key: cord-0924228-6299fbf6
authors: Zeng, Qiqi; Qi, Xin; Ma, Junpeng; Hu, Fang; Wang, Xiaorui; Qin, Hongyu; Li, Mengyang; Huang, Shaoxin; Yang, Yong; Li, Yixin; Bai, Han; Jiang, Meng; Ren, Doudou; Kang, Ye; Zhao, Yang; Chen, Xiaobei; Ding, Xi; Ye, Di; Wang, Yankui; Jiang, Jianguo; Li, Dong; Chen, Xi; Hu, Ke; Zhang, Binghong; Shi, Bingyin; Zhang, Chengsheng
title: Distinct miRNAs associated with various clinical presentations of SARS-CoV-2 infection
date: 2022-04-27
journal: iScience
DOI: 10.1016/j.isci.2022.104309
sha: cf2d4680f6521684049081a576e2ea729eae9a48
doc_id: 924228
cord_uid: 6299fbf6

MicroRNAs (miRNAs) have been shown to play important roles in viral infections, but their associations with SARS-CoV-2 infection remain poorly understood. Here we detected 85 differentially expressed miRNAs (DE-miRNAs) from 2,336 known and 361 novel miRNAs that were identified in 233 plasma samples from 61 healthy controls and 116 COVID-19 patients using the high throughput sequencing and computational analysis. These DE-miRNAs were associated with SASR-CoV-2 infection, disease severity, and viral persistence in the COVID-19 patients, respectively. Gene ontology and KEGG pathway analyses of the DE-miRNAs revealed their connections to viral infections, immune responses, and lung diseases. Finally, we established a machine learning model using the DE-miRNAs between various groups for classification of COVID-19 cases with different clinical presentations. Our findings may help understand the contribution of miRNAs to the pathogenesis of COVID-19 and identify potential biomarkers and molecular targets for diagnosis and treatment of SARS-CoV-2 infection.

The have been conducted to understand the differences between the moderate disease (MD) and the severe 65 disease (SD), the AS remains poorly characterized. In addition, some of the patients were tested 66 positive for the nucleic acid of SARS-CoV-2 by RT-PCR but did not become negative for a longer 67 period of time (≥60 days, herein designated as long-term nucleic acid test positive, LTNP) whereas 68 others turned into negative for the viral nucleic acid in a shorter period of time (≤45 days, designated 69 as short-term nucleic acid test positive, STNP). However, the underlying mechanisms responsible for 70 the LTNP and STNP remain unknown. 71

MicroRNAs (miRNAs) are small single-stranded RNA molecules with a length of 18 to 28 72 nucleotides and have been suggested to play important roles in a variety of physiological and 73 pathological processes in animals and humans, including development, immune responses, 74

inflammation and apoptosis (Bartel, 2004) . In particular, miRNAs were involved in the replication 75

process of a number of viruses, such as hepatitis C virus, HIV-1, and influenza viruses (Lee et al., presentations. We revealed 85 differentially expressed miRNAs that were associated with the SASR-84

CoV-2 infections, disease severity and viral persistence in the COVID-19 patients, such as hsa-miR-85 370-3p, hsa-miR-146a-3p, hsa-miR-885-5p, hsa-miR-214-3p and hsa-miR-10b-5p. We also identified 86 a panel of miRNAs that may directly target the viral genes of SARS-CoV-2. Moreover, we detected a 87 number of miRNAs that may target the cellular genes involved in the life cycle of SARS-CoV-2, 88

including ACE2, AXL, TMEM106B and TOMM70. The gene ontology and KEGG pathway analyses 89 of the significant miRNAs revealed their connections to immune responses, viral infections, and lung 90

diseases. Finally, we established and tested a machine learning model using the differentially 91 expressed miRNAs between various compared groups for classification of COVID-19 cases with 92 different clinical presentations. Our findings may help understand the contribution of miRNAs to the 93 pathogenesis of COVID-19 and identify potential biomarkers and molecular targets for the diagnosis 94

and treatment of SARS-CoV-2 infection. 

