First-Trimester Screening for Fetal Growth Restriction and Small-for-Gestational-Age Pregnancies without Preeclampsia Using Cardiovascular Disease-Associated MicroRNA Biomarkers

The goal of the study was to determine the early diagnostical potential of cardiovascular disease-associated microRNAs for prediction of small-for-gestational-age (SGA) and fetal growth restriction (FGR) without preeclampsia (PE). The whole peripheral venous blood samples were collected within 10 to 13 weeks of gestation from singleton Caucasian pregnancies within the period November 2012 to March 2020. The case-control retrospective study, nested in a cohort, involved all pregnancies diagnosed with SGA (n = 37) or FGR (n = 82) without PE and 80 appropriate-for-gestational age (AGA) pregnancies selected with regard to equality of sample storage time. Gene expression of 29 cardiovascular disease-associated microRNAs was assessed using real-time RT-PCR. Upregulation of miR-16-5p, miR-20a-5p, miR-146a-5p, miR-155-5p, miR-181a-5p, and miR-195-5p was observed in SGA or FGR pregnancies at 10.0% false positive rate (FPR). Upregulation of miR-1-3p, miR-20b-5p, miR-126-3p, miR-130b-3p, and miR-499a-5p was observed in SGA pregnancies only at 10.0% FPR. Upregulation of miR-145-5p, miR-342-3p, and miR-574-3p was detected in FGR pregnancies at 10.0% FPR. The combination of four microRNA biomarkers (miR-1-3p, miR-20a-5p, miR-146a-5p, and miR-181a-5p) was able to identify 75.68% SGA pregnancies at 10.0% FPR in early stages of gestation. The detection rate of SGA pregnancies without PE increased 4.67-fold (75.68% vs. 16.22%) when compared with the routine first-trimester screening for PE and/or FGR based on the criteria of the Fetal Medicine Foundation. The combination of seven microRNA biomarkers (miR-16-5p, miR-20a-5p, miR-145-5p, miR-146a-5p, miR-181a-5p, miR-342-3p, and miR-574-3p) was able to identify 42.68% FGR pregnancies at 10.0% FPR in early stages of gestation. The detection rate of FGR pregnancies without PE increased 1.52-fold (42.68% vs. 28.05%) when compared with the routine first-trimester screening for PE and/or FGR based on the criteria of the Fetal Medicine Foundation. Cardiovascular disease-associated microRNAs represent promising early biomarkers with very suitable predictive potential for SGA or FGR without PE to be implemented into the routine screening programs.

The definitions of early-onset (before 32 weeks of gestation) and late-onset (after 32 weeks of gestation) fetal growth restriction (FGR) in the absence of congenital anomalies derived from various types of cells present in the human body. On the other hand, whole peripheral blood leukocyte lysates usually reflect intracellular expression profile corresponding with actual systemic response of the human body to the current stage (normal or abnormal course of gestation).

Patients Cohort
Singleton pregnancies of Caucasian descent only were involved in the retrospective case-control study held within the period November 2012 to March 2020. The whole peripheral venous blood samples were collected during the first-trimester gestation at 10-13 weeks in the Institute for the Care of Mother and Child, Prague, Czech Republic. Finally, 4469 out of 6440 pregnancies had complete medical records since they were delivered in the Institute for the Care of Mother and Child, Prague, Czech Republic. Out of these 4469 pregnancies, 82 women were confirmed to have growth-restricted fetuses, and 37 women were diagnosed to carry small-for-gestational-age fetuses. In detail, 5 FGR pregnancies were delivered before 32 weeks of gestation (early FGR) and 77 pregnancies after 32 weeks of gestation (late FGR).
The control group included 80 women with a normal course of gestation that delivered healthy infants after completing 37 weeks of gestation with a weight above 2500 g. The control group was selected with regard to the uniformity of gestational age at sampling and storage times of biological samples.
The clinical characteristics of pregnancies are outlined in Table 1.
All participants signed informed consent. The Ethics Committee of the Institute for the Care of the Mother and Child and The Ethics Committee of the Third Faculty of Medicine, Charles University, approved the study (Implication of microRNAs in maternal circulation for diagnosis and prediction of placental insufficiency, date of approval: 7 April 2011). All procedures were in compliance with the Helsinki Declaration of 1975, as revised in 2000.

