Deciphering the similarities and disparities of molecular mechanisms behind respiratory epithelium response to HCoV-229E and SARS-CoV-2 and drug repurposing, a systems biology approach

Abstract

Purpose

Identifying the molecular mechanisms behind SARS-CoV-2 disparities and similarities will help find new treatments. The present study determines networks’ shared and non-shared (specific) crucial elements in response to HCoV-229E and SARS-CoV-2 viruses to recommend candidate medications.

Methods

We retrieved the omics data on respiratory cells infected with HCoV-229E and SARS-CoV-2, constructed PPIN and GRN, and detected clusters and motifs. Using a drug-gene interaction network, we determined the similarities and disparities of mechanisms behind their host response and drug-repurposed.

Results

CXCL1, KLHL21, SMAD3, HIF1A, and STAT1 were the shared DEGs between both viruses’ protein-protein interaction network (PPIN) and gene regulatory network (GRN). The NPM1 was a specific critical node for HCoV-229E and was a Hub-Bottleneck shared between PPI and GRN in HCoV-229E. The HLA-F, ADCY5, TRIM14, RPF1, and FGA were the seed proteins in subnetworks of the SARS-CoV-2 PPI network, and HSPA1A and RPL26 proteins were the seed in subnetworks of the PPI network of HCOV-229E. TRIM14, STAT2, and HLA-F played the same role for SARS-CoV-2. Top enriched KEGG pathways included cell cycle and proteasome in HCoV-229E and RIG-I-like receptor, Chemokine, Cytokine-cytokine, NOD-like receptor, and TNF signaling pathways in SARS-CoV-2. We suggest some candidate medications for COVID-19 patient lungs, including Noscapine, Isoetharine mesylate, Cycloserine, Ethamsylate, Cetylpyridinium, Tretinoin, Ixazomib, Vorinostat, Venetoclax, Vorinostat, Ixazomib, Venetoclax, and epoetin alfa for further in-vitro and in-vivo investigations.

Conclusion

We suggested CXCL1, KLHL21, SMAD3, HIF1A, and STAT1, ADCY5, TRIM14, RPF1, and FGA, STAT2, and HLA-F as critical genes and Cetylpyridinium, Cycloserine, Noscapine, Ethamsylate, Epoetin alfa, Isoetharine mesylate, Ribavirin, and Tretinoin drugs to study further their importance in treating COVID-19 lung complications.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s40199-024-00507-0.

Introduction

Coronaviruses belong to the family of Coronaviridae, which causes several diseases in animals and humans [1]. Some Coronaviruses, including 229E, NL63, OC43, and HKU1, cause mild respiratory infections such as the common cold. However, coronaviruses like MERS and SARS cause more severe respiratory diseases [2]. HCoV-229E is one of the human coronavirus strains isolated in 1966 from students associated with common cold symptoms [3]. The virus damages the lower respiratory system in children and older adults. Severe lower respiratory system infections are reported in some immunocompromised patients [4]. The classification of coronavirus family viruses is shown in Supplementary Fig. 1.

Recently, a novel coronavirus called SARS-CoV-2 started in late 2019 and has caused a severe deadly pandemic worldwide [5]. This virus is of the same lineage as SARS-CoV but genetically distinct [6]. The viruses have common symptoms and manifestations, including cough, fever, sneezing, and shortness of breath [7]. Although it has appeared as a deadly virus, no effective drug has been available to treat it. According to the recent reports of the World Health Organization and the epidemiological point of view, SARS-CoV-2 is still a potentially serious threat since it can potentially obtain new and more dangerous mutations.

Moreover, the immunity duration to COVID-19 vaccines is short and somewhat similar to Flu vaccines. Besides, not everyone is vaccinated, especially in low-income countries. These facts show that the scientific world needs to work vigorously to manage the condition, and investigations to find possible effective treatments seem necessary [8].

The COVID-19 pandemic is caused by a highly pathogenic coronavirus in the human population [6]. The previous members of the HCoV family varied in their distribution in various regions. However, COVID-19 has appeared as a global pandemic [9], leading to a higher mortality rate than other coronaviruses.

However, more information is accessible on the molecular mechanisms of HCoV-229E, and there are still numerous gaps in our knowledge of the host response to HCoV-229E. Although both diseases (HCoV-229E and SARS-CoV-2) were involved in acute respiratory distress, the regulatory motif elements are unknown in HCoV-229E and SARS-CoV-2 host response GRN. Although both diseases rapidly became known for their pathogenesis leading to acute respiratory distress, a single-cell sequencing study showed that SARS-CoV-2 could harm most other organs (heart, kidneys, and brain) [10].

So far, no effective drug has been discovered, possibly due to its complex molecular mechanism. Unfortunately, many patients with severe forms of COVID-19 are still dying every day around the world, which highlights the importance of immediately discovering effective drugs against the infection. The research center is working to find an effective drug against this infection. Their results demonstrate that our understanding of viral mechanisms has remained incomplete in recent months.

This study compares the human response to a coronavirus with mild symptoms (HCoV-229E) with SARS-CoV-2 (in vitro) to decipher and predict the critical target behind these disparities in clinical manifestations from the common cold to a deadly cytokine storm agent and pave the way to repurposing and developing new therapeutics for the identified crucial targets.

Using a systems biology approach to investigate the molecular mechanisms behind the virus can help predict critical molecules mediating in response to this virus. It could be effective in drug repurposing against the identified targets. The Omics’ experimental and computational data may provide helpful information on the structure of the host response network and give us a better view of the interactions among the related biomolecules [11–21]. Various network analyses could be applied to enlighten different molecular aspects of infections. Mercè Llabrés et al., in 2020, aligned SARS-CoV and SARS-CoV-2 viruses-host protein-protein interaction networks and identified shared proteins in these virus-host networks and shared human proteins with functions related to viral infection and shared human proteins with a functional similarity that interact with SARS-CoV and SARS-CoV-2 proteins [22]. Host response to pathogenic agents could be regulated by gene regulatory networks controlled by transcription factors, microRNAs, and other regulatory elements. Applying a systems biology approach could also help investigate similarities and differences between various molecular mechanisms of host responses to diverse pathogens [23]. Scott A. Ochsner et al. in 2020 have identified transcriptional regulations of coronavirus-infected human cells and suggested that SARS-CoV-2 probably repressed the E2F family in airway epithelium cells using a systems biology approach [24]. In 2021, Takayuki Amemiya et al. reported the results of omics networks in drug predicting for pathogenic mosquito-borne viruses. Their study proposed 146 proteins and 77 drug candidates [25]. David E. Gordon et al., in 2021, analyzed the targets of SARS-CoV-2 for drug repurposing using a (host-virus) protein interaction map. They recommended some drugs targeting the viral enzymes that could be used to treat COVID-19 [26]. The advances in omics data at genomics, transcriptomics, and metabolomics biological levels help provide information such as drug targets for repurposing new drugs [27].

This study attempted to identify the molecular mechanisms involved in some lung cells in SARS-CoV-2 and suggested repurposed drugs based on it. This study used a system biology approach and performed PPI and GRN network analysis to identify key host factors in response to HCoV-229E and SARS-CoV-2 viruses, separately. We determined the shared crucial elements of the networks in response to both viruses, identified the specific molecules for each infection, and used the essential target molecules to predict and repurpose drugs to treat the recent pandemic infection utilizing the construction and analysis of a drug-gene interaction network. Finally, we confirmed and validated our crucial target molecules and repurposed drug candidates using previous experimental studies to show that other recommended molecules and drugs are worth investigating experimentally for their possible role in treating the infection.

Material and methods

Study design

This study aimed to investigate the PPIN and GRN in response to the A549 cell line to HCoV-229E and SARS-CoV-2 and compare these two viruses. Figure 1 shows a workflow of methods used in this study.

Fig. 1.

