Identifying microRNA expression alterations in erythrocytes, lymphocytes, and monocytes during severe COVID-19

Full Text

INTRODUCTION

COVID-19 is a complex disease that alters the expression profiles of many molecules in blood cells [1]. The vast majority of studies in this field compare the expression profiles of protein-synthesizing genes. Non-coding RNAs – long non-coding RNAs, piRNAs [2], and especially miRNAs – have previously been shown to play a key regulatory role. In particular, miRNAs are key immune response regulators, influencing the maturation, proliferation, differentiation, and activation of immune cells, antibody production, as well as the release of inflammatory mediators [3]. The defining observations about changes in miRNA levels are typically made in blood plasma and serum. A whole spectrum of these molecules has been acknowledged as markers of severe COVID-19 [4]. However, it remains unclear from which cells the miRNAs are released into plasma. Of particular importance is also the changes in the intracellular signaling processes that occur under the influence of noncoding RNAs. Changes in long non-coding RNAs in monocytes and lymphocytes during COVID-19 have been demonstrated earlier [5, 6]. Nevertheless, the influence of actively synthesizing cells is not the only factor that impacts the plasma miRNA levels. The miRNA profiles in erythrocytes were shown to undergo changes during sickle cell anemia [7], paroxysmal nocturnal hemoglobinuria [8], as well as Parkinson’s disease [9]. Additionally, it has been established that these cells secrete extracellular vesicles that carry numerous miRNAs modulating the immune response [10]. The aim of this research was to compare the miRNA expression profiles in erythrocytes, lymphocytes, and monocytes from control donors and patients with severe COVID-19, as well as evaluate how the signaling pathways in these cells are altered.

EXPERIMENTAL

Patients, blood collection and fluorescence cell sorting

Blood from five control donors (four males and one female aged 29–73 years) and five patients with severe COVID-19 (four males and one female aged 53–76 years) was collected from the median cubital vein via venipuncture. The clinical data of the patients with severe COVID-19 are presented in Table 1 and the Supplementary Materials. Informed consent was secured from all involved in the study. This study was conducted in compliance with the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of City Hospital No. 40 (Sestroretsk, St. Petersburg, Russia) on May 18, 2020 (protocol No. 171).

 

Table 1. Clinical data of severe COVID-19 patients

Patient	Sex	Age, years	Interleukin-6, pg/mL	CRP (C-reactive protein), mg/L	The patient’s condition at the time of blood sampling
1	male	53	34	106.7	Severe
2	male	57	204	45.1	Severe
3	male	73	62	105.7	Severe
4	female	75	24	58.9	Moderate
5	male	76	2398	10.2	Moderate

 

All the cell types were sorted on a MoFlo Astrios EQ cell sorter (Beckman Coulter, USA). For erythrocyte sorting, 2 µL of whole blood was dissolved in 100 µL of PBS and stained with antibodies specific to CD235, CD45, and CD41. After staining, 5 × 10⁶ CD235+ events were sorted. The precipitate was frozen in dry ice. To sort lymphocytes and monocytes, 100 µL of whole blood was treated with 1 mL of a VersaLyse solution and stained with antibodies. A detailed description of the sorting and gating procedure is available in Ref. [2]. After sorting, the suspension was centrifuged for 10 min at 2,000 g.

RNA separation and next-generation sequencing

ExtractRNA reagent (900 μL) was added to the cells frozen in dry ice, and RNA isolation was performed according to the protocol. The RNA concentration was measured using the QuantiFluor® RNA System (Promega, USA). The quality of leukocyte RNA was determined on a TapeStation instrument (Agilent, USA). Only samples with an RNA Integrity Number (RIN) > 7 were used in the study. Small RNA libraries were constructed using the MGIEasy Small RNA Library Prep Kit V2.0 (BGI-940-000196-00). The prepared libraries were purified by electrophoresis in 6% polyacrylamide gel (PAGE) by cutting a 100–130 bp band corresponding to a transcript of 17–47 nt in length. Sequencing was performed on a DNBSEQ-G400 instrument (BGI) using the DNBSEQ-G400RS High-throughput Sequencing Set (Small RNA FCL SE50) (BGI-1000016998).

