**3. Discussion**

Cardiovascular/cerebrovascular disease associated microRNA expression profile was assessed in children at the age of 3 to 11 years. Subsequently, epigenetic profile was compared between groups descending from pregnancies that had normal course of gestation or were complicated by GH, PE, and FGR, and correlated with the severity of the disease with regard to clinical signs (mild versus severe preeclampsia), dates of delivery (< and >34 weeks in case of PE, < and >32 weeks in case of FGR, respectively), and Doppler ultrasound parameters.

Initially, we collected the anamnesis at children and parents, performed medical examination (BP measurement, and BMI assessment) and utilized Doppler echocardiography to check the heart's anatomy and function and diagnose or rule out cardiac problems in the studied cohorts. That is the reason why the control group of children after normal pregnancies consisted just of those children who had normal BP, normal BMI, normal reference values of echocardiographic measurements and no cardiac findings in anamnesis. The second group of offspring born out of normal pregnancies consisted of children either already dispensarized in the department of pediatric cardiology, or those ones who were indicated during our medical examination to have valve problems and heart defects, abnormal BP or BMI. In general, the expression profile of microRNAs was equal between these two groups of children descending from uncomplicated pregnancies with an exception of miR-21-5p, which was up-regulated in a proportion of children with abnormal clinical findings (28.95%). According to our opinion, this group of children would benefit from dispensarization and implementation of primary prevention strategies, since it may be at higher risk of later development of cardiovascular diseases.

Since we identified a large proportion of children with abnormal clinical findings within groups descending from complicated pregnancies (57.41% children with a prior exposure to GH, 47.37% children exposed to PE, and 64.7% children affected with FGR), we analyzed also microRNA gene expression in relation to both aspects, the current presence or absence of cardiovascular risk factors (overweight/obesity and/or prehypertension/hypertension) and cardiovascular complications (valve problems and heart defects) and the previous occurrence of pregnancy-related complications.

As expected, the expression profile of microRNAs differed between children with a history of complicated pregnancies (GH, PE, and FGR) and controls. With respect to particular pregnancy-related complication subtypes, abnormal expression profile of multiple microRNAs was found in children affected with GH (15/29 studied microRNAs: miR-1-3p, miR-17-5p, miR-20a-5p, miR-21-5p, miR-23a-3p, miR-26a-5p, miR-29a-3p, miR-103a-3p, miR-125b-5p, miR-126-3p, miR-133a-3p, miR-146a-5p, miR-181a-5p, miR-195-5p, and miR-342-3p), clinically established PE (6/29 studied microRNAs: miR-1-3p, miR-20a-5p, miR-20b-5p, miR-103a-3p, miR-133a-3p, miR-342-3p) and FGR (3/29 studied microRNAs: miR-17-5p, miR-126-3p, miR-133a-3p).

Interestingly, a set of microRNAs associated with cardiovascular/cerebrovascular diseases (miR-1-3p, miR-17-5p, miR-29a-3p, miR-126-3p, miR-133a-3p, miR-146a-5p, and miR-181a-5p) was dysregulated in both groups of children with a prior exposure to GH regardless of the occurrence of postnatal clinical findings. Seven additional microRNAs (miR-21-5p, miR-23a-3p, miR-26a-5p, miR-103a-3p, miR-125b-5p, miR-195-5p, and miR-342-3p) were observed to be dysregulated in children affected with GH with normal postnatal clinical findings. In addition, higher expression levels of miR-20a-5p were detected in children with a prior exposure to GH, who were found to have abnormal clinical findings.

With regard to miR-133a-3p, a group of children exposed to PE produced similar findings to the group of children affected with GH. MiR-133a-3p up-regulation appeared in 37.04% and 22.56% children descending from GH or preeclamptic pregnancies irrespective of the postnatal clinical

findings. Nevertheless, miR-133a-3p up-regulation was detected mainly in a group of children affected with severe PE (23.58%) and late PE (21.43%). On the other hand, up-regulation of miR-133a-3p appeared only in those children affected with FGR that were identified to have abnormal clinical findings (31.82%).