In this study, we first identified 2,336 known and 361 novel miRNAs in 233 plasma samples from 61 99 healthy controls (HC) and 116 SARS-CoV-2 infected subjects (COV) by high throughput sequencing 100

( Figure 1A and Supplementary Data 2-3). Of note, we identified 2,336 of the 2656 known miRNAs 101

(approximately 88%) in the current database (http://www.mirbase.org/ftp.shtml), suggesting that our 102 experimental procedures and the analytical pipelines were robust and well performed. We 103

subsequently performed a series of computational and statistical analyses on these miRNAs to identify 104 the target genes (both the viral and cellular genes), the differentially expressed miRNAs (hereinafter 105 referred to as DE-miRNAs) between various groups ( Figure 1A suggesting that these DE-miRNAs may play important roles in the pathogenesis of COVID-19. 114

We then compared the miRNA expression levels between the SARS-CoV-2 infected individuals 116

(COV) and the healthy control (HC), and identified 112 up-regulated (fold change>1.5, P≤0.05) and 117 44 down-regulated DE-miRNAs (fold change<0.67, P≤0.05) (Figure 2A ). To improve the accuracy 118

and reliability of the DE-miRNA candidates, we removed the DE-miRNAs with the average 119 expression level lower than 10 copies and retained 24 DE-miRNAs (Supplementary Data 5). Further 120 analysis suggested that 18, 12, 17 of the DE-miRNAs were associated with virus infections, lung 121 diseases and/or immune responses, respectively ( Figure 2B ). Our data also indicate that the age and 122

gender of the study subjects had no significant influence on the expression level and function of the 123 DE-miRNAs ( Figure 2B ). A total of 3108 target genes were mapped to these DE-miRNAs (e.g., 124

CD40 for hsa-miR-370-3p and CXCL8 for hsa-miR-302a-3p) through the target gene prediction 125 analysis (Supplementary Data 6). Moreover, 936 significant GO terms and 59 KEGG pathways were 126 derived from these targeted genes (Supplementary Data 7-8). Here, we displayed the top 20 GO terms 127 (Figure. 2C) and KEGG pathways ( Figure 2D ), which were associated with immune responses (T cell 128 homeostasis and interleukin-2 production), virus infection (Hepatitis B), lung diseases (non-small cell 129 lung cancer) and other processes (HIF-1 signaling pathway), respectively. Furthermore, to test the 130 hypothesis that the DE-miRNAs could be used as biomarkers for diagnosis of SARS-CoV-2 infection, 131

we built an XGBoost machine learning model based on miRNA expression data from the 61 HC and 132

116 COVID-19 patients, leading to prioritization of eleven important miRNA variables ( Figure 2E ). 133

This model reached an area under curve (AUC) of 0.9258 and correctly classified over 83% of COV 134

and 87% of HC subjects in the training set ( Figure 2F ). In the testing set, 24 of 29 COV (83%) and 12 135

of 15 HC (80%) were correctly classified by this model ( Figure 2G ). Finally, we identified four DE-136 miRNAs (i.e., has-miR-34a-5p, has-miR-370-3p, has-miR-193a-5p_R-1 and has-miR-12136-137

P3_1ss19TC) by combining the differential expression data with the XGBoost machine learning, 138

which could be used as potential biomarkers to distinguish the SARS-CoV-2 infected patients (COV) 139 from the uninfected healthy individuals (HC) ( Figure 2H ). 140 J o u r n a l P r e -p r o o f 5

Similar to the analysis described above, we compared the miRNA expression levels between the 142 asymptomatic infected individuals (AS) and the healthy control (HC), and identified 26 up-regulated 143

and 93 down-regulated DE-miRNAs ( Figure 3A ). There are 24 DE-miRNAs left after removing the 144 ones with the average expression level lower than 10 copies (Supplementary Data 5). Further analysis 145

suggested that 13, 8, 20 of the DE-miRNAs were associated with virus infections, lung diseases 146 and/or immune responses, respectively ( Figure 3B ). Our data also indicate that the age and gender of 147 the study subjects had no significant influence on the expression level and function of the DE-148 miRNAs ( Figure 3B ). A total of 3184 target genes were mapped to these DE-miRNAs through the 149 target gene prediction analysis (Supplementary Data 6). Moreover, 952 significant GO terms and 71 150 KEGG pathways were derived from these targeted genes (Supplementary Data 7-8). Here, we 151 displayed the top 20 GO terms ( Figure 3C ) and KEGG pathways ( Figure 3D ), which were associated 152 with immune responses, virus infection, lung diseases, and other processes, respectively. We also built 153

