COVID-19 comorbidity and metabolic syndrome : is there a Molecular basis ?

The risk factors associated with COVID-19 related severity, morbidity, and mortality, i.e., obesity (often associated with NAFLD), hyperglycemia, hypertension and dyslipidemia all cluster together as metabolic syndrome (MetS). Instead of studying association of these risk factors with COVID-19, it makes sense studying the association between MetS on one hand and COVID-19 on the other. This study explores a molecular basis underpinning the above association. Severity of COVID-19 patients with MetS could be due to functional alterations of host proteins due to their interactions with viral proteins. We collected data from Enrichr (https://amp.pharm.mssm.edu/Enrichr/), DisGeNET (https://www.disgenet.org/) and others and carried out enrichment analysis using Enrichr. Various biological processes and pathways associated with viral protein interacting partners are known to involve in metabolic diseases. The molecular pathways underlying insulin resistance, insulin signaling and insulin secretion are not only involved in diabetes but also in CVD and obesity (associated with Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 21 June 2020 doi:10.20944/preprints202006.0245.v1 © 2020 by the author(s). Distributed under a Creative Commons CC BY license. 2 non-alcoholic fatty liver disease; NAFLD). Lipid metabolism/lipogenesis, fatty acid oxidation and inflammation are associated with MetS. Viral interacting host proteins are associated and enriched with terms like hyperglycemia, coronary artery disease, hypertensive disease related to CVD and liver diseases in DisGeNET. Association of viral interacting proteins with disease-relevant biological processes, pathways and disease-related terms suggests that altered host protein function following interaction with viral proteins might contribute to frequent occurrence and/or severity of COVID-19 in subjects with MetS. Such analysis not only provides a molecular basis of comorbidity but also incriminates host proteins in viral replication, growth and identifies possible drug targets for intervention.


Metabolic syndrome and comorbidity of COVID-19
Metabolic syndrome (MetS) consists of a cluster of inherited metabolic traits that center on obesity, insulin resistance, hypertension and atherogenic dyslipidemia and contribute to the prevalence of CVD, T2D, NAFLD and cancer (Figure 1). These traits were more prevalent among people with severe COVID-19 compared to their non-severe counterparts [3,6,11,12]. Detailed information of the co-existence of obesity, CVD, hypertension, diabetes and NAFLD is shown in the supplementary Text (Supplementary text 1.1). It is thus pertinent to consider whether metabolic syndrome may be considered as co-morbid condition for COVID-19.

Altered expression of host genes in virus infected cells
Altered expression of host genes in vitro following infection with SARS-CoV-2 has been observed [15,16]. Analysis of host transcriptional response to SARS-CoV-2 and other respiratory infections through in vitro, ex vivo and in vivo model systems reveal that infection deregulates many genes that serve as unique transcriptional footprints for the virus. Comparison with influenza A and respiratory syncytial viruses reveals an altered response that lacks robust induction of cytokine and chemokine.
Transcriptional profiling in lung biopsy specimens from COVID-19 subjects reveal a unique and inappropriate inflammatory response defined by elevated chemokine expression [16]. Altered expressions of host genes on infection with many viruses including SARS-CoV are catalogued in Enrichr at https://amp.pharm.mssm.edu/Enrichr/ [17,18].

Interactions between host proteins and SARS-CoV-2 coded viral proteins
Viral proteins, including SARS-CoV and other human corona virus coded proteins interact with host proteins [1,[19][20][21]. Such interactions hijack host proteins and modify cellular processes and pathways to evade immune response of the host and survival of viruses. Interaction of viral protein with the host proteins might modulate the functions of the host proteins by (i) changing cellular localization, transport and interaction with other host proteins, (ii) changing activation of the host proteins by post translational modifications and (iii) degradation of the host proteins. SARS-CoV coded ORF6 interacts with STAT1 and prevents entry of STAT1 into the nucleus [22]. M protein of SARS-CoV interacts with TRF3 and inhibits TBK1/IKKepsilon complex assembly [23]. NSP1 interacts with 40S ribosomal subunit and inactivates the translational activity of the 40S subunits. NSP1-40S ribosome complex can modify the 5' region of capped mRNA template inhibiting the translation of the modified mRNA [24] as well as degradation of mRNA [25]. NSP1 inhibits phosphorylation of STAT1 and represses target genes like ISG15, ISG54, and ISG56 of STAT1 [26]. Inhibition of phosphorylation of IRF3 by ORF3b, ORF6 and N proteins of SARS-CoV and alters the activation of IRF3 [27]. Interaction and inhibition of phosphorylation and nuclear translocation of IRF3 by viral Papain-like protease (PLpro) have been observed [28]. Host protein IRF3 by PLproinhibits de-ubiquitination and changes the stability of IRF3 [29]. These interactions of different viral proteins with IRF3 and its downstream STAT1 evade the host immune response. Host-viral protein interactions for SARS-CoV and MERS-CoV are reviewed extensively [19,30].
Interactions between SARS-CoV-2 proteins and the host proteins are not studied extensively. Very ORF9c) and modification of extracellular matrix (NSP9). In addition, host partners were also associated with various innate immune function, protein synthesis and proteasomal degradation [31] .