Prediction of the Occurrence of PE and FGR
Astraia Obstetrics Programme (Astraia -software for women s health, obstetric and gynaecological database, application version 1.27.3; Astraia FMF -first trimester screening for trisomy 21, version 4.4, Germany) was used to calculate the risks for chromosomal aneuploidies (trisomy 21, 18, and 13), preeclampsia, fetal growth restriction, and preterm delivery in the first trimester of gestation using the FMF risk algorithm [11] developed in close collaboration with the Fetal Medicine Foundation.

Processing of Samples
Homogenized leukocyte lysates containing lymphocytes, monocytes, and granulocytes were prepared from 200 µL maternal whole peripheral venous blood samples immediately after collection using QIAamp RNA Blood Mini Kit (Qiagen, Hilden, Germany) according to manufacturer's instructions Firstly, lysis of erythrocytes was performed using EL buffer and then pelleted leukocytes were stored in a mixture of RLT buffer and β-mercaptoethanol (β-ME) at −80 • C until further processing. β-mercaptoethanol eliminates ribonucleases released during cell lysis. Numerous disulfide bonds make ribonucleases very stable enzymes, so β-mercaptoethanol is used to reduce these disulfide bonds and irreversibly denature the proteins. This prevents them from digesting the RNA during its storage and subsequent extraction procedure. Homogenized cell lysates can be stored like this at −80 • C for several months. To keep the requirements of the manufacturer, we performed real-time RT-PCR analyses regularly every six months to process the collection of frozen samples derived from normal and complicated pregnancies.
Subsequently, a mirVana microRNA Isolation kit (Ambion, Austin, TX, USA) was used to isolate RNA fractions highly enriched for small RNAs from whole peripheral blood leukocyte lysates.
mRNAs of appropriate microRNAs were reverse transcribed into cDNA by TaqMan MicroRNA Assays containing miRNA-specific stem loop primers and TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems, Branchburg, USA). The total reaction volume was 10 µL. Furthermore, 3 µL of cDNA were mixed with the components of TaqMan MicroRNA Assays containing specific primers and the TaqMan MGB probes and the components of the TaqMan Universal PCR Master Mix (Applied Biosystems, Branchburg, NJ, USA). The total reaction volume was 15 µL. Reverse transcription and realtime qPCR were performed on 7500 Real-Time PCR System using TaqMan PCR conditions set up in the TaqMan guidelines. Reverse transcription thermal cycling parameters were the following: 30 min at 16 • C, 30 min at 42 • C, 5 min at 85 • C, and then held at 4 • C. Real-time qPCR thermal cycling parameters were the following: 2 min at 50 • C, 10 min at 95 • C, then 50 cycles at 95 • C for 15s and 60 • C for 1 min.
Assessment of microRNA gene expression was performed using the comparative Ct method [22]. The geometric mean of Ct values of selected endogenous controls (RNU58A and RNU38B) was used as a normalization factor [23] to normalize microRNA gene expression. Selection and validation of endogenous controls for microRNA expression studies in whole peripheral blood samples affected by pregnancy-related complications have already been described in our previous study [24]. In brief, expression of 20 candidate endogenous controls (HY3, RNU6B, RNU19, RNU24, RNU38B, RNU43, RNU44, RNU48, RNU49, RNU58A, RNU58B, RNU66, RPL21, U6 snRNA, U18, U47, U54, U75, Z30 and cel-miR-39) was investigated using NormFinder [25]. RNU58A and RNU38B were identified as the most stable small nucleolar RNAs (ncRNA) and equally expressed between patients with normal and abnormal courses of gestation. Therefore, these ncRNAs were selected as the most suitable endogenous controls for normalization of microRNA qPCR expression studies performed on whole peripheral blood samples affected by pregnancy-related complications.