an XGBoost machine learning model based on miRNA expression data from the 61 HC and 16 AS 154

patients, leading to prioritization of nine important miRNA variables ( Figure 3E ). This model reached 155

an area under curve (AUC) of 0.999 and correctly classified 100% of the AS and 96% of the HC 156 subjects in the training set ( Figure 3F ). In the testing set, 3 of 4 COV (75%) and 13 of 1 H4C (93%) 157

were correctly classified by this model ( Figure 3G ). Of note, the lower percentage of correct 158

classification of the AS subjects in the testing set was most likely caused by the smaller sample size.

Finally, we also identified four DE-miRNAs (has-miR-423-5p, has-miR-23a-5p, has-miR-146a-160

3p_2R-2 and has-miR-1-3p) by combining the differential expression data with the XGBoost machine 161

learning, which could be used as potential biomarkers to distinguish the AS from the HC ( Figure 3H ). 162

When we compared the miRNA expression levels in the SD vs MD group using the same cutoff value 164 described above (i.e. fold change>1.5 and P≤0.05 for the up-regulated; fold change<0.67, and P≤0.05 165

for the down-regulated DE-miRNAs), we only observed one significant DE-miRNA (has-miR-214-166 3p_L+1R_4). Therefore, we adjusted the fold change>1.2 for the up-regulated whereas the fold 167 change<0.83 for the down-regulated miRNAs. As a result, we identified 47 up-regulated and 37 168 down-regulated DE-miRNAs ( Figure 4A ). There are 17 DE-miRNAs left after removing the ones 169

with the average expression level lower than 10 copies (Supplementary Data 5). Further analysis 170 suggested that 7, 8, 17 of the DE-miRNAs were associated with virus infections, lung diseases and/or 171 immune responses, respectively ( Figure 4B ). Our data also indicate that the age and gender of the 172 study subjects had no significant influence on the expression level and function of the DE-miRNAs 173 ( Figure 4B ). A total of 2626 target genes were mapped to these DE-miRNAs through the target gene 174 prediction analysis (Supplementary Data 6). Moreover, 901 significant GO terms and 39 KEGG 175 pathways were derived from these targeted genes (Supplementary Data 7-8). Here, we displayed the 176 top 20 GO terms ( Figure 4C ) and KEGG pathways ( Figure 4D ), which were associated with immune 177 responses, virus infection, lung diseases, and other processes, respectively. We also built an XGBoost 178 machine learning model based on miRNA expression data from the 48 SD and 52 MD patients, 179

leading to prioritization of nine important miRNA variables ( Figure 4E ). This model reached an area 180 under curve (AUC) of 0.996 and correctly classified 100% of the SD and 97% of the MD patients in 181

the training set ( Figure 4F ). In the testing set, 10 of 12 SD (83%) and 9 of 13 MD (69%) were 182

correctly classified by this model ( Figure 4G ). Of note, the lower percentage of correct classification 183 of the SD and MD patients in the testing set was most likely caused by the smaller sample size. 184

Finally, we identified five DE-miRNAs (i.e., has-miR-214-3p_L+1R_4, has-miR-143-3p, has-miR-185 J o u r n a l P r e -p r o o f 6 224-5p_L-1, has-miR-452-3p_R+2 and hsa-miR-625-5p) by combining the differential expression 186 data with the xgoost machine learning, which could be used as potential biomarkers to distinguish the 187 SD from the MD patients ( Figure 4H ). Furthermore, we examined the dynamic changes of the DE-188 miRNA expression levels in two of the patients with blood samples collected at multiple time points, 189

including three samples collected from one fatal patient (COV012, COV031, COV042 and COV077) 190

and three samples collected from a recovered patient (COV011, COV041 and COV078). Since the 191 expression level of hsa-miR-122-5p_ R-1 in SD was significantly higher than that in MD, suggesting 192 that this miRNA could be positively correlated with the disease progression. Indeed, the level of hsa-193 miR-122-5p_R-1 in the fatal patient continuously increased as the disease worsened，whereas its 194 expression restored to the normal level in the recovered patient ( Figure 4I ). On the other hand, the 195 expression level of hsa-miR-224-5p_L-1 exhibited opposite changes in these two patients ( Figure 4J ), 196

suggesting that this miRNA could be negatively correlated with the disease progression. In this study, 197

we also completed similar analysis on other compared groups ( Figure 1B 

We compared the miRNA alternations between the LTNP and STNP groups and identified 78 up-202 regulated and 37 down-regulated DE-miRNAs ( Figure 5A ). There are 20 DE-miRNAs left after 203

removing the ones with the average expression level lower than 10 copies (Supplementary Data 5).

