Immune characterization and profiles of SARS-CoV-2 infected patients reveals potential host therapeutic targets and SARS-CoV-2 oncogenesis mechanism

Abstract

The spread of SARS-CoV-2 and the increasing mortality rates of COVID-19 create an urgent need for treatments, which are currently lacking. Although vaccines have been approved by the FDA for emergency use in the U.S., patients will continue to require pharmacologic intervention to reduce morbidity and mortality as vaccine availability remains limited. The rise of new variants makes the development of therapeutic strategies even more crucial to combat the current pandemic and future outbreaks. Evidence from several studies suggests the host immune response to SARS-CoV-2 infection plays a critical role in disease pathogenesis. Consequently, host immune factors are becoming more recognized as potential biomarkers and therapeutic targets for COVID-19. To develop therapeutic strategies to combat current and future coronavirus outbreaks, understanding how the coronavirus hijacks the host immune system during and after the infection is crucial. In this study, we investigated immunological patterns or characteristics of the host immune response to SARS-CoV-2 infection that may contribute to the disease severity of COVID-19 patients. We analyzed large bulk RNASeq and single cell RNAseq data from COVID-19 patient samples to immunoprofile differentially expressed gene sets and analyzed pathways to identify human host protein targets. We observed an immunological profile of severe COVID-19 patients characterized by upregulated cytokines, interferon-induced proteins, and pronounced T cell lymphopenia, supporting findings by previous studies. We identified a number of host immune targets including PERK, PKR, TNF, NF-kB, and other key genes that modulate the significant pathways and genes identified in COVID-19 patients. Finally, we identified genes modulated by COVID-19 infection that are implicated in oncogenesis, including E2F transcription factors and RB1, suggesting a mechanism by which SARS-CoV-2 infection may contribute to oncogenesis. Further clinical investigation of these targets may lead to bonafide therapeutic strategies to treat the current COVID-19 pandemic and protect against future outbreaks and viral escape variants.

Keywords: Bulk RNASeq analysis; Cancer; Gene expression analysis; Immune characterization of COVID19 patients; Immune profiling of COVID19 patients; SARS-COV-2; Single cell scRNAseq analysis.

Figure 1.

Schematic of standard single-cell RNA-seq pipeline in Partek® Flow®. Single cell count matrices were imported from GEO and processed for downstream analysis.

Figure 2.. RNA-seq profile of nasopharyngeal swabs and PBMCs from COVID-19 patients.

(A) Volcano plot of differentially expressed genes in SARS-CoV-2 infected patients versus uninfected patients from GSE152075 analysis. (B) Volcano plot of differentially expressed genes in SARS-CoV-2 infected patients versus uninfected patients from GSE152418 analysis. An adjusted p-value (q-value < 0.05) and fold change (log2 fold change ≥ ±2) were used to determine significantly downregulated or upregulated genes. The log2 fold change of the five topmost significantly upregulated and downregulated genes (FDR <0.01) are highlighted.

Figure 3.. Pathway analysis of DEGs in nasopharyngeal swabs from COVID-19 patients.

(A) Most significantly downregulated canonical pathways in IPA, determined by –log(p-value). The orange and blue-colored bars in the bar chart indicate predicted pathway activation or predicted inhibition, respectively, based on z-score. Gray bars indicate pathways for which no prediction could be made by IPA. (B) Most significantly upregulated canonical pathways in IPA.

Figure 4.. Pathway analysis of DEGs in PMBCs from COVID-19 patients.

Kinetochore Metaphase Signaling Pathway was ranked first in the significant IPA canonical pathways for GSE152418 based on –log(p-value). The orange and blue-colored bars in the bar chart indicate predicted pathway activation or predicted inhibition, respectively. Gray bars indicate pathways for which no prediction can be made for the data provided.

Figure 5.. Potential oncogenesis mechanism of SARS-CoV-2 through interaction of the N protein with the Rb-E2F complex.

This figure was created in BioRender based on the Coronavirus Pathogenesis Pathway from IPA and the G1/S Checkpoint in BioRender, modified to highlight the key genes most significant in the dataset GSE152418. Genes significantly downregulated are indicated in red and genes significantly upregulated are indicated in blue. Genes not significant in this dataset are indicated in gray and viral proteins are indicated in yellow. Adapted from “G1/S Checkpoint”, by BioRender.com (2020). Retrieved from https://app.biorender.com/biorender-templates.

Figure 6.. BALF Clustering Analysis results visualized by Global UMAP.

(A) Patient samples were grouped by disease severity. (B) Cells were classified by four immune cell subsets: B-cells, Macrophages, NK cells, and T cells. N/A defines all unclassified cells. (C) Graph-based analysis was performed in Partek® Flow®. 25 clusters were reported. (D) Top features in each immune cell subset generated by clustering analysis.

Figure 7.. Differential gene expression analysis of RNAseq data from BALF of COVID-19 patients.

(A) 10xGenomics Human Immunology Panel was used to filter differentially expressed genes for GSE145926 and annotated based on function. (B) The top row shows the DEGs for SARS-CoV-2 infected COVID-19 patients (mild+severe) versus healthy patients. The bottom row shows the DEGs for severe versus mild COVID-19 patients. The ten most significantly upregulated genes are ranked in green and the ten most significantly downregulated genes are ranked in red, based on log2(fold change).

Figure 8.. Differential gene expression analysis across immune cell subsets in RNAseq data from BALF of COVID-19 patients.

(A) 10X Genomics Human Immunology Panel was used to filter differentially expressed genes in each immune subset (NK cells, T cells, B cells, and Macrophages) of severe versus mild COVID-19 patients. (B) Scatter plot for IKFZ2, one of the most downregulated genes, in immune cell subsets. (C) Scatter plot for HLA-DQA2, one of the most downregulated genes, in the immune cell subsets.

Figure 9.. Blood Buffy Coat Clustering Analysis results visualized by Global UMAP

(A) Patient samples were grouped by disease severity. (B) Cells were classified by four immune cell subsets: B-cells, Macrophages, NK cells, and T cells. N/A defines all unclassified cells. (C) Graph-based analysis was performed in Partek® Flow®. 25 clusters were reported (D) The top features generated by clustering analysis in each immune cell subset.

Figure 10.. Differential gene expression of RNAseq data from blood buffy coat samples of severe and recovered COVID-19 patients.

10X Genomics Human Immunology Panel was used to filter differentially expressed genes of severe versus mild COVID-19 patients (left panel) and recovered versus severe COVID-19 patients (right panel).

Figure 11.. Differential gene expression of RNAseq data from blood buffy coat of recovered COVID-19 patients.

Differentially expressed genes across immune cell subsets (NK cells, T cells, B cells, and macrophages) found in recovered versus severe COVID-19 patients.