Interferon induced circRNAs escape herpesvirus host shutoff and suppress lytic infection

A first line of defense during infection is expression of interferon (IFN)-stimulated gene products which suppress viral lytic infection. To combat this, herpesviruses express endoribonucleases to deplete host RNAs. Here we demonstrate that IFN-induced circular RNAs (circRNAs) can escape viral-mediated degradation. We performed comparative circRNA expression profiling for representative alpha- (Herpes simplex virus-1, HSV-1), beta- (human cytomegalovirus, HCMV), and gamma-herpesviruses (Kaposi sarcoma herpesvirus, KSHV; murine gamma-herpesvirus 68, MHV68). Strikingly, we found that circRNAs are, as a population, resistant to host shutoff. This observation was confirmed by ectopic expression assays of human and murine herpesvirus endoribonucleases. During primary lytic infection, ten circRNAs were commonly regulated across all subfamilies of human herpesviruses, suggesting a common mechanism of regulation. We tested one such mechanism, namely how interferon-stimulation influences circRNA expression. 67 circRNAs were upregulated by either IFN-β or -γ treatment, with half of these also upregulated during lytic infection. Using gain and loss of function studies we found an interferon-stimulated circRNA, circRELL1, inhibited lytic HSV-1 infection. We have previously reported circRELL1 inhibits lytic KSHV infection, suggesting a pan-herpesvirus antiviral activity. We propose a two-pronged model in which interferon-stimulated genes may encode both mRNA and circRNA with antiviral activity. This is critical in cases of host shutoff, such as alpha- and gamma-herpesvirus infection, where the mRNA products are degraded but circRNAs escape.

Other supporting materials for this manuscript include the following:

Immune stimulation of PBMCs
Human peripheral blood mononuclear cells (PBMCs) were prepared from buffy coats.Buffy coats from two different donors were purified with Ficoll-Paque PLUS (GE Healthcare #17144003).PBMCs were treated with red blood cell lysis buffer (BioLegend #420301) and washed three times with phosphate-buffered saline (Gibco #10010023).10 or 30 ng/mL recombinant human IFN-b (Peprotech #300-02BC) was added to the culture media.After 24 hours RNA was isolated from the cell fraction.

Overrepresentation analysis (ORA)
ORA was performed using a list of genes colinear to circRNAs detected in a given model (>10 raw BSJ read counts).ORA was performed using WebGestalt (5) and plotted using SRplot (https://www.bioinformatics.com.cn/en).

Supporting Dataset 1. Transcriptomic data from HSV-1 human infection models
Transcriptomic data from bulk RNA-Seq data for MRC-5 infected with HSV-1 strain KOS at MOI of 10 plaque-forming unit (PFU)/cell for 12 or 24 hours.Data was mapped to a concatenated genome assembly containing hg38 (gencode.v36),KT899744.1,and ERCC (External RNA Controls Consortium) spike-ins.Tab 1-GenePlotter) Query ERCC normalized gene expression data from the samples by inputting gene symbol.Tab 2-CircPlotter) Query ERCC normalized back splice junction (BSJ) data from the samples by inputting circBase annotation.Tab 3-Summary data) RNA Star 2pass mapping log output file information, including input reads and those mapped to the genome assemblies.Tab 4-Raw counts) CircRNA counts are generated by CircExplorer3 (6) mapping and include only BSJ reads.All other gene counts are outputs from RNA STAR GeneCount (per gene read counts).Tab 5-ERCC norm counts) BSJ or gene counts normalized using ERCC spike-in reads.Tab 6-Log2FC) Log2FC (Infected/Uninfected) for ERCC normalized counts.

Supporting Dataset 2. Transcriptomic data from HCMV infection models
Transcriptomic data from bulk RNA-Seq data for MRC-5 infected with HCMV strain TB40/E at MOI of 3 PFU/cell for 24 or 72 hours (7).Data was mapped to a concatenated genome assembly containing hg38 (gencode.v36)and NC_006273.2.Tab 1-GenePlotter) Query TPM (transcript per million) normalized gene expression data from the samples by inputting gene symbol.Tab 2-CircPlotter) Query TPM normalized BSJ data from the samples by inputting circBase annotation.Tab 3-Summary data) RNA Star 2-pass mapping log output file information, including input reads and those mapped to the genome assemblies.Tab 4-Raw counts) CircRNA counts are generated by CLEAR (CircExplorer3) mapping and include only BSJ reads.All other gene counts are outputs from RNA STAR GeneCount (per gene read counts).Tab 5-TPM norm counts) BSJ or gene counts normalized as TPM.Tab 6-Log2FC) Log2FC (Infected/Uninfected) for TPM normalized counts.