Knowledge gap
COVID-19 patients with MetS appear to have severe disease with higher mortality. When infected with SARS-CoV-2, people with MetS have unmasking and/or aggravation of some of its features.
Alternatively, pre-existing metabolic conditions might allow infected SARS-CoV-2 to grow aggressively or prevent clearance of the virus due to immune modulation, leading to more severe disease manifestations. Increased viral load and slow clearance of the virus have been reported in severe COVID-19 patients [32,33]. Severity being associated with MetS, there might be correlation between viral load and slow clearance. SARS-CoV-2 infection triggers stress conditions leading to release of glucocorticoids and catecholamines that increase blood glucose levels [34]. COVID-19 patients with uncontrolled T2D had higher mortality than those with good glycemic control [35]. Two receptors ACE2 and DPP4, necessary for entry of SARS-CoV-2 and MERS-CoV respectively are involved in inflammatory pathways, cardiovascular physiology and glucose homeostasis [36] showing overlap of viral entry, growth and MetS. However, molecular interconnection between COVID-19 severity and MetS remains largely unknown.

Hypothesis
Severity of COVID-19 patients with MetS could be due to functional alterations of host proteins by interacting with viral proteins.

2.
Material and Methods

SARS viral protein interacting partners of host proteins
From published data, we collected 609 host proteins that interact with corona virus proteins; specific viral proteins are known for 410 host proteins [19-21, 31, 37]. Detailed sources are shown in the Supplementary Text Table ST2 and Supplementary Table S1.

Enrichment analysis
To find the functional associations of viral interacting proteins, enrichment analysis for the genes was carried out using online facility at Enrichr (https://amp.pharm.mssm.edu/Enrichr/) [17,18]. Enrichr is an integrative web-based software application for analysis of a gene-set comparing with various geneset libraries. Given an input list of genes, it provides enrichment for different libraries like KEGG pathways, Gene Ontology (GO) terms for biological processes (BPs). DisGeNET, a discovery platform containing publicly available collections of genes and variants associated to human diseases and is also available at the same site.

Genes/proteins elevated in organs/tissues
Datasets for elevated genes and genes expressed in all tissues/organ were collected for 8 organs namely brain, heart, kidney, liver, lung, pancreas, spleen, thyroid and adipose tissue as reported by [38]. Datasets for tissue/organ-specific proteins also collected from data reported by Wilhelm, et al. [39]. Combing these two datasets, unique proteins/genes were determined and designated as elevated 8 genes in organ/tissue as well as the housekeeping genes. Expression of genes elevated in different tissues and housekeeping genes are shown in the Supplementary Table S7.
For comparison of different set of genes/proteins, we used online facility at http://bioinformatics.psb.ugent.be/webtools/Venn/

Enriched Biological processes (BPs) with host proteins interacting with SARS virus coded proteins.
To

Enrichment of KEGG and Wiki Pathways
Enrichment analysis with 609 interacting proteins revealed that all together 250 KEGG pathways were associated with viral protein interacting host proteins (Supplementary Table S3A); 17 pathways were significantly (adjusted p≤0.05) enriched with viral protein interacting host proteins (Figure 3). In addition to KEGG pathways, Enrichr also provides enrichment for Wiki Pathways