The workflow represents the study design

Data collection

This study retrieved gene expression data related to HCoV-229E from A549 (lung epithelial) and mock cells infected with HCoV-229E. A549 and mock cells were cultured in DMEM supplemented with 10% fetal calf serum, two mM L-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin. Infections of A549 were performed at 33°C using the indicated multiplicities of infection (MOI). The extraction of RNA was performed on four samples (two infected and two uninfected), and finally, transcriptomic was determined using an Agilent 60k microarray [28]. For SARS-CoV-2, the A549 and normal human bronchial epithelial (NHBE) cells were maintained in DMEM with 10% Fetal Bovine Serum, 37 °C, and 5% CO2. Cell infections were performed for 24 h with MOI of 0.2 SARS-CoV-2. Total RNA was extracted from infected and mock cells. TruSeq RNA Library protocol (Illumina) was used to prepare RNA-seq libraries [29].

Data processing

The differentially expressed genes were checked between the mock cells and infected samples. For HCoV-229E, we used the results of the DEGs normalized and published by Poppe M et al. in 2017 for analysis [28]. In this study, the researchers related to the HCoV-229E dataset had previously filtered and analyzed the data, and we used the same resulting DEGs with the same criteria. In the case of SARS-CoV-2, we screened the resulting data based on p-value < 0.05 and |log fold-change > 0.5| to generate sufficient results for constructing networks and comparison.

Protein-protein interaction networks

Construction of protein-protein interaction network

Human Integrated Protein-Protein Interaction rEference or HIPPIE v2.0 (http://cbdm.uni-mainz. de /hippie/) is a web resource to evaluate and generate PPI networks. HIPPIE v2.0 was used from information experimental with a confidence score > 0.7 for the construction network. HIPPIE v2.2 has certainty scores on edges calculated based on their scheme [30]. STRING is a biological web resource for drawing and reviewing protein-protein interactions (https://string-db.org) [31]. The interactions with a confidence score > 0.7 are identified to build interactions. All scores rank from 0 to 1, with 1 being the most noteworthy conceivable certainty. A score of 0.5 would show that, generally, each moment interaction may be wrong (i.e., a false positive). STRING presented thresholds of 0.15 (low confidence), 0.40 (medium confidence), 0.70 (high confidence) or 0.90 (highest confidence) [32]. BisoGenet was used to generate and visualize the PPI networks between genes from the human Protein Reference (HPRD) database in constructing the network maps [33], and we created two protein-protein networks for HCoV-229E and SARS-CoV-2 viruses. Cytoscape is open-source software for visualizing, analyzing, and interacting with bio-molecular interaction networks [34]. Finally, the networks from three sources were fed into Cytoscape software to merge the networks for every virus.

Topological analysis of HCoV-229E and SARS-CoV-2-related networks

The merged protein-protein interaction networks were analyzed using the plugin Network Analyzer. This plugin was used to determine nodes with a maximum degree (hub) and betweenness-centrality (bottleneck). The node degree of a node n is the number of edges linked to n [35]. Betweenness-centrality is the number of paths crossing each node in the network [36]. Next, the 10% of top nodes with a maximum degree and betweenness-centrality scores were selected in the PPI networks of both viruses. The hub genes are nodes with high connectivity in the module. Bottleneck genes are defined as nodes that have many shortest paths. Bottlenecks are essential connector proteins with critical functional properties [37, 38].

MCODE sub-networks

HCoV-229E and SARS-CoV-2 PPI networks were analyzed for clustering separately using the MCODE app in Cytoscape software. This app finds protein complexes and sub-networks highly interconnected in PPI networks. The algorithms used in MCODE select nodes that are more interconnected with neighbors. Finally, the cluster quality improves with haircut and fluff filters [39]. MCODE algorithms were applied to choose highly interlinked regions in the HCoV-229E and SARS-CoV-2 networks with the default setting: Degree Cut off = 2, Node Score Cutoff = 0.5, K-Core = 2, and Max-Depth = 100. Clusters with Scores > 3 and numbers of nodes > 5 are selected as modules and more analysis (enrichment analysis and drug-gene interaction network).

Gene regulatory networks

Construction of TF-miRNA-gene regulatory networks for HCoV-229E and SARS-CoV-2 genes

The DEGs obtained from studies related to HCoV-229E and SARS-CoV-2 are used to plot the regulatory network, separately. The relationships TF-gene, TF-miR, miR-gene, and miR-TF were obtained using databases to construct GRN.

Identification of miRNA repression of genes and transcription factors (miR-gene and miR-TF)

MiRNAs repression of genes and transcription factors obtained from miRecords (http://c1.accurascience.com/miRecords/) and miRTarBase (http://miRTarBase.cuhk.edu.cn/). miRecords is a source of animal miRNA-target interactions [40]. The miRTarBase is a website from annotated and experimentally validated miRNA-target interaction. This web includes > 13.404 miRNA-target interactions [41].

Identification of transcription factors regulating gene/miRNA expression (TF-gene, TF-miR)

TFs regulating genes were obtained from TRANSFAC (http://www.gene-regulation.com/pub/databases.html) and TRRUST (www.grnpedia.org/trrust/) databases. [42, 43]. TransmiR v2.0 (http://www.cuilab.cn/transmir) is a curated database that predicts transcription factors that regulate miRNAs[44].

GRN construction

The four types of regulation (TF-gene, TF-miR, miR-gene, and miR-TF) merged in Cytoscape software for constructing GRN networks. The FANMOD software detected size-3 network motif types in the network [45]. Z-score and p-value were calculated for all the motif types. Finally, 3-node motif types with Z-score > 2.0 and p-value < 0.05 were detected as significant motifs. The significant motifs of every virus merged and built motif-specific sub-networks. separately. The visualization of motif-specific sub-networks is done by Cytoscape software.

Enrichment analysis

The hub-bottleneck genes in PPIN and GRN beside MCODE cluster nodes of both viruses were selected for analyzing the gene ontology (Biological process, Molecular Function, and Cellular Component) and KEGG pathway by DAVID databases. DAVID (https://david.ncifcrf.gov/) is a web for functional annotation, enrichment of gene lists, and visualization with intuitive graphical summaries. This database covers 55464 organisms, including humans, rats, mice, Drosophila, etc. DAVID can drill down the GO hierarchy for any subset of genes with a standard classification. Additionally, DAVID highlights pathway members in the biochemical pathways provided by KEGG [46]. For enrichment analysis by this web server, list genes input in the database and homo sapiens organism selected for analysis. Finally, gene ontology and pathways with significant p-values < 0.05 were chosen for discussion.

Analysis of drug-gene interaction network

We selected shared proteins between both viruses (shared nodes of hub-bottleneck in PPIN and GRN, nodes and seeds in sub-networks HCoV-229E and SARS-CoV-2) and specific proteins of each virus to repurpose the new drug. We used the DGIdb database to repurpose new drug candidates for the critical proteins. We visualized the drug-protein network between crucial proteins and new repurposed drugs using Cytoscape 3.5.1 software. Network analyzer done on gene-drug interaction network identified medications with a high degree. Finally, experimental reports and literature reviews were discussed to verify some repurposed drugs.

Statistical Analysis

The software and web servers used in this study include GEO, Venn diagram, HIPPIE v2.0, STRING, BisoGenet, miRecords, miRTarBase, TRANSFAC, TRRUST, TransmiR v2.0, DAVID, and DGIdb databases, Cytosacape (Network Analyzer, MCODE app), and FANMOD software. The data with a p-value < 0.05 is used as significant in these databases and software.

Results

Data screening

The DEGs were identified based on adjusted p-value (< 0.05) and log₂ (fold change) > 0.5 and < -0.5 obtained 152 DEGs of gene expression data HCoV-229E including 99 up-expressed genes and 53 down-expressed genes and in SARS-CoV-2 (COVID-19) has 452 DEGs of gene expression data including 351 up-expressed genes and 101 down-expressed genes. (Table S1).

PPI network analysis in HCoV-229E and SARS-CoV-2

Using Cytoscape software, we merged and visualized the STRING, HIPPIE, and Bisogenet networks for every virus. The HCoV-229E included 156 nodes and 321 edges, and SARS-CoV-2 had 457 nodes and 1197 edges. Network Analyzer plugin showed hub-bottleneck genes in HCoV-229E and SARS-CoV-2 (Table S2). Figure 2A, B show a 10% top Hub-Bottleneck of HCoV-229E and SARS-CoV-2 PPI networks visualized based on the degree in Cytoscape software. A list of the top 10 Hub-Bottlenecks is presented in Table 1.

