SARS-CoV-2 Infection-Induced Promoter Hypomethylation as an Epigenetic Modulator of Heat Shock Protein A1L (HSPA1L) Gene

Abstract

Numerous researches have focused on the genetic variations affecting SARS-CoV-2 infection, whereas the epigenetic effects are inadequately described. In this report, for the first time, we have identified potential candidate genes that might be regulated via SARS-CoV-2 induced DNA methylation changes in COVID-19 infection. At first, in silico transcriptomic data of COVID-19 lung autopsies were used to identify the top differentially expressed genes containing CpG Islands in their promoter region. Similar gene regulations were also observed in an in vitro model of SARS-CoV-2 infected lung epithelial cells (NHBE and A549). SARS-CoV-2 infection significantly decreased the levels of DNA methyltransferases (DNMT1, DNMT3A, and DNMT3B) in lung epithelial cells. Out of 14 candidate genes identified, the expression of 12 genes was upregulated suggesting promoter hypomethylation, while only two genes were downregulated suggesting promoter hypermethylation in COVID-19. Among those 12 upregulated genes, only HSPA1L and ULBP2 were found to be upregulated in AZA-treated lung epithelial cells and immune cells, suggesting their epigenetic regulation. To confirm the hypomethylation of these two genes during SARS-CoV-2 infection, their promoter methylation and mRNA expression levels were determined in the genomic DNA/RNA obtained from whole blood samples of asymptomatic, severe COVID-19 patients and equally matched healthy controls. The methylation level of HSPA1L was significantly decreased and the mRNA expression was increased in both asymptomatic and severe COVID-19 blood samples suggesting its epigenetic regulation by SARS-CoV-2 infection. Functionally, HSPA1L is known to facilitate host viral replication and has been proposed as a potential target for antiviral prophylaxis and treatment.

Keywords: COVID-19; DNA methylation; HSPA1L; SARS-CoV-2; bioinformatics; epigenetic.

FIGURE 1

Common differential expressed gene signatures in lung autopsies from five COVID-19 patients. (A) RNAseq data set containing expression profile from autopsy samples (Formalin-fixed paraffin-embedded lung tissues) from five patients deceased due to SARS-Cov2 infection, were analyzed. Each patient sample had three or more replicates. Mean expression was calculated and Log2 fold change was computed for each patient sample. Common differentially expressed genes were identified with a threshold of a Log2FC of 1.5 or above. Non-coding genes or pseudogenes were excluded. Total 101 genes were selected based on these criteria and the Venn diagram was constructed to represent the common genes. (B) Database containing 81 RNAseq samples derived from 12 Healthy lung donor subjects was utilized to identify selectivity of these commonly expression 101 genes. Most of the genes were non-selectively expressed. (C) Bar graph showing expression of those 101 genes in COVID-19 lung tissue compared to normal lung. All the values are representing Log2FC ± SEM for the relative mRNA expression in five COVID-19 cases compared vs. healthy controls.

FIGURE 2

Nineteen of the common differentially expressed genes had CpG Island in their promoter region. (A) Data from the Genotype-Tissue Expression (GTEx) database containing normal tissue expression datasets were utilized to analyze the tissue-specific expression levels of all the ubiquitously expressed genes. (B) GTEx database was utilized to present the gene expression profile of 19 short-listed epigenetic genes in normal human lung tissue (TPM: transcripts per millions). (C) A database containing 81 RNAseq samples derived from 12 Healthy lung donor subjects was utilized to identify cell-specific selectivity of these 19 epigenetic genes. (N.B.: Endothelial Cells: ULBP2; Epithelial Cells: FOXJ1, LRRC56; Immune Cells: None; Mesenchymal Cells: MME; Non-selectively expressed: CBR3, CECR7, DERL3, DSC3, HS3ST2, HSPA1L, KLRG2, NIPAL1, NR4A1, PRADC1, SLC19A2, SYT2, ULBP3, USP53, and VMAC).

FIGURE 3

The epigenetic candidate genes are upregulated during COVID-19, but not due to chronic lung inflammation. (A) Relative mRNA expression (Log2FC) of 19 selected epigenetic genes in COVID-19 infected lung tissue were plotted and were plotted against Relative mRNA expression of the same genes in lung tissues obtained from COPD patients (distinctive healthy controls RNAseq data were used as a control for each disease-type). (B) Relative mRNA expression (FC) of 19 selected epigenetic genes in human lung epithelial cell lines infected with SARS-Cov-2. GEO data set containing expression profiling by high throughput sequencing of human lung epithelial cell lines infected with SARS-Cov-2 was searched. Dataset included had independent biological triplicates of primary human lung epithelium (NHBE) mock-treated or infected with SARS-CoV-2 and independent biological triplicates of transformed lung alveolar (A549) cells mock-treated or infected with SARS-CoV-2. (C,D) GEO data set containing expression profiling by high throughput sequencing of human lung epithelial cell lines (NHBE, A549) and monocyte cell line (U937) treated with 0.5 μM Aza-dC (AZA) were analyzed for mean mRNA expression (FC) of these 19 epigenetic genes. Genes that are found upregulated are more likely suspected to be epigenetically regulated in lung epithelial cells or immune cells. All bar graph data represents mean relative mRNA expression for pairwise comparison of infected/treated samples vs. the control and error bars represents SEM.

FIGURE 4

Quantitative Methylation-specific PCR and real-time PCR confirms HSPAIL promoter hypomethylation-overexpression with SARS-CoV-2 infection. Methylation levels of HSPAIL and ULBP2 were measured by qMSP. (A,B) For the HSPA1L gene, the methylation levels were significantly lower but mRNA expression was significantly higher in both asymptomatic and symptomatic COVID-19 compared to healthy controls. (C,D) No difference in ULBP2 methylation levels or mRNA expression was observed between the blood of healthy and COVID-19 samples. *P-value < 0.05 (n = 5 for each group); the red error bar shows standard error of mean (SEM) and the black horizontal bar shows the mean value.