MicroRNAs (miRNAs), small non-coding RNAs, have emerged as important, epigenetic regulators of endothelial function. Metabolic disturbances such as diabetes alter miRNA expression. In adults, the miRNA transcriptome as well as endothelial function differ between the sexes. Here, we hypothesized that metabolic disturbances associated with gestational diabetes (GDM) alter miRNA signatures in feto-placental endothelial cells (fpEC), dependent on fetal sex. We isolated human primary fpEC after normal and GDM-complicated pregnancies with male and female neonates and screened for differential miRNA expression using next-generation miRNA sequencing. To test for miRNAs commonly regulated in fpEC of female and male progeny, data were stratified for fetal sex and maternal body mass index (BMI). Analyses were also performed separately for female and male fpEC, again accounting for maternal BMI as covariate. Potential biological pathways regulated by the altered set of miRNAs were determined using mirPath software. Maternal GDM altered 26 miRNA signatures when male and female fpEC were analyzed together. Separate analysis of male versus female fpEC revealed 22 GDM affected miRNAs in the females and only 4 in the males, without overlap. Biological functions potentially modulated by the affected miRNAs related to ‘Protein Processing in Endoplasmic Reticulum’ and ‘Proteoglycans in Cancer’. Maternal GDM alters miRNA signatures in fpEC, and biological functions affected by these miRNAs relate to well-known adverse functional consequences of diabetes on endothelium. GDM effects were highly dependent on fetal sex with miRNA signatures in female fpEC being more susceptible to metabolic derangements of GDM than miRNAs in male fpEC.

Introduction

MicroRNAs (miRNAs) are an abundant class of small, non-protein coding RNAs that regulate expression of more than 60% protein coding genes [1]. MiRNAs function by binding to complementary mRNA regions, but as their action does not require a perfect base pairing, a single miRNA can bind to various target genes, and vice versa, one target gene may be subject to regulation by several distinct miRNAs. This highlights the complexity of miRNAs as important fine-tune regulators of distinct biological processes including processes involved in endothelial function [2].

Normal endothelial function includes the formation of new blood vessels (angiogenesis), maintenance of the endothelial barrier and regulation of vascular tone. In diabetes, an altered metabolic, hormonal and inflammatory environment modulates or even impairs endothelial function in long term (reviewed by [3]), and an arisen endothelial dysfunction increases the risk for cardiovascular disease [4]. In diabetic adults, also altered miRNA levels have been identified [5] and associated with endothelial dysfunction [6].

In gestational diabetes (GDM) not only maternal, but also fetal circulation/intrauterine environment is characterized by hyperglycemia and hyperinsulinemia, and altered levels of hormones and growth factors can impact the feto-placental endothelial function (reviewed by [7]). For instance, GDM induces increased capillary branching [8] and capillary surface area [9] of the placental vasculature. Moreover, surface expression of several adherens and tight junctional molecules in the feto-placental endothelium is decreased in GDM [10] as is the response to vaso-relaxing agents [11]. Differential miRNA expression between human umbilical vein endothelial cells (HUVEC) isolated after normal versus GDM pregnancy suggests that feto-placental endothelial function is susceptible to the GDM environment via altered miRNA expression [12,13].

The risk for endothelial dysfunction and cardiovascular diseases differs between men and women, and hormonal influences of androgens are regarded as main contributors [14,15]. Also micro- and macrovascular complications of diabetes display sex specific differences, with men having a higher risk for microvascular and women for macrovascular complications [16]. Moreover, although in the general population men are more susceptible towards vascular complications, presence of diabetes renders women more susceptible than men [16,17]. In mammals, gene expression differs between the sexes even before gonad development [18], potentially as a result of sex chromosomal transcripts that regulate autosomal gene expression [19]. The transcriptome of human feto-placental endothelial cells is also characterized by sex differences [20]. Based on these previous reports, we here hypothesized that the intrauterine environment associated with GDM alters miRNA signatures in feto-placental endothelium dependent on fetal sex. We used primary feto-placental endothelial cells (fpEC) isolated after normal and GDM pregnancy with male and female progeny. The feto-placental endothelium is part of and continuous with the fetal vasculature and, hence, exposed to the same normal or diabetic environment in the circulation as the fetus. Moreover, endothelial cells isolated from the placenta are more reactive than endothelial cells isolated from umbilical cord (HUVEC) [7] and therefore, represent a good model for investigating cellular response to diabetic environment. In order to cover the full array of miRNAs expressed in our cell model, the effect of GDM on miRNA expression was analyzed using next-generation miRNA sequencing. Altered miRNA signatures were further subjected to pathway analysis.

Methods

Sample collection

Ethical approval was obtained from the Medical University of Graz (approval reference number 27-268 ex 14/15) and all women provided written informed consent. Control placentas were collected from pregnancies of non-smoking (self-reported) women with a negative 75  g oral glucose tolerance test (oGTT) performed at 25–28 weeks of gestation, free from any medical disorders or pregnancy complications. GDM placentas were collected after GDM was diagnosed after oGTT according to WHO/IADPSG criteria [21]. Again, only women without other medical disorders were included. After positive oGTT, women were either recommended a diet and classified as GDM A1 or additionally treated with insulin (Novorapid plus Insulatard; Novo Nordisk Pharma, Wien, Austria) and classified as GDM A2. Expression of miRNAs that were identified as significantly different between control and GDM fpEC by miRNA sequencing revealed no correlation with the type of diabetes, i.e. GDM A1 or GDM A2. Control and GDM samples were matched for ethnicity. Maternal body mass index (BMI) differed between control and GDM group, since overweight is a major risk factor for GDM. There was no difference in the passage in which cells were used between the control and the GDM groups. Additionally, for the two miRNAs revealing the highest expression difference between control and GDM cells, we demonstrated unchanged miRNA levels throughout culture between passages 2 and 10 (Supplementary Figure S1). Table 1 shows subjects’ characteristics of the discovery cohort used for miRNA sequencing, and Supplementary Table S1 shows subjects’ characteristics of the validation cohort.

Principle component analysis plot based on the 181 miRNAs that were used for statistical analysis

Figure 1
Principle component analysis plot based on the 181 miRNAs that were used for statistical analysis

The orange oval encompasses female control samples, the red oval female GDM samples, except of DF3. CF1-9: female control fpEC; DF1-7: female GDM fpEC; CM1-5: male control fpEC; DM1-7: male GDM fpEC.

Figure 1
Principle component analysis plot based on the 181 miRNAs that were used for statistical analysis

The orange oval encompasses female control samples, the red oval female GDM samples, except of DF3. CF1-9: female control fpEC; DF1-7: female GDM fpEC; CM1-5: male control fpEC; DM1-7: male GDM fpEC.