Bioinformatics and statistics

A file 700–1,200 MB in size (11–20 million reads per file, with a Q30 score of 93–97) was obtained from each sample. Adapter trimming, miRNA annotation, and the search for novel miRNAs were performed using the Mirmaster 2.0 web service [11]. Annotation of known miRNAs was performed using MirBase [12] version 22.1. Raw data are presented in the Supplementary Materials. Nomenclature of the new miRNAs was performed using the built-in algorithm of the Mirmaster 2.0 web service. Graphs were plotted in RStudio version 2023.09.1 + 494.pro2. The reliability of the differences in the levels of both known and newly predicted miRNAs was analyzed using the Wilcoxon–Mann–Whitney test. The miRNA profiles were analyzed using the miRNA Enrichment Analysis and Annotation Tool (miEAA 2.1) [13]. To subject our data to miEAA, we used the Fold change-p-value ratio with the following formula: –log₁₀(p) · sign(log₂(fold change)). Signaling pathways were annotated using miRTarBase (Release 9.0 beta) [14]. Only signaling pathways and targets with a Q-value < 0.05 were selected for analysis. Prediction of novel miRNAs targets was executed using the miRDB service [15].

RESULTS

We performed a NGS analysis for each sorted blood cell type separately: erythrocytes, lymphocytes, and monocytes. Analysis of the differential miRNA expression in erythrocytes revealed two upregulated and five downregulated miRNAs (Fig. 1, Table 2). All the statistically significant 13 miRNAs in lymphocytes appeared as upregulated (Fig. 1, Table 2). Monocytes had a more diverse pattern in terms of upregulation and downregulation: three upregulated and 11 downregulated (Fig. 1, Table 2).

 

Fig. 1. Differential expression of miRNAs in erythrocytes, lymphocytes, and monocytes. Lists of miRNAs with significantly altered normalized expression levels compiled from the obtained data. These lists are presented in Table 2

 

Table 2. Differential expression of miRNAs in erythrocytes, lymphocytes, and monocytes during severe COVID-19

miRNA	Fold change	P-value	Expression
Erythrocytes
hsa-miR-550a-5p	2.25	0.016	downregulated
hsa-miR-12136	2.60	0.016	upregulated
hsa-miR-449a	3.18	0.02	downregulated
hsa-miR-619-5p	3.18	0.021	downregulated
hsa-miR-1277-5p	2.36	0.032	downregulated
hsa-miR-1228-5p	2.11	0.032	upregulated
hsa-miR-3187-3p	2.30	0.036	downregulated
Lymphocytes
hsa-miR-21-3p	6.99	0.029	upregulated
hsa-miR-23a-5p	3.62	0.029	upregulated
hsa-miR-27a-5p	2.18	0.029	upregulated
hsa-miR-24-2-5p	2.16	0.029	upregulated
hsa-miR-25-5p	3.11	0.029	upregulated
hsa-miR-484	3.17	0.029	upregulated
hsa-miR-34a-5p	7.58	0.029	upregulated
hsa-miR-185-3p	2.69	0.029	upregulated
hsa-miR-503-5p	2.72	0.029	upregulated
hsa-miR-3614-5p	3.08	0.029	upregulated
hsa-miR-130b-5p	2.47	0.029	upregulated
hsa-miR-10398-3p	2.80	0.029	upregulated
hsa-miR-7705	2.42	0.029	upregulated
Monocytes
hsa-miR-4732-5p	6.89	0.008	downregulated
hsa-miR-12136	3.98	0.008	downregulated
hsa-miR-450a-2-3p	2.67	0.008	upregulated
hsa-miR-708-5p	11.62	0.008	downregulated
hsa-miR-34a-5p	3.21	0.008	upregulated
hsa-miR-6734-5p	2.34	0.008	downregulated
hsa-miR-8485	2.43	0.008	downregulated
hsa-miR-6125	2.07	0.012	downregulated
hsa-miR-33b-5p	3.52	0.016	upregulated
hsa-miR-151b	2.07	0.016	downregulated
hsa-miR-4484	3.42	0.032	downregulated
hsa-miR-31-5p	2.12	0.032	downregulated
hsa-miR-151a-5p	2.01	0.032	downregulated
hsa-miR-5187-5p	2.39	0.036	downregulated