Statistical Analysis
With regard to non-normal distribution of the data, microRNA gene expression was compared between normotensive pregnancies subsequently diagnosed to carry AGA, SGA, and FGR fetuses using the Kruskal-Wallis one-way analysis of variance. Afterward, a posthoc test for the comparison among multiple groups and Benjamini-Hochberg correction for multiple comparisons was applied (Table 2) [26,27]. Box plots display the median, the 75th and 25th percentiles (the upper and lower limits of the boxes), the maximum and minimum values that are no more than 1.5 times the span of the interquartile range (the upper and lower whiskers), outliers (circles), and extremes (asterisks), respectively. Statistica software (version 9.0; StatSoft, Inc., Tulsa, OK, USA) was used to produce box plots.
Receivers operating characteristic (ROC) curve analyses state the areas under the curves (AUC), the best cut-off points related sensitivities, specificities, positive and negative likelihood ratios (LR+, LR-), sensitivities at 10.0% false positive rate (FPR), respectively (MedCalc Software bvba, Ostend, Belgium). To select the optimal microRNA combinations, logistic regression with subsequent ROC curve analyses were applied (MedCalc Software bvba, Ostend, Belgium).

Results
Expression profiles of cardiovascular disease-associated microRNAs in the whole peripheral venous blood were compared during the first trimester of gestation between women with appropriate-for-gestational-age (AGA) fetuses, small-for-gestational-age (SGA) fetuses, and fetal growth-restricted (FGR) fetuses. Results that reached a statistical significance after Benjamini-Hochberg correction for multiple comparisons are discussed below in detail (Supplementary Figure S1). Statistical non-significant data gained after the application of Benjamini-Hochberg correction are also displayed (Supplementary Figure S2) but not discussed further.
In our group of FGR pregnancies, the first-trimester screening detected 23 out of 82 pregnancies (28.05%) at risk of the development of PE and/or FGR using the criteria of the Fetal Medicine Foundation [10]. Prophylactic use of ASA was subsequently administered in total to 21 out of 82 women (25.61%). ASA was indicated as prevention of PE/FGR development on the basis of the positive results of the first-trimester screening for PE/FGR (n = 19) or with regard to the history of PE or FGR in anamnesis (n = 2). In four patients with the positive results of the first-trimester screening for PE or FGR, the usage of ASA was contraindicated.
The comparison of the predictive results of the routine first-trimester screening for PE or FGR based on the criteria of the Fetal Medicine Foundation [10] and the first-trimester screening for FGR using a panel of seven cardiovascular disease-associated microRNAs only revealed that the detection rate of FGR in our group of pregnancies increased 1.52-fold (42.68% vs. 28.05%).