Fig. 2.

PPI network. A The 10% genes high degree and betweenness-centrality of the HCoV-229E network. B The 10% genes have a high degree of betweenness-centrality in the SARS-CoV-2 network. The visualization is based on a degree in Network Analyzer tools

Table 1.

Top 10 hub-bottleneck proteins in HCoV-229E and SARS-CoV-2 PPI networks obtained by Cytoscape software

Virus	Name genes	Degree	Betweenness centrality
HCOV-229E	UBC	31	0.365912
	NPM1	29	0.227156
	MDM2	28	0.160978
	EEF1A1	18	0.107044
	YWHAG	17	0.086993
	MAPK6	16	0.088947
	NCL	14	0.045147
	RPS15	13	0.102386
	SMN1	10	0.048684
	PSMA5	10	0.045895
SARS-COV2	STAT1	45	0.072245
	HERC5	32	0.032238
	UBE2L6	31	0.026428
	SAMHD1	30	0.054073
	IFI16	23	0.151206
	SP100	22	0.042925
	CXCL1	20	0.08026
	CEP290	19	0.125452
	UBE2V2	17	0.036875
	CCL5	17	0.038962

The clustering of SARS-CoV-2 and HCoV-229E PPI networks

The MCODE algorithm with parameters (Degree Cut off = 2, Node Score Cutoff = 0.5, K-Core = 2, and Max-Depth = 100) and Score > 3 and numbers of nodes > 5 identified 2 and 5 sub-networks in the HCoV-229E and SARS-CoV-2, respectively (Table 2). The HLA-F, ADCY5, TRIM14, RPF1, and FGA were the seed proteins in subnetworks NO. 1, 2, 3, 4, and 5 of the response to the SARS-CoV-2 PPI network. The HSPA1A and RPL26 proteins were the seed in subnetworks NO.1 and 3 of the PPI network of HCoV-229E (Table 2). Figure 3A, B depicted sub-networks related to HCoV-229E and SARS-CoV-2 PPI networks. The Supplementary Table S3 showed proteins related to subnetworks.

Table 2.

The HCoV-229E and SARS-CoV-2 PPI subnetworks with Score > 3 and nodes > 5 were obtained by the MCODE app

Virus	Subnetworks	score	Density nodes	Number of interactions	Seed node
HCoV-229E	1	5	5	13	HSPA1A
	2	3.33	7	13	–
	3	3	7	11	RPL26
SARS-CoV-2	1	23	23	263	HLA-F
	2	11	11	58	ADCY5
	3	8.133	16	65	TRIM14
	4	6.333	7	22	RPF1
	5	6	6	16	FGA

Fig. 3.

SARS-CoV2 and HCoV-229E sub-networks. Subnetworks NO. 1, 2 and 3 are related to HCoV-229E. Subnetworks NO. 1, 2, 3, 4, and 5 are related to SARS-CoV-2. Shapes and different coloring of nodes represent cluster-related proteins (triangle and blue) and seeds proteins (circle and yellow)

Analysis of GRN

This study retrieved miRNAs and their experimentally validated targets from miRTarBase and miRecords. 153 and 452 genes were selected as differentially expressed from HCoV-229E and SARS-CoV-2, respectively. The 1438 and 1503 miRNAs targeted these genes through 5274 and 7928 interactions in HCoV-229E and SARS-CoV-2. After retrieving TFs from TRANSFAC and TRRUST databases, miRNA-TF interactions were also considered. In the HCoV-229E virus, 1523 miRNA targeted 211 TFs with 2388 interactions. The results of SARS-CoV-2 revealed 1,548 miRNA target 281 TFs with 9,754 interactions. The TFs regulating miRNAs were also retrieved from the TransmiR database. The results in HCoV-229E showed that 431 TFs regulated 347 miRNAs with 5148 interactions. In SARS-CoV-2, 433 TFs regulate 347 miRNAs with 5152 interactions. 210 and 297 TF retrieved TRANSFAC and TRRUST databases for 153 and 452 genes of HCoV-229E and SARS-CoV-2, respectively. Finally, these regulatory networks were used to extract motifs. These results are shown in Supplementary Tables S4 and S5.

Motif-specific sub-network analysis

The TF-miRNA-gene network was constructed by combining the miRNA-TF/gene and TF-miRNA/gene interactions. The gene regulatory network (GRN) contained 2539 and 2952 nodes in HCoV-229E and SARS-CoV-2, respectively. In the next step, we selected the motifs with p-value < 0.05 and Z-score > 2 in both viruses’ GRN networks. In HCoV-229E, No.78, 14, and 164 motifs were selected and merged into a sub-network that includes four genes (SMAD3, NPM1, ATF4, and CBX5), 215 TFs, and 1525 miRNAs. The FANMOD software identified 78 significant motifs in the SARS-CoV-2 GRN network. This motif includes three genes (HIF1A, STAT2, and STAT1), 53 TF, and 122 miRNAs. The Cytoscape software 3.5.1 was used for regulatory visualizing of sub-networks. The types of motifs are shown in Fig. 4 for HCoV-229E and SARS-CoV-2 gene networks.

Fig. 4.

Regulatory motifs consist of miRNAs, TFs, and target genes detected in HCoV-229E (A) and SARS-CoV2 (B) Gene networks with their Z-score and p-value. Three types of relationships involved in these motifs included TF-gene (TFs regulating gene expression), miRNA-TF (miRNA represses TF expression), and TF-miRNA (TF regulates miRNA expression)

Gene ontology and pathway analysis

Gene ontology categories analysis was done to identify the biological process, molecular function, and cellular components involved in the lung response to HCoV-229E and SARS-CoV-2 infection. The 10% Hub-Bottlenecks of the HCoV-229E and SARS-CoV-2 PPI network were submitted in DAVID databases, separately. Finally, the statically significant gene ontology terms were extracted. The top biological processes for HCoV-229E included regulating cellular protein, metabolic process, and cellular component assembly. Protein, enzyme, and nucleic binding were the highly scored molecular function terms related to the 10% Hub-Bottleneck of the HCoV-229E PPI network. The top cellular compartment terms were cytosol, organelles, and nucleus. The results are shown in Table 3.

Table 3.

Top five gene ontology of 10% Hub-Bottlneck proteins in HCoV-229E PPI network with default parameters obtained by DAVID database

Terms	GO ID	Terms	p-value	Proteins
Biological process	GO:0032269	negative regulation of the cellular protein metabolic process	2.53E-07	PSMA5, NPM1, SMAD3, ANXA2, …
	GO:0051248	negative regulation of the protein metabolic process	4.07E-07	PSMA5, NPM1, SMAD3, ANXA2, …
	GO:0032268	regulation of cellular protein metabolic process	4.40E-07	NPM1, SMAD3, ANXA2, DERL1, …
	GO:0030163	protein catabolic process	8.12E-07	PSMA5, SMAD3, ANXA2, UBC, …
	GO:0051246	regulation of protein metabolic process	9.41E-07	NPM1, SMAD3, ANXA2, DERL1, …
Molecular function	GO:0019899	enzyme binding	2.88E-07	EEF1A1, NPM1, SMAD3, ANXA2, …
	GO:0003723	RNA binding	1.76E-05	RPS15, EEF1A1, NPM1, ANXA2, …
	GO:0044822	poly(A) RNA binding	1.40E-04	RPS15, EEF1A1, NPM1, ANXA2, …
	GO:0019901	protein kinase binding	1.84E-04	EEF1A1, NPM1, SMAD3, MAPK6, …
	GO:0031625	ubiquitin protein ligase binding	2.07E-04	SMAD3, MDM2, DERL1, SQSTM1, …
Cellular component	GO:0005829	cytosol	5.15E-09	NPM1, SMAD3, ANXA2, HSP90B1, …
	GO:0032991	macromolecular complex	5.67E-08	NPM1, SMAD3, ANXA2, DERL1, …
	GO:0044444	cytoplasmic part	4.44E-07	NPM1, SMAD3, ANXA2, DERL1, …
	GO:0044446	intracellular organelle part	1.39E-05	NPM1, SMAD3, ANXA2, DERL1, …
	GO:0005634	nucleus	1.54E-05	NPM1, SMAD3, ANXA2, HSP90B1, …

In the SARS-CoV-2 PPI network, cellular response to type I interferon and defense response to the virus are the top five related biological process terms. The highly scored five molecular function terms include protein binding, nucleic acid binding, adenylyltransferase activity, and 2'-5'-oligoadenylate synthetase activity. Most SARS-CoV-2 PPI network proteins are presented in the cytoplasm and nuclear. Table 4 shows the top five gene ontologies related to the SARS-CoV-2 PPI network.