Table 1
Subjects’ characteristics of samples for miRNA sequencing
 Controls GDM Female controls Female GDM Male controls Male GDM 
Number of cases 14 14 
Type of GDM (A1/A2)  8/6  4/3  4/3 
Mode of delivery (vaginal/C-section) 5/9 4/10 3/6 2/5 2/3 2/5 
Maternal characteristics       
Age (years) 29.8 ± 7.8 30.5 ± 4.7 28.3 ± 6.9 30.5 ± 5.7 32.2 ± 9.3 30.6 ± 4.2 
Prepregnancy BMI 22.4 ± 3.3 28.4 ± 7.9* 23.2 ± 3.6 29.6 ± 8.6 20.8 ± 2.1 27.1 ± 7.5 
BMI at delivery 26.6 ± 4.2 31.0 ± 6.2 27.4 ± 5.5 31.7 ± 6.7 25.6 ± 2.2 30.1 ± 6.1 
Gestational age (weeks) 39.0 ± 1.0 39.3 ± 1.0 39.2 ± 0.6 39.5 ± 1.1 38.6 ± 1.6 39.2 ± 1.0 
Neonatal characteristics       
Weight (g) 3365 ± 366 3566 ± 218 3264 ± 264 3585 ± 238 3533 ± 512 3547 ± 231 
Length (cm) 51.6 ± 3.0 51.6 ± 2.3 50.2 ± 2.3 51.5 ± 2.4 52.0 ±2.3 51.8 ± 2.5 
Placental characteristics       
Placental weight (g) 618 ± 112 663 ± 153 588 ± 123 763 ± 218 666 ± 78 613 ± 98 
Primary cell characteristics       
Passage number 7.4 ± 1.3 8.1 ± 1.4 7.8 ± 1.4 8.8 ± 0.8 6.8 ± 0.8 7.0 ± 0.8 
 Controls GDM Female controls Female GDM Male controls Male GDM 
Number of cases 14 14 
Type of GDM (A1/A2)  8/6  4/3  4/3 
Mode of delivery (vaginal/C-section) 5/9 4/10 3/6 2/5 2/3 2/5 
Maternal characteristics       
Age (years) 29.8 ± 7.8 30.5 ± 4.7 28.3 ± 6.9 30.5 ± 5.7 32.2 ± 9.3 30.6 ± 4.2 
Prepregnancy BMI 22.4 ± 3.3 28.4 ± 7.9* 23.2 ± 3.6 29.6 ± 8.6 20.8 ± 2.1 27.1 ± 7.5 
BMI at delivery 26.6 ± 4.2 31.0 ± 6.2 27.4 ± 5.5 31.7 ± 6.7 25.6 ± 2.2 30.1 ± 6.1 
Gestational age (weeks) 39.0 ± 1.0 39.3 ± 1.0 39.2 ± 0.6 39.5 ± 1.1 38.6 ± 1.6 39.2 ± 1.0 
Neonatal characteristics       
Weight (g) 3365 ± 366 3566 ± 218 3264 ± 264 3585 ± 238 3533 ± 512 3547 ± 231 
Length (cm) 51.6 ± 3.0 51.6 ± 2.3 50.2 ± 2.3 51.5 ± 2.4 52.0 ±2.3 51.8 ± 2.5 
Placental characteristics       
Placental weight (g) 618 ± 112 663 ± 153 588 ± 123 763 ± 218 666 ± 78 613 ± 98 
Primary cell characteristics       
Passage number 7.4 ± 1.3 8.1 ± 1.4 7.8 ± 1.4 8.8 ± 0.8 6.8 ± 0.8 7.0 ± 0.8 
*

indicates significance between controls and GDM cases when female and male groups were combined.

indicates significance between controls and GDM cases only in the female, or the male group, respectively. Data are given as mean ± SD.

Cell culture

Primary arterial feto-placental endothelial cells were isolated from third trimester human placentas after healthy and GDM complicated pregnancies following a standard protocol [22]. Cells were characterized by immuno-cytochemical analysis and internalization of acetylated low-density-lipoprotein (Biomedical Technologies, Stoughton MA) and cultured on 1% (v/v) gelatin-coated flasks (75 cm²) using Endothelial Basal Medium (EBM, Cambrex, Clonetics, Walkersville, MD) supplemented with the EGM-MV BulletKit (Clonetics) under normoglycemic conditions. For RNA isolation, cells were grown in 75 cm² flasks to approximately 90% confluency, washed with ice cold phosphate buffered saline (PBS) and harvested as described below. Sex of the neonates from whom the fpEC isolations derived was verified by quantitative reverse-transcription PCR (RT-qPCR) for sex-specific genes as described below.

RNA isolation and quality control

One milliliter of TRIzol (Thermo Fisher Scientific, Waltham, MA) was added to each flask and cells were detached by scraping, homogenized by vortexing and stored at −80°C for subsequent RNA isolation. For optimal miRNA isolation, the RNA precipitation step following phase separation was modified: 700 µl 100% ethanol was added to the aqueous phase placed at −80°C overnight. Then, the precipitated RNA was centrifuged at 12000×g and 4°C for 30 min, after which the protocol was followed according to the manufacturer’s instructions. Total RNA concentration was determined using QIAxpert (Qiagen, Hilden, Germany). MicroRNA yield and RNA integrity were evaluated using Agilent 2100 Bioanalyzer system (Agilent Technologies, Santa Clara, CA) and the Agilent RNA 6000 Pico kit (Agilent Technologies). Only samples with RNA integrity numbers between 7.5 and 10 were used further for miRNA analysis.

RT-qPCR for sex determination

Sex of fpEC was confirmed by assaying the expression of the DDX3Y gene (DEAD (Asp-Glu-Ala-Asp) box polypeptide 3, Y-linked) on the Y and of the XIST gene on the X chromosome [23]. Complementary DNA (5 ng) was used for multiplex RT-qPCR with TaqMan assays (Applied Biosystems, Thermo Fisher Scientific). DDX3Y gene was detected by a FAM-labeled TaqMan assay (Hs01079824_m1), and XIST gene by a VIC-labeled assay (Hs00965254_gH). Expression of the housekeeping gene HPRT1 (Hypoxanthine phosphoribosyltransferase 1, Hs02800695_m1) was used for validation of the PCR performance. Sex-specific genes were regarded as expressed when the Ct values were <35.

miRNA sequencing

Total RNA (1 µg) enriched for miRNA was used for generating next-generation sequencing libraries using the Ion Total RNA-Seq Kit v2 (Thermo Fisher Scientific). Preparation was performed following the instructions for generating small RNA libraries (Ion Total RNA-Seq Kit v2 for Small RNA Libraries, Thermo Fisher Scientific). Briefly, adapters were ligated to total RNA molecules followed by reverse transcription. DNA fragments were purified and size selected to enrich fragments of 94 to 114 base pairs. Libraries were post-amplified for 14 cycles and library concentration was determined using the Ion Library TaqMan™ Quantitation Kit (Thermo Fisher Scientific). Equimolar concentrations of up to 16 barcoded miRNA libraries were pooled and sequenced using the Ion Proton System (Thermo Fisher Scientific) with the Ion PI™ Hi-Q™ Sequencing 200 Kit (Thermo Fisher Scientific). A total of 97 million reads was obtained with an average of 3.4 million reads per sample. Identification of miRNA reads was performed using the CAP-miRSeq v1.1 workflow [24] yielding on average 800,000 miRNA read counts per sample.