Further analysis suggested that 10, 8, 11 of the DE-miRNAs were associated with virus infections, 205 lung diseases and/or immune responses, respectively ( Figure 5B ). Our data also indicate that the age 206 and gender of the study subjects had no significant influence on the expression level and function of 207 the DE-miRNAs ( Figure 5B ). A total of 1848 target genes were mapped to these DE-miRNAs through 208 the target gene prediction analysis (Supplementary Data 6). Moreover, 594 significant GO terms and 209 44 KEGG pathways were derived from these targeted genes (Supplementary Data 7-8). Here, we 210 displayed the top 20 GO terms ( Figure 5C ) and KEGG pathways ( Figure 5D ), which were associated 211 with immune responses, virus infection, lung diseases, and other processes, respectively. We also built 212

an XGBoost machine learning model based on miRNA expression data from the 30 LTNP and 33 213 STNP patients, leading to prioritization of 16 important miRNA variables ( Figure 5E ). This model 214

reached an area under curve (AUC) of 0.977 and correctly classified 100% of the LTNP and 92% of 215 the STNP patients in the training set ( Figure 5F ). In the testing set, 5 of 7 LTNP (71%) and 7 of 8 216 STNP (88%) were correctly classified by this model ( Figure 5G ). Of note, the lower percentage of 217 correct classification of the LTNP and STNP patients in the testing set was most likely caused by the 218 smaller sample size. Finally, we identified five DE-miRNAs (has-miR-429, has-miR-574-5p, has-219 miR-483-5p, has-miR-95-3p_R-1 and hsa-miR-378i_R+1_1ss9AT) by combining the differential 220 expression data with the XGBoost machine learning, which could be used as potential biomarkers to 221 distinguish the LTNP from the STNP patients ( Figure 5H ). Moreover, 9 of the 20 DE-miRNAs were 222 positively or negatively correlated with the length (days) of the nucleic acid test positive for SARS-223

CoV-2 (Supplementary Figure. (https://www.genecards.org), we have identified a panel of miRNAs that may target the cellular genes 232

involved in the life cycle of SASR-CoV-2 replication including the entry and post-entry events, such 233 as miRNA let-7b targeting ACE2 receptor, hsa-miR-1-3p targeting AXL receptor, hsa-miR-122-5p 234 targeting ADAM17, hsa-miR-485-5p targeting DC-SIGN, hsa-miR-424-5p targeting furin, hsa-miR-235

1185-1-3p targeting HAT(HAT1), hsa-miR-183-5p targeting integrin (ITGB1), hsa-miR-1-3p 236

targeting Neuropilin-1 (NRP1), hsa-miR-485-5p targeting TMPRSS4, hsa-miR-34a-5p targeting AR, 237

hsa-miR-423-5p targeting Cyclophilin A, hsa-miR-377-5p targeting Cathepsin B, hsa-miR-148a-5p 238 targeting TMEM106B, hsa-miR-193a-5p_R-1 targeting TOMM70 (TOM70), and hsa-miR-1-3p, 239

hsa-miR-7110-3p targeting VMP1( Figure 6B ). Our findings indicate that the cellular miRNAs may which may suppress IAV-induced cytokine storm and prevent the translation of nuclear factor-kappa 260 B (NF-κB) from the cytosol to the nucleus, respectively (Gui et al., 2015) . Both hsa-miR-302a and 261

hsa-miR-302b may target on the C-X-C motif chemokine ligand 8 (CXCL8) gene, which was 262 associated with the patients with acute respiratory distress syndrome (ARDS) by promoting neutrophil 263 migration to lung interstitium and alveolar space (Williams et al., 2017) . CXCL8 was also suggested 264

to play a vital role in the inflammatory diseases of the lungs, such as asthma and chronic obstructive 265 pulmonary disease (COPD) (Huang et al., 2017) . These findings suggested that miR-302 may have a 266

protective effect in SARS-CoV-2 infected patients through targeting at the CXCL8-linked pathways. 267

