Role of Nrf2 in KSHV Biology and Oncogenesis

Abstract

Kaposi’s sarcoma-associated herpesvirus (KSHV), or human herpes virus 8 (HHV8), is a ?2 lymphotropic herpesvirus and is the etiological agent of Kaposi’s sarcoma (KS), primary effusion B-cell lymphoma (PEL), and the multicentric Castleman’s disease (MCD). KSHV malignancies are associated with immune suppression, such as in AIDS or organ transplantation patients, and with certain African and Mediterranean populations.

In an effort to identify host factors involved in the establishment of KSHV infection, we focused on the nuclear factor E2-related factor 2 (Nrf2) a member of the Cap’n’Collar basic leucine zipper (bZIP) family of transcription factors. Nrf2 is induced by ROS, and has been shown to play an important role for infection by several viruses. Moreover, Nrf2 activation leads to the expression of a cohort of genes involved in apoptosis, angoigensis, metastasis, drug resistance, and the proliferative pentose pyrophosphate pathway (PPP) enzymes. Interestingly, several of these pathways are upregulated during KSHV infection through unknown mechanisms. Because KSHV infection, like other viruses that activate Nrf2, induces ROS, a powerful activator of Nrf2, we hypothesized that KSHV infection of target cells activates Nrf2 in order to induce Nrf2 target genes and create a microenvironment conducive to infection and tumorigenesis. To this end, we utilized cellular models to assess Nrf2 activity during de novo and latent KSHV infection, the mechanism of its activation, and its role in host and virus biology.

In the first part of this study, we found that Kaposi's sarcoma (KS) skin tissue exhibited elevated Nrf2 levels compared to healthy skin tissue. De novo infection of endothelial (HMVEC-d) cells showed that ROS were essential for Nrf2 activation during the early stages of infection, but dispensable during latency, where the COX-2/PGE2/PKCζ axis played an essential role in the sustained activation. Interestingly, Nrf2 was essential for optimal COX-2 expression, a major pro-viral agent during KSHV infection, establishing a feed-forward loop between COX-2 and Nrf2 in KSHV biology. Nrf2 activation was also necessary for the KSHV-mediated induction of host Bcl-2, VEGF, and the PPP enzymes. Nrf2 colocalized with LANA-1 and the KSHV genome during latency, and played an important role in proper lytic (ORF50) and latent (ORF73) gene expression. This study demonstrated for the first time that KSHV induces Nrf2 during de novo infection of endothelial cells to aid with establishment of latency.

In the second part of the study, we focused on long-term-infected telomerase-immortalized endothelial cells (TIVE-LTC), which provide a model of prolonged KSHV latency. We determined that ROS did not affect Nrf2 activity in these cells. More interestingly, we identified the existence of two simultaneous Nrf2 activating pathways. The first, the non-canonical pathway, involved the autophagic protein p62-mediated sequestration of the Nrf2 inhibitor Keap1, promoting intracellular Nrf2 protein accumulation. A second activating pathway involving the COX-2/PGE2/PKCζ axis further induced Nrf2 activation and phosphorylation, which was necessary for sustained expression of Nrf2 target genes, including GCS, NQO1, xCT, VEGF and IL6, all important agents in KSHV infection and oncogenesis.

In the third part of the study, we shifted our attention to KSHV latency in PEL models. We determined that histopathological tissue obtained from PEL of the stomach exhibited significant Nrf2 activation. Investigation of PEL-derived cell lines revealed that the COX-2/PGE2/PKCζ axis required prostaglandin E receptor 4 (EP4) activation, which acted as the receptor used by PGE2 to activate Nrf2. Next-generation RNA sequencing (NGS) and qPCR experiments revealed that Nrf2 knockdown or inhibition with the chemical Brusatol resulted in elevated global lytic gene expression. Additionally, we identified a novel regulatory mechanism of the major lytic regulatory gene, ORF50, that involved Nrf2, viral LANA-1 and the host transcriptional repressor KAP1. We determined that Nrf2 played a crucial role in the ORF50-mediated lytic burst during early de novo infection, an effect that is repressed in latency by LANA-1 recruitment of KAP1 to the ORF50 promoter in an Nrf2-dependent manner. (Abstract shortened by UMI.)