MiRNA sequencing data analysis

Differentially expressed miRNAs were identified based on a negative binomial distribution from raw count data using the R/Bioconductor package edgeR [25]. Only miRNAs with 150 counts in at least three samples were further considered. To test differences between GDM exposed and control fpEC, a generalized linear model (GLM) was used including disease state as the main factor. Fetal sex and maternal pre-pregnancy BMI were included as categorical and continuous covariates, respectively, and a likelihood ratio test was performed. To test the interaction between the disease state and fetal sex, a complete GLM was used including both factors and their interaction, and maternal pre-pregnancy BMI was included as covariate. The analyses were also performed separately for female and male fpEC, again accounting for maternal BMI as covariate. P-values were adjusted for multiple testing based on the false discovery rate (FDR) according to the Benjamini–Hochberg method. Reflecting the low sample size, only one of the measured miRNAs reached 5% FDR. Nevertheless, data were validated in a different cohort and we proceeded by using unadjusted P-value <0.05. Normalized data (counts per million; CPM) were log2-transformed (log2(CPM+1)) for further analyses. The variance of microRNA expression explained by pre-pregnancy BMI was estimated by the coefficient of determination (R2) of a linear regression model for each miRNA individually. Raw and processed data were deposited in GEO database (accession number GSE104297). Heat maps were generated using Genesis [26] based on z-score and principal component analysis (PCA) was performed using R.

Validation of miRNA sequencing data

From a different cohort of primary cells (validation cohort, Supplementary Table S1), cDNA was transcribed from 1 µg of total RNA using miScript II RT Kit and 5x miScript HiFlex Buffer (Qiagen) according to the manufacturers’ instructions. This facilitates parallel quantification of mRNAs, mature miRNAs, precursor miRNAs and non-coding RNAs for reverse transcription. Complementary DNA was stored at –20°C. Two nanograms of cDNA per reaction were used for RT-qPCR. PCR was performed on a CFX96 cycler (Bio-Rad Laboratories, Hercules, CA) using miScript Primer Assays (Qiagen) for mir-145-5p (Hs_miR_145_1) mir-139-5p (Hs_miR-139_1), mir-134-5p (Hs_miR-134_1) and mir-324-3p (Hs_miR-324_1) miScript SYBR® Green PCR Kit. PCR efficiency of all primer assays was tested using a 5-point standard curve with a range from 25 to 0.004 ng of pooled cDNA/reaction from male and female control fpEC isolations. All primer pairs met the criteria according to Nolan et al. [27]. Ctrl_miRTC_1 Primer Assay (Qiagen) was included for the assessment of reverse transcription performance on every cDNA sample. Primary data analysis was performed on the CFX Manager 3.1 software (Bio-Rad) using the Regression Mode for Cycle of quantity (Cq) value determination. Relative gene expression was calculated using the 2−ΔΔCT method and results were normalized to the mean of RNU-6 (RNA, U6 small nuclear 6, pseudogene) and mir-103a-3p reference genes (miScript Primer Assays Hs_RNU6_2_11 and Hs_miR-103a_1) [28].

In order to determine whether cell passage affects miRNA expression, fpECs (n=4) were cultured after isolation up to passage 10, cells were harvested and RNA was isolated after each expansion step. Mir-145-5p and mir-139-5p were quantified as described above.

Pathway analysis

Pathway analysis of miRNA targets was performed using DIANA mirPath v.3 online tool (http://snf-515788.vm.okeanos.grnet.gr/) [29]. Target evaluation was based on DIANA TarBase v7.0 database [30], and pathways analysis was based on KEGG pathway database.

Statistical analysis

To identify differences in the subject’s characteristics between control and GDM group, a student’s t-test was used. When fetal sex was included as a factor in the analysis, two-way ANOVA with Bonferroni as post-hoc test was used. The same tests were performed for the validation of miRNA sequencing results using the ΔCt values of the RT-qPCR. The influence of cell passage and culture was analyzed by comparing ΔCt values using one-way ANOVA with repeated measures. Spearman correlation of ΔCt values was used to identify a potential effect of GDM type, GDM A1 and GDM A2, respectively, on miRNA expression.

Results

GDM alters miRNA signatures in feto-placental endothelial cells depending on fetal sex

In order to investigate whether GDM affects the expression of miRNA in fpEC, and whether this is dependent on fetal sex, we performed next-generation miRNA sequencing of primary fpEC with male and female progeny after normal (9 females and 5 males) and GDM pregnancies (7 females and 7 males).

Next-generation sequencing analysis identified 453 miRNAs in fpEC with 181 miRNAs having more than 150 counts in at least three samples. In order to visualize group samples according to their miRNA expression profiles, the dimension reduction method PCA was applied (Figure 1). By visualization of the first two components, samples were separated according to fetal sex, except for one female control fpEC (CF7) and one female GDM isolation (DF3) clustering with male isolations. PCA analysis further separated controls and GDM samples only in the female group whilst samples of male progeny did not show such separation. Among the 181 analyzed miRNAs, 26 were significantly regulated by GDM when control fpEC were compared with GDM fpEC, and fetal sex and maternal pre-pregnancy BMI were used as covariates. As evident from Supplementary Figure S2 for some microRNAs, the estimated variance of expression by pre-pregnancy BMI was >30%. The significantly altered miRNAs, the expression fold change and the adjusted P-values are listed in Supplementary Table S2 and depicted in a volcano plot (Supplementary Figure S3). When male and female fpEC were analyzed separately with maternal BMI included as covariate, only four miRNAs were significantly altered by GDM in cells of male progeny whilst 22 were differentially expressed in cells of female progeny (Supplementary Table S2). Only two of the miRNAs sensitive to GDM in the male cells, and 14 miRNAs sensitive to GDM in the female cells were among the list of miRNAs obtained when male and female samples were analyzed together (Figure 2). Four significantly altered miRNAs were validated by RT-qPCR in an independent cohort of 21 controls (11 females and 10 males) and 18 GDM exposed cells (8 females and 10 males).

MicroRNAs significantly altered by maternal GDM in fpEC in the entire cohort, and in cells of female (pink circle) or male (blue circle) progeny only.

Figure 2
MicroRNAs significantly altered by maternal GDM in fpEC in the entire cohort, and in cells of female (pink circle) or male (blue circle) progeny only.

The black circle encompasses miRNAs significantly altered when female and male fpEC were analyzed together. The total number of significantly altered miRNAs in these three groups is indicated under the respective circles.

Figure 2
MicroRNAs significantly altered by maternal GDM in fpEC in the entire cohort, and in cells of female (pink circle) or male (blue circle) progeny only.

The black circle encompasses miRNAs significantly altered when female and male fpEC were analyzed together. The total number of significantly altered miRNAs in these three groups is indicated under the respective circles.

Heat map illustrating expression levels of miRNAs significantly altered in fpEC by GDM

Figure 3
Heat map illustrating expression levels of miRNAs significantly altered in fpEC by GDM

Heat map is based on normalized expression levels (z-scores) of miRNAs that are significantly altered by GDM when female and male fpEC were analyzed together (A), or when female (B) and male fpEC (C) were analyzed separately. Heat maps were generated using Genesis based on the z-score of the signals after adjustment for maternal BMI (A, B and C) and fetal sex (A); CTRL, signal of control samples; GDM, signal of GDM samples.

Figure 3
Heat map illustrating expression levels of miRNAs significantly altered in fpEC by GDM

Heat map is based on normalized expression levels (z-scores) of miRNAs that are significantly altered by GDM when female and male fpEC were analyzed together (A), or when female (B) and male fpEC (C) were analyzed separately. Heat maps were generated using Genesis based on the z-score of the signals after adjustment for maternal BMI (A, B and C) and fetal sex (A); CTRL, signal of control samples; GDM, signal of GDM samples.

