Introduction

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which causes coronavirus disease 2019 (COVID-19), can lead to a wide range of clinical outcomes, from asymptomatic infection to severe illness and death1. Severe cases are often associated with respiratory pathology2. As COVID-19 is a relatively new disease, long-term follow-up data are limited, though many studies have assessed its acute phase3. Notably, COVID-19 has been linked to frequent cardiac complications4. The pathophysiology of SARS-CoV-2 infection may involve several factors, including modifications to the immune system, persistence of viral fragments or proteins in various tissues, and vascular injury leading to blood clotting5. Moreover, SARS-CoV-2 infection has long-term cardiovascular effects even in individuals without prior cardiovascular conditions, indicating a strong affinity for the cardiovascular system6. The interplay between cardiomyocytes, cardiac fibroblasts, endothelial cells, and cytokines is a complex network that significantly influences heart function7. Cardiomyocytes act as the heart’s contractile units, orchestrating synchronized contractions for effective blood circulation8. Cardiac fibroblasts contribute to the structural framework by secreting extracellular matrix proteins, maintaining myocardial integrity9, and facilitating tissue healing and scar production when the heart is injured or stressed10. Cytokines are key mediators in this network, playing roles in immune responses, inflammation, and cellular communication within the cardiac microenvironment11,12. (Fig. 1) provides a visual representation of the specific cellular pathways that are altered as a result of the interaction between SARS-CoV-2 and cardiomyocytes.

Figure 1
figure 1

Putative signaling pathway of SARS-COV-2 interaction with heart tissue causing heart damage and heart failure.

Full size image

Under physiological conditions, cytokines contribute to tissue homeostasis13 and repair processes14. However, in the context of cardiac pathology, dysregulated cytokine signaling may lead to chronic inflammation15, adverse remodeling, and ultimately, heart failure16.

Therefore, the function of the implicated genes is not well understood in terms of their particular pathogenic significance, and multiomics analysis17 is necessary to identify the malfunction caused by the virus18. So, investigation of genetic changes and molecular pathways in various diseases, including neoplastic and non-neoplastic disorders19, has new avenues because of the recent and rapid advancement of transcriptome sequencing technology20 21. Since HF and COVID-19 genes linked to CVD are of increasing interest, a growing amount of RNA sequencing (RNA-seq) and microarray datasets22 have been uploaded to the Gene Expression Omnibus (GEO) database23, offering prospects for thorough meta-analysis using bioinformatics data mining. The cardiovascular disease etiology leads to a deep molecular level. This study sheds light on the nexus of COVID-19 inflammation and endothelial dysfunction. The study required a comparison of normal human transcriptome vs. COVID-19 infected cell transcriptome data from a publicly submitted sample. In the current investigation, DEG analysis has been carried out initially using 54 different samples of RNA-seq datasets in the GEO database to identify common upregulation and downregulation in HF normal samples and COVID-19 infected samples. Higher infection is connected to a pro-inflammatory immune phenotype known as “inflammation,” in which T cells develop a pro-inflammatory24 phenotype more innate than that of NK cells and are linked to the elevation of markers for both T cell senescence and exhaustion25. Additionally, elevated chemokine receptor expression, including CXCL2 on T cells, has been demonstrated in the iPSC-derived cardiomyocytes (iPSC-CMs) model26. The study was also followed by Kyoto Encyclopaedia of Genes and Genomes (KEGG) enrichment analysis27, Gene Ontology (GO)28, for the DEGs, which validate the role for TNF-A-NKB inappropriate pathway activation.

Materials and methods

Data collection

Gene expression datasets for this study were obtained from the National Center for Biotechnology Information’s (NCBI) Gene Expression Omnibus (GEO) database. We downloaded RNA-seq datasets to identify genes shared between COVID-19 and heart failure (HF).The search used keywords such as “SARS-CoV-2,” “cardiomyocytes,” and “cardiac.” The Fastq files from GEO accession IDs GSE15039229, GSE15187930, and GSE15675431 were downloaded. These datasets included RNA-seq expression data from various tissue types, including fibro fatty plaque, human pluripotent stem cell-derived cardiomyocytes (hPSC-derived CMs), and adult cardiomyocytes (CMs), as well as iPSC-derived cardiomyocytes, cardiac fibroblasts (CFs), and endothelial cells (ECs). A total of 54 samples from iPSC-CM model studies were analysed, comparing infected samples with mock samples. Table S1 provides a comprehensive overview of the RNA-seq data utilized in this study, encompassing both mock-infected and infected cell samples across various studies. The table includes detailed information on GEO IDs, sample titles, run accessions, and the specific study classifications. Please refer to Supplementary Table S1 for a detailed listing of the RNA-seq samples used.