Table 4.

Top five gene ontology of 10% hub-bottleneck proteins in SARS-CoV-2 network with default parameters obtained by DAVID database

Terms	GO ID	Terms	p-value	Proteins
Biological process	GO:0071357	cellular response to type I interferon	3.74E-41	IFITM3, IFITM1, SP100, IFITM2, IFI6, …
	GO:0060337	type I interferon signaling pathway	3.74E-41	IFITM3, IFITM1, SP100, IFITM2, IFI6, …
	GO:0034340	response to type I interferon	1.67E-40	IFITM3, IFITM1, SP100, IFITM2, IFI6, …
	GO:0051607	defense response to virus	2.27E-33	IFITM3, IFITM1, IFITM2, ADAR, GBP1, …
	GO:0009615	response to virus	6.37E-33	IFITM3, IFITM1, IFITM2, ADAR, …
Molecular function	GO:0003725	double-stranded RNA binding	5.26E-09	IFIH1, OAS1, DDX58, OAS2, OAS3, …
	GO:0005515	protein binding	1.85E-08	IFITM3, SNAP25, IFITM1, CXCL8, …
	GO:0001730	2'-5'-oligoadenylate synthetase activity	2.17E-07	OAS1, OAS2, OAS3, OASL
	GO:0003723	RNA binding	5.13E-07	SSB, RPL21, DDX58, EIF2AK2, …
	GO:0070566	adenylyltransferase activity	1.04E-04	OAS1, OAS2, OAS3, OASL
Cellular component	GO:0005829	cytosol	8.57E-08	SNAP25, UBE2L6, IFI35, SMC3, …
	GO:0005737	cytoplasm	1.12E-04	IFITM3, SNAP25, UBE2L6, ADAR, …
	GO:0005634	nucleus	2.44E-04	SP100, BEX2, UBE2L6, ADAR, …
	GO:0044428	nuclear part	2.57E-04	SP100, UBE2L6, ADAR, SAMHD1, …
	GO:0005622	intracellular	2.94E-04	IFITM3, SNAP25, CXCL8, UBE2L6, …

The significant pathways related to 10% hub-bottleneck proteins HCoV-229E and SARS-CoV-2 PPI networks were submitted in the DAVID database, separately. Table 5 shows the top five pathways related to these proteins. In the HCoV-229E PPI network, the Cell Cycle, Protein processing in the endoplasmic reticulum, and RNA transport are significant pathways. The RIG-I-like receptor signaling pathway is a pathway involved in SARS-CoV-2 infection. The gene ontology data for HCoV-229E and SARS-CoV-2 PPI networks are shown in Supplementary Table S6 and S7.

Table 5.

shows the significant pathways related to 10% hub-bottleneck proteins HCoV-229E and SARS-CoV-2 PPI network with default parameters obtained by the DAVID database

Number of sub-network	ID	Terms	p-value	Proteins
HCoV-229E	hsa04110	Cell cycle	0.02546	SMAD3, MDM2, YWHAG
	hsa04141	Protein processing in the endoplasmic reticulum	0.04499	DERL1, HSP90B1, HSPA1A
	hsa03013	RNA transport	0.046445	EEF1A1, STRAP, SMN1
SARS-CoV-2	hsa04622	RIG-I-like receptor signaling pathway	2.79E-05	IFIH1, CXCL8, DDX58, IRF7, TRIM25, …

We performed KEGG pathway analyses of clustered genes using the DAVID database. The KEGG pathways are shown in Table 6. In HCoV-229E, KEGG pathway terms show that most genes participate in the proteasome. The top five KEGG pathways resulting from the DAVID tool analysis included the Chemokine signaling pathway, Cytokine-cytokine receptor interaction, NOD-like receptor signaling pathway, TNF signaling pathway, and Viral carcinogenesis in SARS-CoV-2.

Table 6.

The significant pathways related to cluster genes HCoV-229E and SARS-CoV-2 PPI network with default parameters obtained by the DAVID database

Number of sub-network	ID	Terms	p-value	Proteins
HCoV-229E	hsa03050	Proteasome	0.043943	PSMA5, PSMB3
SARS-CoV-2	hsa04062	Chemokine signaling pathway	2.44E-06	CXCL8, CCL20, CCL5, STAT2, CXCL1, CXCL3, CXCL2, CXCL5, ADCY5
	hsa04060	Cytokine-cytokine receptor interaction	1.50E-04	IL6, CXCL8, CCL20, CCL5, CXCL1, CXCL3, CXCL2, CXCL5
	hsa04621	NOD-like receptor signaling pathway	1.52E-04	IL6, CXCL8, CCL5, CXCL1, CXCL2
	hsa04668	TNF signaling pathway	1.61E-04	IL6, CCL20, CCL5, CXCL1, CXCL3, CXCL2
	hsa05203	Viral carcinogenesis	0.003079	C3, SP100, IRF7, HLA-B, HLA-F, IRF9

Gene ontology and pathway enrichment were performed separately for DEGs related to both viruses in regulatory motifs. Most of the DEGs are transcription factors, so target genes of these TFs were extracted from the TF-gene Table and finally were enriched.

Based on biological process enrichment analysis, the DEGs of HCoV-229E played roles in localization and transportation. The protein binding and RNA binding activities are significant terms for the molecular function of DEGs in HCoV-229E. So, most of the DEGs of the HCoV-229E network are presented in the cytosol, cytoplasm, and vesicle. In SARS-CoV-2, the enrichment analysis reveals the interferon signaling pathway, defense response to viruses, and response to cytokine as the top significant biological processes. Most DEGs SARS-CoV-2 involve transition metal ion binding, nucleic acids binding, and hormone receptor binding. The intracellular and nucleus are significant terms of cellular components. (Table 7).

Table 7.

Top five gene ontologies related to DEGs regulatory motifs of HCoV-229E and SARS-CoV-2 using the DAVID database, separately