This study showed similar findings to our previous study [44], where we observed increased expression of miR-133a-3p in umbilical cord blood from patients with PE and/or FGR that had decreased values of the CPR indicating a protective reaction of the fetus against hypoxia. We sugges<sup>t</sup> that the increased expression of miR-133a-3p found in a proportion of children previously exposed to GH, PE and/or FGR may be a long-term consequence of pregnancy-related complications. We suppose that up-regulation of miR-133a-3p may have rather compensatory than harmful effect, but postnatal miR-133a-3p screening may stratify a risky group of children predisposed to later cardiovascular disease development.

Interestingly, the data resulting from our previous studies [44,45] indicated that miR-1-3p was significantly up-regulated in placental tissues or umbilical cord blood of PE and/or FGR patients with an abnormal index of pulsatility in the umbilical artery or the signs of centralization of the fetal circulation, or in late PE, apparently as a consequence of an incapacity of maternal cardiovascular system to deal with the demands of an advanced pregnancy. That's why the increased expression of miR-1-3p present in circulation of children born out of pregnancies complicated with GH or late PE may also be associated with previous occurrence of hypertensive disorders of pregnancy. A large proportion of children with an up-regulated miR-1-3p profile with a prior exposure to GH (47.83% children with normal clinical findings and 45.16% children with abnormal clinical findings) or late PE (35.14% children with abnormal clinical findings) was identified. We suppose that children descending from complicated pregnancies with aberrant expression profile of miR-1-3p are at a higher risk of later onset of cardiovascular diseases, and should be carefully monitored in the long term.

This study demonstrated that the dysregulation of at least two microRNAs (miR-1-3p and miR-133a-3p) caused by pregnancy complications in placental tissues and/or umbilical cord blood is present as well in circulation of children with hindsight (3 to 11 years after the birth) after the exposure to GH, PE, or FGR. It is obvious that changes in epigenome induced by pregnancy complications in placental tissue and umbilical cord blood may cause later development of cardiovascular and cerebrovascular diseases in offspring. An impact of the environment on early life epigenetic programming might support the phenomena known as developmental programming and explain the developmental origins of diseases. It is well known that complex diseases, such as diabetes, obesity, and heart disease, result from the interaction between genetic and environmental factors. Alternatively, other explanations to the aberrant microRNA expression in pregnancy-related complications may be taken into consideration. For instance, it may be that complicated pregnancies induce changes in proportional representation of any of cell subpopulations including endothelial stem cells and immune cells, which may be the cause of different microRNA expression.

However, it is also evident that epigenetic profiles of other microRNAs have been changing with time by force of various circumstances, since several microRNAs which were up-regulated in children exposed to GH (miR-17-5p, miR-20a-5p, miR-21-5p, miR-23a-3p, miR-29a-3p, miR-146a-5p, and miR-181a-5p), PE (miR-20a-5p, and miR-20b-5p) or FGR (miR-17-5p) were not observed to be dysregulated in placental tissues and/or umbilical cord blood during the clinical manifestation of pregnancy-related complications [44,45].

In parallel, several microRNAs which were up-regulated in children exposed to GH (miR-26a-5p, miR-103a-3p, miR-125b-5p, miR-126-3p, miR-195-5p, and miR-342-3p), PE (miR-103a-3p, and miR-342-3p) or FGR (miR-126-3p) were observed to be down-regulated in placental tissues and/or umbilical cord blood during the clinical manifestation of pregnancy-related complications [44,45].

However, existing data suggests that these microRNAs play a role in the pathogenesis of cardiovascular/cerebrovascular diseases (Table 2) [21,23,25,29,39,46–138].




MiR-1-3p is from the precursors for miR-1-1 and miR-1-2 encoded by distinct genes located on chromosome 20q13.3 and on chromosome 18q11.2 [46]. MiR-1 is abundantly expressed in cardiac and skeletal muscles, especially in myocardium. Extracellular miR-1 levels are significantly increased in patients with acute MI and highly correlate with circulating troponin T, a reliable marker of cardiac damage [47]. MiR-1 also represents a promising therapeutic target in treatment of cardiovascular diseases, heart ischemia, and post-MI complications. Inhibition of miR-1 with antisense oligonucleotides is cardioprotective, since it leads to reduction of apoptosis, increase of resistance to oxidative stress, and attenuation of spontaneous arrhythmogenic oscillations [48–50].

