Cardiovascular diseases (CVDs) are the leading cause of death worldwide. An estimated 17.9 million people died from CVDs in 2016, representing 31% of all global deaths. Of these deaths, 85% are due to heart attack and ischemic stroke (1). Most CVDs can be prevented by addressing lifestyle-related risk factors: tobacco use, physical inactivity, hypertension, diabetes, and hyperlipidemia (2). The treatment of choice for hypercholesterolemia is the use of statins, which inhibit the enzyme 3-hydroxy-methylglutaryl coenzyme A (HMG-CoA) reductase, which limits the biosynthesis of liver cholesterol (3). Experimental and clinical studies suggest that statins may exert cardiovascular protective effects by mechanisms independent of cholesterol lowering called “pleiotropic effects” (4, 5). It has been described that the mechanism by which statins produce their pleiotropic effects is through inhibition of the mevalonate pathway. Mevalonate is a precursor not only of cholesterol but also of several other isopropenoids (6). These include farnesylpyrophosphate (FPP) and geranylgeranylpyrophosphate (GGP), which are lipids needed for posttranslational prenylation of Rho GTPase proteins, which act as biological switches through signal transduction (6, 7). The inhibition of mevalonate synthesis prevents the activation of Rho and the subsequent activation of Rho-associated protein kinase (ROCK) (8, 9). Several studies have reported the importance of the RhoA/ROCK pathway in endothelial function affecting vascular tone, platelet aggregation, smooth-muscle cell proliferation, leukocyte adhesion, production of nitric oxide synthase (eNOS), and bioavailability of nitric oxide (NO). The mechanisms associated with eNOS are of great importance, since the decrease of NO corresponds to one of the first manifestations of atherosclerosis (10, 11). Therefore, inhibition of the RhoA/ROCK pathway by statins influences NO signaling in endothelial cells (12). This has been demonstrated by two mechanisms: stability of eNOS messenger RNA (mRNA) and an increase in its activity through activation of phosphatidylinositol 3-kinase (PI3K)/AKT (13, 14). For this reason, inhibition of the RhoA/ROCK kinase pathway might be an attractive atheroprotective treatment for the development of CVD.

MicroRNAs (miRNAs) are important regulators of gene expression that bind complementary target mRNAs and repress their expression. The miRNAs are small non-coding RNA molecules, evolutionarily conserved. They have approximately 22 nucleotides, and their functions are gene regulation through posttranscriptional mechanisms. The miRNAs are involved in a variety of biological processes (15). Over the past few years, a total of 2,300 miRNAs have been reported in humans and it is estimated that human mRNA information exceeds 25,000 (16, 17). For this reason, current laboratory technologies do not allow to test globally every possible interaction between gene and miRNAs. However, the use of bioinformatics approaches for miRNA target prediction is used as a guide for laboratory validation experiments to more quickly elucidate gene regulation networks (18). The bioinformatics development combined with diverse experiments has allowed them to be used as potential biomarkers for diagnosis, prognosis, and personalized treatment (19).

Several studies in the area of cancer have demonstrated the regulation of transcriptional expression by miRNAs of the ROCK1 and ROCK2 isoforms of the RhoA/ROCK kinase pathway, reporting beneficial effects such as decreased proliferation and migration by modulating the expression of these miRNAs (20, 21). Zambrano et al. (22) showed that treatment with atorvastatin and simvastatin at 10 micromolar (μM) in HepG2 cells for 24 h deregulated the expression of 13 miRNAs by atorvastatin and 2 miRNAs by simvastatin. Other studies also show that atorvastatin can lower miR-221 and miR-222 levels in cell endothelial parents of patients with coronary disease (23). In particular, the effects observed by statins are heterogeneous, appearing to be independent of their chemical composition (lipophilic vs. hydrophilic) (24). It has also been reported that miRNAs would have the ability to influence endothelium atheroprotective effects, such as increasing nitric NO. Cerda et al. (25) described that these miRNAs increase the levels of NO and the expression of eNOS3 in endothelial cells. This background reveals that statin-modulated miRNAs influence their response and effectiveness. For this reason, the objective of this study was to identify by bioinformatic analysis possible new miRNAs that could modulate the pleiotropic effects exerted by statins through the inhibition of ROCK1 and ROCK2 kinases, and also to explore the pathways associated with the atheroprotective capacity of endothelial cells through an in-silico approach. Therefore, a bioinformatics study was performed comparing the differential expression of miRNAs in human umbilical vein endothelial cells (HUVEC) under two conditions: control and treated with simvastatin at 10 (μM) for 24 h.