The High Accuracy of First Trimester Combined MicroRNA Screening to Differentiate between SGA and Normal Pregnancies
The combined screening of four microRNA biomarkers (miR-1-3p, miR-20a-5p, miR-146a-5p, and miR-181a-5p) was able to detect at 10.0% FPR in early stages of gestation aberrant microRNA expression profile in 75.68% SGA pregnancies (AUC 0.868, p < 0.001, 75.68% sensitivity, 90.0% specificity, cut off >0.3664) (Figure 2). In our group of SGA pregnancies, the first-trimester screening detected 6 out of 37 pregnancies (16.22%) at risk of the development of PE and/or FGR using the criteria of the Fetal Medicine Foundation [10]. Prophylactic use of ASA was finally administered to only 2 out of 37 women (5.40%) on the basis of the positive results of the first-trimester screening for PE or FGR.
The comparison of the predictive results of the routine first-trimester screening for PE or FGR based on the criteria of the Fetal Medicine Foundation [10] and the first-trimester screening for SGA using a panel of four cardiovascular disease-associated microRNAs only revealed that in our group of pregnancies the detection rate of SGA increased 4.67-fold (75.68% vs. 16.22%).
To our knowledge, no data on the prediction of SGA or FGR using microRNA profiling in maternal whole peripheral blood cell lysates during the early stages of gestation were reported. Parallel, few studies were dedicated to the assessment of the predictive potential of early circulating microRNA biomarkers for the development of FGR or SGA, which differs from biomarkers explored under the current study.
The study of Pei et al. reported a correlation of plasma miR-590-3p expression levels with FGR risk, but in middle (20th and 21st weeks of gestation) and late pregnancies (33th and 34th weeks of gestation) only, not in early stages of gestation (10th and 11th weeks of gestation). Increased plasma miR-590-3p levels were observed to be associated with an increased FGR risk via interaction with genes for vascular endothelial growth factor (VEGF), placental growth factor (PIGF), and matrix metalloproteinase 9 (MMP9), whose expression levels were significantly decreased [29].
Our group previously demonstrated that maternal plasma exosomal profiling of selected C19MC microRNAs revealed a novel microRNA biomarker (miR-520a-5p) that was down-regulated in early stages of gestation in pregnancies subsequently diagnosed to carry FGR fetuses [30]. Unfortunately, no circulating C19MC microRNA biomarkers were identified to predict the occurrence of FGR if maternal plasma samples were screened during the first trimester of gestation [31].
Recently, we have observed upregulation of miR-1-3p in a significant proportion of patients with chronic hypertension (51.72% at 10.0% FPR) [30]. The current study revealed that miR-1-3p expression is also dysregulated during the early stages of gestation in more than half of pregnancies carrying SGA fetuses (56.76% at 10.0% FPR).
Parallel, not long ago, we found out upregulation of miR-20a-5p and miR-146a-5p during early stages of gestation in women with chronic hypertension (44.83% and 65.52% at 10.0% FPR) and normotensive women with later occurrence of PE (33.33% and 42.42% at 10.0% FPR) [32]. At present, it is evident that upregulation of miR-20a-5p and miR-146a-5p also appears during the first trimester of gestation in a substantial proportion of pregnancies carrying SGA (59.46% and 64.86% at 10.0% FPR) or FGR fetuses (30.49% and 37.80% at 10.0% FPR).
Alongside, it is evident that miR-181a-5p upregulation may be detected in early stages of gestation not only in normotensive women subsequently developing hypertensive pregnancy-related complications such as GH (22.89% at 10.0% FPR) or PE (40.91% at 10.0% FPR) [32] but as well as in pregnancies affected with SGA (32.43% at 10.0% FPR) or FGR (30.49% at 10.0% FPR).
At the same time, upregulation of miR-145-5p and miR-574-3p is present in the early stages of gestation in a smaller proportion of normotensive pregnancies with subsequent onset of PE (21.21% and 27.27% at 10.0% FPR) [32] and those ones later diagnosed with FGR (18.29% and 26.83% at 10.0% FPR).
First-trimester screening in the whole peripheral venous blood based on the assessment of gene expression of these four selected cardiovascular disease-associated microRNAs (miR-1-3p, miR-20a-5p, miR-146a-5p, and miR-181a-5p) only led to the undeniable increase in the detection rate of SGA pregnancies when compared with the routine first-trimester screening for PE and/or FGR based on the criteria of the Fetal Medicine Foundation [11]. The detection rate of SGA pregnancies increased using this molecular biology approach 4.67-fold (75.68% vs. 16.22%). The combination of four upregulated microRNA biomarkers (miR-1-3p, miR-20a-5p, miR-146a-5p, and miR-181a-5p) seems to be superior over other microRNA combinations or the performance of individual microRNA biomarkers since it is able to reveal during early stages of gestation, majority of pregnancies carrying SGA fetuses (75.68% at 10.0% FPR).