A large proportion of children with an up-regulated miR-1-3p profiles with a prior exposure to GH (47.83% children with normal clinical findings and 45.16% children with abnormal clinical findings) or late PE (35.14% children with abnormal clinical findings) was identified. We suppose that children descending from complicated pregnancies with aberrant expression profiles of miR-1-3p are at a higher risk of later onset of cardiovascular diseases, and should be carefully monitored in the long term.

MiR-17-5p, a member of miR-17-92 cluster, located on the human chromosome 13q31.3 [51,52], is overexpressed in endothelial cells and lowly expressed in vascular smooth muscle cells [53]. MiR-17p~92 microRNAs were found to play a major role in cardiac development, since the hearts of miR-17p~92 deficient mutant embryos presented a clear ventricular septal defect [54]. Moreover, multiple studies have also confirmed the involvement of miR-17-5p in regulating ischemia/reperfusion-induced cardiac injury (I/R-I). Up-regulation of miR-17-5p has been reported to promote apoptosis induced by oxidative stress via targeting Stat3 in in vivo I/R-I mouse model and in vitro cellular model of oxidative stress induced by H202 [55]. The inhibition of miR-17-5p by its specific inhibitors preserved cell survival and rescued cell death in both in in vivo I/R-I mouse and in vitro cellular oxidative stress models [55] and improved cardiac function after acute myocardial infarction via weakening of apoptosis in endothelial cells in SD rat model [56]. Kaucsar et al. reported that miR-17-5p was activated during kidney ischemia-reperfusion injury in mice [57]. In humans, circulating miR-17-5p represents one of the potential up-regulated biomarkers for diffuse myocardial fibrosis in hypertrophic cardiomyopathy [58], in patients with acute ischemic stroke [59], and for the severity of coronary artery disease [60]. Our data showed that miR-17-5p was overexpressed in a proportion of children with a prior exposure to GH, regardless of normal or abnormal clinical findings (30.43% versus 29.03%), and in children previously affected with FGR, however only with abnormal clinical findings (22.73%). In view of the fact that the overexpression of miR-17-5p is associated with a high cardiovascular risk, we suppose that children with aberrant expression profile of miR-17-5p need to be stratified as soon as possible to prevent them from later cardiovascular disease development.

MiR-20a-5p belongs to the miR-17 family and is also transcribed from the miR-17-92 cluster located on the human chromosome 13q31.3 [61]. MiR-20a is involved with inflammatory signaling in pulmonary hypertension [62]. Intraperitoneal injections of antagomiR-20a significantly down-regulated the expression levels of miR-20a-5p and restored functional levels of bone morphogenetic protein receptor type 2 (BMPR2) in pulmonary arteries in hypoxia-induced pulmonary hypertension mouse model [62]. A previous pilot study of Zhu et al. demonstrated increased plasma levels of miR-17-5p and miR-20a-5p also in patients diagnosed with gestational diabetes mellitus (GDM) at 16–19 weeks of gestation [63]. Our results support the involvement of miR-20a-5p in pathogenesis of pathologies enhancing a cardiovascular risk, since overexpression of miR-20a-5p was identified only in a proportion of children with abnormal clinical findings that were previously exposed to GH (25.81%) or preeclampsia terminated after 34 weeks of gestation (13.51%). The early identification of this risky group of children may improve their future cardiovascular health.

MiR-20b-5p also belongs to the miR-17 family, however is transcribed from the miR-106a-363 cluster located on chromosome Xq26.2 [61]. Recent study showed that increased plasmatic levels of miR-20b served as one of selected biomarkers in hypertension-induced heart failure in adult male Dahl salt-sensitive rats [64]. In addition, higher levels of miR-20b-5p were found in second trimester maternal sera of pregnancies with small-for-gestational age fetuses when compared with appropriate-for-gestational age fetuses [65]. In our study, mir-20b-5p represented a unique biomarker that was up-regulated in a proportion of children with contemporary normal clinical findings that were previously affected with mild preeclampsia (36.36%). Due to the lack of information on aberrant expression profile of miR-20b-5p in cardiovascular diseases, its role in children descending from complicated pregnancies remains for us unclear.

