Transcriptome of nasopharyngeal samples from COVID-19 patients and a comparative analysis with other SARS-CoV-2 infection models reveal disparate host responses against SARS-CoV-2

Abstract

Background: Although it is becoming evident that individual's immune system has a decisive influence on SARS-CoV-2 disease progression, pathogenesis is largely unknown. In this study, we aimed to profile the host transcriptome of COVID-19 patients from nasopharyngeal samples along with virus genomic features isolated from respective host, and a comparative analyses of differential host responses in various SARS-CoV-2 infection systems.

Results: Unique and rare missense mutations in 3C-like protease observed in all of our reported isolates. Functional enrichment analyses exhibited that the host induced responses are mediated by innate immunity, interferon, and cytokine stimulation. Surprisingly, induction of apoptosis, phagosome, antigen presentation, hypoxia response was lacking within these patients. Upregulation of immune and cytokine signaling genes such as CCL4, TNFA, IL6, IL1A, CCL2, CXCL2, IFN, and CCR1 were observed in lungs. Lungs lacked the overexpression of ACE2 as suspected, however, high ACE2 but low DPP4 expression was observed in nasopharyngeal cells. Interestingly, directly or indirectly, viral proteins specially non-structural protein mediated overexpression of integrins such as ITGAV, ITGA6, ITGB7, ITGB3, ITGA2B, ITGA5, ITGA6, ITGA9, ITGA4, ITGAE, and ITGA8 in lungs compared to nasopharyngeal samples suggesting the possible way of enhanced invasion. Furthermore, we found comparatively highly expressed transcription factors such as CBP, CEBP, NFAT, ATF3, GATA6, HDAC2, TCF12 which have pivotal roles in lung injury.

Conclusions: Even though this study incorporates a limited number of cases, our data will provide valuable insights in developing potential studies to elucidate the differential host responses on the viral pathogenesis in COVID-19, and incorporation of further data will enrich the search of an effective therapeutics.

Keywords: COVID-19; Genome variations; Host transcriptional response; Immune response; Integrins; SARS-CoV-2.

Fig. 1

A brief workflow of the whole study

Fig. 2

Genomic information of the sequenced SARS-CoV-2 isolates. a Genome coverage normalized density map for the four sequenced SARS-CoV-2 isolates. b Pie-chart illustrating the different types of variations found within these four isolates. c Genome location-wise representation of the mutations and their associated frequency. d Isolate-wise variation information. e Gene-wise amount and type of mutations. f Annotated impacts of the different mutations (only those are shown which have frequencies more than 1). g Frequencies of selected unique mutations observed in these isolates

Fig. 3

Differential gene expression analysis of the studied nasophryngeal samples of COVID-19 patients. a Variance plot. This plots the standard deviation of the transformed data, across samples, against the mean, using the variance stabilizing transformation. The vertical axis in the plots is the square root of the variance over all samples. b Sample to sample distance plot. A heatmap of distance matrix providing an overview of similarities and dissimilarities between samples. Clustering is based on the distances between the rows/columns of the distance matrix. c Principal component analysis plot. Samples are in the 2D plane spanned by their first two principal components. d Clustered heatmap of the log₂ converted normalized count matrix RNA-seq reads, top 50 genes, of nasopharyngeal samples. e Normalized Log₂ read counts of the genes encoding SARS-CoV-2 receptor and entry associated proteins. Enrichment analysis and comparison between deregulated genes and the genes of some selected processes in SARS-CoV-2 infected nasopharyngeal samples and SARS-CoV-2 infected lung biopsy samples using f GOBP module, g KEGG pathway, h Bioplanet pathway module. Selected significant terms are represented in heatmaps. Significance of enrichment in terms of the adjusted p-value (< 0.05) is represented in color-coded P-value scale for all heatmaps; Color towards red indicates higher significance and color towards yellow indicates less significance, while grey means non-significant. Normalized Log₂ converted read counts are considered as the expression values of the genes and represented in a color-coded scale; Color towards red indicating higher expression, while color towards green indicating little to no expression. Here, Up, down and DE denote Upregulated, Downregulated and Differentially expressed, respectively

