPα was removed from a tetradecanoyl phorbol acetate-treated JSC-1 primary effusion lymphoma cell lysate, the levels of association of RTA and RAP with the RAP promoter were reduced 3- and 13-fold, respectively. Finally, RTA also proved to physically interact with both C/EBPα and RAP, as assayed both in vitro and by immunoprecipitation. Binding to C/EBPα occurred within the N-terminal DNA binding domain of RTA, and deletion of a 17-amino-acid basic motif of RTA abolished both the C/EBPα and DNA binding activities as well as all RTA transactivation and the cooperativity with C/EBPα. Therefore, we suggest that RTA transactivation of the RAP RRE is mediated by an interaction with DNA-bound C/EBPα but that full activity requires more than just the core C/EBP binding site.