25-Hydroxycholesterol Inhibits KSHV and EBV Infections and Activates Inflammatory Cytokines

We previously demonstrated that 25HC (25-hydroxycholesterol, a derivative of cholesterol) blocks KSHV (Kaposi Sarcoma-associated Herpesvirus/ Human Herpesvirus-8) de novo infection of primary endothelial cells at a post-entry step and decreases viral gene expression. Herein we sought to determine transcriptomic changes associated with 25HC treatment of primary endothelial cells and primary B cells using RNA sequencing. To understand how 25HC inhibits KSHV infection, we examined global gene expression changes induced by 25HC in HUVECs, with or without KSHV infection. We pre-treated HUVEC with 25HC or vehicle control, then infected with KSHV (four conditions: -/+ 25HC, -/+ infection). At 2 days post-infection (dpi), we harvested RNA and performed RNA-seq to determine differentially expressed genes. We found that 25HC treatment inhibited KSHV gene expression globally and induced several inflammatory cytokines (CXCL8, IL1?), AP-1 components, and innate immunity receptors (RIG-I, TLR2) . Likewise, we also observed 25HC to block EBV de novo infection by inducing apoptosis and blocking transformation into lymphoblastoid cell lines. To determine transcriptomic changes that contributes to this, we purified B cells from healthy donors and infected them with EBV after 25HC pre-treatment with 4 similar conditions mentioned above. Samples were sent for RNA-Seq at 10-12 dpi. In contrast to KSHV infection, we found that only a subset of EBV viral transcripts are downregulated in the 25HC-treated samples. LMP1, which is the primary oncogenic factor, was suppressed and several downstream host targets that would prevent apoptosis during infection were likewise dysregulated. Additionally, several viral noncoding RNAs were also highly downregulated. Further analyses revealed that some 25HC-induced genes in our KSHV data set overlap with antiviral genes that were previously identified in various published screens involving other viruses, demonstrating the broad activity of 25HC. Together, these results answer some important questions about a widely acting antiviral (25HC), with implications for multiple viral and bacterial infections. Four biological replicates of KSHV de novo infected HUVEC were submitted for total RNA sequencing. Sixteen RNA samples were depleted for ribosomal RNA and pooled for sequencing on four lanes on HiSeq4000 with Illumina Hiseq3000-4000 chemistry. All the samples have excellent yield with yield above 146 million reads. Alignment was performed with reference human hg19 genome with KSHV [hg19v24 with KSHV(NC_009333.1)] and gencode v24 transcripts using STAR software. Quantification was carried out with RSEM using transcriptome bam file created by STAR and gencode v24 transcriptome as index for RSEM. Four biological replicates of EBV de novo infections and 3 biological replicates of uninfected (mock) control of B cells (purified using MACSExpress from Buffy coats) were submitted for total RNA sequencing. Fourteen RNA samples were depleted for ribosomal RNA using Illumina Ribo-Zero rRNA removal kit and submitted for sequencing using HiSeq 4000 or equivalent machine. All the samples have excellent yield with yield above 270 million reads. Alignment was performed with reference human hg38 genome with EBV [hg38v30 with EBV (NC_007605.1)] and gencode v30 transcripts using STAR software. Quantification was carried out with RSEM using transcriptome bam file created by STAR and gencode v30 transcriptome as index for RSEM.