 

Evaluation of the signaling pathways altered by miRNAs in erythrocytes, lymphocytes, and monocytes

To investigate the potential of changes (statistical enrichment or depletion) in signaling pathways during severe COVID-19, we performed a miRNA enrichment analysis using the miEAA and miRTarBase (Gene Ontology) tools. The pathways with the smallest Q-values are shown in Fig. 2. The complete list of pathways (Q-value < 0.05) is provided in the Supplementary Materials. These algorithms did not identify any signaling pathways whose profile had been significantly altered in erythrocytes during COVID-19. However, microRNAs were found to influence various signaling pathways in monocytes and lymphocytes.

 

Fig. 2. Evaluation of the major signaling pathways significantly (Q-value < 0.05) altered by miRNAs in monocytes and lymphocytes

 

Identification of novel miRNAs with altered expression during severe COVID-19

The evaluation of novel miRNAs in the three cell populations studied revealed that only four such miRNAs were present in monocytes. No significant changes in such miRNAs were detected in other cell types (Fig. 3). Table 3 summarizes the sequences of these microRNAs and their precursors.

 

Fig. 3. Predicted novel miRNA differential expression in erythrocytes, lymphocytes, and monocytes

 

Table 3. Predicted four novel miRNAs that alter their expression in monocytes during severe COVID-19

miRNA	Fold change	P-value	Expression	miRNA, nucleotide sequence	Pre-miRNA, nucleotide sequence
hsa-miR-3-3p	2.67	0.032	upregulated	GGGUGCGGGCCGGCGGGGUCCU	GACCUCGCCGUCCCGCCCGCCGCCUUCUGCGUCGCGGGUGCGGGCCGGCGGGGUCCU
hsa-miR-120-5p	2.29	0.01	downregulated	UGGGGGAGGAGGAAGAGGAGA	UGGGGGAGGAGGAAGAGGAGAUGGGGAGGCAGGUGAGCCUGACCAAGCAGCCUGCUCCCUUUCUCCCUCCCCUUCCCCCUC
hsa-miR-593-3p	2.37	0.032	downregulated	UUUGGGGAUUCUAAGAGGAAG	AACUCUUAGAAUCCCCAAAGCAUUCUGUGAAGUGGUUUGGGGAUUCUAAGAGGAAG
hsa-miR-388-5p	2.67	0.008	downregulated	GUCCCAGCAACUCAGGAGGCUAAGG	GUCCCAGCAACUCAGGAGGCUAAGGUGGGAGGAUCACUUGAGCCCAGGAGUUCUGGGCUG

 

In order to assess how these miRNAs may affect signaling in cells, we attempted to locate possible target genes using the miRDB service [15]. The most confident targets (score > 95) are presented in Table 4. A complete list of possible targets (score > 50) is provided in the Supplementary Materials. No high-confidence targets (score > 95) were identified for hsa-miR-3-3p.

 

