In silico prediction of cellular gene targets of herpesvirus encoded microRNAs.

Herpesviruses have evolved to encode multiple microRNAs [viral miRNAs (v-miRs)], a unique feature of this family of double stranded DNA (dsDNA) viruses. However, functional role of these v-miRs in host-pathogen interaction remains poorly studied. In this data, we examined the impact of oral disease associated v-miRs viz., miR-H1 [encoded by herpes simplex virus 1 (HSV1)] and miR-K12-3 [encoded by Kaposi sarcoma-associated herpesvirus (KSHV)] by identifying putative targets of viral miRNAs. We used our published microarray data (GSE107005) to identify the transcripts downregulated by the v-miRs. The 3' untranslated region (UTR) of these genes were extracted using BioMart tool on Ensembl and subjected to RNA:RNA interaction employing RNA Hybrid. We obtained hundreds of potential and novel miR-H1 and miR-K12-3 binding sites on the 3'UTR of the genes downregulated by these v-miRs. The information can provide likely regulatory mechanisms of the candidate v-miRs through which they can exert biological impact during herpesvirus infection and pathogenesis.

Specifications Table

Value of the data

The data presented is valuable for the reasons listed below:

The data provided here enlists human genes that were downregulated by herpesvirus derived miRNAs viz., miR-H1 (Herpes simplex virus 1) and miR-K12-3 (Kaposi sarcoma-associated herpesvirus) and harbor potential v-miR binding sites.

These genes can provide new avenues to begin focused research on the role of viral miRNAs viz., miR-H1 and miR-K12-3 in the pathogenesis of oral mucosal diseases.

Due to lack of online tools that can predict viral miRNA binding sites with high confidence, this methodology can provide a starting point to share large datasets examining global impact of v-miRs to identify more reliable candidate targets or facilitate development of algorithms to predict v-miR targets with a high degree of confidence.

Data

Human Herpesviruses (HHV) are dsDNA viruses that are highly prevalent worldwide [1]. A key feature of all herpesviruses is their capability to encode microRNAs [2]. These small non-coding RNAs are implicated in wide range of biological functions that govern host-pathogen interaction [2]. Recent evidences show a likely association of herpesvirus in oral diseases, however a role of viral components in the oral pathogenesis remains unknown [3], [4]. We recently identified four viral miRNAs that were upregulated in human subjects with inflamed pulps and diseased gingival biopsies compared with healthy tissues [5], [6]. Our recent transcriptome and miRnome analysis showed v-miRs can profoundly impact a specific set of genes in oral keratinocytes which are targeted by herpesviruses [6], [7]. However, the direct gene targets of these viral miRNAs will shed light on the possible pathways through which viral miRNAs can modulate host cell functions. The data presented here provides a list of potential miR-H1 and miR-K12-3 binding sites on the 3'UTR of host transcripts that were significantly downregulated by these v-miRs in our previously published microarray (GSE107005). Table 1, Table 2 provides list of some representative interaction for miR-H1 and miR-K12-3, respectively, identified in our screening. The remaining interactions are listed as supplementary information in the Supplementary text file 1 (for miR-H1) and Supplementary text file 2 (for miR-K12-3).

Table 1

Predicted miR-H1-5p binding sites on the downregulated host genes. Sequence alignment of selected potential miR-H1-5p binding sites is shown. Only the binding sites with mfe<−20 kcal/mol are shown.