Sequence alignment and transcriptome assembly

For the fifty-four downloaded RNA-seq runs, a standardized RNA-seq analytic process was used. In the first phase of quality control, RNA-seq runs were downloaded and analysed for sequencing errors using FastQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc).

Second, quality control involved trimming the transcript abundance from single samples using Fastp (version 0.23.0)32. Specifically, we used quality-controlled RNA-seq runs with parameters of phred scores of 15 and a minimum length requirement of 2033, considering the presence of infected and mock cells from various studies. We then utilized transcripts per million (TPM) to calculate the expression levels in each sample34. We compiled the quantitative data of all samples into an expression matrix and used a trimmed mean of M values (TMM) to normalize the expression levels between samples35. We determined the τ value for every gene to assess the level of tissue specificity in gene expression36. For the analysis, we used the R Studio version 3.4 (Bioconductor Package 3.18). The heatmap was created using the “GSVA” R package (https://bioconductor.org/packages/release/bioc/html/GSVA.html)37, and the bar plot showcasing the top 20 significantly enriched pathways was produced using the “enrichplot” package (https://bioconductor.org/packages/release/bioc/html/enrichplot.html)38. A schematic bioinformatics pipeline is provided as a supplementary file attachment (Bioinformatics workflow). This workflow is designed to conduct transcriptome research on raw RNA-seq data for both mock-infected and infected cell samples.

Results

Identification of shared differentially expressed genes

Differential expression between COVID-19 infected cardiomyocytes and mock-infected cells was determined. The number of significant differentially expressed transcript clusters is compared in Table S2. Analysis was performed using DESeq239 with an adjusted p-value < 0.05 in R Studio. Upon mapping genomic reads from infected hiPSCs, macrophages, cardiac fibroblasts, and endothelial cells across multiple heart areas, heatmaps of differentially expressed genes demonstrated clustering based on conditions (mock or infected) and transcriptomic profiles. Figure 2 shows a volcano plot where each gene is represented as a point in a two-dimensional space of statistical significance (P-value or FDR) vs. log fold change, allowing visualization of expression changes and their significance across the entire gene set. Figure 3 presents a heatmap of differentially expressed genes, providing a detailed visual representation of expression patterns. Hierarchical clustering was used to cluster data by both columns (samples) and rows (genes).

Figure 2
figure 2

Volcano plots of the differential gene expression data from GSE150392, GSE151879, and GSE156754. In the volcano plots, the red points show up-regulated genes (log2FC ≥ 0.5 and adjusted P-value < 0.05). The Venn map of the differentially expressed genes, however, shows the up-regulated genes as red dots. The number within each circle denotes the quantity of genes that exhibit differential expression across the various comparisons. We took into account only the designated genes. The genes specific to each condition are indicated by the non-overlapping numbers, whereas the overlapping number represents the genes that are mutually differentially expressed between the various comparisons.

Full size image
Figure 3
figure 3

Heat map of the top differentially expressed genes based on GSE150392, GSE151879, and GSE156754. The colour intensity suggests the higher to lower expression pathway and pathway enrichment analysis.

Full size image

Gene ontology of genes and genome pathway analyses

In all three investigations, 297 genes were found to be up-regulated in common by a differential expression analysis among infected cardiac cells. A venn diagram suggested that more genes were up-regulated in common among different groups. Together, GO enrichment analysis was conducted for these three gene categories, revealing genes were strongly linked to infected cells and have a significant role in the network regulated by NF-κB signaling. Biological processes frequently include the regulation of small GTPase-mediated signal transduction, as shown by nine GO annotations. It displays that one of the main mechanisms controlling inflammation is the nuclear factor-kappa B (NF-κB) pathway. The GO bar chart and ppi network shown in Fig. 4 represent the top functional category derived from elevated genes according to KEGG analysis.

Figure 4
figure 4

GO analyses of genes that are elevated and downregulated and analysis of the GO function enrichment (including BP, MF, and CC). B. Analysis of KEGG pathway enrichment. The gene ratio is displayed on the x-axis, while the GO category is displayed on the y-axis. The size of the bar represents the number of genes included in the relevant pathway, and the significance of enrichment progressively grows from blue to red.

Full size image

Involvement of inflammatory genes in the regulation of cardiac inflammation and remodeling in heart failure

The SARS-CoV-2 infection induced significant changes in gene expression within cardiomyocytes. The most notable expression change, as indicated by adjusted p-values, was observed in a gene encoding an immune cytokine known to be transcriptionally upregulated during SARS-CoV infection—CXCL2, which recruits immune cells to sites of inflammation. Increased levels of inflammatory cytokines may contribute to cardiac dysfunction. Additionally, NF-κB plays a key role in regulating inflammation, and its activation has been linked to cardiac inflammation and remodeling in heart failure. The NF-κB pathway plays a central role in the regulation of immune responses, inflammation, cell survival, and other cellular processes.