Table 4. The potentially important target genes of the predicted miRNAs

Predicted miRNA	Target gene, conditional prediction confidence, a.u.	Target gene, symbol	Target gene-encoded protein
hsa-miR-120-5p	100	MECP2	methyl-CpG binding protein 2
	100	PPP1R9B	Protein phosphatase 1 regulatory subunit 9B
	99	SLC6A17	Solute carrier family 6 member 17
	99	WIZ	WIZ zinc finger
	98	NFIX	Nuclear factor I X
	98	CASTOR2	Cytosolic arginine sensor for mTORC1 subunit 2
	98	NR1D1	Nuclear receptor subfamily 1 group D member 1
	97	NLGN2	Neuroligin 2
	97	ELK1	ELK1, ETS transcription factor
	97	SHISAL1	Shisa like 1
	96	CAMK1D	Calcium/calmodulin dependent protein kinase ID
	96	ACTB	Actin beta
	96	HEYL	Hes related family bHLH transcription factor with YRPW motif-like
	96	SRF	Serum response factor
	96	EPB41L1	Erythrocyte membrane protein band 4.1 like 1
hsa-miR-593-3p	98	CEP135	Centrosomal protein 135
	98	ELK1	ELK1, ETS transcription factor
	98	ZNF629	Zinc finger protein 629
	97	SUSD2	Sushi domain containing 2
	97	ARGFX	Arginine-fifty homeobox
	97	IGF2	Insulin-like growth factor 2
	97	ZDHHC8	Zinc finger DHHC-type containing 8
	97	BCAM	Basal cell adhesion molecule (Lutheran blood group)
	96	GDF11	Growth differentiation factor 11
	96	TAB3	TGF-beta activated kinase 1 (MAP3K7) binding protein 3
	96	PKNOX2	PBX/knotted 1 homeobox 2
	96	ZBTB39	Zinc finger and BTB domain containing 39
hsa-miR-388-5p	99	STK4	Serine/threonine kinase 4
	98	FOXK1	Forkhead box K1
	98	PCBP1	poly(rC) binding protein 1
	97	BSDC1	BSD domain containing 1
	97	LRTOMT	Leucine rich transmembrane and O-methyltransferase domain containing
	96	KIAA0930	KIAA0930
	96	TUB	Tubby bipartite transcription factor
	96	NRP1	Neuropilin 1
	96	CRP	C-reactive protein

 

DISCUSSION

A few important remarks are in order prior to discussing the result’s validity. Most of the changes in the expression profiles are most likely unrelated to the effects of hypoxia, due to the fact that the O₂ saturation was above 93% in all the patients with severe COVID-19, which is not significant enough to have any effect on blood cells. These changes are not associated with serious comorbidities in COVID-19 patients, since no such patients were added to the sample. Furthermore, follow-up sequencing was performed to assess the validity of the changes in miRNA concentrations. As a result, expression levels of less than ten reads were shown to be poorly reproducible by such sequencing. Hence, we considered significant changes in levels only for those miRNAs where at least one group (control or COVID-19) had median levels of a particular miRNA above ten reads. All the miRNAs reported in this study as having significant differences met our stated criterion.

The mechanism of miRNA profile changes in erythrocytes is not fully understood. Mature erythrocytes lack nuclei [16] and are therefore unable to synthesize pre-miRNA. Therefore, expression regulation should be carried out at the stage of either pre-miRNA or miRNA excision. The expression profile can also be regulated at the stage of an immature erythrocyte possessing a nucleus. However, the latter mechanism is unlikely, since the COVID-19-induced cytokine storm develops rapidly, before any substantial renewal of peripheral blood erythrocytes. Given the fact that the average lifetime of an erythrocyte is 120 days [17], during severe COVID-19 it is likely that their lifespan will be shortened by rapid hemolysis and erythrocyte renewal. It is still not exactly clear how quickly these cells are renewed during a SARS-CoV-2 infection. There are few examples of conditions under which the miRNA profile in erythrocytes is altered, something that is typically due to chronic diseases or other long-term adverse influences on the organism. It has been demonstrated that the miRNA expression pattern changes during sickle cell anemia [7], paroxysmal nocturnal hemoglobinuria [8], as well as Parkinson’s disease [9]. Additionally, people living in high-altitude mountain regions also exhibit changes in miRNA expression [18]. miRNAs contained in erythrocytes cannot influence the synthetic ability of these cells, since erythrocytes lack ribosomes and, therefore, lack translation and miRNA-mediated silencing. However, it has been shown that erythrocytes can release vesicles that circulate in the blood. The contents of these vesicles can enter another cell, and miRNA will change the gene expression in that cell.