The hsa-miR-146a-3p was significantly up-regulated in AS subjects compared with HC. Previous 268 studies suggested that hsa-miR-146a-3p was a key factor in the regulation of innate immunity, viral 269

infection, inflammation and tumour development ( asymptomatic infection and serve as a biomarker for diagnosis and therapeutic approach for AS.

We also identified a number of De-miRNAs that may help differentiate the SM from AS. In 278 particular, hsa-miR-122-5p, hsa-miR-1246 and hsa-miR-885-5p were suggested to be involved in the Another important aspect of this study is the identification of 20 DE-miRNAs in the LTNP vs 299 STNP groups, nine of which were significantly correlated with the length of the viral persistence in 300 COVID-19 patients, including hsa-miR-483-5p and hsa-miR-429. The expression level of hsa-miR-301

483-5p was significantly higher in the LTNP subjects, indicating that hsa-miR-483-5p was positively 302 correlated with the disease course ( Figure 4E ). In this study, we have identified 63 miRNAs that may directly target the viral sequences of SARS-317

CoV-2 genome and a panel of miRNAs that may target the cellular genes involved in the life cycle of 

• This manuscript doesn't report original code. R is a free software environment for statistical computing and 432 graphics (https://www.r-project.org/). Machine learning using XGBoost method with the R package "xgboost" 433 (https://dl.acm.org/doi/10.1145/2939672.2939785#pill-authors__content).

• Any additional information required to reanalyze the data reported in this paper is available from the lead 435 contact upon request. 

The peripheral blood was collected into the standard EDTA-K2 Vacuette Blood Collection Tubes (Jiangsu Yuli 460 Medical Equipment Co., Ltd, China; Cat.Y09012282) and stored at room temperature or 4  C until processed.

The plasma was prepared after centrifugation of the whole blood sample at 2500 rpm for 20 minutes and stored 

The Norgen Plasma/Serum Circulating and Exosomal RNA Purification Kit (Slurry Format, Cat#51000) was 469 used for isolation of the cell-free RNA from the plasma samples inside the BSL-2 laboratory described above.

Approximately 3μg of total RNA from each sample was used for the library construction by using the NEBNext 

The DNA fragments were examined by using Qsep100TM bio-fragment analyser (Bioptic Inc., Taiwan). The 477 high throughput sequencing was performed on the Novaseq 6000 sequencer (PE150 model, Illumina).

The miRNA-seq data analysis 

To identify the differentially expressed miRNAs (DE-miRNAs), miRNA differential expression based on the 506 normalized deep-sequencing counts was analysed among the eleven groups by using Student't test ( Figure 1A ).

Nine of the eleven groups (COV vs HC, AS vs HC, SM vs HC, SM vs AS, MD vs AS, SD vs AS, SD vs MD, 508 SD22 vs MD12 and LTNP vs STNP) were analysed using the independent samples' t-test ( Figure 1B) , whereas 509 two of them (MD12 vs MD-R and SD22 vs SD-R) were analysed using the paired t-test. The threshold for 510 significant difference was set to be P value 0.05 and the fold changes >1.50 or < 0.67. The miRNA with copy 511 number less than 10 was considered to be low expression level. The miRNAs with copy number higher than 10 512 in at least one sample but less than the average copy number in the data set (the sum of copy numbers / (sample 513 J o u r n a l P r e -p r o o f 15 number × total number of miRNAs) was considered to be middle expression level. The miRNA with copy 514 number higher than the average copy number in the data set was considered to be high expression level. 

. S denotes the number of significant genes that were mapped to a specific GO term 538 or pathway, B denotes the number of genes mapped to a specific GO or pathway, TB represents the total number 539 of genes mapped to all GO or pathways, TS represents the total number of the significant differentially 540 expressed genes. The flowchart for our analysis was showed in Figure 1A . 

