Human Papillomavirus-Encoded microRNAs as Regulators of Human Gene Expression in Anal Squamous Cell Carcinoma: A Meta-Transcriptomics Study

Abstract

Introduction: Anal squamous cell carcinoma (ASCC) is a rare but increasingly common gastrointestinal malignancy, mainly associated with oncogenic human papillomaviruses (HPVs). The role of non-coding RNAs (ncRNAs) in tumorigenesis is recognized, but the impact of viral ncRNAs on host gene expression remains unclear. Methods: We re-analyzed total RNA-Seq data from 70 anal biopsies: 31 low-grade squamous intraepithelial lesions (LGSIL), 16 high-grade SIL (HGSIL), and 23 ASCC cases. Microbial composition was assessed taxonomically. Novel viral miRNAs were predicted using vsRNAfinder and linked to host targets using TargetScan and expression correlation analyses. Results: Microbial profiling revealed significant differences in abundance, with Alphapapillomaviruses types 9, 10, and 14 enriched across lesion grades. We identified 90 novel viral miRNAs and 177 significant anti-correlated miRNA-mRNA interactions. Target genes were enriched in pathways related to cell cycle, epithelial-mesenchymal transition, lipid metabolism, immune modulation, and viral replication. Discussion: Our findings suggest that HPV-derived miRNAs, including those from low-risk types, may contribute to neoplastic transformation by modulating host regulatory networks. Conclusion: This study highlights viral miRNAs as potential drivers of HPV-related anal cancer and supports their utility as early biomarkers and therapeutic targets in ASCC.

Keywords: anal squamous cell carcinoma; de novo miRNAs; human papillomavirus; oncomiR; squamous intraepithelial lesion; target prediction; transcriptomics; viral miRNAs.

Figure 1

Differential abundance analysis of viruses across the four comparisons: anal squamous cell carcinoma (ASCC) vs. high-grade squamous intraepithelial lesions (HGSILs), ASCC vs. low-grade squamous intraepithelial lesions (LGSILs), HGSILs vs. LGSILs, and ASCC vs. LGSILs and HGSILs. Circles represent downregulated viruses, while hexagons represent upregulated viruses among the samples. The size of each point corresponds to the logFC (fold change) value. False discovery rate (FDR) is represented using a color gradient.

Figure 2

Volcano plots showing differentially expressed HPV miRNAs across the following comparisons: (A) high-grade squamous intraepithelial lesions (HGSILs) vs. low-grade squamous intraepithelial lesions (LGSILs), (B) anal squamous cell carcinoma (ASCC) vs. HGSILs, (C) ASCC vs. LGSILs, and (D) ASCC vs. combined HGSILs + LGSILs. Colored dots indicate significant miRNAs (FDR < 0.05), yellow dots represent upregulated miRNAs, and blue dots indicate downregulated ones. For each comparison, the most significantly upregulated or downregulated miRNAs were listed according to their fold changes (|logFC|).

Figure 3

Volcano plots represent differentially expressed protein-coding genes (PCGs) in three comparisons: (A) anal squamous cell carcinoma (ASCC) vs. high-grade squamous intraepithelial lesions (HGSILs), (B) HGSILs vs. low-grade squamous intraepithelial lesions (LGSILs), and (C) ASCC vs. combined HGSILs + LGSILs. Colored dots represent statistically significant miRNAs (FDR < 0.05): yellow points represent upregulated PCGs, while blue points depict downregulated PCGs. Labeled genes are those with the largest absolute logFC (fold change) values and statistical significance. For each comparison, the most significantly upregulated or downregulated genes were listed according to their fold changes (|logFC|).

Figure 4

Networks representing the correlation patterns between differentially expressed human papillomavirus (HPV) miRNAs and their significantly anti-correlated target protein-coding genes (PCGs) (p < 0.05) across all comparisons: (A) anal squamous cell carcinoma (ASCC), high-grade squamous intraepithelial lesions (HGSILs), and low-grade squamous intraepithelial lesions (LGSILs); (B) ASCC vs. HGSILs; (C) ASCC vs. LGSILs; and (D) ASCC vs. HGSILs + LGSILs. Empty circles denote HPV-miRNAs, while blue circles represent target genes. Edges indicate significant negative correlations, calculated using the Spearman correlation coefficient, with edge color intensity corresponding to the strength of the correlation. The ten most highly anti-correlated genes and the five HPV-derived miRNAs with the highest degree of connectivity are labeled in each network. For enhanced visualization and interactive exploration, these networks are shown in HTML versions of the figures, available at http://bioinfo.ipb.csic.es/Papers/ncRNA_HPV/, accessed on 1 June 2025.

Figure 5

Functional enrichment analysis of differentially expressed PCGs predicted to be target genes regulated by differentially expressed HPV-miRNAs. Ten most significantly enriched GO biological processes pathways ordered by gene ratios in the comparisons (A) anal squamous cell carcinoma (ASCC) vs. high-grade squamous intraepithelial lesions (HGSILs), (B) ASCC vs. low-grade squamous intraepithelial lesions (LGSILs), and (C) ASCC vs. HGSILs + LGSILs are shown. The size of the dots represents the number of predicted target genes in the gene list associated with the GO term and the color of the dots represents the p-value.