FOS and JUN were also significantly increased in all infected samples, as these genes are components of the AP-1 transcription factor complex. They are involved in regulating gene expression in response to various stimuli, including stress and inflammation. AP-1 activation has been associated with cardiac hypertrophy and heart failure. Analysis of the top common genes in each sample revealed a highly upregulated proinflammatory and inflammatory modulation pathway. Moreover, the higher expression of TLR2 leads to the recruitment of various downstream signaling molecules, such as IRAKs (Interleukin-1 receptor-associated kinases) and TRAF6 (TNF receptor-associated factor 6), which further propagate the signaling cascade. This process also involves the recruitment and activation of protein kinases, such as MAPKs (Mitogen-Activated Protein Kinases), leading to the TNF-NF-κB pathway activation of transcription factors and the induction of pro-inflammatory gene expression, as illustrated in Fig. 5.

Figure 5
figure 5

Differentially expression of TNF-NF-κB pathway genes from DEG analysis.

Full size image

Discussion

By integrating bulk RNA-sequencing data from various studies, we have created an innovative immune signature atlas of COVID-19infected cardiomyocytes and mock cells, comprising 54 samples. This comprehensive analysis allowed us to identify key mechanisms that could drive the severity of the SARS-CoV-2 infection and its impact on heart health.

Our study primarily highlights the significant role of TLR2 and downstream NF-κB signaling in the context of SARS-CoV-2 infection. TLR2 (Toll-like receptor 2) is known to be essential for recognizing pathogens and initiating immune responses. Previous studies have highlighted the protective function of NF-κB in the heart during stressful conditions, such as myocardial infarction or cardiac hypoxia40. NF-κB, a transcription factor, regulates genes involved in inflammation, immune response, and cell survival. However, our findings suggest that hyperactive TLR2/NF-κB signaling significantly contributes to the onset and progression of heart failure through several pathways, including oxidative stress, inflammation, myocardial fibrosis, endothelial damage, myocardial hypertrophy, and apoptosis41.

Oxidative stress is a critical factor in the pathogenesis of cardiovascular diseases. During SARS-CoV-2 infection, excessive production of reactive oxygen species (ROS) can lead to oxidative damage in cardiomyocytes. This oxidative stress can further activate NF-κB signaling, creating a vicious cycle that exacerbates inflammation and cell death. Inflammation, another major contributor, involves the activation of various cytokines and chemokines. Our analysis identified several inflammatory mediators, such as CXCL2, which recruits immune cells to sites of inflammation, and MCP-1 (CCL2), which attracts monocytes45,46. These inflammatory responses can lead to chronic inflammation, adverse cardiac remodeling, and ultimately heart failure.

Myocardial fibrosis, characterized by excessive deposition of extracellular matrix proteins by cardiac fibroblasts, can lead to stiffening of the heart tissue and impaired cardiac function. NF-κB activation in fibroblasts promotes the production of pro-fibrotic factors, contributing to fibrosis. Endothelial damage, another critical aspect, involves injury to the blood vessels, leading to impaired vascular function and increased risk of thrombosis. The activation of NF-κB in endothelial cells can exacerbate this damage by promoting inflammation and coagulation.

Myocardial hypertrophy, the enlargement of cardiomyocytes, is a compensatory response to increased workload or stress. However, sustained hypertrophy can lead to heart failure. Our data indicate that NF-κB signaling, along with other transcription factors like AP-1 (formed by FOS and JUN), plays a significant role in regulating genes associated with hypertrophy43. Apoptosis, or programmed cell death, is another critical factor. Excessive apoptosis of cardiomyocytes can lead to a reduction in functional heart tissue, contributing to heart failure.

In addition to the heart, TLR4/NF-κB signaling is crucial for the functioning of other organs, such as the kidneys, neurological system, digestive system, and lungs. This widespread involvement underscores the importance of NF-κB as a central regulator of immune and inflammatory responses. An intricate network of genes is involved in the TNF-NF-κB pathway, highlighting its role as a critical regulator of cellular functioning. MYC, initially discovered as a proto-oncogene, becomes a key actor capable of stimulating the NF-κB pathway and promoting cell division42. FOS and JUN work together to generate the AP-1 transcription factor, which responds to a variety of stimuli, including TNF, and plays a vital role in the precise control of target genes for NF-κB43.