GO ID	Terms	Proteins	p-value
		HCoV-229E
	Biological process
GO:0046907	intracellular transport	KHDRBS1, INSIG1, DERL1, ATP5I, ATP5G3,…	1.03E-08
GO:0051641	cellular localization	KHDRBS1, INSIG1, DERL1, ATP5I, ATP5G3,…	1.27E-07
GO:0033554	cellular response to stress	GSTP1, INSIG1, DERL1, CRIP1, PTPRF, HSP90B1, …	1.35E-07
GO:0051179	localization	MIDN, CXCL1, ATP5G3, SPINT2, TCIRG1,…	2.30E-07
GO:0051649	establishment of localization in cell	KHDRBS1, INSIG1, DERL1, ATP5I, ATP5G3,…
	Molecular function
GO:0019899	enzyme binding	MIDN, DCUN1D5, GSTP1, DERL1, TCIRG1, …	4.40E-05
GO:0003723	RNA binding	KHDRBS1, ZC3H4, HSP90B1, RPS15, TUBA1B, …	8.37E-05
GO:0044822	poly(A) RNA binding	KHDRBS1, TSFM, NPM1, ANXA2, ZC3H4, NONO,… …	9.81E-05
GO:0005515	protein binding	MIDN, RPS15, LGALS1, FTH1, CFL1, PIM3, …	1.01E-04
GO:0045296	cadherin binding	TMEM2, KRT18, CTTN, ANXA2, FXYD5, HSPA1B, …	3.46E-04
	Cellular component
GO:0005829	cytosol	MIDN, BBC3, RPS15, TRMT112, POTEKP, AP1G1, …	6.69E-11
GO:0005737	cytoplasm	MIDN, SPINT2, RPS15, POTEKP, LGALS1, FTH1, …	1.64E-07
GO:0044446	intracellular organelle part	MIDN, ATP5G3, TCIRG1, BBC3, HERPUD1, RPS15, …	2.71E-07
GO:1,903,561	extracellular vesicle	FKBP2, GSTP1, BOLA2B, PTPRF, HSP90B1, RAB21, …	5.35E-07
GO:0043230	extracellular organelle	FKBP2, GSTP1, BOLA2B, PTPRF, HSP90B1,RAB21, …	5.40E-07
		SARS-CoV-2
	Biological process
GO:0060337	type I interferon signaling pathway	STAT1, STAT2, IRF7, ADAR, IFIT1, OASL	1.89E-07
GO:0071357	cellular response to type I interferon	STAT1, STAT2, IRF7, ADAR, IFIT1, OASL	1.89E-07
GO:0034340	response to type I interferon	STAT1, STAT2, IRF7, ADAR, IFIT1, OASL	2.47E-07
GO:0071345	cellular response to cytokine stimulus	PNPT1, STAT1, FLRT3, STAT2, IRF7, ADAR, IFIT1, HIF1A, OASL	2.41E-05
GO:0019221	cytokine-mediated signaling pathway	STAT1, FLRT3, STAT2, IRF7, ADAR, IFIT1, HIF1A, OASL	4.31E-05
	Molecular function
GO:0046914	transition metal ion binding	ZCCHC7, MORC3, ZNF92, USP16, ZNF184, ERI2, HLTF, PHF11, PLOD2, TRIM6-TRIM34	0.001583
GO:0008270	zinc ion binding	ZCCHC7, MORC3, ZNF92, USP16, ZNF184, ERI2, HLTF, PHF11, TRIM6-TRIM34	0.001852
GO:0035257	nuclear hormone receptor binding	STAT1, NCOA7, HIF1A, OASL	0.002194
GO:0003676	nucleic acid binding	ZCCHC7, MORC3, PNPT1, ZNF184, STAT1, STAT2, ADAR, IFIT1, HIF1A, PARP12, SYNE1, OASL, ZNF92, ZSCAN30, ERI2, HLTF, IRF7	0.003139
GO:0051427	hormone receptor binding	STAT1, NCOA7, HIF1A, OASL	0.003363
	Cellular component
GO:0005634	nucleus	ZCCHC7, MORC3, USP16, ZNF184, STAT1, STAT2,	0.021327
GO:0005622	intracellular	ZCCHC7, USP16, ADAR, PLOD2, IFIT1, HIF1A,	0.047976

To gain mechanisms for these genes, we enriched the KEGG pathways for DEGs obtained from regulatory motifs of both viruses separately. Protein processing in the endoplasmic reticulum and Cytosolic DNA-sensing pathway were significantly enriched for the related DEGs of regulatory motifs in HCoV-229E and SARS-CoV-2, respectively. These results are described in Table 8. All gene ontology and pathway enrichment results are reported in Supplementary Table S8 and Table S9.

Table 8.

KEGG pathways' significant DEGs were obtained from regulatory motifs of both viruses using the DAVID database

ID	Terms	Proteins	p-value
	HCoV-229E
hsa04141	Protein processing in the endoplasmic reticulum	HSPH1, MAN1B1, HSP90B1	0.019284
	SARS-CoV2
hsa04623	Cytosolic DNA-sensing pathway	IL6, IRF7, ADAR	0.029443

Comparing PPIN and GRN of both viruses

The comparison of PPIN between HCoV-229E and SARS-CoV-2 identified CXCL1 and KLHL21 as common DEGs. The intersection motifs between HCoV-229E and SARS-CoV-2 have revealed 167 shared nodes, which include (122 miRNAs and three genes (HCoV-229E (SMAD3) and SARS-CoV-2 (HIF1A, STAT1)) and 42 TFs). The NPM1 was identified as a specific critical node for HCoV-229E and was a hub&bottleneck shared between the PPI and GRN networks of HCoV-229E. TRIM14, STAT2, and HLA-F played the same role for SARS-CoV-2. The biological process and functional pathway comparison between HCoV-229E and SARS-CoV-2 represented no similarities.

Repurposing new drugs and drug-protein interaction analysis

We evaluated protein-drug interactions for shared proteins between both viruses and the specific proteins for each virus in PPIN and GRN (Supplementary Table S10) to repurpose new FDA-approved drugs available in the drug-gene interaction databank (DGIdb) (www.dgidb.org). These results are represented in Table 9. We visualized protein-drug interactions using Cytoscape. Among the five shared proteins (CXCL1, KLHL21, SMAD3, STAT1, HIF1A), three (SMAD3, STAT1, HIF1A) had drug interactions. Eight and eighteen proteins encompass drug interactions with 25 and 90 specific proteins related to HCoV-229E and SARS-CoV-2. Noscapine, Isoetharine mesylate, Cycloserine, Ethamsylate, Cetylpyridinium, Tretinoin, Ixazomib, Vorinostat, Venetoclax, Vorinostat, Ixazomib, Venetoclax, Ribavirin, and epoetin alfa medications were identified as candidate drug in treatment COVID-19 patients. The Network Analyzer app in Cytoscape shows that Ribavirin, cisplatin, and tretinoin drugs have the most degrees. Ribavirin is a meditation that interacts with HLA-B, IL-6, OASL, and CXCL8. The MDM2, IL-6, and STAT1 genes are vital molecules that interact with cisplatin. Tretinoin is another drug that interacts with the NPM1, HIF1A, and CXCL8 genes. The interaction of these genes and medications is shown in Fig. 5.

Table 9.

The repurposed FDA-approved medications for identified SARS-Cov-2 and HCoV-229E crucial proteins obtained by DGIbd and Cytoscape software

Proteins	Number of proteins interacting with drug	Repurposed drugs
NPM1	10	Crizotinib, Midostaurin, Daunorubicin, Venetoclax, Alectinib, Ceritinib, Tretinoin, Vorinostat, Ixazomib, Lorlatinib
MDM2	4	Nivolumab, Cisplatin, Pembrolizumab, Ipilimumab
MAPK6	1	Sorafenib
YWHAG	2	Daunorubicin Hydrochloride, Doxorubicin Hydrochloride
HSPA1A	1	Carbamazepine
SMN1	12	Nabumetone, Thiabendazole, Niclosamide, Amikacin, Nitazoxanide, Anisindione, Phenylbutanoic Acid, Hydralazine, Hydralazine Hydrochloride, Mycophenolic Acid, Lovastatin, Leflunomide
PSMA5	3	Ixazomib Citrate, Carfilzomib, Bortezomib
C3	1	Clozapine
SMAD3	22	Mitoxantrone, Dasatinib, Hexachlorophene, Lanatoside C, Cetylpyridinium, Epirubicin Hydrochloride, Chlorquinaldol, Azacitidine, Ibrutinib, Pyrithione, Triclosan, Thimerosal, Digitoxin, Dexamethasone, Vincristine, Lopinavir, Triclocarban, Fluorescein, Fluspirilene, Imatinib, Ouabain, Doxorubicin Hydrochloride
SNAP25	4	Onabotulinumtoxina, Incobotulinumtoxina, Abobotulinumtoxina, Diazoxide
FLCN	2	Sirolimus, Everolimus
B2M	3	Thyroglobulin, Pembrolizumab, Amikacin
HLA-B	24	Amoxicillin, Oxcarbazepine, Fosphenytoin, Lamivudine, Ribavirin, Carbimazole, Clozapine, Ticlopidine, Pazopanib, Clavulanic Acid, Clindamycin, Abacavir, Stavudine, Acetazolamide, Thalidomide, Sulfasalazine, Trichloroethylene, Phenytoin, Carbamazepine, Floxacillin, Minocycline, Methazolamide, Dapsone, Methimazole
SAMHD1	1	Stavudine
ISG15	1	Irinotecan
FBXW7	6	Temsirolimus, Sirolimus, Belinostat, Vorinostat, Docetaxel, Regorafenib
IL6	17	Siltuximab, Etanercept, Ifosfamide, Ribavirin, Levofloxacin, Gemfibrozil, Nelfinavir, Saquinavir, Rituximab, Infliximab, Cisplatin, Fenofibrate, Fentanyl, Adalimumab, Insulin, Linezolid, Metronidazole
CDH1	5	Bicalutamide, Capecitabine, Erlotinib, Lapatinib,
OASL	1	Ribavirin
HIF1A	42	Clotrimazole, Amcinonide, Desoximetasone, Isoproterenol, Vincristine Sulfate, Niclosamide, Benzbromarone, Hydroquinone, Isoetharine Mesylate, Ethamsylate, Phenoxybenzamine, Hydrochloride, Dequalinium, Epinephrine, Promazine, Sunitinib, Oxatomide, Mefenamic Acid, Oxytetracycline, Triamterene, Epoetin Alfa, Loratadine, Tolfenamic Acid, Sorafenib, Epinephrine Bitartrate, Piretanide, Diclofenac Sodium, Cycloserine, Nitroglycerin, Levonordefrin, Hydrocortisone, Deferoxamine, Inamrinone, Topotecan Hydrochloride, Tretinoin, Pimozide, Noscapine, Flufenamic Acid, Nifedipine, Dopamine, Sulfasalazine, Axitinib, Norepinephrine Bitartrate
S100A10	1	Dexamethasone
PSMB3	3	Bortezomib, Carfilzomib, Ixazomib Citrate
CP	11	Isoproterenol, Vincristine, Estradiol, Hydroxyurea, Penicillamine, Progesterone, Dexrazoxane, Nicotine, Anakinra, Danazol, Ampicillin
STAT1	1	cisplatin
FGG	4	Eptifibatide, Abciximab, Fibrinolysin, Tirofiban
FGA	7	Reteplase, Fibrinolysin, Eptifibatide, Alteplase, Tirofiban, Abciximab, Anistreplase
C5	2	Ravulizumab, Eculizumab
CXCL2	2	Alteplase, Deferoxamine
SAA1	2	Anakinra, Naproxen

