Expression Proles of Plasma IFN Signaling-Related miRNAs (ISR-miRNAs) at the Acute and Recovery Phase of COVID-19

and from the and recovery of 29 and 29 Expression levels of 12 IFN signaling-related miRNAs were analyzed using RT-qPCR. The receptor-binding domain (RBD) IgG antibody in the convalescent plasma samples was detected using competitive ELISA. infection, this study was conducted to analyze the expression characteristics of circulating ISR-miRNAs at the acute and convalescence phase of COVID-19 patients.

pandemic one of the worst pandemics in recent years, and certainly worse than previous coronavirus pandemics such as SARS and MERS.
Type I interferon (IFN-I) exists in vertebrates and triggers the Januskinase/signal transducer and activator of transcription (JAK/STAT) signaling pathway with subsequent induction of IFN-stimulated genes (ISGs) [3]. Mounting studies have suggested that IFN-I could affect the life cycle of the virus by regulating the expression level of microRNAs (miRNAs), which are key post-transcriptional regulators in various cellular biological processes. For example, Aboulnasret al. reported that IFN-α/β could induce the expression of miR-122 in hepatocytes. However, the reduction of miR-122 expression level could weaken the effect of IFN-a/β in inhibiting hepatitis C virus (HCV) [4]. However, many viruses could develop strategies to alter miRNA expression, thereby inhibiting the activity of IFN-I signaling pathway. The result from the sequence alignment illustrates that the presence of putative miRNA target sites for the IFN-I-induced miRNAs located in strictly conserved areas of the HCV genome. Pedersen et al. further con rmed that miR-196, miR-296, miR-351, miR-431 and miR-448 could bind to complementary sequences in the HCV genome and inhibit virus replication [5].
During the past year, several thousand studies have investigated the epidemiologic, clinical, biological and radiological characteristics of COVID-19 patients [6-8]. However, the mechanism of the new coronavirus infection has not yet been fully understood. Based on the prediction from miRDB (http://www.mirdb.org/) and miRPathDB (https://mpd.bioinf.uni-sb.de/overview.html), our recent analysis indicated that the genome of SARS-CoV-2 contains 12 candidate binding sites for IFN-I signaling-related miRNAs (ISR-miRNAs) (Table S1) [9,10]. Thus, to determine the role of ISR-miRNAs in the host response to SARS-CoV-2 infection, this study was conducted to analyze the expression characteristics of circulating ISR-miRNAs at the acute and convalescence phase of COVID-19 patients.