Fig. 4

Comparison of the gene expression patterns in different SARS-CoV-2 infection models. a Venn-diagram showing the observed deregulated genes (with their respective control) in the different cell types. Enrichment analysis and comparison between deregulated genes in different SARS-CoV-2 infection models using b GOBP module, c Bioplanet pathway module, d KEGG pathway module. Selected significant terms are represented in heatmaps. Color scale/schemes are similar to Fig. 3

Fig. 5

Gene expression analysis using different SARS-CoV-2 infection models. a Variance plot, b Sample to sample distance plot, c Principal component analysis plot, d Clustered heatmap of the count matrix of the normalized RNA-seq reads of different SARS-CoV-2 infection samples using to 50 genes. e Gene expression heatmap showing global gene expression profiles in the individual infected samples of the various infection system. Heatmap is clustered based on Pearson’s distance with genes that vary across the sample, leaving out genes that do not vary significantly

Fig. 6

Heatmaps representing the sample level absolute expression of Integrin related genes. a across the different SARS-CoV-2 infection models, b in only nasopharyngeal samples and lung samples; Cytokine signaling related genes c across the different SARS-CoV-2 infection models, d in only nasopharyngeal samples and lung samples; and Inflammation related genes e across the different SARS-CoV-2 infection models, f in only nasopharyngeal samples and lung samples; g Expression profiles of genes encoding SARS-CoV-2 receptor and entry associated proteins. Normalized (DESeq2) Log₂ converted read counts are considered as the expression values of the genes and represented in a color-coded scale; Color towards red indicating higher expression, while color towards green indicating little to no expression

Fig. 7

Multifactorial differential gene expression analysis using differentially expressed COVID-19 lung and nasal data. a Variance plot, b Sample to sample distance plot, c Principal component analysis plot, d Clustered heatmap of the count matrix of the normalized RNA-seq reads (top 50 genes) of the SARS-CoV-2 infected nasopharyngeal and lung samples. e Common dispersion plot or the biological coefficient of variation plot. Here we are estimating the dispersion. The square root of the common dispersion gives the coefficient of variation of biological variation. Here the coefficient of biological variation is around 0.8. f MA plot. Plot log-fold change against log-counts per million, with DE genes are highlighted. The blue lines indicate twofold changes. Red and blue points indicate genes with P-value less than 0.05. g Expression profiles of genes encoding Integrins. Log₂ (fold change) values are considered as the expression values of the genes and represented in a color-coded scale; Color towards red indicating higher expression, while color towards green indicating little to no expression

Fig. 8

Enrichment analysis and comparison between deregulated genes and the genes of some selected processes in SARS-CoV-2 infected nasopharyngeal samples versus SARS-CoV-2 infected lung biopsy samples, using a GOBP module, b KEGG pathway, c Bioplanet pathway module, d Reactome pathway module. Selected significant terms are represented in heatmaps. Color schemes are similar to Fig. 3. For individual processes, blue means presence (significantly differentially expressed gene) while grey means absence (not significantly differentially expressed genes for this module for this experimental condition). Here, Up and down denote Upregulated and Downregulated, respectively

Fig. 9

Interactions between SARS-CoV-2 proteins and differentially expressed genes of host. a Venn diagram showing the commonly deregulated genes between deregulated genes in our nasopharyngeal samples and Gordon et al. reported viral protein-interacting high confidence host proteins. Network representing the interactions between genes in b. Deregulated genes in nasopharyngeal samples along with SARS-CoV-2 proteins and Gordon et al. described viral interacting host proteins, and c Differentially expressed Integrin related genes in lungs compared to the nasal samples along with SARS-CoV-2 proteins and Gordon et al. described viral interacting host proteins. Hexagon, ellipse, rounded rectangle represents viral proteins, process-related genes, and proteins that interact with viral proteins, respectively. Expression values of the genes and represented in a color-coded scale. Color towards red indicating higher expression, while color towards green indicating little to no expression

Fig. 10

Schematic representation of putative mechanisms of acute lung damages in COVID-19. Red arrow suggesting the increasing expression values, while the green arrow indicating the decreasing expression from the nasopharyngeal region to the lung