When observing the significant hits in erythrocyte miRNAs, we found only two miRNAs to be upregulated: miR-1228-5p and hsa-miR-12136. Both have been identified as differentially expressed miRNAs in various studies related to COVID-19. In particular, miR-1228-5p was found among 246 differentially expressed miRNAs in plasma exosomes from patients, an indication of its potential involvement in the disease process and response to the infection [19]. hsa-miR-12136 was among the top ten differentially expressed and upregulated miRNAs in patients with COVID-19. Additionally, a ROC analysis demonstrated that the levels of hsa-miR-12136, along with other miRNAs, can help differentiate hospitalized COVID-19 patients from healthy uninfected controls with high efficiency [20]. This fact aligns with our observations of hsa-miR-12136 demonstrating the second-highest upregulated fold change state among seven other differentially expressed miRNAs. On the other hand, the downregulated hsa-miR-449a demonstrated significant differential expression and statistical significance. hsa-miR-449a has been identified as a tumor suppressor in various cancers, including neuroblastoma and endometrial cancer [21]. It is known to inhibit cancer cell proliferation by inducing cell differentiation and causing cell cycle arrest. For instance, in neuroblastoma, hsa-miR-449a overexpression leads to the differentiation of cancer cells and downregulation of key cell cycle regulators such as CDK6 and LEF1 [22].

In lymphocytes, we observed the upregulation of several miRNAs. One of these was hsa-miR-21-3p. It has previously been shown that six miRNAs, including miR-21-3p, can directly bind to the RNA of all human coronavirus genomes, including SARS-CoV-2, and regulate viral gene expression. Among these, miR-21-3p exhibited the highest binding affinity to the human coronavirus genome [23]. Our data align with those of previously published research indicating that hsa-miR-21-3p is significantly upregulated during the SARS-CoV-2 infection. This upregulation is associated with a delayed immune response, which may foster viral survival and replication. Specifically, hsa-miR-21-3p has been shown to interact with the viral polyprotein 1a mRNA, a conserved feature across human coronaviruses [24, 25]. Another significantly upregulated miRNA was hsa-miR-9-5p, which has been identified as a miRNA that can target the 3’-untranslated region (3’UTR) of the ACE2 gene. The ACE2 protein (angiotensin-converting enzyme 2) is essential for SARS-CoV-2 entry into host cells. By targeting ACE2, hsa-miR-9-5p may potentially influence susceptibility to the infection and the severity of COVID-19 symptoms [26]. Furthermore, Haldar et al. reported hsa-miR-23a-5p to be associated with the host protein genes involved in the SARS-CoV-2 infection; SERPING1 in particular [27]. The SERPING1 gene encodes the C1 inhibitor (C1-INH) protein, a crucial member of the serpin superfamily of serine protease inhibitors [28]. C1-INH helps control inflammation by inhibiting plasma kallikrein and factor XIIa, both of which are involved in the production of bradykinin, a peptide that increases blood vessel permeability and promotes inflammatory responses [29]. By binding to these proteins, C1-INH prevents excessive bradykinin production, thereby regulating fluid transport into tissues during inflammatory responses [30]. The bradykinin activation theory is one of the most credible hypotheses seeking to explain the cardiovascular complications that occur during severe COVID-19 [31]. Our data on the involvement of hsa-miR-23a-5p in the pathogenesis of severe COVID-19 agree with the aforementioned studies [27].

Among the three upregulated miRNAs, hsa-miR-33b-5p is worthy of note. miR-33b is known to influence various cellular functions that are related to the immune response and inflammation. It is involved in the regulation of pro-inflammatory cytokine production and macrophage polarization. For example, miR-33b was shown to modulate expression of the genes involved in inflammatory pathways, thereby affecting monocyte survival and functioning during inflammatory responses [32, 33]. Furthermore, another microRNA identified in our study of monocytes, hsa-miR-151a-5p, was shown to bind directly to SARS-CoV-2 RNA transcripts [26], specifically targeting the spike protein gene [34]. hsa-miR-151a-5p was implicated in the modulation of the inflammatory response during the SARS-CoV-2 infection [35]. As stated by various researchers, the typical dysregulation of the immune response observed during COVID-19 tends to be associated with changes in the expression levels of several miRNAs, hsa-miR-151a-5p being among them [35, 36].