Materials And Methods
Study population This study was approved by the ethics committee of Huai'an Fourth Hospital (Huai'an, China), and conducted in accordance with the Declaration of Helsinki. Between January 2020 and May 2020, 29 COVID-19 patients at the acute phase of infection and 29gender and age (± 5years) matched healthy controls were recruited from Huai'an Fourth hospital. All participants signed an informed consent form and the participants coinfected with other viruses were excluded. The diagnosis of COVID-19 was based on the "New Coronavirus Pneumonia Prevention and Control Program (5th version)" published by the National Health Commission of China [11]. The demographic characteristics of acute COVID-19 patients and controls, including weight, height, and blood pressure were collected by face-to-face interview. In addition, the data about clinical signs, symptoms, and potential comorbidities were extracted retrospectively from electronic medical records. Blood sample collection and laboratory analysis Five milliliters of acute phase blood sample was collected from each patient after at least 8 hours fasting during patients' rst admission to the hospital. Three months after discharge, the patients were invited to participate in the following-up survey and the convalescence fasting blood samples were collected. In addition, ve milliliters of fasting blood samples were collected from the recruited age-and gender-matched healthy controls. All blood samples were centrifuged at 3000 g for 10 minutes at room temperature. Plasmas were separated and inactivated in a water bath at 56°C for 30 minutes, and then frozen at -80 °C for storage as quickly as possible. The routine laboratory tests, including lymphocyte (LYM) count, neutrophil (NEUT) count, white blood cell (WBC) count, platelet (PLT) count, ALT and AST were determined using commercial reagents according to the manufacturer's instructions. ISR-miRNA selection In this study, the complete genome of SARS-CoV-2 strain (NC_045512.2) was retrieved from the GenBank database and used as a reference sequence. The miRDB (http://www.mirdb.org/) software was rstly used to identify miRNAs which can target the genome of SARS-CoV-2 (NC_045512.2). The miRNAs with more than 95 of the target score were primarily included [12,13]. Then, we used miRPathDB (https://mpd.bioinf.uni-sb.de/overview.html) to identify miRNAs related to the JAK-STAT pathway [14].
Furthermore, we conducted a systematic literature review to identify the ISR-miRNAs using the following terms "JAK", "STAT", and "JAK/STAT" in PubMed. Finally, twelve miRNAs were selected as ISR-miRNAs for the present analysis: let-7c-5p, miR-15a-5p, miR-15b-5p, miR-29b-3p, miR-30b-5p, miR-146b-3p, miR-148a-3p, miR-186-5p, miR-409-3p, miR-497-5p, miR-548c-5p and miR-1246. The detailed information about the selected 12 ISR-miRNAs were shown in Table S1. Plasma RNA extraction and ISR-miRNA quantitation analysis Total RNA was isolated from 300 μL plasma with a commercial RNA extraction and puri cation kit (MACHEREYNAGEL SA, France) according to the manufacturer's protocol. To warrant consistency in the experimental procedures, exogenous cel-miR-39 was spiked into each serum sample before RNA extraction and used as an internal control for normalizing miRNA expression levels. The concentration of RNA was measured at OD260/280 by a NanoDrop ND-1000 spectrophotometer (Thermo Scienti c, Wilmington, Delaware). All real-time PCR was performed on a Q6 Real-Time System (Applied Biosystems) using the SYBR Green-based real-time detection method. The miDETECTA TrackTM miRNA qRT-PCR Starter Kit, the upstream and downstream primers for selected ISR-miRNAs quantitation were ordered from RiboBio Corporation (Guangzhou, China) [15]. As described above, cel-miR-39 miDETECTTM miRNA External Control (RiboBio, China) was selected as an endogenous control for miRNA expression analysis. In this study, the reaction system for ISR-miRNA Poly(A) tailing contained 1 μg of total small RNA, 2 μL 5X Poly(A) polymerase buffer, 1 μL Poly(A) polymerase and RNase-free water up to 10 μL. The reaction system for reverse transcription contained 4 μL RTase mix, 4 μL 5X RTase buffer, 2 μL miDETECT A TrackTM Uni-RT Primer and 10 μL Poly(A) Tailing product. 20 μL reaction system for real-time quantitative PCR contained 0.5 μL miDETECT A TrackTM miRNA Forward Primer (10μM), 0.5 μL miDETECT A TrackTM Uni-Reverse Primer (10μM), 10 μL 2X SYBR Green Mix, 0.04 μL ROX Reference Dye, 2 μL cDNA and RNase-free water. The cycle threshold (CT value) was de ned as the number of cycles required for the uorescent signal to cross the threshold. The expression of ISR-miRNA relative to cel-miR-39 miRNA was reported as dCT (ΔCT), which was calculated by subtracting the Ct of cel-miR-39 from the Ct of target ISR-miRNA. The relative quantitative of each ISR-miRNA was expressed as 2(-ΔCT), and was transformed into their natural logarithm to eliminate heteroscedasticity. The signi cance of ISR-miRNA expression between different groups was de ned as a difference of at least 2fold when compared with healthy controls. Competitive ELISA for receptor-binding domain (RBD) IgG antibody detection The competitive ELISA steps were carried out using "RBD-IgG antibody detection kit" (Beijing, China) according to the manufacturer's introduction. Brie y, 96-well Corning Costar high binding plates were coated with SARS-CoV-2 spike RBD protein at a concentration of 0.1 μg per well. For the RBD-IgG antibody measurement, 50 μL of 1:50 diluted plasma sample and 50 μL of HRP-conjugated ACE2-mFc (0.2 μg/ml) were added into each well. Meanwhile, two negative and two positive plasma controls and two blank wells incubated with HRP-conjugated ACE2-mFc were included on each plate, respectively. After incubated for 30 minutes, the plate was washed three times and TMB substrate solution was added.

Demographic and clinical characteristics of included participants
The demographic and clinical characteristics of the included COVID-19 patients were summarized in Table 1. The mean (±standard deviation) age of the patients with COVID-19 was 47.45±15.72 years, and 58.62% were male. Moreover, three, nine and one patient had pre-existing diabetes, hypertension and renal insu ciency, respectively. At the acute phase, the most common symptom at the onset of illness was fever (75.86%). In addition, the median incubation period was 5.0 days. The virus nucleic acid test turned negative approximately 7 days after admission.
Compared with the healthy controls, patients with COVID-19 presented lower LYM counts (Z=-3.86, P=0.001) and PLT counts (Z=-2.80, P=0.005) at the acute phase of disease. In addition, LYM, PLT and CD8+ T cell counts at the recovery phase were signi cantly higher than those at the acute phase. However, we did not observe the difference in other clinical parameters among the three groups.
Expression pro les of ISR-miRNAs at the acute phase of COVID-19 Changes in the relative expression of twelve miRNAs were measured and presented in Figure 1.

Expression pro les of ISR-miRNAs at the recovery phase of COVID-19
As the results shown in Figure 1, the expression levels of miR-29b-3p and miR-1246, which signi cantly elevated at the acute phase, were not different between individuals at the recovery phase and healthy controls. However, the results showed that the expression levels of miR-30b-5p, miR-409-3p, miR-497-5p and miR-548c-5p in convalescent plasma samples were signi cantly lower than those in healthy controls. In addition, the concentration of miR-186-5p in convalescent plasma samples was signi cantly higher than that in healthy controls and patients with acute infection. No signi cant difference was found in the relative expression levels of other ISR-miRNAs among the recovery individuals, patients who acute SRAS-CoV-2 infection and healthy controls.