Table 2
Validation of miRNA sequencing data by RT-qPCR
 miRNA-seq RT-qPCR in original cohort RT-qPCR in validation cohort 
 FC P-value FC P-value FC P-value 
mir-145-5p       
C vs GDM 2.19 0.023 3.48 0.009 2.26 0.001 
CF vs DF 2.79 0.007 6.16 0.002 3.51 0.002 
CM vs DM 1.96 0.251 2.03 0.356 1.93 0.187 
mir-139-5p       
C vs GDM 2.98 0.018 2.27 0.039 2.25 0.009 
CF vs DF 4.17 0.069 3.01 0.043 3.41 0.023 
CM vs DM 1.92 0.133 0.93 0.439 1.48 0.203 
mir-134-5p       
C vs GDM 1.60 0.018 2.05 0.049 1.42 0.019 
CF vs DF 1.79 0.017 1.94 0.089 2.28 0.038 
CM vs DM 1.35 0.328 1.05 0.479 1.04 0.472 
mir-324-5p       
C vs GDM –1.81 0.006 –1.44 0.231 –1.19 0.119 
CF vs DF –2.12 0.011 –1.49 0.196 –1.11 0.503 
CM vs DM –3.68 0.286 –1.57 0.341 –1.31 0.093 
 miRNA-seq RT-qPCR in original cohort RT-qPCR in validation cohort 
 FC P-value FC P-value FC P-value 
mir-145-5p       
C vs GDM 2.19 0.023 3.48 0.009 2.26 0.001 
CF vs DF 2.79 0.007 6.16 0.002 3.51 0.002 
CM vs DM 1.96 0.251 2.03 0.356 1.93 0.187 
mir-139-5p       
C vs GDM 2.98 0.018 2.27 0.039 2.25 0.009 
CF vs DF 4.17 0.069 3.01 0.043 3.41 0.023 
CM vs DM 1.92 0.133 0.93 0.439 1.48 0.203 
mir-134-5p       
C vs GDM 1.60 0.018 2.05 0.049 1.42 0.019 
CF vs DF 1.79 0.017 1.94 0.089 2.28 0.038 
CM vs DM 1.35 0.328 1.05 0.479 1.04 0.472 
mir-324-5p       
C vs GDM –1.81 0.006 –1.44 0.231 –1.19 0.119 
CF vs DF –2.12 0.011 –1.49 0.196 –1.11 0.503 
CM vs DM –3.68 0.286 –1.57 0.341 –1.31 0.093 

RT-qPCR was performed in the original cohort that was used for miRNA sequencing and in a novel validation cohort.

FC: fold change; CF, CM: control fpEC of female or male progeny; DF, DM: GDM exposed fpEC of female or male progeny. ‘C vs GDM’ indicates results obtained when male and female samples were analyzed together, the lines ‘CF vs DF’ and ‘CM vs DM’ reveal data obtained when male and female groups were analyzed separately.

These miRNAs were selected as they were significantly regulated by GDM in the overall group and possessed high basal expression. Moreover, when analyzed sex dependently, three of the miRNAs selected for validation, i.e. mir-145-5p, mir-134-5p and mir-324-5p, were only regulated by GDM in the female, but not in the male cohort, thus enabling validation also of the sex-specific GDM effects. RT-qPCR confirmed general and sex-specific GDM effects for mir-145-5p and mir-134-5p, whilst data did not reach significance for mir-324-5p. According to miRNA sequencing, mir-139-5p was among the miRNAs altered by GDM independently of fetal sex, but RT-qPCR revealed that mir-139-5p was also significantly affected by GDM in the female group only (Table 2).

Two microRNAs with the highest fold change between normal and GDM exposed fpEC, mir-145-5p and mir-139-5p were chosen to demonstrate that the observed differences in miRNA expression were not due to differences in cell passage. Expression of both miRNAs was unaltered between passage 2 and 10 in four different fpEC isolations (Supplementary Figure S1). Heat maps based on expression levels (z-scores) with sex and maternal BMI as covariate illustrate the GDM effect on miRNA regulation (Figure 3). MicroRNAs that were significantly altered by GDM in the analysis with both sexes combined showed a clear GDM effect in the female group (Figure 3A). The heat maps in Figure 3B and C illustrate the miRNAs significantly altered by GDM in the female and male cells only, with again, a clear GDM effect in the cells of female progeny.

Pathway analysis of GDM regulated miRNAs

One specific miRNA may target mRNA of several genes, and one mRNA can be targeted by distinct miRNAs as well. In order to predict biological functions that may be affected by differently expressed miRNAs in GDM, we performed pathway analysis using mirPath online software tool based on TarBase target prediction and KEGG pathways. Most significantly altered pathways in GDM were ‘MicroRNAs in Cancer’, ‘Proteoglycans in Cancer’ and ‘Protein Processing in Endoplasmic Reticulum’ (Table 3). In female fpEC, miRNAs altered by GDM clustered to ‘Proteoglycans in Cancer’, ‘Hippo Signaling Pathway’ and ‘Protein Processing in Endoplasmic Reticulum’ with highest significance. The four miRNAs significantly altered by GDM in male fpEC were most significantly attributed to ‘ECM Receptor Interaction’ pathway (Table 3).

Table 3
Functional pathways associated with differentially regulated miRNAs in fpEC by GDM
KEGG pathways P-value #genes #miRNAs 
Control vs GDM exposed fpEC 
MicroRNAs in cancer 2.02E-77 133 25 
Proteoglycans in cancer 4.35E-17 144 25 
Protein Processing in ER 1.96E-13 130 24 
Hippo signaling pathway 1.30E-08 102 25 
Renal cell carcinoma 1.54E-08 55 24 
Control vs GDM exposed fpEC of female progeny 
Proteoglycans in cancer 1.32E-11 147 17 
Hippo signaling pathway 8.30E-11 109 16 
Protein processing in ER 1.69E-08 118 15 
Adherens junction 1.18E-06 54 15 
TGF-β signaling pathway 4.10E-06 57 14 
Control vs GDM exposed fpEC of male progeny 
ECM–receptor interaction 1.33E-30 22 
Fatty acid biosynthesis 5.81E-14 
Viral carcinogenesis 1.08E-11 54 
Colorectal cancer 8.44E-07 25 
Proteoglycans in cancer 7.90E-06 47 
KEGG pathways P-value #genes #miRNAs 
Control vs GDM exposed fpEC 
MicroRNAs in cancer 2.02E-77 133 25 
Proteoglycans in cancer 4.35E-17 144 25 
Protein Processing in ER 1.96E-13 130 24 
Hippo signaling pathway 1.30E-08 102 25 
Renal cell carcinoma 1.54E-08 55 24 
Control vs GDM exposed fpEC of female progeny 
Proteoglycans in cancer 1.32E-11 147 17 
Hippo signaling pathway 8.30E-11 109 16 
Protein processing in ER 1.69E-08 118 15 
Adherens junction 1.18E-06 54 15 
TGF-β signaling pathway 4.10E-06 57 14 
Control vs GDM exposed fpEC of male progeny 
ECM–receptor interaction 1.33E-30 22 
Fatty acid biosynthesis 5.81E-14 
Viral carcinogenesis 1.08E-11 54 
Colorectal cancer 8.44E-07 25 
Proteoglycans in cancer 7.90E-06 47 

The upper panel shows regulated pathways when cells of male and female progeny were analyzed together and stratified for fetal sex (upper panel), the middle and lower panel show pathways regulated in fpEC of female (middle panel) or male progeny only (lower panel). Pathway analysis was performed using mirPath software.