Fig. 5.

Drug-protein interactions. Nodes with dark colors (red) and big sizes are genes with the highest degree

Discussion

Knowing the molecular biology concepts of the SARS-CoV-2 infection requires sufficient knowledge of viral replication, host responses, and disease progression for developing treatment methods. This study aimed to analyze the host gene expression response to SARS-CoV-2 infection compared to HCoV-229E, which will help understand the differences and similarities of the host response to respiratory viruses. This study is the first comparative study of the two viruses, and analysis of host DGEs through SARS-CoV-2 infection among common genes highlighted the role of some of these genes.

This study compared PPIN and GRN related to lung epithelial cell line (A549) treated with HCoV-229E and SARS-CoV-2. We constructed PPI and TF-miR-gene regulatory networks to identify the crucial nodes (hubs, bottlenecks, seeds, MCODE cluster members, and motif members), gene ontology, and functional pathways mediating the two infections’ pathogenesis. The shared and differentiated crucial genes between HCov-229E and SARS-CoV-2 effect on lung cells could identify lung conditions’ molecular mechanisms in COVID-19 patients. Therefore, we determine and validate some of the predicted genes and mechanisms probably mediating lung manifestations of COVID-19 using other experimental literature.

The study investigated the DEGs shared or specific to HCoV-229E, and SARS-CoV-2 PPI and GRN networks expressed differentially in the epithelial lung cells when infected. The CXCL1, KLHL21, SMAD3, HIF1A, and STAT1 genes are shared between HCoV-229E and SARS-CoV-2 PPI and GRN networks. On the other hand, the NPM1 gene and TRIM14, STAT2, and HLA-F genes were observed as a hub and bottleneck in PPI and GRN networks, HCoV-229E, and SARS-CoV-2, respectively. Herein, we have hypothesized that some of these in silico-identified crucial genes and molecular processes can show severe manifestations of COVID-19 in lung cell patients compared with those infected with HCoV-229E.

The analysis of biological processes and functional mechanisms related to crucial genes in response to HCoV-229E and SARS-CoV-2 identified that regulation of the cellular protein process, cell cycle, protein processing in the endoplasmic reticulum, RNA transport, and the proteasome are essential molecular functions in lung cells infected with HCoV-229E. However, in response to infection with SARS-CoV-2, the function type I interferon signaling pathway, RIG-I-like receptor signaling pathway, Chemokine signaling pathway, Cytokine-cytokine receptor interaction, NOD-like receptor signaling pathway, TNF signaling pathway, Cytosolic DNA-sensing pathway, and Viral carcinogenesis are probably activated.

The results of repurposing drug showed Noscapine, Isoetharine mesylate, Cycloserine, Ethamsylate, Cetylpyridinium, Tretinoin, Ixazomib, Vorinostat, Venetoclax, Vorinostat, Ixazomib, Venetoclax, and epoetin alfa as medications to treat pulmonary symptoms in COVID-19 patients.

The CXCL1, HIF1A, SMAD3, STAT1, and KLHL21 genes are shared genes in lung cells infected with both viruses. CXCL1 was selected as a crucial gene in in-silico analysis upregulated in infection with both viruses. This gene is elevated along with CCL11, CCL27, CXCL12, and growth factors in weak to severe patients based on a report focusing on the association between disease pathogenesis and the over-release of cytokines and chemokines [47–49].

SMAD3 and HIF1A are other crucial genes upregulated in lung cells infected with both viruses [50]. The SMAD3 gene in HCoV-229E was one of the primary genes, while in SARS-CoV-2 infection, it was a TF component. Afsar B et al. 2020 hypothesized that HIF-1(an upstream regulator of the TGFβ / Smad3 pathway) promotes cancer growth. ACE in the pulmonary endothelium has two opposing functions: ACE2 as a vasodepressor and ACE1 as a vasoconstrictor in equilibrium, but in COVID-19-induced hypoxia, ACE1 is increased by HIF1α, a hypoxia-induced factor. HIF-1, a medication target for anemia, can reshape hypoxia by reducing ferritin and increasing iron and hemoglobin levels. This factor in non-small cell lung cancer is increased progression of the disease and can be used as a prognosis in these patients [51]. Mingfu Tian et al. in 2021 reported that HIF1α promotes SARS-CoV-2 infection and is vital in inducing an inflammatory response to this virus [52].

The STAT1 gene is also primary in SARS-CoV-2 infection and a TF component in HCoV-229E. This gene was upregulated in lung cells infected with both viruses. STAT1 is part of the JAK / STAT signaling pathway that controls innate and acquired immune function and mediates anti-tumor immunity. Like other members of the STAT family, it increases tumorigenesis, metastasis, and resistance to chemotherapy in various cancers [53–56]. The study of Hector Rincon-Arevalo et al. in 2022 showed impaired STAT1 transcriptional upregulation in severe COVID-19 infection. This factor increases infection through JAK/STAT signaling and interferon pathway [57].

KLHL21, another crucial gene, is essential for cytokinesis and acts as a negative regulator by inhibiting IKKB Kinase signaling in regulating Ikkb, followed by NFKB [58]. This gene has upregulated and downregulated lung cells infected with HCoV-229E and SARS-CoV-2, respectively, which can strengthen the role of this gene in the pathogenesis of SARS-CoV-2. It is predicted that by decreasing the expression of KLHL21 and subsequently increasing IKKB kinase, an increase is witnessed in NF-KB and progression of infection and JAK-STAT pathway, and increasing KLHL21 expression and inhibition of IKKB kinase prevented this pathway from being blocked [59, 60].

The NPM1 is a private hub and bottleneck upregulated in lung cells infected with HCoV-229E. NPM1 is a multifunctional cytoplasmic factor that mediates in transferring human immunodeficiency proteins (HIV1), including Rav and Tat, between the cytoplasm and the nucleus [61, 62]. Besides interacting with the HDV antigen, it also stimulates RNA replication. This adenovirus involves NPM1 bindings to primary virus proteins such as DBP, viral polymerase, and virus assembly [63].

