The miRNA neuroinflammatory biomarkers in COVID-19 patients with different severity of illness

Introduction The expression of specific miRNAs and their mRNA targets are changed in infectious disease. The aim of this study was to analyze the expression of pro-neuroinflammatory miRNAs, anti-neuroinflammatory miRNAs, and their mRNA targets in the serum of COVID-19 patients with different grades. Methods COVID-19 patients with different grades were enrolled in this study and the expression of pro-neuroinflammatory miRNAs, anti-neuroinflammatory miRNAs, and their target mRNAs was analyzed by q-PCR. Results The relative expression of anti- neuroinflammatory miRNAs (mir-21, mir-124, and mir-146a) was decreased and the relative expression of their target mRNAs (IL-12p53, Stat3, and TRAF6) was increased. Also, the relative expression of pro-neuroinflammatory miRNAs (mir-326, mir-155, and mir-27b) was increased and the relative expression of their target mRNA (PPARS, SOCS1, and CEBPA) was decreased in COVID-19 patients with increase of disease grade. A negative significant correlation was seen between mir-21 and IL-12p53 mRNA, mir-124 and Stat3 mRNA, mir-146a and TRAF6 mRNA, mir-27b and PPARS mRNA, mir-155 and SOCS1 mRNA, and between mir-326 and CEBPA mRNA in COVID-19 patients (P < 0.05). Conclusions This study showed that the relative expression of anti- neuroinflammatory miRNAs was decreased and the relative expression of their targeted mRNAs was increased in COVID-19 patients from asymptomatic to critical illness. Also, this study showed that the relative expression of pro-neuroinflammatory miRNAs was increased and the relative expression of their targeted mRNA was decreased in COVID-19 patients from asymptomatic to critical illness.


Introduction
Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), known as COVID-19, is a new infectious disease first seen in late December 2019 in Wuhan, China, and similar outbreaks occurred in the hospital in neighboring countries. Major clinical symptoms include fever, dry cough, diarrhea, muscle aches, pneumonia, and in severe cases death. 1,2 COVID-19 also is associated with neurological manifestations such as encephalopathy and encephalomyelitis, ischemic stroke and intracerebral hemorrhage, anosmia, neuromuscular diseases, and neuroinflammation diseases. 3 Since COVID-19 is a new disease, complete information on its etiology, cellular mechanisms, and possible risk factors is not available. COVID-19 may be similar to recent acute respiratory syndromes, such as SARS and MERS. 4 Theoretically, after the SARS-CoV-2 enters the human body, different types of immune cells are stimulated. These cells trigger the proper immune response by producing different cytokines, chemokines, antibodies, etc. SARS-CoV-2 can infect the CNS following the entry of the virus into the nose or the eye. The viral particles are transmitted to the olfactory bulb and then to the brainstem, and then all parts of the brain. 5 In addition to the direct attack of nerve cells, the SARS-CoV-2 can systematically cross the BBB through the blood vessels and reach the CNS. The main feature of systemic infection in COVID-19 is the massive increase in pro-inflammatory factors in the blood, which is described as a ''cytokine stor''. 6 This leads to BBB permeability and transmission of SARS-CoV-2 and peripheral immune cells. Once the coronavirus enters the CNS, it is the turn of the astrocytes and microglia to fight it. The immune response of astrocytes and microglia is regulated by different microRNAs (miRNAs). Previous studies showed inflammatory processes in CNS are guided by pro-neuroinflammatory miRNAs (such as mir-155, mir-27b, mir-326) and anti-neuroinflammatory miRNAs (such as mir-146a, mir-124, and mir-21). 7,8 This study aimed to analyze the expression of proneuroinflammatory miRNAs, anti-neuroinflammatory miR-NAs, and their mRNA targets in the serum of COVID-19 patients with different grades.

Materials
All primers were provided from Bioneer, South Korea. MirPremier microRNA isolation kit was sourced from Sigma-Aldrich, USA. Mir-X miRNA First-Strand Synthesis kit and cDNA matermix were purchased from Takara bio inc, USA. SYBR ® Green Real-Time Master Mix was from Invitrogen, UK.

Bioinformatics
In this study, to determine the miRNAs associated with the COVID-19, we used online bioinformatics Softwares. 9 In the first step, mirTarP (https://mcube. nju.edu.cn/jwang/mirTar/docs/mirTar/) was used to the list of appropriate miRNAs. 10,11 In the second step, to reduce the number of selected miRNAs, we selected some limited pro-neuroinflammatory and anti-neuroinflammatory miRNAs that were previously reported in other studies. In the third step, the miRDB online database (http://mirdb.org/) was used to find the target of selected miRNAs. 12 Target genes of the differentially regulated miRNAs were predicted using the mirPath tool (version 3.0). 13 KEGG molecular pathways were also retrieved using the same tool. 14 Pathways and processes regulated with P values lower than 0.05 were considered significant.