In the present in-silico study, we analyzed data collected from the GEO database (GSE126290) in which the differential expression of miRNAs between two conditions was evaluated: HUVEC cells treated with simvastatin at 10 μM and control cells (without treatment) for 24 h. The bioinformatic analysis on differential expression analysis of miRNAs showed the deregulation of a group of seven miRNAs (hsa-miR-515-5p, hsa-miR-18a^*, hsa-miR-640, hsa-miR-618, hsa-miR-363, hsa-miR-548h, and hsa-miR-297) considering as cutoff criteria a logFC > 1 and FDR p-value < 0.05. Among the group of seven miRNAs, four are downregulated (hsa-miR-515-5p, hsa-miR-18a^*, hsa-miR-640, and hsa-miR-548h) and three are upregulated (hsa-miR-618, hsa-miR-297, and hsa-miR-363). Liang et al. (38) performed a differential expression analysis from the same data set presented in this paper. Their results showed that miRNA hsa-miR-652-3p inhibits ISL1, a protein involved in the decrease of NO bioavailability in endothelial cells. However, this miRNA is not upregulated in our analysis. One of these differences is due to their cutoff values considered: p-value of 0.1 and fold change ± 0.2. This would also establish a difference in the amount of deregulated miRNAs, since they present a total of 167, in relation to our analysis which were 7.

Our work supports previous reports on the effect of statins on the dysregulation of miRNA expression and also proposes the way in which statins can exert pleiotropic effects associated with the RhoA/ROCK kinase pathway. These miRNAs are modulating various processes associated with CVDs (39, 40). We observed that simvastatin can modulate seven miRNAs in endothelial cells, and two that are upregulated could modify their expression through binding to the 3′UTR region of the ROCK2 kinases. Nowadays, there are several tools available to predict miRNA targets based mainly on detecting the complementarity of the miRNA sequence with the 3′UTR region of the target gene (41). However, many of these predictions show false negatives in the experimental validation. One strategy to maximize performance and minimize false results in vitro and in vivo functional experiments is the union of bioinformatics tools for miRNA-target prediction that will allow increasing the specificity and precision of the analysis (42). For this reason, three prediction programs were used together: TargetScan, miRDB, and mirDIP 4.1. All three tools showed that hsa-miR-618 and hsa-mir-297 had ROCK2 as a target. The hsa-miR-618 miRNA shows complementarity in three regions of the 3′UTR of ROCK2. Studies have shown that the more seed bonds it has in the 3′UTR, the stronger the miRNA-induced regulation (43). On average, the change in Log2 was linearly correlated with the number of seed regions, suggesting that the effect of seeds is independent and multiplicative (44). For this reason, we can suggest that hsa-miR-618 would have a greater effect on the 3′UTR.

Considering that many of the pathologies associated with the vasculature are related to morphological changes that can be regulated by the interaction of hemodynamic forces, one of the limitations of our work is that the database used contains data from an in vitro model and not from an in vivo model. Also, in the vasculature there is presence of different cell types that contribute to the physiopathology. An example of this, in the adhesion of leukocytes, is the production of inflammatory cytokines by macrophages, etc. However, many authors agree that HUVEC is a model of representative study of the vasculature and also it has allowed the study of physiological and pathological effects in isolation as in coculture with leukocytes and smooth-muscle cells (45, 46). For this reason, it is important to consider validating these results in vitro and in vivo models.

ROCK2 is a 160-kDa serine/threonine kinase protein, and its functions are the regulation of smooth-muscle contraction, organization of the actin cytoskeleton and stress fibers, formation of focal adhesions, and retraction of neurons and cellular adhesions (47). Therefore, its function in different cells and how it contributes to different pathologies through mechanisms mediated by miRNAs have been studied. Liu et al. (48) demonstrated that hsa-miR-122 can inhibit the proliferation of prostate carcinoma cells due to the negative regulation of ROCK2 expression. Furthermore, the hsa-mir-185/ROCK2 pathway was shown to have potential to improve therapies in hepatocellular carcinoma, through metastasis inhibition (49). However, miRNAs hsa-miR-618 and hsa-miR-297 have not been experimentally validated as ROCK2 targets. The hsa-miR-618 has been reported to be unregulated in pathologies as epilepsy (50) and squamous cells from head and neck carcinomas (51). The case of hsa-miR-297 has been associated with prostate cancer (52) and with colorectal carcinoma, specifically with drug resistance (53).

The set of differentially expressed miRNAs and their gene ontological analysis showed the following biological processes: regulation of biosynthesis processes, regulation of biosynthetic macromolecule processes, and positive regulation of nitrogen compound metabolic processes; this last process could be related to the gene's enrichment involved in the mechanisms associated with the ability of statins to increase the bioavailability of NO, through the inhibition of ROCKs (9). The increased expression of ROCKs reduces the expression of eNOS, and the inhibitors (Y-27632 and fasudil) have been shown to increase eNOS (54). In the case of the statins, the increased expression of eNOS was not due to FPP and LDL-C; it was due to inhibition of GGP involved with RhoA and ROCK signaling (6).