MiR-21-5p and miR-21-3p are encoded by the gene located on chromosome 17q23.2 [66] and mediate the homeostasis of the cardiovascular system [67]. Overexpression of miR-21 promotes cardiac fibrosis and development of heart failure with preserved left ventricular ejection fraction. Inhibition of miR-21 by miR-21 antagonists led to amelioration of cardiac atrophy and cardiac fibrosis [68]. MiR-21 is upregulated by HIF-1 α under hypoxia in cardiomyocytes and silencing of HIF-1 α and inhibition of miR-21 increase the apoptosis of hypoxic cardiomyocytes [69]. Extracellular miR-21 can be used as a biomarker for the diagnosis and prognosis of heart failure. Serum levels of miR-21 were higher in patients with heart failure than in controls, and correlated with ejection fraction and brain natriuretic peptide [70]. In our study miR-21-5p was up-regulated in a proportion of children descending from normal pregnancies with abnormal clinical findings (28.95%) and in children descending from GH pregnancies with normal clinical findings (39.13%). According to our opinion, this group of children would benefit from dispensarization and implementation of primary prevention strategies, since it may be at higher risk of later development of cardiovascular diseases.

MiR-23a, encoded by a gene located at chromosome 19p13.12, has two mature microRNAs: miR-23a-5p and miR-23a-3p. MiR-23a regulates cardiomyocyte apoptosis, a key pathogenesis factor of heart failure, by targeting manganese superoxide dismutase [71] and the vasculogenesis of coronary artery disease via targeting epidermal growth factor receptor [72]. Extracellular miR-23a may be a new biomarker for coronary artery disease, since increased levels of miR-23a may be used to predict the presence and severity of coronary lesions in patients with coronary artery disease [72]. MiR-23a-3p also suppressed oxidative stress injury in a mouse model of focal cerebral ischemia-reperfusion [73]. Since our study demonstrated overexpression of miR-23a-3p in a proportion of children with normal clinical findings born of GH complicated pregnancies only (34.78%), we suppose that compensatory effect of miR-23a-3p may appear more likely in these children to normalize cardiomyocyte state and vasculogenesis.

MiR-26a-5p is produced by miR-26a-1 and miR-26a-2, whose genes are located on chromosomes 3p22.2 and 12q14.1 [74]. MiR-26a-5p was demonstrated to regulate the autophagy in cardiac fibroblasts by targeting a key component of autophagy pathway, ULK1 (unc-51 like autophagy activating kinase 1). Overexpression of miR-26a-5p reduces the expression of ULK1 and decreases the activation of LC3-I to LC3-II (microtubule-associated protein 1 light chain) participating in the formation of autophagosome membranes during autophagy [75]. Autophagy plays a protective role in heart failure and cardiac hypertrophy by removing damaged proteins [65]. Nevertheless, in some cases inhibited autophagy may also improve cardiac function [75]. Since our study showed overexpression of miR-26a-5p in a proportion of children with normal clinical findings that were affected with GH (21.74%), we sugges<sup>t</sup> that this phenomenon may have protective effect against potential development of cardiac fibrosis.

MiR-29a-3p, a member of miR-29 family, is encoded by a gene located on chromosome 7q32.3. Antagomirs against miR-29a significantly increased Mcl-2 expression and significantly reduced myocardial infarct size and apoptosis in rat hearts subjected to ischaemia-reperfusion injury [76]. Increased expression of miR-29a-3p was also observed in cardiac cachexia, a common complication of heart failure, in Wistar rat models [77]. In patients with atrial fibrillation, overexpression of miR-29a-3p found in the biopsies collected from the right atrial appendage during the general surgical procedure was associated with underexpression of calcium voltage-gated channel subunit alpha 1C (CACNA1C) [78]. Circulating miR-29a-3p represents one of potential up-regulated biomarkers for diffuse myocardial fibrosis in hypertrophic cardiomyopathy [58]. Previous pilot study of Zhu et al. demonstrated increased plasma levels of miR-29a-3p in GDM patients at 16–19 weeks of gestation [63]. Serum mir-29a-3p levels were also shown to be elevated in patients with recent diagnosis of type 2 diabetes mellitus (T2DM) [23]. Moreover, overexpression of miR-29a-3p was found in resistance arterioles obtained by biopsy from T2DM patients [79]. Our study indicated the presence of overexpression of miR-29a-3p in both groups of children with normal and abnormal clinical findings descending from pregnancies complicated by GH (39.13% versus 32.26%). We believe that this group of children is endangered by later development of cardiovascular diseases and would benefit from early prevention programs to decrease a cardiovascular risk.