Association of ISR-miRNAs with RBD-IgG antibody at the recovery phase
Among 28 patients who provided enough convalescent plasma samples, RBD-IgG antibodies were detected in 27 COVID-19 patients using competitive ELISA. The highest and lowest PI values were 93.0% and 41.2% (median PI: 77.5%), respectively. The potential association of circulating ISR-miRNA levels with RBD-IgG antibody response was further analyzed in the present study. As the results showed in Fig 3, the plasma level of miR-497-5p at the acute phase positively correlated to RBD-IgG antibody response (r=0.48, P=0.038). However, no correlation was observed between other ISR-miRNAs and RBD-IgG response in the analyzed patients.

Discussion
Previously, many studies have reported that ISR-miRNAs were signi cantly associated with viral infection [3][4][5]. This study revealed that the expression levels of miR-186-5p and miR-15a-5p signi cantly decreased at the acute phase of COVID-19 patients. Following in silico target prediction and pathway enrichment analyses, Zhao et al. suggested that miR-186-5p was depleted in retroviral infection. However, the increased miR-186-5p expression could inhibit HIV infection by immunoregulation and T cell regulation [19]. Moreover, Wu et al. reported that overexpressed miR-186 could inhibit the JAK/STAT signaling pathway in vitro [20]. The results from the most recent study reported that hsa-miR-15b-5p were signi cantly downregulated in hamster lung samples infected by SARS-CoV-2 [21]. Considering that the miR-15 family members (i.e., miR-15a, 15b) possess the same seed sequence and have the same target genes, the present results further suggested that downregulated expression of miR-186-5p and miR-15a-5p might be helpful for the IFN-I signaling pathway activation at the acute phase of SARS-CoV-2 infection.
In the early phase of viral infection, the host's innate immunity, including IFN-I signaling pathway, is the rst defense mechanism [22]. However, since the excessive cytokines produced by IFN-1 signaling can cause a cytokine storm and damage the body [23], the effective and precise regulation of JAK/STAT signaling activity is very important for the patients recovered from acute infection [20,[24][25][26][27]. Therefore, it is not a surprise that the expression levels of miR-30b-5p, miR-409-3p, miR-497-5p and miR-548c-5p signi cantly decreased, while miR-186-5p expression signi cantly increased in convalescent plasma samples when compared with those of healthy controls and acute infected patients. Although the exact timing during IFN-I-activated feedback regulation and control of JAK/STAT signaling is currently unclear, the present results implicated that a timely and appropriate JAK-STAT signaling regulation should be necessary and helpful for the recovery of patients with SARS-CoV-2 infection.
A large number of studies have con rmed that the interaction between RBD located at the spike protein of SARS-CoV-2 and the receptor ACE2 on host cells is essential for viral entry [28,29]. Therefore, antibodies against RBD at the recovery phase of COVID-19 present neutralizing activity because they can block the interaction between ACE2 and viral spike protein [30]. Previously, Premkumar et al. reported that antibodies targeting RBD accounted for more than 90% of neutralizing activity in the convalescent serum [31]. In the present study, competitive ELISA results showed that 27/28 patients developed RBD-IgG antibodies at the recovery phase. Moreover, results of the present study suggested that the plasma level of miR-497-5p at the acute phase positively correlated to RBD-IgG antibody response at the recovery phase, indicating that miR-497-5p might serve as a candidate ISR-miRNA for the prediction of SARS-CoV-2 neutralization antibody.
This study has several limitations. First, considering that only 29 patients were included in this analysis, caution should be taken when interpreting the present ndings. Moreover, the difference in plasma ISR-miRNAs between patients with mild and severe infection could not be explored because of the small sample size. Second, although we, for the rst time, reported that miR-29b-3p, miR-497-5p, miR-1246, miR-186-5p and miR-15a-5p were signi cantly associated with acute SARS-CoV-2 infection, more miRNAs related to JAK/STAT pathway still need to be thoroughly analyzed in the future. Moreover, to understand the questions behind the observed associations, the role of ISR-miRNAs involved in the pathogenesis of COVID-19 should be investigated in both in vivo and in vitro studies. Finally, since the disease progression and antibody response might be in uenced by various risk factors (age, comorbidity disease such as hypertension and diabetes) [32][33][34], the effects of the ISR-miRNAs and interactions of other risk factors on SARS-CoV-2 infection should also be carefully veri ed in the future.

Conclusions
In summary, this study is the rst to report that appropriate regulation of ISR-miRNA expression plays a critical role during both acute and recovery phases of SARS-CoV-2 infection. Furthermore, the circulating miR-497-5p level was positively correlated to RBD-IgG antibody response in COVID-19 patients. In the future, further studies with large study samples are needed to understand the biological signi cance of ISR-miRNAs during SARS-CoV-2 infection.   (7):1062-79. Figure 1 Fold changes of relative expression of twelve ISR-miRNAs in patients at the acute and recovery phase of COVID-19. The signi cance of ISR-miRNAs expression was de ned as a difference of at least 2-fold when compared with healthy controls.