Association of interacting host partners of viral proteins with various metabolic disease/trait catalogued in DisGeNET database
To identify whether interacting partners of viral proteins are associated with different diseases or disease related conditions/trait, we explored DisGeNET database. This database catalogs gene-disease relations from diverse experimental data like Genome Wide association studies (GWAS) and others [40] as catalogued in Enrichr at https://amp.pharm.mssm.edu/Enrichr/ [17,18]. Such analysis with viral interacting protein partners revealed that 55 diseases/conditions were significantly (adjusted p≤0.05) associated with viral protein interacting partners; 5308 diseases/conditions were associated with host proteins (Supplementary Table S5A  only involved in diabetes [41][42][43][44] but also in cardiovascular diseases [45] and obesity [46]. The metabolic products of angiotensin I and II play an important role in cardiovascular pathophysiology [47].     (Figure 8). The viral proteins common for 3 diseases (viz. E proteins, ORF3a, ORF14, and NSP6) are shown in Pink. Light yellow squares represent the viral proteins associated with 2 diseases (NSP15, ORF7b and ORF10). The rest of the viral proteins (Blue squares) are unique for a particular disease. This analysis has strengthened our hypothesis that the frequently occurring comorbidity of MetS could be due to interactions of viral protein with the host proteins.

Organ/tissue enriched expression of host protein interacting partners of SARS-CoV coded protein
To identify the enriched genes, defined as genes whose mRNA level, determined by RNA sequencing, in a particular tissue is at least five times higher than those in all other tissues, in different organ/tissue like brain, heart, kidney, liver lung, pancreas, spleen and thyroid together with adipose tissue, we used result reported by Uhlén et al. [38]. Gene whose mRNA level is detected in all tissue with fragments per kilobase of exon model per million reads mapped greater than1 is considered to be housekeeping genes [38]. We used these organ/tissue enriched and compared with viral protein interacting partners,

Altered expression of host genes after infection with SARS-CoV
Combined proteome and transcriptome datasets from infection of SARS-CoV in human lung epithelial cells in culture are utilized to predict regulatory genes involved in the host response. After infection expression of genes were determined at different time points [49]. Increased and decreased expressions of the genes are taken from Enrchr. Summary of the result is shown in the Table and the detail is shown in the supplementary Text Table T2. Overall about 5% of the increased expression and similar proportion of decreased expression of host genes in SARS-CoV infected cells also interact with viral proteins. Out of 609 viral protein interacting host proteins, expression of about 28% genes coded for the viral protein interacting partners increased and expression of 36% genes decreased; overall expression of 64% of genes altered ( Table 3). Mechanism of deregulation of host genes by viral infection remains largely unknown. Deregulation and physical interaction of deregulated host genes, support the general notion that deregulated genes products interact with each other and carry out biological functions.

Mechanism (s) of severity of COVID-19 in patients with MetS
Severity and mortality of COVID-19 is associated with MetS. Increased viral load and decreased clearance of viruses are major contributors to COVID-19 related morbidity and mortality [32,33].
Viral load for COVID-19 patient comorbid with MetS is not known. Possible increased viral load and slow clearance in COVID-19 patients with MetS, if observed, might explain the higher death rate and poor outcomes in them. Infection with SARS-CoV-2 in subjects with MetS, a state of chronic low grade inflammation, may trigger the release of catecholamines and glucocorticoids that further increase blood glucose levels and increase glucose variability [34]. Hyperglycemia and insulin resistance are known to promote enhanced synthesis of glycation end products (AGEs), proinflammatory cytokines and adhesion molecules that may interact with viral proteins and worsen outcome in COVID-19 [50,51]. Thus, infection enhances the metabolic conditions which in turn increase pathogenicity of the virus by modulating AGEs and pro-inflammatory cytokines.
In MetS, immune system is compromised due to defects in inflammation. This condition might delay in clearing viruses and recovery. MERS-CoV infection to diabetic mice results in immune deregulation and enhances disease severity of the infection [52]. Virus RNA was detected in lungs of 100% of the COVID-19 patients, while in about 40% of the patients viral RNA was detected in heart, liver and kidney, possibly reached through blood [53]. Increased levels of troponin levels, a sensitive marker for acute cardiac injury, is commonly observed in severe cases of COVID-19 and is strongly associated with mortality [54]. Acute coronary syndrome and myocardial infarction were noted to occur after SARS [55]. So far no convincing reports are available to show that SARS-CoV-2 infects pancreatic cells [56] or causes acute pancreatitis by the infection [57]. However, effects of direct infection of SARS-CoV in pancreas inducing diabetes in non-diabetic persons are known [58].