Similarly, the detection rate of FGR pregnancies was increased 1.52-fold (42.68% vs. 28.05%) when compared with the results of the risk analysis using the criteria of the Fetal Medicine Foundation [11]. Nevertheless, in case of FGR pregnancies, the combination of seven microRNA biomarkers (miR-16-5p, miR-20a-5p, miR-145-5p, miR-146a-5p, miR-181a-5p, miR-342-3p, and miR-574-3p) was needed. The combination of these 7 selected microRNA biomarkers showed the best performance over other combinations of microRNA biomarkers or individual microRNA biomarkers and revealed at 10.0% FPR at early stages of gestation 42.68% of pregnancies diagnosed during the third trimester of gestation with FGR without PE.
Multiple predicted targets of appropriate dysregulated microRNAs have already been reported to be involved in particular human biological pathways related to abnormal pregnancy outcomes and their associated cardiovascular risk. These biological pathways include, for example, apoptosis pathway, an inflammatory response pathway, senescence and autophagy pathways [120], insulin signaling pathway, type 1 diabetes mellitus pathway, and type 2 diabetes mellitus pathway [121].
In addition, interactions between microRNAs dysregulated in the whole peripheral blood of preterm-born children and specific genes related to human disease ontologies, such as hypertension, congenital heart defects, patent foramen ovale, patent ductus arteriosus, and heart valve disease, were also demonstrated in our previous study [122].
Furthermore, we previously reported an extensive file of predicted targets of all microRNAs aberrantly expressed in the whole peripheral blood of children descending from GDM complicated pregnancies. A large group of these genes is involved in ontologies of human diseases (obesity, hypertension, glucose intolerance, lipid metabolism disease, type 2 diabetes mellitus, heart septal defects and heart valve disease, heart disease, heart failure, venous insufficiency, and pulmonary embolism) [123].
However, it seems probable that microRNAs represent promising biomarkers with very suitable predictive potential for SGA or FGR and might be implemented into current routine first-trimester screening programs to predict the later occurrence of SGA or FGR. The comparison of the predictive results of the routine first-trimester screening based on the criteria of the Fetal Medicine Foundation [10] and the first-trimester screening using panels of selected cardiovascular disease-associated microRNAs only revealed that the detection rates of SGA and FGR increased 4.67-fold and 1.52-fold. The combination of four microRNA biomarkers (miR-1-3p, miR-20a-5p, miR-146a-5p, and miR-181a-5p) seems to be optimal to predict the later occurrence of SGA without PE. The combination of seven microRNA biomarkers (miR-16-5p, miR-20a-5p, miR-145-5p, miR-146a-5p, miR-181a-5p, miR-342-3p, and miR-574-3p) is needed to identify in early stages of gestation subsequent occurrence of FGR without PE.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/biomedicines10030718/s1, Figure S1: Gene expression of cardiovascular disease-associated mi-croRNAs in peripheral blood leukocytes in early stages of gestation-comparison between appropriatefor-gestational-age (AGA) fetuses, small-for-gestational-age (SGA) fetuses, and fetal growth-restricted (FGR) fetuses; statistical significant data after Benjamini-Hochberg correction., Figure S2: Gene expression of cardiovascular disease-associated microRNAs in peripheral blood leukocytes in early stages of gestation-comparison between appropriate-for-gestational-age (AGA) fetuses, small-for-gestational-age (SGA) fetuses, and fetal growth-restricted (FGR) fetuses-statistical non-significant data after Benjamini-Hochberg correction.  Institutional Review Board Statement: The approval with the study was gained initially from the Ethics Committee of the Third Faculty of Medicine, Charles University (Implication of placentalspecific microRNAs in maternal circulation for diagnosis and prediction of pregnancy-related complications, date of approval: 7 April 2011). Ongoing approvals with the study were gained from the Ethics Committee of the Third Faculty of Medicine, Charles University (Long-term monitoring of complex cardiovascular profiles in mother, fetus, and offspring descending from pregnancy-relatedcomplications, dates of approval: 27 March 2014) and the Ethics Committee of the Institute for the Care of the Mother and Child, Charles University (Long-term monitoring of complex cardiovascular profiles in mother, fetus, and offspring descending from pregnancy-related-complications, dates of approval: 28 May 2015, number of approval: 1/4/2015).

Informed Consent Statement:
Informed consent with the study was gained from patients during the first trimester of gestation when the collection of peripheral blood samples for the first-trimester screening was held. Informed consent was signed by all pregnant women involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to rights reserved by funding supporters.