MiR-103 and miR-107 are paralogous microRNAs binding the same target sites. MiR-103 is encoded by two different genes located on chromosomes 5q34 and 20p13 [80]. Both genes generate miR-103a-3p. MiR-103/107 regulate programmed necrosis and myocardial ischemia/reperfusion injury via targeting FADD (Fas-associated protein with death domain). Both miR-103 and miR-107 are overexpressed in the ischemic zone of ischemic heart. Plasma miR-103a concentration is also significantly elevated in patients with high blood pressure and acute myocardial infarction [81]. Injection of miR-103/107 antagomir led to a reduction of FADD and induced a reduction in the myocardial necrosis and myocardial infarct sizes [82]. MiR-103/107 is also involved in hypoxia-induced pulmonary hypertension through hypoxia-induced proliferation of pulmonary arterial smooth muscle cells via targeting HIF-1β. Nevertheless, in this model miR-103/107 induced overexpression led to inhibition of hypoxia induced pulmonary arterial smooth muscle cell proliferation and hypoxia-induced pulmonary hypertension [83]. In addition, miR-103/107 play the central importance in regulation of insulin sensitivity. Overexpression of miR-103/107 is present in obese mice and silencing of miR-103/miR-107 leads to the improvement of glucose homeostasis and insulin sensitivity [84]. Our data are similar to findings of Huang et al. [81] and Trajkovski et al. [84]. Since we identified overexpression of miR-103a-3p in a proportion of children with abnormal clinical findings who were exposed to severe preeclampsia (12.77%) and or preeclampsia developing after 34 weeks of gestation (13.51%), we consider up-regulation of miR-103-3p in circulation of children as a highly risky feature of potential development of cardiovascular diseases.

There are two paralogs, miR-125b-1 on chromosome 11q24.1 and miR-125b-2 on chromosome 21q21.1, both producing miR-125b-5p [85]. A set of upregulated circulating microRNAs including miR-125b-5p was identified to be associated with acute ischemic stroke [86] and acute myocardial infarction [87]. Overexpression of miR-125b-5p was shown to protect endothelial cells from apoptosis under oxidative stress via negative regulation of SMAD4 (SMAD family member 4) [88]. In addition, miR-125b-5p was identified as an ischemic stress-responsive protector against cardiomyocyte apoptosis. Cardiomyocytes overexpressing miR-125b-5p have increased prosurvival signaling and protected the heart from acute myocardial infarction by repressing pro-apoptotic bak1 and klf13 [89]. Our results are consistent with the data of Wei et al. [88] and Bayoumi et al. [89], since we observed overexpression of miR-125b-5p in a significant proportion of children with normal clinical findings that were previously affected with GH (47.83%). We sugges<sup>t</sup> that a protective effect of miR-125-5p may be present just in children previously exposed to minor pregnancy-related complications, while in children exposed to severe pregnancy-related complications it apparently vanished.

MiR-126, producing miR-126-3p, is an intronic microRNA located in intron 7 of EGFL7 (epidermal growth factor-like protein 7) gene on chromosome 9q34.3 [90]. MiR-126 regulates endothelial expression of vascular cell adhesion molecule 1 (VCAM-1) and controls vascular inflammation. Overexpression of miR-126 decreases VCAM-1 expression and in opposite transfection of endothelial cells with miR-126 antagomirs increases TNF α-stimulated VCAM-1 expression [21]. MiR-126-3p was down-regulated in the sera derived from patients with acute myocardial infarction [91] and in the plasma of type 2 diabetes patients [92]. Recent experiments also demonstrated that intercellular transfer of miR-126-3p by endothelial microparticles reduced vascular smooth muscle cell proliferation and limited neointima formation via inhibition of LRP6 (LDL receptor related protein 6) [93]. It seems that compensatory role of miR-126-3p also appears in a proportion of children with both normal or abnormal clinical findings with a prior exposure to GH (39.13% versus 22.58%) or FGR (31.82%), most

likely with the aim to induce suppression of cytokine activated endothelial cells and vascular smooth muscle cell proliferation.

