COVID-19 Virulence in Aged Patients Might Be Impacted by the Host Cellular MicroRNAs Abundance/Profile

The World health organization (WHO) declared Coronavirus disease 2019 (COVID-19) a global pandemic and a severe public health crisis. Drastic measures to combat COVID-19 are warranted due to its contagiousness and higher mortality rates, specifically in the aged patient population. At the current stage, due to the lack of effective treatment strategies for COVID-19 innovative approaches need to be considered. It is well known that host cellular miRNAs can directly target both viral 3'UTR and coding region of the viral genome to induce the antiviral effect. In this study, we did in silico analysis of human miRNAs targeting SARS (4 isolates) and COVID-19 (29 recent isolates from different regions) genome and correlated our findings with aging and underlying conditions. We found 848 common miRNAs targeting the SARS genome and 873 common microRNAs targeting the COVID-19 genome. Out of a total of 848 miRNAs from SARS, only 558 commonly present in all COVID-19 isolates. Interestingly, 315 miRNAs are unique for COVID-19 isolates and 290 miRNAs unique to SARS. We also noted that out of 29 COVID-19 isolates, 19 isolates have identical miRNA targets. The COVID-19 isolates, Netherland (EPI_ISL_422601), Australia (EPI_ISL_413214), and Wuhan (EPI_ISL_403931) showed six, four, and four unique miRNAs targets, respectively. Furthermore, GO, and KEGG pathway analysis showed that COVID-19 targeting human miRNAs involved in various age-related signaling and diseases. Recent studies also suggested that some of the human miRNAs targeting COVID-19 decreased with aging and underlying conditions. GO and KEGG identified impaired signaling pathway may be due to low abundance miRNA which might be one of the contributing factors for the increasing severity and mortality in aged individuals and with other underlying conditions. Further, in vitro and in vivo studies are needed to validate some of these targets and identify potential therapeutic targets.

virally loaded respiratory droplets [3]. However, the transmission may also occur by touching contaminated surfaces and subsequently touching one's eyes, nose, or mouth [4,5]. The incubation period, the period before the first symptoms present itself, is on average of 4-5 days, with 95% presenting within 12.5 days [6,7].
The World Health Organization (WHO) reported 2,920,905 confirmed infected, and 203,269 COVID-19 related deaths globally as of April 25th, 2020 [8]. Drastic measures to combat SARS-CoV-2 are warranted due to its contagiousness, viability, and higher death toll compared to its predecessor SARS-CoV. Although SARS-CoV had a significantly higher case-fatality (around 10%) [9], the virus was not viable enough to remain in the human population and never spread in the United States and other countries like SARS-CoV-2 [10]. SARS-CoV-2, on the other hand, appears to rapidly spread, which is contributing to its global spread and a significantly higher number of cases.
Severe and critical cases of COVID-19 disproportionately affect middle-aged and older/aged populations, with increased mortality in aged adults [11]. Illness severity ranges from asymptomatic, mild, severe, and critical. Mild illness is characterized as no to mild pneumonia [11]. The severe disease presents with dyspnea, hypoxia, or >50% lung involvement, and critical illness occurs when there is respiratory failure, shock, or multiorgan failure [11]. The Centers for Disease Control and Prevention (CDC) and the WHO estimate the mortality rates of SARS-CoV-2 to be around 3.4% and case fatality rates (CFR) may increase up to 10-27 % in individuals 80 years old or older [8,12]. SARS-CoV-2 is a novel virus, and very little is known about it. In such a scenario, it is crucial to understand the COVID-19 pathobiology to identify the innovative treatment strategy. Our laboratory and others have reported that miRNAs play a critical role in age-related complications [13][14][15][16].
MiRNAs play a vital role in the pathogenesis of various diseases, including viral infections, disease progression, and inhibition [17][18][19][20][21][22][23]. MicroRNAs are small noncoding RNAs, bind to 3'UTR of mRNA, and inhibit translation or induce degradation of mRNAs [19,21,23]. Recent studies suggested that host cellular miRNAs can directly target both viral 3'UTR and coding region of the viral genome to induce antiviral effect [19,21,23]. For example, the number of groups previously reported that host miRNAs (miR-323, miR-491, miR-485, miR-654, and miR-3145) bind to influenza PB1 gene coding region, degrade RNA and inhibit viral translation and reduce the accumulation of viral particles [19,24,25]. Furthermore, the host cellular miRNA-29a inhibit Human immunodeficiency virus type 1 (HIV-1) nef protein expression and thus, inhibit viral replication [26]. On the contrary, some groups also suggested the positive effect of host miRNAs on viral replication. For example, miR-122 binding to 3' and 5' UTR of hepatotropic virus RNA and increase viral RNA stability leads to viral propagation [18,20,22]. Based on the above reports, we did in silico analysis of miRNAs targeting SARS and COVID-19 (recent isolates from different regions) to understand the pathophysiology and identify novel therapeutic targets.