# miRNAs: number of miRNAs of the list that are involved in the pathway. # genes: number of genes within the pathway that are targeted by these miRNAs.

Discussion

As epigenetically acting fine-tuners of endothelial function, the expression of miRNAs is sensitive to metabolic derangements of diabetes. For instance, in adults, diabetes alters miRNA profile in plasma [31] and in circulating angiogenic cells [32]. Moreover, in GDM, intrauterine diabetic environment alters expression of selected miRNAs in HUVEC [12,13]. Metabolic derangements, and thus potential mediators of miRNA regulation, are multifactorial in diabetes, but elevated glucose levels are likely to play a role in miRNA regulation, since high glucose was shown to affect miRNA expression in endothelial cells [33,34].

Recent research has indicated that miRNA patterns depend on the sex of the individual. For instance, sex-dependent differences in free circulating miRNAs and endothelium-derived microparticles have been identified in children [35] and healthy adults [36]. Sex-dependent miRNA secretion occurs even prenatally as shown in bovine embryos [37]. However, only few recent studies investigated whether metabolic derangements affect miRNA expression in a sex-dependent manner: animal models showed sex-dependent miRNA expression in obesity in adipose tissue [38,39], in the liver [40] as well as in the brain [41,42]. In human, placental mir-210 expression was shown to respond to maternal obesity depending on fetal sex [43]. In HUVEC, no such comparison has been undertaken.

We here investigated whether GDM affects miRNA expression in fpECs, and whether this change depends on fetal sex. Our data show that fetal sex critically determines the GDM-induced changes in miRNA signatures. In fact, there was no overlap in GDM affected miRNAs in fpEC of male versus female progeny. A strikingly higher number of miRNAs was affected by GDM in the female cells than in the male cells, indicating a higher susceptibility of female cells towards metabolic derangements of GDM. In fact, a study investigating the effect of maternal nutrient restriction on fetal cardiac structure and miRNA expression in baboons identified less miRNAs altered in male fetuses when compared with the female group [44], further highlighting that vascular miRNA expression is more responsive to maternal metabolic derangements in female fetuses.

We further performed pathway analysis to gain insight into biological pathways potentially affected by the altered pattern of miRNA expression. When male and female fpEC were analyzed together, stratified for fetal sex, GDM regulated miRNAs were most significantly attributed to the pathways ‘MicroRNAs in Cancer’, ‘Proteoglycans in Cancer’ and ‘Protein Processing in Endoplasmic Reticulum’.

Proteoglycans are proteins highly glycosylated by covalent addition of glycosaminoglycan chains. They are stabilizing components of the extracellular matrix and mediate connective tissue structure. In fact, adult diabetes is associated with altered proteoglycan levels in the vascular wall [45]. Also, GDM affects proteoglycans and alters proteoglycan composition in the placenta [46]. In the vasculature, proteoglycans modulate angiogenesis and endothelial cell behavior, and changes in their expression or composition might promote diabetic complications (reviewed by [47]).

The biological process ‘Protein Processing in Endoplasmic Reticulum’ was also affected by GDM. The rough endoplasmic reticulum (ER) is the place of protein production and folding. Adverse stimuli, such as oxidative stress, can disrupt ER homeostasis and cause increased formation of misfolded proteins that accumulate in the ER unless properly folded by chaperones, or degraded in the proteasome (reviewed by [48]). Sustained condition of ER stress can lead to the induction of apoptosis. Thus, ER stress is implicated in endothelial dysfunction [49] and Type 2 Diabetes (T2D) [50] and was shown to associate with diabetic retinopathy in mice [51].

When only miRNAs that were altered by GDM in fpEC of female progeny were subjected to pathway analysis, again ‘Proteoglycans in Cancer’ and ‘Protein Processing in Endoplasmic Reticulum’ were among the highest ranked affected functional pathways. Additionally, the ‘Hippo Signaling Pathway’ was most significantly affected. The Hippo pathway is a signaling cascade that controls organ size, proliferation, apoptosis and fate determination of stem cells (reviewed by [52]). Moreover, the Hippo pathway regulates vascular development and function of vascular cells, modulates endothelial cell quiescence and inflammation in atherosclerosis [53] and is implicated in impaired angiogenesis in diabetes [54].

Overall, the effect of GDM on miRNA-mediated cellular functions related to well-known processes affected by diabetes in adults. It therefore seems tempting to speculate a potential effect of GDM, via miRNA action, on endothelial function.

In fact, studies evaluating fetal and offspring markers of endothelial dysfunction and cardiovascular risk profile when mothers developed GDM revealed increased endothelial nitric oxide synthase and decreased superoxide dismutase levels in cord blood [55], and altered adiponectin levels in the offspring within 1 year postpartum [56]. Maternal T2D in pregnancy is associated with increased levels of circulating E-selectin and vascular cell adhesion molecule 1 in the offspring at the age 6–13 [57], reflecting an adverse endothelium perturbation. This was confirmed by a follow-up study investigating the effect of maternal Type 1 Diabetes in pregnancy on offspring at similar age. This second study also revealed increased cholesterol, HDL-cholesterol and plasminogen activator inhibitor 1 [58]. All these studies, however, did not analyze sex specific differences at the cellular level. Nevertheless, in a rat model of late-gestation diabetes mimicking human GDM, hypertension was found only in male adult offspring, with no effect in females [59], supporting the notion of sex-specific programming of endothelial (dys)function.

None of the miRNAs that were regulated by fetal sex was located on the sex chromosomes, indicating either an epigenetic regulation by marks established already in utero or a regulation by genes encoded by the sex chromosomes. An epigenetic regulation, however, seems more likely, since chromosomal location of the GDM regulated miRNAs indicated physical proximity for some of them: Of the 26 miRNAs regulated by GDM in fpEC of both, male and female progeny, 6 are part of the 14q32 miRNA cluster. Within the 22 miRNAs sensitive to GDM only in the female group, 8 miRNAs belong to this cluster. Location of miRNAs within this cluster enables common, epigenetic regulation. Indeed, various miRNAs of the 14q32 cluster target genes mediating cell–cell and cell–ECM contact, and cytoskeletal organization (reviewed in [60]) and are involved in neovascularization [61] and endothelial damage [62].

We acknowledge the low sample size as limitation of our study. Only one of the altered miRNAs remained statistically significant following FDR adjustment for multiple testing, reflecting other omics studies with small sample size [63]. This, however, does not exclude the potential for phenotype-specific associations, which we also demonstrated by validating the results in an independent cohort by RT-qPCR. Our main aim was to identify biologically relevant functions rather than individual miRNAs that are altered in GDM, and we are aware that, although the affected pathways seem nicely related to diabetic complications, further studies are required to demonstrate such functional effect. For the same reason, we did not evaluate expression of individual potential target mRNAs. The differentially expressed miRNAs in GDM exposed cells are predicted to target a total of 16,876 mRNA signatures (TarBase v.8; [64]), and potential targets often overlap between distinct miRNAs, highlighting the complex and multifactorial mechanism of fine-tuning biological functions by miRNAs.

The miRNA patterns in different samples revealed a strong biological variance. In fact, miRNAs are fine-tuners of biological functions and thus, highly responsive to and moldable by environmental factors. For instance, body composition affects miRNA expression in distinct species and tissues [38,39, 43] and thus, maternal BMI was used here as a covariate for statistical analysis.