Predicting effective microRNA target sites in 601 mammalian mRNAs

The 603 Prediction of miRNAs in SARS-CoV-2 Genomes: hsa-miR Databases Identify 7 Key miRs Linked to Host 604 Responses and Virus Pathogenicity-Related KEGG Pathways Significant for Comorbidities

MicroRNAs: genomics, biogenesis, mechanism, and function

Stressin' Sestrins take an aging fight. EMBO molecular 609 medicine 2

Occurrence and transmission potential of asymptomatic and presymptomatic SARS-CoV-2 infections: 612 A living systematic review and meta-analysis

COVID-19: fighting the invisible enemy 614 with microRNAs

XGBoost: A Scalable Tree Boosting System. Kdd'16: Proceedings of the 616 22nd Acm Sigkdd International Conference on Knowledge Discovery and Data Mining

Novel findings from determination of common expressed plasma exosomal microRNAs in patients with 620 psoriatic arthritis, psoriasis vulgaris, rheumatoid arthritis, and gouty arthritis

MicroRNA targets in 622

MiR-146a down-regulates inflammatory response by targeting TLR3 and TRAF6 in Coxsackievirus B 625 infection

Early prediction of COVID-19 severity using extracellular vesicle 628 COPB2

New coronavirus pneumonia prevention and control program

Clinical Characteristics of Coronavirus Disease 2019 in China

Mir-302c mediates influenza A virus-induced IFNβ expression by targeting NF-κB inducing kinase

Biomarkers for Diagnosis of Hepatocellular Carcinoma Post-Hepatitis C Infection in Egyptian Patients. The 640

Implications of SARS-CoV-2 Mutations for Genomic RNA 642 Structure and Host microRNA Targeting

p300 and C/EBPβ-regulated IKKβ 644 expression are involved in thrombin-induced IL-8/CXCL8 expression in human lung epithelial cells

Plasma miR-22-3p, miR-647 642b-3p and miR-885-5p as diagnostic biomarkers for pancreatic cancer

Treatment of HCV infection by targeting microRNA. The New 651 England journal of medicine

Interactions between microRNA-200 family 653 and Sestrin proteins in endometrial cancer cell lines and their significance to anoikis. Molecular and cellular 654 biochemistry 459

miRBase: from microRNA sequences to function

Delta variant triggers new phase in the pandemic

Importance of miR-125a-5p and miR-122-5p expression in 660 patients with HBV infection

The Role of MicroRNA in Pathogenesis and as Markers of HCV 662 Chronic Infection

associates with serum apolipoprotein B but not liver fibrosis markers in CHC genotype 1 infection

High expression of miR-483-5p aggravates 667 sepsis-induced acute lung injury

Downregulation of miR-214-3p May Contribute to Pathogenesis of Ulcerative 669

Colitis via Targeting STAT6

Regulation of somatic cell 671 reprogramming through inducible mir-302 expression

HOTAIR/miR-214-3p/FLOT1 axis plays an essential 674 role in the proliferation, migration, and invasion of hepatocellular carcinoma

MiR-122-5p protects against acute 677 lung injury via regulation of DUSP4/ERK signaling in pulmonary microvascular endothelial cells

Multimodal tumor suppression by miR-302 cluster in melanoma and colon cancer. The international journal of 681 biochemistry & cell biology

Cutadapt removes adapter sequences from high-throughput sequencing reads

Antitumor function of microRNA-122 against hepatocellular 685 carcinoma

Unique Circulating MicroRNA Profiles in HIV Infection

SARS-CoV-2-Specific T Cells Exhibit Phenotypic Features of Helper Function, Lack of 691 Terminal Differentiation, and High Proliferation Potential

MicroRNA-302 Cluster Downregulates Enterovirus 71-Induced Innate Immune Response by Targeting KPNA2

Hospitalization and Mortality among Black 697 Patients and White Patients with Covid-19

HPV16(+) -miRNAs in cervical cancer and 700 the anti-tumor role played by miR-5701

Inflammation potentiates miR-939 expression and packaging into small extracellular vesicles

