Single-Cell RNA-Seq Profiling of Transposable Element Expression in Human Peripheral Blood Cells During Viral Infections

Abstract

Transposable elements (TEs) are key regulators of immunity in both health and disease. It has been proven that the activity and transcriptional expression levels of TEs increase during viral infections, correlating with the antiviral response. This study investigates non-LTR TE (LINE, SINE, and SVA) transcriptomic signatures in human PBMCs during infections caused by influenza A virus, HIV, and SARS-CoV-2 (Delta/Omicron variants) using single-cell RNA sequencing (scRNA-seq) data from 98 patients. In the HIV and SARS-CoV-2 patient cohorts, unique cell-specific TE expression patterns were identified that allow for the differentiation of disease severity, prediction of disease progression, and assessment of the therapy's efficacy. The expression of LINE elements was found to be more dependent on the nature and course of the disease than that of SINE elements. The most variable TE expression profile was observed in precursor cytotoxic T-lymphocytes (T CD8+ Naive cells), which depended on the virus type and the severity of the viral disease. For this cell type, a bioinformatic analysis of the co-expression regulation of TE transcriptional networks and transcription factors during viral infections was performed. This analysis identified key players among those most involved in virus-specific responses, which could serve as diagnostic biomarkers or therapeutic targets for treating diseases caused by influenza A virus, HIV, and SARS-CoV-2. This work confirms the involvement of non-LTR TEs in mediating antiviral responses. Further research into the mechanisms of TE participation in antiviral defense is necessary to recommend them as potential biomarkers for the diagnosis, monitoring, and assessment of antiviral therapy, or as therapeutic targets for viral infections of various origins.

Keywords: COVID-19; HIV; influenza A; scRNA-seq; transposable elements; viral infection.

Figure 1

Single-cell transcriptomic profiling and annotation of PBMCs from patients with viral infections. (A) Study design schematic: merging in-house and public cohorts for comparative single-cell transcriptomics. (B) UMAP visualization of clustering based on gene expression profiles, showing annotated cell types. (C) Percentage and absolute count of each cell type in the analyzed samples. (D) Correlation matrix (Wilcoxon test) of gene expression profiles across different cell types. (E) Expression levels of known marker genes (according to the literature data) in the identified cell populations.

Figure 2

Landscape of TE expression in PBMCs during viral infections. (A) UMAP visualization of clustering based on TE expression with assigned cell type annotations. (B) Correlation matrix (Wilcoxon test) of TE expression profiles across different annotated cell populations. (C) Expression levels of TEs specific to certain conditions, compared across different cohorts.

Figure 3

Cell-type and cohort-specific dynamics of LINE and SINE retrotransposon expression. (A) Distribution and expression level of LINE elements in blood cells, where the color scale reflects the normalized expression score. (B) Distribution and expression level of SINE elements in blood cells, where the color scale reflects the normalized expression score. (C) Quantitative summary of TE dysregulation across cell types and infections. The number of significantly upregulated (red) or downregulated (blue) LINE and SINE retrotransposons is shown for each PBMC subtype and patient cohort. Counts were obtained from single-cell differential expression testing (each cohort vs. healthy donors) within annotated cell clusters, preserving single-cell resolution in the statistical model.

Figure 4

Co-expression modules of TEs in T CD8+ Naive cells are linked to distinct immune pathways and infection outcomes. (A) Identification of TE modules and analysis of their expression correlations in T CD8+ Naive cells. (B) Expression profiles of the identified TE modules across different cohorts within T CD8+ Naive cells. (C) Association between TE module expression and the activity of immunological functional pathways in T CD8+ Naive cells. (D) Expression of TE sets characteristic of specific modules across different cohorts within T CD8+ Naive cells.

Figure 5

LINE and TF co-expression networks in viral infections. (A) Study design includes: obtaining the list of LINE TEs from Figure 3C data; using RMSK data and the GRCh38 reference genome to obtain TE coordinates and sequences; extracting promoter regions from the reference genome; running Homer (findMotifsGenome.pl); and obtaining a list of motifs (top 10) for each unique cohort and cell type combination. (B,C) Heatmaps displaying high and low enrichment of TF gene-LINE element interaction motifs in PBMCs from patients with viral infections. (D) Reconstructed co-expression networks of LINE elements and TOP-5 differentially expressed TFs, identified from the analysis of heatmaps B and C, in T CD8+ Naive cells. (E) Co-expression networks of LINE elements and TOP-5 differentially expressed TFs, identified for T CD8+ Naive cells of the “COVID: Delta variant” group. Orange lines indicate positive expression correlation; blue lines indicate negative correlation. Green borders highlight TFs characteristic of both highly and lowly regulated LINE elements. The purple border marks a TF characteristic; in this cell type, these are exclusively shown for the “COVID: Delta variant” cohort.