v-miRNA	Target gene	vmiR and target gene sequence alignment
hsv1-miR-H1–5p	PREPL	Position 2928
		Target 5′ A UU G A 3′
		UCAUUUC GU UCUUCUAUU
		GGUGAAG CA GGAAGGUAG
		miRNA 3′ GG 5′
hsv1-miR-H1–5p	TTC33	Position 899
		Target 5′ U AA AA A 3′
		CCA UUUU CCUUUCGUC
		GGU AGGG GGAAGGUAG
		miRNA 3′ GA CA 5′
hsv1-miR-H1–5p	ATG16L1	Position 1965
		Target 5′ A AG U A A 3′
		CUACU C CUG CCUUCCAU
		GGUGA G GGC GGAAGGUA
		miRNA 3′ A A G 5'
hsv1-miR-H1–5p	NOTCH2NL	Position 2443
		Target 5′ G G G U G 3′
		CA U CCC UCCUUCCAUU
		GU A GGG AGGAAGGUAG
		miRNA 3′ G G A C 5′
hsv1-miR-H1–5p	ZNF106	Position 1227
		Target 5′ G A U 3′
		UCGCUUUCC G CCUUUUGUU
		GGUGAAGGG C GGAAGGUAG
		miRNA 3′ A 5′
hsv1-miR-H1–5p	CHML	Position 212
		Target 5′ A AC AU A 3′
		UCAC CUC UUCUUUCAUC
		GGUG GGG AGGAAGGUAG
		miRNA 3′ AA C 5′
hsv1-miR-H1–5p	CCDC91	Position 464
		Target 5′ A CC AC G 3′
		CAUU CCC UCUUUCCAU
		GUGA GGG AGGAAGGUA
		miRNA 3′ G A C G 5′
hsv1-miR-H1–5p	RABEP1	Position 88
		Target 5′ C A 3′
		CCAUUUUUC UUUUUCUGU
		GGUGAAGGG AGGAAGGUA
		miRNA 3′ C G 5′
hsv1-miR-H1–5p	TGFBR1	Position 4034
		Target 5′ G A 3′
		UACUUUCUG UUUUCUGU
		GUGAAGGGC GGAAGGUA
		miRNA 3′ G A G 5′
hsv1-miR-H1–5p	TRIM52	Position 453
		Target 5′ G C UU A 3′
		UACU C GUUUUUCUGUU
		GUGA G CAGGAAGGUAG
		miRNA 3′ G A GG 5′
hsv1-miR-H1–5p	DYM	Position 8446
		Target 5′A A A G 3′
		UACUU UG UCUUUCCAUU
		GUGAA GC AGGAAGGUAG
		miRNA 3′ G G 5′
hsv1-miR-H1–5p	NDUFS1	Position 1742
		Target 5′ A A CA C 3′
		GCUGU UGUUU CAGAGUGUG
		CGACG GCAGG GUCUUACAC
		miRNA 3′ AG A U 5′
hsv1-miR-H1–5p	SLC4A7	Position 1345
		Target 5′ U UG G G 3′
		UACU UUU GUCCUUUUAU
		GUGA AGG CAGGAAGGUA
		miRNA 3′ G G G 5′
hsv1-miR-H1–5p	PRRC1	Position 47
		Target 5′ G C U 3′
		UAC UUCC UCCUUUUGUU
		GUG AGGG AGGAAGGUAG
		miRNA 3′ G A C 5′
hsv1-miR-H1–5p	IL1RAP	Position 2670
		Target 5′ U A U A 3′
		UACUU UU UCUUUCCAU
		GUGAA GG AGGAAGGUA
		miRNA 3′ G G C G 5′

Table 2

Predicted miR-K12-3 binding sites on the downregulated host genes. Sequence alignment of selected potential miR-K12-3 binding sites on the predicted targets is shown. Only the binding sites with mfe<−20 kcal/mol are listed.

v-miRNA	Target gene	vmiR and target gene sequence alignment
kshv-miR-K12-3	CBX5	Position 8806
		Target 5′ U AUC G U 3′
		UC UUGUU U UUGGAAUGUGA
		AG GACGG G AGUCUUACACU
		miRNA 3′ CCA G 5′
kshv-miR-K12-3	GOLGA3	Position 3592
		Target 5′ G AGU GA A 3′
		GC GU UUCU UAGGAUGUGA
		CG CG AGGA GUCUUACACU
		miRNA 3′ AG AGC 5′
kshv-miR-K12-3	UIMM21	Position 55
		Target 5′ AU 3′
		GCUGCC UUC CAGAAUGUG
		CGACGG AGG GUCUUACAC
		miRNA 3′ AG C AU 5′
kshv-miR-K12-3	UBL1X	Position 2516
		Target 5′ GU A G U A 3′
		GCU U GUCUU A GAAUGUGA
		CGA G CAGGA U CUUACACU
		miRNA 3′ AG C G G5′
kshv-miR-K12-3	FKBP14	Position 1178
		Target 5′ AAA AG U U 3′
		CUG GUU C GGGUGUGG
		GAC CAG G CUUACACU
		miRNA 3′ AGC GG GA U 5′
kshv-miR-K12-3	DSUN	Position 659
		Target 5′ A A GAG A C 3′
		UC UUGU UGUCUUC G GAAUGUG
		AG GACG GCAGGAG U CUUACAC
		miRNA 3′ CU 5′
kshv-miR-K12-3	ORC2	Position 372
		Target 5′ G GU A 3′
		UGU UUGUUC CAGAGUGUGG
		GCG GGCAGG GUCUUACACU
		miRNA 3′ A AC A5′
kshv-miR-K12-3	COPA	Position 33
		Target 5′ A CC AG U 3′
		UGUU CC CC AGAAUGUG
		GCGA GG GG UCUUACAC
		miRNA 3′ A CCA AG U 5′
kshv-miR-K12-3	POLR3B	Position 228
		Target 5′ G UAU A AG U C 3′
		GCUGC UG UC C A GGAUGUGA
		CGACG GC AG G U CUUACACU
		miRNA 3′ AG AG 5′
kshv-miR-K12-3	RAB3D	Position 260
		Target 5′ C UU 3′
		UUGCUGCU UCC AGGGUGUG
		AGCGACGG AGG UCUUACAC
		miRNA 3′ CAG U 5′
kshv-miR-K12-3	SLC1A4	Position 2435
		Target 5′ G G GC 3′
		G UGCU UCC AGAGUGUG
		C ACGG AGG UCUUACAC
		miRNA 3′ AG G CAG U 5′
kshv-miR-K12-3	CCND2	Position 1742
		Target 5′ AAA CA C 3′
		GCUGU UGUUU CAGAGUGUG
		CGACG GCAGG GUCUUACAC
		miRNA 3′ AG AU 5′
kshv-miR-K12-3	CD101	Position 48
		Target 5′ A GAA A 3′
		UUG GCU CC AGGGUGUGA
		AGC CGG GG UCUUACACU
		miRNA 3′ GA CA AG 5′
kshv-miR-K12-3	RAB40B	Position 98
		Target 5′ G AA GC U 3′
		CG UGCUG CUU GAAUGUG
		GC ACGGC GGA CUUACAC
		miRNA 3′ A G AGU U 5′
kshv-miR-K12-3	PIUPNM3	Position 2628
		Target 5′ A GG GU U 3′
		GUUG CG U U GAGUGUG
		CGAC GC G A CUUACAC
		miRNA 3′ AG GA G GU U 5′