Small RNA isolation, first-strand cDNA synthesis, and quantification of miRNAs and mRNAs by qPCR
Here, Small RNA was isolated from blood samples using mir-Premier microRNA isolation kit. Briefly, 1000 L of the lysis buffer was added to 100 L of serum samples, vortexed for 2 min, and incubated at 55 • C for 5 min. The samples were then centrifuged for 5 min at 14,000 × g to remove cellular debris, genomic DNA, and large RNA. The lysate supernatant was filtered through the filtration column and binding column. After binding, the column was first washed with 700 L of 100% ethanol and centrifuged at 14,000 × g for 30 s and again the flow-through was discarded. The second wash was done by adding 500 l of binding solution into the column and centrifuged at maximum speed (14,000 × g) for 1 min. Subsequently, 500 ml of the ethanol-diluted wash solution 2 was added to the column for a third wash. After centrifugation at maximum speed (14,000 × g) for 30 s, the flow-through was discarded. Next, the column was dried by centrifuging at maximum speed (14,000 × g) for 1 min. The column-tube assembly was carefully removed from the centrifuge to avoid splashing of the residual flow-through liquid to the dried column. Small RNA was eluted from the column using 50 ml elution solution and by centrifugation at 16,000 × g and the process was repeated to improve small RNA yield. The purity of the RNA samples was analyzed by NanoDrop ND-1000 UV-VIS spectrophotometer. The A260 nm/A280 nm ratio of all samples was between 1.8 and 2.1. The quantity of RNA samples was analyzed by agarose gel electrophoretic separation. For first-strand cDNA synthesis, small RNAs were polyadenylated and reverse transcribed using the Mir-X miRNA First-Strand Synthesis kit. Briefly, 5 l mRQ buffer (2×), 5 g RNA and 1.25 l mRQ enzyme was mixed in a reaction volume of 10 l and incubated in a thermocycler for 1 h at 37 • C, then terminate at 85 • C for 5 min to inactivate the enzymes. After reverse transcription, the cDNA was diluted. For quantification of miRNA by qPCR, Mir-X miRNA qPCR SYBR Kit was used. Briefly, 10 l PCR reaction mixture was prepared to comprise of 1× SYBR advantage premix, 0.2 mM of both forward and reverse primers, and 50 ng of the first-strand cDNA. qPCR reactions were incubated in a 96 well plate at 95 • C for 2 min, followed by 40 cycles of 95 • C for 10 s and 60 • C for 20 s. Amplification cycles were followed by a melting curve analysis ranging from 56 to 95 • C. Finally, the threshold cycle (Ct) values were recorded. For mRNA, total RNA was extracted using an RNA extraction kit. Then, the cDNA was synthesized in the presence of the superscript enzyme and hexamers. For real-time PCR, 2 L of cDNA, 2 L of forward primer, and 2 L of reverse primer of each gene were added to 10 L of SYBR ® Green Real-Time Master Mix. In this study, the relative expression of mir-155, mir-27b, mir-326, mir-124, mir-146a, mir-21, IL-12p53, Stat3, TRAF6, PPARS, SOCS1, and CEBPA was analyzed. The expression of microRNA and mRNA was normalized to RNU 48 and GAPDH, respectively.

Statistical analysis
All data were reported as the mean ± standard deviation. To find significant differences between groups, a one-way ANOVA method was applied. A P-value of less than 0.05 was considered statistically significant. Also, Spearman's correlation coefficient was used to correlate the expression of miRNAs and their mRNA targets.

Bioinformatics analysis
Five-top human mRNA targets for pro-neuroinflammatory miRNAs (mir-155, mir-27b, and mir-326) and anti-neuroinflammatory miRNAs (mir-124, mir-146a, and mir-21) are shown in Table 2. It should be noted that each miRNA has many targets, but here we have listed only 5 important mRNA targets with the highest target score. Theoretically, all of them can be affected by pro-neuroinflammatory and anti-neuroinflammatory miRNAs. Based on KEGG database (Table 3), we found that both pro-neuroinflammatory miRNAs and anti-neuroinflammatory miRNAs are significantly enriched in important cellular pathways, such as PI3K-Akt signaling pathway, mRNA surveillance pathway, mTOR signaling pathway, MAPK signaling pathway, Wnt signaling pathway, and AMPK signaling pathway.