The TRIM14, STAT2, and HLA-F were identified as a specific hub and bottleneck that upregulated in lung cells infected with SARS-CoV-2. Several studies have shown that many TRIM proteins have controlled viral infections in recent years [64, 65]. TRIM proteins could regulate the signaling pathways of antiviral innate immune responses and affect viral replication [66, 67]. New evidence shows that viruses can change TRIM proteins for their replication. TRIM14 uses USP14 to inhibit ubiquitylation of cGAS, which acts as a DNA sensor in response to the innate immune response and critical component of IFN type-1 signaling by viruses, and a decrease in TRIM14 causes further damage to the virus infection [68]. TRIM14 positively regulates IFN signaling mediated by RIG-I-MAV3 due to RNA-infected virus infection. As a result, its deficiency disrupts IFN I production in response to the virus in the early and late stages, acting as an adapter [69, 70]. Gayani Krishanthi et al. in 2023, using a hybrid method that combines the binary reptile search algorithm (BRSA) with the LASSO regression method, showed that ACE2, IFIT5, and TRIM14 genes related to COVID-19 [71]. In the nucleus, the transcription factors regulate the expression of interferon I and pro-inflammatory cytokines. After the viral defense’s initial response, interferon-1 activates JAK-STAT signaling, and STAT1 and STAT2 are phosphorylated to form the IRF9 complex and move to the nucleus to initiate the transcription of IFN-stimulated genes. As a result, interferon I must suppress the early stages of cell proliferation and diffusion [70, 72]. The study by Robbert Boudewijns et al. in 2020 showed STAT2 has a dual role in COVID-19 infection, causing severe lung injury on the one hand but limiting systemic viral spread on the other [73]. HLA-F is another upregulated molecule in lung cells in response to SARS-CoV-2. The expression of HLA-F is involved in various physiological and pathological processes such as cancer, viral infection, autoimmune diseases, and transplantation. Recent data has shown that HLA-F, an immune regulatory molecule, can activate and inhibit immune cell receptors. So, studying this molecule can help understand the lung manifestations of COVID-19 patients [74].

Many viruses can induce metabolic processes in host cells. These modifications of the metabolic process by infected cells can provide energy for virus replication, create a virus, and increase the survival of infected cells. Cellular metabolic pathway inhibition can be a novel therapeutic target [75]. On the other hand, the activation and proliferating of immune system cells rely on metabolic pathways, including glycolysis, lactate metabolite, and O2 tension [76]. Several viruses encode proteins that regulate the cell cycle for viral replication. Many DNA viruses induce the cell cycle in cells and thus help viral replication. In contrast, some viruses arrest the cell cycle in a cell cycle phase. The cell cycle arrest may allow cells to evade the immune response, inhibit the early cell death of infected cells, and promote virus assembly [77].

Viruses regulate the function of intracellular organelles, such as endoplasmic reticulum (ER) and RNA transport systems. Viruses are essential to these organelles to form replication factories (RFs) and assemble infectious virus particles [78]. Mohammed Samer Shaban et al. in 2020 reported that reprogramming of ER stress pathways could suppress coronavirus (SARS-CoV-2, HCoV-229E, and MERS-CoV) replication, inhibited virus-related translational, and downregulation of IRE1α and BiP chaperone [79]. The proteasome is a component of the cell essential for the presentation of antigenic peptides on MHC class I molecules and the activation of NF-Kβ pathways. Many viruses interfere with immunoproteasome function by inhibiting the transcription of immunoproteasome subunits, and viral proteins interact with immunoproteasome components [80]. Qin Wang et al.; reported the interaction of proteasome subunit p42 with SARS-CoV NP. This molecular mechanism regulated SARS-CoV pathogenicity and promoted the interaction of SARS-CoV with host cells [81].

Type I interferons (IFNs) affect innate and adaptive immune cells during viruses, bacteria, parasites, and fungi infections through direct or induction of other mediators. These molecules can suppress the immune response to chronic viral infections [82]. In 2021, Ahmed Abdulwahid Salman et al. reported that IFN could increase the effects of invading viruses during the infection process. SARS-CoV and MERS infection induce weal IFN responses. Using recombinant IFN with other antiviral drugs in clinical trials can improve results in shortening the duration of illness [83]. The clinical studies of type I IFNs by Fuyu Lin et al. in 2020 demonstrate the efficacy of using IFNs via aerosol inhalation to treat and prevent COVID-19 [79]. Acid-inducible gene I (RIG-I)-like receptors (RLRs) play vital roles in innate antiviral immunity.

Taisho Yamada et al. 2021 found that all-trans retinoic acid can increase the expression of RIG-I protein and inhibit the early phase of SARS-CoV-2 infection in human lung cells [84]. Chemokines are bioactive peptides that regulate leukocytes’ migration, recruitment, and activation. These molecules are important in inflammation and essential in controlling viral infections [85]. In COVID-19, chemokines may cause acute respiratory disease syndrome and lead to death in about 40% of severe cases. Several clinical studies reveal that chemokine is involved in different stages of COVID-19 infection and can be selected as a therapeutic target in treating COVID-19 [86].

NOD-like receptors (NLRs) are members of the nucleotide-binding domain leucine-rich repeat family of cytosolic pattern-recognition receptors that regulate the host antiviral immune response. NOD-like receptors have several functions, including maturation of IL-1β and IL-18 following virus infection, modulating signaling pathways initiated by Toll-like and Rig-like helicase receptors, enhancing pro-inflammatory pathways, and IFN and NFκB signaling. A subgroup of regulatory NLRs regulates inflammation negatively. These inhibitory NLRs interact with TRAF molecules and various kinases and modulate cellular processes [87]. The NLR studies are limited to COVID-19 infection. The studies of in vitro and in vivo NLRs can suggest these molecules as novel and promising therapeutic strategies. TNF signaling pathways have a crucial role in the pathogenesis of viruses [88]. Dong Qiu et al. in 2022, using bioinformatics technology, showed a similar gene set of immune or inflammation between the patients with COVID-19 and ones with MS, including IL1B, P2RX7, IFNB1, IFNB1, TNF, and CASP1. These genes are associated with NOD-like receptor (NLR) signaling [89]. The deficiency of TNF can lead to excessive production of the cytokines TGFβ, IL-10, IL-6, and IFN-γ and causes severe lung pathology in patients with COVID-19 [90]. The immune sensors can recognize virus components (lipids, sugars, and nucleic acids) and induce an innate immune response against viruses. The intracellular sensors and adaptor molecules interact with virus-derived nucleic acids, leading to host cells’ pro-inflammatory cytokines, chemokines, and type-I IFNs [91]. Cytosolic DNA sensing via a stimulator of interferon (STING)/INF-β pathway can induce the production of IFNs and NFκB, but RNA viruses also stimulate STING [92].

On the other hand, cytosolic DNA sensing suppresses، effector helper T-cell response and activates Tregs to have functional pathways. Viruses are responsible for 10–15% of human cancers worldwide. Viruses lead to the cumulation of mutations, deviations, DNA damage, and genome instability [93]. Serhiy Souchelnytskyi et al. in 2021 reported that the COVID-19 network was linked to tumorigenesis. Clinical markers of the COVID-19 network, such as EGF, VEGF, TGFβ, and FGF, can regulate the immune system’s proliferation, migration, and death [94].

Noscapine, Isoetharine mesylate, Cycloserine, Ethamsylate, and Epoetin alfa were the repurposed drugs affecting HIF1A. The HIF1A has an essential role in the control of inflammatory and infectious diseases. This factor stimulates the secretion of pro-inflammatory cytokines and growth factors in various infections [95]. Noscapine is often used as an antitussive medication [96]. Noscapine has anti-inflammatory effects and reduces pro-inflammatory factors such as IL-6, IFN-c, and IL-1 β [97]. In 2020, Neeraj Kumar et al. reported that noscapine has a binding affinity to protease (Mpro) of SARS-CoV-2 and is crucial in virus infection and progression [97]. This drug is a novel molecule for inhibiting HIF1A [98]. Isoetharine mesylate increases cAMP production and promotes the relaxation of smooth muscle cells. This drug treats emphysema, bronchitis, and bronchodilator [99]. Isoetharine mesylate is a B-adrenergic receptor agonist and can be used as a fast-acting aerosolized bronchodilator for COVID-19 patients’ respiratory organs [100]. Isoetherine mesylate is a selective beta-2-adrenergic agonist and fast-acting aerosolized bronchodilator for treating respiratory distress in COVID-19 [100].