Common genes in COVID-19 and MetS
Sixteen genes are related to infectious diseases, CVD, diabetes and liver diseases (Figure 7).
Relevance of ACE2, ALB, ITGB1, IRF3, MARK2, IL17RA, HMOX1, SCARB1, GPX1 and BCL2 genes for infections and MetS are known. ACE2 has been implicated in obesity, hypertension, diabetes, myocardial infarction, heart failure and inflammatory lung disease [59][60][61]. ACE2 has been proposed to be the receptor for entry into the host cells by interacting with the spike protein of SARS-CoV-2 [62] and SARS-CoV, possibly through renin-angiotensin system as shown for SARS-CoV [63]. Serum albumin levels are decreased in infection, liver dysfunction, obesity, diabetes and CVD due to its ability to act as antioxidant [64,65], modify inflammation, cytokine production [66] and maintain capillary membrane stability and fluid balance across the capillary [67]. ITGB1, a member of integrin beta chain family, promotes entry of different viruses [68], although its role in SARS-CoV-2 remains unknown. ITGB activation enhances secretion of insulin and inhibited by excess glucose [69].
Interferon regulatory factor 3 (IRF3) modulates innate immune response to viral and bacterial infections [70], inflammation [71], glucose and lipid homeostasis [72,73] by interacting with viral proteins and regulating interferon genes. IL17RA by altering downstream cytokines and chemokine involves in infectious disease and reviewed [74], diabetes [75] and heart failure [76]. Evidences that the IL-17 signaling pathway contributes to the pathogenesis of liver diseases by inhibiting fatty acid βoxidation and altering cell death [77], ERK1/2/p65 signaling pathway [78], obesity, inflammation and reviewed [79] are available. Modified IL-17 signaling pathway has been observed in chronic viral hepatitis [80]. SCARB1, a receptor for high density lipoproteins, facilitates uptake of cholesteryl esters from HDL to the liver. It has been implicated in entry of HCV [81]. Level of serum HDL-Cholesterol is risk factor for CVD [82,83], hypertension, hyperlipidemia, insulin resistance and obesity [84].
Microtubule Affinity Regulating Kinase 2 (MARK2) involves in motility and uncoatng HIV-1 by locally phosphorylating the viral cores [85]. It is one of the kinases that phosphorylates AMPK and involves in epithelial mesenchymal transition [86], metabolic rate, adiposity, and insulin hypersensitivity [87]. HMOX1 possess cytoprotective, anti-inflammatory and antioxidant properties and modulates HBV infection [88], fatty liver, NASH, NAFL, insulin resistance, hypertension, vascular dysfunction, diabetes, atheroscerosis and others [89]. Glutathione peroxidase 1 (GPX-1) is an enzyme having antioxidant properties and involves in coxsackie virus induced myocarditis [90], influenza A virus-induced lung inflammation [91], HCV induced liver fibrosis and heptocellular carcinoma [92] and HBB replication [93]. Increased oxidative stress in MetS and a decreased antioxidative defense were correlated with triglycerides, high-density lipoprotein cholesterol, waist circumference, blood pressure components of MetS [94]. There are reports to show that mortality due to CVD is associated with low serum level selenium. Low selenium levels also correlate with risk of myocardial infarction. Contradictory result of no effect or increased risk of ischemic heart disease ((reviewed in Bellinger, et al. [95]) has also been reported. Even though various proteins identified here have common role in viral infections and MetS, whether these are relevant for SARS-CoV-2 infection remains unknown and further studies are necessary. Taking together, our analysis provides molecular clues for over representation of MetS in COVID-19 patients, especially for severe disease and/or or non-survivors. Altered gene expression from peripheral blood in MetS shows a general systemic inflammation condition mediated by the innate immune system. NF-κB activation, activation of T cells, macrophages and increased expression of CXCL14, IL4R, GPX1 and many others are observed in MetS, diabetes and coronary heart disease [96]. In pre-existing MetS condition, when infection with SARS-CoV-2 occurs, viral coded proteins interact with host proteins that have already been compromised or activated. Such interactions might result in functional interactions synergistically between the infection and pre-existing conditions. This condition might delay in clearing viruses and recovery as observed for MERS-CoV infection to diabetic mice [52]. It remains unknown whether infection with SARS-CoV-2 in individual without any comorbidity induces acute injury to heart, liver or pancreas either due to direct infection in these organ or by systemic changes in inflammation and immune modulation. Acute cardiac injury is strongly associated with mortality and severity [54].
SARS-CoV coded structural protein 7a can cause direct liver injury by inducing apoptosis in vitro studies [97]. Liver injury has also been observed in COVID-19 patients [98,99].