MiR-133a-3p belongs to the miR-133 family. MiR-133a-1 gene is located on chromosome 18q11.2 and miR-133a-2 gene on chromosome 20q13.33, respectively. Both genes produce miR-133a-3p [94]. MiR-133 was down-regulated with hypertrophy [95], heart failure [96], and down-regulation of miR-133 also contributed to arrhythmogenesis in the hypertrophic and failing hearts. Induction of overexpression of miR-133 reduced hypertrophy as well as lead to correction of conduction abnormalities [97,98]. Furthermore, circulating miR-133a-3p provides one of up-regulated candidates as potential biomarkers for diffuse myocardial fibrosis in hypertrophic cardiomyopathy [58] and coronary artery calcification [99]. We sugges<sup>t</sup> that the overexpression of miR-133a-3p observed in our study in a proportion of children previously exposed to GH, PE and/or FGR may be a long-term consequence of pregnancy-related complications and we believe that postnatal miR-133a-3p screening may stratify a risky group of children predisposed to later cardiovascular disease development.

The miR-146 family consists of 2 members, with nearly identical sequences, miR-146a-5p and miR-146b-5p [100]. MIR146A gene was found within a larger long noncoding RNA host gene, MIR3142HG, on chromosome 5q33.3 [101]. MiR-146a-5p is actively involved in multiple oncological processes such as antitumor immune suppression, metastasis, and angiogenesis [102]. MiR-146a-5p is an anti-inflammatory microRNA, since it functions as a negative regulator of inflammation by targeting interleukin-1 receptor-associated kinase 1 (IRAK1) and tumor necrosis factor (TNF) receptor associated factor 6 (TRAF6), resulting in inhibition of NF-κB activation [103]. Furthermore, miR-146a is one of the microRNAs that is most sensitive to hypoxia. It seems that miR-146a overexpression can protect the myocardium from I/R damage. Lentivirus expressing miR-146a transfected into mouse hearts decreased I/R-induced myocardial infarct size and prevented I/R-induced decreases in ejection fraction and fractional shortening [104]. Increased circulating plasma levels of miR-146a also correlated with the severity of coronary atherosclerosis in patients with subclinical hypothyroidism and have been suggested as good predictor for coronary heart disease development among individuals with elevated TSH levels [105]. Nevertheless, a significant reduction of miR-146a expression was observed in acute ischemic stroke. A down-regulation of miR-146a was suggested to be a self-protective response of the brain against the consequences of acute cerebral ischemia injury via the up-regulation of Fbxl10 expression, which protects neurons from ischemic death [103]. Since we observed the overexpression of miR-146a-5p in both groups of children with normal and abnormal clinical findings, whose pregnancies were complicated by GH only (26.09% versus 12.9%), we believe that a protective role of miR-146a-5p is applied at least at a lesser group of children with the aim to reduce inflammation and its negative consequences.

MiR-181a and miR-181b cluster together on chromosomes 1q32.1 and 9q33.3. MiR-181a generates several mature microRNAs involving miR-181a-5p, miR-181a-3p and miR-181a2-3p [106]. The miR-181 family plays a key role in both acute and chronic inflammatory disease states such as atherosclerosis, type 2 diabetes, and obesity [106]. Nevertheless, contradictory data are reported concerning expression levels of miR-181a in various pathological conditions. Decreased expression of miR-181a observed in monocytes of obese patients was associated with the metabolic syndrome and coronary artery disease [107]. However, another study reported that overexpression of miR-181a-5p in adipocytes upregulated insulin-stimulated AKT activation and reduced TNF α-induced insulin resistance [108]. Moreover, increased hepatic miR-181a impaired glucose and lipid homeostasis by silencing sirtuin 1 in non-alcoholic fatty liver disease [109]. Several contradictory data are also reported concerning serum levels of miR-181a in diabetic patients. While serum levels of miR-181a-5p were decreased in obese and diabetic patients [108], circulating levels of miR-181a were increased in type 1 diabetic children and adolescents [110]. Nevertheless, circulating miR-181a-5p levels were increased in patients with ischaemic stroke, transient ischaemic attack and acute myocardial infarction [111,112]. Our study revealed overexpression of miR-181a-5p in a proportion of children with both normal and abnormal clinical findings descending from GH pregnancies (39.13% and 25.81%). In view of the inconsistency between studies, the role of miR-181a-5p is not clear. Nevertheless, we believe that increased levels of miR-181a-5p in whole peripheral blood of children previously exposed to GH may predispose to later development of cardiovascular diseases.