In addition to studying the differential expression of miRNA in erythrocytes, lymphocytes, and monocytes, we subjected our sets of miRNA to a miRNA Enrichment Analysis (miEAA), which allowed us to analyze the various pathways in which the miRNAs are involved (Fig. 2). The miEAA operates in GeneOntology (GO) terms [37]. Erythrocytes demonstrated no statistically significantly enriched or depleted pathways. This result derives from the fact that a mature erythrocyte has neither a strong signal transduction system nor serious transcriptional activity. A 2020 study of miRNA pathways showed that miR-4732-3p targets components of the TGF-β signaling pathway (SMAD2 and SMAD4) [38], which are involved in erythropoiesis and promote cell proliferation during erythroid differentiation. A 2024 study of hematopoiesis regulation demonstrated how miR-7145 enhances erythropoiesis, while inhibiting myeloid progenitor cell differentiation through the JAK1/STAT3 signaling pathway [39]. Its expression correlates with that of GATA1, a key transcription factor in erythrocyte development. Nevertheless, the aforementioned miRNAs are involved in the erythrocyte progenitor stages, which allows us to suggest that the discovery of these miRNAs in mature erythrocytes is more likely to be evidence of processes occurring during erythropoiesis. Hence, the observed absence of signaling pathways in mature erythrocytes can be regarded as confirmation of low or absent synthetic activity. Regarding statistically significantly enriched pathways in monocytes, one of the most credible pathways appeared to be the positive regulation of metanephric mesenchymal cell migration. Certain miRNAs have been identified as positive regulators of cell migration. For example, miR-200 family members are known to influence epithelial-to-mesenchymal transition (EMT), a process relevant to mesenchymal cell migration [40]. Given that severe COVID-19 leads to immune hyperactivation and immunological dysfunction [41], it is reasonable to posit that the observed enrichment in this pathway is a consequence of these processes. Positive regulation of vascular-associated smooth muscle cell (VSMC) differentiation is important for vascular development and remodeling. The miRNAs involved in VSMC are miR-143/145. This miRNA cluster is crucial for VSMC differentiation. It promotes the expression of contractile proteins, while inhibiting pathways that lead to dedifferentiation and proliferation. The loss of miR-143/145 leads to impaired VSMC differentiation and contributes to vascular pathologies. Additionally, these miRNAs are positively regulated by the serum response factor (SRF) and myocardin, which are critical in promoting VSMC differentiation by suppressing the factors that inhibit the process [42]. Regulation of astrocyte differentiation (another pathway that we have identified) is important for the development of the central nervous system (CNS). Key regulatory factors include RNF20 [43] – an E3 ubiquitin ligase; TAZ (WW domain-containing transcription regulator 1) and YAP (Yes-associated protein) – transcriptional co-activators involved in the Hippo signaling pathway [44]; and the transcription factor PITX1 [45], which controls astrocyte differentiation by regulating SOX9 expression. The involvement of miRNAs in this process had not previously been demonstrated. Recent studies have also highlighted the regulatory role of miRNAs in modulating ribonucleoside-diphosphate reductase (RNR) activity, particularly through the regulation of its RRM2 subunit. hsa-miR-125b-5p and hsa-miR-30a-5p have been shown to negatively correlate with the RRM2 expression in various cancer types. Their regulation suggests that these miRNAs may play a role in the maintenance of appropriate dNTP levels by modulating RNR activity [46]. DH domain binding emerged as another enriched pathway. MicroRNAs interact with RNA-binding proteins (RBPs) that contain specific domains, such as DH. By modulating miRNA levels and activity, RBPs can influence target gene expression, impacting cellular functions such as proliferation, differentiation, and response to stress. The movement of miRNAs into extracellular vesicles is mediated by specific RBPs, and, consequently, this process plays an important role in intercellular signaling and tissue interactions [47]. Corticospinal tract (CST) morphogenesis is another process shown to be regulated by miRNAs. miRNAs often work in conjunction with RNA-binding proteins, which facilitate their loading into the RNA-induced silencing complex (RISC). The miR-34/449 family has been shown to fine-tune the expression of the genes critical for spinal interneuron development. Studies involving mutant mice lacking miR-34/449 revealed a notable disruption in the genetic profiles of spinal cord neurons, indicating that these miRNAs are essential for the proper circuit formation necessary for motor control [48]. Furthermore, miRNAs are key regulatory molecules in motor neuron axon fasciculation [49]. In motor neurons, miRNAs are involved in several developmental processes. For example, modifying the miR-9 expression has been shown to affect motor neuron subtype specification and spinal cord development. Additionally, miR-17-3p regulates the stability of Olig2 transcription factor mRNA, which is critical for spinal motor neuron differentiation [50]. miRNAs also play crucial roles in the early stages of kidney development, particularly in the differentiation of nephron progenitor cells, which give rise to glomerular structures. Specific miRNAs, such as members of the miR-30 family, have been shown to target key transcription factors like Lhx1, which is vital for nephrogenesis [51, 52]. Furthermore, neuron fasciculation is significantly influenced by miRNAs. Several specific miRNAs have been implicated in the fasciculation process. In particular, miR-8 has been shown to regulate the expression of the cell adhesion molecules critical for synapse formation and may also play a role in axon guidance during synaptogenesis in Drosophila [53].