In summary, altered miRNA expression due to GDM is more pronounced in female than in male fpEC, suggesting a higher sensitivity to diabetic derangements in female than male fpEC. This may either indicate a better adaptation of female fpEC to the disturbed environment or a more pronounced dysfunction. The first option reflects the fact that, potentially through better adaptation, female fetuses have a lower risk for peri- and postnatal mortality than male fetuses [20]. However, in order to address this particular question, additional follow-up studies are required in GDM offspring.

Clinical perspectives

  • miRNAs have emerged as important regulators that fine-tune biological processes including endothelial function, and were shown to contribute to diabetes induced endothelial dysfunction. The risk for endothelial dysfunction and cardiovascular diseases differs between adult males and females.

  • Our study demonstrated that miRNA expression pattern of female fpEC is more susceptible or responsive to the GDM environment than that of male fpEC. The effect of GDM on miRNA mediated cellular functions related to well-known processes affected in endothelial cells by diabetes in adults.

  • The effect of diabetic environment on miRNA regulated endothelial dysfunction is sex dependent.

Author Contribution

J.S., A.T. and H.M.A. performed experiments. H.H. and K.K. performed data analysis. U.H. was responsible for conception and design, manuscript writing and final manuscript approval. G.D. reviewed the manuscript. P.K., J.S. and S.C. were also responsible for conception and design of the study, and reviewed the manuscript.

Funding

S.T. is supported by the Anniversary Fund of the Österreichische Nationalbank (Project No. 16442).

Competing Interests

The authors declare that there are no competing interests associated with the manuscript.