The integral inhibitor of NF-κB, IκBα, encoded by NFKBIA, functions by confining NF-κB within the cytoplasm and obstructing its activation44. This regulatory mechanism is crucial for maintaining a balance between immune activation and inhibition. Interleukin-8 (IL-8) or CXCL8, facilitates the recruitment of immune cells to areas of inflammation, further amplifying the inflammatory response45. The transcription factor ATF3 shapes the cellular response to inflammation by regulating NF-κB activation. This modulation is critical for controlling the extent and duration of inflammatory responses.

MCP-1 (CCL2) is a chemokine that draws monocytes to areas of inflammation, playing a key role in the recruitment of immune cells during the inflammatory process46. Genes involved in cell proliferation and differentiation are regulated by EGR1, which is activated by TNF. EGR1 plays a pivotal role in the cellular response to injury and stress. TNFAIP3 (A20) appears as a negative regulator, providing feedback inhibition to reduce NF-κB activation and avoid over-inflammation47. This feedback loop is essential for preventing excessive inflammatory responses that could lead to tissue damage.

The phosphatase-encoding gene DUSP1 inhibits MAP kinases, which affects how cells react to TNF. This regulation is important for modulating the cellular responses to inflammatory stimuli48. SIRT1, a member of the sirtuin family, has anti-inflammatory properties through NF-κB activity modulation49. SIRT1’s role in reducing inflammation highlights its potential as a therapeutic target for inflammatory diseases. The transcriptional repressor BCL6 modifies NF-κB signaling in a subtle way, ensuring precise control overinflammation responses. Members of the AP-1 transcription factor family, JUNB and FOSB, actively participate in the control of NF-κB target genes. These transcription factors integrate various signaling pathways, contributing to the complexity of the inflammatory response.

SOCS3, a cytokine signaling suppressor, has inhibitory effects on pathways such as NF-κB, providing another layer of regulation to prevent excessive inflammation. NR4A1 (Nur77) regulates NF-κB activity, affecting apoptosis and cellular responses. The tumour suppressor PTEN interacts with NF-κB signaling to negatively control the PI3K/AKT pathway, highlighting its role in balancing cell survival and apoptosis50. The toll-like receptor TLR2 is essential for identifying pathogens and has the ability to trigger NF-κB in reaction to infections. Lastly, integrin ITGAM influences NF-κB activation and is involved in cell adhesion, further contributing to the regulation of immune responses51.

The intricate relationships between these genes demonstrate how they cooperate to subtly and crucially regulate the TNF-NF-κB pathway for inflammatory responses, immune responses, and cellular survival. Therefore, to prevent heart failure, cardiomyocytes, cardiac fibroblasts, and endothelial cells must maintain a delicate balance and coordinate their responses. Understanding the mechanisms behind both clinical disorders and physiological cardiac function requires an understanding of the TNF-NF-κB pathway and cytokines.

Furthermore, the limited number of existing datasets hinders the ability to draw definitive conclusions. To bridge these knowledge gaps, additional data is needed. Current therapeutic strategies targeting these pathways have shown limited success. However, further research employing single-cell proteomics and genomics to investigate cardiomyocyte infection by COVID-19 holds promise for developing more selective and targeted approaches, potentially leading to improved clinical outcomes. Future research focusing on novel therapeutic agents, personalized medicine, and combination therapies may offer new hope for heart failure patients. These follow-up studies would be crucial to solidifying our current understanding. Also, the existing datasets may be limited in terms of sample diversity, so there is a need to reduce gaps in dataset availability to draw more precise conclusions. While current therapies targeting these pathways have had limited success, more single-cell proteomic and single-cell genomics research on cardiomyocyte infection by COVID-19 into more selective and targeted approaches holds promise for improved clinical outcomes. Future research focusing on novel therapeutic agents, personalized medicine, and combination therapies may offer new hope for heart failure patients.

Conclusion

The complicated links between these genes show how they work together to carefully and precisely manage the TNF-NF-κ pathway. This pathway is in charge of immune system responses, inflammatory responses, and cell survival. Hence, in order to avert heart failure, it is imperative for cardiomyocytes, cardiac fibroblasts, and endothelial cells to uphold a fragile equilibrium and synchronize their reactions. To comprehend the workings of clinical diseases and physiological heart function, one must have a grasp of the cytokines in the TNF-NF-κB pathway. Moreover, the scarcity of available statistics impedes the capacity to reach conclusive findings. In order to fill these knowledge gaps, more data is required. Existing treatment approaches aimed at these pathways have demonstrated minimal efficacy. Additional research utilizing single-cell proteomics and genomes to study the COVID-19 infection of cardiomyocytes shows potential for establishing more precise and focused strategies, which could result in enhanced therapeutic results. Subsequent investigations that prioritize innovative therapeutic substances, tailored medical treatments, and the use of several therapies may provide renewed optimism for those suffering from heart failure. These subsequent investigations will be critical in consolidating our current understanding.