Cetylpyridinium was a medication that interacted with SMAD3. This drug is used as an antimicrobial agent against a variety of pathogens. This drug disrupts the viral envelope through physicochemical interactions with a lipid bilayer and is used to treat respiratory infections [101]. SMAD3 is upregulated in human lungs in response to HCoV-229E; therefore, this transcription factor probably regulates the expression of genes in the lungs of SARS-CoV-2-infected patients. Decreased SMAD3 expression reduces transcription of TGFB1-regulated target genes and finally reduces inflammation [102]. Filippo D'Amico et al. reported in 2023 that cetylpyridinium can be effective against SARS-CoV-2 salivary virus in vivo. Using a mouthwash containing Cetylpyridinium chloride in SARS-CoV-2-positive individuals may reduce the infectivity and severity of COVID-19 [103]. Therefore, Cetylpyridinium could be suggested as a novel therapeutic in treating SARS-CoV-2 for further in-vitro investigations. Ethamsylate (2,5-dihydroxy-benzene-sulfonate dimethylammonium salt) is a hemostatic agent that prevents capillary bleeding [104]. Epoetin alfa is a version of human erythropoietin. Erythropoietin improves severe acute respiratory syndrome in SARS-CoV-2 virus infection through leukocyte release from bone marrow, cytokine modulation, and cytokine modulation [105]. Epoetin alfa has been used to treat anemia for over a decade in application in patients with HCV infections [106]. Rhonda Souvenir et al. reported that this drug could inhibit HIF1A [107]. Cycloserine is an antibiotic used to treat urinary infections [108].

NPM1 is another crucial molecule that interacts with Tretinoin, Ixazomib, Vorinostat, and Venetoclax drugs. NPM1 interacts with a nucleocapsid of viruses and protects them against proteolytic cleavage, thus enhancing cell survival [109]. Tretinoin is a vitamin A derivative and an essential cell reproduction, proliferation, and differentiation regulator. Two studies reported that retinoic acid compounds could inhibit the replication of viruses by upregulating the innate immune response elements [110, 111]. This drug could be a therapeutic strategy as an ion channel blocker and virus assembly inhibitor [112]. In 2022, Emine Müge Acar et al. showed that retinoids could be a safe treatment for COVID-19 patients [113]. Retinoic acid compounds can also help degrade NPM1 [114] and be an effective drug in treating HCoV infections. Vorinostat is a histone deacetylase inhibitor that reduces viruses’ latent reservoir in vivo [115] and is also used in treating acute myeloid leukemia (AML) [116, 117]. Histon deacetylases (HDACs) control innate and adaptive immunity on HDAC inhibitors and can be a pharmacological treatment strategy for COVID-19 patients [118]. Chiara Ripamonti et al. in 2022 reported HDAC inhibition can use as a therapeutic strategy in COVID-19. As HDAC Inhibition, Vorinostat decreases the pro-inflammatory cytokine expression activation of TLR4 and increases CD40 + /CD86 + monocytes and secretion of IL-1β [119]. Ixazomib could decrease the expression of NPM1 and induce apoptosis in cancer cells [120]. Chih-Chieh Chen et al. in 2022 reported that Venetoclax can bind to spike protein of SARS-CoV-2. This drug inhibits the interaction of spike protein with the ACE2 receptor [121]. Venetoclax is another drug that can cause the elimination of NPM1 in AML patients [122]. Chih-Chieh Chen et al. in 2022 shown that Venetoclax degrade expression of spike protein through amino acids Q493 and S494. This process prevent interaction spike protein with ACE2 receptor. Ribavirin is another drug that shown in results of repurposing drug. Based on Ribavirin's direct antiviral activity against SARS and MERS outbreaks and also 2019-nCoV in vitro, this drug can use for end persistent outbreaks of SARS-CoV-2 [123]. This drug interacted with HLA-B, IL-6, OASL, and CXCL8 proteins.

Identifying essential functional genes in the lung infection of people infected with SARS-CoV-2 and predicting drugs that target these genes can suggest treatment strategies with fewer complications for improving these patients. The present in-silico study and its findings can lead clinical studies for the treatment of SARS-CoV-2 lung infections to be done in a more targeted and accurate direction.

Conclusions

The systems biology approach can help predict the molecular pathology of diseases. This study provided a systems biology approach for deciphering the molecular mechanism of HCoV-229E and SARS-CoV-2 and drug repurposing through their PPIN and GRN. Our study revealed critical genes, biological processes, and functional pathways that play crucial roles in the pathogenesis of HCoV-229E and SARS-CoV-2.

Using the limited available data on lung cells infected with HCoV-229E and SARS-CoV-2, we predicted the molecular mechanisms behind the respiratory pathogenesis of SARS-CoV-2 infection compared to HCoV-229E. Previous experimental studies validated some of the in-silico findings, and others should be further examined experimentally.

The CXCL1, KLHL21, SMAD3, HIF1A, and STAT1 were identified as shared DEGs/TFs between HCoV-229E and SARS-CoV-2. The CXCL1gene with over-release of cytokines and chemokines, HIF-1α with induce hypoxia, STAT1 with control innate and acquired immune function involved in pathogenesis disease. The downregulation of KLHL21 activates NF-KB and JAK-STAT pathways and increases the progression of infection. The top critical nodes specific to only one infection included the NPM1 gene and TRIM14, STAT2, and HLA-F genes in HCoV-229E and SARS-CoV-2, respectively. NPM1 and TRIM14 regulate virus replication, the immune system in the virus, and STAT2 and HLA-F genes are immune regulatory molecules in viral infections. The top identified enriched pathways included the cell cycle and Proteasome in HCoV-229E and RIG-I-like receptor, Chemokine, Cytokine-cytokine, NOD-like receptor, and TNF signaling pathways in SARS-CoV-2 infection. We suggested Cetylpyridinium, Cycloserine, Noscapine, Ethamsylate, Epoetin alfa, Isoetharine mesylate, Tretinoin, Ixazomib, Vorinostat, and Venetoclax as repurposed-drug candidates since they interact with essential molecules (NPM1, HIF1A, and SMAD3) in SARS-CoV-2 condition and seem logical and suitable for further investigations in vitro and in vivo as therapeutics. These drugs improve COVID-19 infection in patients by regulating virus replication and the immune system and inhibiting interaction spike protein with the ACE2 receptor. This study relies on limited data, with a few samples available on lung cells infected with both viruses. Therefore, using data with more samples can identify critical genes with higher accuracy, and naturally, repurposing drugs can be done with a higher degree of confidence. A more significant number of data samples had better be considered in future studies to obtain more accurate results related to COVID-19. Although the results of our research have determined the essential genes that cause respiratory problems and then suggested pharmaceutical solutions to reduce these complications, creating animal models of SARS-CoV-2 and confirming the expression of crucial genes obtained from this study in the lung samples of these animals, as well as prescribing these drugs to determine their performance to reduce pulmonary complications in these animals, can be effective treatment methods for improving pulmonary infection in COVID-19 patients.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

SARS-CoV-2

Severe Acute Respiratory Syndrome Coronavirus 2

HCOV-229E

Human Coronavirus 229E

COVID-19

Coronavirus Disease 2019

MERS

Middle East Respiratory Syndrome

MOI

Multiplicities of Infection

NHBE

Normal Human Bronchial Epithelial

DEGs

Differentially Expressed Genes

HIPPIE

Human Integrated Protein-Protein Interaction rEference

PPI

Protein-Protein Interaction

HPRD

Human Protein Reference Database

MCODE

Molecular Complex Detection

GRN

Gene Regulatory Network

DGIdb

Drug Gene Interaction Database

INFs

Interferons

RLRs

(RIG-I)-Like Receptors

NLRs

NOD-Like Receptors

Funding

This work was supported by the Shiraz University of Medical Science (grant number 25859).

Declarations

Conflicts of interest

The authors declare no conflict of interest.

Ethics approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

Not applicable.