Abbreviations

     
  • BMI

    body mass index

  •  
  • ECM

    extra cellular matrix

  •  
  • ER

    endoplasmic reticulum

  •  
  • FDR

    false discovery rate

  •  
  • fpEC

    feto-placental endothelial cells

  •  
  • GDM

    gestational diabetes mellitus

  •  
  • GLM

    generalized linear model

  •  
  • HUVEC

    human umbilical vein endothelial cells

  •  
  • miRNA

    micro RNA

  •  
  • oGTT

    oral glucose tolerance test

  •  
  • PCA

    principle component analysis

  •  
  • T2D

    Type 2 diabetes

References

References
1
Friedman
R.C.
,
Farh
K.K.
,
Burge
C.B.
and
Bartel
D.P.
(
2009
)
Most mammalian mRNAs are conserved targets of microRNAs
.
Genome Res.
19
,
92
105
[PubMed]
2
Santulli
G.
(
2016
)
MicroRNAs and endothelial (Dys) function
.
J. Cell. Physiol.
231
,
1638
1644
[PubMed]
3
De Vriese
A.S.
,
Verbeuren
T.J.
,
Van de Voorde
J.
,
Lameire
N.H.
and
Vanhoutte
P.M.
(
2000
)
Endothelial dysfunction in diabetes
.
Br. J. Pharmacol.
130
,
963
974
[PubMed]
4
Versari
D.
,
Daghini
E.
,
Virdis
A.
,
Ghiadoni
L.
and
Taddei
S.
(
2009
)
Endothelial dysfunction as a target for prevention of cardiovascular disease
.
Diabetes Care
32
,
S314
S321
[PubMed]
5
Shantikumar
S.
,
Caporali
A.
and
Emanueli
C.
(
2012
)
Role of microRNAs in diabetes and its cardiovascular complications
.
Cardiovasc. Res.
93
,
583
593
[PubMed]
6
Prattichizzo
F.
,
Giuliani
A.
,
Ceka
A.
,
Rippo
M.R.
,
Bonfigli
A.R.
,
Testa
R.
et al.  (
2015
)
Epigenetic mechanisms of endothelial dysfunction in type 2 diabetes
.
Clin. Epigenetics
7
,
56
[PubMed]
7
Sobrevia
L.
,
Abarzua
F.
,
Nien
J.K.
,
Salomon
C.
,
Westermeier
F.
,
Puebla
C.
et al.  (
2011
)
Review: differential placental macrovascular and microvascular endothelial dysfunction in gestational diabetes
.
Placenta
32
,
S159
S164
[PubMed]
8
Jirkovska
M.
,
Kubinova
L.
,
Janacek
J.
,
Moravcova
M.
,
Krejci
V.
and
Karen
P.
(
2002
)
Topological properties and spatial organization of villous capillaries in normal and diabetic placentas
.
J. Vasc. Res.
39
,
268
278
[PubMed]
9
Teasdale
F.
(
1981
)
Histomorphometry of the placenta of the diabetic women: class A diabetes mellitus
.
Placenta
2
,
241
251
[PubMed]
10
Babawale
M.O.
,
Lovat
S.
,
Mayhew
T.M.
,
Lammiman
M.J.
,
James
D.K.
and
Leach
L.
(
2000
)
Effects of gestational diabetes on junctional adhesion molecules in human term placental vasculature
.
Diabetologia
43
,
1185
1196
[PubMed]
11
Figueroa
R.
,
Martinez
E.
,
Fayngersh
R.P.
,
Tejani
N.
,
Mohazzab
H.K.
and
Wolin
M.S.
(
2000
)
Alterations in relaxation to lactate and H(2)O(2) in human placental vessels from gestational diabetic pregnancies
.
Am. J. Physiol. Heart Circ. Physiol.
278
,
H706
H713
[PubMed]
12
Floris
I.
,
Descamps
B.
,
Vardeu
A.
,
Mitic
T.
,
Posadino
A.M.
,
Shantikumar
S.
et al.  (
2015
)
Gestational diabetes mellitus impairs fetal endothelial cell functions through a mechanism involving microRNA-101 and histone methyltransferase enhancer of zester homolog-2
.
Arterioscler. Thromb. Vasc. Biol.
35
,
664
674
[PubMed]
13
Tryggestad
J.B.
,
Vishwanath
A.
,
Jiang
S.
,
Mallappa
A.
,
Teague
A.M.
,
Takahashi
Y.
et al.  (
2016
)
Influence of gestational diabetes mellitus on human umbilical vein endothelial cell miRNA
.
Clin. Sci.
130
,
1955
1967
[PubMed]
14
Lupu
C.
,
Zhu
H.
,
Popescu
N.I.
,
Wren
J.D.
and
Lupu
F.
(
2011
)
Novel protein ADTRP regulates TFPI expression and function in human endothelial cells in normal conditions and in response to androgen
.
Blood
118
,
4463
4471
[PubMed]
15
Perez-Lopez
F.R.
,
Larrad-Mur
L.
,
Kallen
A.
,
Chedraui
P.
and
Taylor
H.S.
(
2010
)
Gender differences in cardiovascular disease: hormonal and biochemical influences
.
Reprod. Sci.
17
,
511
531
[PubMed]
16
Maric-Bilkan
C.
(
2017
)
Sex differences in micro- and macro-vascular complications of diabetes mellitus
.
Clin. Sci.
131
,
833
846
[PubMed]
17
Zhang
M.
,
Mao
J.
,
Tuerdi
A.
,
Zeng
X.
,
Quan
L.
,
Xiao
S.
et al.  (
2018
)
The constellation of macrovascular risk factors in early onset T2DM: A Cross-Sectional Study in Xinjiang Province, China
.
J. Diabetes Res.
2018
,
3089317
[PubMed]
18
Penaloza
C.
,
Estevez
B.
,
Orlanski
S.
,
Sikorska
M.
,
Walker
R.
,
Smith
C.
et al.  (
2009
)
Sex of the cell dictates its response: differential gene expression and sensitivity to cell death inducing stress in male and female cells
.
FASEB J.
23
,
1869
1879
[PubMed]
19
Carruth
L.L.
,
Reisert
I.
and
Arnold
A.P.
(
2002
)
Sex chromosome genes directly affect brain sexual differentiation
.
Nat. Neurosci.
5
,
933
934
[PubMed]
20
Cvitic
S.
,
Longtine
M.S.
,
Hackl
H.
,
Wagner
K.
,
Nelson
M.D.
,
Desoye
G.
et al.  (
2013
)
The human placental sexome differs between trophoblast epithelium and villous vessel endothelium
.
PLoS One
8
,
e79233
[PubMed]
21
Metzger
B.E.
,
Gabbe
S.G.
,
Persson
B.
,
Buchanan
T.A.
,
International Association of D, Pregnancy Study Groups Consensus P
et al.  (
2010
)
International association of diabetes and pregnancy study groups recommendations on the diagnosis and classification of hyperglycemia in pregnancy
.
Diabetes Care
33
,
676
682
[PubMed]
22
Lang
I.
,
Schweizer
A.
,
Hiden
U.
,
Ghaffari-Tabrizi
N.
,
Hagendorfer
G.
,
Bilban
M.
et al.  (
2008
)
Human fetal placental endothelial cells have a mature arterial and a juvenile venous phenotype with adipogenic and osteogenic differentiation potential
.
Differentiation
76
,
1031
1043
[PubMed]
23
Staedtler
F.
,
Hartmann
N.
,
Letzkus
M.
,
Bongiovanni
S.
,
Scherer
A.
,
Marc
P.
et al.  (
2013
)
Robust and tissue-independent gender-specific transcript biomarkers
.
Biomarkers
18
,
436
445
24
Sun
Z.
,
Evans
J.
,
Bhagwate
A.
,
Middha
S.
,
Bockol
M.
,
Yan
H.
et al.  (
2014
)
CAP-miRSeq: a comprehensive analysis pipeline for microRNA sequencing data
.
BMC Genomics
15
,
423
[PubMed]
25
Robinson
T.J.
,
Dinan
M.A.
,
Dewhirst
M.
,
Garcia-Blanco
M.A.
and
Pearson
J.L.
(
2010
)
SplicerAV: a tool for mining microarray expression data for changes in RNA processing
.
BMC Bioinformatics
11
,
108
[PubMed]
26
Sturn
A.
,
Quackenbush
J.
and
Trajanoski
Z.
(
2002
)
Genesis: cluster analysis of microarray data
.
Bioinformatics
18
,
207
208
[PubMed]
27
Nolan
T.
,
Hands
R.E.
and
Bustin
S.A.
(
2006
)
Quantification of mRNA using real-time RT-PCR
.
Nat. Protoc.
1
,
1559
1582
[PubMed]
28
Schwarzenbach
H.
,
da Silva
A.M.
,
Calin
G.
and
Pantel
K.
(
2015
)
Data normalization strategies for MicroRNA quantification
.
Clin. Chem.
61
,
1333
1342
[PubMed]
29
Vlachos
I.S.
,
Zagganas
K.
,
Paraskevopoulou
M.D.
,
Georgakilas
G.
,
Karagkouni
D.
,
Vergoulis
T.
et al.  (
2015
)
DIANA-miRPath v3.0: deciphering microRNA function with experimental support
.
Nucleic Acids Res.
43
,
W460
W466
[PubMed]
30
Vlachos
I.S.
,
Paraskevopoulou
M.D.
,
Karagkouni
D.
,
Georgakilas
G.
,
Vergoulis
T.
,
Kanellos
I.
et al.  (
2015
)
DIANA-TarBase v7.0: indexing more than half a million experimentally supported miRNA:mRNA interactions
.
Nucleic Acids Res.
43
,
D153
D159
[PubMed]
31
Zampetaki
A.
,
Kiechl
S.
,
Drozdov
I.
,
Willeit
P.
,
Mayr
U.
,
Prokopi
M.
et al.  (
2010
)
Plasma microRNA profiling reveals loss of endothelial miR-126 and other microRNAs in type 2 diabetes
.
Circ. Res.
107
,
810
817
[PubMed]
32
Olivieri
F.
,
Spazzafumo
L.
,
Bonafe
M.