Viral genome sequence retrieval, homology, and phylogenetic analyses
The complete genome sequences of the SARS and COVID-19 isolates were retrieved from the GenBank database. We retrieved four SARS and 29 COVID-19 sequences from NCBI and GISAID for for homology and phylogenetic analysis. Details of sequence identification are summarizing in Table 1. The sequences were aligned using the Multiple Sequence Alignment tool at Clustal Omega 1.2.3 on Geneious Prime 2020.1.1. The phylogenetic alignment tree generated using the neighborjoining method.

COVID-19 genome and human MiRNA target analysis
As previously mentioned above, miRNAs are known to target 3'UTR and coding sequences and prevent mRNA translation or degrade RNA. Keeping this in consideration, we used the whole viral genome sequence for miRNA target analysis. We used miRDB (http://www.mirdb.org/) software to identify novel human miRNAs targeting the COVID-19 viral genome [27,28]. Furthermore, we used this data to correlate with the existing literature.

GO, and KEGG pathway analysis
Gene Ontology (GO) and KEGG signaling pathway analyses were performed on human microRNAs targeting the COVID-19 genome using DIANA-miRPath v 3.0 (http://diana.imis.athena-innovation.gr/DianaTools/ index.php) [29]. We used miRNAs with a target score above 90 because these miRNAs have high probability of being real targets [27,28].

MicroRNAs targeting SARS and COVID-19 genome
As we mentioned above, we performed miRNA analysis on the whole genome of SARS and COVID-19 due to the efficiency of miRNAs targeting both 3'UTR and coding region. Our analysis found some interesting results. We found 848 common miRNAs targeting the SARS genome (four isolates, NC_004718.3, AY291451, AY338175, AY348314) and 873 common microRNAs targeting the 29 isolates of COVID-19 genome (Supplementary Table  2). Out of a total of 848 miRNAs from SARS, only 558 commonly present in all COVID-19 isolates (Fig. 2, Supplementary Table 2). Interestingly, 315 miRNAs are unique for COVID-19 isolates and 290 miRNAs unique to SRAS (Fig. 2). Furthermore, the COVID-19 targeting miRNAs with a higher target score (above 94) showed more than ten target sites and maximum complementary miRNA-RNA seed paring (~6-8 nucleotide seed base pairing) ( Table 3). We also noted that out of 29 COVID-19 isolates, 19 have identical miRNA targets (Table 4). Ten isolates have unique miRNAs targets. Among ten isolates, Netherland (EPI_ISL_422601), Australia (EPI_ISL_413214), and Wuhan (EPI_ISL_403931) showed six, four, and four unique miRNAs targets respectively. The detail is given in the table (Table 4).

