Date Published: January 9, 2014
Publisher: Public Library of Science
Author(s): Da-Jiang Li, Dinesh Verma, Tim Mosbruger, Sankar Swaminathan, Paul Ling.
Kaposi’s sarcoma-associated herpesvirus (KSHV) is a human herpesvirus that causes Kaposi’s sarcoma and is associated with the development of lymphoproliferative diseases. KSHV reactivation from latency and virion production is dependent on efficient transcription of over eighty lytic cycle genes and viral DNA replication. CTCF and cohesin, cellular proteins that cooperatively regulate gene expression and mediate long-range DNA interactions, have been shown to bind at specific sites in herpesvirus genomes. CTCF and cohesin regulate KSHV gene expression during latency and may also control lytic reactivation, although their role in lytic gene expression remains incompletely characterized. Here, we analyze the dynamic changes in CTCF and cohesin binding that occur during the process of KSHV viral reactivation and virion production by high resolution chromatin immunoprecipitation and deep sequencing (ChIP-Seq) and show that both proteins dissociate from viral genomes in kinetically and spatially distinct patterns. By utilizing siRNAs to specifically deplete CTCF and Rad21, a cohesin component, we demonstrate that both proteins are potent restriction factors for KSHV replication, with cohesin knockdown leading to hundred-fold increases in viral yield. High-throughput RNA sequencing was used to characterize the transcriptional effects of CTCF and cohesin depletion, and demonstrated that both proteins have complex and global effects on KSHV lytic transcription. Specifically, both proteins act as positive factors for viral transcription initially but subsequently inhibit KSHV lytic transcription, such that their net effect is to limit KSHV RNA accumulation. Cohesin is a more potent inhibitor of KSHV transcription than CTCF but both proteins are also required for efficient transcription of a subset of KSHV genes. These data reveal novel effects of CTCF and cohesin on transcription from a relatively small genome that resemble their effects on the cellular genome by acting as gene-specific activators of some promoters, but differ in acting as global negative regulators of transcription.
Infection with Kaposi’s sarcoma-associated herpesvirus (KSHV, HHV8) is causally associated with Kaposi’s sarcoma (KS), primary effusion lymphoma (PEL) and multicentric Castleman’s disease (for a review, see reference ). KSHV maintains a persistent latent infection as an episome in B lymphocytes, from which it occasionally reactivates, enters a lytic cycle of replication, and produces infectious virions. Released virions infect other lymphocytes to maintain the latent reservoir or are transmitted from person-to-person in saliva. Cell-mediated immunity is essential for limiting KSHV reactivation and pathogenesis, but cellular epigenetic regulatory mechanisms may also play an important role in limiting viral replication. The balance between lytic and latent infection is an important determinant of pathogenicity.
In this study we report several novel aspects of the role of CTCF and cohesin as regulators of KSHV virus production. First, both CTCF and Rad21 act as host restriction factors for lytic KSHV replication as depletion of either protein resulted in markedly increased production of infectious virions. Rad21 appears to exert a greater effect, as Rad21 knockdown resulted in nearly 100-fold increases in virus yield, approximately five times more than the increase caused by CTCF knockdown. We also demonstrate that both CTCF and Rad21 dissociate from viral genomes during the process of lytic KSHV replication. Rad21 binding is lost earlier and more completely than CTCF after lytic KSHV replication begins. The almost complete loss of Rad21 from the majority of KSHV genome sites indicates eviction from latent episomes early during lytic replication as well as a lack of binding to newly replicated genomes. Conversely, the persistence of Rad21 at the major latency region and the terminal repeats indicates that Rad21 not only remains bound to template genomes but that it binds to nascently replicated genomes at these two sites. CTCF also exhibited site-specific changes in KSHV genome occupancy during lytic replication. CTCF occupancy was decreased by 3 days, and by 5 days, the relative occupancy at most sites was reduced by over 50%, indicating that CTCF binding also does not occur to newly replicated genomes at these locations. The finding that CTCF depletion results in increased virus production is in contrast to those of Chen et. al. who did not observe any effects of CTCF knockdown on KSHV lytic transcription in PEL cells . These differences may be due to the different cell lines employed, and to the fact that knockdown in the our experiments was essentially complete, with no detectable CTCF remaining at the time of lytic induction.