Transmission of 2019-nCoV Infection from an Asymptomatic Contact in Germany

Identification of Novel SARS-CoV-2 Drug Targets by Host 708

MicroRNAs and Transcription Factors Co-regulatory Interaction Network Analysis. Frontiers in genetics 11

miR-483-5p plays a protective role in chronic obstructive 711 pulmonary disease

Multiple targets of miR-302 and miR-372 promote reprogramming of human fibroblasts to induced pluripotent 714 stem cells

miRNA-1246 suppresses acute lung 716 injury-induced inflammation and apoptosis via the NF-κB and Wnt/β-catenin signal pathways

Immune Response and Tumor Development

Long-term safety and efficacy of microRNA-722 targeted therapy in chronic hepatitis C patients

Effect of 724 miR-483-5p on apoptosis of lung cancer cells through targeting of RBM5

COVID-19) Dashboard

Evidence for chemokine synergy during neutrophil migration in ARDS

COVID-19: Surveillance, case investigation and epidemiological protocols

Laboratory testing for 2019 novel coronavirus (2019-nCoV) in suspected 734 human cases

Characteristics of and Important Lessons From the Coronavirus Disease 737 2019 (COVID-19) Outbreak in China: Summary of a Report of 72314 Cases From the Chinese

Disease Control and Prevention

Transcriptomic characteristics of bronchoalveolar lavage fluid and peripheral blood mononuclear cells in 741 COVID-19 patients

MiR-214-3p exacerbates kidney damages and 743 inflammation induced by hyperlipidemic pancreatitis complicated with acute renal injury

Reciprocal regulatory mechanism between miR-214-3p and FGFR1 in FGFR1-amplified lung cancer

MiRNA let-7b promotes the 749 development of hypoxic pulmonary hypertension by targeting ACE2. American journal of physiology. Lung 750 cellular and molecular physiology 316

The miRNA: 752 a small but powerful RNA for COVID-19

The Downregulation of 754

protein: hsa-miR-103a-3p, hsa-miR-139-5p, hsa-miR-15b-5p, hsa-miR-21-3p, hsa-miR-382-5p, hsa-miR-497-5p_R+1, hsa-miR-505-3p_R+2, hsa-miR-627-3p, hsa-mir-3650-p5, hsa-miR-2020_061-3p, etc. hsa-miR-382-5p, hsa-miR-214-3p_L+1R-4, hsa-miR-103a-3p, hsa-miR-107_R-4, hsa-miR-29b-3p, hsa-miR-29c-3p_R-1, hsa-miR-4288_L+1_1ss12AG, hsa-miR-4524a-5p_L-2R+2, hsa-miR-4659a-3p_L-1R+1, hsa-miR-497-5p_R+1, hsa-mir-1282-p5_1ss11GC, etc. hsa-miR-194-5p_R-1, hsa-miR-1226-3p, hsa-miR-187-5p_L-1R+2, hsa-miR-299-5p, hsa-miR-330-3p_R+2, etc. hsa-miR--503-5p_R-4, hsa-mir-548l-p3. hsa-mir-1976-p5, hsa-miR-214-3p_L+1R-4, hsa-miR-15a-5p_R-1, hsa-miR-219a-2-3p_R-2, hsa-miR-29b-1-5p

(1) ACE2 receptor: miRNA let-7b (2) AXL receptor: hsa-miR-1-3p, hsa-miR-34a-5p, hsa-miR-122-5p, hsa-miR-10b-5p;(3) ADAM17: hsa-miR-122-5p, hsa-miR-302b-3p, hsa-miR-302a-3p;(4) DC-SIGN: hsa-miR-485-5p;(5) furin: hsa-miR-424-5p;(6) HAT: hsa-miR-1185-1-3p, hsa-let-7b-3p, hsa-miR-381-3p;(7) ITGB1: hsa-miR-183-5p;(8) Neuropilin-1 (NRP1): hsa-miR-1-3p;(9) TMPRSS4: hsa-miR-485-5p;(10) AR: hsa-miR-34a-5p, hsa-miR-381-3p, hsa-miR-625-5p;(11) Cyclophilin A: hsa-miR-423-5p, hsa-miR-423-5p, hsa-miR-342-5p ;(12) Cathepsin B: hsa-miR-377-5p, hsa-miR-183-5p;(13) TMEM106B: hsa-miR-148a-5p, hsa-miR-1255a, hsa-miR-574-5p;