Other biological processes and cellular components highlighted in the analysis are neurogenesis, development of the nervous system, and cellular development. Several studies in animal models of spontaneous intracerebral hemorrhage that have been treated with statins have improved neurological function, reduced cerebral edema effects, increasing angiogenesis, and neurogenesis, and have decreased the infiltration of inflammatory cells. This supports the possible neuroprotective effects of statins (55–57).

Statins may also exert their pleiotropic effects through Kruppel-like factor-2 (KLF2). Statins induced the increase in mRNA expression of KLF2 in endothelial cells, which is necessary for eNOS expression (58). This could be associated with the results of molecular functions obtained like the transcription regulator activity in ontological gene analysis.

Inside the analysis associated with the biological pathways, the following ones might be highlighted: signaling pathways regulating pluripotency of stem cells, axon guidance, PI3K-AKT signaling pathways, and miRNAs in cancer. The signaling pathways regulating pluripotency of stem cells has been associated with the transforming growth factor (TGF-B) superfamily, performing important functions during the differentiation of vascular progenitor cells derived from mouse embryonic stem cells (59). Statins can exert pleiotropic effects by increasing the mobilization of endothelial progenitor cells. Damaged endothelial progenitor cells are associated with impaired endothelial function and reduced NO levels. Studies show that prehypertension and hypertension in patients are related to early senescence of progenitor cells and impaired endothelial function (60). Another route mentioned is the axon guidance, which could represent a key stage in the formation of the neural network. This mechanism has been associated with the Rho GTPase pathway through the reorganization of the cytoskeleton that determines the direction that will guide the growth cone (61). A study conducted on brain slices of oxygen-glucose-deprived cells that were exposed to fasudil (ROCK inhibitor) showed an improvement in neuronal viability (62, 63). These findings could be linked to the neuroprotective effects that statins would have on the neurovascular system. Furthermore, the PI3K-AKT signaling pathway corresponds to one of those affected by ROCK inhibition by statins as mentioned above. ROCK is a negative regulator of AKT, possibly through activation of phosphatase and the homolog of tensin (64). Also, the regulation of AKT is known to influence the expression of eNOS and thus an increase in the bioavailability of NO. This effect has been shown to contribute to a decrease in the proliferation of smooth-muscle cells, a key mechanism in the development of atherosclerosis (65). Studies have revealed that statins modulate miRNAs associated with influence the change from senescent contractile phenotype to proliferative phenotype. This could be associated with the enrichment of the miRNA pathway in cancer (66).

Finally, in-silico analysis has shown that statins might exert pleiotropic effects through the modulation of miRNAs that would inhibit ROCK2 kinase and also generate atheroprotective effects in endothelial cells (Figure 4). Although our work proposes new miRNAs that regulate the RhoA/Rock kinase pathway through an in-silico approach, it is necessary to continue with the experimental validation of the miRNAs expressed differentially in endothelial cells through in vitro and in vivo study models. In addition, the functional role of hsa-miR-618 and hsa-miR-297 miRNAs as inhibitors of ROCK2 gene expression and in the mediation of pleiotropic effects should be evaluated.

In this work, we report that simvastatin treatment deregulates the expression of seven miRNAs (hsa-miR-515-5p, hsa-miR-18a^*, hsa-miR-640, hsa-miR-618, hsa-miR-363, hsa-miR-548h, and hsa-miR-297) in HUVEC cell culture. In-silico analysis shows that the miRNAs hsa-miR-618 and hsa-miR-297 are upregulated, and the 3′UTR region of ROCK2, an important protein of the RhoA/Rock kinase pathway involved in modulation of pleiotropic effects exerted by statins, is predicted as a potential target for both miRNAs. In addition, functional analysis and enriched pathways revealed an important association with the pleiotropic effects produced by statins and with the inhibition of RhoA/ROCKs. These results open a new way to understand how statins, through the deregulation of the miRNA expression, might cause an atheroprotective effect through the inhibition of ROCK2 kinase.

In summary, our work through the use of bioinformatics tools contributes with new potential candidates that could regulate the pleiotropic effects in response to statin treatment. However, the functional role of hsa-miR-618 and hsa-miR-297 miRNAs and their mediation of pleiotropic effects must be evaluated by validating in vitro and in vivo models.

Publicly available datasets were analyzed in this study. This data can be found here: Accession number GSE126290.

KL and CR performed the bioinformatics analysis. KL wrote the manuscript. KS and LS contributed in writing—review and editing. All authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.