The positive regulation of RhoGEFs by miRNAs is particularly relevant in cancer biology. Aberrant expression of specific miRNAs can lead to enhanced RhoGEF activity, promoting cancer cell migration and invasion. This has been observed in various cancer types where dysregulated miRNA profiles correlate with increased metastatic potential due to modified Rho GTPase signaling pathways [54].

The inner ear development pathway is one of the most prominent signaling pathways in monocytes. The miR-183/96/182 cluster is one of the groups of miRNAs whose functions and expression are best-studied in inner ear development. Studies using animal models have shown that knockout of these miRNAs causes severe damage to hair cell development and hearing loss [55, 56]. Additionally, several miRNAs that regulate the MAPK signaling pathway were identified in monocytes. In particular, miR-203 regulates BCR-ABL levels and inhibits cell proliferation in chronic myeloid leukemia (CML) by silencing the mRNA of MAPK pathway components [57]. It is also known that miR-155 is involved in the regulation of the SOS and KRAS proteins, influencing MAPK/ERK pathway activity. Furthermore, miR-19a regulates RAF1 and other components of the MAPK cascade, affecting the overall signaling dynamics of this pathway. Meanwhile, miR-128 affects c-Met/PI3K/AKT signaling, which is linked to MAPK pathways, particularly in lung cancer [58].

It is known that some miRNAs can directly target mRNAs encoding cyclases, leading to their degradation or translational repression. For instance, miR-282 has been identified as a regulator of adenylate cyclase in the nervous system, suggesting modulation of pathways associated with cAMP signals important for neuronal function [59].

The miRNA sets in lymphocytes demonstrated less involvement in GO terms pathways; however, they seem to be more specific. Among the enriched GO terms, Phospholipase A1 (PLA1) is an enzyme that hydrolyzes phospholipids, playing a crucial role in lipid metabolism and cell signaling within lymphocytes. The regulation of the PLA1 activity is linked to the function of some miRNAs. For instance, the miR-17~92 cluster regulates B-cell survival by targeting the mRNAs of pro-apoptotic factors such as BIM [60]. miRNAs are crucial for the maturation of T and B cells, ensuring that autoreactive cells are eliminated during development to prevent autoimmunity [60]. During T cell activation, specific miRNAs such as miR-155 are upregulated, enhancing effector functions like cytokine production. Conversely, other miRNAs can suppress activation to maintain homeostasis [61, 62]. These miRNAs are likely to serve as mediators in the immunological misfiring that occurs during the cytokine storm induced by severe COVID-19 [41]. Cellular response to dithiothreitol (DTT) is another signaling cascade associated with the miRNAs identified in our study. It has previously been shown that miR-101 regulates SEL1L expression, which is involved in endoplasmic reticulum (ER) stress response. Under conditions of ER stress, for example induced by DTT, regulation of SEL1L by miR-101 may influence neuronal cell death pathways [63, 64].