Supporting Dataset 3. Transcriptomic data from KSHV infection models
Transcriptomic data from bulk RNA-Seq data for i.LEC infected with KSHV strain BAC16 at MOI of 1 PFU/cell for 24 or 72 hours or ii.iSLK-BAC16 reactivated with 1 mM Sodium Butyrate 1 ug/mL Doxycycline for 24 or 72 hours.Data was mapped to a concatenated genome assembly containing hg38 (gencode.v36),NC_009333.1,and ERCC spike-ins.Tab 1-GenePlotter) Query ERCC normalized gene expression data from the samples by inputting gene symbol.Tab 2-CircPlotter) Query ERCC normalized BSJ data from the samples by inputting circBase annotation.Tab 3-Summary data) RNA Star 2-pass mapping log output file information, including input reads and those mapped to the genome assemblies.Tab 4-Raw counts) CircRNA counts are generated by CLEAR (CircExplorer3) mapping and include only BSJ reads.All other gene counts are outputs from RNA STAR GeneCount (per gene read counts).Tab 5-ERCC norm counts) BSJ or gene counts normalized using ERCC spike-in reads.Tab 6-Log2FC) Log2FC (Infected/Uninfected) for ERCC normalized counts.

Supporting Dataset 4. Transcriptomic data from HSV-1 mouse infection models
Transcriptomic data from bulk RNA-Seq data for i. mice trigeminal ganglia (TG) latently infected with HSV-1 strain 17 for 4 weeks, or ii.TG explants from mice infected with HSV-1 strain 17 for 5 weeks and grown in the presence or absence of 2 uM JQ1 (Selleckchem, Houston TX) for 12 hours.Data was mapped to a concatenated genome assembly containing mm39 (gencode.vM29),NC_001806.2,and ERCC spike-ins.Tab 1-GenePlotter) Query ERCC normalized gene expression data from the samples by inputting gene symbol.Tab 2-CircPlotter) Query ERCC normalized BSJ data from the samples by inputting circBase annotation.Tab 3-Summary data) RNA Star 2-pass mapping log output file information, including input reads and those mapped to the genome assemblies.Tab 4-Raw counts) CircRNA counts are generated by CLEAR (CircExplorer3) mapping and include only BSJ reads.All other gene counts are outputs from RNA STAR GeneCount (per gene read counts).Tab 5-ERCC norm counts) BSJ or gene counts normalized using ERCC spike-in reads.Tab 6-Log2FC) Log2FC (Infected/Uninfected) for ERCC normalized counts.

Supporting Dataset 5. Transcriptomic data from MHV68 infection models
Transcriptomic data from bulk RNA-Seq data for (i) NIH 3T3 infected with MHV68 strain H2B-YFP MOI of 5 PFU/cell for 6 or 18 hours, or (ii) A20 HE-RIT reactivated with 5 ug/ml Doxycycline and 20 ng/ml 12-O-tetradecanoylphorbol-13-acetate (TPA) for 6 or 24 hours.Data was mapped to a concatenated genome assembly containing mm39 (gencode.vM29),MH636806.1,and ERCC spikeins.Tab 1-GenePlotter) Query ERCC normalized gene expression data from the samples by inputting gene symbol.Tab 2-CircPlotter) Query ERCC normalized BSJ data from the samples by inputting circBase annotation.Tab 3-Summary data) RNA Star 2-pass mapping log output file information, including input reads and those mapped to the genome assemblies.Tab 4-Raw counts) CircRNA counts are generated by CLEAR (CircExplorer3) mapping and include only BSJ reads.All other gene counts are outputs from RNA STAR GeneCount (per gene read counts).Tab 5-ERCC norm counts) BSJ or gene counts normalized using ERCC spike-in reads.Tab 6-Log2FC) Log2FC (Infected/Uninfected) for ERCC normalized counts.

Supporting Dataset 6. Transcriptomic data from interferon stimulation experiments
Transcriptomic data from bulk RNA-Seq data for MRC-5, LEC, or Akata-treated with IFN-b and -g for 48 hours.Data was mapped to a concatenated genome assembly containing hg38 (gencode.v36)and ERCC spike-ins.Tab 1-GenePlotter) Query ERCC normalized gene expression data from the samples by inputting gene symbol.Tab 2-CircPlotter) Query ERCC normalized BSJ data from the samples by inputting circBase annotation.Tab 3-Summary data) RNA Star 2-pass mapping log output file information, including input reads and those mapped to the genome assemblies.Tab 4-Raw counts) CircRNA counts are generated by CLEAR (CircExplorer3) mapping and include only BSJ reads.All other gene counts are outputs from RNA STAR GeneCount (per gene read counts).Tab 5-ERCC norm counts) BSJ or gene counts normalized using ERCC spike-in reads.Tab 6-Log2FC) Log2FC (Infected/Uninfected) for ERCC normalized counts.