,
Recchioni
R.
,
Prattichizzo
F.
,
Marcheselli
F.
et al.  (
2015
)
MiR-21-5p and miR-126a-3p levels in plasma and circulating angiogenic cells: relationship with type 2 diabetes complications
.
Oncotarget
6
,
35372
35382
[PubMed]
33
Li
X.X.
,
Liu
Y.M.
,
Li
Y.J.
,
Xie
N.
,
Yan
Y.F.
,
Chi
Y.L.
et al.  (
2016
)
High glucose concentration induces endothelial cell proliferation by regulating cyclin-D2-related miR-98
.
J. Cell. Mol. Med.
20
,
1159
1169
[PubMed]
34
Olivieri
F.
,
Bonafe
M.
,
Spazzafumo
L.
,
Gobbi
M.
,
Prattichizzo
F.
,
Recchioni
R.
et al.  (
2014
)
Age- and glycemia-related miR-126-3p levels in plasma and endothelial cells
.
Aging
6
,
771
787
[PubMed]
35
Lizarraga
D.
,
Huen
K.
,
Combs
M.
,
Escudero-Fung
M.
,
Eskenazi
B.
and
Holland
N.
(
2016
)
miRNAs differentially expressed by next-generation sequencing in cord blood buffy coat samples of boys and girls
.
Epigenomics
8
,
1619
1635
[PubMed]
36
Ameling
S.
,
Kacprowski
T.
,
Chilukoti
R.K.
,
Malsch
C.
,
Liebscher
V.
,
Suhre
K.
et al.  (
2015
)
Associations of circulating plasma microRNAs with age, body mass index and sex in a population-based study
.
BMC Med. Genet.
8
,
61
37
Gross
N.
,
Kropp
J.
and
Khatib
H.
(
2017
)
Sexual dimorphism of miRNAs secreted by bovine in vitro-produced embryos
.
Front. Genet.
8
,
39
[PubMed]
38
Link
J.C.
,
Hasin-Brumshtein
Y.
,
Cantor
R.M.
,
Chen
X.
,
Arnold
A.P.
,
Lusis
A.J.
et al.  (
2017
)
Diet, gonadal sex, and sex chromosome complement influence white adipose tissue miRNA expression
.
BMC Genomics
18
,
89
[PubMed]
39
Mentzel
C.M.
,
Anthon
C.
,
Jacobsen
M.J.
,
Karlskov-Mortensen
P.
,
Bruun
C.S.
,
Jorgensen
C.B.
et al.  (
2015
)
Gender and obesity specific microRNA expression in adipose tissue from lean and obese pigs
.
PLoS One
10
,
e0131650
[PubMed]
40
Fornes
D.
,
White
V.
,
Higa
R.
,
Heinecke
F.
,
Capobianco
E.
and
Jawerbaum
A.
(
2018
)
Sex-dependent changes in lipid metabolism, PPAR pathways and microRNAs that target PPARs in the fetal liver of rats with gestational diabetes
.
Mol. Cell. Endocrinol.
461
,
12
21
[PubMed]
41
Edlow
A.G.
,
Guedj
F.
,
Pennings
J.L.
,
Sverdlov
D.
,
Neri
C.
and
Bianchi
D.W.
(
2016
)
Males are from Mars, and females are from Venus: sex-specific fetal brain gene expression signatures in a mouse model of maternal diet-induced obesity
.
Am. J. Obstet. Gynecol.
214
,
623e1
623e10
42
Morgan
C.P.
and
Bale
T.L.
(
2017
)
Sex differences in microRNA-mRNA networks: examination of novel epigenetic programming mechanisms in the sexually dimorphic neonatal hypothalamus
.
Biol. Sex Differ.
8
,
27
43
Muralimanoharan
S.
,
Guo
C.
,
Myatt
L.
and
Maloyan
A.
(
2015
)
Sexual dimorphism in miR-210 expression and mitochondrial dysfunction in the placenta with maternal obesity
.
Int. J. Obes.
39
,
1274
1281
[PubMed]
44
Muralimanoharan
S.
,
Li
C.
,
Nakayasu
E.S.
,
Casey
C.P.
,
Metz
T.O.
,
Nathanielsz
P.W.
et al.  (
2017
)
Sexual dimorphism in the fetal cardiac response to maternal nutrient restriction
.
J. Mol. Cell Cardiol.
108
,
181
193
[PubMed]
45
Wilson
P.
,
Drennon
K.
and
Tannock
L.R.
(
2007
)
Regulation of vascular proteoglycan synthesis by metabolic factors associated with diabetes
.
J. Investig. Med.
55
,
18
25
[PubMed]
46
Chen
C.P.
,
Chang
S.C.
and
Vivian Yang
W.C.
(
2007
)
High glucose alters proteoglycan expression and the glycosaminoglycan composition in placentas of women with gestational diabetes mellitus and in cultured trophoblasts
.
Placenta
28
,
97
106
[PubMed]
47
Hiebert
L.M.
(
2017
)
Proteoglycans and diabetes
.
Curr. Pharm Des.
23
,
1500
1509
[PubMed]
48
Klausner
R.D.
and
Sitia
R.
(
1990
)
Protein degradation in the endoplasmic reticulum
.
Cell
62
,
611
614
[PubMed]
49
Galan
M.
,
Kassan
M.
,
Kadowitz
P.J.
,
Trebak
M.
,
Belmadani
S.
and
Matrougui
K.
(
2014
)
Mechanism of endoplasmic reticulum stress-induced vascular endothelial dysfunction
.
Biochim. Biophys. Acta
1843
,
1063
1075
[PubMed]
50
Ozcan
U.
,
Cao
Q.
,
Yilmaz
E.
,
Lee
A.H.
,
Iwakoshi
N.N.
,
Ozdelen
E.
et al.  (
2004
)
Endoplasmic reticulum stress links obesity, insulin action, and type 2 diabetes
.
Science
306
,
457
461
[PubMed]
51
Shruthi
K.
,
Reddy
S.S.
and
Reddy
G.B.
(
2017
)
Ubiquitin-proteasome system and ER stress in the retina of diabetic rats
.
Arch. Biochem. Biophys.
627
,
10
20
[PubMed]
52
Yu
F.X.
,
Zhao
B.
and
Guan
K.L.
(
2015
)
Hippo pathway in organ size control, tissue homeostasis, and cancer
.
Cell
163
,
811
828
[PubMed]
53
Xu
S.
,
Koroleva
M.
,
Yin
M.
and
Jin
Z.G.
(
2016
)
Atheroprotective laminar flow inhibits Hippo pathway effector YAP in endothelial cells
.
Transl. Res.
176
,
18e2
28e2
54
Yuan
L.
,
Mao
Y.
,
Luo
W.
,
Wu
W.
,
Xu
H.
,
Wang
X.L.
et al.  (
2017
)
Palmitic acid dysregulates the Hippo-YAP pathway and inhibits angiogenesis by inducing mitochondrial damage and activating the cytosolic DNA sensor cGAS-STING-IRF3 signaling
.
J. Biol. Chem.
,
212
,
15002
15015
,
55
Mordwinkin
N.M.
,
Ouzounian
J.G.
,
Yedigarova
L.
,
Montoro
M.N.
,
Louie
S.G.
and
Rodgers
K.E.
(
2013
)
Alteration of endothelial function markers in women with gestational diabetes and their fetuses
.
J. Matern. Fetal Neonatal Med.
26
,
507
512
[PubMed]
56
Retnakaran
R.
,
Ye
C.
,
Hanley
A.
,
Sermer
M.
,
Connelly
P.
,
Zinman
B.
et al.  (
2013
)
Effect of maternal gestational diabetes on the cardiovascular risk factor profile of infants at 1 year of age
.
Nutr. Metab. Cardiovasc. Dis.
23
,
1175
1181
57
West
N.A.
,
Crume
T.L.
,
Maligie
M.A.
and
Dabelea
D.
(
2011
)
Cardiovascular risk factors in children exposed to maternal diabetes in utero
.
Diabetologia
54
,
504
507
[PubMed]
58
Manderson
J.G.
,
Mullan
B.
,
Patterson
C.C.
,
Hadden
D.R.
,
Traub
A.I.
and
McCance
D.R.
(
2002
)
Cardiovascular and metabolic abnormalities in the offspring of diabetic pregnancy
.
Diabetologia
45
,
991
996
[PubMed]
59
Katkhuda
R.
,
Peterson
E.S.
,
Roghair
R.D.
,
Norris
A.W.
,
Scholz
T.D.
and
Segar
J.L.
(
2012
)
Sex-specific programming of hypertension in offspring of late-gestation diabetic rats
.
Pediatr. Res.
72
,
352
361
[PubMed]
60
Benetatos
L.
,
Hatzimichael
E.
,
Londin
E.
,
Vartholomatos
G.
,
Loher
P.
,
Rigoutsos
I.
et al.  (
2013
)
The microRNAs within the DLK1-DIO3 genomic region: involvement in disease pathogenesis
.
Cell. Mol. Life Sci.
70
,
795
814
[PubMed]
61
Dimmeler
S.
and
Yla-Herttuala
S.
(
2014
)
14q32 miRNA cluster takes center stage in neovascularization
.
Circ. Res.
115
,
680
682
[PubMed]
62
Martinez-Micaelo
N.
,
Beltran-Debon
R.
,
Aragones
G.
,
Faiges
M.
and
Alegret
J.M.
(
2017
)
MicroRNAs clustered within the 14q32 locus are associated with endothelial damage and microparticle secretion in bicuspid aortic valve disease
.
Front. Physiol.
8
,
648
[PubMed]
63
Pollard
K.
,
Dudoit
S.
,
van der Laan
M.
,
2005
.
Multiple testing procedures: the multtest package and applications to genomics
. In
Bioinformatics and Computational Biology Solutions Using R and Bioconductor
, pp.
249
271
,
Springer
64
Karagkouni
D.
,
Paraskevopoulou
M.D.
,
Chatzopoulos
S.
,
Vlachos
I.S.
,
Tastsoglou
S.
,
Kanellos
I.
et al.  (
2018
)
DIANA-TarBase v8: a decade-long collection of experimentally supported miRNA-gene interactions
.
Nucleic Acids Res.
46
,
D239
D245
[PubMed]

Author notes

*

These authors contributed equally to this work.