GO and KEGG pathway analysis of miRNAs targeting COVID-19
To identify the functional relevance of human miRNAs targeting COVID-19, we performed Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotation and GO analysis. The KEGG annotation analysis data reveal that these miRNAs plays important in various Cancer signaling (e.g Renal, Pancreatic, Prostate, Colorectal, Melanoma ), Hippo signaling pathway, cardiomyopathy, Wnt signaling, Circadian rhythm, stem cell signaling, Adrenergic signaling in cardiomyocytes and number of other signaling pathways (Table 5). GO analysis data showed more than 72 biological processes were associated with the miRNAs targeting COVID-19 ( Table  6). The biological processes such as organelle, ion binding, cellular, biosynthesis process, protein complex, immune response, and viral process are regulated by human miRNAs targeting the COVID-19 genome. Details of KEGG annotation and GO analysis are shown in Table 5, and Table 6, respectively.

DISCUSSION
Host cellular miRNAs are known to play an antiviral role in the number of published studies [18][19][20][21][22][23]. In this study, we performed in silico analysis of human cellular miRNAs targeting SARS and COVID-19 (isolates) genome and identified some novel miRNAs. We identified number (558) of common human cellular miRNAs targeting both SARS and COVID-19 genome. Top 10 common miRNAs have a target score of ≥95 (target score between 99-95), and each miRNA contains more than at least ten sites in the targeting viral genome (Table 3), indicating possible antiviral property for coronavirus infection (for both SARS and COVID-19). It will be interesting to verify these miRNA in-vitro and invivo animal models to use as a therapeutic target in the future. Cocktail of multiple miRNAs mimics through the intranasal route will be useful in coronavirus infection. These human miRNAs targeting coronavirus can be useful to combat any future outbreaks. In a previous report, host cellular miRNAs-181 binds to the ORF-4 region at the viral genome of porcine reproductive and respiratory syndrome virus (PRRSV) to inhibit its replication [17]. One step further, Guo et al. (2013) delivered intranasal inhalation of miR-181 mimics to slow down the progression of PRRSV in an experimental porcine model [17]. Another study used intranasal inoculation of miR-130 mimic to protect piglets from lethal challenge of PRRSV [30]. Similarly, the intranasal administration of chemically modified five miRNA mimics protected mice from H1N1 viral infection [31].  In humans, coronavirus infection is pervasive and usually causes common cold-like symptoms, but some of them can lead to serious illness. Previous coronavirus outbreak in 2002, known as SARS-CoV spread worldwide but was contained quickly [9]. On the other hand, the current coronavirus (COVID-19) is highly contagious and spread worldwide rapidly and became a global pandemic [8,11]. It will be interesting to know what separates these two viruses based on the genomic sequence. Ours (Table 2 and Fig. 1) and other data [32][33][34] of SARS-CoV and COVID-19 genome sequence analysis showed sequence homology of around ~78.7%, which was obvious, but our miRNA target scan data showed striking results. We found considerable changes in the number of human miRNAs targeting SARS-CoV (848 miRNAs) and COVID-19 (~873 miRNAs). Moreover, we also found unique miRNAs for SARS-CoV (290 miRNAs) and COVID-19 (315 miRNAs) isolates. At nucleotide ( Table 2) and phylogenetic levels (Fig. 1), both viruses look closely related, but host cellular miRNAs target comparison differences are much more significant. This might be one of the reasons for COVID-19 to increase infectivity and easy viral propagation compared to SARS-CoV. The COVID-19 recent isolates (twentynine different isolates) are closely related at nucleotide, phylogenetic ( Fig. 1 and Table 2), and host cellular miRNAs target level ( Fig. 2 and Supplementary Table 2). The Netherland (EPI_ISL_422601), Australia (EPI_ISL_413214), and Wuhan (EPI_ISL_403931) isolates have more mutation and few different host cellular miRNAs targeting sites compared to other 26 COVID-19 isolates (Table 4). In our study, we used 29 isolates from different geographical reagion, which informs that COVID-19 nucleotide sequences and thus the human miRNAs target sites are not changing among the isolates/viruses (Table 4). In most of the COVID-19 cases, the virus entered a country from different geographical locations. At the time of writing this paper, only the USA and China had sequenced multiple viruses isolates from a different region. There is a need to sequence more of the COVID-19 viral genome from different parts of each country before coming to any conclusion.