Discussion
This study showed that the relative expression of antineuroinflammatory miRNAs (mir-21, mir-124, and mir-146a) The correlation between the relative expression of mir-21 and IL-12p53 mRNA (a), mir-124 and Stat3 mRNA (b), and mir-146a and TRAF6 mRNA (c) in COVID-19 patients with different grades. Spearman's correlation coefficient was used to correlate these parameters.
was decreased and the relative expression of their target mRNAs (IL-12p53, Stat3, and TRAF6) was increased in COVID-19 patients with increase of disease grade from asymptomatic to critical illness. Also, this study showed that the relative expression of pro-neuroinflammatory miRNAs (mir-326, mir-155, and mir-27b) was increased and the relative expression of their target mRNA (PPARS, SOCS1, and CEBPA) was decreased in COVID-19 patients with increase of disease grade. A negative significant correlation was seen between each miRNA and its target mRNA. Based on bioinformatics analysis, some important pathways are affected by these pro-neuroinflammatory and anti-neuroinflammatory miRNAs, including PI3K-Akt, mRNA surveillance, mTOR, MAPK, Wnt, and AMPK signaling pathways. What we have found is that in patients with high severity of illness, the expression of pro-inflammatory miRNAs is increased, and conversely, the expression of anti-inflammatory miRNAs is decreased. Of course, it is clear that this situation follows a cytokine storm. Unfortunately, we have to say that this special condition not only causes serious damage to the brain but also causes damage to several organs and leads to multiple organ failure. We think that when immune cells are highly stimulated, cytokines and miRNAs can travel through the bloodstream to the whole body. This phenomenon has been mentioned by some researchers. 15,16 Mir-155 is a central pro-inflammatory mediator in CNS by NF-B dependent TLR signaling. It is synthesized inside macrophages and microglia. [17][18][19] mir-155 targets anti-inflammatory regulators such as SOCS1, 17,19 SHIP1, 20 C/EBP-ˇ2 1 and IL13R˛1. 22  transcription factor p53, and it targets the c-Maf transcription factor, which induces differentiation and inflammatory responses. 23 Mir-146a is an anti-inflammatory regulator in nerve cells, microglia, and astrocytes. It activates by NF-B dependent TLR signaling. 24,25 The Mir-146a targets MyD88 signaling complex, including IRAK1 and TRAF6, and acts as an NF-B signaling regulator. In addition, Mir-146a targets other pro-inflammatory mediators including STAT-1, 26,27 IRF-5 27 and CFH. 28,29 The polarization of macrophages and microglia are also altered by mir-146a. 30 mir-124 is also an anti-inflammatory miRNA and has a major role in neuronal differentiation 31 and is highly expressed in microglia under normal conditions, but is not expressed in peripheral macrophages. 32 Expression of mir-124 in microglia leads to anti-inflammatory effects 33 Figure 6 The correlation between the relative expression of mir-27b and PPARS mRNAs (a), mir-155 and SOCS1 mRNAs (b), and mir-326 and CEBPA mRNA (c) in COVID-19 patients with different grades. Spearman's correlation coefficient was used to correlate these parameters.
that mir-124 has anti-inflammatory activity by reducing inflammatory mediators and limiting microglia to activity. The role of mir-21 is very prominent in different types of CNS cells such as microglia 35 and astrocytes, 36 neurons, 37 and oligodendrocytes. 38 Mir-21 is an anti-inflammatory regulator activated by TLR signaling. This induces the expression of the anti-inflammatory cytokine such as IL-10. 39 In addition, mir-21 decreases TNF-␣ secretion in macrophages and microglia. 40 mir-27b targets an antiinflammatory transcriptional activator, PPAR-; in human macrophages, this interaction blocks the induction of an anti-inflammatory phenotype. Inhibiting mir-27b also limits inflammatory signaling. It leads to produce inflammatory cytokines including IL-6 and TNF-␣. 41 mir-326 is another pro-inflammatory miRNAs and can affect on differentiation of IL-17-producing Th17 cells. It was found that silencing mir-326 reduced EAE pathology. 42 miRNAs have a cumulative effect on neuronal signaling and act together in inflammatory or anti-inflammatory pathways. For example, both mir-146a and mir-21 target different components of the TLR/MyD88/NF-B and JAK-STAT pathways. 26,28 In contrast, mir-155, mir-27b, and mir-326 activate the JAK-STAT pathway by targeting SOCS1 and SHIP1. 19 It is interesting to note that miRNAs are also present in extracellular exosomes and can participate in intercellular communication. 43 For example, mir-124, mir-21, and let-7 are found in exosomes and stimulate and regulate adjacent cells such as microglia and contribute to inflammatory signaling. 44 One of main limitations of this study was to find and to collect COVID-19 patients with no comorbidities, no inflammatory autoimmune diseases, and no drug treatments. Theoretically, these factors can affect the expression of mRNAs and miRNAs. Second limitation was that we did not include COVID-19 patients caused by different variants of SARS-COV-2. Here, only COVID-19 patients with English variant (Lineage B.1.1.7) were included. We think that the expression of mRNAs and miRNAs may also be affected by virus variants. The third limitation was that we evaluated only 6 neuroinflammatory miRNAs in COVID-19 patients and it is suggested that other neuroinflammatory miRNAs could be studied in future studies.

Conclusions
This study showed that the relative expression of antineuroinflammatory miRNAs (mir-21, mir-124, and mir-146a) was decreased and the relative expression of their mRNAs (IL-12p53, Stat3, and TRAF6) was increased in COVID-19 patients from asymptomatic to critical illness. Also, this study showed that the relative expression of proneuroinflammatory miRNAs (mir-326, mir-155, and mir-27b) was increased and the relative expression of their mRNA (PPARS, SOCS1, and CEBPA) was decreased in COVID-19 patients from asymptomatic to critical illness. A negative significant correlation was seen between mir-21 and IL-12p53 mRNA, mir-124 and Stat3, between mir-146a and TRAF6, between mir-27b and PPARS, between mir-155 and SOCS1, and between mir-326 and CEBPA mRNA in COVID-19 patients (P < 0.05).