Experimental design, materials and methods

Primary gingival human oral keratinocyte (HOK) culture

Primary HOK (human gingival epithelial cells) were purchased from ScienCell Research Laboratories (Carlsbad, CA). Cultures were tested for HOK markers by immunofluorescent methods using antibodies to cytokeratine-8, -18 and -19 and were negative for Human Immunodeficiency Virus 1 (HIV-1), Hepatitis B Virus (HBV), Hepatitis C Virus (HCV), mycoplasma, bacteria, yeast and fungi. Cells were cultured using DermaLife K Keratinocyte Medium Complete Kit (Lifeline Cell Technology, Frederick, MD).

Transient miRNA transfections and total RNA isolation

Transient viral miRNA (miR-H1 or miR-K12-3) or control mimic transfections in HOK were performed using Lipofectamine 2000 reagent (Life Technologies, San Diego, CA) as described previously [8], [9]. Cells were transfected with viral miRNA mimics (Qiagen, Gaithsburg, MD, USA) at a final concentration of 15 nM for 36 h. Total RNA was isolated using the miRNeasy kit (Qiagen).

Microarray analysis

We used our published microarray data deposited in the Gene Expression Omnibus public database under Accession Number GSE107005 for the identification of putative viral miRNA target transcripts [6]. Array data were in compliance with Minimum Information About a Microarray Experiment (MIAME) guidelines.

V-miR target prediction of differentially downregulated genes

To identify miR-H1 and miR-K12-3 gene targets with high confidence, we first selected downregulated genes. The 3′UTR of these genes were extracted using BioMart tool on Ensembl (http://www.ensembl.org/biomart/martview/aa867419c3c6fd64f94af6d4a6549d3c). Briefly, we selected Ensembl Genes 87 database and Human Genes dataset (GRCh38.p7). Next, the "Filters" were selected to match the input genes list. In the "Gene" tab set the "ID list limit" filter to "HGNC symbol(s)". Finally, to procure the 3'UTR sequences “Attributes” were set. In the "Attributes", select "Sequences" and then select 3′UTR start and 3'UTR end, click "Ensembl Gene ID" and "Associated Gene Name". The results were exported to by selecting "File", "FASTA" and "Unique results only”. This was done separately for miR-H1 and miR-K12-3 datasets.

v-miR-target 3’UTR interaction was assessed by target prediction tool RNAHyrbid software (https://bibiserv2.cebitec.uni-bielefeld.de/rnahybrid?id=rnahybrid_view_submission). The procured 3′UTR sequences and miR-H1 and miR-K12-3 sequences (extracted from miRbase v.21) were provided as input for RNA Hybrid analysis. The stringency parameters were set-up for individual sequences and we opted for three hits per target to highlight any probable v-miR binding sequence present on the target.

We considered the following parameters to select putative v-miR regulated genes. (i) There should be high sequence complementarity in the seed region (positions 2–8 nt from 5′ of miRNA), with only 1 mismatch allowed. (ii) For stringency, we picked v-miR-target interactions where more than 11 nts of the v-miR sequence are involved in the interaction. (iii) If there is any mismatch in the seed regions, this should be compensated by strong binding beyond the seed region. (iv) The bulge in the interaction region should not involve more than 3 nucleotides. (v) Entropy of the v-miR-target interaction was set at stringent level with cut-off <22 kcal/mol.

Subject area	Biology
More specific subject area	Molecular Virology
Type of data	Text file
How data was acquired	Microarray and Bioinformatics
Data format	Filtered and analyzed
Experimental factors	Cells were transfected with v-miRs or control mimics
Experimental features	Genes downregulated by v-miRs were scanned for putative miRNA binding sites on the 3'UTR using RNA Hybrid tool.
Data source location	NA
Data accessibility	Data is presented as supporting file text with this manuscript. Microarray data of transcriptome wide changes in miR-H1 and miR-K12-3 overexpressing human oral keratinocytes compared to control mimics is deposited in the Gene Expression Omnibus public database under Accession Number GSE107005 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE107005).