Possible target genes controlled by the newly predicted miRNAs in monocytes are also of peculiar interest. Thus, hsa-miR-388-5p presumably controls the synthesis of C-reactive protein, one of the proteins of acute inflammation (Table 4). This protein is synthesized by liver cells. However, it is possible that monocytes secrete this miRNA in extracellular vesicles with the purpose of signaling to the liver cells. In the investigated system, the hsa-miR-388-5p level in monocytes during severe COVID-19 is decreased compared to that in healthy donors. This may be an additional factor increasing the CRP blood level in patients with severe COVID-19.

Summarizing our work, we would like to emphasize that its main result is that the spectrum of miRNAs being altered in monocytes, erythrocytes, and lymphocytes during severe COVID-19 has been identified. This can be used to search for potential predictors of severe COVID-19. Additionally, our work would help elucidate the changes in the molecular mechanisms in blood cells not only during COVID-19-induced cytokine storm, but also during other viral infections.

This work was supported by St. Petersburg State University (project ID: 95412780).

Supplementary materials are available at https://doi.org/10.32607/actanaturae.27610

About the authors

A. A. Artamonov

S. M. Kirov Military Medical Academy

Email: kondratovk.kirill@yandex.ru
Russian Federation, St. Petersburg, 194044

Yu. V. Nikitin

S. M. Kirov Military Medical Academy

Email: kondratovk.kirill@yandex.ru
Russian Federation, St. Petersburg, 194044

A. A. Velmiskina

City Hospital No. 40, St. Petersburg; St. Petersburg State University

Email: kondratovk.kirill@yandex.ru
Russian Federation, Petersburg, 197706; St. Petersburg, 199034

S. V. Mosenko

City Hospital No. 40, St. Petersburg; St. Petersburg State University

Email: kondratovk.kirill@yandex.ru
Russian Federation, St. Petersburg, 197706; St. Petersburg, 199034

V. S. Shimansky

City Hospital No. 40, St. Petersburg; St. Petersburg State University

Email: kondratovk.kirill@yandex.ru
Russian Federation, St. Petersburg, 197706; St. Petersburg, 199034

A. Yu. Asinovskaya

City Hospital No. 40, St. Petersburg; St. Petersburg State University

Email: kondratovk.kirill@yandex.ru
Russian Federation, St. Petersburg, 197706; St. Petersburg, 199034

S. V. Apalko

City Hospital No. 40, St. Petersburg; St. Petersburg State University

Email: kondratovk.kirill@yandex.ru
Russian Federation, St. Petersburg, 197706; St. Petersburg, 199034

N. N. Sushentseva

City Hospital No. 40, St. Petersburg

Email: kondratovk.kirill@yandex.ru
Russian Federation, St. Petersburg, 197706

A. M. Ivanov

S. M. Kirov Military Medical Academy

Email: kondratovk.kirill@yandex.ru
Russian Federation, St. Petersburg, 194044

S. G. Scherbak

City Hospital No. 40, St. Petersburg; St. Petersburg State University

Email: kondratovk.kirill@yandex.ru
Russian Federation, St. Petersburg, 197706; St. Petersburg, 199034

K. A. Kondratov

S. M. Kirov Military Medical Academy; City Hospital No. 40, St. Petersburg; St. Petersburg State University

Author for correspondence.
Email: kondratovk.kirill@yandex.ru
Russian Federation, St. Petersburg, 194044; St. Petersburg, 197706; St. Petersburg, 199034

References

Supplementary files