MiR-195 gene is located on the chromosome 17p13.1 and generates two microRNAs, miR-195-5p and miR-195-3p [113]. MiR-195 is increasing in cardiac hypertrophy, and cardiac miR-195 overexpression results in heart failure [29,114]. MiR-195 is also a powerful regulator of the aortic extracellular matrix. Administration of miR-195 antagomirs led to significant elevation of elastin and collagens in the murine aorta, but has no effect on survival and aortic diameter size. Nevertheless, in plasma samples an inverse correlation between miR-195 and the presence of abdominal aortic aneurysms and aortic diameter was observed. Surprisingly, miR-195 plasma levels were decreased in abdominal aortic aneurysms [115]. On the other hand, aortic stenosis caused by leaflet calcification of the bicuspid aortic valve was associated with down-regulation of miR-195 [116]. The mechanism of miR-195 action is not completely understood. Since we observed overexpression of miR-195-5p in a proportion of children with normal clinical findings previously affected with GH only (34.78%), we sugges<sup>t</sup> that overexpression of miR-195-5p may have rather protective role than harmful effect.

MiR-210-3p, encoded by a gene located on chromosome 11p15.5, is the most prominent microRNA consistently stimulated under hypoxic conditions. A significant increase of mir-210 levels was detected in placentas of women with preeclampsia, a condition that is characterized by hypoxia resulting from inadequate blood supply to the placenta [39]. Several microRNAs involving miR-210 were reported to be involved in atherosclerotic plaque formation through the regulation of endothelial apoptosis [25,117]. Nevertheless, some authors reported that up-regulated levels of miR-210 positively correlated with the level of endothelial cell apoptosis [117], while the others found that miR-210 blockage only led to induction of endothelial cell apoptosis and cell death in hypoxia [128]. Another study showed that up-regulation of miR-210 had cytoprotective effects, mainly in cardiomyocytes [119] and the skeletal muscle [118]. Increased serum levels of miR-210 may indicate early stages of atherosclerosis obliterans [120] and heart failure [121]. MiR-210 also regulates angiogenesis in response to ischemic injury to the brain. Overexpression of miR-210 may activate the Notch signaling pathway, which probably contributes to angiogenesis after cerebral ischemia [122]. In view of the fact that we observed a strong positive correlation between gene expression of miR-210-3p in whole peripheral blood of children with a history of preeclampsia and/or FGR and the pulsatility index in the ductus venosus, we believe that children descending from preeclampsia and/or FGR affected pregnancies who had high postnatal levels of miR-210-3p in their circulation represent a risky group, that is endangered by the onset of cardiovascular and cerebrovascular diseases in the future, since increased PI in the ductus venosus is a sign of a poor perinatological outcome.

MiR-342-3p is encoded by a gene located on chromosome 14q32.2. MiR-342-3p is considered as an obesity-associated microRNA, since it positively regulates adipogenesis of adipose-derived mesenchymal stem cells by suppressing CtBP2 and releasing the key adipogenic regulator C/EBP α from CtBP2 binding [123]. A set of up-regulated microRNAs expressed in peripheral blood mononuclear cells involving mir-342-3p was shared among patients with type 1 diabetes mellitus, type 2 diabetes mellitus and gestational diabetes mellitus [124]. Urinary exosomal miR-342 was also expressed at significantly elevated levels in type 2 diabetes mellitus patients [125]. However, miR-342-3p expression was significantly reduced in endothelial cells isolated from lung and heart tissues of type 2 diabetes mellitus mice and this inhibition blocked vasculogenesis in vivo by repressing endothelial proliferation and migration [126]. Decreased expression of miR-342 was also observed in T regulatory cells of patients with type 1 diabetes mellitus [127]. Moreover, plasma miR-342-3p was identified as a potential down-regulated biomarker of children aged 5–10 years with endothelial dysfunction [128]. Our study demonstrated decreased expression of miR-342-3p in a proportion of children with abnormal clinical findings that were previously exposed to early preeclampsia, which required the termination of gestation before 34 weeks (26.92%). Therefore, we sugges<sup>t</sup> that this group of children is a highly risky group, which would benefit from implementation of early primary prevention strategies and long-term follow-up.