The most important and striking feature of COVID-19 is the increased case fatality rate in aged individuals. The CDC reports 45% of cases requiring hospitalization are patients ≥65 years old [12]. Furthermore, 80% of COVID-19 related deaths occur in these patients (≥65 years old) [12]. Moreover, individuals with underlying conditions such as cardiovascular, chronic lung disease, diabetes, kidney, liver diseases, and cancer are at higher risk for severe illness and mortality. On the contrary, <1% mortality is observed in patients between 20-54 years of age or younger [12]. Based on miRNA target analysis and published literature, we hypothesized that the COVID-19 genome replicates and propagates viral particles easily in the aged individuals compared to younger due to the overall low abundance of miRNA expression with age. Recent human studies reported that the overall expression of miRNAs levels goes down with age [16,35,36]. For example, Huan et al. (2018) reported that 81% (103 miRNAs) microRNAs were negatively, and 19% (24 miRNAs) positively correlated with age in human peripheral blood [36]. We speculate that in younger individuals with COVID-19 infection, host cellular miRNAs bind to the complementary site of the viral RNA genome and prevent its replication; however, in an aged individual, this mechanism is not that efficient leads to accumulation of viral particle and severe illness. This might be true for underlying conditions as well. Some of the miRNAs targeting the COVID-19 genome showed low abundance (down-regulated) with some underlying conditions (Table 7). For example, miR-15b-5p (Coronary artery disease) [37], miR-15a-5p (Kidney disease) [38], miR-520c-3p (obesity/diabetes) [39], miR-30e-3p (Myocardial Injury) [40], miR-23c (hepatocellular carcinoma) [41], miR-30d-5p (non-small cell lung cancer) [42], miR-4684-3p (colorectal cancer) 43], and miR-518a-5p (Gastrointestinal stromal tumors) [44] down-regulated in pathophysiological condition.
To identify the biological relevance of COVID-19 targeting human cellular miRNAs, KEGG pathway annotation, and GO analysis was performed on target gene pools. KEGG annotation data showed that COVID-19 targeting human miRNAs plays an important role in various cancer signaling, Hippo signaling pathways, cardiac (e.g cardiomyopathy, Adrenergic signaling in cardiomyocytes), cell cycle, FoxO and number of other signaling pathways (Table.5). Furthermore, GO pathway annotation analysis showed that the COVID-19 targeting human miRNAs involved in important signaling dysregulated during age-related complications, such as biosynthesis, metabolic, cellular protein modification, and cellular component assembly. Most importantly, we also found that signaling related to immune response, Fcepsilon receptor signaling pathway, viral process, and symbiosis, encompassing mutualism through parasitism signaling affected by the COVID-19 targeting human cellular miRNAs (Table.6). The signaling mentioned above is important to fight against viral infections [45][46][47][48]. Both KEGG and GO pathway analysis revealed that COVID-19 targeting human cellular miRNAs are involved in the number of age-related complications. Impaired GO and KEGG pathways may be due to low abundance miRNA and might be one of the contributing factors for the increasing severity and mortality of the COVID-19 infection in aged individuals and with other underlying conditions.
The dual role of host cellular miRNAs cannot be denied in the viral replication or inhibition. Host miRNAs can have both functions, antiviral, which is beneficial to host, or miRNA-viral genome interaction can increase stability and beneficiary for viral propagation [18,20,22]. In a viral infection, we hypothesized that host cellular miRNAs predominantly have antiviral activity. Though possible, positive regulation of viral genome translation by miRNAs is less likely. Our data and hypothesis are based on in silico analysis and realize that not all miRNAs targets will be real. Considering the many miRNAs targets identified in our in-silico analysis, even if 5% of total miRNAs targets are real (based on target score), it is possible that these could be used for therapeutic purposes. Further, in vitro and in vivo studies will be needed to validate some of these targets. However, in view of the current lack of effective treatment strategies, innovative approaches need to be considered.